THE GIANT A F R I C A N SNAIL THE G I A N T AFRICAN SNAIL: A PROBLEM IN ECONOMIC MALACOLOGY A L B E R T R. MEAD THE UNIVERSITY OF CHICAGO PRESS LIBRARY OF CONGRESS CATALOG CARD NUMBER: 61-14949 THE UNIVERSITY OF CHICAGO PRESS, CHICAGO & LONDON THE UNIVERSITY OF TORONTO PRESS, TORONTO 5, CANADA © 1961 BY THE UNIVERSITY OF CHICAGO. PUBLISHED 1961 COMPOSED AND PRINTED BY THE UNIVERSITY OF CHICAGO PRESS CHICAGO, ILLINOIS, U.S.A. TO MY W I F E ELEANOR W H O WAS MY F I R S T AND B E S T S T U D E N T PREFACE There is no comprehensive work in any language encompassing the area of our knowledge that can be defined as Economic Malacology. In fact the subject is just beginning to be recognized as a discipline in its own right. The mission of the present work is to bring together its widely scattered literature, which goes all the way from typewritten, mimeographed, or dittographed official reports to publications in obscure foreign journals. It has taken over ten years of searching, examining, evaluating, digesting, collating, and editing to bring this work to its present status. During this process many essentially unimportant or irrelevant papers were encountered, particularly in the area of snail control. In nearly every case these were not included in the bibliography because either they made little in the way of an original contribution or they contained information not relevant to the general problem of the economics of terrestrial gastropods. To this extent, the bibliography is a selected bibliography. And, in a broader sense, the subject of economic malacology embraces the problems presented by the freshwater snails. Such problems and the literature that treats of them, however, are vast indeed-- so vast that it is completely impractical to consider them here, except for an occasional reference. Also, the problems and their solutions vary so much between the two groups of snails that there is little in the way of common applicability or of comparative value. Further, because many of the problems involving freshwater snails have a distinct medical flavor, far greater attention has been given to them-- witness, for example, the vast literature concerning the snail hosts of the schistosomiases. Because the original sources of information will not be available to the greater segment of the readers, the essence of the contributions, in each case, has been incorporated in the text in so far as it has been practicable to do so. Some authors quoted or referred to in this work may be judged as not being truly qualified to make sound reports. It should be borne in mind, therefore, that the inclusion of certain of vii viii PREFACE these works was for the completeness of the record; inclusion is not an indorsement. Where it is believed that a given author is in error, this is pointed out. But under no circumstances was any reference excluded simply on the basis of its being at variance with the conclusions set forth in this book. The pertinent information in the literature has been combined with considerable field data which I collected in Africa, Asia, and Oceania. Supplemental information has been obtained from many correspondents who have had firsthand experience with terrestrial snail problems. Throughout this book, credit is given on the spot where credit is due; and sincere efforts have been made not to imply by omission that originality rested with me, even though I may have independently arrived at the same conclusion. The reader therefore can trace back to the original source almost any item of information. No matter how hard one tries, inevitable errors, omissions, and misinterpretations will creep into any work of this size. These are as regrettable as they are unavoidable. Certainly every effort has been made to keep them to a minimum. New contributions in the field of economic malacology are continuing to appear in the literature just often enough that it has not been possible to keep pace with them right up to the last minute. As a compromise, reference to recent, pertinent works is made only in the bibliography. It has been a temptation to prolong the editing of the manuscript in the expectation that all points of difficulty would be removed. But it is much more realistic and practical to get all of this information as soon as possible into the hands of those who need it and have been asking for it for years. Hence, with a good measure of apprehension and with a full awareness of the shortcomings involved, this work is being released in the fond hope, not that it will serve the needs perfectly, but that it will serve them well. During the years that it has taken to produce this work, there has accumulated a vast indebtedness in a multitude of ways; and the task of attempting to indicate here anything approaching the gratitude that is felt, seems overwhelming and irrevocably foredoomed to inadequacy. Those for whom I feel the greatest gratitude, I would like to mention: Dr. Yoshio Kondo of the B. P. Bishop Museum, whose sound j u d g m e n t has tempered many ideas presented in this book, and whose companionship in the field is without equal; Dr. Joseph C. Bequaert of Harvard University, whose interest, encouragement, and advice in Africa sixteen years ago provided a great turning point in my life; Dr. Harold J. Coolidge, executive director of the Pacific Science PREFACE ix Board, whose faith in my research, when others grew doubtful, gave me encouragement when I needed it most; Dr. George A. Baitsell of Yale University, whose advice precipitated two $500 Sigma Xi--RESA grants-in-aid which made it possible to prepare the greater share of the original manuscript; Dr. W. Wayne Boyle of Pennsylvania State University, whose many months of help in Hawaii brought forth some of the most valuable data in this work concerning the great need for, and the value of, long-range studies on the giant African snail; Dr. W. Harry Lange, Jr., of the University of California, who critically examined the chapter on chemical control and offered many profitable suggestions and leads in this complex aspect of the subject; and A. P. Messenger and H. M. Armitage of the California State Department of Agriculture, who have been my constant guides in all matters involving quarantines. Grant funds from several sources have permitted me to gather field data and conduct research without which the present work would not have been conceived. The Pacific Science Board of the National Research Council, through the National Academy of Science, provided ONR funds in 1948 to study the anatomy of a number of achatinids at Harvard University; and additional funds were allotted in 1957 to permit me to chair a symposium on the giant African snail at the Ninth Pacific Science Congress in Bangkok, Thailand. The Office of Naval Research (NR 161 472) paid for all expenses in an extensive survey of the problem of the giant African snail in the Trust Territory of the Pacific Islands during the three summer months of 1949. The National Science Foundation made one of its early grants (G-519) in 1953 in support of the successful search in Ceylon for a predicted disease in the giant African snail. Dr. A. D. Ross, Honorable Secretary of the Pan Indian Ocean Science Association, made it possible for me to attend the Second Pan Indian Ocean Science Congress in Perth, Australia, in 1954. The National Institutes of Health have provided funds (E-1245[C3j) since 1957 which have financed considerable research at the University of Arizona and the University of Hawaii on the disease syndrome in the giant African snail. Ada P. McCormick of Tucson, Arizona, furnished funds and secretarial help which were indispensable in the early stages of preparing the manuscript. Edward A. Steinhaus and Yoshinori Tanada of the University of California patiently discussed at length with me my proposed research and hypotheses on the disease syndrome in the giant African snail. Chemical and bio-assays of the snail meal were conducted by A r t h u r R. Kemmerer, Mitchell G. Vavich, and Edward L. Breazeale; x PREFACE and the more complex mathematical computations were made by Donald L. Webb and Samuel R. Browning--all of the University of Arizona. In Hawaii, the following provided all sorts of help in many ways over a period of several years: Henry A. Bess, George D. Butler, Walter Carter, Q. C. Chock, C. H. Edmondson, Jim Kim, C. E. Pemberton, Alexander Spoehr, Alan D. Thistle, and the late Paul W. Weber. Many investigators in foreign countries generously contributed, among other things, valuable information which helped tremendously in piecing together the otherwise exasperatingly fragmentary data in some of the literature. The following were of especial help: A. F. Caldwell of the University of Malaya in Singapore; Silverio M. Cendana of the University of the Philippines; R. E. Dean, superintendent of gardens, Hong Kong; G. S. Dun of the Lowlands Experiment Station, New Britain; K. C. Ghose of the City College of Calcutta; John R. Hendrickson of the University of Malaya in Kuala Lumpur; Mr. and Mrs. Peter J. R. Hill of Koror, Palau Islands; Alan J. Kohn of Florida State University, who supplied information from the Maldive Islands during the 1957 Yale-Seychelles Expedition; J. C. van der Meer Mohr of the Deli Proefstation, Medan, Sumatra; and George D. Peterson, Jr., the entomologist of the government of Guam. During the five weeks that Yoshio Kondo, the late Dan Langford, and I were in the Bonin Islands in 1949, we became greatly indebted to the people of Chichi Jima and, particularly, to the head councilman, Roderick Webb, and the entire Savory family. For nine months in Ceylon, I received help from many more than I could possibly list here; but I must acknowledge help particularly from the following members of the Department of Agriculture in Peradeniya: Henry E. Fernando and Yasatileka Elikawela of the Division of Entomology, Dr. J. W. L. Peiris of the Division of Plant Pathology; and A. Bandaranyake and W. Fernando of the Veterinary Research Laboratory. Assistance in translating from the Dutch was given by L. J. M. Butot of the Museum Zoologicum Bogoriense in Java and by H. Reerink, agronomist of FAO in Ceylon. M. Dale Arvey of Long Beach State College translated one long, important article from the Japanese; and William Osuga of the University of California East Asiatic Library gave considerable help in locating and translating portions of several other articles in Japanese. Donald M. Powell and Lutie L. Higley of the University of Arizona Library patiently ferreted out a number of elusive references. I am indebted to several of my colleagues for the use of photo- PREFACE xi graphs. Specific credit has been given in each case where the illustration is not mine. I am especially indebted to Donald B. Sayner, who is the instructor in scientific illustration in the Department of Zoology at the University of Arizona, for the two excellent line drawings and to Robert Broder, the departmental photographer, for the greater share of the photographic work. For the jacket photograph, I wish to thank Drs. William J. Clench and Ruth D. Turner of the Harvard Museum of Comparative Zoology. My sister Jennie S. Burnett and Virginia A. Miles typed the early drafts of the manuscript; and Ruth I. Spiller miraculously typed the entire final draft in just a few days. Charles D. Miles and Robert J. Drake cheerfully assisted in the long, monotonous task of proofreading the several drafts, copies, and proofs. The constant help of my wife and the willing hands and sharp eyes of Ruth and Jim in the field have been my greatest blessings in the long task that now lies behind me. xiv CONTENTS THAILAND / 15 VIETNAM / l6 III. Factors Favoring Dispersal and Survival MAN--THE PRINCIPAL AGENT OF DISPERSAL / 17 LONGEVITY / 21 REPRODUCTIVITY / 21 VARIABILITY / 24 ESTIVATION AND HIBERNATION / HARDINESS / 28 PAUCITY OF NATURAL ENEMIES / PERIOD OF ACTIVITY / 33 33 34 36 AVAILABILITY OF CALCIUM / 32 25 17 ESCAPE REACTION AND MIGRATION CYCLE / IV. Economic Status CURRENT OPINION / 36 DAMAGE TO PLANTS / 39 EVALUATION OF DAMAGE / 49 INDIRECT DAMAGE / 52 NUISANCE FACTOR / 54 HEALTH FACTOR / 55 ROLE AS SCAVENGER / 57 UN JUST CHARGES / 59 V. Chemical Control CHEMICALS AND COMPOUNDS / 62 CHEMICAL CONTROL--AN EVALUATION / 84 VI. Control through Mechanical Devices BARRIERS / 92 BURNING-OVER / 94 CLEAN CULTURE / 95 DROWNING / 97 HAND COLLECTING AND DESTROYING / 98 SHIFTING CULTIVATION / 1OO TRAPS / 1OO VII. Biological Control AMPHIBIANS / 1O2 ANTS / 105 61 92 102 CONTENTS xv BEETLES / 1O6 BIRDS / CRABS / 113 1l6 FLIES / 118 HELMINTHS / 12O MAMMALS / 121 MICRO-ORGANISMS / 123 MITES / 125 REPTILES / 125 SNAILS / 126 MISCELLANEOUS / 136 BIOLOGICAL CONTROL--AN EVALUATION / 136 VIII. Control through Human Use H U M A N CONSUMPTION / 146 155 156 REPUTED POISONOUS PROPERTIES / 146 EXPERIMENTS IN EATING THE GIANT SNAIL / FOOD FOR POULTRY / 159 l6o 164 165 CHEMICAL ASSAY OF SNAIL MEAL / VITAMIN ASSAY OF SNAIL MEAL / AMINO ACID ASSAY OF SNAIL MEAL / FOOD FOR LIVESTOCK / FISH BAIT / 167 FERTILIZER / 167 l68 167 PRODUCTION OF SNAIL MEAL / IX. Control through Legislative Action X. Population Decline POPULATION SENILITY / 182 STERILITY / l82 172 180 STARVATION / 183 EXPOSURE / 183 TRAUMATIC BREAKS / PREDATORS / GENETICS / DISEASE / 184 184 185 193 183 MULTIPLE FACTORS / xvi CONTENTS XI. Outlook DISPERSAL / 195 BIONOMICS / 199 CONTROL / 199 DECLINE / 203 195 Bibliography Index 205 247 CONTENTS List of Illustrations I. Introduction II. Dispersal of the Giant African Snail BISMARCK ARCHIPELAGO / 6 BONIN ISLANDS / 6 BURMA / 7 CAROLINE ISLANDS / 7 CEYLON / 8 CHINA / 8 FORMOSA / 8 HAWAIIAN ISLANDS / HONG KONG / INDIA / 9 INDONESIA / 9 JAPAN / 11 MALAYA / 11 MALDIVE ISLANDS / MAURITIUS / 13 NEW GUINEA / 13 NORTH BORNEO / 13 PALAU ISLANDS / 14 PHILIPPINE ISLANDS / 14 REUNION / 14 R Y U K Y U ISLANDS / 15 SARAWAK / 15 SEYCHELLES / 15 SINGAPORE / 15 11 MARIANA ISLANDS / 11 9 8 xvii 1 4 xiii ILLUSTRATIONS FIGURE FACING PAGE 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. MAP The largest living Achatina fulica Bowdich ever photographed Army Hill population of giant African snail, 1949 Achatina fulica and several possible control snails Giant snails sold and used as human food Giant snail on trunk of Artocarpus heterophyllus Shells showing characteristic damage Variability in Achatina fulica Bowdich The "bent-nose" type of giant African snail The common hermit crab preys on giant African snail Evidence of widespread disease found among older snail populations Thousands of giant snails in Rota tomato patch Dead snails as source of lime Full-grown larva of Lamprophorus tenebrosus feeding on snail Giant snail with two natural enemies Inspectors discover giant snails in Los Angeles 46 47 47 78 78 79 79 79 110 110 111 111 142 142 143 5 137 Dispersion of the giant African snail CHART The giant African snail and its environment xvii CHAPTER 1 INTRODUCTION The study of economic malacology is largely the story of the giant African snail, Achatina fulica Bowdich. More than anything else, the relatively rapid development of this species into the most serious land snail pest plaguing man today has precipitated the need for creating the separate discipline of economic malacology. But although the giant African snail dominates the subject, there are a surprising number of other land snail species, particularly those that have wandered from their native heath, which have become serious and even major garden pests. The problems presented by them will be with us for a long time to come. The giant snail and most other pestiferous species have defied all efforts to eradicate them and they seem to have the capacity to resist control measures indefinitely. Some of these snails are destined to become worse pests in the future. And new ones will make their appearance. For example, the announcement of Torres (1950) that the giant South American snail Strophocheilus oblongus has been attacking the coffee plant seems patently prognosticative. The groundwork from which economic malacology has emerged was laid largely during the multifold campaign to control the giant African snail in the Indo-Pacific region. Initial research was conducted in Southeast Asia, particularly in Ceylon, Malaya, Singapore, and Java. In the Pacific, it was the Insect Control Committee for Micronesia (which later became the Invertebrate Consultants Committee for the Pacific) of the Pacific Science Board of the National Research Council that led the way in the program of delving into some of the deeper problems associated with the economics of land snails, and of the giant African snail in particular. It was not known how far the giant snail had spread; how it was spreading; why it was 2 INTRODUCTION spreading so rapidly; why it was building up in unprecedented numbers; to what extent it was really causing horticultural and agricultural damage; whether it ever practicably could be used for human consumption; or what control measures--chemical, mechanical, biological, legislative--have been attempted with what measure of success. The early investigations led to new and unsuspected findings. The enigmatic phenomenon of population decline has been shown to produce dramatic results in some of the older snail populations. Disease as a factor in natural control has just entered the increasingly more complex ecological picture. Other phenomena are hanging in the near shadows. In an effort to determine the amount of research currently in progress in the problem of the giant African snail, letters of inquiry were sent to all major infested areas, either to known investigators or to directors of agriculture. The response was disheartening in the extreme. With only a few exceptions, the answers presented one or more of a host of good reasons why no work, except perhaps for a small amount of routine control, was being done on the problem. Irrespective of the validity of most of the good reasons, the situation is indeed deplorable, especially when it is compared with that of other types of major agricultural pests. Time after time, however, there could be sensed the tacit confession that, regardless of the excuses offered, too little was actually known or could be discovered about snail problems. They would complain: "We are fighting in the dark." "We do not know what control measures have been tried." "Our experimental results have been inexplicably inconsistent." It is little wonder that many investigators have become discouraged and have actually turned their backs to the problem. With the emergence of the discipline of economic malacology out of the grief caused by the giant African snail, we must look both to the past and to the future. We must survey critically the many and varied contributions regarding both the giant snail and other snails; and we must look at what lies ahead so that we will be in better position to tackle the predictably worsening problems and still other problems yet to come. As a matter of fact, we have right now weapons of control which ironically are not generally known by those who are charged with trying to effect controls. In some cases, discoveries of a complementary nature in research have not yet been brought together. Wholly unpredictable avenues of approach will be discovered when all pertinent information extant is brought together and synthesized. Even to this day, problems in economic malacology are invariably passed on to the economic entomologist, who finds his INTRODUCTION 3 knowledge of malacology too limited (if indeed it exists at all) and the insect control methods almost completely useless to him. This places a premium on putting into the proper hands all available pertinent information on the economics of terrestrial snails. Attempts are made in the following pages to do just this. Now that there has been brought together at last the rather substantial fund of information tucked away in the far corners of the literature, and this has been combined with the discoveries in recent research, it is hoped that investigators in many quarters will be stimulated with renewed enthusiasm to take vigorous steps in seeking the answers to a host of major and fundamental questions surrounding the giant African snail and other land snail pests. When this becomes a reality, economic malacology truly will be on its way. CHAPTER 2 DISPERSAL O F THE GIANT AFRICAN SNAIL Much has been written about the travels that Achatina fulica has made since it left its original home in East Africa. The most outstanding treatises on the subject are: Bequaert (1950b), Lange (1950), Pemberton (1954), Rees (1951), and van Weel (1948) in the English language; Kalshoven (1950) in Dutch; Merle (1949) in French; Boettger (1951) in German; Sonan (1936) and Esaki and Takahashi (1942) in Japanese; and Aguayo (1950) in Spanish. An examination of the rather considerable literature on the subject has revealed many disturbing discrepancies and conflicts. In so far as it is possible to do so, these have been resolved, clarified, or explained in order to arrive at a better understanding of what has taken place and what will probably take place in the next relatively few years. Chief among the items of misinformation being perpetuated in the literature are those which concern the presumed establishment of this snail pest in the "South Pacific," New Caledonia, Yap, the Marshall Islands, the continental United States, and the West Indies. There is also the frequent misstatement that A. fulica has been eradicated in Hawaii. It should be emphasized that the only place in the U n i t e d States where this snail is established is Hawaii, although it has been intercepted and destroyed in Arizona (Mead 1959a, b) and 4 During the nineteenth century the giant African snail moved slowly eastward across the Indian Ocean from its East African home. Upon reaching Southeast Asia, however, it moved at an increasing rapid pace, appearing almost simultaneously in many primary sites of infestation in East Asia and the Pacific Islands. With the advent of the Second World War, it spread to many other Pacific islands and was intercepted alive at American and Australian ports. This snail pest is still on the move and predictably will become established in agriculturally important sites farther east in the next relatively few years. Broken lines indicate interceptions; solid lines indicate establishment; ca. = circa; p. = prior to. (Drawn by D. B. Sayner.) 6 DISPERSAL OF THE GIANT AFRICAN SNAIL in a number of ports in the continental United States. It also should be emphasized that, so far as we know, the giant snail has not become established anywhere in the Marshall Islands. Because people have been alerted to the dangers and the rapid spread of this snail, any large snail has been immediately suspected by the uninitiated as being A. fulica. For example, it was possibly an Amphidromus which caused van Brero (1933) to miscalculate the time A. fulica had been in Java (Leefmans 1933b). A large Vivipara in New York caused an "Achatina invasion" scare which was dramatically played up in the local papers. The campaign in Los Angeles in 1951 to rid itself of a Mediterranean edible snail, Otala lactea, ran headlong into trouble because the announced common name of the snail, "the African snail," created the impression in many parts of the country, through newspaper channels, that the giant African snail had already become established in California (cf. Wibberley 1951). In the following paragraphs there has been set up under each appropriate geographical title, as far as the available information permits, a concise statement of the history of the establishment of the giant African snail, its development, and its present status. Special efforts have been made to avoid any secondary source of reference. Bismarck Archipelago The first record of the giant African snail appearing in New Britain and New Ireland brought the news that this pest was present in great quantities (Anon. 1947a, c}. It seems apparent that the original introductions were made by the Japanese forces prior to 1945 (Anon. 1948e). G. S. Dun of the Lowland Experiment Station at Keravat in New Britain has conducted considerable valuable research on the giant snail in this region. Unfortunately, much of it has not yet found its way into the literature. He reports (in litt. April 21, 1950) that the snail has also become established in the Duke of York Islands. Bonin Islands (Ogasawara Gunto) Mead and Kondo (1949) with the late D. B. Langford made the first survey of these islands to discover the snails well established on Chichi Jima, Haha Jima, and Ani Jima. The snails were introduced in Chichi Jima from Japan in 1937-38 as an item of primitive medicine (see p. 151). Five to six years later the snails were seen quite commonly in the lowlands about the harbor (Futami Ko). Jerry Savory reported that they were p l e n t i f u l in June, 1943, when he was evacuated from Chichi Jima to Japan. On April 2, 1946, a high tidal wave washed over a good portion of these lowlands and for the rest of that year and the succeeding year there were relatively few snails seen. In 1948, however, the DISPERSAL OF THE GIANT AFRICAN SNAIL 7 snails appeared in unprecedented numbers and in 1949 they were still more abundant (Mead 1950b). Ten years later the snails were reported to be continuing as a serious pest. Wilson Savory stated that the snails were introduced on Haha Jima soon after they were introduced on Chichi Jima. The farmers on that island, however, soon recognized that the snails were a pest and they did everything they could to prevent their spread. The island was completely evacuated in 1944. The absence of man as a disseminating agent and the presence of the hermit crab, Cenobita perlatus, as a predator probably account for the fact that the elevento twelve-year-old population was found in 1949 to be limited to a surprisingly small area in the abandoned port village of Okimura. The Ani Jima population in 1949 appeared to be limited to the southwest portion of the island. The giant snails introduced on the tiny island of Higashi Shima were reportedly wiped out by the large endemic population of hermit crabs. The fishermen of Chichi Jima reported that in their numerous trips to Muko Jima and Ototo Jima, no giant snail had ever been seen; and they seriously questioned that the snails had ever been introduced in these islands. Burma Essentially nothing is known of the status of the giant African snail in Burma except that it has been established there for a number of years. The infestation started either from the nearby populations in Bengal or, more likely, from Malaya through the agency of the Japanese occupation forces. Caroline Islands According to Esaki and Takahashi (1942), the giant snails were first introduced into Ponape, near the village of Kolonia, in the fall of 1938; the specimens were reportedly brought from Okinawa by a Mr. Junki Miyahira. By 1949, the snail was found well established in a number of localities in the vicinity of Kolonia (Mead 1950b). At that time, searching parties found no sign of the giant snail on the nearby islands of Pingelap, Mokil, or Kusaie. Dublon was apparently the first of the Truk Islands to become infested (Townes 1946). The snails were probably brought in prior to 1940; and because of the concentration of health resorts on this island, it is possible that the snails were intended to be used for their presumed medicinal properties (Mead 1950b). As nearly as could be determined from questioning the Trukees, the giant snails were introduced into Moen and Romalum from Dublon by the Japanese some time prior to 1945; but they did not show up in Uman, near Sapota, and Fefan, near Sannuk, until 1948. Esaki and Takahashi (1942) were responsible for the often quoted, erroneous report that the giant snails are on Yap Island. 8 DISPERSAL OF THE GIANT AFRICAN SNAIL Ceylon In 1900 Oliver Collett, an enthusiastic conchologist, introduced into his estate at Rozelle (near Watawala) specimens of the giant snail which he obtained either from Mauritius or India (Green 1910b). On the advice of E. E. Green, as many specimens as possible were collected and destroyed, but not before some had accidentally been carried on vegetables to the coastal town of Kalutara. In ten years' time, the "Kalutara snail" (Singhalese: gombela or golubela) was found so well established in both the original sites and adjacent areas that eradication was considered impossible. Today it is found in every province in Ceylon (Mead 1955b). China Herklots (1948) found full-grown specimens of the giant snail in the Amoy University compound in June, 1931. Upon inquiry, he learned that they had been brought on plants imported from Singapore. Jarrett (1931, 1949) identified the specimens and made the official announcement. No recent word has been received about this infestation; but it is more than likely that the snails have spread from Amoy Island to the mainland. From the initial mainland infestation, other sites of infestation inevitably will become established. Formosa A dozen specimens of the giant African snail were introduced into this island from Singapore by a Formosan government official, Mr. Kumaichi Shimojo, in January, 1932. Because of the cold weather and improper rearing techniques, all the specimens died. In April, 1933, twelve additional specimens were brought in and these survived and propagated to form breeding stock for subsequent introductions to Japan and Micronesia (Esaki and Takahashi 1942). By 1940, the giant snail had spread almost throughout the island (Kaburaki 1940). Vosburgh (1950) made interesting observations on this snail in the interior of Formosa. Hawaiian Islands Pemberton (1938) has given us a complete and detailed report of the introduction of the giant snail on Oahu and Maui. In 1936, a young lady returning from Formosa brought two specimens of the giant snail in her baggage and released them in her Oahu garden for aesthetic reasons. The specimens were neither declared nor discovered by the port inspectors. In November of that same year, a man imported specimens through the mails from Japan to breed them in his home town of Makawao in Maui and to sell them as materia medica. Esaki and Takahashi (1942) were not correct in indicating that the imported specimens came from Okinawa. The infestations were not discovered until June, 1938 (Fullaway 1939), and in spite of intensive and continuous control measures, eradication proved impossible, notwithstanding the frequent state- DISPERSAL OF THE GIANT AFRICAN SNAIL 9 ments in the literature to the contrary. On Oahu, the snail quickly spread from Pauoa Valley to Kaneohe on the windward side; and on Maui it spread from Pauwela Gulch to other parts of the island. By 1944, there were eight foci on Oahu and six on Maui (Fullaway 1941, et seq.). These populations were successfully corralled for several years; but by 1951, new foci many miles away were making their appearance (Wong 1951, Thistle 1953b, Weber 1954). Today, the populations on Oahu are numerous and scattered. The infestation on Roundtop near Honolulu is clearly among the most dense on record (Mead 1959b). Although snails were found in the possession of two men on Lanai, there has been no report of an infestation on that island (Pemberton et al. 1939). In March, 1958, however, these snails were found well established near Mana in the northwest tip of Kauai; and two months later an infestation was first noticed near Hilo on Hawaii (Mead 1958a, b, 1959b; Thistle 1959a). Hong Kong On April 3, 1941, Jarrett positively identified as A. fulica several snail specimens from the Happy Valley--Sookunpoo area in Hong Kong Island (Herklots 1948). It was his belief that the pest had been brought from Amoy, China, by Chinese duck farmers four years previously. This would establish the date of the original infestation as 1937 (Jarrett 1949). Within a decade, the infestation had spread to a number of places on the mainland (Dean 1950). India In February, 1847, W. H. Benson (1858) collected specimens of A. fulica in Mauritius and released them the following April in the Chouringhie Gardens near Calcutta. Later he learned that they had spread a great deal. He also learned that all the specimens released by Captain Hutton in Mussoorie (Masuri) at 6,600 feet altitude at the foot of the Himalayas in the Dehra Dun district of the United Provinces died because of the severe winter. By 1907 they were common in the gardens of Calcutta (Annandale), and by 1910 the infestation reportedly had spread all over northern Bengal and new foci had developed in Bombay (Green 1910c, d) and further north in Rajkot (Comber). Ghose (1960) indicates that this snail is found in some districts of Eastern Pakistan and in the northern and eastern parts of West Bengal, but it is rare in the western parts of West Bengal. The snail is also to be found in parts of Orissa and Bihar (Behura 1955). H. C. Ray of the Zoological Survey of India adds to this list the Central Provinces and Berar, although neither he nor Ghose was able to obtain information regarding the present status of the Bombay infestation. Indonesia Probably more has been written about the introduction and spread of the giant snail in Java than about the situation 10 DISPERSAL OF THE GIANT AFRICAN SNAIL in any other place except possibly Hawaii; unfortunately, however, the records are not always in agreement. Kalshoven (1950), for example, reported that the giant snail was first brought into Buitenzorg (Bogor) from Singapore in 1922; but it has also been stated (Anon. 1925) that they arrived in 1925 in a shipment of grass plants from Peradeniya, Ceylon. Leefmans (1933a, b) said that specimens were introduced clandestinely into Sukabumi, a suburb of Batavia (Djakarta), in 1930-31 and were taken into Batavia by plant breeders in 1933. Riel (1933) discovered the snails near Batavia in 1933 and learned from the local inhabitants that they came from Singapore. Van Weel (1948) and Jaski (1953) add confirmatory evidence. Benthem Jutting (1952b), Djaenoedin (1942), Franssen (1936), van der Goot (1939), and Waterschoot (1933) contribute further information on the continued spread of the snail. The most complete early paper on the subject is that of Leefmans and van der Vecht (1933a, b, c). L. J. M. Butot, formerly of the Museum Zoologicum Bogoriense, makes the following statement regarding the giant snail population in Java (in litt. Aug. 1, 1952), "I venture to say that the whole of Java is infested except the mountain region upwards of 1000M and some uncultivated parts below that level." Van der Meer Mohr (1935, 1940, 1941, 1948) and van Weel (1948) give the best accounts of the spread of the snail in Sumatra. This pest apparently first made its appearance in northeastern Sumatra near Tandjong Poera, northwest of Medan, in 1930 or 1931, although it has been speculated that the snail may have been on the island since 1921. It showed up in southeastern Sumatra near Palembang in 1931 or 1932. Shortly after that, it appeared almost simultaneously in a number of localities (Anon. 1941, Heubel 1937, Kalshoven 1950). Latif (1933a) reported achatina from Poelau Bintan, Riouw Archipelago, and explained that the inhabitants had seen the snail in that region since about 1903. Van Weel (1948) justifiably questions the initial date of establishment and sets it conservatively at 1932. The original specimens apparently came from Singapore (Leefmans and van der Vecht 1933a, b, c). Latif (1933c) reported the giant snail for the first time from East Borneo, indicating that it was found in Balikpapan in 1921 and ten years later it was found in Samarinda. It was discovered in Koetai in 1938 (Witkamp 1941) and in Tenggarong about two years after that (van Weel 1948). G. S. Dun has stated (in litt. Aug. 12, 1949) that the giant snail is found in Celebes and in Halmahera of the Moluccas Islands; but this information has not been confirmed. DISPERSAL OF THE GIANT AFRICAN SNAIL 11 J a p a n The giant snail (Japanese: katatsumori) was imported a number of times into Japan proper from Formosa. Esaki and Takahashi (1942) believed that the first shipments started in 1935; but Tokubei Kuroda stated (in litt. Dec. 20, 1949) that they started ''previous to 1933." This would set the date close to the time when the snails made their first appearance in Formosa; for this reason, Boettger's (1951) date of 1925 would seem to be out of line. A number of magazine and newspaper articles popularizing the raising of snails for food and medicinal purposes were instrumental in precipitating a rather large-scale importation of snails into Kobe, Osaka, Nagoya, and Nara. In May, 1936, it was recognized by the Ministry of Agriculture and Forestry that the giant snail was a menace. Regulations were set up immediately to confiscate and destroy all live specimens in the country and prohibit the entry of further shipments. The timely, thorough measures, along with the severe winter climate, were completely effective in preventing the establishment of this snail pest; and to this day, it is believed not to be present anywhere in Japan. Malaya Both South (1923b, 1926b) and Jarrett (1923, 1949) agree that the giant snail first entered Malaya in its northernmost section, in Kedah, in 1911. From Kedah the snail was apparently taken in 1922 to Province Wellesley by Chinese duck farmers who used the snails as duck feed. From these northern infestations, the snail quickly spread south into many areas of the peninsula (Birkinshaw 1928, Corbett 1929, Doscas 1929, South 1922, et seq.). There has been much speculation as to the origin of the initial infestation. It may have been imported directly from India or Ceylon, or it may have come from India via Burma. In a relatively recent colonial publication (Anon. 1954), it is stated that the snail was probably introduced from Ceylon in 1911. Maldive Islands During the Yale Seychelles Expedition in 1957, Alan J. Kohn collected specimens of A. fulica on Hitadu Island and Gan Island in the Addu Atoll; but he encountered no snails on Ile du Coin, Peros Banhos Atoll, in the Chagos Archipelago. The comparatively large amount of commerce between these islands and Ceylon makes it most probable that the giant snail has been introduced from Ceylon not once but several times. Mariana Islands All available evidence points to the conclusion that the giant snail first made its appearance in the Marianas almost simultaneously in Rota, Saipan, and Tinian some time between 1936 and 1938 (Mead and Kondo 1949). Esaki and Takahashi (1942) state that Hiroshi Kuwahata of Saipan 12 DISPERSAL OF THE GIANT AFRICAN SNAIL made an investigation of the snail problem on that island in April, 1941, and concluded that the snails had been introduced the previous month. But the very widespread nature of the infestation at that time makes his conclusion untenable. Other estimates place the date nearer 1940 (Abbott 1949, Bequaert 1950, Lange 1950). Mr. Joaquin Guerrero, who was acting director of the Naval Government Agricultural Experiment Station in Guam in 1949, gave the author permission to quote the following from his report of December 26, 1945 which was addressed to the head of the Commerce and Industry Department: "Mr. Jose Roberto volunteered the statement that during the Japanese administration, he was shown a sample of these snails preserved in alcohol by the Japanese authorities and was told they were deliberately introduced into Saipan and Rota by the Okinawans for use as food from one of the islands of the East Indies." Guerrero's report announced that the presence of the giant snail on Guam first came to the attention of the American authorities just eight days previous to his report and that an investigation brought to light a single area of infestation at Santa Rita. Guerrero states, "In order to determine the possible time when this plant pest made its invasion into Guam, I made an inquiry from Mr. Jose I. Shimizu, a Japanese half-caste and old resident of the island, and was told that the subject snails first made their appearance in Guam in 1943 when the Japanese shipped into Guam sweet potatoes from Rota island." Inquiries made by the author and Kondo did not bear out the often quoted alternative suggestion that the giant snails were brought into Guam in 1946 on pandanus leaves shipped from Saipan for native handicraft (Abbott 1948, 1949). By July, 1946, three separate infestations were found on Guam. At least seven in north and central Guam existed in the fall of 1949 (Kondo 1950a, c). Four years later, infestations appeared in the southern part of the island and it was predicted that it was only a matter of time until the rich farming valleys in that section of Guam would be overrun with snails (Peterson 1957). The giant snail undoubtedly was taken from Tinian or Saipan to the small island of Agiguan (Aguijan), just off the southwest coast of Tinian, shortly before World War II; for it was found to be well established when it was first discovered on this uninhabited island in July, 1949 (Mead and Kondo 1949). The experimental introduction of the predatory snail Gonaxis kibweziensis on this island has caused more to be writen about its giant snail population than that of almost any other island in the West Pacific. Davis, Kondo, Mead, Owen, and Peterson have been the main contributors. DISPERSAL OF THE GIANT AFRICAN SNAIL 13 During the period September 21-October 1, 1949, Kondo (1950a, c) made an intensive survey of Pagan Island and found it to be the most northern of the infested islands in the Mariana chain. He states, "The snail was first introduced in 1939 by Sato Gumi whose w i f e had an infected lung. He raised the snails for medicinal purpose but a man named Sonohara raised many of them for food." Mead and Kondo visited the inhabited islands of Agrihan and Alamagan in 1949 and found no verbal or environmental evidence of the giant snail. M a u r i t i u s Bequaert (1950b) offers convincing evidence in support of his belief that A. fulica did not become established on the island oŁ Mauritius much before 1800. He translates the following from the earliest published record of the infestation here (Bosc 1803): "I have heard from an inhabitant of Mauritius that the wife of a governor of the island, ailing of the chest, had on doctor's orders fetched from Madagascar many of these snails, since there were none in this part of the colony. She died shortly after and the snails spread over the island, increasing to the extent of becoming a calamity. They have been hunted several times, but they are even now very common." This at last sets straight an often quoted story which Benson (1858) erroneously associated with the introduction of this pest on Reunion. Since the original home of this species is in East Africa (Bequaert 1950b), it has been assumed that the snails were taken from East Africa to Madagascar in recent times by natives who intended establishing them as an accustomed source of food (Dollfus 1899). Probably from there they were taken to the Comoro Islands (Morelet 1860). A short time before 1847, a second species, Achatina panthera, was introduced into Mauritius by Sir David Barclay (Benson 1858). Recent reports from Mauritius indicate that this newcomer has taken over the lowlands from sea level to 1,200 feet, forcing A. fulica to maintain its populations at the 1,200-2,000-foot level. New Guinea There is little doubt that the Japanese forces were responsible for the introduction of the giant snail into this island prior to 1945, about the same time that it was introduced in nearby New Britain and New Ireland (Anon. 1948e). Allan (1949), Morrison (1950b) and De Wilde de Ligny (1953) have reported on the New Guinea infestation. North Borneo G. S. Dun of the Lowlands Agricultural Experiment Station in New Britain, reports (in litt. Nov. 25, 1949) that he had learned that the giant snail first appeared in British North Borneo about 1939 and that it was "introduced to rubber plantations outside Beaufort on infested budwood brought from Malaya." 14 DISPERSAL OF THE GIANT AFRICAN SNAIL P a l a u Islands In their search for information about the history of the giant snail populations in the Palau Islands, Esaki and Takahashi (1942) learned from the Palau Office of the Micronesian government that the first introduction was made about May, 1938, by Shoichi Nishimura, a farmer on Babelthaup (Asahi-mura), who planned to breed them for use as food. Because for two years the giant snail had been prohibited in Japan and Formosa, it is presumed that he obtained his specimens from Okinawa. In November of that year a flood washed the snails into the lowlands of the Garumisukan River. They survived the flood and began multiplying. But before this some of the breeding stock sent by Nishimura to several people on Koror made the inevitable escape from inadequate cages, and natural populations were soon developing in the field. Although in 1939 Hatai and Kato (1943) could hardly find any specimens on Koror, by 1940 drastic control measures had to be instituted in a vain attempt to halt the manifestly rapidly expanding population. In addition to Babelthaup and Koror, Kondo (1950a, c) found Auluptagel (Aurapushekaru) and Peleliu infested with the snails. Mead (1950a, c) added Anguar and Urukthapel to this list; and Lange (1950) added Malakal and Arakabesan. Other infested islands unquestionably exist. P h i l i p p i n e Islands There still remains considerable uncertainty regarding when and how the giant snails made their way to the Philippines. G. H. Halden writes (in litt. Nov. 1, 1949) that he observed them in Pampanga and Santo Tomas, Luzon, during the Japanese occupation, and he sent to the author a shell specimen which he had collected at that time. It has been suggested that they were brought from Formosa by Japanese soldiers (Anon. 1946a). Pangga (1949), who gives the best account of the problem in this area, follows this suggestion and sets the date of entry as 1942, although he states that people claim specimens were collected in Pasay, Rizal Province, before the war. By 1949 most of the provinces of Luzon were infested, including the following: Batangas, Bulacan, Camarines Sur, Cavite, Laguna, Nueva Ecija, Nueva Vizcaya, Pampanga, and Rizal. Reunion Bequaert (1950b) concluded that the giant snail reached Reunion early in the nineteenth century, shortly after it became established on Mauritius. Ferussac found it present in 1821. It was Lesson's (1830) feeling that this species had been brought to DISPERSAL OF THE GIANT AFRICAN SNAIL 15 Reunion from Madagascar; but Bequaert stated that it came from Mauritius. Ryukyu Islands Nothing in the printed record has come to light to indicate when the giant snail was first introduced into the Ryukyus. It is most likely, however, that specimens from Formosa were brought to Okinawa some time in 1934 or 1935, that is, shortly a f t e r this pest became established in Formosa. It should be recalled t h a t it reportedly was from Okinawa that specimens were introduced into some of the Mariana and Palau islands several years before the war. The latest information is that the snail has become established in Amami Oshima, which is close to half way between Okinawa and Kyushu, the south island of Japan (Mead 1958a, b). Sarawak According to Jarrett (1931, 1932), the giant snail was introduced into Kuching in 1928 by Chinese farmers who intended using the snails as feed for poultry. Tom Harrisson of the Sarawak Museum writes (in litt. Aug. 25, 1952), "The snails have steadily spread in the last 20 years but still have not reached far up most of the rivers nor into the uplands. Centers of introduction seem to have been in the southeast and southwest, and the distribution gets lower and lower as one goes north." Seychelles It is not known when the giant snails were taken to this group of islands. Dufo recorded them as being present in 1840. Rees (1951) concluded that the snails were introduced from Mauritius. Dupont (1935) listed A. fulica as being on Mahe, Praslin, and Silhouette. Singapore In February, 1922, Jarrett (1923) received several snail specimens from the Balestier District of Singapore for identification. These he identified as A. fulica. From an examination of the infestation, he concluded that they could not have been introduced into the island before 1917 (Jarrett 1949). This places the Singapore infestation six years later than the first infestation in northern Malaya and sets straight some of the confusion in the earlier literature (cf. South 1926b). It is of the highest likelihood that Singapore received its infestation from Malaya, and not once, but many times. Thailand Living specimens from Thailand were obtained in 1938 by Boettger (1951); from this he concludes that at least by 1937, the first snails entered from infested areas in Malaya. This agrees with the estimates of Abbott (1949) and Rees (1951). Ariyant Manjikul of the Central Research Station in Bangkhen writes (in litt. Feb. 14, 1952): "In Thailand this snail is confined to the southern prov- 16 DISPERSAL OF THE GIANT AFRICAN SNAIL inces of the peninsular strip of this country. It is supposed to have been introduced by the Chinese from Malaya. It makes a very good feed and the duck raisers collect them for the purpose." In December, 1957, Mead (1958a, b) found thoroughly intrenched infestations of this snail in central Bangkok. Vietnam Both Boettger (1951) and Rees (1951) agree that the giant snail must have reached Vietnam by 1937. Efforts to obtain more recent information on the status of the infestation have failed. CHAPTER 3 FACTORS FAVORING DISPERSAL A N D SURVIVAL Man--the Principal Agent of Dispersal There are many factors, both intrinsic and extrinsic, which either directly favor or otherwise predispose to a greater spread of the giant snail. Chief among these is the human factor--to the extent that man might justifiably be considered the only effective disseminator. There are many references in the literature to wilful introductions of achatinas into uninfested areas. In fact the history of the spread of Achatina fulica is in its larger portion little more than a series of one intentional introduction after another. Puteh (1939), for example, relates that "several peasants from the inland [Singapore] who had never seen these snails, came to Cherang Ruku and took some specimens for the purpose of breeding them, saying that they looked very beautiful!" The stories about one of the Oahu introductions (Pemberton 1938) and one of the Djakarta introductions (van Brero 1933) read almost the same way. A check of the literature indicates that to a great extent the intentional introductions of A. fulica have been cases of Oriental people introducing the snails for duck feed (e.g., Jarrett 1949). Live specimens of this species were either sent or carried from Siam to Berlin (Boettger 1951); from East Africa to London (Burton 1949); from Rajmahal, India, to London (Godwin-Austin 1908); from Singapore to London (Jarrett 1932); from Mombasa, Kenya, to the Belgian Congo (Anon. 1927a); and from Kenya to London (Rees 17 18 FACTORS FAVORING DISPERSAL AND SURVIVAL 1951).1 In another case, specimens were sent from Singapore to Java in a tin container labeled "flower seeds" (van Leeuwen 1932). This same intent to import the giant snails is reflected in a letter from an Italian correspondent who requested of the author information as to how he might successfully introduce the giant snail into Italy "where food is very dear and scarce." Specimens of an unidentified species of Achatina were sent from Zanzibar to Germany (Semper 1890); and Longstaff (1921) returned to England with three specimens of A. zebra from Cape Colony. Andrews (1948) purchased from a Chicago dealer a specimen of the huge A. achatina which was kept for some time in his home in Baltimore. It escaped once but was found; shortly afterward it laid a number of eggs which fortunately were improperly cared for and none hatched. He suggested that these snails might be developed in Florida as an additional source of food for us. This drew an immediate sharp rebuke from Hanna (1948). Similarly, the taking of this species into California by Dickinson (1946) was exposed and censured by Hanna (loc. cit.) and Mead (1949b). At least until a very few years ago, specimens of A. fulica, A. achatina, and A. ventricosa were to be found alive in the National Zoological Park in Washington, D.C. A similar situation exists in the larger zoological parks in Europe and elsewhere. As for the unintentional introduction of giant snails, Connolly (1912) reports specimens of A. fulica arriving in Durban, Natal, in flower pots sent from Mauritius. Other specimens were found in Java in the soil of a shipment of grass plants from Ceylon (Anon. 1925, Leefmans and van der Vecht 1933a). Latif (1933a) records an instance in Riouw Archipelago in which orchid shipments, apparently bound for Java, were found infested. Many other interceptions of this species are on record. Other species of giant snails have also been taken under similar circumstances. Archachatina marginata was found in Berlin on a bunch of bananas originating in the Cameroons (Boettger 1937, 1938). Paravicini (1926) states that Achatina craveni was being spread in East Africa through shipments of plants. Still other records of the interception of achatinids have been compiled by Boettger (1947). It would be utterly impossible at this point to give any conception of the multiplicity of reports in the literature of introductions, intentional and otherwise, of live mollusks into 1 Dartevelle (1952b) reports that a small shell of A. fulica hamillei has been found in what was then called the Belgian Congo. Since this species is not endemic in this area, he assumes that the specimen was imported as an amulet from East Africa by one of the natives. FACTORS FAVORING DISPERSAL AND SURVIVAL 19 non-endemic areas. This has been going on for years and will continue indefinitely. It thus remains as one of the big forces with which to reckon in setting up control programs. The reason for most of the unintentional introductions of snails, and of A. fulica in particular, may be found in the sheer mechanics of the many diverse and varied activities of man. As one extreme, Campbell (1897) relates in an unintentionally humorous vein how snails even can be "carried about unconsciously by persons on their clothing." The single greatest human factor contributing to the spread of the giant snail is the desire and necessity for members of the human species to be frequently on the move. Coupled with this is the propensity of the snail for crawling back into recesses--recesses which all too often are in the possessions of man. These are carried from place to place and, thus, so are the snails. For this reason, commercial traffic in produce can be particularly threatening and it is therefore the main avenue checked in carrying out quarantines. The author has witnessed in Micronesia the transportation of A. fulica in bunches of bananas, a favorite among native peoples for gifts and trading, and this evidence is offered as the most likely explanation for the spread of this species to several islands in the Truk group (Mead 1950b, c). Shipments of inedible plants or plant products (e.g., grass thatching) provide ideal retreats for the snails (cf. Jarrett 1923). Even soil, especially that containing nursery stock, is not to be overlooked as it may harbor both eggs and snails. It is thus not surprising that Corbett (1933) found the giant snails in the exhibition ground of the Malayan Agri-Horticultural Association. Estivating snails often attach themselves to stored or abandoned war equipment. This helps to explain why war salvage equipment has consistently yielded the greatest share of the giant snails intercepted by quarantine authorities in the United States. Messenger (1952:253-54) has shown illustrations of this type of contamination. This also explains why railroad yards (Corbett 1933), shipyards, and the like have often been found to be the primary site of establishment of new populations. Pereira (1926) has stated that the giant snail was carried into new areas in Ceylon through the agency of motor vehicles. Abortive attempts to effect a control have in some cases succeeded only in spreading the snails still further, for example, attempting to drown the snails in streams (Green 1910c). It is significant that the giant snail population on uninhabited Haha Jima in the Bonin Islands was found to be restricted to a relatively small area, whereas the equally old snail population on inhabited Chichi Jima was found 20 FACTORS FAVORING DISPERSAL AND SURVIVAL scattered all over the northern part of the island from valleys to mountain tops (Mead and Kondo 1949). The pronounced appetite for rotting and decaying materials is the main reason that the snails have frequently been reported congregating in garbage dumps and refuse heaps. And for the same reason, the normal accumulation of material of this sort around the habitations of people living in the tropics has fostered the maintenance, spread, and sometimes restriction of these snails to inhabited areas. Abandoned native crops provided a similar situation, particularly when the exigencies of the war caused many to abandon their land just at the crest of the invasion of the giant snail. In other cases, the crops were abandoned because of the depredations of the snails, thus contributing directly to the expanding snail population. One thing is certain: whenever man enters the picture, he introduces undetermined, altering, ecological factors which make the environment more suitable for snail invasion and maintenance. The effect of these factors seems to carry on for some time after man leaves the area; but as the abandoned land gradually returns to native bush, it strangely becomes less and less acceptable to the snails. Among the more subtle factors operative in man's contributing to the dispersal of the snail is the almost universal strong objection in the non-endemic areas to the very thought of eating snails. But, even more than a matter of innate repulsion, in some areas eating the flesh of such a lowly earth-dwelling animal as a snail is a distinct taboo. The dominance of the Buddhist and Hindu religions in many of the Far Eastern snail-infested areas is directly responsible for the actual "protection" of these pestiferous snails. One good Buddhist acquaintance of the author complained that even in the city where she lived, the snails were so abundant that every night she collected two baskets full, and by the next night, there were as many more in her small garden. Knowing that she certainly had not killed the snails, I asked what she does with the baskets of snails after they are collected. To this, she answered, "Oh, I just take them across the street and dump them on the other side." If her Buddhist neighbor across the street is as fastidious as she, it can be imagined that these two ladies have been trading snails for a long, long time! As one might guess, attempting to set up any sort of control program in these Far Eastern countries is frustrating in the extreme; for even though one may have the co-operation of the well-educated local official, the most orthodox laborers steadfastly refuse to be parties to the crime. Hence, if control measures are set up at all, they more often are half-hearted and ineffectual. It can be seen, then, that al- FACTORS FAVORING DISPERSAL AND SURVIVAL 21 though man is basically the single greatest factor known in the "natural" control of the giant African snail, there are whole areas of the world where this factor has virtually been eliminated. Man, therefore, once again has unwittingly fostered the spread of the giant snails. Longevity The specimen of A. fulica which van Leeuwen (1932) kept in captivity for nine years is clearly the record for this species. It is very possible, however, that his specimen spent a large portion of its life in estivation. Because the body processes are at their lowest ebb during estivation, it is altogether logical to reason that the resultant economy in the physiological mechanisms delays the normal gerontologic changes that bring on senescence and death. Longevity could thereby conceivably be extended by an amount comparable to the capacity to withstand estivation, although excesses in frequency or duration in estivation could produce just the opposite effect. As is shown below, this species will readily retreat into quiescence for protracted periods. It is felt, therefore, that van Weel (1948) was probably correct in assuming that van Leeuwen's specimen represents "an exceptional case." But as a case it does seem to emphasize the fact that estimates of the longevity of this species have probably been on the low side. According to Pelseneer (1935:620), Gibbons reported that A. fulica lives for at least two and a half years. The closely related A. zebra, however, lived for six and a half years in captivity in London (Longstaff 1921). The only other record for an achatinid is that of Flower (1922), who states that a small Limicolaria from Sudan lived in captivity for one year and twenty-one days. In strong contrast are the records of longevity of much smaller pulmonates (e.g., twelve years for Rumina decollata [Vignal 1919], thirteen years for Helix spiriplana [Vignal 1923] and twenty-three years for Oxystyla capax [Baker 1934]). Considering the length of life in other pulmonate gastropod species (cf. Woodward 1880, Pelseneer 1935), it is felt that an average life span of five to six years is a conservative estimate for A. fulica. Where specimens are in areas perennially offering optimum conditions for protracted activity, the normal life span possibly may be shortened by a year or more. R e p r o d u c t i v i t y In determining the reproductive potential of A. fulica, one must know how long it lives, how soon it begins to lay eggs, how often it lays eggs, and how many eggs are laid at a time. One additional factor has been well known from the first: This animal is hermaphroditic and therefore each individual is a pro- 22 FACTORS FAVORING DISPERSAL AND SURVIVAL ducer of eggs; further, any two sexually mature individuals have the capacity for being mutually receptive in cross fertilization. Van der Meer Mohr (1949b) reported that he had repeatedly observed "self-fertilization" in A. fulica in Sumatra. Rees (1951) interpreted the report as a warning with the words, "This means that the introduction of a single snail to a new country is enough to start a colony." In correspondence with the author, van der Meer Mohr later stated that his conclusions had been based on the fact that snails of seven and a half whorls kept in isolation for as long as 382 days still laid viable eggs. To check his conclusions, van der Meer Mohr (in litt. Nov. 5, 1951) kept "virgin" specimens in isolation for two years. None laid eggs. He therefore retracted his original conclusion and reinterpreted his data to mean that a single copulation was sufficient for the fertilization of a number of batches of eggs laid i n t e r m i t t e n t l y over a period of many months. It would now appear that van Leeuwen's specimen kept in captivity for nine years without laying eggs had apparently never copulated. In the absence of autofertilization, Rees's warning still is to be taken almost as seriously as it was given; for a single, fertilized, young specimen is capable of laying a great many viable eggs, even after months of estivation. Information regarding the period necessary for attaining sexual m a t u r i t y was given by G. S. Dun (in litt. Nov. 25, 1949), who reported that snails raised by him from the egg laid their first batch of eggs at the tender age of five to eight months. This concurs almost exactly with the unpublished findings of Daniel B. Langford in Guam. Garnadi (1951) independently reports it as eight months. Earlier, Leefmans (1933c) stated that with an abundant supply of CaCO3 the first eggs were laid in six months; and without it, nine months were required. According to observations in Micronesia, the number of eggs laid at a time by the average, normal appearing adult snail was found to be close to 300. In Ceylon, the number was found to be nearer 200. Mature pygmy individuals in the Isley Field population in Saipan, however, produced only 30-60 eggs at each laying. Kondo has indicated that the maximum number may go over 500 eggs. Atoda's maximum of 733, as reported by Hatai and Kato (1943), quite probably represents total productivity during the period of observation as his largest snail specimen was only 66 mm. long from base to apex. The whole question of total productivity of A. fulica still rests very much in the realm of speculation as the existent figures are based on specimens held in captivity for only a portion of their life or FACTORS FAVORING DISPERSAL AND SURVIVAL 23 u n t i l they happened to die. Field observations, however, indicate strongly that under optimum conditions, this species will lay a batch of eggs every few weeks for apparently an indefinite period of time or until the advent of unfavorable conditions. There probably is a natural tapering-off in productivity with advanced age; but this also has not been determined for certain. Green (1911b) assumed that A. fulica lived two years and that it laid 100 eggs the first year and 200 the second. From these pure assumptions, he estimated that one gravid individual would give rise to 10,930,442,400, or approximately 11 billion snails, in a period of five years. This figure has been quoted many times in the literature. Van Weel (1949), however, points out that the annual output of eggs would actually be nearer 900-1200; but he makes no estimate as to how these higher figures would alter Green's total. Morrison's estimate (1950b) was ten times that of Green, but it is still too conservative. In the light of our present knowledge of the life history of this snail, and figuring very definitely on the conservative side, let it be assumed that the snails mature in nine months after hatching, produce four batches of eggs a year of 150 eggs each, and live for at least five years. At the end of a three year period, the progeny from one gravid individual, if all lived, would total 7,783,764,301 or nearly 8 billion. With numbers building up exponentially in multiple geometric series, another two years would show an increase to 16,121,432,399,695,050 or somewhat over 16 quadrillion! In order to grasp fully the significance of the latter figure, it is almost necessary to interpret it in terms of time and distance. Thus, if we conservatively assume that each individual is 4 inches long and that all the progeny is arranged in a straight line, there would be 15,840 snails per mile for 1,017,767,196,950 miles. This distance would be equivalent to over 2,130,494 round trips to the moon or over 5,477 round trips to the sun. And with light traveling 186,300 miles per second, it would take well over two months for it to travel the distance from the first to the last individual. On the other hand, if one were to count these 16 quadrillion snails, day and night, at the rate of one a second, it would take over half a billion years. This in reverse would take the counter back to the mid-Cambrian, during which period most of our oldest fossils were formed. And as a final comparison: This number of snails would permit each man, woman, and child in the world today to have over 80 million all to himself --or somewhat over 125 acres of snails packed just as closely as possible in a single layer. This prodigious reproductive potential of A. fulica significantly is 24 FACTORS FAVORING DISPERSAL AND SURVIVAL in strong contrast to that of predatory snails. According to Kondo (1951b), the predatory snail Gonaxis kibweziensis gives rise oviparously to only three or four encapsulated, advanced embryos at a time. These are brought forth at an apparent maximum average rate of every ten days during the breeding season for a "reproductive potential of 11 maximum per year (in laboratory)" (Kondo 1956). V a r i a b i l i t y Among the most striking impressions one gets as one examines the various populations of A. fulica is the pronounced difference in appearance of the snails. This point has been emphasized and elaborated upon by both Kondo (1950a, c, 1952) and Mead (1950b, c). Jaski (1953) has been so impressed with the variability in the specimens in Java that he has suggested an explanation might be found either in "recent spontaneous mutation" or "cross-breeding" of the Japanese introduced stock with the stock issuing from the original infestation. In the Santa Rita and Laguna populations on Guam, the shells are commonly close to six inches in length and are exceedingly thick. Not far away, giant specimens have been found that have exceeded seven inches in length. Probably the record specimen was found by G. D. Peterson; it measured just one-eighth of an inch under eight inches! In contrast, the Isley Field population in Saipan, in the midst of an abundant supply of calcium carbonate, was made up largely of ''pygmy" but mature specimens in a size range of two to three inches in length. Shells from Anguar Island in the Palaus were found to have a deep, rich, contrasting color pattern, a thick periostracum, and a high nacreous gloss. Farther north, on Koror Island, a large share of the snails had shells that were dull, chalky, devoid of a periostracum, and so pale that they appeared almost white. One out of every three or four shells in the Army Hill population in Saipan were also pale and devoid of a periostracum; in addition, however, they not only had an arcuate axis, which gave them a "bent-nose" appearance, but had so many lamellate layers of calcareous material deposited around the aperture that the snail could withdraw only partway into its shell. The average specimen encountered in Micronesia was found to be of moderate size (31/2-41/2inches), to be so thin that they could be broken with ease, and to have a body whorl that was unicolorous olivaceous-tan. Specimens from Manila and India have the proportionately larger body whorl as does A. fulica hamillei from East Africa. On the other hand, specimens from some populations in Ceylon and Hawaii, besides being considerably darker in color, have such a long axis and reduced body whorl that they always roll with the aperture upward when they are placed on a flat surface. Nearly all FACTORS FAVORING DISPERSAL AND SURVIVAL 25 populations have a small percentage of anomalous forms; Germain (1921) has made an interesting study of these. An explanation for these and other apparent differences is not difficult to discover. Environmental influence, of course, is being expressed; but the extent has not been determined. In particular, the works of Moore (1936), Ino (1949), and Wagge (1952) bring forth convincing evidence to demonstrate that diet has a pronounced effect upon the appearance, and especially the color, of the snail shell. Because some variable physical characters persist in the same expression in individuals of populations under wholly different environmental conditions, it is assumed that these characters are of genetic origin. The genetic makeup of a given population is predetermined to the largest extent by the genetic complement possessed by the invariably few specimens which started the population. As these first few specimens give rise to subsequent generations, the various genetic combinations make their appearance. The environment, in turn, has its own complement of variable ecological factors. This complement will literally "select'' from the genetically variable snails the combination of genetic characters which is best suited to it. Had the original first few snails been of a different genetic makeup, other genetic combinations would have been available, and the environment might therefore have selected differently. Another environment would have selected still differently. The many different appearing snail population types, then, would seem to be explained by the simple combination of small initial population, isolation, inbreeding, considerable genetic variablity, phenomenal reproductivity, and environmental diversity. Variability thus is of survival value and it helps to explain the success this species has had in becoming established in the many different environments in which it now is found. We do not know the limits of genetic potentiality in producing still different and still more hardy individuals. Estivation and Hibernation In 1930 Duval made a significant finding when he determined that the freezing point was lower in hibernating Helix pomatia than in active individuals of the same species. This work was followed quickly by that of von Brand (1931, 1932) confirming the predictable fact that the H2O content of H. pomatia is actually lower during hibernation than during periods of normal activity. Kamada (1933) corroborated with the report that the blood of hibernating H. pomatia was isotonic with 0.69 per cent NaCl whereas that of active individuals was isotonic with 0.50 per cent NaCl. These findings meant to Howes and Wells (1934a, b) that hibernation as a physiological phenomenon requires "elaborate met- 26 FACTORS FAVORING DISPERSAL AND SURVIVAL abolic preparations," whereas estivation, which is also a resting phase, is a comparatively simple physiological phenomenon. The two phenomena are elaborated upon by Pelseneer (1935). Estivation in A. fulica is of the commonest occurrence. In fact, of the great many populations of this species that have been examined by the author in almost every conceivable type of environment there never has been one in which an estivating individual could not readily be found. A direct correlation between estivation and reduced moisture in the environment has been independently arrived at by a number of investigators. But it was not until the observations of Howes and Wells (1934a, b) that an interpretation was advanced to explain the presence of estivating individuals even under optimum conditions for snail activity. It was their conclusion that "there was a tendency for phases of estivation to alternate with phases of activity during the life of any one individual, even under approximately constant external conditions, so that at any time some of the animals would be in either phase." They concluded further "that estivation is the result of a low water content, and that the latter may be brought about either by dryness of the environment or else by the natural hydration cycle of the animals." The significance of these conclusions is tremendous; for in them is found the explanation for the fact that no combination of even the most rigorous control measures, under apparently ideal conditions, have ever been effective in eradicating A. fulica. In other words, it is the well-secreted, estivating individuals that are responsible for restocking an infested area after the effects of normal or man-made adverse conditions have dissipated. Although estivating snails are most often found under protective conditions that are very superficial, some individuals work their way into remarkably inaccessible niches. They have been found far under sizable rocks, deep in hollow trees, in the center of rotten logs, high in the crotches of trees, far into plant debris, and under many protective layers of man-made litter. The inhabitants of Chichi Jima stated that during the winter months A. fulica specimens are found four to five inches below the ground. In the former Belgian Congo A. rugosa has been found four inches below the surface of the ground in a dry, open, sunny spot (Bequaert 1919). Others have been found "several inches" in the ground (Lang 1919). In Ceylon it was observed that during mildly adverse conditions, such as a brief dry spell during the rainy season, specimens of A. fulica to a great extent tended to remain in a quiescent state on the tree trunks. Usually they were in a range of 3-9 feet from the ground, although specimens were observed to be as high up as 30-35 feet in the dadap trees. At FACTORS FAVORING DISPERSAL AND SURVIVAL 27 these times the ground-dwelling predators remained active and it was therefore a temptation to ascribe a protective function to this behavior of the snails. When environmental conditions were favorable, a large percentage of the snails would make nocturnal visits to the ground but would retreat to the tree trunks with the approach of day. Under prolonged adverse conditions, however, the snails would leave the tree trunks to seek normal estivation sites on or in the ground. At such times many of their predators, significantly, also were inactive. During estivation a thin, fragile, muco-calcareous epiphragm is formed over the aperture of the shell. This epiphragm is normally complete except for a thin grooved slit over the pneumostome permitting the reduced respiration to take place. The slightest disturbance may cause the estivating snail to withdraw farther into its shell. This invariably ruptures the epiphragm. The snail may then repair it, replace it with a new one, or go into a period of activity. The general subject of epiphragm formation has been treated by Hora (1928) and Hora and Rao (1927). The epiphragm of A. immaculata has been described by Smith (1899); and Williams (1951) has described the epipragm of A. albopicta. Of particular interest is his description and illustration of the tubular epiphragm formed by specimens estivating on tree trunks. This apparently has never been observed in A. fulica, although specimens adhering to vertical surfaces have very often been observed to attach themselves at the aperture by a more parchment-like secretion before forming the normal, flat, complete epiphragm. It is definitely not known whether A. fulica can actually undergo hibernation with all its attendant complex physiological adjustments. In Hong Kong, Herklots (1948) stated, "In the cold, dry winter months they hibernate. . . ." It is possible, though, that the quiescent individuals to which he referred were really in a continuing state of estivation which had been initiated by the dryness of the late fall months. Specimens in that area examined by the author in January, 1955, behaved no differently from those known to be in estivation in other areas. The most northerly, well-established populations of this species are in Ani Jima and Chichi Jima of the Bonin Islands. The "shivering cold" weather in those islands, however, is quite comparable to that of Hong Kong inasmuch as weather records for Chichi Jima indicate a low of 45° F. during a recent thirty-year period (Kondo in litt. Jan. 6, 1950; cf. Clayton 1927). But irrespective of whether it is hibernation or estivation, the fact remains that A. fulica in these cooler areas not only survives, but thrives. 28 FACTORS FAVORING DISPERSAL AND SURVIVAL Hardiness Coupled with the phenomena of estivation and hibernation is the physiological faculty of being able to survive prolonged periods without food or water. There is strong evidence that the duration of a period of quiescence is closely correlated with temperature. Under tropical conditions, A. fulica has been reported to have survived periods of estivation, without food or water, for 5 1/2 months (Lange 1950), 7 months (Mead 1950b,c}, 10 months (Corbett and Pagden 1941), and 12 months (G. S. Dun in litt. Nov. 25, 1949). G. S. Butler in Hawaii placed 27 specimens in a jar without food or water in November, 1958; 10 months later, 10 specimens remained alive. Archachatina degneri was still living after having been kept under similar conditions for nearly 6 months (Mead 1950a). In the literature, the record for duration of estivation is 15 years for an unnamed snail species (Bingley 1829); the scientific accuracy of this record, however, is open to question. The classical example, often referred to, is that of Xerarionta veatchii which estivated for 6 years (Stearns 1868). Other records worth noting are Paludestrina ulvae (a freshwater snail), 5 years (Quick 1924); Helix desertorum, 4+ years (Baird 1850); Otala lac tea, 4+ years (Gaskoin 1850); "many Helices," 3+ years (Calkins 1877); Bulimulus pallidior, 2+ years (Stearns 1877); and several common European helicine species, approximately 1 year (Hartley 1898, Lockwood 1880, Sivers 1872, Ward 1879). Kew (1893) gives a general treatment of the subject. Even though the record so far for A. fulica lines it up with the lesser lights in the literature, the faculty for estivating for a whole year without food or water equips this species well for surviving even the slowest, most devious trip to an uninfested land. Dun (loc cit.) stated that after twelve months in estivation his specimens of A. fulica had lost approximately 60 per cent of their original weight. Specimens emerging from estivation were observed both in the Pacific islands and in Ceylon to "drink'' water by rasping at drops. In this manner, other species have been reported to take up 80-100 per cent of their weight in water in a short period of time (Zimmerman 1931). Anatomical examinations of estivating A. fulica have shown that several of the organ systems are drawn upon for sustenance. One of the first to be affected is the reproductive system; this becomes atrophied and proportionately attenuated, resembling superficially the condition in the juvenile specimen (Mead 1950a, b, c). This same phenomenon has been reported in other species (Laviolette 1950). The general subject of resistance to desiccation has been ably elucidated by Gebhardt-Dunkel (1953). Important as a consideration in the present study is the classical work of Shelford FACTORS FAVORING DISPERSAL AND SURVIVAL 29 (1913) which establishes the fact that the rate of evaporation in an environment is the result of the combined action of wind, temperature, isolation, and dryness of the air. It is this rate of evaporation which closely determines, inversely, the rate of activity of terrestrial gastropods in general; and A. fulica is manifestly no exception. In fact, it was observed innumerable times in Ceylon and in the Pacific islands that this species is extremely sensitive to even a slightly elevated evaporative rate. Food, as a vital factor in the environment, is quantitatively and qualitatively infinitely more flexible in the life of A. fulica than in almost any other animal that could be mentioned. As pointed out below, this snail has a tremendous range of acceptable foods; now it is apparent that it can go without all food, including water, for months on end. The environmental factor of temperature operates apparently in a much more limited range than the factors of food and water. A. fulica dies quickly when exposed to the direct sunlight; however, a humid, tropical climate in general seems to offer the nearest to optimum conditions for this species. Such a climate is also conducive to typically luxuriant plant growth which affords ample protection for the giant snail during the daylight hours. As this snail has rapidly extended into many new areas in the past few years, it has "taken hold" in some environments which seem far indeed from what might have been predicted as acceptable. For instance, beach populations were encountered in the Bonin and Mariana Islands, which had such formidable barriers as open stretches of barren, hot sand, and adjacent vegetation covered with fine salt crystals from the constant ocean spray (Kondo 1950c). The fact that this snail was able to crawl over salt-covered vegetation with apparent impunity attests to its remarkable adaptability and hardiness. But as far as the excessively high temperature of the beach areas is concerned, it was the normal nocturnal habits of the snail which permitted it to forage under the suitable conditions of reduced temperatures and concomitantly higher relative humidity. This, incidentally, is a point completely overlooked by those who are prone to preconception in deciding whether or not a given environment will be acceptable to the giant snail. As a parallel, Jarrett (1922) announced on the basis of a small amount of empirical evidence that the European Helix aspersa could not take hold in Malaya. Others had similarly predicted that this species could not survive in the Sonoran southwestern part of the United States; yet it was recently found to be well established in southern Arizona (Mead 1952a, 1953a). 30 FACTORS FAVORING DISPERSAL AND SURVIVAL The capacity of A. fulica to withstand reduced temperatures is an almost completely unknown quantity. The experiments of Chock and Nakao (1951) provide the only information extant on this subject; but even these were very limited in their scope and, therefore, misleading. Their experiments consisted of placing snail and egg specimens in "standard refrigerators of various models" set at 20° F. at r.h. 52-56, 30° F. at r.h. 55-65, 40° F. at r.h. 35-40, and 45° F. at r.h. 40-50. Specimens were set up in the following groups: eggs on the surface of soil, eggs buried in one inch of soil, active adults on the surface of soil, active adults buried with three inches of soil around them in every direction, and dormant adults on the surface of soil. The groups were exposed to a given reduced temperature for a varying number of hours or days. The lethal minimum exposure for 45° F. was reported as 4 days for active snails on soil, 4 days for active snails in soil, and 7 days for dormant snails. The eggs proved to be very susceptible to all reduced temperatures. Some investigators unfortunately have accepted the reported results of these exploratory experiments as definitive and have speculated as to just where in the world A. fulica could become successfully established. Both as a note of caution and as a guide in future experiments in the problem of reduced temperature tolerance, certain points should be emphasized. In the experiments of Chock and Nakao, there are some inconsistencies in results which seem significant. Nineteen out of 20 dormant snails, exposed to a temperature of 40° F. for 24 hours, remained alive after 5 days; 7 out of 10 active snails, exposed to a temperature of 30° F. for 8 hours, remained alive after 5 days; and 1 out of 10 dormant snails, exposed to a temperature of 20° F. for 8 hours, remained alive after 5 days. The inference is that larger samples might have told a somewhat different story, especially regarding survival at the population level. In strong contrast to these survival rates in the preliminary series of experiments, the supplementary series conspicuously showed consistently complete killing throughout the experimental results. No explanation was given for this. Although humidity was controlled to an extent, it was not treated as a constant; and this fact doubtless has influenced the results as another variable. An important consideration which was apparently overlooked in the experiments is the distinction between estivation and hibernation. The "dormant" specimens were obviously in estivation; yet they were subjected suddenly to conditions demanding a hibernation response. Although it should have been the dormant specimens rather than the active specimens that were buried in soil during these experiments, burying them in the manner that they did, with FACTORS FAVORING DISPERSAL AND SURVIVAL 31 three inches of soil all the way around, did not duplicate conditions in the field where heat loss takes place essentially from a single plane. Naturally, there would be in the field an additional insulating effect produced by superincumbent logs, rocks, plant debris, rubble, and the like. The most important factor which has not been taken into consideration in these experiments is that which concerns the acclimatization or conditioning of the animal as it gradually is exposed to lower and lower temperatures. Actually, after a snail has been feeding actively for several days, it extends so far beyond the lip of its shell in the fully contracted condition that it cannot possibly form the protective epiphragm even though sudden adverse conditions may demand it. Normally, a period of less favorable conditions will effect a gradual water loss which, among other things, will permit of epiphragm formation. The entomologists have demonstrated many times that certain insects will adjust to remarkably low temperatures if they are arrived at slowly enough. Cockroaches put suddenly in a cold chamber will die in a short time; and yet if several days are taken to reach gradually the temperature in the cold chamber, these insects will undergo physiological changes, including the alteration of the water content of the cells, and withstand successfully the lower temperatures for prolonged periods of time (Wigglesworth 1953). It therefore brings no surprise that many insect pests are infinitely more abundant after a mild winter than they are after an unseasonably sudden cold snap. It doubtless is this phenomenon of a physiological adjustment that permits the giant snail to maintain vigorous populations in the Bonin Islands despite the relatively cold winters. Apropos of this, it has been suggested that a possible sensitivity to the greater diurnal-nocturnal fluctuations in temperature at altitudes above 5,000 feet have kept A. fulica conspicuously only at the lower altitudes (see p. 197). As an interesting and possibly significant sidelight, Longstaff (1921) kept specimens of Achatina zebra in London for six and a half years, during which time the temperature in the conservatory where they were kept "did not fall below 45° F." He observed that during the colder months, they remained buried for many weeks at a time. The faculty of A. fulica to withstand trauma and even an extensive loss of its shell attests still further to its general physical hardiness. Particularly in the islands of the Pacific, specimens are frequently encountered that have shells almost unbelievably damaged. On inspection, it can be seen that the shells have suffered many breaks. Some- 32 FACTORS FAVORING DISPERSAL AND SURVIVAL times the broken shell pieces are not dissociated and are repaired in place. But, more often, the pieces break off completely or become badly dissociated; under such conditions, repair produces an unsightly shell that gets progressively worse with each subsequent break. Occasionally the mantle becomes damaged and this adds to the distortion through malformation of the new shell. Kondo (1950c) has reported instances where specimens have survived the loss of the lower 2-2 1/2 whorls of their shell; and that is considerably over half the bulk of the shell. As might be expected, there is a direct correlation between the instances of damaged shells and the barren, hard, or rocky nature of the terrain. If there is a cover of humus or a thick undergrowth to break the fall of the foraging snail, there will be little or no damage to the shell even though it is relatively thin. In the regions in Ceylon where the jungle crow was commonly encountered, it was not unusual to find giant African snail specimens which had sustained a considerable loss of flesh as a result of pecks by these malacophagous birds. Most often, it was the posterior portion of the foot, that is, the last part to be withdrawn into the shell, that was amputated and regenerating. Occasionally, however, specimens were found with one or both ommatophores removed, including a portion of the head. Specimens were also found that were regenerating tissue lost in unsuccessful or incomplete attacks by the predatory glowworms. The large, heavy shell and the generous supply of mucus of a mature giant African snail affords considerable protection both from the adverse physical factors in the environment and from natural enemies. Even such avid snail feeders as ducks and geese are unable to handle the snails if the specimens are more than a few months old. As will be seen in the discussions below, this species has demonstrated a remarkable capacity to withstand submersion in both fresh and salt water. The typhoon and flood in Manila (Pangga 1949), the flood in Babelthaup (Esaki and Takahashi 1942), and the tidal wave in Chichi Jima (Mead and Kondo 1949), if anything, seemed to speed up the increase and spread of the giant snail population after the initial, temporary setback. Paucity of Natural Enemies As one reviews the entire topic of the biological control of A. fulica, one is left with the profound conviction that this animal has remarkably few natural enemies indeed. Even the natural enemies it does have are of questionable value in effecting any real control. And this is one of the reasons that some investigators are now in the process of attempting to discover and import foreign predators which promise to become new "nat- FACTORS FAVORING DISPERSAL AND SURVIVAL 33 ural" enemies of the giant snail. A. fulica even seems insensitive to competition from the endemic species of snails in the areas it has invaded (e.g., the large Rhysota in the Truk Islands). On the contrary, it is felt that some of the endemic species, such as Partula gibba, are actually losing in the competition for the gastropod niche in the environment (Kondo 1950c). Period of Activity The facts that A. fulica is normally nocturnal and crepuscular in its habits, and that it will become active in the daytime when it is raining or overcast, indicate that light, as well as temperature, moisture, and food, is a vital factor in its activity. To a great extent, activity during these periods permits the giant snail to escape a number of basically unfavorable environmental factors operative under normal diurnal conditions. A definite homing instinct appears to be present in this species. The experiments of Hatai and Kato (1943) indicated that this may be demonstrated by as much as 70 per cent of the active members of a giant snail population. Such an instinct is of survival value in that it permits the snail to locate readily a protective retreat. A v a i l a b i l i t y of Calcium In a study of the relationships of snails to soil, Van Cleave (1953) concluded that "the presence of the snails is an indicator of available lime s u p p l y . . . . " Usually, the malacologist looks at the correlation from the other direction, namely, that if a good lime supply is present, snails will be found. With respect to the achatinas in particular, Pilsbry and Bequaert (1927) found in the former Belgian Congo a direct correlation between limy soil and abundance of snails. Similarly, Oughton (1948) concluded that the sole nutritional factor limiting the distribution of snails is the availability of calcium. It is not understood why the findings of Hagen (1952) were in direct contrast. Adding calcium to the diet of snails was demonstrated by Oldham (1929a, 1934) to produce in several European snails a shell somewhat larger and heavier by three to four times. With information of this type coupled with the knowledge of the diversified eating habits of A. fulica, it is understandable why the coralline islands of the Pacific have managed to support inordinately large populations of this snail pest. The mineral soils of these islands is reported to have a pH range that is on the alkaline side (e.g., pH 7.2-8.7 in the Arno Atoll [Stone 1951]). CaCO3 in such soils would be stable. This is precisely the explanation for the findings of Atkins and Lebour (1923a, b) wherein snails showed a preference for slightly alkaline soils, being most abundant in the pH range of 7.0-8.0. These facts, however, made it all the more difficult to understand 34 FACTORS FAVORING DISPERSAL AND SURVIVAL how the strongly acid soils of Ceylon could support equally large populations of A. fulica. The soil in the tea-growing areas, for example, is characteristically near a pH of 4.5. CaCO3 could not possibly last long in such acidity, especially in the presence of the usual abundant moisture. The calcium which is present is locked up in silicates and combinations of iron and aluminum. Plants are able to ''unlock" this calcium as manifested by the fact that plant leaf analyses in the acid soil areas of Ceylon revealed as much as 40 per cent calcium in the ash. When the nocturnally feeding A. fulica were seen to feed extensively on fallen leaves in these areas, it was then understood where they were obtaining the lime for their shells. But of even greater importance was the fact that here were snails independent of an "available" supply of lime in the soil. The calcium cycle continues as shells of dead snails are rasped completely away by those that survive. Even the living snails will rasp their own shells, especially along the lip of the shell, and particularly right after a period of estivation. In the Pacific islands, specimens were encountered with the surface of the shell, which had been exposed during estivation, rasped by other snails until it was paper-thin. Snails deprived completely of an external source of calcium are able to draw upon their own shell for their calcium needs for a long period of time (Wagge 1952). Acid soils and a concomitant lack of available CaCO3 in the soil, then, are no real deterrents to A. fulica. Ironically, man in Ceylon unwittingly converts a situation which is tolerable for the giant snail into one that is distinctly favorable. Cacao is being grown more and more in that country because on the world market it is more dependable than rubber. Cacao, however, requires a less acid soil (i.e., in a pH range of 6.0-7.0). Alkalizing the soil is accomplished by adding dolomitic lime. This increases the productivity of the cacao; but it also makes the environment more favorable for the giant snails. Little wonder that the largest populations of this snail in Ceylon were encountered in cacao plantations! Escape Reaction and Migration Cycle Specimens of A. fulica from the Mahinui population in Oahu were marked with white paint and after a period of confinement were released in their home area under essentially the same weather conditions in which they were collected. Observations over the following hours and days revealed an unmistakable "escape reaction" of sustained crawling to the periphery of the experimental area in surprisingly near-equal numbers in all directions of the compass. Further experiments are being conducted. On the basis of what has been observed, however, FACTORS FAVORING DISPERSAL AND SURVIVAL 35 it has been suggested (Mead 1959b) that there exists a "migration cycle" comparable to the "hydration cycle" of Howes and Wells (1934). It is proposed that even under essentially ideal conditions, specimens from time to time will undergo sustained movement out of the home area. Stimuli adequate to invoke this reaction may be less intense in an inverse proportion to the time element since the last "migration." The most subtle stimulus may be sufficient to initiate the reaction in some individuals that have been sedentary for a protracted period of time. On the other hand, a sufficiently strong stimulus might surpass the threshold of a large segment of a snail population in an erstwhile reasonably stable environment and precipitate a "mass migration." Such migrations have been observed many times in the field by a number of investigators, but they have continued to remain essentially unexplained. It is more than likely that intercurrent stimuli (e.g., prevailing wind [Chamberlin 1952a, b]), would have a directive influence on the movement. Proof of the existence of a "migration cycle" would throw much light on the factors of dispersal and population density. CHAPTER 4 ECONOMIC STATUS Current Opinion As attested by the following quotations, the majority of earlier investigators in the field have generalized from their observations that Achatina fulica presents a problem of very real if not actually serious proportions: "They eat nearly everything in the garden" (Jarrett 1923); "this snail is a very destructive visitor [which will devour] anything that is not too hard or distasteful" (Jarrett 1931); "vast amount of damage to agricultural crops resulting from their attack" (Bias and Thamotheram 1939); "a major pest on a number of larger islands'' (Tones 1946). Many other quotations of a similar nature could be given. Most unfortunately, various writers who have had no practical experience in the field aspects of the problem have seized upon these and similar statements, blown them up, embellished them with the products of overactive imaginations in true Sunday supplement style, confused them with information about other agricultural pests, and highlighted them with the sheerest of fantasy until almost terrorizing proportions have been reached. Undoubtedly in many instances these confusions and distortions have been unwitting, but the end results nonetheless have been very damaging to progress in creating a better general understanding of the true proportions of the problem. Green gave probably the first account of this impending trouble w i t h his note (1910a) about an article which appeared in the August 28, 1910, issue of the New York Herald (continental edition). He states that the illustration showing many snails on a coconut stem was superscribed as representing "Snails Destroying Trunk of Coconut Tree." But the situation was to become worse. Statements which 36 ECONOMIC STATUS 37 accused A. fulica of "chewing to shreds much of the wild vegetation of the islands'' and of having ''denuded much of the vegetation in the Hawaiian Islands" were to give way to even greater exaggerations, such as, "countless millions . . . are gobbling down crops on the Hawaiian Islands" and "billions upon billions have already eaten every blade of grass and all the jungle growth on scores of Pacific islands" (Symontowne 1949). An unchallenged climax was finally reached in the magazine section of a newspaper in the western United States (Rhodes 1950). Starting almost immediately in a reinfortsando of "astronomical numbers have devastated parts of Hawaii, India and China." the tempo of exaggeration quickened with an account of a supposed infestation of Ceylon where "the pests appeared in vast numbers one summer morning and by nightfall several thousand acres of farmland were literally crawling with the creatures which ate everything in sight, including the thatched roofs of farmhouses." Soon all credibility was abandoned and a tremendous crescendo built up to the explanation that the "next day not a live snail could be f o u n d on the island" and that "it is guessed the snails, impelled by some strange urge, committed mass suicide in the sea[!]" In the light of all this excessive exaggeration, it is not only of little wonder but definitely predictable that just the reverse reaction has been engendered in many quarters. Skepticism has grown first into disbelief and finally into complete repudiation of the very idea that A. fulica has provided anything more than a field day for reporters and free-lance writers. Such an attitude on the part of some people has provided the single greatest impediment in setting up protective legislation and otherwise meeting adequately the threat presented by this confused problem. To make matters worse, this skepticism has apparently found its justification in the existence of a fair amount of indisputable evidence that the damage caused by this snail is actually considerably more limited than originally believed. Statements in the literature minimizing the problem have seemed to support further the stand of the skeptics. As early as 1910, Green (1910c:56) pointed out and later reiterated (1911b) that in spite of millions of snails in a heavily infested area ''evidence of injury could be found only upon careful search" and that "the casual visitor might pass through the affected area without noticing anything unusual." Hutson (1920), van Weel (1948), and others have supported this stand. Abbott (1951c) has indicated that the achatinas in their homeland of Kenya and Zanzibar cause apparently slight destruction to agricultural produce despite their large populations. Tones (1945) reported that there was "little damage to vegetation in general" in his 38 ECONOMIC STATUS survey of the problem in Micronesia. More recently, J. W. Hes (in litt. Feb. 5, 1952) stated that in Java the economic importance of this snail has been overestimated; Chamberlin's investigation in Tinian (1952a) indicated only "slight damage" to crops on that island and that the rats and insects provided a far worse threat. Olaf Ruhen of New South Wales (vide Morrison 1950a) concluded that there was not even enough evidence against these snails to provide the basis for an article and that "people are just naturally destructive, and want things killed for no other reason than that there are a lot of them." More recently, there has been a tendency even to glamorize the whole subject, thus minimizing still further the problem at hand (e.g., Poling 1954). This swing of opinion from one extreme to the other is kept in a constant state of vacillation through the appearance in the literature of a great number of conflicting reports, inconsistencies, and apparent discrepancies. These would seem to reflect inaccuracy or incompleteness of observations of those making the reports, and in some instances such is probably the case. In this respect, it is worth while to point out that Hutson in 1920 reported that there was no damage to rubber trees through the action of A. fulica. A reversal of opinion started with the report of Weir in 1929 which indicated that the damage had not been sufficient to attract attention. But Beeley, who took over Weir's position, reported six years later positive damage to seedlings and buddings and, shortly thereafter (1938), branded A. fulica a pest of "serious economic importance in Malaya." Examples of this sort have quite naturally caused a questioning of other negative reports. Bias is also undoubtedly an important factor which has insidiously flavored certain other reports. The other factors which have helped create confusion are even more subtle. For instance, in some areas, a certain food plant is not touched at all by the giant snail and is so reported; in other areas, that same species of food plant may be the only one which is at all acceptable and may therefore suffer severe damage. And even in this latter area the damage may be negligible during certain seasons of the year when other food items become temporarily available. During the dry season, damage may be reduced considerably through the fact that a great majority of the snails go into a state of estivation. In some areas advantage is taken of this "snail-free period," especially in the growing of vegetables in small irrigated plots. In contrast, South (1926b) reports greater damage during the dry season because of concentration of snails in the perennially green gardens. The size of the snail population, of ECONOMIC STATUS 39 course, will greatly influence the amount of damage. And the size of the population in turn will be determined by many other factors such as the availability of food, moisture, CaCO3, and temperature extremes. In the early stages of the invasion of a new area, the population quickly reaches its peak and damage is at a maximum; in the later stages the phenomenon of ''decline'' (discussed below) may reduce the population in a very dramatic fashion with a proportionate reduction in damage. In much of the recent literature, there is very apparent a growing indifference to the problem of the giant African snail. Undoubtedly a number of reasons are contributing to this trend: the people have lived with the problem long enough to begin to accept it among their other curses; very limited funds in many cases have permitted the problem to continue without recent rechecking and the lack of official reports has been interpreted elsewhere as evidence of cessation of importance in the area in question (India is a case in point in this respect); entomological and mycological problems are so much more acute in many areas that they demand all or nearly all the attention of a seriously limited number of specialists; and lastly, the phenomenon of "decline" has convinced the authorities in some of the older areas of infestation that the "worst is over" and that the problem will continue to diminish. Damage to Plants At this point, it is more than obvious that there is a genuine need for determining as accurately as possible the true economic status of the giant African snail. The first step in making such a determination is to check the literature exhaustively for reports of investigators who have made field observations of damage in areas invaded by this giant snail. But at the outset one is frustrated by the frequency of extreme contrast in the nature of the reports made by observers of varying authoritative stature. In an attempt to resolve these differences, certain writers in the past have rejected some, minimized other, and championed still other data which gave emphasis to their own interpretations. In the present account, however, an attempt is made to embrace all data and then reconcile the differences through an explanation of their apparent underlying causes. In this way, it is hoped that the existing knowledge will be brought into better perspective than has been the case heretofore. In the following list an honest attempt has been made to prepare an unbiased resume, giving as concisely as possible the essence of the findings reported in the literature or in correspondence. In every category where this author has made field observations, a note concerning them has been placed first in the series of data and followed 40 ECONOMIC STATUS by an asterisk (*) when the observations were made in the infested central and western Pacific islands (Hawaiian, Micronesian, and Bonin Is.) and by a dagger (f) when the observations were made in Ceylon. The data of other investigators are identified by appropriate letters; these in turn are interpreted in the footnote legend.1 In preparing this list there was no attempt to make it completely exhaustive; instead, there was omitted in every case all information which clearly was quoted from earlier original works already covered or which concerned plants of little or no significance or economic importance. Where there was doubt, the information was included in this list. Air plant (Bryophyllum pinnatum): The common, dense stands of this plant in the uncultivated areas of the Hawaiian Islands are providing an almost unlimited source of preferred food.* Albizzia: Where unprotected from snails, only 30 per cent survival of Albizzia falcata in E. Java (AV); bark of A. falcata attacked in Sumatra (He); A. lebbek a preferred food plant in Saipan (L). Amaranthus spp.: A preferred food in Saipan (L); young plants completely destroyed, older ones seriously attacked (W). Amaryllidaceae: Crinum favored in Malaya (Ja); serious damage to many species (W). Ampalaya (Momordica charantia): Seriously attacked in Philippines Arrowroot (Canna edulis): Seriously attacked in Philippines (Pg). Bananas and plantain (Musa paradisiaca): The snails are frequently found on the leaves and trunks; small specimens may work their way deep into the bunches of fruit. Occasionally young bananas are rasped; when they mature the skin appears very badly scarred and is therefore commercially less valuable, though the fruit itself 1 (A)-- Anonymous, 1947a; (AV)-- van Alphen de Veer, 1954; (Be)-- Beeley, 1935; (Be1)-- 19386; (Be2)-193&z; (BJ)-Benthem Jutting, 1934, 1952; (Br)-Bertrand, 1928; (Br^-1941; (C)-- Chamberlin, I952a; (C and G)-- Charmoy and Gebert, 1922; (Ca)--A. Campbell in litt. Nov. 19, 1951; (Co)-Corbett, 1933; (Co^-1937; (Co2)-- 1941; (Ct)-- Cotton, 1940; (D)-- G. S. Dun in litt. Aug. 12, 1949 to Dec. 17, 1953; (Da)-- Dammerman, 1929; (E and T)-- Esaki and Takahashi, 1942; (F)-- Fairweather, 1937; (Fe)-- Feij, 1940; (G)-- Green, 1910c; (G1)-1910&; (G2)-19116; (H)-Hutson, 1920; (H and K)-Hatai and Kato, 1943; (Ha)-- Tom Harrisson in litt. Aug. 25, 1952; (He)-Heubel, 1937, 1938; (Ho)-Holmes, 1954; (Hs) -Hes, 1949, 1950; (Ja)--Jarre tt, 1923; (Jk)-Jaski, 1953; (K1)-Kondo, 1950c; (K2)-1950a; (K3)-- 1952; (L)-- Lange, 1950; (Le)-- Leefmans, 1933a,b; (L and V)-- Leefmans and van der Vecht, 19330, 6; (M)-van der Meer Mohr, 1949a; (M-)-- 1924; (Ma)-Macmillan, 1943; (Mi)-- Milsum, 1950; (N)-R. C. L. Notley in litt. Dec. 7, 1950; (O)-Otanes, 1948; (Pb)-- Philbrick, 1949; (Pe)-Pemberton, 1938; (Pg)-Pangga, 1949; (Ph)-E. Phillis in litt. Jan. 31, 1950; (Pv)-Paravicini, 1922; (R)-Rappard, 1949; (Ri)-Riel, 1933; (Ry)-H. C. Ray in litt. July 22, 1952; (S)-South, 1926&; (S J-1923&; (T)-Townes, 1946; (W)-van Weel, 1948-49; (We)-Weber, 1954. ECONOMIC STATUS 41 is unaltered. If the bananas split open on the tree, they will almost invariably be consumed by the achatinas frequenting the treetops. Several native reports indicated that young plantings may be killed by the attacks of the snails. Total damage is small.* Inhabitants of Tinian report no damage (C); young leaves eaten in Saipan (E and T); leaves and steins of "pisang" attacked only slightly in Sumatra (He); unverifiable reports in Ceylon that leaves and apparently blossoms of plantain eaten, therefore fruitless (G); preferred by larger snails in Palau Islands (H and K); a chief food plant in Koror, ripening fruit liable to damage (K2); fruit, leaves, and new shoots are damaged in Saipan (L); young leaves eaten (Le); attacked in Philippines, peelings eaten (Pg); attacked in Java (Ri); could not verify damage (W). Beach morning-glory (Ipomoea pes-caprae): This very common vine serves not only as one of the main shelter plants for achatina but as a good source of food;* a preferred food plant in Saipan (L). Beans: Fed upon in Saipan (L); eaten voraciously in Philippine Islands (Pg); Phaseolus radiatus seedlings totally destroyed, foliage of older plants skeletonized, but the yam bean (Pachyrhizus tuberosus) suffered no damage (W). B e t e l pepper (Piper betle): Untouched (G); no damage (W). Bird's-nest-fern (Asplenium nidus): A preferred food plant in Saipan (L). Blimbing (Averrhoa bilimbi): Fruit completely stripped in Ceylon Bougainvillea: Both large and small snails attacked in Riouw Archipelago (L and V). Breadfruit (Artocarpus spp.): The fruit of this tree, whether it be green, overripe, rotting, or dried, is a real favorite of the giant snails. Although the snails have been seen up to about twelve feet in these trees, the fruit which had not fallen was never observed being attacked or showing any signs of attack. Since the native people of many Pacific islands have abandoned the breadfruit for the more easily prepared imported staples, the breadfruit crop in some areas is being allowed to contribute 100 per cent to the maintenance of a larger and still larger snail population which otherwise could not exist. Robert E. Burton of the Agricultural Development Station on Ponape reported that all of a number of sprouted breadfruit clippings brought from the Truk Islands in 1948 were killed by having their bark removed by achatinas.* Inh a b i t a n t s of Tinian report no damage (C); a chief food plant in Koror (K2); a preferred food plant in Saipan (L). (G)-. 42 ECONOMIC STATUS Cabbage (Brassica spp.): This and other Cruciferae are the most preferred food plants in the Philippines (Pg); severe damage in Saipan (L). Cacao (Theobroma cacao): In Ceylon the giant snails were frequently seen congregating in great numbers in the areas where the cacao pods were being harvested; they were observed feeding on the flowers, broken pods, seeds, and seedling plants in nurseries; damage varied from slight to severe;H fed on young shoots and flowers in Indonesia (BJ); young seedlings up to two months of age readily attacked and killed in New Britain (D); no damage (H); direct damage by consuming freshly planted seeds, indirect damage by destruction of cuttings of dadap cover crop in Ceylon (N); killed young plants and injured blossoms of bearing trees, impossible to plant new clearings in Ceylon (S); seriously attacked and damaged (W). Cactus: Opuntia sp. eaten (BJ); Opuntia and Cereus suffered damage in Java (L and V); damaged (Pg); attacked in Java (Ri). Calophyllum inophyllum: A definite preference for the leaves of this plant. * Canna sp.: Eaten voraciously in the Philippines (Pg). Carambola (Averrhoa carambola): Fruit reported damaged in Ponape.* Carrots (Daucus carota): Severe damage to both the tops and the tuberous roots was reported by the inhabitants of Chichi Jima;* fed upon in Saipan (L). Cassava (Manihot esculenta): Damage is largely restricted to the young plants, reported to be a serious problem in Romalum, Truk Islands;* Tinian inhabitants report no damage (C); badly damaged, young stems heavily peeled off and killed in Java (Fe); leaves and bark damaged in Sumatra (He); a chief food plant in Koror (K2); attacked in Philippines (Pg); no damage (W). Chili peppers (Capsicum spp.): Untouched (G); the fruit, leaves, and bark of C. annuum are attacked in Sumatra (He); stem and leaves of C. grossum eaten in Rota but damage not serious (K3); fed upon in Saipan (L); no damage (W). Citrus sp.: Near-ripe fallen fruit and foliage of sweet orange (C. sinensis) were observed to be eaten on Agiguan;* seedlings seriously attacked in Philippines (Pg). Coconut (Cocos nucifera): No damage observed;*H swarming on fronds but no evidence of injury (G); no damage (H, S). Coffee (Coffea spp.): Slight damage to berries in Malaya (Co); in Su- ECONOMIC STATUS 43 matra, young leaves attacked only when other food is not present (He); observed to attack in Philippines (Pg). Corn (Zea mays): Only slightest damage to leaves and kernels in Chichi Jima, decaying leaves eaten readily;* inhabitants of Tinian report near destruction of field of very young seedlings, but no appreciable damage to larger plants (C); only slight damage to leaves in Sumatra (He); young seedlings readily eaten in Guam (K2); fed upon in Saipan (L); seedlings sometimes attacked but no severe damage (W). Cosmos sp.: Damaged in Philippines (Pg). Cotton (Gossypium sp.): In Mauritius "it was responsible for a good deal of damage to cotton seedlings, when the attempt was made to plant cotton, on a large scale, in 1911" (C and G). Cowpea (Vigna sinensis): Cover crop in Ceylon almost completely destroyed (Br); the fruit, leaves, and stems are damaged in Sumatra (He); young plants completely destroyed, older plants skeletonized Cucumber (Cucumis sativus): Very little damage in Koror;* inhabitants of Tinian report serious damage often to entire plant (C); one of favorite foods in Sarawak (Ha); fed upon in Saipan (L); seriously attacked in Philippines (Pg). Cucurbitaceae: Especially liable to damage (D). Eggplant (Solanum melongena): Plants stripped of bark in Ceylon (G); fed upon in Saipan (L); attacked in Philippines (Pg). Elephant ear (Alocasia sp.): Foliage only sparingly eaten in Ceylon (G); "ape" loaded with snails and eaten (Ki). Euphorbiaceae: Eaten (BJ); Euphorbia trigona eaten in Java (L and V); inflorescences eaten, but no serious damage (W) (vide Rubber). Ferns: Fronds of tree ferns (Alsophila lunulata) eaten in Palaus;* young snails especially destructive to garden ferns (Gi). Flower gardens (general): A wide variety of garden flowers are readily eaten, some of them being quite seriously damaged;*H! suffering extensively in New Britain if they are not carefully protected (D); the damage killed young Salvia plants in Sumatra (Fe); "whole gardens devastated" (Jk); do not seem to attack Salvia, Torenia, or Coleus in Ceylon (Ma); serious pest in Seychelles (Mi); decided preference for ornamentals in Philippines (Pg); all agree that greatest damage is to gardens (S); much damage in Malaya (Si); a real pest in horticulture (W). Gourd (Lagenaria leucantha): "Upo" seriously attacked in Philippines (Pg). Grasses: All seem to be nearly completely immune to attack.*H (w) 44 ECONOMIC STATUS Gynandropsis speciosa: Complete planting failed in Malaya in spite of controls (S). Hibiscus spp.: A favorite plant, damage to both flowers and leaves noted; * hedges of this plant were invariably loaded with giant snails; all parts of the plant were subject to attack;H fallen flowers preferred in Palaus (E and T, H and K); damaged in Java (Fe); attacked in Philippines (Pg) (vide Okra). Jak (Artocarpus heterophyllus): The fallen fruit is a great favorite; whole areas of bark, even on mature trees (Fig. 5), may be consumed; young seedlings very susceptible to damage;*H "chew up whole seedlings up to three to four feet in height overnight" (Ho). Leguminosae (general): A number of unidentified species consumed in the field attest to the wide preference for this group; * damage to the cover crop plant Pueraria thunbergiana was essentially complete on some estates in Ceylon causing indirect damage to Cacao; the more hardy Desmodium triflorum was only slightly attacked; bark of Gliricidia maculata and dadap (Erythrina lithosperma) subject to extensive attack, even in trees in excess of three inches in diameter; flowers of the dadap are a favorite;H large areas in Pueraria cover crop in New Britain barren and not reseeding (D); Centrosema pubescens damaged in Java (Fe); young dadap trees killed by bark removal in Ceylon (H); in Sumatra, the young leaves of C. pubescens and the leaves and bark of Deguelia sp. are damaged, and although under experimental conditions Cassia multijuga was not touched, the following were readily eaten: Cassia mimosoides, Crotalaria anagyroides, C. striata, Indigofera suffruticosa, Mimosa invisa, Parkia sp., Tephrosia Candida and T. vogelii (He); indirect damage to tobacco as young plants of cover and green manure crop (Mimosa invisa) were stripped of their bark in Sumatra (M); particularly destructive to dadap cuttings in cover crop of cacao (N); could not grow leguminous ground cover crops in Ceylon when achatina was at its height (Ph); young dadaps girdled and killed (S). Lettuce (Lactuca sativa): Native people of Tinian indicate no damage (C); severe damage in Saipan (L); plants of all ages completely defoliated, but only seedlings of L. indica destroyed (W); food of greatest choice in the laboratory (We). Leucaena glauca: This exceedingly common introduced species in its characteristic dense stands is clearly a preferred food plant from the Hawaiian Islands (where it is known as "Haole koa") to Chichi Jima; all parts are avidly consumed, including the water-soaked seeds and the exposed roots; the slender stems are frequently seen ECONOMIC STATUS 45 bent far over under the weight of snails feeding on the leaflets and tender bark;* occasionally fed upon in Tinian (C); not touched in Sumatra (He); defoliation of young "lamtoro" trees placed in a teakwood stand in Java (R). Liliaceae: Nearly all lilies seem susceptible to attack; *H Crinum and other lilies preferred (Da); favored plants (Ja); Caladium not attacked (Ma). Malungay (Moringa oleifera): Observed being attacked in Philippines (Pg). Melons (Cucumis melo; Citrullus vulgaris): The fruit and vines of all types are on the preferred list; on Rota, they cause such destruction of watermelon vines that the native people walk several miles to plant their seeds in an achatina-free area; attacks on the young fruit will badly disfigure it, interfere with proper development, and make it worthless for consumption;* inhabitants of Tinian report serious damage often to entire plant (C); severe damage in Saipan (L); rinds consumed (Pg). Montanoa hibiscifolia: The bark and pithy stems deeply and extensively grooved in Hawaii.* Morinda (Morinda citrifolia): Both the leaves and the mushy, white f r u i t were among the most preferred items in Koror;* preferred in Saipan (L). Oil palm (Elaeis guineensis): Prefer overripe or underripe fruit in Malaya (Co1, Co2); attack mainly fallen fruit but also ripening f r u i t on branches (F); will attack leaves only if very hungry (He). Okra (Hibiscus esculentus): Plant practically defoliated and fruit injured to such an extent as to make it valueless in Ceylon (G); severe damage in Saipan (L). Onion (Allium spp.): No damage (W). Orchids: Phalaenopsis and Vanda damaged in Riouw Archipelago and Java (L and V); Phalaenopsis spp. damaged in Philippines (O); attacked in Java (Ri). Pandanus spp.: No damage observed;* occasionally fed upon (L). Papaya (Carica papaya): A great many times the snails were seen high up in the trees; both fruit and bark are damaged; in Ponape missionaries complained bitterly of attacks on the fruit; *H a fallen tree was completely consumed in a week, a near-prostrate tree suffered damage to buds and flowers, snails seldom seen in top of erect trees in Tinian (C); older trees in New Britain are deprived of their bark until they wilt and die, seedlings are eaten away completely (D); ripe fruit damaged in Saipan (E and T); leaves eaten, young stems ringed and killed (Fe); flowers consumed 46 ECONOMIC STATUS in Ceylon (G); fruit, leaves, and bark of young trees damaged in Sumatra (He); preferred by larger snails in Palaus (H and K); chief food plant in Koror, fruit liable to damage (K2); preferred food plant in Saipan (L); eaten in Java (L and V); fruit eaten in Philippines (Pg); climb trees and eat ripe fruit (T). Passion flower (Passiflora spp.): One of the commonest plants of choice in the Pacific; fruit, flowers, and leaves eaten; in many areas it is the main if not the sole food item;* greatest preference in Tinian (C); the leaves and ripe fruit are preferred items in Saipan (L); most of the snails' food is the old yellowing leaves and ripe f r u i t (T). Peanut (Arachis hypogaea): A planter in New Guinea abandoned peanut cultivation largely because the snails made continual appreciable inroads in his fields (D); attacked the leaves, stems, and nuts in Sumatra (He); young plants completely destroyed, older ones seriously defoliated (W). Pepper vine (Piper nigrum): Young pepper vines planted at the base of Gliricidia trees in an unthrifty cacao plantation in Ceylon suffered 100 per cent kill because the palm frond shields, set up to protect the vines from the sun, provided shelter for great quantities of hungry achatinas;f plant killed by eating outer layers of stem; Tinian inhabitants report vines often killed by snails (C); leaves attacked in Sumatra (He). Pineapple (Ananas comosus): No damage observed or reported in the East Caroline Islands, Hawaiian Islands, or Ceylon; *f no record of feeding on pineapple (Pe). Pipturus albidus: The fallen leaves of the common "mamaki" of the Bonin Islands were observed to be a definite favorite.* Pumpkin (Cucurbita pepo): In Chichi Jima, the pumpkins had to be raised in protected boxes and then transplanted to prevent them from being chewed off as fast as they came up;* inhabitants of Tinian reported serious damage often to entire plant (C); species introduced by Japanese in New Britain entirely eliminated by the snails in some areas (D); leaves skeletonized and stems barked in Ceylon (G); fed upon in Saipan (L). Radish (Raphanus sativus): All parts of the plant attacked in Sumatra (He); fed upon in Saipan (L). Ramie (Boehmeria nivea): Attacked in Philippines (Pg). Rice (Oryza sativa): No damage (G, H, He, W); no harm except breaking down plants with their weight (S). Roses (Rosa spp.): Do not seem to attack in Ceylon;H attacked in Philippines (Pg). FIG. 1.--The largest living Achatina fulica Bowdich ever photographed. This specimen, collected in Guam by George D. Peterson, Jr., has an over-all length of nearly one foot and a shell length just short of eight inches. (Photo courtesy of G. D. Peterson, Jr.) FIG. 2.-The Army Hill population of the giant African snail in Saipan in 1949 FIG. S.-Achatina fulica (ca. 85 mm.) with three of several predatory snails that have been considered in its biological control; left to right: Euglandina rosea, Gona quadrilateralis, and Natalina sp. (Photo courtesy of Alan Thistle.) ECONOMIC STATUS 47 Rubber (Hevea brasiliensis): Occasional specimen seen feeding on latex at tapping site; disfiguring or killing damage to young plants in nurseries considerable in some areas ;H feed on young seedlings and buddings, continued destruction of buds produces distinctive type of fasciation (Be); drink latex and eat sweet cambium layer of bark exposed by tapping thus causing wounds (Be2); serious economic importance in Malaya (Be1); feed on young shoots and flowers (BJ); damage to young rubber rather serious in Ceylon (Br1, Ct); G. A. C. Herklots indicates no very considerable losses in rubber industry (Ca); young trees severely damaged in Sumatra; bark removed, stems ringed, leaves consumed, growing tips and replacement growth damaged to such an extent that death resulted in some cases; older trees not seriously affected although the snails' consumption of latex (proved by crushing snails and observing the latex in the gut) contaminated the latex, sidetracked its flow, and caused the collecting cups to fall to the ground (Fe); no damage (H); young leaves and bark damaged only when more acceptable food is not present (He); bark stripped off, killing young plants; achatinas seen by the hundreds in Ceylon sucking sap in latex cups (N); "learned to drink rubber latex" (Pb); strip succulent bark from young plants, drink latex (Ph); drink a considerable amount of latex (Pv); impossible to plant new clearings in Ceylon in 1916 (S); seriously attacked (W). Scaevola frutescens: Both leaves and flowers are preferred items.* Soursop (Annona muricata): No apparent damage in Dublon, Truk Islands;* a preferred food plant in Saipan (L). Squash (Cucurbita spp.): Attacked in Philippines (Pg). Staple crops (general): Left unharmed in Malaya (Be2); no appreciable damage in Malaya (Si); damage only a fraction if at all (W). Sugar cane (Saccharum officinarum): Often found congregating in great numbers on the leaves but damage was practically nil;* damage to leaves on Saipan (E and T); damage in Java, especially to the leaf axils where there is much less concentration of silica cells, but of no real economic importance (Hs); a chief food plant in Koror (K2); no record of damage (Pe). Sweet potato (Ipomoea batatas): The vines provide cover often for phenomenal numbers of snails but damage to leaves and exposed tubers was invariably inappreciable;* not eaten in Tinian according to inhabitants (C); many snails but no damage was noticeable in Ceylon (G); a chief food plant in Koror (K2); attacked in Philippines (Pg); foliage sometimes attacked but no serious damage (W); foliage a food of choice in the laboratory (We). 48 ECONOMIC STATUS Taro (Colocasia esculenta): Foliage is not infrequently attacked, though seldom skeletonized; rarely, damage is caused to exposed roots; total damage is surprisingly almost inconsequential even in areas overrun with snails;* a lot of damage in Kabunga, New Guinea (A); feed only on fallen leaves, but according to the inhabitants of Tinian, the rarely exposed tubers are eaten (C); only slightly affected in New Britain as it is particularly hardy and fast growing (D); foliage eaten very sparingly in Ceylon (G); fallen leaves eaten in Palaus (H and K); not affected in Guam (Ki); a chief food plant in Koror (K2); only occasionally fed upon in Saipan (L). Tea (Camellia sinensis): Every report in Ceylon was negative;H young shoots and flowers are eaten (BJ); no damage in Ceylon (H); young leaves attacked only when there is little else to eat (He). Teakwood (Tectona grandis): Year-old plants damaged in S. Sumatra (AV); up to 90 per cent lethal damage to young plants in Java (R). Thespesia populnea: A preferred food plant in Saipan (L). Tobacco (Nicotiana sp.): Bark removed from base of plant, stalk weakened so that it blows over easily; leaves of young plants eaten; total damage negligible; will never become a serious pest (M). Tomato (Lycopersicon esculentum): Foliage seriously damaged in Guam immediately following the first big rain of the season;* not attacked in Ceylon (Ma); decaying fruit consumed (Pg); indifferent to plants (W). Tree nettle (Laportea crenulata): Stems and branches completely denuded, killing the trees; practically exterminated in some localities (G). Vegetables (general): Most varieties were found to be attacked at least to some extent; in some cases serious damage was caused, especially in small plots adjacent to abandoned or uncultivated areas; *t suffering extensively in New Britain if they are not carefully protected (D); certain plants defoliated and others denuded in small vegetable patches (G2); damage to fresh vegetables caused mainly by the smaller specimens (H and K); a pest (H); farmer in Sinajana-Ordot area of Guam reported that it was not possible to raise vegetables as they were too vulnerable to attack by snails (Ki); serious pest in Seychelles (Mi); decided preference for succulent vegetables in Philippines (Pg); a serious pest in Ceylon (Pv); "great depredations" in Calcutta (Ry); in some localities in Malaya the growing of certain vegetables has become almost impracticable (S); particularly destructive (T); seriously attacked (W). ECONOMIC STATUS 49 Vegetable sponge (Luffa spp.): "Patola" seriously attacked in Philippines (Pg). Weeds and uncultivated plants: Many species are attacked; some are definitely preferred and others are only occasionally eaten; *f Lange (1950) has prepared an impressive list for Saipan and as far as known, it is the only such list extant. Xanthosoma braziliense: Leaves and stems of "Tahitian spinach" eaten in Hawaii (Ki). Yam (Dioscorea alata): The leaves are an obvious favorite; eaten extensively in Ceylon;H no appreciable damage in Tinian (C); young plants eaten down to the ground in Ceylon (G); impossible to plant yams in Rota when snails were abundant (K3); severe damage in Saipan (L); "ubi" seriously attacked in Philippines (Pg). Zinnia linearis: Attacked in Philippines (Pg). Evaluation of Damage In examining critically the above list, one is left with the very definite conviction that considerably more in the way of careful observation is required before there can be made an accurate evaluation of A. fulica as an agricultural and horticultural pest. A review of the reports of damage caused by this snail discloses the fact that the greater share of them are qualitative only. That is, we know what the snail will attack, but we do not know to what extent the snail will attack it. Studies of quantitative damage therefore must be made. Then, when it is known, for example, that the maximum damage to a certain type of plant is negligible, the presence of A. fulica in stands of that plant will not cause the needless initiation of elaborate and costly control measures. In spite of these difficulties, certain generalizations and a number of rather specific conclusions can be made at this time. These at least will assist in arriving at a reasonably satisfactory tentative evaluation. Garden flowers and ornamentals of a great many varieties are among the most susceptible to the attacks of A. fulica; often all parts of the plant in all stages of development are eaten. Most vegetables, but especially those belonging to the Cruciferae, Cucurbitaceae, and the Leguminosae, often suffer severe damage. Cuttings and seedlings, even of plants not attacked in the mature state (e.g., breadfruit, cassava, teakwood, etc.), are also preferred food items of this snail; damage to these is caused by their complete consumption or by the removal of bark. According to Jaski (1953), the young snails, up to about four months, feed almost exclusively on young shoots and succulent leaves. Cover crops, and especially those that are leguminous, represent the f o u r t h and last major category of plants which may be severely dam- 50 ECONOMIC STATUS aged. In most cases the effects of the damage rest not so much in the physical destruction of the cover crop plants as in the secondary damage to the main crop for which the cover crop is used for green manure, shade, soil retention, moisture retention, and/or nitrogen restoration. A case in point is the almost complete destruction of the leguminous cover crop Pueraria thunbergiana in the Dodangoda Post of the Godahene estate in Ceylon. The cacao plants suffered from the reduction of available nitrogen in the soil; and the soil, in turn, suffered marked erosion particularly in the steeper denuded areas. The more woody, hearty legume, Desmodium triflorum was substituted because it was more resistant to the attacks of the giant snail; but this cover plant grows more slowly, produces a less dense cover, adds proportionately less nitrogen, and succumbs more readily to one of the plant diseases. A. fulica, however, comes rightfully by this type of destruction as Spence (1938) has reported related achatinids (Limicolaria zebra and L. numidica) in their native home of British Cameroons destroying leguminous cover crops of Calopogonium, Centrosema, and Pueraria, which were interplanted with oil palms. Not only did the cover crops suffer from the attacks, but the palm fruit itself was eaten. Similarly, there has been reported damage to sesame and coffee by Achatina craveni Smith (Salaam 1938); damage to sisal and cotton by A. zanzibarica (Tomaszewski 1949); damage to papaya and other plants by A. albopicta (Williams 1951); and damage in nurseries by Limicolaria distincta and L. lucalana (Dartevelle 1954). Although many plants of little or no economic importance are attacked by A. fulica, it is significant that the majority of plants which suffer the greatest damage are conspicuously of the cultivated type. Taylor (1894) quotes from Gain data which similarly show that cultivated plants form nearly two-thirds of those "acceptable" to the snail Cepaea hortensis. Several reports of damage listed above are considerably more significant than they might appear to the casual reader. The observed damage to citrus fruit and seedlings sounds an ominous warning of what might happen should these giant snails become settled in the great citrus growing areas of the world. It has already been established that the climatic conditions particularly suitable for citrus are very nearly equally suitable for A. fulica. And, for the skeptics, stranger things than this have happened in the field of malacology. For example, although it was well known that Theba pisana had a strong propensity for citrus fruit (both leaves and bark) in its native home (de Stefani 1913) and in California (Basinger 1923a, c, 1927; ECONOMIC STATUS 51 Gorton 1919; Smith 1919a, b, 1922), most malacologists would have been somewhat hesitant to announce that its near-relative, Helix aspersa, if given the opportunity, could become an almost equally serious pest. But within recent years, just that has happened in Southern California (Anon. 1929; Basinger 1931, 1940; Gammon 1943; Lewis and LaFollette 1941, 1942a, b; Persing 1943-45; Woglum 1943). The over-all damage that results from the feeding of A. fulica on rubber trees, fortunately, is not critical in any real sense of the word. Under certain conditions, however, there can be significant and even serious local damage to the young saplings, especially in nurseries. R. C. L. Notley, chairman of the Planters' Association of Ceylon, and E. Phillis, director of the Rubber Research Scheme of Ceylon, have reported in correspondence numerous incidents of serious damage of this sort. In the literature, the works of Beeley and Feij are particularly convincing. The snails' strange but not unique appetite for the fresh rubber latex makes them a nuisance and sometimes a pest on the larger trees when they are being tapped. Green (1909, 1910a), Keuchenius (1914), and Meer Mohr (1924) have shown that the endemic sluglike snails Mariaella dussumieri and Parmarion reticulatus can be, in the same respects, equally serious pests on rubber. Paravicini (1922) adds the well-known plant pest Bradybaena similaris to this list. All parts of the papaya tree seem inordinately susceptible to the attacks of the giant African snail; and as its fruit is a universal favorite wherever it is available for human consumption, losses would be particularly felt in the areas where the snail is abundant. Reports of damage to peanut and pepper vines seem to be more than a little significant as these plants figure very prominently in world commerce. In both cases the extent of damage needs to be determined accurately. In the meantime, it will be hoped that the lack of more reports of this sort is indicative of only slight general damage. Consistent reports of very little or no damage to certain other crops are, in many respects, even more significant than reports of appreciable damage. For example, all members of the grasses, such as sugar cane, corn, rice, and the "grains," are very conspicuously immune to all but occasional, relatively insignificant damage. And this sems to hold pretty well in all regions of the world, although Herrstrom (1953) reports slug damage to stored winter wheat. Considering the tremendous importance of these products in the economy of the world, it is a blessing indeed that there is such universal immu- 52 ECONOMIC STATUS nity. Other staple starchy crops such as the taros and sweet potatoes are almost as free from serious damage, although the cassava suffers some damage, especially in the young stage; and in some areas the yam is seriously attacked. Newspaper accounts and other reports (e.g., Zuk 1949) to the contrary notwithstanding, the giant African snail cannot be considered in any possible way a threat to the coconut industry. The same can be said for other economically important "woody" plants like the pineapple and the pandanus. The betel pepper, onions, most varieties of chili peppers, and many other highly aromatic or irritating plants are understandably not among the preferred items of A. fulica; but one cannot completely generalize from this, as the tree nettle, cactus, and tobacco are appreciably damaged in some instances. The literature is conspicuously almost completely devoid of anything but the most incidental report of damage to coffee and tea plants. From this we can infer that the giant snail can be written off as a possible pest of these plants since otherwise its abundance in the tea- and coffee-growing areas of the East would surely have caused many reports of damage to have been made by this time. Other species of snails, however, have been listed as principal pests of coffee in Brazil (Fonseca and Autuori 1932). The other great beverage plant, the cacao, unfortunately suffers a fair amount of direct and indirect damage. Of fruits themselves, only the papaya seems to be damaged to a serious extent; the others either have not been reported as damaged or have been reported as damaged only slightly. Perhaps an explanation for this can be found in the fact that there is a definite preference for rotting and therefore fallen fruit. At any rate, it seems safe at this point to assume that another whole category of plant products can be considered very nearly free from significant damage. It should be borne in mind, however, that many of these apparently "immune" plants can and do sometimes suffer complete destruction in the unprotected young stages. Indirect Damage So far, the only example that has been given of indirect damage to plants concerns the destruction of cover crops. There are others which must be taken into consideration if a fair evaluation of the problem is to be made. Quite probably, many of these have not found their way into the literature because they have not been recognized as damage resulting from A. fulica. For example, rice and other plants which are not eaten by the giant snail nevertheless may be badly damaged under conditions where the sheer weight of the snails causes the plants to break. Another example, with an ironic twist, is reported by Dr. E. Phillis who states (in ECONOMIC STATUS 53 litt. Jan. 31, 1950) that in Ceylon some land on tea estates has been rendered unfit for growing tea because the snails were collected and buried in such quantities that the acid soil, demanded by the tea plant, became too strongly alkalized. An equally provocative situation is found in Sarawak where Tom Harrisson states (in litt. Aug. 25, 1952) that the Chinese market gardeners and others are required, in order to minimize damage by A. fulica, to keep their plots so clean all the year around that excessive soil erosion has occurred. Because the giant snail competes with other animals for food, and because it, in turn, provides an increasing source of food for certain carnivorous animals, including paradoxically some with which it is competing, the entering of A. fulica into an environment invariably produces a series of ecological chain reactions which in some cases result, via the most devious channels, in damage of a completely unpredictable nature. The snail's potentiality in this respect should never be underestimated. This whole topic of environmental interrelationships is discussed under several headings below and summarized in the figure on page 137. Under the heading of indirect damage must also be considered the possibility that A. fulica might carry plant diseases. This possibility finds support in our knowledge of some predisposing aspects of the biology of this snail, viz., the formation of plant wounds through rasping, the proclivity for decaying and rotting vegetation, and the contact with many plants over a relatively short period of time. But although there is no positive evidence in the literature of this snail being implicated in this manner, the lack of experimental evidence one way or the other clearly indicates both the need for such experimentation and the danger of attempting to draw any conclusions at this time. It is apropos, however, to mention at this point references which establish a vector role for gastropod mollusks. Wagner (1896) demonstrated that snails could act as vectors of fungous spores. Later, Gravatt and Marshall (1917) reported that snails were capable of spreading white pine blister rust in greenhouses; on the other hand, slugs apparently were not able to transmit tobacco mosaic to healthy plants (Purdy 1928). Rands (1924) reported that he had repeatedly confirmed the fact that Zonitoides arbor ens is a vector of "root rot" of sugar cane in Louisiana; and Robbs (1946) listed Deroceras reticulatum as one of the disseminators of black rot of cabbages and other Cruciferae. Plate and Fromming (1951) offer additional information on the vector role of slugs in the transmission of fungous diseases of plants. More recently, Fromming (1955) has implicated various species of slugs and snails in the dissemination of potato diseases. Cer- 54 ECONOMIC STATUS tainly the disease vector aspect of the achatina problem needs badly the attention of a qualified investigator, especially since a danger of this sort could exist without raising the suspicions of the average observer. Nuisance Factor As a nuisance factor, the giant African snail at the height of its invasion can be listed among the worst which plague man. Corbett (1933) was one of the very first to recognize the importance of this factor. The snails multiply in such unbelievable numbers that they crawl all over and into everything, they crush or slip out from under foot almost wherever one steps, they cover things with their excreta and sticky slime trails, and they die in great quant i t y for various reasons and create rank odors. In this manner they very seriously interfere with normal living and thus assume an importance which may far outweigh any damage they may do to the vegetation. Therefore, in judging the undesirability of A. fulica, the nuisance factor must not be lost sight of as it alone more than justifies efforts to keep this snail out of uninfested areas. Unfortunately, those who are prone to minimize completely the threat that this snail presents seem invariably to overlook this very important point. The highways and other thoroughfares may be lined for miles with living and dead snails. Individuals attempting to cross to the other side may be crushed into a slimy mess, and as such they provide one of the choicest food items for those that survive. But, ironically, those that feed upon the crushed carcasses all too often become crushed themselves and in turn attract other snails which may suffer the same f a t e . In this manner a genuine road hazard may be formed, especially on the curves and grades. This has been observed in Saipan by Abb o t t (1949), the present writer, and others. Apropos of this, the following words of Kalshoven (1950) do not need translating, "Toen de Amerikaanse troepen er in 1945 landden, waren de slakken er zo overvloedig, dat de jeeps er slipten." R. C. L. Notley (in litt. Dec. 7, 1950) reports them similarly forming "stinking nightmares" on the highways of Ceylon. Their appetite for whitewash (which contributes lime to their shell) permits them almost to exceed their capacity as a nuisance; for although damage to painted surfaces is rarely extensive, it is disfiguring and adds substantially to maintenance costs. A number of investigators have reported damage of this sort, but Witkamp (1941) has illustrated and made more recent special mention of it. In Guam, the giant snail so incessantly interfered with the deratization program, by consuming with apparent impunity the warfarin rat bait, that it was necessary either to prebait with a molluscicide or ECONOMIC STATUS 55 to use concurrently both a snail bait and a rat bait (Peterson 1957b). R. K. Enders observed in his survey of the rat problem in Micronesia in 1949 that rat traps set with ordinary baits, such as peanut butter, cheese, bacon, etc., often caught more giant snails than rats. It will be left to the reader's imagination the slimy mess that is created when a rat trap snaps on a snail! Health Factor By their sheer multiplicity the snails will often find their way into open wells, creeks, and other bodies of water. Their rotting carcasses not only render the water unfit for drinking but present a hazard to health. But the hazard does not rest there alone. The thousands upon thousands of dead and dying snails in a heavily infested area form potent breeding sources for filth flies which are carriers of enteric diseases. Annandale (1919) reported the breeding of "blue bottle flies (Pycnosoma or Lucilia dux)" in this way and associated their appearance in Silpur, India, with epidemics of "enteric." Smedley (1928) similarly reported in Malaya the breeding of Megaselia xanthina (i.e., Aphiochaeta) which he indicated can produce a myiasis in the human intestine. Lange (1947), Williams (1951), and others have made comparable observations. In Chichi Jima and Haha Jima of the Bonin Islands, Musca domestica and other filth flies were found breeding in such great quantities that keeping them from one's food even while eating was almost an impossible task. In this particular situation, however, their breeding was made much worse because the inhabitants gathered the snails in large barrels and allowed them to die in the sun and become maggot infested before crushing and using them as fertilizer. Cockroaches presented an equally serious problem in these islands. Since they were often seen at night feeding on dead achatinas, it is assumed that at the very least the snails have aggravated this problem if not actually precipitated it. The health hazard under such conditions is a very great one but its full epidemiological potentiality will not apparently be realized unless and until an enteric disease agent, for which the native people have little or no resistance, is accidentally introduced by outsiders. The propensity of these snails for feeding on the excrement of humans, household pets, and livestock presents a somewhat different type of problem (Lange 1950; Mead 1950b, c; Garnadi 1951; Williams 1951). Green (1910c) apparently was the first to make this observation; but he believed at the time that by consuming such material, the snails were providing a real source of service. Van Weel (1948), however, interpreted this information in a different way, viz., that the snails would, in addition to their other sins, cause the spread 56 ECONOMIC STATUS of "contagious diseases." Actually, this would only be the case if (1) the snails were used for human consumption in the raw or insufficiently cooked state, or (2) flies fed upon the dead, infected snails and in turn contaminated human food, or (3) the disease producing entities were able to pass through the snail and be deposited in an infective state some distance from the original source of infection. It can be seen that none of these alternatives contributes anything of a significant nature to the epidemiological factors involved. In the case of the latter two alternatives, the filth flies themselves are so effective in carrying enteric disease agents that any possible part the snails might have would be inappreciable indeed. Although the first alternative is an extremely unlikely one, Schnell (1919) has indicated that it is not an impossibility as he reports on a case of mass infection of the dwarf tapeworm (Hymenolepis nana) in prisoners of war who had eaten raw Weinbergschnecken (Helix pomatia). It is of interest to note in passing that Lebour (1915) demonstrated that the slugs Deroceras reticulatum and Arion circumscriptus would naturally consume proglottids of the tapeworms Moniezia cxpansa of sheep and Cittotaenia pectinata of rabbits. Since the ova were found to pass through the digestive tract in a viable state, it is believed that these slugs act as incidental disseminators of disease agents. In the following year, Railliet similarly found fair evidence that carrier pigeons were becoming infected with the tapeworm Bertiella delafondi by consuming D. reticulatum which had earlier fed on infectious material. This same species of slug as well as three different species of land snails were shown by Taylor (1935) to be capable of acting as the intermediate host of the gapeworm of poultry (Syngamus trachea). Both D. reticulatum and Arion ater were found to be carrying on the surface of their bodies viable ova of two nematode parasites of humans, Ascaris lumbricoides and Trichuris trichiura (Galli-Valerio 1918). Hanson (1930) and Blaisdell (1950) report, respectively, that the slugs D. reticulatum and Prophysaon andersoni are the intermediate hosts of the microscopic tapeworm of chickens (Davainea proglottina) and the slug Arion circumscriptus is an intermediate host of the cat lungworm (Aelurostrongylus abstrasus). Similarly the common land snail pests Bradybaena similaris and Subulina octona are intermediate hosts of the cecum fluke of poultry (Postharmostomum gallinum) (Alicata 1938, Thistle 1959a). The rat lungworm, Angiostrongylus cantonensis (Chem), was found to be carried normally by unidentified species of garden slugs (Mackerras and Sandars 1954). Recently, Fromming (1952b) empha- ECONOMIC STATUS 57 sized that the coprophagous habits of slugs strongly implicates them in spreading the diseases of humans and plants. But these items are only suggestive in the case of A. fulica. Under no circumstances can there be found any grounds for the fears of Morgenstern (1949) who believes that these snails could spread the several dread fluke diseases of humans. In East Africa, Williams (1951) observed mosquitoes breeding in empty achatinid shells and thus brought attention to a factor which conceivably could be of the utmost importance in areas where disease-carrying mosquitoes prefer this type of breeding site. Role as Scavenger It has been mentioned that A. fulica provides at least an incidental benefit by consuming human and livestock excrement. A closer examination, however, shows that this is only one example of many which attest to the fact that this snail contributes much to the plus side of the ledger by acting as a general scavenger par excellence. For example, even a dead black rat near Agafia, Guam, was observed by the author to have been eaten by these snails (cf. Dalgliesh 1907). The fact that they avidly consume the shell and flesh of dead and even putrifying individuals of their own and other species is of but the commonest knowledge. Pangga (1947) has stated, "Growing plants furnish the major portion of its food." The majority of investigators do not concur in this. Instead, they feel that rotting, soggy, and decaying vegetables, fallen fruits, leaves, and even entire plants are definitely the most frequently selected food items, even in a great many cases where the plants are never touched in the fresh or growing state (Chamberlin I952a; Esaki and Takahashi 1942; Green 1910^ 19116; Hatai and Kato 1943; Townes 1946; etc.). Jaski (1953) states that after a snail reaches the age of four months, it shows a definite preference for decaying matter. Garbage and trash of almost any sort, especially if they are wet (e.g., water-soaked cardboard boxes), are also preferred food items (Esaki and Takahashi 1942, Pangga 1949)--so much so, in fact, that Rees (1951) has been moved to conclude that this explains the abundance of snails about native villages. It would seem, however, that their abundance in such places more clearly reflects the fact that the natives themselves have provided the best means for spreading these otherwise very slow snails from one village to another and that their garbage, trash, and waste material are only contributory factors. It is assumed that the frequent consumption of moist soil by these snails is still another expression of scavenging which not only brings in an additional amount of decomposing organic material but assures under normal conditions an adequate intake of certain essential inorganic compounds. 58 ECONOMIC STATUS Barnes (1949b) minimizes the importance of the scavenger role of terrestrial gastropods with the suggestion that all the material they eat would normally be broken down by other biological agents. But even in the tropics, agents to which he doubtless refers, viz. bacteria and fungi, reduce the material at a much slower rate. Further, they do not build it back up into an available source of highly nutritious proteins as do the snails, the economic significance of which is emphasized below. Although Green (1910c), Corbett (1933), and a very few others have emphasized the importance of this scavenger role, it is a topic which in general has been ignored by the majority of investigators, probably because the disappearance of dead and decaying material, by its very nature, is neither noticed nor contemplated. On the other hand, even with full cognizance of this scavenger role, it is difficult to evaluate. And, further, because of the unwanted nature of the material consumed, it apparently has seemed pointless to attempt such an evaluation. It will be shown later, however, that a careful evaluation is of worth in that it automatically suggests a very constructive method of control. But, beyond this, it must be concluded that it is the dominance of this scavenger role which unequivocally explains the apparent paradox of great concentrations of these snails existing in uncultivated areas with little or no patent signs of damage to the vegetation. It is quite another story, however, when such concentrations exist in or adjacent to cultivated areas planted with flowers, vegetables, cover crops, or young seedlings or cuttings. In such areas, it is both the normally reduced amount of plant debris and a high preference for succulent plant tissues which decrease scavenging and invite damage. From these statements it can be inferred that in the uncultivated areas A. fulica, through its scavenging, may actually contribute more good than harm and that j u s t the reverse obtains in cultivated areas where the plantings are especially preferred. The nuisance and health factors, of course, will detract in either case to a degree depending upon the nature of the human element. Because this snail spreads largely through the human agency and because it spreads relatively very slowly on its own, there is a strong tendency for snail populations to build up and stay close to inhabited areas. This explains why in the majority of instances the giant African snail must be looked upon as detrimental rather than beneficial. And even in inhabited areas where it causes only a minimum amount of harm, its potentiality for becoming worse argues against maintaining an attitude of indifference. At any rate, there is at this point clearly neither evidence nor excuse for the tall tales which liken A. fulica to a plague of migratory ECONOMIC STATUS 59 locusts. It is only in the most localized areas, under unusually favorable conditions, that damage by these snails will approach anything near the absolute. A not uncommon example of this is found in the small, unprotected, native garden plot surrounded by snail infested bush. The first rains after the dry season may cause the snails to come out of estivation, move into the garden plot en masse and virtually clean it out overnight. But, even under these conditions, the alert native can anticipate this damage and take effective steps to reduce it very considerably. Such measures, however, demand unrelenting effort, and the snail menace therefore still is reduced only to a most undesirable, persistent harassing agent. And in fact it is in this very capacity that the snail assumes its usual role with respect to the susceptible crops of man. Unjust Charges A final aspect in attempting to determine the actual economic status of the giant African snail is found in the subject of ''unjust charges." First of all, the nocturnal feeding and secretive habits have given an insidious flavor to the damage caused by this snail. This in turn has caused the snail to fall victim to a great deal of circumstantial evidence. As in many other systems where trouble is afoot, there is need for a scapegoat, and A. fulica has been a "natural" for this role wherever it has gone. Its large size, recentness of arrival, and ubiquitousness have all contributed to the "case" against it. In areas where this snail has invaded, it is not difficult to find one or more individuals adjacent to plant damage of any sort. The frequency of such observations has persuaded the native people in many quarters to blame the giant snail for almost everything. A closer inspection may show that some smaller, less conspicuous animal is either largely or entirely to blame for the damage. As an example, the purported damage by A. fulica to sweet potato vines in several of the Micronesian islands proved quite clearly to be simple cases of circumstantial evidence. The snails were found in abundance among the obviously chewed vines; however, a more critical examination by the author revealed that it was actually a cutworm and a small flea-beetle which were producing the observed damage. Green (1910c) made comparable observations. The mere presence of many snails seeking the shade of banyan trees, sugar cane, coconut palms, and other essentially immune plants has persuaded native people and certain incautious observers to conclude that even though no discernible damage could be found, damage somehow must be there-- and it has been so reported. In Guam some of the native peoples bitterly accused the snails of 4 'poisoning'' their cats and dogs when in reality death was caused 60 ECONOMIC STATUS from biting the commonly introduced giant toad, Bufo marinus, which is protected by very toxic skin glands. Other, less conspicuous pulmonate gastropods in certain areas join A. fulica in creating damage, but actually may escape accusation. Cases of joint damage of this sort have been reported on tobacco (van der Meer Mohr 1949a), oil palm (Corbett 1941), corn (Mead 1950b), gardens (Corbett 1933), and rubber (Beeley 1938a, b; Robson 1914). In certain areas at least, damage to fruits, such as the papaya, may be caused more frequently by birds and rats than by the giant snail (Chamberlin 1952). The many exaggerated claims and cases of misinformation alluded to earlier are also included under the heading of "unjust charges." In contrast to the statement that they will "eat anything that will hold still," there is an impressively long list of economically important plants which are seldom or never attacked. Actually, their only lack of dietary specificity rests in the eating of dead and decaying material. CHAPTER 5 CHEMICAL CONTROL There is embraced under this heading a vast and widely scattered literature consisting of every possible intergradation between authority and quackery, detail and superficiality, circumstantial evidence and proof. It has taken the utmost effort and an inordinate amount of time to examine, edit, condense, and work into some semblance of suitability the multitude of facts, which together give us a fair conception of which chemicals, formulations, and methods hold the greatest promise for bringing under control our pestiferous terrestrial gastropods. Probably a fair share of responsibility for the many discrepancies in the literature in the field of economic malacology rests in the fact that too little is known about the bionomics of terrestrial mollusks; hence there is little in the way of "standards" upon which to base results of toxicological experimentation. Most often, it is the economic entomologist who is given the task of probing into the essentially unexplored field of malacological toxicology; it is therefore not strange that some of the results have a strong entomological flavor. Because there are so few who have worked in the field, and because of the general reluctance to do work in a strange field, there has been a tendency to "pass on" as authoritative the results, conclusions, and recommendations of earlier workers, and not always with proper credit, until only by virtue of frequency of appearance in the literature have certain methods become accepted as basic and standard. To make matters worse, some authors will incorporate into the recommendations, without adequate experimental testing, certain modifications which occur to them as being likely improvements. From another angle, 61 62 CHEMICAL CONTROL nurserymen and even garden enthusiasts will make their own "discoveries," for which they do not have adequate testing facilities, and report upon them all too often in obscure publications. Discoveries of this sort not infrequently have real merit and warrant further investigation. A very large portion of the literature in malacological toxicology concerns the problem of controlling medically important freshwater snails (Mozley 1952). Because of the aquatic medium and the special nature of the fluviatile and lacustrine environments, the requirements for a suitable molluscicide1 are vastly different than for those of terrestrial snails and slugs. Hence, with only a rare exception, the present coverage of the literature has been limited to that which concerns the terrestrial forms. But even this has been difficult to classify. Under the circumstances it has seemed best to make summarizing comments under an alphabetically arranged series of subheadings based on the more important toxicants, with a small measure of cross reference. In the majority of cases, the less important toxicants have been grouped under such collective subheadings as "Attractants," "Contact Poisons," "Repellents." And, finally, an evaluation is made on the basis of the survey of the literature. This has been made with a twofold purpose: First, as far as possible, to put at the disposal of the reader the very best of, and all the help that, the literature has to offer so that he may determine what combination of known measures will best meet his needs under the circumstances in which he is working--especially where the giant snail is involved; and, second, to emphasize the disturbing paucity of truly dependable information, the serious chasms in our knowledge, and the genuine need for funds and qualified investigators to bring mollusk toxicology at least somewhere near equality with insect toxicology. For over twenty years, no molluscicide of a spectacular nature has been discovered. Progress has largely been limited to modifications in formulation and application. This is indeed a sad state of affairs in a field destined to play a proportionately greater role in our agricultural economy! Chemicals and Compounds Alum (potassium aluminum sulphate): Anderson and Taylor (1926) dusted this chemical on the slugs Deroceras reticulatum with completely negative results, but sprinkling with an aqueous solu1 In the American literature especially, there is a tendency to spell this word "molluscacide"; but, both orthographically and etymologically, such spelling is not acceptable. The recent introduction of the hybridized word "snailicide" is unfortunate. CHEMICAL CONTROL 63 tion was reported to be lethal to both snails and slugs (Anon. 1930). According to Merrill (1929, 1930), in regions where calcium arsenate-bran baits are in competition with green vegetation, Storer found that a spray, consisting of 1/2-l lb. of either potassium alum or ammonium alum in a gallon of water, was superior as a killer of slugs. Smeaton and Smeaton (1906) compounded and patented a vicious but apparently unsuccessful formulation of powdered alum, coconut fiber, silver sand, and--ground glass! Aluminum Sulphate: The use of this chemical as a molluscicide was first recommended by Durham (1920). Hodson (1924) suggested that by combining 1 lb. with 5 gal. of saturated solution of quicklime in water, the scorching effect on the vegetation could be overcome. Later (1925), he recommended a 1:2 by weight powder mixture with lime, broadcast at the rate of ca. 56 lb. per acre, as giving "very efficient control" of D. reticulatum; but he emphasized that it was not practical on a large scale both because of the high cost and the damage to young foliage. Similarly, MacDougall (1931) found aluminum sulphate impractical as a molluscicide; and in the same year, Miles et al. failed to find it lethal to the slug Milax sowerbii. ARSENICALS2 Calcium arsenate: Lovett and Black (1920) tried a great many different chemicals in the control of D. reticulatum and found Ca3(AsO4)2, both as a dust and a spray on chopped lettuce leaves, to be the best. Basinger (1923a, b, 1927) modified the recommendations of Lovett and Black and developed, in the control of Theba pisana, a highly effective calcium arsenate-bran bait (1:16 by weight, made moist, but not wet, by water and scattered 1 lb. of bait to approximately 1,458 sq. ft.). He warned that this is a slow acting poison and that its effectiveness should not be judged until two to three days after it has been scattered. This period should apparently be extended to four days in the case of A. fulica (van Weel 1949). Basinger found, however, that calcium arsenate used as a dust was of no value in controlling T. pisana. He later showed (1931) that the calcium arsenate-bran bait could be used with equal success on Helix aspersa (cf. Anon. 1949h). Lewis and LaFollette (1941) labeled it the best bait for this species, although Hely (1946) warns that it will cause considerable fruit and leaf fall if, as recommended by Lewis and LaFollette (1942a, b), it is scattered in the trees during periods of high humidity. A combination 2 See also Paris green. 64 CHEMICAL CONTROL of 25 lb. of bran, 2 lb. of calcium arsenate, 3/4 gal. of molasses, and 1 gal. of water (or 2 lb. of sugar and 1 1/2 gal. of water), produced a reported 99 per cent kill of T. pisana in South Africa (Joubert and Walters 1951). If slugs were also involved, it was suggested that there be added to the above formulation 1/2-1 lb. of meta fuel or 6 pints of methylated spirits. Lange and MacLeod (1941) found that the mixture of calcium arsenate and metaldehyde was more effective than either chemical alone. The failure of Persing (1944b) to concur in this conclusion was later explained by Lange and Sciaroni (1952) when they warned that increasing the metaldehyde from 1 per cent to 5 per cent by weight in arsenical baits produces a repellent effect. More and more, however, it is being recommended that metaldehyde baits be fortified either with calcium arsenate or sodium fluosilicate (e.g., USBEPQ 1953). A number of investigators have explored the possibilities of using calcium arsenate in the control of A. fulica. Beeley (1938a) recommended Basinger's formula for the control of this species in Malaya and gave comparative cost figures. FitzGerald (1947) supported this recommendation. Pangga (1949), however, found it not very attractive to the giant snail. He tried the powdered calcium arsenate on banana peel, but it had no effect; in contrast, van Weel (1949) found the powder very effective. Since the snails are attracted to and will consume lime, Corbett (1937) suggested the use of balls of lime poisoned with calcium arsenate and, similarly, Rees (1951) recommended a 1 per cent solution of this poison sprayed on lime covered walls. The creation of a longer lasting bait by the addition of cement to the calcium arsenate--lime mixture was first suggested and favorably reported upon by Leefmans (1933^). Its use was strongly recommended by Garnadi (1951); but Fairweather (1937) had "little success" with this new type of bait. Beeley (1938a) recommended a fairly strong bait consisting of one part of calcium arsenate, six parts of slaked lime, and two parts of cement by volume (or 1:4:2 by weight) and sufficient water to form a consistency of ordinary concrete mix. This was dried in thin slabs, broken into small pieces and scattered in the areas frequented by A. fulica. A considerably stronger formulation (8:11:1 by weight plus water to make a thin paste) has been effectively used for the past several years as one of the chief means of combating this snail in Hawaii (Fullaway 1949). This mixture is painted on rocks, posts, tree trunks, and the like. Where there are no such objects, stones or cinders are coated with the mixture and broadcast; or wooden lath is painted with the mixture and CHEMICAL CONTROL 65 placed on the ground as a barrier (Thistle 1954b). According to Thistle (1953b) and Weber (1954), better results have been obtained when small coated stones (1/2-l inch in diameter) were placed two feet apart than when the same weight of large stones (about 2 by 4 inches) were placed ten feet apart. Similar methods are currently being used in a number of other areas including Ceylon (Mead 19556) and New Guinea (where they use crushed coral instead of lime). Because of the high coralline content of the soil in Guam, however, the snails were not sufficiently attracted to the poisoned whitewash to make it effective as a method of control (Peterson 1957). Chipman and Seibert (1939) have patented a molluscicide which combines calcium arsenate and Paris green. Lead arsenate: Lovett and Black (1920) found that slugs would readily eat this poison, but few were killed by it. Torres (1950) recommends its use with sugar and wheat bran (1:1:10 ratio by weight) as a bait to kill the giant South American snail (Strophocheilus oblongus) in coffee plantations. Cameron (1951) and Araujo (1952) recommend similar formulations. A weaker formulation (1:2:17) is claimed by Pereira and Gon^alves (1949) to give an 88 per cent kill of slugs and snails; the addition of 2.5 per cent metaldehyde and the substitution of honey for sugar made no significant improvement in the bait. Sodium arsenate: It is not readily eaten by slugs and tends to blacken lettuce bait (Lovett and Black 1920). Sodium arsenite: This chemical ("penite" is a 40 per cent commercial solution) was used in the form of a spray in early, futile attempts in 1946 to eradicate A. fulica in Guam. More recently it has been used quite successfully as a 1 per cent spray (i.e., a 2 1/2 per cent "penite" solution) against this same pest in Hawaii, especially where infestations are heavy (Lennox 1953, Weber 1954). Later, the spray was used as a 0.5 per cent solution with no reduced effectiveness (Thistle 1953b). Experiments conducted by Q. C. Chock (in litt. Jan. 7, 1946) gave strong indications that poisoning was effected through absorption by the foot, although Weber assumes that some have been killed by ingesting the poison-covered leaves. Since much of the vegetation dies within a few days after being sprayed, this chemical is generally considered a "weed killer" (cf. Basinger 1927). Its highly toxic nature makes it dangerous to use under most conditions where the snail is found. Furthermore, it is expensive to use, even under limited conditions, and a rain quickly dispels its effectiveness. 66 CHEMICAL CONTROL White arensic: This was found not to be eaten readily by slugs; when it was eaten, the kills were low (Lovett and Black 1920). Pangga (1949) found it ineffective in the control of A. fulica. Ashes: In the control of A. fulica, ashes or a suitable substitute (e.g., sawdust, charcoal, cinders, etc.)as a repellent barrier have been recommended by many, including Green (1911c), Hutson (1920), South (1926b) and Corbett (1933). Most often, it is suggested that they be used in combination with copper sulphate (q.v.). The continued use of ashes introduces the danger of moving the soil pH too strongly in the alkaline direction. ATTRACTANTS3 Bran of wheat, rice, or other grains is almost the universal attractant (vide Diluents) in snail and slug baits in spite of the fact that its availability and cost vary considerably from area to area. Lewis and LaFollette (1941, 1942a, b) and Persing (1944a) found in southern California that fresh orange pulp (in a 20--25:1 by weight ratio with calcium arsenate) was about equally effective in controlling H. aspersa in citrus groves but only one quarter as expensive as when bran was used. A combination of equal parts of bran and pulp was shown to be more effective than either one alone. The dried pulp however proved to be inferior to bran. On the other hand, dried citrus peels have been demonstrated to be effective both as an attractant and a diluent. Lime and calcium carbonate in various forms have been used as attractants in a number of poison baits, but particularly in those containing calcium arsenate and/or metaldehyde (q.v.)· Metaldehyde itself has been considered by many to be a specific attractant for snails and slugs under proper conditions. A great many different substances were recently tested in Hawaii in an effort to find an effective, practical attractant for A. fulica. Metaldehyde emerged as the only one with any real possibility (Thistle 1953b, 1954b); and even this in the dry form or in solution proved to be "entirely negative" in these experiments, as was wheat bran in either the dry or fermented state. Many commercial bait preparations contain amyl acetate as an attractant. Barium Fluosilicate: A spray of this chemical was used in attempt to control H. aspersa in citrus groves in Australia but was not 3 The less commonly used term "attrahent" is not used here because of its medical connotation suggesting, antithetically, an irritant. CHEMICAL CONTROL 67 f o u n d to be effective enough to be of value (Hely 1946). Levy (1938) patented a snail poison containing this chemical and metaldehyde. Benzene Hexachloride (BHC}: Pereira and Goncalves (1949) mentioned without elaboration that for killing snails and slugs, this chemical was inferior to metaldehyde. On the other hand, Rao et al. (1953) reported that a 5 per cent dust broadcast about 25 lb. per acre of drained paddy was superior to a 2 per cent metaldehyde bait in that it "killed most of the snails in the fields." Pappas and Carman (1955) found it slightly less than moderately effective in killing H. aspersa. Bordeaux: This mixture of lime and copper sulphate was shown to be an ''excellent repellent for slugs and non-injurious to foliage" (Lovett and Black 1920). Miles et al. (1931) similarly recommended its use. It was found to be only of limited use however in repelling H. aspersa (Lewis and LaFollette 1941). Similarly, Basinger (1927) discontinued its use against T. pisana because it did not give desired results. On the other hand, a spray consisting of a Bordeaux mixture (2:2:80), white oil emulsion (1 gal.), and nicotine sulphate (34 pint) was found to be more effective than any other control measure of H. aspersa in citrus groves in New South Wales (Hely 1946). Later reports (Anon. 1949h) suggested the omission of the nicotine sulphate and pointed out that the spray not only had a considerable knockdown effect on the small snails, but this effect was retained in a residual fashion for a period of several months. Pangga (1949) reported that a spray of this mixture was toxic to only the young specimens of A. fulica. Carbon Disulfide: This gas was used in Sidney, Australia, to refumigate (after hydrocyanic acid gas treatment) a shipment of copra infested with A. fulica (Harrison 1951). Chlordane: Pappas and Carman (1955) demonstrated in field tests that this insecticide is of essentially no value in the control of H. aspersa. Pangga (1949) reported without elaboration that the "Octaklor spray" had toxic effect upon only young A. fulica. Coal tar: The use of tarred coconut fibers around the trunks of fruit trees to protect them from the attacks of A. fulica was apparently first reported by Green (1910c) and reiterated by Hutson (1920) and van Weel (1949). G. S. Dun (in litt. April 21, 1952) put a ring of coal tar around plantings in his own yard in New Guinea; this very successfully and inexpensively repelled the giant African snails. The rapid growth of weeds, however, demanded replacement about every two weeks. In Ceylon, R. C. L. Notley (in litt. 68 CHEMICAL CONTROL Dec. 7, 1950) succeeded in keeping the tar moist by mixing it with old engine oil. Boards painted with this mixture and laid about the garden formed an effective barrier against this snail. CONTACT POISONS Carbolic acid (vide Phenol), cedar oil emulsion, clove oil emulsion, gasoline, nicotine sulphate, nicotine resinate, oil of lemon, oil of tar emulsion, sodium hydroxide, sulphur and tobacco dust were used by Lovett and Black (1920) in various combinations and concentrations on the slug D. reticulatum with varying results. Even with this small naked gastropod, these substances were considered to be of "decidedly minor value in practical field work." Another series of chemicals was tried by Hodson (1924) in an attempt to find a good contact poison for slugs. The following were without effect: dichlorbenzene (10 per cent sol.), potassium bichromate (10 per cent sol.), derris powder, derris solution, mustard (brown) solution, and sodium silicofluoride (10 per cent sol.). Potassium xanthogenate solution, sodium hyposulphite (10 per cent sol.) and, very strangely, copper sulphate (10 per cent sol.) were reported as having only transient effects. Chloral hydrate (10 per cent sol.) proved to be irritant, but its cost made it economically impractical; on the other hand, borax was "extremely lethal" but it was destructive to vegetation. Miles et al. (1931) tried still other substances on the slug Milax sowerbii. Dusting specimens of this species with the following did not prove lethal: aluminum sulphate, ammonium chloride, copper carbonate, copper sulphocyanide, flake naphathlene, flowers of sulphur, green sulphur, lead sulphocyanide, potassium permanganate, precipitated chalk with chlorcresylic acid, precipitated chalk with creosote, sodium nitrate, thiourea, and "used'' calcium carbide. On the other hand, they found that the following substances would kill the slugs within a few minutes after being dusted on them: ammonium sulphate, ammonium sulphocyanide, barium sulphocyanide, calcium carbide, calcium cyanide, corrosive sublimate, drained creosote salts, potassium sulphocyanide, sodium carbonate and sodium sulphocyanide. Of these, only calcium cyanide was sufficiently lethal at practicable concentrations to be considered for use in field conditions; its very deadly nature, however, makes serious consideration completely out of the question. Mixing ammonium sulphate with steamed bone flour gave negative results as a contact poison for slugs (Anderson and Taylor 1926). More recently, Lange and Sciaroni (1952) reported that CHEMICAL CONTROL 69 dusts of hydrated lime, cupric oxide, and nicotine were effective as contact molluscicides for only a short period of time. The insecticide hexaethyl tetraphosphate (HETP) was ineffective as an aerosol in fumigating greenhouses infested with snails (Smith et al. 1948). A Russian patent suggests that a spray consisting of at least a 0.1 per cent aqueous solution of ethylene chlorohydrin (2-chloroethanol) will kill slugs (Paikin et al. 1949). Copper sulphate, Bordeaux mixture, and metaldehyde may be classified as contact poisons, but they have been treated in detail under separate headings. COPPER COMPOUNDS4 The following compounds of copper were used in aqueous solution of 1 per cent or less by Lovett and Black (1920) in an attempt to find a control for the slug D. reticulatum: acetoarsenate, benzoate, carbonate, chloride, chromate, cyanide, ferrocyanide, and sulphate. Only the chloride and sulphate compounds produced a burning of foliage. The results of their experiments caused them to dismiss summarily all of these compounds as having no promise in slug control. Basinger (1927) found the use of CuSO4 impractical in the control of T. pisana because it caused damage to plants. Using lower concentrations of this chemical, MacDougall (1931) reported effective control of slugs with no damage to plant foliage. Anderson and Taylor (1926) similarly recommended as a "deadly" control for slugs a 4-6:100 by weight mixture of CuSO4 in the less toxic kainite fertilizer (hydrous potassium-magnesium chlorosulphate), distributed 2-3 hundredweight per acre. Miles et al. (1931), after considerable experimentation with this chemical, cautiously suggested that a precropping treatment with crystalline CuSO4 at the rate of one hundredweight per acre might bring protection from the slug pest M. sowerbii; but they warned against damage to foliage if this chemical is broadcast after the crops have come up. To prevent slug migration, they recommended sprinkling peripheral ditches with CuSO4 or a mixture of this chemical and ground limestone. For slug control Lange and Sciaroni (1953) suggested a dust containing monohydrated copper sulphate and hydrated lime in a 20:80 ratio. These and other experiments with CuSO4 focused attention upon this chemical in early attempts to control A. fulica. Green (1910c), Hutson (1920), South (19266), Corbett (1933), and Fernando (1952) have recommended its use in almost every conceiv4 See also Bordeaux, Paris green. 70 CHEMICAL CONTROL able manner; shallow ditches around gardens were filled with wood ashes, coconut fiber dust, or sawdust treated with a 4-10 per cent solution of CuSO4; some of this same mixture was spread around small beds and individual special plants; coir ropes were treated with 10 per cent solution and placed around vegetable gardens and around the bases of trees; small crystals of CuSO4 were placed at the surface of the ground near plants needing special protection; a solution consisting of one lb. of CuSO4 to ten gal. of water was sprinkled on the ground where the giant snails were numerous; stone walls specially constructed around vegetable gardens were watered two to three times a week with a solution of one lb. CuSO4 in a gal. of hot water; and snails collected by hand were first drowned in a 4 per cent solution before being discarded. To insure against attacks by A. fulica in Ceylon, the sticky seeds of the cacao are rolled in a mixture of CuSO4 and ashes before being planted (R. C. L. Notley in litt. Dec. 7, 1950). Pangga (1949) tried 1-10 per cent solutions of CuSO4 on giant snails and found that, although the young individuals were readily killed, only a few adults were affected. The copper-based commercial fungicides, Greenol, Omazene, Crag 658, Cunimene 2243 and Corona 53 all failed to control slugs without phytotoxic effects (Karlin and Naegele 1958). Corrosive Sublimate: A 1:1000 aqueous solution proved "highly efficient as a repellent" in combating M. sowerbii (Miles et al. 1931). This recommendation has subsequently been indorsed by a number of investigators. Even in low concentrations, however, this is a dangerous chemical to use. Creosote: One per cent mixture of creosote in precipitated chalk gave "outstanding" results as a slug repellent (Miles et al. 1931). Cryolite (sodium fluoaluminate): This spray was shown by Lewis and LaFollette (1941, 1942a, b) to be of little value in controlling H. aspersa in California citrus groves. The same conclusions were announced by Hely (1946) in Australia. DDT: The few experiments which have been performed to determine the possible molluscicidal properties of this insecticide indicate definitely that further experimentation is needed. Buckhurst (1947) had at first only slightly encouraging results with the 0.1 per cent aqueous emulsion, the 5 per cent dust, and the "DDT bait." Slugs did not touch the bait and only very slight kills were produced by the dust and the emulsion; there was however a very clear deterrent effect. But when he increased the strength of the CHEMICAL CONTROL 71 emulsion to 0.2 per cent and applied it at the rate of 2 gal. per 100 sq. yd., he found it produced "an excellent control of slugs without damage to growing cabbage plants." It was determined that the killing power of the application was retained for at least two weeks. Thomas (1948) tried a DDT-bran bait and found that although the slugs were attracted to it (undoubtedly because of the bran, vide Attractants), they were not killed by it. This insecticide in a kerosene emulsion was tried in an undisclosed percentage on A. fulica by Pangga (1949); young specimens were killed but the adults remained unaffected. Fromming (1949, 1950) and later Fromming and Riemschneider (1952) and Fromming and Plate (1952) failed to demonstrate any molluscicidal action in preparations of DDT and other insecticides. Hely (1946) found DDT of essentially no value in controlling H. aspersa in citrus orchards in Australia; Pappas and Carman (1955) concurred in this. Pereira and Gon^alves (1949) and Karlin and Naegele (1958) tried it on slugs and snails but did not recommend it. When it was used as an aerosol to fumigate snail infested greenhouses, it proved to be ineffective (Smith et al. 1948). Dieldrin: W. H. Lange, of the University of California at Davis, has indicated in a preliminary report (in litt. Jan. 29, 1952) that, even when combined with metaldehyde (q.v.), the action of dieldrin on H. aspersa suggests that it has relatively little value as a molluscicide. Pappas and Carman (1955) concur in this. Karlin and Naegele (1958) found it "ineffective" in controlling slugs in greenhouses. DILUENTS The most desirable diluent in snail baits is one that acts as an attractant (q.v.) and is relatively inexpensive. Wheat bran has long been the diluent of choice, especially in metaldehyde and calcium arsenate baits. More recently, apple pomace has been successfully used, particularly in a 1:1 combination with wheat bran. In the Orient, in particular, rice bran is more commonly used; and although it has been reported to be less good than wheat bran, its much lower price makes it more economical to use (Callan 1941). Coconut meal, corn meal, rice husk, sawdust, and other diluents have been used with lesser varying degrees of success. Sawdust in particular is of uncertain value because of the aromatic nature of some wood, even when thoroughly seasoned. In cases where only a mediocre diluent is available, some compensation may be made through the addition of an attractant, for example, amyl acetate. 72 CHEMICAL CONTROL Ethylene Dibromide: This was shown by Balock (1951) to be seventeen times as effective as methyl bromide in killing immature stages of the oriental fruit fly in papaya. It is immediately apparent that serious consideration should be given to the replacing of methyl bromide, carbon disulphide, and hydrocyanic acid gas (q.v.) by this more toxic chemical in the fumigation of snail infested cargoes. Ferrous Sulphate: When specimens of T. pisana were placed on a surface dusted with an approximately 80 per cent proprietary mixture of this chemical, they displayed great irritation, withdrew into their shells for four weeks, but remained alive (Basinger 1927). Pierce (1931) enthusiastically recommends dusting this chemical about the garden to kill slugs and snails indicating that not only are good kills obtained but the chemical itself is a plant food and stimulant. Gammexane (gamma isomer of hexachlorocyclohexane): Two investigators report on the use of this chemical in the control of A. fulica. Pangga (1949) states without elaboration that gammexane was "ineffective." In direct contrast, J. A. Tubb states (in Hit. Jan. 8, 1952) that it is "effective" in North Borneo. Since this insecticide imparts an off-flavor to fruits and vegetables, it at best would be of limited use. Fromming and Riemschneider (1952) report that HCH was ineffective on limacine slugs both as a contact poison and a stomach poison. Hydrocyanic Acid Gas: The Bureau of Plant Quarantine in California has used this gas to fumigate the holds of ships infested with A. fulica. It was used for the same purpose in Vancouver, British Columbia (in litt. W. Reed Mar. 9, 1951; cf. Zuk 1949), and in Sydney, Australia (Harrison 1951). I solan: This toxicant was found to be considerably more effective as a molluscicide than any of 31 other promising chemicals tested by Pappas and Carman (1955). Both as a spray and a bait it approached metaldehyde in effectiveness. Kerosene Emulsion: This was tried, in a manner not indicated, by Pangga (1949) on A. fulica and was found unsatisfactory inasmuch as it was toxic only to the young snails. Lime: Dusting hydrated lime, Ca(OH)2, about slug infested plants (Massee 1928) or even on the slugs as a contact poison (Anderson and Taylor 1926) gave unsatisfactory results, especially since some plants may be damaged in the process (Anon. 1930). Blauvelt (1952) suggested that CuSO4 be added to the lime; and similarly, Lange and Sciaroni (1953) report that lime has only transient CHEMICAL CONTROL 73 value unless it is combined with 20 per cent monohydrated copper sulphate (vide Bordeaux). Pangga (1949) combined it with sulphur and tested it on A. fulica, but it proved toxic only to immature specimens. Where lime is lacking or low in the soil, it can be used successfully as an attractant (q.v.) in snail baits; but its prolonged use in acid soils, or the practice of "liming" soils, in general creates a more favorable environment for snails (Atkins and Lebour 19230, b). Metaldehyde: According to Gimingham (1940), this chemical was first used as a molluscicide in 1934 in South Africa, although it apparently was not used in this manner in England until 1936 (Jary 1939). In an earlier work by Gimingham and Newton (1937) and in a fairly complete summary by Barnes and Weil (1942), there is indicated much uncertainty, approaching the legendary, as to j u s t how the molluscicidal properties of metaldehyde were discovered. It is an inflammable polymerized form of acetaldehyde with a chemical formula of (CH3CHO)4, and is the chief constituent of the solid, commercial "meta fuel," a form in which it is often sold. Its solubility in water is low in the extreme--being 0.018 per cent at 0.5° C., 0.020 per cent at 17° C. and 0.026 per cent at 30° C.--with no detectable tendency to depolymerize in neutral or alkaline aqueous solutions even after standing for over six months (Cragg and Vincent 1952). There is no unanimity among investigators as to just what is the action of metalhyde on various mollusks. Many agree that it is a specific attractant. This was early questioned by Jary (1939). It was further questioned by Corbett and Pagden (1941), who announced that individuals of A. fulica will crawl toward bait containing metaldehyde and then change their course just before they get to it. Weber (1954) reported that in carefully controlled experiments dry metaldehyde and metaldehyde "in solution'' failed to attract A. fulica. Lewis and LaFollette (1942a, b") went a step further by concluding that under certain conditions it is actually repellent to H. aspersa. Lange and Sciaroni (1952) explained this and warned that in higher concentrations, metaldehyde becomes a repellent. Lewis and LaFollette further concluded that slugs are more susceptible to metaldehyde than are the snails. Gammon (1943) drew the same conclusion; but Thomas (1948) suggested that just the reverse is the case. Jary and Austin (1937) were apparently the first to conclude that metaldehyde is both a contact and a stomach poison. This conclusion was supported by Cameron (1939) and Lange and 74 CHEMICAL CONTROL MacLeod (1941). In contrast, Thomas (1948) believed that the action of metaldehyde, as far as snails are concerned, is entirely that of a stomach poison. And inversely, he explained that the action on slugs is threefold, viz., an irritant effect, causing excessive production of mucus, from which the slug will recover if it is washed off and kept from desiccating air; an anaesthetic effect causing complete immobility except under the most severe stimulation; and last, an irreversible, lethal, toxic effect which produces a characteristic transparency in the gut wall. The classical work of Cragg and Vincent (1952) confirms some of Thomas' conclusions and disproves others. They demonstrate indisputably that metaldehyde is not only a contact and a stomach poison, but that it has a progressively greater toxic action in the haemocoele, in the crop, and on the surface of the body; hence they recommend its greater use as a contact agent. They could not confirm a "characteristic" transparency of the gut wall emphasized by Thomas. Nor could they detect any fumigant action. However, they did indicate that as a physiological reaction to metaldehyde, there is a loss of water which continues even under conditions of 100 per cent relative humidity. Lange and MacLeod (1941) stated that metaldehyde will produce greater kills of slugs and snails in unshaded areas because of the killing action of the sun on the "stunned" individuals. Essentially the same explanation was made by Woglum (1943) and later by Persing (1945b), who reported that a substantial percentage of H. aspersa will recover from metaldethyde poisoning during cloudy weather or in the shade (cf. Lewis and LaFollette 1942a, b). Although the same reason is not given by Corbett and Pagden (1941), they imply reaching the same conclusions through their recommendation that less meta is needed for A. fulica when the bait is placed in exposed places. Later Lange and Sciaroni (1952) stated that "in protected situations metaldehyde does not seem to be an active enough poison to give high mortalities when used alone in baits." Lange subsequently determined that an excessively high concentration of metaldehyde stuns the snail before it can consume a lethal dose. Under conditions of high relative humidity, Lange and MacLeod obtained greater kills, but Stringer (1946) reported less favorable results because of the greater recovery of the slugs. Thomas (1948) explained that when metaldehyde is used during warm, wet weather, there will be a greater number of slugs caught but only a small percentage of them will be killed; and inversely, during cold, CHEMICAL CONTROL 75 dry nights, the number caught will be smaller but the percentage of kill will be larger. From this he concluded that the maximum kill can be obtained when metaldehyde is used during wet, warm nights which are followed by dry, warm days. Again, Cragg and Vincent (1952) beautifully elucidated the several points of apparent confusion. First, they indicated that metaldehyde intoxication causes immobilization, inhibition of feeding, and a loss of water through secretion of mucus. Under optimum conditions of high humidity and low temperature, the snails and slugs may recover. But with low humidity, there is a greater loss of water and death ensues. High temperatures similarly produce a greater incidence of death through the greater uptake of poison, the increased toxic action, and the greater activity of the snail. It should be pointed out, however, that increased activity of the snails, and therefore increased opportunities for contacting the metaldehyde, depends not only upon a higher temperature but a concomitant high humidity--which high humidity, ironically, reduces the killing effect of metaldehyde! Moreton (1953) takes advantage of this information by suggesting that during dry weather, lower concentrations of metaldehyde may be used effectively. And conversely, it is recommended that during cool, wet weather metaldehyde baits be fortified either with calcium arsenate or sodium fluosilicate (USBEPQ 1953). The more recent works on this subject unfortunately seem to neglect the important factor of the killing action of the sun; hence the ideal conditions, as set forth by Thomas, should be amended to read, "clear, dry warm days." Apparently because of the confused picture presented by the interplay of several of the irregularly variable ecological factors, some German investigators have come to question seriously the reliability and practicability of metaldehyde baits (Fromming 1951, 1952b; Fromming and Plate 1952; Plate and Fromming 1952), although Trappmann (1952) in rebuttal stoutly defends their use. Lange (in litt. Feb. 14, 1956) adds a welcome clarifying note with the following words: "In my experiments I can get from practically no control with metaldehyde to 100 per cent kill within a few feet difference in location, depending upon whether the affected mollusks are exposed to desiccatory conditions or to high humidities. The variations in results with metaldehyde I am convinced are due to the multiple and variable climatic conditions prevailing at any particular time." A variety of "carriers" were tried with metaldehyde by Barnes and Weil (1942) in an attempt to formulate a more effective bait. Considering practicability and percentage of kill, none proved 76 CHEMICAL CONTROL more promising than the old standby, wheat bran. At first, an excessively strong formulation of 1:2-1:2 1/2 was used (Newton 1937). Shortly after that, however, it was decided that a bait of approximately the following proportions was adequate: 1/3 oz. "meta," 1 lb. bran, and 1 pt. water (Cameron 1939, Esslement 1938, Thomas 1944, et al.). Without the water, the bait is considerably less effective (Pereira and Goncalves 1949). It has been suggested (Anon. 1949h) that to the above formulation 1 oz. of calcium arsenate or y2 oz. of Paris green be added for a greater kill. Reflecting a very definite, recent trend, a stronger bait has been recommended by the USDA through the USBEPQ Farmers Bulletin No. 1895 (1953) in that to a mixture of 1 oz. of metaldehyde and 2 lb. of wheat bran, corn meal, or similar material, there be added either 2 oz. of calcium arsenate or 1 oz. of sodium fluosilicate. This mixture is moistened with water just before use and applied to small gardens at the rate of 1 lb. per 1,000 sq. ft. or 40-50 lb. per acre in larger areas. On large scale operations, it is recommended that the bait be machined into pellets and broadcast at the rate of 5-10 lb. per acre. The addition of lead arsenate and honey to the metaldehyde-bran mixture apparently makes no improvement in the bait (Pereira and Goncalves 1949). Putting sugar or molasses in the basic formula did not increase significantly its attractiveness as a bait but it did increase the percentage of kill (Thomas 1948). But baits of this sort understandably spoiled more quickly. In an attempt to overcome some of the difficulties of a mealy bait, Thomas made bait "biscuits" by combining plaster of Paris (dry), metaldehyde, and bran in a 1:1:10 proportion by volume, mixing, adding water to permit molding into 3 by % inch discs, and sun-drying until completely hard. These proved to be more attractive to slugs but they produced less of a lethal effect than the standard mix. Substituting casein glue for plaster of Paris made the "biscuits" more attractive and more lethal only to slugs with carnivorous proclivities. A formula for a more conventional biscuit is under British patent (Boot's et al. 1939). In addition to metaldehyde, the biscuits contain bran, flour mucilage, lard, and a leavening agent, and are baked at about 130° F. to avoid volitalization of the metaldehyde. Another novel, but dubiously effective, approach to the problem of making a ''lasting'' bait was made by Beekler (1944) who suggested that finely divided metaldehyde be suspended in paraffin! Ever since shortly after its discovery as a molluscicide, considerable use of metaldehyde has been made in combating A. fulica. In CHEMICAL CONTROL 77 fact, there undoubtedly is no infested area of any appreciable size in which metaldehyde has not been tried in some form or combination. Beeley (1938a) in Malaya was probably the first to suggest its use for this pest. He recommended a 1:64-1:32 (1.6-3.1 per cent) mixture, by weight, with rice bran, moistened and distributed three pounds to the acre. Dias and Thamotheram (1939) and Fernando (1952) in Ceylon made similar recommendations but indicated more specifically the addition of approximately 26 oz. of water (about 11/4pt.) per lb. of the bait mix. Similar formulations were used in Mauritius (Anon. 1942), Sumatra (Feij 1940), and elsewhere. The relatively unimportant catches made by this bait (van Weel 1949) suggested the need for a stronger formulation. The percentage of metaldehyde was increased to 5-10 per cent by Pangga in the Philippines (1949) with 'Very satisfactory" results. In the same area, Cendafia (in litt. Feb. 12, 1952) used a 10 per cent mixture on slices of ripe papaya and managed to produce a kill of 33,863 giant snails with two kilos of bait. He complains, however, that the bait lasted only about three days; but that is not a new problem. Dr. J. J. H. Szent-Ivany reported at the Ninth Pacific Science Congress that Dr. Bridgeland of New Britain successfully protracted the effectiveness of metaldehyde by mixing it with paraffin oil. Corbett (1938) attempted to protect metaldehyde from rain and to keep it from poisoning poultry by putting it in impractical cigarette-can cages. Dias and Thamotheram (1939) suggested the use of a hood of corrugated iron. A shield of bamboo has been suggested by others (Beeley 1938a, Anon. 1942). Inverted wooden blasting powder boxes were used successfully in Guam (Peterson 1957). R. C. L. Notley (in litt. Nov. 17, 1951) met the problem in Ceylon by putting the metaldehyde-bran bait on the underside of curved tiles. Elsewhere in Ceylon the author observed the use of metaldehyde bait shields made by interlacing two large leaves, petiole-to-midrib. When the potency of the bait had been spent, however, these shields afforded excellent sanctuary from the killing sun. In some cases, the vast accumulation of snail excreta vouched for the fact that the leaves had provided effective shelter for a considerable period of time. To get a more durable bait, Altson (1950b) mixed metaldehyde with cement and lime to form a so-called "brick bait." This he found durable but insufficiently attractive. The addition of rice bran as a fourth ingredient, however, produced a bait which was attractive both to snails and slugs and was clearly superior to the 78 CHEMICAL CONTROL cement-lime-calcium arsenate "brick bait" in areas where lime was naturally abundant. Numerous experiments showed that increasing the bran increased the attractiveness of the bait; and increasing the cement increased its durability. It was suggested that metaldehyde, lime, rice bran, and cement be compounded by weight in the ratio of 1:2:6:6. Since rice husk is cheaper, three parts of it can be substituted for two parts of the rice bran and the cement can then be reduced by one part. A number of modifications of the basic metaldehyde-bran bait have been made by various investigators. Powdered CaCO3 has been used in place of, or in addition to, the bran with greater success in areas where the soil is low in lime (Notley loc. cit.). Banana peelings (Pangga 1949) and papaya fruits (Cendafia in litt. Feb. 12, 1952) similarly have been used as attractants with metaldehyde. C u t t i n g the cost of the bait by substituting sawdust for the bran produced a less attractive bait, but one that was effective for 8-10 weeks (G. S. Dun in litt. April 17, 1950). The use of the straight metaldehyde in either the pellet or powdered form has been advocated by many (FitzGerald 1947, Townes 1946, etc.); this is not only more expensive than the mixed baits, but it is actually less effective. Speyer's conclusions (1954) bear out this assumption. If it is effective at all, the only thing that can be said in its favor is that it does not require costly preparation. To increase the spreading effect as a contact poison, Jary (1939) mixed the powdered metaldehyde with sand and obtained impressive results. In an effort to meet a similar need in protecting orchids with a thin cover of metaldehyde, Alicata (1950) used a 1 per cent aqueous suspension spray with a far better killing effect than was found with the metaldehyde baits. Blauvelt (1952) emphasized that a 10 per cent metaldehyde dust on orchids killed more slugs by contact than would be attracted to baits. These reports persuaded Jefferson (1952) to make comparative studies of sprays and dusts. He prepared his wettable powder by grinding the metaldehyde with a silicate clay in a hammer mill. The 66.6 per cent metaldehyde-clay concentrate powder, in turn, was mixed with pyrophyllite to produce metaldehyde ''dusts" of varying concentrations. After considerable experimentation, he found that a 15 per cent dust with a coverage of 1-2 lb. per 1,000 sq. ft. was most economical and more effective than heavier coverages of lower per cent dusts. The dust should be applied at night and in at least three applications at 7-10 day intervals. A 10 per cent dust is reported to be effective in contact killing not only the adult slug, FIG. 4.--In many parts of central west Africa giant snails of several species provide the greatest single source of protein in the diet of humans The snails may be purchased in the market place ( l e f t ) are either prepared for immediate consumption or removed from the shell, put on wooden skewers smoked, dried, and stored for consumption during the dry season (right) (Photo courtesy of IFAN. A Cocheteux, Ivory Coast.) FIG. 5.-The tmnk of the jak fruit tree Artocarpus hetei ophijllus piovides refuge foi the giant Afncan snails duimg the day and its sweet b a i k supplies one of the choicest food items during the night Fieshly r a s p e d aieas appeal low on the tiunk. FIG. 6.--Shells showing characteristic damage by the jungle crow (Centropus chlororhynchus) in Ceylon. This type of damage is in contrast to the destruction of the apical whorls by the bandicoot and the flecking-off of the body whorl by the rat, the hermit crab, and the coconut crab. FIG. 7.--Variability in Achatina fulica Bowdich. From upper left to lower right, the specimens were collected in: Oahu, Hawaii (ca. 140 m m . ) ; Mombasa, Kenya; Saipan, Mariana Is.; Pallekelle, Ceylon; Koror, Palau Is.; Agiguan, Mariana Is.; Saipan; Luzon, Philippine Is.; and Calcutta, India. FIG. 8.---Approximately one out of four specimens of the Army Hill population of the giant African snail in Saipan in 1949 were of this "bent-nose" type with an arcuate axis, highly irregular lower whorls, heavy lamellate deposition of calcareous material on the columella, and an almost complete CHEMICAL CONTROL 79 D. reticulatum, but the eggs as well (Lange and Sciaroni 1952). According to Jefferson, 10--15 per cent metaldehyde dusts were not satisfactory for H. aspersa and Oxychilus cellarius. Dusts composed entirely of 42-50 per cent wettable powder, however, were lethal to the snails, but obviously were economically unfeasible. A spray consisting of 6 lb. of 50 per cent wettable powder to 100 gal. of water and distributed at the rate of 1 gal. per 30-35 sq. ft. proved lethal to snails. It was recommended, however, that in addition to this spray metaldehyde baits be used as a supplementary measure. Jefferson pointed out that the sprays have a disadvantage in that they are more costly to prepare and more difficult to apply than the dusts. Doucette (1954) determined that a 15 per cent dust, in the absence of rain, was effective for a maximum of only 6-7 days, in contrast to a 2 per cent pellet bait which was effective for 12 days. He therefore concluded that the baits should be used on fallow ground; but in gardens, where the bait would be in competition with green foliage, metaldehyde dusts should be applied in bands in such a way that they would have to be crossed to get to the plants. In contrast to Jefferson's spray application equivalent to ca. 40 lb. of metaldehyde per acre, and more in keeping with the findings of Cragg and Vincent, Moreton (1953) recommended a spray application equivalent to 2.5 lb. of metaldehyde per acre, with the comment that increasing it to 5 lb. per acre brought in only slightly higher catches. But even at the lower amount, he contrasted it with the 0.5-0.75 lb. of metaldehyde per acre required in the distribution of "standard bran baits" and pointed out the greater cost must be weighed against the potentially greater effectiveness of a contact poison as compared with a stomach poison. In quite the other direction, Leoni (1953a, b) recommended an impractical and admittedly expensive spray containing linseed oil and whole milk in addition to an excessively high percentage of metaldehyde. Jefferson determined that wettable powders and dusts of metaldehyde deteriorated rapidly and therefore must be prepared only a short time before use. Cragg and Vincent and Moreton demonstrated experimentally that metaldehyde sprays and dusts after application lose their effectiveness within a very few days; hence, treatment must be repeated periodically. These findings are in sharp contrast to the report of Blowers (1954) wherein he states, without presentation of experimental evidence, that a recently released commercial suspension of metaldehyde will leave a semipersistent coating of metaldehyde "sufficient . . . for many weeks." 80 CHEMICAL CONTROL Similarly, Karlin and Naegele (1958) found a 15 per cent metaldehyde dust superior to a 20 per cent spray in that it remained effective for a period of three weeks in killing slugs. The relatively evanescent property of metaldehyde in sprays and dusts, and the consequent need for repeated frequent application, sharpen just that much more the contrast to the proportionately small amount of metaldehyde required when standard baits are used. The experimental work with metaldehyde teaches us that, in the final analysis, the most economical type of application can be determined only when the density of the snail population, relative humidity, temperature, sunlight, available shade, extremes in the diurnal-nocturnal temperature cycle, and other environmental conditions are known or can be discovered. At the present time, only in the Hawaiian Islands have metaldehyde dusts and sprays been used against A. fulica. Although it has been used in strength equal to that of the dust, the spray has given much better results (Weber 1954). Regarding the spray, Thistle (1953b, 1954b) reports that "while transitory in effect, it is useful . . . where arsenical or other highly toxic sprays cannot be used." Lange and MacLeod (1941) reported that the following aldehydes mixed with bran in 3 per cent strength were not observably attractive to slugs and snails: acetaldehyde, paraldehyde, hexaldehyde, butyraldehyde, propionaldehyde, valeraldehyde, and heptaldehyde. In carefully controlled experiments with D. reticulatum, Cragg and Vincent (1952) found no contact, haemal, or stomach poison effect in acetaldehyde and paraldehyde. Leoni (1953a, b) submitted corroborative evidence. Methyl Bromide: Higher and higher concentrations of this deadly gas have had to be used by the California Bureau of Plant Quarantine in order to insure a 100 per cent kill of the giant snails in the cargoes of ships coming in from infested areas (A. P. Messenger in Hit. Jan. 4, 1950). There still is much uncertainty as to the amount actually needed under a given set of conditions. Naphthalene: Uncertain and variable results were obtained by Massee (1928) when this chemical was used in an attempt to control slugs. Miles et al. (1931) found it neither lethal nor appreciably repellent to the slug M. sowerbii. In contrast, out of a number of chemicals tested Dustan (1927) recommended naphthalene flakes as the best general repellent. To a large extent, his recommendation was based on the fact that naphthalene flakes are resistant to the leaching action of rain. CHEMICAL CONTROL 81 Parathion: A 2 per cent parathion and 5 per cent metaldehyde contact dust was found to be very effective in controlling slugs (Lange and Sciaroni 1952). There are suggestions of a synergistic action in this combination. Karlin and Naegele (1958) branded parathion as "unreliable" and warned of its extreme toxicity. Unimpressive results were obtained in field tests on H. aspersa (Pappas and Carman 1955). Paris Green: Lovett and Black (1920), in their experiments with D. reticulatum, showed that Paris green had a definite repellent effect upon this slug and that it was eaten only under stress; but even though it was eaten sparingly, it was usually fatal. Apparently the recommendations of Hodson (1925) were incorporated in the slug bait formula of Douence (1929), viz. 1 lb. of Paris green, 20 lb. of bran, and sufficient water to make a "damp" bait. This was reported to be sufficient for an acre. The 1:30 formulation of Miles et al. (1931) gave uncertain results against M. sowerbii. Thomas (1944) tried a 1:28 formulation as a slug bait and produced "SOSO per cent kills." Cameron (1939) added to a 1:25 formulation 1/2 pt. of molasses, instead of "a little sugar" as recommended by Hodson; but he obtained less good results than he did with metaldehyde baits. Plaster of Paris, bran, and Paris green "biscuits" were made by Thomas (1948) but they did not prove effective in controlling slugs. Pangga (1949) found Paris green ''ineffective" against A. fulica though he did not indicate the nature of his experiments. Phenol: Along with a number of other suggestions for slug control in Australia, French (1906) recommended the use of "carbolized sawdust." Ewart (1910) claimed an improvement in this repellent by moistening a bucket of sawdust with a solution made up of 1-2 cups of "phenyle" in 10-20 large cups of water. Severe burning of foliage by "crude carbolic acid" was demonstrated in the experiments of Lovett and Black (1920) and it was therefore considered to be of little practical use in the control of slugs. A 1:500 aqueous solution was reported by Miles et al. (1931) to be effective in controlling M. sowerbii. The formulation of Ewart was suggested by Green (191 la) as a possible control measure for A. fulica. Potassium Permanganate: The experimental use of this chemical as a contact poison for slugs was claimed by Hall (1932) to be "excellent." Miles et al. (1931) found it not lethal to the slug M. sowerbii. Potassium Cyanide: This exceedingly poisonous chemical was found by Lovett and Black (1920) to darken various baits and render them unattractive to slugs and snails. 82 CHEMICAL CONTROL Pyrethrum: A mixture of this insecticide and bran (Thomas 1948) was shown to be attractive to slugs (probably largely if not entirely because of the bran), but no kills were produced. Pangga (1949) found pyrethrum toxic to young A. fulica but not to the adults. REPELLENTS Although a great many chemicals and compounds, including several listed here under separate headings, are repellent in nature, by far the majority produce only transitory irritant effect, or effects which last only while the snail or slug remains in contact with them. In cases where the irritant effect is so severe that death is produced, the substance is more correctly referred to as a contact poison (q.v.). From this, it is obvious that the difference between a repellent and a contact poison may be simply a matter of different concentration of one and the same chemical. And, further, repellents may be considered most effective when they are also contact poisons. Lovett and Black (190) conducted the first extensive experiments to find an efficacious slug repellent. Certain combinations of carbolic acid, sulphur, gasoline, lime, and sodium hydroxide were found to be temporarily repellent but the damage to foliage made them impractical. Combinations of hellebore (Veratrum), nicotine sulphate, sulphur, lime, and tobacco dust similarly had a temporary repellent effect, but it was lost or greatly reduced with rain. In their early experiments, nicotine sulphate solution (1:800 aqueous) and air slaked lime and tobacco dust (5:1 by weight) were recommended as "excellent" repellents. Later, Paris green, copper sulphate, and especially Bordeaux mixture (q.v.) were shown to be the best repellents. Unidentified slugs were exposed to eleven different substances by Dustan (1927) in an effort to find a good chemical repellent. He found that naphthalene flakes, because of their resistance to heavy rains, proved to be the best of the lot. In order of decreasing importance, he recommended four others, viz., sodium fluoride, magnesium fluosilicate, sodium chloride, and creolin (5 per cent aqueous). Tanglefoot, hydrated lime, fish oil emulsion, flowers of sulphur, tobacco dust, and calcium fluosilicate were insufficiently repellent to warrant further consideration. Miles et al. (1931) ran other extensive tests and found that aluminum sulphate, ammonium chloride with phenol, ammonium sulphate, copper sulphate, drained creosote salts, potassium permanganate, and sodium carbonate were definitely repellent to the slug M. sowerbii. Phenol in a 1:500 aqueous solu- CHEMICAL CONTROL 83 tion was particularly good. The best repellents were stated to be creosote in powdered chalk and 1:500-1:1000 aqueous solutions of mercuric chloride. Chlorcresylic acid in chalk, naphthalene, and nitrate of soda were shown to have little or no effect as repellents. Tryon (1899) recommended the use of tobacco waste as a repellent and contact poison in the control of vaginulid slugs. Ferrous sulphate is reported by Basinger (1927) to be repellent to T. pisana. A solution of rubber in crude oil effectively repels A. fulica in Malaya (Altson 1950a). Trenches sprayed with crude oil were completely effective in repelling the giant snail in Guam (Peterson 1957). In Ceylon this snail is reportedly repelled when the aromatic Khas Khas (Vetiveria zizanoides) is used. Several authors (e.g., Hall 1932) have recommended the use of sheets of zinc as repellent barriers. Others suggested using plain sawdust (e.g., "R.B." 1952). An early investigator (Miege 1906) recommended hog bristles! Soap solution: Soap solution has proved to be only moderately promising in controlling slugs (Lovett and Black 1920). Similar results with A. fulica have been reported by Pangga (1949). Sodium Chloride: Common table salt has long been a household method of killing slugs and snails. Although Anderson and Taylor (1926) used it against slugs in field tests without satisfactory results, Dustan (1927) and Vandenberg (1929) found it effective. Latif 1933a) used it as a repellent to protect potted orchid plants from the giant African snail. Hall (1932) recommended the use of an aqueous solution rather than the dry salt; but Pangga (1949) was able to produce kills with the solution in only the young specimens of A. fulica. In Guam, salt water sprays proved practical only in areas where there would be incurred no damage to the soil, for example, in beach zones and along roadways (Peterson 1957). Sodium Dinitro-orthocresylate: G. S. Dun (in lift. Nov. 25, 1949) quite accidentally discovered that this weed killer produced a contact killing effect on A. fulica in New Britain. This observation warrants investigation. Sodium Fluoride: The slug D. reticulatum was shown to do very little feeding on bait poisoned with this chemical (Lovett and Black 1920). Dustan (1927) and MacDougall (1931) were able to obtain good slug kills with it; but MacDougall did not recommend its use because of the burning effect it had upon plant foliage. Hodson (1924) had earlier come to the same conclusion. Sodium Fluosilicate: In the revised edition of the USDA Farmers 84 CHEMICAL CONTROL Bulletin No. 1895, it is recommended that metaldehyde baits be fortified with either this chemical or calcium arsenate. Soot: This has been recommended for controlling slugs and snails, but with the warning that unless carefully applied to plants, there may be damage (Anon. 1930). Earlier, Durham (1920) found that even when it was combined with lime, it was of little use in killing slugs. STOMACH POISONS, MISCELLANEOUS Baits containing diethylparanitrophenyl thiophosphate (i.e., Rhodiatox) and dinitro-o-cyclohexylphenol were found by Pereira and Gon^alves (1949) to be less effective than metaldehyde and therefore were not recommended. Levy (1938) patented a bait containing rotenone; but there is no evidence in support of its questionable effectiveness. Strychnine Sulphate: In spite of the fact that this was readily eaten by slugs, it did not have a killing effect (Lovett and Black 1920). Sulphocyanides: Although some of these were very lethal to the slug M. sowerbii, they were so deliquescent that they were considered useless in any large scale operation (Miles et al. 1931). Tartar Emetic (potassium antimonyl tartrate): Lewis and LaFollette (1942a) report high kills of H. aspersa with a spray of tartar emetic. Persing (1944a) states that such sprays "show exceptional promise" especially as an emergency measure to stop immediately damage to citrus trees and fruits. He recommends adding 2 lb. of tartar emetic and 4 lb. of white or brown sugar to 100 gal. of water and applying it with a boom gun or boom sprayer at the rate of 3-4 gal. per orange tree. A 6:12:100 ratio is used if the solution is applied with a spray-duster. Warfarin: In Guam, achatinas consumed with apparent complete impunity rat bait containing warfarin (Peterson 1957). Whitewash, Poisoned: The lack of natural limestone in Ceylon encourages the giant snails to remove the whitewash from the houses. This has suggested the use of bags of poisoned whitewash as a control measure (Connolly 1931). Because some parts of Java are equally poor in limestone deposits, Benthem Jutting (1934) has recommended the same type of control. Chemical Control--an Evaluation Any attempt to evaluate on a comparative basis the various chemicals and compounds listed above is frustrated from the outset because there has been reported in the literature too little of an exact nature upon which to base CHEMICAL CONTROL 85 scientifically a comparison of any great extent. Although a number of the experiments with molluscicides have been adequately controlled, the conditions under which they have been conducted have been so variable that their results are not reducible to any sort of common denominator. For example, the animals used in the various experiments have been different species of slugs and snails, with varying tolerances to the different molluscicides. This is complicated by the general belief that there is a fundamental marked difference between slugs and snails in their capacities to withstand toxic effects. Therefore merely because a certain formulation is shown to be effective in controlling slugs, it does not necessarily follow that it will be equally effective in controlling the giant African snail. On the other hand, many of the reports in the literature are inadequately supported by factual detail. A case in point is the work of Pangga (1949) on A. fulica. Without giving a single detail as to the nature of many of his experiments, he simply states that certain chemicals have little or no effect upon this giant snail. In other cases in the literature, there is very strong evidence of insufficient or no controls in the experimentation. Such, of course, are fruitful grounds for biased or specious conclusions. So little is known about the normal physiology of terrestrial gastropods that to a great extent, the toxicological effects of the various molluscicides are simply not understood. But in spite of these drawbacks, much of a positive natare can quite safely be concluded from the collected and collated data. First of all, it is more than apparent that a poison, even approaching the ideal for the giant African snail, has not yet been found. "Repellents" as such, by their very nature, are of the most limited value and they therefore can almost completely be dismissed at this point. They are of use only on a small scale; for example, to protect individual plants or small garden plots. They therefore are to be considered more in terms of a protective measure rather than of a control per se. In contrast, "poison baits'' have been put to considerable use in control programs and they still continue to provide much promise of help as new combinations are being formulated. The requirements of a good poison bait, however, are very great and unf o r t u n a t e l y in practice they are sometimes basically antagonistic to each other. Ideally, a poison bait should contain a powerful attractant and toxicant; it should consist largely of a readily available, inexpensive diluent; it should have a high degree of stability; and its application should provide the very minimum of difficulty. With the appearance of the comprehensive work of Kieckebusch (1953), there no longer can be justifiable doubt about the existence 86 CHEMICAL CONTROL of a very keen olfactory sense in snails. But finding a suitable attractant, which will not be neutralized or made impracticable by some local or environmental factor, is anything but a simple problem. For example, lime is generally considered a good snail attractant, but in areas where there is a natural great abundance of calcareous material, it is very much less effective. Lewis and LaFollette (1941, 1942a, b) showed that substituting fresh orange pulp for bran as the attractant-diluent in arsenical baits used to control H. aspersa in California citrus groves not only appreciably increased the effectiveness, but reduced the cost 75 per cent. An attractant of this sort however has the disadvantages of being obtainable in only a very restricted area and of spoiling quite rapidly under usual field conditions (Persing 1944a). Sugar similarly increases the attractiveness of baits but it hastens their spoilage (Thomas 1948). Bran itself quickly loses its attractiveness under conditions of high humidity. Metaldehyde is an attractant, but if it is used in excess, it may actually serve to "protect" the snail from the poison bait by interfering with the amount it can consume. The discovery of a positive anemotaxis in A. fulica (Chamberlin 1952a, b) suggests that the attractant qualities of a bait can be taken fullest advantage of if the bait is placed upwind to the greatest concentration of the snails. Obtaining a suitable toxicant presents equally difficult problems. Some of them chemically alter the other constituents of the bait. Others have been demonstrated to be highly toxic to snails, but they are of such a nature that they are seldom consumed and hence they act more like repellents than toxicants. To find a toxicant which will be effective under all conditions of weather is even more difficult. Quite naturally, optimum concentration of the chemical is of prime importance and it is therefore necessary to take into consideration the dilution factor of environmental moisture. Some toxicants produce the greatest total kills under warm, moist conditions; but such conditions favor rapid spoilage of the bait--especially through the development of mold. Other quite effective toxicants when excessively diluted, become mere irritants from which the snails may survive by throwing off a great deal of slime. This is especially the case during rains, which may actually wash the toxicant completely away in a few minutes' time. On the other hand, the supplemental killing effect of the sun, through desiccation, upon individuals stunned by the toxicant may be rather badly offset by the fact that in the meantime the bait has dried out and become ineffective. Even though some baits, such as those containing metaldehyde, can be rejuvenated a f t e r dehydration simply by remoistening, the mechanics of the proc- CHEMICAL CONTROL 87 ess on a large scale would become a considerable item both in time and cost. It is an obvious fact that tropical conditions in general are basically antagonistic to any measure of stability in snail baits. Providing some sort of shelter for the bait, no mater how simple, inevitably is impractical except on the smallest scale. Moreover, shelter for the bait often means shelter for the snail; thus damage to plants may actually be increased through the unwitting provision of an attractive shelter from the killing action of the sun, persisting long after the poison has lost its effectiveness. The use of cement in baits has provided a very welcome element of stability; but this has its shortcomings in making the bait less attractive and in some cases less toxic. In general, oil base baits hold up better under moist conditions. In the final analysis, the practicability of application of a specific bait or toxicant is the decisive factor. No matter how effective a snail poison is, if it requires anything more than a minimal amount of labor in its application, if it is expensive in the least, if it provides dangers to the lives of other animals, if it produces appreciable scorching of plant foliage, or if it is difficult to prepare, its use most definitely will be very limited. Even if each of these problems is adequately met, the actual effectiveness of a molluscicide is influenced to a considerable extent by such environmental factors as humidity, sunlight, contrast in the temperatures of night and day, rainfall, soil type, terrain, and plant cover. It becomes necessary then to determine optimum conditions under which snail poisons can be used. For example, it would be a senseless waste of time and materials to disseminate a poison, no matter how effective it is, at a period when environmental conditions have forced a majority of the snails into estivation. If under the same conditions, however, local irrigation brought the snails out of estivation, disseminating the poison would be clearly indicated. But the interplay of the various environmental factors can produce strange results. In certain regions in Ceylon, and in the Pallekelle District in particular, there was the strongest evidence that an intensive poisoning campaign (metaldehyde bait put out weekly for eight months) actually brought about an increase of the population of giant snails! Metaldehyde and calcium arsenate unquestionably are to date the chemicals of choice in attempts to control A. fulica. However, there is among investigators no general agreement as to which is the more useful or even whether there is any synergistic factor in their combination. Some (e.g., Lange and MacLeod 1941, Anon. 1949h, USBEPQ 1953) indicate that in controlling slugs and snails, metaldehyde-bran baits are more effective, especially under dry conditions, than cal- 88 CHEMICAL CONTROL cium arsenate-bran baits, and that the addition of calcium arsenate, sodium fluosilicate, or Paris green to a metaldehyde bait increases its effectiveness. Others (e.g., Lewis and LaFollette 1942a, b; Persing 1944a) insist that, at least as far as the control of H. aspersa is concerned, calcium arensate baits are better than the metaldehyde baits and that, except during hot weather, the greater cost of combining the two chemicals in a bait cannot be justified. Apropos of this general subject, it is unfortunate that Altson (1950b) did not extend his experimentations to include the addition of calcium arsenate to his "brick baits" of cement, lime, metaldehyde, and rice bran. Thomas (1944) convincingly reports on the superiority of meta-bran baits over Paris green-bran baits. In screening thirty-two possible molluscicides, including the promising Isolan, Pappas and Carman (1955) found none superior to the commercial pelleted bran bait containing 6.75 per cent calcium arsenate and 1.5 per cent metaldehyde. Peterson (1957) used this formulation in Guam and makes the following report regarding its action on the giant African snail: "Following ingestion of the bait, snails usually became paralyzed within 10 to 15 minutes and died within 30 minutes to 1 hour. It was found that migrating snails killed by the bait and left on the ground would be partially eaten by other snails the following night and additional snails would be killed." Recent extensive correspondence with investigators in the field has revealed the fact that a rice bran-metaldehyde bait is still considered in most areas infested with A. fulica to be the only practicable chemical control method. According to Pangga (1949), calcium arsenaterice bran bait was much less effective on this species than was the meta-bran bait. In the Hawaiian infestations, however, the metaldehyde baits were found less practical and in the long run more expensive than the calcium arsenate-slaked lime-cement mixture. Broadcasting straight metaldehyde in the form of pellets is a simple, expensive, and relatively much less effective method of controlling snails, and its use in this form has quite understandably been discontinued in many areas. Disseminating metaldehyde in sprays and dusts similarly is costly, but very much more effective; the unfortunately transient nature of the effectiveness of such applications, however, increases the labor and cost factors far beyond practicability in most cases. No matter what chemicals or baits are used, invariably there are disadvantages and even dangers involved which must be taken into consideration and which must condition their use. Of prime concern, of course, is the possibility of endangering the lives of humans, live- CHEMICAL CONTROL 89 stock, and poultry (cf. Douence 1929). This is particularly the case when the relatively effective sprays of sodium arsenite or corrosive sublimate are used. The indiscriminate use especially of arsenicals and metaldehyde introduces very real problems in this direction. And although the element of danger in the use of arsenicals is fairly well understood and appreciated, the same probably cannot be said for metaldehyde. Within months after the initiation of its use as a molluscicide, J. P. Hudson (1937) announced with obvious concern t h a t birds were seen to be removing dead slugs and snails from metaldehyde baits. A few weeks later, the fears he expressed were confirmed in the report of R. H. Hudson (1937) who stated that a blackbird and a starling, after feeding on metaldehyde bait, soon "fluttered about screaming in obvious agony, and . . . died about half-an-hour later. . . ." To obviate this danger, Carbett (1938) suggested the construction of special bait cages that would permit the entry of snails but would exclude fowl. In an attempt to determine the poisonous nature of metaldehyde to pets, Shewell-Cooper (1938) checked with the Royal Society for the Prevention of Cruelty to Animals and obtained a significant report, a part of which follows: ". . . in several instances dogs have died from eating meta, which was apparently put down to destroy slugs. . . . In one case we had an analysis made of a dog . . . and the result of the analysis clearly showed that the dog died from consuming meta." The following year, Jary indorsed the stand of exercising great caution in using metaldehyde and introduced the suggestion that it might be harmful to earthworms--a point which seems to have been ignored since then. That same year, Lewis et al., echoing the earlier warnings of Gimingham and Newton (1937), presented the medical history of a thoroughly convincing case of death in a child from metaldehyde poisoning. According to their report, accidental death of other children and attempted suicides of adults are on record. They warn, without presentation of evidence, that ". . . the method of mixing it with bran has been responsible for the innocent slaughter of many of our wild birds." In direct contrast to these reports, Cameron (1939) fed two chickens for four weeks on a total of two pounds of a 1:50 meta-bran mixture along with their regular feed. Even though it is believed t h a t this amount is in excess of what they would be picking up under normal field conditions, the chickens remained unaffected. Fernando (1952) similarly minimized the dangers of metaldehyde by stating: ". . . dogs, cats and poultry will sometimes taste the bait and reject it without ill effects." Among many other toxicants used against invertebrates, Tilemans and Dormal (1952) list metaldehyde 90 CHEMICAL CONTROL as a nerve poison with an LD50 of 170-200 mg/kg for man and 250 m g / k g for dogs. For intoxication, the symptoms of which they describe, they recommend vomitives and an antispasmodic of chloral chlorhydrate. When either metaldehyde sprays or dusts are used, there apparently is no phyticidal effect even on such delicate plants as orchids (Jefferson 1952, Moreton 1953). If bait containing toxic agents is broadcast instead of being placed in heaps, any possible hazard is considerably reduced--but unfortunately so is the efficacy of the bait. This leaves still unaltered the problem of soil contamination; for example, the prolonged use of arsenicals eventually interferes with normal agricultural practices in that the accumulation in the soil builds up beyond the tolerance point of some arsenic-sensitive crop plants (Pierce 1931). In areas subjected to periods of drought, there is danger of a surface concentration of molluscicides through the siphon action of deeper soil moisture being drawn up by capillarity to replace evaporated surface moisture. If copper exists as a contaminant in the arsenical, subsequent insecticidal fumigation of the plants with cyanide may produce severe injury to the foliage (Gammon 1943). In Hawaii, it was not the poisonous nature of the calcium arsenate-cement--lime mixture to which some of the complaints were directed; it was the fact that stones, painted with this mixture, were broadcast in grassy areas where they caused damage to lawnmowers! Unfortunately, the promising tartar emetic sprays henceforth will be used in the United States only on a very limited basis because of the discontinuance of the large scale commercial production of potassium antimonyl tartrate. Copper sulphate has seemed to be inordinately popular in the recommendations of many who have been faced with the control of the giant African snail. To a fair extent, this is unfortunate. The frequent rains in the tropics quickly wash away this chemical and require the addition of more. Its continued use not only builds up an abnormally high concentration of the copper ion, but it very appreciably increases the acidity of the soil. In the strongly basic soils of the coralline islands, this change in pH might be an advantage; on the other hand, it might be quite deleterious to crops which are sensitive to acid soils. Miles et al. (1931) have suggested the concurrent addition of an equal quantity of lime to offset this acidifying effect. This brings up the point that the molluscicidal use of lime will conversely alkalinize the soil. This directly favors the survival of the snails but not certain crops, such as tea, which demand an acid soil. Plants which have been injured by the feeding of snails absorb CuSO4 CHEMICAL CONTROL 91 rapidly at the wound site and die. But even healthy plants may be burned severely by this chemical if it is applied in too high a concentration. Pangga's experiments (1949) suggest that there has been an overemphasis of the use of CuSO4 against A. fulica. The value of ashes has similarly been overrated. For one thing, except under the most unusual conditions, ashes in any quantity will not be found in the tropics. When it becomes soaked with rain, it is useless and will need replacing. Besides, the washing of ashes into the soil will alkalinize it with the resultant disadvantages indicated above. Of the poison gases used in fumigation of cargoes infested with the giant snail, ethylene dibromide seems to hold much greater promise than methyl bromide, hydrocyanic acid, or carbon disulphide. Accurate determination of adequate concentration of any of these gases to insure a 100 per cent kill has yet to be made. Some of the newer insecticides may be found, after much more extensive experimentation, to possess sufficiently great molluscicidal properties to be of use in giant snail control programs; but the results so far have been neither spectacular nor promising. For example, the work of Pappas and Carman (1955) appears to have quite convincingly eliminated from further serious consideration, except for some possible synergistic effect, such common insecticides as aldrin, chlordane, DDT, dieldrin, endrin, isodrin, malathion, and parathion. W. H. Lange, of the University of California at Davis, however, believes that at elevated temperatures, some of these insecticides become effective. He is currently conducting tests which promise to bring to light new organic toxicants of unprecedented molluscicidal properties. CHAPTER 6 CONTROL THROUGH MECHANICAL DEVICES Various mechanical methods have been devised to control Achatina fulica, and other pestiferous snails, especially in the absence of suitable chemicals. Some of these provide under a great variety of conditions the only, and sometimes the best, means whereby snail pests can be held in check. At the very least, however, they serve as valuable adjuncts to other types of control. Barriers Fences made of closely placed bamboo sticks are used in many places to create a mechanical barrier against the giant snails. But even where the fence is six to eight feet high, the barrier effect is only temporary; for the larger snails scale the fence and the smaller ones manage to work their way through the gaps. Nonetheless, erecting a fence in conjunction with hand picking and destroying, is probably the most common method of control that is used, other than just hand picking alone. G. S. Dun (in litt. Dec. 17, 1953) found that painting a board fence with crude creosote increased considerably its effectiveness as a barrier. Effective barrier fences have been constructed of other materials. Large experimental vegetable plots at the Agricultural Development Station in Ponape are surrounded by a very high fence of closely placed strips of wood. Torres (1950) similarly suggests erecting a wooden fence to protect coffee plants from the attacks of the giant South American snail, Strophocheilus oblongus. The people in Chi92 CONTROL THROUGH MECHANICAL DEVICES 93 chi Jima strapped together large sheets of corrugated iron around their garden plots to keep out the giant snail. The fact that these become excessively hot in the bright sunlight should not suggest added protection, as the heat quickly dissipates before the crepuscular and nocturnal snails come in contact with them. Actually, the corrugated iron sheets can be more easily and more quickly scaled than wooden or bamboo fences. Because of vast surplus stores of these sheets in Guam, it was recommended by high government officials in 1949 that they be used for snail barriers. Sheet zinc is recommended by Hall (1932) as being an absolute deterrent to snails and slugs; but of course their use as a mechanical barrier is quite out of the question from the standpoint of cost in anything but the most small-scale control measures. As early as 1933, Corbett suggested without elaboration that flower beds could be protected from the giant African snail by surrounding them with an 18 inch high, 1/4 inch mesh wire netting. Bryan (1949) repeats the suggestion. Independently of these recommendations, Peter J. R. Hill of the War Memorial Laboratory in Koror (Palau Islands) erected around his 65 by 45 foot vegetable garden a 6-8 inch high, 18-mesh copper wire screen barrier. The screen was buried in the ground at a 45° angle so that the top of the screen leaned away from the garden. The horizontal wires were removed from the upper inch of the screen, the last one of which was removed at an oblique angle so that the vertical wires were alternatingly vertical and almost horizontal. This produced a double row of inch-long sharp projections along the top of the wire fence. This type of barrier is very effective because of the following advantages: the screen helps to break the suction of the foot and the snail's own weight tends to pull it off; the angle of the fence increases the effectiveness of the pull on the foot and causes the snail to fall away from the fence; the sharp vertical wires force the snail to loop way out in an attempt to circumvent it, leaving such a small portion of the foot in contact with the screen that purchase is invariably lost; and in the areas where a double barrier of this sort was set up, the snails which did manage to get over the first fence were imprisoned between two fences and died because of exposure to the hot sun the following day. But these advantages were found to be offset by disadvantages. With the fence only 6--8 inches high, the snails were found to pile up along the fence in certain areas to such an extent that others could crawl upon them and drop over the other side. Even though the fence is relatively inexpensive, it still is considerably beyond the means of most native people. And since the necessity of removing the snails, as they accu- 94 CONTROL THROUGH MECHANICAL DEVICES mulate along the fence, is still not obviated, the native people find it more economical not to erect the fence and simply resort to frequent hand collecting. A possible improvement of this fence might be effected if it were constructed along this same design but built to stand one foot high. The upper four inches of this could then be bent away from the garden plot at a 90° angle. This would force the snail to crawl upside down for a way even before it got to the barrier of the double row of vertical wires. The drop-off should be 100 per cent. But, again, this would significantly increase the cost. Kondo (1952) announced that his description and figures of the "Peter Hill snail fence" were mimeographed in 1949 by naval personnel and distributed in the Trust Territory. As an apparent outcome of this, Manuel Mendiola of Rota used the fence to protect his extensive bell pepper plantings. Kondo gives a vivid and detailed account of the effectiveness of this approximately 1,400 foot fence, reemphasizing the advantages and the inherent difficulties in the use of this type of barrier (see Fig. 11). It has been suggested by Lange (1947) that brush taken from clearings be piled up around the periphery as a mechanical barrier to the giant snail. Such accumulations of decaying plant material are particularly attractive to these snails because of the almost ideal conditions of food, moisture, and protection from the sun. It would seem then that such a practice might invite the build-up of dangerously high peripheral populations. On the other hand, this method coupled with other control measures, such as the use of molluscicides, might be very effective. Rattan or coir fiber tied to the trunks of trees susceptible to the attacks of the giant snails has proved effective (Feij 1940, Bertrand 1941). Even coir fiber rope was considered by South (1926b) to be of value in setting up a mechanical barrier around gardens. But experimental plantings of thorny mimosa plants did not produce any barrier effect at all (Feij 1940). Trenches filled with sawdust have also been recommended (South 1924b). In the Pallekelle estate in central Ceylon, freshly planted cacao seeds and young seedlings were protected from the ravages of the giant snail by specially constructed wire cages; but this practice soon became economically prohibitive. Technically, repellents are chemical barriers and should be considered here; however, they have already been treated above in the discussion of chemical control. Burning-Over This practice has been resorted to by native peoples in many parts of the world primarily because it is the easiest and quickest means of clearing land before planting. It is very effec- CONTROL THROUGH MECHANICAL DEVICES 95 tive in killing a high percentage of snails present in spite of the fact that this benefit is only incidental. Burning-over is most often done during the dry season and at this time a great share of A. fulica are in estivation. Estivation often takes place right on the surface of the ground under vegetation or other superficial types of protection. To a great degree, snails under such conditions will be killed in the burning-over process. Other specimens, however, will go down into the ground as much as four or five inches before estivating. These will survive and re-establish the population with the return of normal conditions. Other species of the giant snail were shown by Lang (1919) to survive in the same manner the annual grass fires in the former Belgian Congo. No matter how thoroughly it is done, burning-over is not eradicative by any means. But besides this, snails from contiguous high-pressure population areas will quickly invade the burned-over area and speed up repopulation just that much more. Tsetkov (1940) emphasizes this point. Early unsuccessful attempts to "burn out" A. fulica in Agafia, Guam, gave further convincing proof. The systematic use of the flame thrower, after clearing the land of vegetation, proved much more effective than simple burning-over in controlling T. pisana in California (Basinger 1923a, Gammon 1943). But even this method, in efforts to control H. aperta in southern California, was disappointing because this species has a greater propensity for estivating at comfortable distances below the ground. In contrast, but with similar results, Henderson (1936) found Polygyra uvulifera survived grass fires where other species did not, simply because it habitually retreated to large, thick protective leaves. Joubert and Walters (1951) found that burning-over fields heavily infested with T. pisana was very effective in South Africa. Clean Culture As early as 1926, it was suggested by Pereira that the responsibility for a renewed outbreak of A. fulica in Ceylon rested in the fact that the people had failed to destroy weeds and to clean up accumulated refuse. Leefmans (1933c), Corbett (1933), Feij (1940), Lange (1947), Anon. (1947a), van Weel (1949), Rees (1951), and Peterson (1957) have further emphasized the importance of reducing the numbers of the giant African snail through clearing and weeding. There is no question about it; rubbish, debris, wild plant growth, and the like do afford a sanctuary for these snails. A premium therefore is put on their removal. Lovett and Black (1920) produced "remarkable results" in reducing slug and snail damage through the simple expedient of cleaning away all possible refuge in areas adjacent to crops. Although keeping the ground free from plant debris may be conducive to good snail control, it is contrary to the long 96 CONTROL THROUGH MECHANICAL DEVICES established agricultural practice of allowing the trimmings especially of green manure plants (e.g., Gliricidia and Erythrina) and the leaves of crop plants (e.g., cacao) to accumulate as a natural mulch. Through this practice, weed growth is kept down, moisture is retained, nutrients are added to the soil, and, unfortunately, insect and snail pests are increased. The nocturnal and crepuscular habits of A. fulica normally permit it to escape the killing effect of the sun. It has been observed over and over again in the field that snails which were unable to find protection from the sun, after a night's foraging, were quickly immobilized and soon killed. In fact, in many places, by far the majority of deaths in the snails can be attributed to overexposure to the sun. The drying effect and the excessive heating of both the snail and the ground are probably the causes of death; but the work of Carmichael (1928, 1931) suggests that it may be the ultraviolet rays which produce the lethal effect. At any rate, advantage can be taken of this knowledge by removing as much as practicable of anything which affords protection from the sun. This was tried on a large scale on an experimental farm in Tinian by Nelson Young. A large field was plowed over, leaving exposed to the sun both snails and their eggs. This essentially neutralized area was then planted and by harvest time, the snails in the adjacent uncultivated areas had been able to move only a short way into the big field. Damage was limited to the peripheral areas. Of course, similar peripheral damage in a smaller plot would exact a much higher percentage of loss. In fact, the usual, very small native garden plot is so small that the snails can often reach its center in a few hours' time during a single night. This does emphasize, however, the importance of clearing around the plot as large a buffer area as practicable. The daily removal of snails from this buffer area, especially in the early morning, would greatly increase its effectiveness. Clearing through the use of weed killers has been tried in only a few instances (e.g., in Guam in 1949), and it is possible that with the introduction of more effective weed killers we will see them more frequently used for this purpose. But as far as large-scale farming is concerned, little help of a practical nature can be expected. The necessity for mechanized farm equipment, for one thing, would put it completely out of the question for the majority of native peoples. Whether the clearing is done on a large or a small scale, it introduces still another difficulty, viz., soil divested of its normal plant cover not only quickly loses its nutrients through leaching, but it is subject to rapid erosion. In the coralline islands especially, where the soil is often exceedingly thin, this would be a CONTROL THROUGH MECHANICAL DEVICES 97 real tragedy. Where the cover of vegetation is thick, as in the case of grass, not removing it entirely, but cutting it down to a stubble is a compromise measure put to effective use at the Agricultural Experiment Station in Peradeniya, Ceylon. Drowning Anyone who has attempted to drown a snail or slug in order to preserve it in an extended condition, knows full well how difficult it is to produce a kill by this method (cf. Fromming 1929, Courtois and Duval 1927). Even when the dissolved oxygen had been driven off by boiling the water, Rees (1951) was able to demonstrate that it took 48 hours for young A. fulica hamillei to die after complete submersion. Earlier, van Weel (1948) stated that it took 20-24 hours for the giant snail to drown (cf. Meer Mohr 1949b:5); but the water was probably maintained at a higher temperature. He introduced evidence to show how this capacity has permitted the giant snail to survive apparent drownings and spread more rapidly to adjacent uninfested areas (c.f. van der Goot 1939). He stated that snails dumped into the Mahakam River at Tenggarong, East Borneo, were apparently able to withstand a long trip in the water; for just a year later a healthy infestation was found in Djambajan, over eleven miles downstream. Scarcely one-tenth this distance was covered upstream via other means of transportation. Meer Mohr (1950) observed this snail apparently "deliberately" entering pools of water during the rain and drowning; on the other hand, Davis (1958a) observed a small specimen successfully negotiate a trip across standing water. Submerging snails in sea water has not seemed to produce appreciably better results. One of the classical experiments was that of M. Aucapitaine of Corsica in which he plunged specimens of several species of land pulmonates in the ocean for fourteen hours (Caziot 1928). He obtained 87 per cent survival! Dartevelle's observations (1952b) were less dramatic but just as significant. A. fulica has similarly demonstrated its capacity to withstand such treatment, Somanader's remarks (1951) to the contrary notwithstanding. In 1923, Jarrett reported that specimens of this species were being destroyed in Malaya by dumping them in the ocean. Later (1931), however, he indicated that such specimens were found to wash ashore in a viable condition. Of course, a number of factors, such as temperature of the water, salinity, oxygen tension, and onshore wind, would undoubtedly seriously affect survival rates; but even when such factors are far from favorable, there is great enough survival to warrant the use of caution when resorting to this method. Crushing the snails before dumping them in the water would seem to eliminate all 98 CONTROL THROUGH MECHANICAL DEVICES danger of this sort. Under any circumstances, the possibility of fouling the water must be taken into consideration. Floods provide only a temporary setback in the snail population as witnessed by South (1925, 1927) in northern Malaya. A tidal wave in 1946 in Chichi Jima similarly caused a delay of less than two years in the expanding snail population on that island (Mead 1950b). Hand Collecting and Destroying Without question, no single control method has gained a broader general acceptance than the simple expedient of collecting the snails and eggs by hand and destroying them. An examination of the literature shows over and over again that this, almost without exception, is the method resorted to in the very earliest phases of an invasion by A. fulica. Frequently, in fact, participation is stimulated by offering bounties for both the snails and eggs. Invariably, the bounties are discontinued after a relatively short time because the rapid increase in the snail population, in spite of the collections, soon exhausts even the most generous funds that have been set up for this purpose. Such has been the experience of many, many areas, including quite recently Hawaii (Wong 1951). For the same reason, people despair of the seemingly endless and ineffective collecting, discontinue their efforts, and let the snails ''take over" (vide, e.g., Bertrand 1928). One thing is certain, no matter how assiduously this collecting is done, it is never eradicative; that is, if countless unsuccessful attempts in the past are any criterion. Nor can it ever be eradicative except on the very smallest scale, despite some of the optimistic statements in the literature to the contrary (e.g., Peterson 1957). But this does not j u s t i f y dismissing summarily this method as an impractical one. Actually, when the giant snail population is vigorously building up, as in the early stages of invasion, this method is least effective. On the other hand, after the snail population has reached stability or a "climax," collecting and destroying the snails, especially on an extensive and intensive scale, has been shown in a number of recent instances to reduce the population to a point where it never again completely recovers. Factors of population senility and "decline" doubtless enter into the picture at this phase. These points are discussed below. But quite aside from whether or not the population is increasing or decreasing, hand collecting and destroying on the largest practicable scale is still the only inexpensive, effective method to which the poor native peoples can turn. For this very reason, it is currently being used on a far greater scale than any other method, especially in and around flower and vegetable gardens--where the greatest CONTROL THROUGH MECHANICAL DEVICES 99 damage is done. Because the eggs and young are easily overlooked and because of the tremendous reproductive potential of these snails, the task of collecting resolves itself into an indefinite, daily harvesting. This is usually done in the early morning when the snails are still quite active, although occasionally it is done at night with the aid of a light. Rough terrain or dense vegetation makes this method impossible or impracticable (Corbett 1937). Unfortunately such conditions are common in areas where A. fulica abounds. It is of interest and perhaps significance at this point to note the effect of hand collecting in the control of other terrestrial gastropods. Both Basinger (1923a, 1927) and Gammon (1943) strongly indorse this method; in fact, Gammon goes so far as to say that it "in many instances has constituted the difference between success and failure" in the control of T. pisana in California. Lewis and LaFollette (1941), on the other hand, report that it is of some value in controlling H. aspersa in California citrus groves but, in general, is "not satisfactory." Barnes (1949b) removed between 10,000 and 17,000 slugs per year for four successive years in a garden in Harpenden and "in spite of removing these numbers no reduction in population was observable." The real problem in this method, however, is not the mechanics of collecting the snails but what to do with them after they are collected. At first thought, this problem would not seem to be a big one; but one reconsiders when one learns that large-scale collections will bring in tons of snails almost indefinitely. What then? South (1926b) was the first to make a number of suggestions in an attempt to answer this question. These and other suggestions should be considered here for what they are worth. Usually, the snails are dumped into pits, crushed, and covered with soil. In Java during the early phases of the invasion by the giant African snail, 2,105,743 snails and 491,350 eggs were collected in a few weeks' time and disposed of in this manner (Leefmans 1933c). In some cases, they had not even been crushed; but a two-foot layer of superincumbent, well-packed soil insured a complete kill. Plunging the snails into boiling water; dusting them with CaO; letting them soak in solutions of CuSO4 (4 per cent) or NaCl before burying them--all have proved successful. Such methods, however, add needless expense and man-hours of labor. Further, the addition of these chemicals to the soil in any great quantity can produce unfortunate changes in the soil and therefore the plant cover it can support. And, as pointed out above, even the addition of CaCO3 in the form of snail shells will change the soil pH in a basic direction. In strongly 100 CONTROL THROUGH MECHANICAL DEVICES acid soils, this might have an advantage; but in cases where the crops, such as tea, demand an acid soil, a change of this sort could reduce crop yields. Such actually has been the case, for example, in some parts of Ceylon. Dumping the live snails into the ocean or rivers has already been shown to be inadvisable. Crushing them before dumping them in the ocean would seem in many respects to be the best way to dispose of them. Unfortunately, collections of snails made even a reasonably short way inland manifestly could not practicably be disposed of by this means, to say nothing about the danger of starting new foci of the infestation along the route. Burning the snails in incinerators or on piled brush has been frequently resorted to and, although this method has merits, the offensive smell and its effect upon health have to be taken into consideration. Obviously, the best method of "disposing" of the collected snails involves making use of them for the benefit of man. This whole subject is discussed in detail in chapter eight. Shifting Cultivation In many areas, for example in the colony of North Borneo, the natives practice shifting cultivation to insure continued high yields in their crops. They clear off an area and put in their crops. The giant snails gradually invade the preferred cultivated plots from adjacent second growth areas. Hand collecting by the natives simply and effectively impedes this invasion. By the time the snail population has built up sufficiently to cause concern, the few harvests and the tropical rains will have removed the greater share of the soil nutrients, and it will be time to move on anyway to another area which has not recently been cultivated and which naturally supports only a small snail population--or perhaps none at all. In contrast, some native people will travel long distances to plant crops in achatina-free areas. A case in point was found in Rota in 1949. According to Frank L. Brown of that island, the inhabitants travel almost ten miles on foot to the opposite side of the island to plant their watermelons, of which the giant snails are inordinately fond. In that achatina-free area, examined by Kondo and the author, the watermelons grow to full maturity with practically no care. Such was almost impossible in the snail-infested areas. Traps So far as can be determined, no elaborate trap has been devised to control A. fulica. In fact, trapping as such has been quite incidental to other control measures even in intensive programs. Typical of trapping methods in general, the little trapping that has been done has been designed to take advantage of a partic- CONTROL THROUGH MECHANICAL DEVICES 101 u l a r proclivity or appetite. The snail's manifest fondness for decaying vegetation has encouraged some people to pile plant debris peripherally around an area to be protected and then periodically either collect and destroy the snails that this debris has attracted or b u r n the debris to kill the snails. Fortifying the debris with a molluscicide is an obvious alternative and although this would increase the cost, it would reduce the labor of maintenance. In fact, to neglect u n f o r t i f i e d peripheral piles of plant debris "traps" would seem only to invite greater snail damage. Appealing to the snail's propensity for crushed individuals of its own kind has also been used in setting up traps. A group of snails is piled in a protected place and crushed, and when these have attracted other snails, they in turn are crushed only to increase the a m o u n t of "bait." A system of periodic visits to a series of such traps to kill any live snails in the vicinity has been reported as being very effective. An ever increasing accumulation of rapidly putrifying flesh and the flies it would attract are only two of several obvious drawbacks to a control measure of this sort. Any device which would cause the snails in quantity to become stranded in the hot sun would be effective, as, for example, setting up temporary shelters in clearings and then removing them in midmorning. Again, however, the problem of the rapid accumulation of dead snails introduces great difficulties. An interesting though unintentional modification of this method was observed on Agiguan Island. Giant snails had sought refuge under large pieces of corrugated iron, only to be cooked to a turn the following day as the sun's rays heated the iron like a stove. Chamberlin's work in Tinian (1952a, b) strongly suggests that the positive anemotaxis in A. fulica should be taken advantage of in setting out baits and traps by placing them upwind of the snail concentrations. CHAPTER 7 BIOLOGICAL CONTROL Since before the turn of the century, the study of the predators and parasites of mollusks has been a matter of at least academic interest. As a consequence, some excellent, detailed, and comprehensive papers have appeared in the literature, including those of Bequaert (1926), Fromming (19546:311-14), Gain (1896), Keilin (1919, 1921), Pelseneer (1928, 1935), Pilsbry and Bequaert (1927: 470-79), Plate (1951), Simroth and Hoffmann (1928), Taylor (18941900), and Wild and Lawson (1937). Now that there has arisen a genuine need for a suitable predator or parasite of the giant African snail, however, this study has taken on a very practical aspect. Even more than that, it actually demands an inventory and reassessment of the information on record. With this thought in mind, there has been prepared below a series of discussions, under the subheadings of the main predator or parasitic types, which will bring the subject of biological control more sharply into focus. An examination of this, in turn, will permit of a more accurate evaluation of the biological means of control--especially as it compares with other types of control. Amphibians Noel (1891) effectively demonstrated the fact that toads and frogs introduced into gardens would have a telling effect upon the slug and snail population. It was knowledge of this sort that persuaded the American authorities in 1937 to introduce the giant Central American toad, Bufo marinus, into Guam from Hawaii in an attempt to control the destructive, large, black slug, Veronicella leydigi. This biological control measure was very effective and, as a consequence, the Japanese were persuaded subsequent102 BIOLOGICAL CONTROL 103 ly to introduce the toad into Rota, Tinian, Saipan, and Yap (Townes 1946). In 1938, it was introduced into Mauritius from Puerto Rico in the hopes that it would control the sugar cane pest, Phytalus smithi Arrow (Anon. 1939). More recently, this toad has been purposely and in some cases perhaps accidentally introduced in a number of the Micronesian islands. In Ceylon, Bertrand (1928) reported t h a t the only control of Achatina fulica in Madagascar was a "big bull-frog." In spite of his recommendation, the Director of Agriculture in Ceylon decided against introducing this giant amphibian predator. Pemberton (1938) announced that Bertrand was apparently in error, that it was Mauritius and not Madagascar, and that the amphibian in question was not a "bull-frog" but probably a ''large toad common in Mauritius." Pemberton stated further, "It is probable that Bufo marinus would feed on the young snails, should the pest ever become established in regions where the toad occurs." This prediction was shown by the author in 1949 to be correct--at least on the island of Ponape where the stomach contents of this giant toad were examined. Not only were small specimens of A. fulica and Opeas sp. found, but pieces of the flesh and shell of large specimens of the giant snail were even more commonly encountered (cf. Lever 1939). The presence of dead fly maggots with the latter quickly told the story. The toads were apparently attracted to the crushed larger snail specimens on the roads because of the activity of the maggots. The abundance of the giant snails in all sizes, especially in the Jokaj region of Ponape, and the markedly emaciated condition of the numerous giant toads strongly suggested that B. marinus was providing little in the way of biological control. Such were also the conclusions of Lange (1950) on Saipan. More recently, however, Gressitt (1954:90, ICCP 1953) reports that it is believed by some that B. marinus is responsible for the decreasing snail population in Ponape. This is questionable, notwithstanding the fact that Jaski (1953) found "with an astonishing frequency" achatinas in the stomachs of a species of toad in Java. Dartevelle (1954) has reported that certain amphibia in the former Belgian Congo seek out the eggs of the endemic achatinas. Frogs would be even less effective than the toads, as they are less independent of the aquatic environment. Garnadi (1951), however, in Indonesia found the frog Rana tigrina Daud. in several instances in the vicinity of egg masses of A. fulica; one dissected specimen revealed pieces of achatina egg shell in the stomach. Earlier information from Pilsbry and Bequaert (1927:473) and from Kirkland (1904:8), wherein only 1 per cent of the stomach contents of the 104 BIOLOGICAL CONTROL American toad was shown to consist of snails, suggests that amphibians figure in only a minor way in the consumption of snails. To sum up the effects of introducing B. marinus, however, simply by stating that it does effectively control V. leydigi but not A. fulica, is to ignore the indirect effects, some of which are of considerable importance but infinitely subtle in their connection with the original introduction. As an example on the plus side of the ledger: The great quantity of dead achatinas in a heavily infested area causes the cockroach population to build up often to serious proportions. B. marinus will thrive on cockroaches (Pemberton 1949); both are nocturnal. It should be recalled at this point that the toad will also feed on fly maggots. In the absence of experimental evidence, it is q u i t e safe to reason a priori that, even though the introduction of B. marinus will not produce an appreciable effect upon the giant snail population, it may reduce substantially some of its side effects of public health importance. On the other side of the ledger, however, we find an indirect effect which is felt through a wondrously devious chain reaction. The early traders accidentally introduced the pestiferous, disease-carrying rats. In the zoos in Japan, the giant monitor lizard (Varanus) fed avidly on rats. The monitor lizard was therefore introduced, as a biological control measure, into the Micronesian islands where it had not earlier become established. Then came an astounding discovery. The monitor lizard is diurnal and the rat is nocturnal. To make matters worse, the biological faux pas proved irreversible. The monitor lizard quickly became a pest by consuming eggs and young chickens--precious items indeed in Micronesia, especially since chickens, too, will do their bit in consuming young achatinas. The introduction of B. marinus, strangely enough, brought some relief to this problem. The giant toads do not hide as effectively as rats during the day and the monitor lizard would easily find them in its search for food. The toads, however, have potent poison glands in their skin and the meal would prove fatal (cf. Gressitt 1954:142). The prey, paradoxically, was controlling the predator! This was an unexpected benefit which seemed to give further support to the advisability of introducing the toad; in fact, after this was discovered, the toad was purposely introduced into other islands to control the lizard biologically. The benefit, however, was not without serious side effects; for the monitor lizards in the meantime had developed a healthy appetite for the number one agricultural enemy of the Pacific, the grubs of the rhinoceros beetle, which kill the coconut palm--a far more important agricultural item than chickens (Gressitt 1952). BIOLOGICAL CONTROL 105 Also, the lizards have been demonstrated to feed on another enemy of the coconut palm--the coconut crab (Birgus latro) which incidentally, in turn, feeds on the giant snail, as does the lizard itself (vide infra). But the complications did not stop even there. When the barnyard pigs caught the toads and ate them, they became sick or died (Townes 1946). The dogs and especially the cats, as they are largely nocturnal, also discovered the giant toads and similarly were killed when they bit them or attempted to eat them. This was a tragedy, for the dogs and cats were the best rat catchers in the islands! Who could have guessed that introducing B. marinus would, in addition to reducing the black slugs, aggravate the rat problem, kill the monitor lizard, reduce natural control of coconut pests and the giant snail, bring some relief to the poultry industry, kill pigs and house pets, and ameliorate a public health problem of cockroaches and flies brought on by the introduction of giant African snails? As a final ironic twist, the native peoples are convinced that their dogs and cats have died from eating the "poisonous" giant African snails! Ants Green (1910c, 1911b) was the first to suggest that ants might be a factor in the control of A. fulica by reporting that he had observed a predaceous ant "Phidelogeton affinis [sic: Pheidologeton] swarming in a batch of eggs that was just hatching." As it is stated, there is only circumstantial evidence of any attack per se on the snails. This evidence nonetheless has been accepted as fact and even embellished by subsequent investigators (Hutson 1920, South 1926b, van Weel 1949). This assumption, however, has proved to be a safe one, for in Indonesia this same species of ant has very recently been shown by Butot (1952b) to carry to its nest other species of snail (Opeas gracile, Gulella bicolor^ and Geostilbia moellendorffi). Other species of Pheidologeton have also been shown to indulge in snail robbing. Meer Mohr (1931b,c) reported P. diversus was similarly observed carrying away specimens of Bulimulus sp. (cf. Rothney 1889). In a letter to the author (Dec. 16, 1949), Professor Silvario M. Cendana of the University of the Philippines wrote that a species of ground ant common in Los Bafios helped check the rapid increase of the giant African snail in that area. Apparently the same species of ant is referred to by Pangga (1949) as Solenopsis geminata Fabr.; he reports that they sometimes attack newly hatched A. fulica. It is of interest to note that Martinson (1929) reports that in Ghana the worst of the natural enemies of the local achatinids are the "red and black driving ants." A number of times during the survey of Micronesia conducted by the author and Kondo ants were seen to be swarming over dead 106 BIOLOGICAL CONTROL or dying achatinas. In every case, it seemed quite clear that the snail was either dead before it was attacked by the ants or was dying from overexposure to the hot sun. The same species of ant would completely ignore healthy snails in protected areas. At no time were any ants seen to show the slightest interest in the young snails or the thick-shelled eggs. Occasionally, eggs would be laid in an empty shell and ants were seen to crawl over these to get into the upper whorls of the shell where there still was some decaying flesh. An incautious observer might have been led to believe that the ants were attacking the eggs. It is probable that Lawson (1920b) was similarly misled in his observations and that Bequaert's skepticism was justified (1926). From the evidence accumulated so far, it is questionable that ant raids on young A. fulica are producing an appreciable control except under the most limited conditions. Beetles Several beetle families are notorious for having species that live largely or entirely upon pulmonate gastropods. Bequaert (1925, 1926) has summed up admirably the widely scattered literature on this subject. Significant contributions that have appeared since Bequaert's works, other than those directly pertaining to predation on A. fulica, are the following: Cros (1926), Drilus mauritanicus consuming Rumina decollata; Moore (1934), Carabus violaceus killing Arion hortensis; Tomlin (1935), C. violaceus carrying off Agriolimax agrestis (Deroceras agreste), Milax gagates, and small H. aspersa; Ingram (1950), Calosoma sp. feeding on Triodopsis albolabris and Ventridens intertextus; Metteo (1946), Ablattaria laevigata consuming Eobania vermiculata and Helicella variabilis; Fincher (1947), Lampyris noctiluca attacking Arion ater; Schwetz (1950), Luciola sp. attacking Planorbis tanganyicensis. Clausen (1940) reports that Lampyris noctiluca has been imported into New Zealand from England for the control of H. aspersa. U n t i l recently, at least, the beetle most famous for its predation on A. fulica is the so-called India glowworm, Lamprophorus tenebrosus (Walker) (Lampyridae), endemic in Malaya, Ceylon, and India. Paiva (1919) was the first to draw attention to this nocturnal predator by giving considerable significant information about its life history. Additional studies of this type were reported upon by Hutson and Austin (1924), Austin (1924), Jacobson (1936), Fernando (1952), and Bess (1956). A study of the egg laying of the closely related L. dorsalis was made by de Hass (1937). The work of Hutson and Austin, wherein they report that a male larva will consume 20-40 achatinas and a female larva will consume 40-60 achatinas during their development, has in particular been responsible for a BIOLOGICAL CONTROL 107 growing optimism as regards the controlling effect the larva of this beetle can and does have. Hutson himself (1920) anticipated this by announcing that it was his belief that L. tenebrosus was an import a n t factor in the control of the giant African snail in India. South (1926b), Jarrett (1931), Philbrick (1949), Rees (1950), Somanader (1951), and others, including several of the author's correspondents in southeastern Asia, in the absence of further supporting evidence have continued to add to this optimism until this beetle has been given, in the minds of many, at least the greater share of the responsibility for the increasing sharp decline in the A. fulica populations in Ceylon. Understandably, this whole matter headed the agenda in the author's investigations in Ceylon. Five species of lampyrids were found to be predatory upon both A. fulica and the endemic snails. The two most abundant were L. tenebrosus and an unidentified species of the genus Diaphanes. The other three species were considerably smaller, they were only rarely encountered, and they were not successfully reared to the adult stage and hence could not be identified. Of the two larger species, L. tenebrosus (Singhalese: adult is "kalamadiriya"; glowworm larva is "rabadulla") was by far the more common. In contrast to local reports, it was not restricted to the higher altitudes of the interior but was found at a few hundred feet of altitude in Mankulam in the north and Ratnapura in the south. The glowworm of Diaphanes sp. was found in drier environments and its ruptive coloration plus its habit of feigning death made it difficult to find. Unlike L. tenebrosus^ the glowworms of Diaphanes not infrequently co-operated in their attack on the snails, as many as five being found feeding at the same time on a single snail. In both L. tenebrosus and Diaphanes sp., the attack on the snail was initiated by stabbing and pinching with the long, ice-tongs-like mandibles. The area of attack was invariably the flesh on the left side of the foot near the base of the columella of the retracted snail. When the snail specimen was large or the glowworm was well fed, the attack was limited to the removal of a small amount of flesh in this region. In such cases, the snails would survive the attack and subsequently regenerate the lost flesh. In contrast to the report of Somanader (1951), this suggested very strongly that there was no toxic, proteolytic substance injected into the bite site, as reportedly is the case in Luciola (Alicata and Bess 1952). Pieces of flesh lacerated from the snail were seen to be squeezed by the bowed mandibles as one might hug a pillow. The mandibles were seen to be kept in a constant state of alternating with each other, first anterior and then 108 BIOLOGICAL CONTROL posterior, after the manner of sharpening two knives together. As this motion continued, the pieces of flesh became smaller and smaller. When no more tissue juices could be extracted, the small wad of flesh was gradually worked onto the dorsal surface of the head and then pushed aside. Once again, then, the glowworm would plunge at the equally rapidly retracting snail, remove another piece of flesh, and work it over as before. This would normally continue until all or nearly all of the snail flesh had been removed. The chances that a snail specimen would be attacked were found to be in inverse proportion to its size. A f t e r feeding, the glowworm apparently crawls only a short distance away before curling up in some shallow refuge to digest its meal. Glowworms in this state were often found in the immediate vicinity of achatina egg masses; but there was no positive evidence that they consumed the eggs. Similarly, there was only circumstantial evidence of cannibalism among the glowworms. Moist snail "retreats" were often found to be harboring several glowworms in addition to a great many snails. During prolonged dry spells, the glowworms sought refuge under rocks and in other deeper retreats. With the advent of the rains, the snails made a noticeably quicker return to activity than the glowworms, hence giving the snails a predatorfree period in which to forage. There is strong evidence that L. tenebrosus is not limited to the single seasonal cycle suggested by Hutson and Austin (1924). Somanader (1951) reports that the glowworm dies after a straight diet of A. fulica. Even when starved, full grown glowworms would not attack specimens of achatina that were larger than 40 mm. in length (Peterson 1957a). Since 1953, the Board of Agriculture and Forestry in Hawaii has made several attempts to introduce L. tenebrosus in Oahu as a biological control agent (Pemberton 1957, Thistle 1957, 1959a, b). Probably the earlier introduction into Hawaii of other lampyrids (Luciola cruciata, L. lateralis, and Colophotia praesta) to control the aquatic snail hosts of disease-producing flukes has done much to encourage further work of this type (Alicata and Bess 1952, Fullaway 1952, Bess and Alicata 1953, cf. Lutz 1927). The first shipment of the live glowworms from Ceylon was kept under subquarantine conditions as attempts were made to rear the specimens to maturity. Eventually, all specimens died (Thistle 1953a, Weber 1954). Subsequent shipments of L. tenebrosus from Ceylon were made by Bess (1956). Altogether, he sent over 1,000 glowworms of this species to Oahu for immediate release. Similar shipments directed in 1955 to FOA entomologist Edgar Dresner in Djakarta, Java, were kept un- BIOLOGICAL CONTROL 109 der observation and all specimens died before they could be released. Still other shipments were sent to George Peterson, government entomologist of Guam, and during the first few months of 1955, 933 specimens were released on that island (Peterson 1957a, b). In October, 1958, additional releases were made on Oahu and Maui (Davis 1959). Considering the nature of the ecological conditions and the extent of the releases, it might be assumed that L. tenebrosus has become successfully established both on Oahu and Guam. Subseq u e n t recoveries of specimens in the release areas in Guam lend substance to this assumption; however, to date, no recoveries have been made on Oahu or Maui. In attempting to evaluate the possible effectiveness of this predator in the control of the giant snail, it might be kept in mind that in Area Seven of the Pallekelle division of the Pallekelle estate in Central Ceylon, both L. tenebrosus and A. fulica have been together for a minimum of twenty-five years, and yet the giant snail still remains common to abundant (Mead 1955b, 1956a). All specimens of the lampyrid Diaphanes from Ceylon, intended for release on Oahu, died while under observation in the laboratory; and renewed efforts were not made to determine the value of this species as a biological control agent (Thistle 1957). Beetles of the family Drilidae also have come into serious consideration in proposed biological control measures. The work of Desmarest (1824), Mielzinsky (1824), and Lucas (1842, 1870, 1871) brought early attention to the pronounced malacophagous1 proclivities of the drilid beetles. According to Bequaert (1926), further observations of their biology were made by: Bellevoye (1870), Crawshay (1903), Bayford (1906), Rosenberg (1909), Schmitz (1909), and Deubel (1913). The work of Cros (1926, 1930) should be added to this list. It was not until the work of de Peyerimhoff (1914), however, that beetles of this family were known specifically to attack achatinids, at least in East Africa. The observations of Williams (1951), set forth in considerable detail, leave no doubt as to the readiness with which the drilid beetle larvae of East Africa will attack and consume the endemic achatinids, including A. fulica hamillei. Achatinas up to 115 mm. in length were observed to be attacked by nearly mature drilid larvae. Living specimens that were sent to Hawaii by Williams failed to multiply and consideration of their use in the biological control of the giant snail was thus abandoned (Pemberton 1954). Entering into this consideration was also the fact that the drilid life cycle is a long one. 1 Not "malacovorous" of some authors. 110 BIOLOGICAL CONTROL In his investigations in East Africa, Williams also found that large, black, voracious beetles of the genus Tefflus (Carabidae), in both the larval and adult stages, would consume the giant African snail. Several live adult specimens of this beetle, along with larval specimens of drilid beetles, collected by Williams in and near Mombasa, Kenya, were sent in May, 1948, to the Hawaiian Board of Agriculture and Forestry. The announced purpose for this shipment was to enable the Hawaiian authorities to observe under subquarantine conditions the biology of these beetles so that their worth in the biological control of A. fulica might be ascertained (Williams 1953). N. L. H. Krauss, the entomologist of the Board of Agriculture and Forestry, made a second trip in 1951-52 to East Africa and sent back to Hawaii 107 additional live specimens of two species of Tefflus. At first, the larger species, ca. 45 mm. long, was reported to be T. hacquardi; and the smaller species, ca. 25 mm. long, was reported to be T. carinatus (ICCP 1952). Krauss informed the author, however, that P. Basilewsky of the Museum van Belgish Congo, Tervuren, Belgium, has identified the larger species as Tefflus zanzibaricus alluaudi Steinberg and the smaller species as Tefflus purpureipennis wituensis Kolbe (cf. Krauss 1955). Q. C. Chock of the Hawaiian Board of Agriculture and Forestry successfully raised in cages several hundred specimens of the smaller species (Lennox 1953). The highlights of their biology have been reported upon by Weber (1954). Of particular interest is their ability to emit an irritant, when disturbed, which will burn the skin if it is not soon removed. Unfortunately, novel research of this sort, that is, enlisting large beetles to attack, kill, and consume giant snails, has been too much for some overenthusiastic newswriters. The ridiculously humorous article of Milhon (1948, 1949) only inferred that Tefflus would be used against the giant snail. But other articles, for example that of Ferguson (1948), treated the matter of their release as a fait accompli with such statements as: ". . . and soon Guam resounded with the crunching sound of beetles dining on giant snails." Efforts were made to correct this misinformation (Mead 1949d). At least partially in response to unrelenting pressure from the people of Hawaii and in spite of earlier decisions to the contrary, it was decided by the Board of Agriculture and Forestry in June of 1952 to release in the giant snail infestation in the Kaneohe area (three-quarters of a mile south of Mahinui) ten marked adult Tefflus zanzibaricus (Weber 1953). A few days later, ten more were released (Lennox 1953). The next year, twenty more were released (Thistle 1953a). During 1953 the beetles being raised in cages became less and less thrifty until FIG. 9.--The common amphibious hermit crab of Micronesia, Cenobita perlatus, not only consumes the flesh of the giant snail, but adds insult to injury by using the shell of its prey as a home. Length of shell ca. 125 mm. FIG. lO.--Achatina fulica from Mahinui, Oahu, Hawaii, showing multiple leukodermic lesions of a widespread disease found in high incidence in the older snail populations. The disease is suspected of having a viroid etiology and of being the decisive factor in the observed phenomenon of population decline. FIG. 11.--Tens of thousands of sun-bleached giant snails in this Rota tomato patch demonstrate several months' cumulative effects of clearing, poisoning with metaldehyde, and erecting a screen barrier. The live snails immediately to the left of the fence and the sparse tomato plants, to say nothing of the need for clearing away the dead shells, suggest that these combined measures have been something less than successful. (Photo courtesy of Yoshio Kondo ) FIG. 12.-The empty shells of dead giant snails form a tempting source of lime for those that remain, particularly in areas where the soil is acid. BIOLOGICAL CONTROL 111 in 1954, it was decided to release all that remained--eleven specimens (Thistle 1954a). During the following several years, only one unmarked adult Tefflus was recovered (Chong 1954, Weber 1954); and it was therefore believed by many that the beetle had not become successfully established (Dwight 1955). However, in September, 1959, a second live, unmarked specimen was found (Thistle 1959b). This discovery, nearly five years after the first unmarked specimen was f o u n d , has raised new hopes in certain quarters that this beetle has been able to maintain at least a beachhead population. Arrangements have been made with Richard LePelley, senior entomologist of the Department of Agriculture in Nairobi, Kenya, to make additional large shipments of Tefflus to Hawaii for immediate liberation in the expectation that they will eventually become firmly established. An earlier tentative decision to introduce Tefflus concomitantly with the predatory snail Gonaxis in the experimental island of Agiguan was fortunately postponed, in spite of the fact that it was done in Oahu. More recently, the giant carabid beetle Damaster b. blaptoides Kollar from Fukuoka, Kyushu, Japan, has been considered a possibly more suitable predator for achatina because of its spectacularly large size. Specimens under observation in 1958 in Hawaii avidly consumed larger snail specimens than Tefflus could manage. On July 3, 1958, fifty specimens were released on the Old Pali Road in Oahu and on July 9, 1958, twenty-five specimens were released in Haiku, Maui. Forty-six specimens of the closely related D. b. rugipennis Motchulsky from Sapporo, Hokkaido, Japan, were released on Tantalus, Oahu, on July 28, 1958. Although it is still too soon to determine whether or not establishment has taken place, it should be kept in mind that these beetles come from a temperate zone and the Hawaiian climate may prove inadequate for breaking the diapause. The same may be offered as an explanation for the fact that Scaphinotus striatopunctatus (Chaudoir) and Scaphinotus sp. from California have not been recovered since their release on Oahu in November, 1956. It is probably too early to determine whether or not specimens of an unidentified species of Tefflus from the Congo and Thermophilum hexasticum Gerstaecker from Kenya have become established since their release in October, 1956, and May, 1957, respectively (Thistle 1959a). But, even at this early date, both Tefflus and Damaster already have become involved in other aspects of the over-all biological program. Tefflus is a night feeder, as is its snail prey. This is good. But the introduced toad, Bufo marinus is also a night feeder and it 112 BIOLOGICAL CONTROL is particularly fond of beetles. Is this the reason why Tefflus has been recovered only twice in five years in the release areas? Damaster should escape this threat, as it is a day feeder; but at that time, the giant snails characteristically retreat to the tree trunks out of harm's way. It has been suggested that an excessive immediate scattering of a released biological control agent may preclude its establishment. In an effort to offset this, Damaster was released in Oahu in a sizable inclosure. The introduced mongoose is also a day feeder and in addition is fond of beetles. One got into the inclosure and put a finish to the experiment. Perhaps the mongoose did a better job on Tefflus and L. tenebrosus than it did on the rats, for which it was introduced to control. But, as will be shown below, both the mongoose and the rat will consume the giant snail and probably other species, including the introduced predatory snails, for which they may have a preference! The advisability of releasing any predatory beetle, as will be seen below in the discussion of the predatory snails, is still problematical and therefore pretty much of a controversy. The experiments in rearing Tefflus in Hawaii have been carried on for only a relatively short time and have revealed little more than the fact that these beetles can be reared in captivity with difficulty. The extent to which they are possibly effecting a control of achatinas in East Africa has not been determined in the slightest. Even though it can be reasoned that the cause for the achatinas being less abundant in East Africa might rest in the fact that there are also at hand predatory beetles and predatory snails, it is as risky as it is unscientific to reason a priori that the introduction of these predators elsewhere will produce similar results and that therefore introductions are justified in the absence of experimental confirmation. There is already evidence that although the predators do not manifest a high prey-specificity, they do show a preferential selection in what they will attack and consume. The fact that the drilid larvae will apparently more readily attack the East African predatory snails (Gonaxis and Edentulina, vide infra] than they will the achatinas (Williams 1951, Krauss 1951) warns that in a similar fashion, introduced predatory beetles might consume the endemic snails, or even some other invertebrate, in preference to A. fulica. Thus a new problem may be created in addition to the unimproved original one. In a partial answer to this problem, the information has been offered that both adult and larval Tefflus purpureipennis not only seem to show a preference for A. fulica, but they have little proclivity for climbing trees, thus giving a certain degree of immunity to such important endemic, arboreal BIOLOGICAL CONTROL 113 genera as Partula and Achatinella. It should be recalled at this point, however, that it was T. zanzibaricus, not T. purpureipennis, which was released in Hawaii. In reference to this specific matter, Kondo (1951 b) has pointed out that what is of infinitely greater significance is the danger to such terrestrial gastropod families as Amastridae, Endodontidae and Zonitidae. Small, unidentified lampyrid beetle larvae were seen by Williams to feed on quite small and newly hatching achatinas; but their quick dismissal from all but a brief mention suggests that they cannot or are not to be considered in a biological control program. Krauss (1951) similarly gives them only a slight mention in the reports of his East African investigations. A very detailed account of the feeding habits and life history of this type of beetle is given by Newport (1857). A quite common coprine beetle (Scarabaeidae) in East Africa was observed by Williams to frequent the sites of dead achatinas; but since it was only the disintegrating snail flesh which attracted the beetles, they were obviously not a factor in biological control. Birds The work of Kleiner (1931, 1936) has demonstrated the fact that examination of bird stomachs on an extensive program may show snail remains not only in a majority of cases, but in a great many different avian species. Further, he interestingly demonstrated that in spite of the high frequency of appearance, snails bulk small in the total content of the stomachs (2.69 per cent). Similarly, McAtee (1918) found that mollusks formed only 5.73 per cent of the total food intake of the mallard, and Collinge (1921) found that they formed 6.5 per cent of the animal food ingested by the starling. From these one cannot safely conclude that birds in general find snails uninteresting; for not just the relative degree of appetite for snails but the general availability of snails would determine the percentage consumed. Actually, Kleiner's most important finding is that a wide variety of birds will consume snails, notwithstanding the fact that in some cases the snails may have been taken into the digestive tract accidentally. But this conclusion is independently reached when one surveys the vast literature that has accumulated especially in the past century on the subject of birds consuming snails. In discussing the natural biological control factors of the pestiferous Theba pisana in Italy, de Stefani (1913) lists crows, magpies, and owls. To this list, Basinger (1927) in California added the Hudsonian curlew, pigeon, English sparrow, chickens, and ducks. The domesticated duck has the reputation of being the most avid consumer of snails (e.g., cf. Panos Marti 1952). In fact, there is in correspondence and in the literature the persistent report (e.g., 114 BIOLOGICAL CONTROL South 1923b, Jarrett 1931) that the giant African snail was transported into uninfested areas, especially in Malaysia, with the intent that they would provide suitable food for ducks. Just the reverse process is concurrently being undertaken; that is, the ducks are being enlisted in local control measures to keep down the population of the snail. G. A. S. Barnacle (in litt. Jan. 15, 1950) states that ducks were "introduced" in this manner in various areas in Ceylon. Essentially the same report was given by R. C. L. Notley (in litt. Dec. 5, 1950) with the addition that some ducks got pieces of snail shell stuck in their throats. In Thailand, Ariyant Manjikul (in litt. Feb. 14, 1952) states that the giant African snail makes 'Very good feed and the duck raisers collect them" for this purpose. A similar report comes from Guam (Peterson 1957b). In Singapore, R. E. Dean writes (in litt. Jan. 17, 1952) that he has been informed that the common practice there is to allow the local strain of domestic ducks free range over the compounds and that they will attack the giant snails with gusto. In that same region, A. F. Caldwell gives supporting evidence (in litt. April 23, 1953) in the following statement: "I used to keep a few ducks and fed them a number of these [giant African] snails daily as part of their diet. If the shells are broken ducks eat the snails with obvious relish." A similar report is made by Jaski (1953). The practice of using ducks in an attempt to control H. aspersa in California citrus groves, according to Lewis and LaFollette (1941), proved to be of "some value," but it was not recommended. On the other hand, Hely (1946) recommended it as "excellent." In flower and vegetable gardens, however, the ducks may cause more damage than the snails, especially if succulents and seedlings are present. Not only ducks, but chickens will feed on achatinas (Hutson 1920, South 1926b). Lang (1919) also observed this in the Congo. The larger, thick-shelled specimens, however, are not effectively attacked by either ducks or chickens unless they are crushed (van Weel 1949). In South Africa, Joubert and Walters (1951) report that turkeys as well as ducks will eat large numbers of the serious snail pest T. pisana. The utilization of the giant snails as a food supplement in poultry has been treated in detail below, under the discussion of control through human use. As early as 1911, Green proposed the idea that insectivorous birds might be important in controlling A. fulica. South (1926fo) and later Philbrick (1949) announced that Centropus chlororhynchus, the "jungle crow" (also called "pheasant crow" or "cockoo") attacks and consumes the giant snail in Ceylon. This was recently verified by the -BIOLOGICAL CONTROL 115 author during field observations in Ceylon. In areas frequented by jungle crows, shells with characteristic damage were commonly encountered. Of 98 live achatina specimens examined on the Warriapolla estate near Matale, 9 (9.2 per cent) were found with diamond- or triangle-shaped wounds. These indubitably were caused by bird pecks. Even in the cases of extensive trauma, tissue regeneration was sufficiently advanced so that it seemed apparent that the snails would survive. Keeping them under observation in the laboratory confirmed this assumption. An explanation for their escaping fatal injury seems to be found in the words of E. Phyllis (in litt. Dec. 11, 1950), who states that he has seen the jungle crow in Ceylon h u n t i n g for the giant snail for hours on end and that it does not consume more than a part of the snail after it finds one. Introducing the j u n g l e crow into a non-endemic area as a biological control agent to control the giant African snail is completely counterindicated; for this bird is a foraging rogue with the strongest raiding and robbing proclivities, eating in particular the eggs and young of ground-dwelling birds. Krauss (1952) reports that in the northwestern part of Madagascar, a large native bird, the famakankora ("snail-breaker"), Anastomus madagascariensis (Ciconiidae), is said to feed on achatina and other animal food. It may have been the anvils of this type of bird which Jaski (1953) saw in South Africa. He reports that scores of A. achatina were found hammered to smithereens. Observing the snail-eating habits of some of the larger birds has persuaded a few people to recommend introducing them into areas infested with the giant African snail. R. S. Gardiner states (in litt. Nov. 22, 1949) that in Chile the thick-kneed plovers or so-called "queltehue" birds (Burhinus superciliaris), kept in inclosed gardens by clipping their wings, are exceedingly effective in removing all sorts of terrestrial mollusks. Her suggestion that they be used in controlling A. fulica, however, is no more practical, for several obvious reasons, than Eyerdam's recommendation (1952a, b) that the New Guinea bush hen or bush turkey (Megapodius spp.), the rhea (Rhea spp.), and the cassowary (Casuarius spp.) be used. With Eyerdam's suggestion in mind. It was of more than passing interest to read in a recent list of offerings of a dealer in live animals that a cassowary could be purchased for a mere $3,000! Gressitt (1952) and others have reported megapode birds on some of the western Pacific islands (e.g., the northern Mariana Islands) which are not infested with the giant snail. It is possible, though, that both snails and megapodes 116 BIOLOGICAL CONTROL have come together in some of the Palau Islands. This is a matter distinctly worth investigating. Crabs In Micronesia, the ubiquitous hermit crab, Cenobita perlatus Edwards (called "humpa" by the Bonin Islanders) has been observed (Mead and Kondo 1949, Mead 1950b, c) not only occupying as high as 21 per cent of the empty A. fulica shells encountered in beach populations but actually consuming the live snail by pinching off small pieces of the flesh with the chelae. This species spends long periods of time on land; specimens were found in Chichi Jima at an altitude of approximately six hundred feet and over one and a half miles from the nearest seashore. There is, however, an apparently geometrically progressive reduction in their numbers in an inland direction. Even in achatina shells abandoned by the hermit crab, there are unmistakable evidences of their work. The shell becomes highly polished, not just on the underside but all over, due to the rough treatment it gets; and invariably the columellar surface of the ultimate whorl is characteristically abraded away, making more room for the hermit crab. Wilson Savory of Chichi Jima stated to the author that he released half of a flour sack of small achatinas on the tiny island of Higashi Shima, just off the northeast coast of Chichi Jima in 1942. He said that when the island was visited again the following year the hermit crabs were found to have taken over the achatina shells entirely. A mysterious "achatina-free" area, approximately three-quarters of a mile long and at least five hundred feet wide, bordered by dense populations of A. fulica along the east coast of Rota, proved upon examination to be nothing more than a region where the hermit crabs were completely dominant (Mead 1950b, c}. Conditions along the adjacent coast seemed to be particularly favorable for hermit crab reproduction and existence, and the recent immigrant A. fulica had not yet successfully invaded and infiltrated that region. Burningover the land to plant watermelons, had apparently neutralized the area; but the hermit crabs had relatively rapidly reinvaded the area and kept it free of live achatinas. Paradoxically, the hermit crabs caused greater damage to the melons than did the achatinas; and yet it was the snails which were indirectly responsible for this damage as it was their abundant empty shells that provided vital protection for the soft bodies of a greater population of hermit crabs than otherwise could have built up. A brief reinspection by Kondo (1952) of this achatina-free area and two other similar areas on Rota, emphasized the need for a thorough ecological study to determine what decisive factors are actually operative. BIOLOGICAL CONTROL 117 So far as can be determined, no other endemic predator in Micronesia, with the possible exception of the coconut crab, produces a greater kill of the giant snail. Even so, the effect upon the snail population is undoubtedly a minor one, except in localized beach areas or on relatively small islands, where the hermit crabs may have a decimating or even exterminating effect. Davis (1954) concluded that "Cenobita was not a predator of Achatina." This conclusion, however, was based to the largest extent upon the undependable behavior of caged specimens. Somanader (1951) describes and pictures unidentified hermit crabs which take over the shells of A. fulica in the Kalkudah beach area of eastern Ceylon; but he does not suggest that there is predation. On the island of Auluptagel (Aurapushekaru) in the Palau group, there was found a large, conspicuous cave of a robber crab or coconut crab (Birgus latro L.) on the slope of a hill. Well over fifty broken shells of A. fulica were found strewn three to four feet below the opening of this cave. Sharp, angular breaks in the very thick, large shells clearly indicated that these had been broken open by the coconut crab. The characteristic droppings gave further convincing evidence. The coconut trees have become virtually extinct on this island because of the work of the introduced coconut beetles (Oryctes rhinoceros and Brontispa mariana), and it seems obvious that the coconut crab has had to turn to some other source for its food. It is a curious thing that the introduction of beetle pests has predisposed to the destruction of a snail pest! Kondo (1952) reports that on Agiguan, this crab and the rats together kill more than twice as many giant snails as does the predatory snail Gonaxis. Direct and indirect evidence of the consumption of achatinas by this crab were reported upon by Davis (1954). Because the coconut crab is so highly prized as an article of food and therefore diligently hunted for by the Micronesians, and because the eating of snails by the coconut crab is only incidental in its catholic diet, Birgus latro unquestionably has but the smallest part in the biological control of A. fulica. An unidentified land crab in the Diani Beach area of Kenya was observed by Williams (1951) to consume live achatinas; and, in at least one case, the snail was considerably larger than the crab. Its predation, however, is apparently limited to the peripheral beach zone. Dartevelle (1954) tentatively identifies this crab as belonging to the genus Ocypode. The so-called "paddy crab" was reported in several instances to feed on the giant snail in Ceylon. Although this was not verified in 118 BIOLOGICAL CONTROL the field, it is altogether possible that this crab is of some local, minor value as a biological control agent. Flies Sarcophagids, sciomyzids, phorids, and other diptera have been reported by a number of observers to come from the dead bodies of various land snails. In by far the majority of the cases, it has been established, or very strongly suggested, that the fly larvae are saprophagous rather than parasitic. And in only a very few instances (e.g., Rostand 1920, Mercier 1921, Mokrzecki 1923, Berg 1953, 1955, Muma 1954, 1955) has there been more than circumstantial evidence of a possible normal parasitic role. Keilin (1919, 1921), Seguy (1921, 1935), Bequaert (1925, 1926), and Pelseneer (1928) have made the major contributions and have surveyed the literature for reports on the general subject of diptera-mollusk associations. With the exception of those that concern the giant snails, more recent works which should be mentioned are those of Bhatia and Keilin (1937), Lopes (1940), Metteo (1946), Berg (1953, 1955), and Muma (1954, 1955). A. R. Main of the University of Western Australia is conducting significant investigations of dipterous parasites which he has reared from the endemic land snails Bothriembryon spp. Muma in particular has done a significant piece of work on the biological control factors apparently operative in keeping down the numbers of a presumed "beneficial" tree snail in Florida citrus groves. He reports that dipteran parasites are the most important agents in reducing the snail population. His rearing experiments convinced him that he had found true parasites and not just saprozoites. From the "parasitized" snails, he raised four sarcophagids, one phorid, and one chloropid, viz., Sarcophaga lambens Wd., S. morionella Aid., Johnsonia elegans Aid., /. cf. frontalis Aid., Megaselia sp., and Hippelates dissidens (Tuck.). As a complication he found that some of the sarcophagids were being attacked, but apparently not seriously, by the epiparasites Aphaereta auripes (Prov.) and Melittobia sp. The wingless phorid flies of the genus Wandolleckia were apparently first observed on achatinas by Cook (1897) in Liberia, where he saw them running about on the surface of living Achatina variegata (i.e., A. achatina). The following year, Wandolleck offered the suggestion that the flies feed on the slime of the snails. Bequaert (1919) not only supported this suggestion but advanced the idea that they are perfectly harmless to the snails. He gave further information on the history and biology of this species (W. achatinae] and in subsequent papers (1925, 1926), he discussed two other species of this genus, viz. W. indomita and W. biformis (cf. also Pilsbry and Be- BIOLOGICAL CONTROL 119 quaert 1927:472; Brues 1907). The latter species has been shown to live on A. rugosa chapini (Bequaert and Clench 1934). Baer (1952) considers the "commensal" W. biformis synonymous with W. achatinae and reports finding it on A. achatina L. and Archachatina ventricosa Gould in the Ivory Coast, Liberia, Cameroons, and the former B. Congo. Rodhain and Bequaert (1916) reported a number of larvae of Mydaea bivittata Macq. devouring the achatinid Burtoa nilotica Pfeiffer. As far as can be determined, Senior-White (1924) was the first one to report phorid flies from A. fulica. These, which were bred from a dead snail, he described as new--Megaselia achatinae. Brues (1942) has suggested that it is possible Senior-White actually had the ubiquitous Megaselia xanthina Speiser. It was this latter species which Smedley (1928) in Malaya bred from the eggs of A. fulica. Because he reports the flies as being capable of producing human intestinal myiasis, Smedley emphasizes the public health implications rather than those of biological control. The phorids bred from a dead specimen of A. fulica in Hawaii by Yoshio Kondo were described by Brues (1942) as new--Megaselia biformis. At that time, Brues advanced the suggestion that in spite of the lack of experimental evidence, some of the phorids "may be true internal parasites of living snails." In 1949, Van Emden reported Ochromusca trifaria Bigot to attack A. craveni E. A. Smith at Fort Johnston, Nyasaland. Captain W. A. Lamborn, the collector of the specimens at Fort Johnston, later reported to Krauss (1951) that the flies were bred from dead snails and had not been observed to come from live individuals. This report removed earlier optimism about the fly being of use as a biological control weapon. An unidentified fly is described by van Weel (1949) as being a possible parasite of the eggs or young of A. fulica in Java. The weak spot in his evidence is that, despite the fact that the "narrow cleft" in the breeding container was not wide enough to permit the adult fly to enter and lay eggs, it was certainly sufficiently wide to allow oviposition within the container at the site of the cleft. Hence it is not possible to determine whether or not the fly larvae had access to the snails or eggs before death from some other possible cause. Hardy (1952) later announced that it was a new species of the phorid genus Pericyclocera. A slightly larger phorid was commonly encountered by the author in the giant snails in Ceylon. Specimens were sent for identification to the British Museum, through the Commonwealth Institute of Entomology. According to D. E. Hardy (in litt. April 18, 1956), these 120 BIOLOGICAL CONTROL are being described by C. N. Colyer as a new species of Spinophora. Under the many circumstances in which these flies were observed, there was never evidence of anything but a saprozoic role with respect to both the endemic and the giant snails. Snails killed in the laboratory would attract in a few hours numerous phorid flies from the nearby bush. Snails killed and being fed upon by the glowworm were observed to contain at the same time female flies busily laying eggs in characteristic patterns on the lip of the shell. Typical larvae were f o u n d in insects which had been killed and allowed to decompose. It has been suggested that these phorid flies may be implicated in the transmission of a disease in A. fulica (Mead 1956a). In Saipan, Lange (1950) reared Sarcophaga gressitti Hall and Bohart and S. dux Thomsen from dying and dead A. fulica. Near Diani Beach, Kenya, East Africa, Krauss (1951) found fly larvae and pupae in the shells of Achatina "in which the snails were dead and decomposing, and often reduced to a black foul liquid." These were later identified by Van Emden as: Aethiopomyia steini Curr., Alluaudinella bivittata Macq. (Muscidae); Sargus sp. (Stratiomyidae); and Discomyza similis Lamb (Ephydridae). In the same area, Williams (1951) was able to find species of the muscoid genera Sarcophaga^ Panaga^ and Aethiopomyia in only the dead or dying achatinas. Other saprophagous but unidentified flies were observed on dead achatinas by Pangga (1949) in the Philippines. The large maggots of filth flies were not uncommonly seen by Kondo and the author in several of the islands in Micronesia and especially in the Bonin Islands; unfortunately, a tightly packed itinerary did not permit rearing them for identification. There was, however, absolutely no discernible evidence of parasitism in any of the thousands of snails specimens examined. Krauss similarly found no evidence of dipteran parasitism in his investigations in Kenya, Zanzibar, Tanganyika, and Madagascar. Recent initial efforts to use Johnsonia elegans, which attacks Drymaeus in Florida, and a Tetanocerid fly from New York in the biological control of A. fulica in Hawaii have failed completely (Thistle 1959a). H e l m i n t h s There has been found in the literature no report of helminths of any type being found in the achatinid snails. Pilsbry and Bequaert (1927:472), Pelseneer (1928, 1935), and Adam and Leloup (1943) refer to records of helminths being found in other gastropods. The work of Chitwood and Chitwood (1934) is valuable in that it lists the nematodes encountered in gastropods. An unidentified rhabditoid nematode was encountered by the author during BIOLOGICAL CONTROL 121 the examination of the lower one-quarter of the intestinal tract of several specimens of A. fulica in Ceylon. However, all evidence suggests that this worm is only an incidental symbiont of low incidence and little consequence. Dying specimens of A. fulica found in 1957 in Maui, Hawaiian Islands, by Lew Akaka, were discovered to contain many nematodes; but, in the highest probability, these were saprophagous and not parasitic. Actually the whole subject of gastropod helminthiasis has been seriously neglected and it is hoped that in the near future attention will be given to it by qualified specialists. Mammals As in the case of bird stomach examinations, the contents of the stomachs of a great many species of mammals have been found to contain the remains of terrestrial or aquatic snails. Without much question, by far the larger share of such cases clearly represent incidental, accidental, or subsistence consumption of the snails. It therefore would be as pointless as it would be misleading to attempt to compile an exhaustive list of mammals known to have consumed snails at one time or another and offer it as a list of the mammalian enemies of snails. There are however certain mammals which have a manifest appetite for snails. Chief among these probably is the shrew (Hamilton 1930, Ingram 1942b, 1944), which Clench (1925) brands as one of the worst enemies of land mollusks. Significantly important among the others are: rabbits (Lawson 1929, 1930, Oldham 1929b, Wright 1909), mice (Coghill 1909), and rats (Adams 1938, Hoffman 1936, Lawson 1920a). Of the mammals mentioned so far, only the rat has been linked definitely with A. fulica. Meer Mohr (1935) illustrated damage to giant African snail shells which was suspected to be caused by rats or birds. Abbott (1951c) and Williams (1951), similarly cautious, implicate the rat. The survey in Micronesia conducted by Kondo and the author, however, removed all doubt. Rat nests containing the chewed remains of A. fulica shells were found in one or more instances in Guam, Chichi Jima, Haha Jima, and Tinian. In each case, the shells were numerous, fairly small in size, and showed unmistakable evidence of rat work. Numerous arcuate shell flecks in the nest debris indicated that the shells were brought to the nest before being eaten. As the entire columella was destroyed in many specimens, it is probable that all of the soft parts were removed. In northern Saipan (Magpi) two instances were witnessed where rats were feeding on crushed achatinas on the road; but the rats were emaciated and apparently using the snails for subsistence. After considerable field study, Kondo (1952) concluded that on Agiguan the rats are more effective predators of the giant snail than the experimentally introduced carnivo- 122 BIOLOGICAL CONTROL rous snails. The report of Davis (1954) was in sharp contrast. However, it is felt in general that, with certain fairly clear-cut, localized exceptions, the over-all controlling effect the rats have on the giant snails is relatively minor. On the contrary, the giant snails are often to be considered contributory to the rat problem by providing themselves as an additional source of food. As an ironic twist, the giant snails on Guam contributed still further to the rat problem by consuming with apparent impunity the warfarin bait intended for the rats (Peterson 1957b). The omnivorousness of domestic pigs suggests correctly that they will consume the giant snails and that their foraging near habitations provides some small measure of control. South (1926b) lists the wild pig among the enemies of: A. fulica. The mongoose, Herpestes mungo, is said to eat the giant snail in Ceylon (Philbrick 1949, Rees 1950). While the author was in Ceylon, a great deal of circumstantial evidence was collected implicating this predator in the destruction of achatinas. For example, on several occasions freshly broken snail shells and partly consumed snail carcasses were found in the immediate vicinity of burrows which the local inhabitants declared were those of the mongoose. Equally often, piles of broken shells were found scattered around a prominent rock indicating that the mongoose had used the rock as a "hammer stone." Some of the shells were remarkably broken almost exactly longitudinally. The hepatopancreas or "liver" of the snail not infrequently was left untouched, suggesting that it was distasteful or at least less appetizing than the rest of the snail carcass. In numerous instances in the field in Ceylon, there were found shells which had been flecked open, hence indicating that they had been attacked by a small mammal of some sort. Informants stated that rats, both the "wild" type and the introduced type, were known to eat the giant snails. In other cases they stated that it was the giant squirrel, "Dondolena" (Sciurus macrurus Pennant). In still other cases, for example on the Godahene estate near Kalutara, there was the strongest evidence that it was the Bandicoot, "Uru-miya" (Bandicota malabarica) which was breaking the shells and eating the snails. The "mongoose" of East Africa to which Abbott (1951c) refers is quite probably the civet cat, Bdeogale tennis and B. crassicauda, whose snail-eating habits are interestingly described by Williams (1951). The jackal in Ceylon (Philbrick 1949, Rees 1950) and the baboon in East Africa (Abbott 1951c) are also reported to eat the giant snail. In Ghana, an endemic wild cat, referred to by the natives as "odompo," is believed to feed on giant snails. Since the snails are BIOLOGICAL CONTROL 123 an important food item to the natives, the "odompo" is hunted and killed at every opportunity. And as an interesting elaboration, a snake--the deadly horned Cerastes--is known to attack the "odompo" and has been seen in the company of the giant snails; it is therefore believed by the natives to be the protector of the snails (Martinson 1929). The large musk shrew, Suncus murinus (L.) became established on Guam in 1953 and in two years' time it had spread considerably (Peterson 1959). For a time, it was hoped that this predator would prove to be a potent new biological control agent in the battle against the giant snail; but field and laboratory observations indicated that there was little hope for anything of this sort (Peterson 1957b). As a strange effect of mammals on snails, Peterson (1954) and Davis (1954) report that the great many feral goats on Agiguan Island are apparently responsible for killing and injuring the giant snails by trampling and crushing them as they get under foot. Micro-organisms The most neglected aspect in the problem of the giant African snail, and in fact in the entire field of malacological biology, is the study of the role of micro-organisms in molhiscan symbiosis and pathology. A search of the literature reveals few references indeed on this important subject, and none of these concerns a virological investigation. The work of Drz (1913), which has not gone unchallenged, reports the presence of bacteria in special cells between the kidney and stomach in Cyclostoma. Wurtz and Gray (1939) found in the intestine of Triodopsis albolabris what appeared to be eight new species or varieties of bacteria belonging to the genera Escherichia, Alkaligenes, and Bacillus. Edward Steinhaus recently reported to the author that in France an Aerobacter infection in colonies of commercially raised Helix. Pan (1956) found in tissue sections of the freshwater snail Australorbis glabratus an obligate, intracellular, acid-fast bacillus in various organs of the body, and a "fungus spore or yeast-like agent" in the nervous tissue. Spirochetes were found by Fantham (1921) in the hepatopancreas of South African pond snails; their role however was not determined. Dartevelle (1954) similarly reports the spirochete Borrelia in the hepatopancreas of 10 per cent of the achatinas examined in a study in the Congo. For an early basic work on the protozoa of snails, one should turn to that of Kuhn (1911). Pelseneer's survey of this portion of the literature (1935) brings the subject more nearly up to date although he does not mention the work of Hegner and Chu (1930), wherein 124 BIOLOGICAL CONTROL the ciliate Balantidium haughwouti was reported to be found in Philippine fresh water snails. More recently, the species of the sporozoan genus Klossia, found in the kidneys of several genera of snails and slugs, have been discussed by Nabih (1938). The contributions of Kozloff (e.g., 1946) on the mastigophoran symbionts of gastropods stand out among the best in this whole field. The mastigophoran Trypanoplasma isidorae and an amoeba have been reported from the seminal receptacle of the pond snail Bulinus (Isidora) tropica; undescribed amoebae were also found in the slug Arion fuscus (Fantham 1923, 1925). As far as can be determined, the only protozoan (and, for that matter, the only micro-organism) that has been described from the giant snails is Trichodina achatinae found in the seminal receptacle of Achatina zebra (Fantham 1924, Fantham and Robertson 1927); and, although it has been referred to as a "parasite," there is not at present sufficient experimental evidence to warrant such a classification. At two different times, diflagellate protozoans were found in smears of the intestinal tract of A. fulica in Ceylon; but contamination could not be ruled out as a possible explanation for their presence. There was absolutely no evidence of pathogenesis. Gain (1896) refers to Laurent's work wherein the eggs of Deroceras reticulatum are claimed to have been found infected with a fungus even before they were deposited. The more careful work of Tervet and Esslemont (1938) revealed the fact that the eggs of this slug were infected with the fungus Verticillium chlamydosporium Goddard. Even though they felt that this fungus exerted "a strong natural control" of the slug, they considered "impracticable" its use in biological control. The significantly high percentage of infection in the eggs which they collected in the field might find an explanation in the possibility of a sequela of fungous infection subsequent to mechanical injury to the eggs during field collecting and transporting to the laboratory. It is at this instant apropos to point out that Lovett and Black (1920) found fungous diseases "particularly active and virulent" in their breeding cages of slugs; but their field observations indicated that such diseases were "of minor importance under natural conditions." In looking back over the reports extant in the field of malacological microbiology, it is immediately apparent that the surface of the multifold problems has hardly been scratched. In most cases, the reports have been made by investigators unqualified to do definitive work in microbiology. Hence, if the organism is identified at all, the identification is still very much open to question, as is the particular type of symbiosis (sensu lato), that is, whether it is commensalism, BIOLOGICAL CONTROL 125 m u t u a l i s m , or parasitism. Or, for that matter, even the possibility of the association being incidental, accidental, or contaminatory in nature seems hardly or not at all to have been considered. For the problem at hand, however, there is needed not only the information that parasitism exists but that a pathogenesis per se is detectable. Then an investigation of such problems as mode of transmission, epizootiology, methods of culture, possible epiparasitic contamination, and introduction into the field can be undertaken. It is obvious that there is still a long way to go in this phase of the investigations; this notwithstanding, in the long run it will probably prove to be the most practical, productive, and revolutionary method of control. In fact, the possibility of a spontaneous biological control of the giant snail rests almost completely in the field of microbiology. The possible role of micro-organisms in the so-called "diseased snails" and in the phenomenon of "decline" is discussed under the proper headings below. Mites Anyone attempting to raise snails or slugs may all too soon find the specimens infested with small, swarming mites. As was the author's experience in raising the giant slug Ariolimax, the mites can become so abundant that they even interfere with normal locomotion and feeding. Banks (1915) made early, inconclusive observations on malacophilus mites. Andre and Lamy (1930, 1931) have given us our only comprehensive work on the mites of mollusks. There are no known records of mites being found on A. fulica. But Bequaert (1925, 1926; Pilsbry and Bequaert 1927:472) reports an unidentified, "ecto-parasitic" mite on a live achatina in the former B. Congo; he also cites from that same area, Stuhlmann's records of mites on A. schweinfurthi and A. stuhlmanni. Nothing is offered as to the possible parasitic effects upon the snails; however, Turk and Phillips (1946) in their monograph of the slug mite Riccardoella limacum (Schrank) (i.e., Ereynetes limacum) give convincing evidence that mite and mollusk live together in commensalism of a fairly high order. Baker and Wharton (1952) support this interpretation. On the other hand, a closely related species is believed to be at least contributory to an unthrifty, malformed condition in Arianta arbustorum (Oldham 1934). Simroth and Hoffmann (1928) list from the literature a questionable record of the tick Amblyomma variegatum being found on the achatinid Limicolaria adansoni. Reptiles In the former B. Congo, the large monitor lizard Varanus niloticus (Linne) has been observed by Lang (1919:55) to feed principally upon half-grown achatinas. This observation was 126 BIOLOGICAL CONTROL supported by Pilsbry and Bequaert (1927:473) and elaborated upon w i t h the information that the stomach of one specimen was found to contain four large snails. They quote from Schmidt, who earlier pointed out that the teeth of Varanus niloticus are adapted for crushing snails. Species of this reptilian genus occur endemically or have been introduced on a number of the snail-infested islands of the Pacific (Loveridge 1945). There has been no adequate examination of the stomach contents of these lizards to determine the extent to which they are feeding upon A. fulica. As far as is known, the only studies of the feeding habits of this lizard in the Pacific area are those of Kondo (1952) and Davis (1954); but, unfortunately, negative stomach contents in two specimens and the reluctance of two other captive specimens to feed on achatinas cannot be considered anything but inconclusive. It has already been pointed out however that the lizard, because of its appetite for young chicks and eggs, presents a fair threat to poultry, especially where other food is not available (Mead 1949d). This has caused Eyerdam (1952a, b) to conclude that Varanus in the Pacific is at best pretty much of a neutral value. Nor does the fact that the lizard will readily consume the coconut crab (Birgus latro) materially change the picture; for, although the crab is a pest of the coconut palm, it will attack and consume the giant snails--undoubtedly far more than will the lizard. But with the recent announcement that the lizard will consume the grubs of Oryctes rhinoceros (L.), which is in many respects the most serious agricultural pest in the Pacific, the scales seem to be tipped very much in the favor of Varanus (Gressitt 1952). Although V. monitor and V. salvator were commonly encountered in some snail-infested areas of Ceylon, there was no evidence that they were anything more than of incidental value in controlling the giant snail. The only chelonian known to attack A. fulica is the pond turtle Nicoria trijuga thermalis of Ceylon (Green 1911b). During the author's investigations in Ceylon, it was determined that because of the strongly aquatic affinities of this turtle, it was essentially of no value as a biological control agent. Recently, Vianney (1953) reports that a small Javanese snake Pareas c. carinatus, in captivity, fed frequently on very young achatinas, swallowing them shell and all, although somewhat older specimens were ' 'seized in a peculiar way and . . . extracted from their shells by the freely moving mandibles." S n a i l s In spite of the fact that A. fulica will readily and even avidly consume the flesh of injured, dying, dead, and even putrifying individuals of its own species, there is absolutely no evidence of BIOLOGICAL CONTROL 127 predation per se. Even the eggs and delicate young are apparently completely safe in the presence of larger individuals. Green (1911b) was the first to emphasize this point; and it should be re-emphasized here so that it will be understood that except for competition for food, the giant snails do not provide any direct limiting factor among themselves. It is of real interest to note, however, that an injured or dying snail has a very positive attraction which brings out the latent cannibalistic tendencies of foraging snails. This suggests that there is released in the injured or dying snail some substance which is not apparent in the normal, healthy individual. Many pulmonate gastropod species have been reported in the literature to display occasionally a propensity for cannibalism (e.g., Elliot 1918). In 1927, Pilsbry and Bequaert (p. 469) indicated that perhaps the most important predacious enemies of the African land mollusks were the rapacious streptaxid snails. Bequaert later reiterated this point (1950a). To determine more specifically what the natural enemies of A. fulica were in East Africa, where this species is autochthonous, the Pacific Science Board of the National Research Council and the Office of Naval Research in the fall of 1947 sent F. X. Williams to Kenya and Zanzibar for several months. Interesting accounts of the information gathered in the field are on record (L. C. Williams 1949, F. X. Williams 1951, 1953). The following May, live specimens of the two endemic predacious snails in the Diani Beach area of Kenya, Edentulina affinis C. R. Boettger and Gonaxis kibweziensis (E. A. Smith) (Streptaxis), were sent by Williams to the Board of Agriculture and Forestry in Honolulu, Hawaii (ICCM 1948, Bryan 1949). The near maximum length of these species is 50 mm. and 22 mm., respectively. The purpose behind sending these alive was to observe under subquarantine conditions their biology so that the possibility and practicability of using them in the biological control of the giant African snail could be determined. While these observations were still in progress in the summer of 1949, Kondo and the author made a survey of Micronesia, under the same auspices, to determine among other things what island, infested with A. fulica,, would be most suitable on which to conduct an experimental introduction of the predatory snails. Reports from these two investigators (Mead and Kondo 1949, Kondo 1949, Mead 1949a, 1950b, c) emphasized the complexity of the problem and recommended that in the selection of an island for the proposed experiments the following factors be kept in mind: size, topography, proximity, available transportation facilities, accessibility, inhabitance, isolation, quarantinability, horticultural and sylvicultural crops at 128 BIOLOGICAL CONTROL stake, relative abundance of A. fulica, types and abundance of indigenous snails, and practicability of possible subsequent eradication of the introduced predatory snails. After reviewing the recommendations in March, 1950, it was determined at the Fourth Annual Meeting of the Insect Control Committee for Micronesia, of the Pacific Science Board, that the small (3 by 1 mi.) uninhabited island of Agiguan (Aguijan), about five miles off the southwest coast of Tinian in the Mariana Islands, would be most suitable (cf. Gressitt 1954:53). Kondo, R. E. Enders of Swarthmore College, and Mead had determined for the first time that A. fulica was actually on this island and that for other reasons the island had possibilities. At the same ICCM meeting, it was decided that R. T. Abbott should go to East Africa, collect Gonaxis and Edentulina, and return to Guam with them for release on the selected experimental island which was to be maintained in quarantine (ICCM 1950). This mission was accomplished during the following May and June (Abbott 1951 b, c). On May 31, 1950, and within approximately ten days after 545 living specimens of Gonaxis kibweziensis had been sent via air express to Guam by Abbott, about 400 surviving individuals2 were released on the first terrace of the southwestern end of Agiguan by Robert P. Owen, entomologist and staff quarantine officer of the Trust Territory of the Pacific Islands (Owen 1950, 1951). On August 30, 1951, Owen returned on the third expedition to Agiguan to determine what progress had been made in the trial field experiment. He was accompanied by George Peterson, Jr., the entomologist of the government of Guam, and J. Lockwood Chamberlin, who was conducting an ecological study of A. fulica in Tinian. Due to an unfortunate set of circumstances, they were able to spend only three hours examining the area where Gonaxis had been introduced the previous year; but they were able to obtain evidence indicating that this predatory snail had survived, that it was reproducing, and that it had spread at least 300 feet beyond the point of release (Owen 1951). There remained undetermined such important points as: whether or not Gonaxis was actually feeding on the giant snail; whether there was being effected any appreciable control of the giant snail; whether any inroads were being made on the endemic snails; and whether any deleterious side effects were being demonstrated as a result of the introduction of Gonaxis. The ICCM, which had now become the Invertebrate Consultants Committee for the Pacific, decided at its Sixth Annual Meeting in February, 1952, that these and other points should be investigated over a period of a number of days by Kondo 2 This number was later reported to be "229" (Owen 1953) and "407" (Davis 1954). BIOLOGICAL CONTROL 129 during the following summer. In accord with the suggestions of the ICCP, Kondo took with him 100 live specimens of Gonaxis, raised in Hawaii, for release in a high, forested area near the old village on the third terrace of the east end of Agiguan. At least 95 per cent of these were alive at the time of release. During his seventeen days on Agiguan, in the company of four other assistants and investigators, including Owen and Peterson, Kondo (1952, Anon. 1953a) made a n u m b e r of discoveries; among the most significant are the facts that Gonaxis had actually been feeding on A. fulica, that it had effected only about a 19 per cent kill, and that in two years time, the original lot of approximately 400 Gonaxis had increased to an estimated 21,750. This great population build-up notwithstanding, Kondo announced that Gonaxis could not be considered a strong factor in the biological control of A. fulica. But before this time, Kondo had eminently qualified himself for this field work by spending a number of months in Hawaii making careful observations of the biology of several predatory snails, including Gonaxis. The predatory snails sent by Williams to the Hawaiian Board of Agriculture and Forestry were transferred to the laboratory of the Hawaiian Sugar Planters' Association in January, 1950, where Kondo was asked to take charge of them. Their numbers had dwindled considerably and it was thought wise to put them under the care of a trained malacologist. By the spring of 1952, the number of Gonaxis had increased from the low of 22 late in 1949 to 439 living specimens. Under a joint agreement between the ICCP and the Hawaiian Board of Agriculture and Forestry, N. L. H. Krauss, entomologist of the board, was sent to East Africa, Australia, New Caledonia, and adjacent areas for a period of several months in 1950-52 to seek further into the problem of finding a suitable predator or parasite of A. fulica (Coolidge 1950, 1951, 1952). Additional live specimens of G. kibweziensis and Edentulina affinis were sent to the Board of Agriculture and Forestry (Krauss 1951-52, ICCP 1953, Lennox 1953). Besides these and specimens of the beetle Tefflus, Krauss sent specimens of several other predatory mollusks so that their potential as biological control wapons might be determined by Kondo. All of approximately two hundred specimens of the omnivorous Oxychilus cellaria from Sydney, Australia, died within four months; the cause is attributed by Kondo to the inability of this temperate zone snail to adapt to the summer heat of tropical Hawaii. A similar fate was suffered by eight specimens of Strangesta capillacea (Rhytida) and 45 specimens of Paryphanta compacta (Victaphanta), both from Sydney, 130 BIOLOGICAL CONTROL as well as eleven specimens of Rhytida spp. from the dense rain forests of New Caledonia. Kondo (1950a) reports that, of the latter, Rhytida inaequalis (Ouagapia) quite unlike the smaller predatory species attacked some of the large specimens of A. fulica. When Williams was in East Africa, he found that carnivorous snails belonging to the genus Gulella were "probably effective enemies of quite young Achatina" (1951, 1953). Their small size, however, eliminated them from consideration. Williams also examined the paryphantids in New Caledonia and found them frequenting moist, dense ravines. G. S. Dun (in litt. April 17, 1950) concluded from this that predator and prey might not overlap sufficiently to effect any appreciable control, especially since A. fulica tends to be more abundant in the less dense areas frequented by man. As the giant snail is fortunately not in New Caledonia, there is no way of determining what the predatorprey relationships might be. Recently, de Wilde de Ligney (1953) made the following interesting report on the spread of A. fulica in New Guinea: "It is believed that this snail is unable to cross the patches of jungle isolating the small farms from the town and from each other. This assumption is supported by the fact that a concentration of empty shells was often found in the outer regions of the forest surrounding the town. In a few cases naked snails, probably parasitic [sic!], were observed on the shells of the giant snail. Experiments will be made to ascertain the carnivorous habits of the naked snail." The Edentulina sent to Hawaii by Williams, including an additional 48 specimens sent by Abbott, maintained a steady decline in spite of plenty of food (young A. fulica) and moisture until in June, 1951, the colony numbered but eleven individuals (Kondo 1951b). At least in the laboratory, this species proved itself less hardy than Gonaxis. Perhaps the same explanation is behind the fact that Williams, Abbott, and Krauss encountered it in the field far less commonly than Gonaxis. At any rate, it is not difficult to understand why earlier tentative plans to introduce Edentulina in Agiguan, after the establishment of Gonaxis, were given up. In June, 1957, a fresh shipment of Edentulina was received from Mombasa, Kenya, and was released on Oahu; but to date there have been no recoveries (Thistle 1959a). At this point, it is appropriate to review the very interesting experiments of McLauchlan (1949), which concerned a biological study of the carnivorous snails Strangesta capillacea (Ferussac) and Helicella cellaria (i.e., Oxychilus) in Mosman, New South Wales. The fact that these snails will attack and consume the introduced Helix BIOLOGICAL CONTROL 131 aspersa is of particular interest in the present discussion. However, it is McLauchlan's belief that a diet made up largely or exclusively of H. aspersa causes a high mortality in these predators either because of excessive gorging or because of some toxic factor in the prey. He presents the following additional factors which tend to mitigate against the effectiveness of especially S. capillacea as a biological control agent: cannibalism among the predators, a primary preference for the native snails (species of Paralaoma and Egilomen), a secondary preference for only the young H. aspersa, the absence of a tendency for young Strangesta to attack young H. aspersa, the relatively very slow growth and development of the predator as compared with the prey, the general retiring nature of the predator, and the tendency for H. aspersa to move out of the range of Strangesta. Benthem J u t t i n g (1952b) reports on, but does not elaborate upon, a communication from J. Hope Macpherson to the effect that a similar study was conducted in New Britain to determine the predation of H. eellaria and an Australian Rhytida on the eggs and young of A. fulica. Kondo (1952) found in field observations of feeding Gonaxis that the largest specimens of A. fulica which would be attacked were almost never greater than 35 mm. in length. This immediately brings up the question as to whether or not the size of the predator is of any importance. G. S. Dun (in litt. Dec. 29, 1950) feels that a small predator is about as effective as a large one since the older specimens in his breeding cages were shown to be sterile. This is not a safe conclusion, however, as it has been found that dietary imbalance in caged specimens of the giant slug Ariolimax will invariably produce sterility and genital anomalies (Mead MS). The combination in A. fulica of a tremendous reproductive potential and immunity from attack after the first relatively few weeks of growth suggests at least t h a t Gonaxis is not an ideal predator and, further, implies that the role of Gonaxis might ultimately be reduced merely to that of consuming the multitudinous young which would have been eliminated anyway by other causes. It was reasoning of this sort which encouraged Mead and Kondo (1949) to recommend that if predatory snails are to be used in the biological control of A. fulica the larger snail predators should be considered, for example, the giant paryphantids of Australia. Subsequently it was pointed out (Mead I950b, c) that, since in predation the degree of prey-specificity is not as high as is host-specificity in parasitism, the search for a suitable predator of A. fulica need not and should not be limited to the area where this snail pest is indigenous, viz., East Africa. That is, merely because the predator has the same endemicity as its prey, it does not 132 BIOLOGICAL CONTROL necessarily follow that no better predator can be found. The field work of Krauss and Williams, the observations of Dun, and the reports of Kondo all indicate that the giant predatory paryphantids of the temperate regions are not as promising as hoped. Their inability to adjust to tropical climate suggests automatically that a large tropical molluscan predator might be tried. None is found in the area now occupied by A. fulica. The large species of the voracious, tropical American Euglandina would seem to be worth serious consideration if biological control is to be investigated further. The eggs of Euglandina are proportionately gigantic and the hatching individuals are therefore so large that by far the majority of the endemic molluscan fauna of the Pacific islands, consisting of small or minute species, would escape their ravages. In addition, the juveniles and adults could attack larger specimens of A. fulica, eat more, and live longer than the small endemic predators of East Africa. It is of interest to note that Euglandina was imported into France nearly fifty years ago in an attempt to control snail pests of truck crops and was said to confer "immense benefits on market gardeners" (Anon. 1913). Euglandina rosea (Ferussac) of our own Gulf States area is well known for its vicious attacks upon other snails (Pilsbry 1946:2:1:188). As will be seen below, this relatively small species has already been introduced into Hawaii. Muma (1954, 1955) has observed it feeding upon a reputedly beneficial tree snail Drymaeus dormani (Binney) in the citrus groves of Florida (cf. Norris 1952, Nunn 1953). He reports unenthusiastically about it, though, with these words, "Experiments with Euglandina . . . have demonstrated that it is not a heavy feeder and probably not of great importance." The only other frontier in the search for a molluscan predator is in Mauritius. Both A. fulica and A. panthera have been introduced on that island (Dupont 1878) and there is recent strong evidence that in the ensuing competition the latter species is quite successfully taking over in the areas below 1000-1200 feet in altitude whereas A. fulica remains dominant in the area 1200-2000 feet (J. Vinson in litt. Dec. 24, 1949). Deep in the interior, where significantly the achatinas have not yet successfully penetrated, there are found about thirty native species of predatory streptaxids. Efforts by G. S. Dun and the author to stimulate by correspondence a scientific investigation of the relationships between the achatinas and the streptaxids have failed. Coolidge (1951) reported that a conchologist, Dorothy Getz, planned to visit Mauritius and would be given assistance to permit her to obtain information there about the problem of the BIOLOGICAL CONTROL 133 giant African snail. Her findings, however, have thrown no new light on the possibility of natural biological control being in operation in Mauritius (Coolidge 1952). On June 3, 1952, while Kondo was still in Agiguan in the process of attempting to determine among other things the nature of precautions which might properly be taken in introducing Gonaxis elsewhere, the Hawaiian Board of Agriculture and Forestry made a decision and took action which has stimulated a considerable amount of controversy. The preliminary results of the trial field experiment in Agiguan and the laboratory studies in Hawaii, coupled with extreme pressure from the people of Hawaii to "do something" about the new outbreaks of A. fulica in Oahu and Maui, persuaded the Board to introduce Gonaxis kibweziensis into the Hawaiian Islands without waiting for additional results from the projected program. Twenty specimens, along with ten specimens of the predatory beetle Tefflus, were released near Mahinui in the Kaneohe district of Oahu, where the giant snail has extended its range considerably (Lennox 1953, Weber 1953). Through a singularly unfortunate oversight, however, stones covered with a cement-lime-calcium arsenate bait, scattered there to control the giant snail, proved equally tempting and fatal to all of the introduced Gonaxis specimens. A second introduction of 200 specimens of Gonaxis was made in this same area in September, 1954. At the time 300 specimens were introduced only a relatively short distance away, near the insane asylum. In January, 1955, this latter site was inspected by the author and there was found to be every evidence of successful establishment of this predatory snail. It was the plan then to move 100 of the specimens from that site to a third location on the east side of Oahu as an initial step in a projected plan to spread Gonaxis as rapidly as possible. A few weeks later, specimens were released on Maui (Weber 1956). In 1953, efforts were made to introduce Gonaxis from Kenya into the achatina-infested Seychelles Islands; but because of a prevailing precaution on the part of the administration, the plans were tabled indefinitely. During the period January 12-16, 1954, G. D. Peterson (1954) accompanied by three assistants, conducted the fifth expedition to Agiguan. He concluded that considerable changes were taking place in the various animal populations on that island, that Gonaxis was apparently exacting a greater toll on Achatina than estimated by Kondo, and that if the later point could be proved during a subsequent expedition, immediate steps should be taken to establish Gonaxis on other infested islands. From July 21 to August 11 of that 134 BIOLOGICAL CONTROL same year, Davis (1954), accompanied by Peterson and two assistants, conducted a follow-up survey with the results essentially re-emphasizing and elaborating upon Peterson's earlier conclusions. The elaborations underscored the complexity of the ecological conditions which were obviously in a significant state of flux. The most important announcements were: That the 100 Gonaxis released two years earlier by Kondo had increased to an estimated population of 80,800; and that the ca. four hundred Gonaxis released four years earlier, and which had built up to an estimated 21,750 in two years' time, had died out almost completely. In September, 1954, 88 specimens of Gonaxis from Agiguan were marked with plastic paint and introduced by Peterson into Guam (Peterson 1957, b). A few months later, young specimens of the third generation were found, indicating that successful establishment had taken place. By this time, the authorities in charge of the biological control program in the Pacific area were convinced that a full scale "Gonaxis Program" should be undertaken. Accordingly, a seventh expedition to Agiguan was planned for November, 1955, with the specific mission of obtaining as many Gonaxis specimens as possible for introduction into other islands of the Trust Territory, into other areas on Oahu and Maui, and into California as a biological control agent for two introduced helicine pests (Mead 1955c, Anon. 1956b, c). Reportedly, over 5,000 live Gonaxis were collected (Coolidge 1955, Kondo 1956, Anon. 1956a). Of these, 2,000 were sent to Hawaii and 200 were sent to California. The remaining specimens were divided into lots and released in one or more sites on Saipan, Tinian, Rota, Ponape, Truk, and the Palaus. Subsequently, specimens were introduced in New Britain. Introductions into still other areas soon followed, as implied in the words of Pemberton (1956), ". . . our Committee has been besieged from many sources requesting colonies of Gonaxis." G. kibweziensis, however, still has as its main drawback the fact that its small size restricts it to attacking only the smaller achatina specimens. In an effort to improve the situation, the Hawaiian Board of Agriculture and Forestry introduced on Oahu in June, 1957, and subsequently on Maui, the east African Gonaxis quadrilateralis (Preston), which is almost exactly twice the size (Thistle 1959a). Two years after its release, it was still holding a weak second place; and, in general, the slow, uncertain developments were disappointing. But even before this, more and more attention had been turned to Euglandina rosea as a third possible molluscan predator. A number of live specimens from Florida, collected by Krauss, were sent to BIOLOGICAL CONTROL 135 Hawaii in 1955 as the initial step in introducing this species as a biological control agent. Their introduction was speedily approved and 616 specmens were released later that same year in Oahu (Weber 1956, Kondo 1956). In contrast to both species of Gonaxis, this aggressive, vigorous, rapacious snail took hold immediately and spread rapidly. By 1957, it was considered clearly the most promising of all biological control agents being used in the giant snail control program. In addition to feeding on Achatina, it was found to feed on the snail pest Bradybaena similaris and the liver fluke snail, Lymnaea (Fossaria) ollula Gould (Thistle 1959a). In both Makiki and Hauula release sites, the snail had moved far afield, even into Gonaxis release sites. Ironically, Gonaxis have been found in the egg clutches of Euglandina. By July, 1958, the Hauula population had built up to the point where it was possible to remove 12,000 specimens for release in many other sites in Oahu, Kaui, Maui, Hawaii, New Guinea, Okinawa, and the Palau, Philippine, and Bonin Islands. In January, 1956, 25 specimens of a fifth molluscan predator, Oleacina oleacea straminea (Deshayes) from Cuba, were released on Oahu (Kondo 1956, Thistle 1957). About four hundred specimens of the predatory snail Gulella wahlbergi (Krauss) from South Africa were released in January, 1957, on Oahu and subsequently others were released on Maui. A number of specimens of Gonaxis vulcani Thiele and an unidentified species of Gulella, both from the former B. Congo, were released on Oahu in November, 1956 (Thistle 1959a). There is no evidence to date that any of these four predators has become successfully established. Similarly, little hope is held that the giant predatory snail Natalina sp. (i.e., Rhytida), under examination in Hawaii, will prove to be effective enough to warrant its release, in spite of the fact that its body attains a length of 7-8 inches. Another possible candidate, Varicella similis from Jamaica, died in the laboratory before it could be decided that it might be released on Oahu. Gulella bicolor Hutton from the Philippine Islands proved to be of negative value so far as preying upon Achatina is concerned; but it proved to be such an effective predator of the snail pest Subulina octona, which is also the host of the cecum fluke of poultry (Postharmostomum gallinum), that it was released in large numbers on Oahu and Hawaii in 1957 and 1958 (Thistle 1959a, Davis 1959). Those absorbed in the economic aspects of the problem have judged this expanded program of predatory snail releases as a strategic, timely, and reasonably safe one, although some naturalists have condemned it as unwise, premature, and irrevocably threatening to the unique native mollusk fauna of the Pacific Islands. For example, 136 BIOLOGICAL CONTROL Fosberg (1957) deplores the "almost hysterical program of importations of predators in an attempt to control the giant African snail" in Hawaii and concludes that "these will almost surely bring about destruction of many members of the extraordinary Hawaiian land snail fauna." It undoubtedly will take years to determine positively the actual value of these irreversible introductions. One thing is certain. There is to date absolutely no scientific justification for such a statement as, "The snail [Gonaxis kibweziensis] ultimately proved very effective in the control of the giant African snail" (Peterson 1957b). M i s c e l l a n e o u s In New South Wales, McLauchlan (1949) reports that unidentified "tiny insects in the soil also destroy the eggs" of the predatory snail Strangesta capillacea (Ferussac). B i o l o g i c a l Control--an Evaluation South (1926b) was probably the first to consider seriously the possibility of attempting to control A. fulica by enlisting the help of some of its "natural enemies" from East Africa. He accordingly made contact with authorities in Kenya, Uganda, and Tanganyika; but for some unexplained reason, the program died in its tracks. Since then, the subject has been alluded to from time to time; but it was not until Townes (1946) made his survey of the problem in the Pacific that the matter clearly came out into the open once again with the statement that "the most feasible control for it is to find and introduce a natural enemy." This recommendation was picked up as a keynote by the Pacific Science Board of the National Research Council and subsequently by its Insect Control Committee for Micronesia (Ryerson 1947, ICCM 1947). In this manner, it has influenced to a very great extent the character of the research on the giant snail in the Pacific. The first major undertaking in the investigation of the biological control of A. fulica was the exploration of the native heath of this snail in Kenya by Williams (1951, 1953). The early reports that no parasites but several predators were found brought definitely mixed reactions (ICCM 1948), the record of which might be considered a crystallization of the basic philosophy underlying the current biological control program. Had specific parasites of the giant snail been f o u n d , the method of undertaking the project would have been clearly indicated. But with only non-specific predators, some felt that at least a fair amount of precaution was needed. It was this feeling t h a t caused Rees (1950) to announce that "the American authorities are reluctant to release these predators in new countries lest they should become greater pests than Achatina." Since that time, however, the program of biological control of the giant African snail has GIANT A F R I C A N SNAILS Representation of the basic environmental interrelationships of the giant African snail and its primary associated organisms. Solid lines represent lines of feeding, predation, or attack; broken lines represent a poisoning effect upon the feeder or predator. Each newly introduced organism or factor inordinately increases the complexity in the existing interrelationships, particularly through ecological chain reactions, and it may be responsible for reversing completely the definitive beneficial or deleterious effect of any other organism or factor. (Drawn by D. B. Sayner.) 138 BIOLOGICAL CONTROL become well established. Not one, but nineteen predatory invertebrates have been introduced in the various infested areas. And more apparently are to follow, as there are currently several others under consideration. As this program continues to develop, there still prevails a mixed feeling as to the element of risk involved. All agree, however, that this so-called "natural control'' is exceedingly attractive from the standpoint of its being self-perpetuating. Any other method of control that has been devised requires maintenance; and, at best, maintenance on an indefinite basis is a serious consideration because of the cost, either in money or man-hours of labor, or both. The literature is replete with reports on the general subject of the dangers of introducing alien species (e.g., Kew 1893; Thompson 1922; Schlesch 1928; Anderson 1934; Storer 1931, 1934, 1949; Hanna 1948; Wodzicki 1950; Rees 1955; and Heim 1956). On the specific subject of the introduction of alien animals to control the giant snail, Aguayo (1950), Clench (1949), Jaski (1953), Mead (1949b, c, d, 1950b, c, 1955a, b, c, 1956b, c), Morrison (1950b), and others have sounded a warning. Lennox (1949) and Pemberton (1956), in turn, have minimized the dangers and emphasized the importance of cont i n u i n g with the biological control program. In many ways, Morrison (1950b) comes close to the heart of the dilemma with the following words: "The man who introduces a new predator into a country undertakes a grave responsibility. Before it is regarded as safe, it must have been proved conclusively that the candidate for introduction will rather die than touch native fauna or flora or local economic products. Such proofs take a long time to establish, and meanwhile the giant snail is eating its way through the Pacific." As pointed out earlier, the predatory snails Gonaxis kibweziensis and Euglandina rosea have taken the limelight almost entirely in the program of the biological control of A. fulica. Contrary to various current reports (Anon. 1956b,c) y there is no evidence that Gonaxis effects any real control of the achatinas in East Africa (Abbott 1951c). Similarly, the several endemic predators of A. fulica in Ceylon have been reported to be of relatively little value as "natural controls" (Fernando 1952, Mead 1955b, I956d). But this seems to be the usual story. For example, Ingram (1942a) was able to determine that the native carnivorous snail Haplotrema minimum in California had an inappreciable effect upon the introduced H. aspersa even though it preyed freely upon its young. An explanation for this phenomenon seems fairly simple. It is probable that the foreign species has a biological advantage in that it is in an environ- BIOLOGICAL CONTROL 139 ment lacking its own specific natural controlling factors, while the endemic predators continue to be held in check by their own natural controlling factors, despite the increase in acceptable prey. This explanation becomes more tangible if it is assumed that the more effective ''natural controlling factors" are specific disease agents that become proportionately more limiting in their effect as the host population increases. Although the idea has not been presented in support of the predatory snail program, it would appear, by extending this same reasoning, that if in a given environment both predator and prey are recently introduced foreign species the predator would be at an advantage. If under these conditions, however, disease of either the predator or prey should enter the ecological picture, it could be completely decisive in its effects, one way or the other. On the other hand, if disease of both prey and predator should enter the picture, the outcome obviously would depend upon the differential of such factors as the severity of the disease, immunity, tolerance, physical hardiness, and a whole host of other factors. But disease, as a complicating factor, has scarcely entered consideration in the "Gonaxis Program," in spite of its importance. Even without it, the picture is far from simple despite the glib accounts, especially in the more popular articles and newspaper accounts, which suggest that after consuming all the achatinas Gonaxis will t u r n cannibalistically upon its own members and finally exterminate itself (e.g., Mahoney 1955). As an aside, Bodenheimer and Schiffer (1952) have convincingly demonstrated mathematically why there almost never is extermination under natural conditions, even in parasitism. Durham (1920), however, puts over the point more humorously with the following: "I may point the tale with reference to the flea and the louse. With great and enduring pertinacity their 'natural enemies/ the monkey, the dog and cat, nay even on to mice and men, are hunting them day by day and into the night season, but they still abide in their haunts." Some have pointed out that the predicted predation-to-extinction and self-predation-to-extinction already have taken place on Agiguan Island since all Achatina and all Gonaxis have been found dead in the area of the first release of the predator. If that is so, how then can one explain the similar disappearance of Achatina in other areas on that island where Gonaxis has not penetrated? (cf. Davis 1954:15, 23) First or all, it should not be forgotten that animal populations under adverse conditions may apparently disappear, only to reappear months or even years later. Hence, merely because specimens could not be found, it cannot safely be assumed that extinction has taken 140 BIOLOGICAL CONTROL place in either population. The question, however, does emphasize the fact that in the present case control factors other than Gonaxis are effectively operative. If we assume that both Gonaxis and Achatina have not become extinct, but have been reduced in numbers beyond the point where they can be encountered in the field, we should naturally turn our attention to the factors influential in bringing about a recovery of the populations of both prey and predator. Some of the more obvious of these factors are: general hardiness, versatility in appetite, intraspecific cannibalism (Kondo [1951b] has shown this often to be severe in Gonaxis), longevity, duration of estivation, reproductive potential, and adaptability to varying types of environment. In every case, Achatina has a distinct advantage. From this it can be assumed that at least Achatina would make a faster "comeback." As still another facet in the complex ecological setting in which we find the Gonaxis-Achatina problem is the role of the endemic snails. Kondo found that Gonaxis shows a definite preference for several species of terrestrial and arboreal endemic snails in Oahu (1950b, 1951b) and for Omphalotropis erosa on Agiguan (1952). As long as the arboreal snails remain in the vegetation they will be safe, as Gonaxis manifests little tendency to leave the ground. The terrestrial snails, however, would surely suffer (Anon. 1956d). This is regrettable; for the terrestrial snails "because of their secretive habits form precious 'keys' indeed for unlocking the vast zoogeographic storehouse of the Pacific" (Mead 1955c). The question has been raised regarding the possibility of Gonaxis t u r n i n g to some other prey, such as earthworms, after the achatinas and the endemic snails have become scarce. Although other carnivorous pulmonate gastropods are known to consume earthworms, Kondo (1950b) failed to demonstrate in the laboratory any tendency of this sort in Gonaxis. Nothing is known about the extent to which other invertebrates might be attacked. There is strong evidence, however, that certain other terrestrial invertebrate populations may undergo substantial change because of the indirect effects of introducing Gonaxis. The reports of Peterson (1954) and Davis (1954) show that four years after Gonaxis was introduced on Agiguan, the hermit crabs (Cenobita perlatus] decreased in numbers markedly, whereas the coconut crabs (Birgus latro] definitely increased. Concurrently, the vertebrate populations went through corresponding changes. The monitor lizard (Varanus indicus] went into a definite decline; the rats virtually disappeared; and the ferral goats increased in a very pronounced fashion. The plant communities also under- BIOLOGICAL CONTROL 141 w e n t changes, a more significant one of which is the serious increase of the poisonous plants Jatropha gossypifolia and /. curcas (Peterson 1954). In attempting to explain this whole series of changes, one finds one's self face to face with the exceedingly ramified and little understood subject of ecological chain reactions. Some notable examples of ecological chain reactions have been presented above. A general treatment of the subject as it concerns the biota of the islands of the Pacific is to be published soon (Mead in MS). Because of the high state of flux of the fauna and flora of Agiguan, some have felt that ecological chain reactions are actually in progress and that greater changes are in store before an equilibrium once again is attained. Unfortunately, however, what has happened so far has not met with unanimity of interpretation by the several investigators. For example, when Enders, Kondo, and Mead made the initial survey of Agiguan, there was a manifest abundance of rats; Kondo announced later that the rats in combination with the coconut crabs were killing more giant snails than the predatory snails were; but when Davis and Peterson arrived on the scene, they found such an "extreme scarcity of rats" that they concluded Kondo had "weighed the evidence too strongly in favor of the rats" (Peterson 1957b). So far, Gonaxis has demonstrated twice on Agiguan that in two years it can go from a few dozen individuals to populations estimated to be in the thousands. It is not clear, though, just what takes place following the population buildup. The first population went to an estimated 21,750 in two years; but in an additional two years, practically no living individuals could be found during two different intensive searches. The second population, in another area on that island, went to an estimated 80,800 in two years. At its apparent population peak, over 5,000 individuals were recently removed for introduction into other areas. If this second population is due to continue its increase, the removal of 5,000 individuals should not have any appreciable effect; if however the population has gone into a decline, as the first one had at this stage of its development, then there could conceivably be a very material effect. (It is apropos to recall that hand collecting as a control method for A. fulica is most effective after the population has gone into the apparently inevitable "decline.") But we dare not extrapolate from one population to the other; for the populations are not only in different areas, but the populations of the other animals on the island have continued to undergo such pronounced changes that they are bound to cause different repercussions in this second population of Gonaxis, and even in the first population, if it is still recoverable. 142 BIOLOGICAL CONTROL The "Euglandina Program" similarly is producing in some cases truly remarkable and even enigmatic results. For example, Euglandina has been in the Makiki area on Oahu for two years, yet in some of the sections where the giant snail is most abundant, this predator, along with G. kibweziensis, is surprisingly only seldom encountered. Specimens that have been collected more often than not are emaciated in spite of the abundance of food. Adult Gonaxis have been found consuming the egg clutches of Euglandina; but on the other hand, Euglandina in the laboratory will often select Gonaxis and the small, introduced snail pest Bradybaena similaris in preference to the giant snail. In some cases, Euglandina steadfastly refused to attack an acceptably small Achatina even after fasting two to three weeks. In other cases, specimens which did consume giant snails in the laboratory mysteriously died within a day or two in contrast to others in the same lot that refused to feed. This brings to mind McLauchlan's suggestion of a "toxic factor." The emaciation has suggested the possibility of disease. Individuals have been found with leucodermic lesions apparently identical with those of "diseased" giant snails. It is possible that Euglandina is more susceptible to the disease than Achatina. Another differential between these two species is the giant snail's i n f i n i t e l y greater capacity to withstand prolonged estivation with a consequent lack of food and water (Mead 1959b). A protracted dry period could therefore enforce estivation and cause death through starvation in spite of living food being nearby in abundance. Essentially nothing is known of the population dynamics of Euglandina and it is altogether possible that this cannibalistic species, as is characteristic of predators in general, will not maintain high populations in limited areas. This may explain why the highest concentrations of this predator have consistently been found at the periphery as an advancing crest in a relatively rapidly expanding population. Threading through this entire problem as an element of still further uncertainty is the genetic factor. The study of genetics teaches us that each species has its own spectrum of hereditary traits and rate of genetic mutability. From the study of population genetics we learn that different populations of the same species may have diff e r e n t complements of the total possible hereditary traits. In some cases, the differences are great enough so that two populations of the same species may be considered distinguishable as two subspecies. The selecting factors operative in the different environments quite understandably vary in their relative value. It is the sum total of the effects of the selecting factors in a given environment that determines FIG. 13.--A full-grown larva of the India glowworm, Lamprophorus tenebrosus feeding on a ca. 75 mm. specimen of the giant snail. FIG. 14.--Achatina fulica with two of its natural enemies from East Africa, Edentulina affinis (ca. 40 mm. long) and Gonaxis kibweziensis (ca. 20 mm. long). FIG. 15.-- California Plant Quaranitine inspectors in Los Angeles discover giant snails in a cargo of war salvage material from Guam. BIOLOGICAL CONTROL 143 which available hereditary traits will be selected and perpetuated as most suitable in the population in that environment. In the ten to fifteen years that A. fulica has been in the western Pacific Islands, it has displayed such a remarkable variability from one population to another, with respect to size, color, thickness of shell, and even as to what plants it will preferably attack, that a taxonomist, unacquainted with the history of this species would be tempted to designate the populations at the very least as those of different "races" (Mead 1951a, 1955b). When these same thoughts are applied to the "Gonaxis Program," we cannot but wonder what this predatory species will do in a like period of time. It has been suggested that even a greater versatility in appetite in Gonaxis is not outside the realm of possibility (Mead 1955, 1956b), notwithstanding the remarks in the literature to the contrary (Pemberton 1956). This carries the disquieting implication that the environment might well select unsuspected and undetectable undesirable traits. There is no way in which this risk can be eliminated; for even though the animals may be "screened" in the laboratory for their feeding proclivities, there cannot be duplicated in the laboratory the population dynamics or the multiplicity of environmental factors operative in a natural population. Therefore, one may not really safely conclude that, merely because all of, say, 2,000 experimental animals refused to eat "Alpha" and died when it was their only source of food, "Alpha" will not be jeopardized in any way with the release of the experimental animals. The 2,000 experimental animals may have been too small a sample to express the full spectrum of genetic variability in character determiners influencing feeding habits. But grant that this was an adequate sample. How can one anticipate ecological chain reactions which might threaten "Alpha" indirectly? Actually, there have been developed in insect biological control, methods which substantially reduce the greater share of the elements of risk. But these methods demand such an inordinate amount of precious time that, where the life cycle is long or the economic pressure is great, compromise or short-cut measures may be adopted. Such measures have been strongly opposed. Some have questioned the advisability of introducing multiple predators of the giant African snail in Hawaii. To date, there have been introduced nine predatory beetles, two presumably parasitic flies, and eight predatory snails. In insect biological control, the concurrent introduction of two or more parasites of an insect pest is a common practice (e.g., in the fruitfly control program in Hawaii). The rationale behind this practice is that if one parasite does not 144 BIOLOGICAL CONTROL effect a control perhaps another one will; or, together, the multiple parasites will do the required job; or, one parasite may take over in one type of environment and another will take over in a second type of environment. Although these thoughts make good sense, putting the idea into practice introduces new difficulties. In the case where predators instead of parasites are introduced, predators may consume predators, even in preference to the intended prey. Although Peterson (1957 a, b) reported that in the laboratory Lamprophorus would not attack Gonaxis it has been pointed out that Tefflus can consume Lamprophorus, Lamphrophorus can consume Euglandina, Euglandina can consume Gonaxis,, and Gonaxis can consume its own kind (Mead 1955b). In addition, a number of the factors, discussed above in terms of the "Gonaxis Program" and the "Euglandina Program," would also pertain to these other predators. The complexity becomes geometric in its proportions. Still other difficulties are introduced when different methods of control are concurrently used. Molluscicides and other agents of snail destruction will kill predatory snails quite as quickly as they will A. fulica. This was convincingly demonstrated at the time Gonaxis was first released in the Hawaiian Islands; and it surely must have been demonstrated again when both Euglandina and metaldehyde were used in concentric, contiguous zones of defense in the Mana district of Kauai when the giant snail was first discovered on that island. In the same manner, insecticides used in the field are a threat to the beetle predators of the giant snail. As an apparent "double threat," some snail baits are fortified with insecticides; in other cases, the molluscicides are known to be insecticidal in their effects. And, as a provocative additional thought, there was evidence in Ceylon that a snail dying of metaldehyde poisoning provided a temptingly lethal morsal for the marauding Lamprophorus. It is perfectly obvious at this point that, despite the attractive features of molluscan biological control, the method has more than its share of difficulties. If true specific parasites of A. fulica can ever be found, a great share of the intercurrent problems will be removed automatically. The recent works of Muma (1954) and Berg (1953, 1955) on malacophagous fly larvae are also encouraging; but Berg's reference to the fly larvae as "relatively unselective snail predators" seems to sound a familiar note of warning. The work on diseases of snails (vide infra) has only just begun (Mead 1956a); and, although it is exciting to contemplate, it is obviously too early to do more than speculate as to how much promise micro-organisms offer as potential biological control weapons. A BIOLOGICAL CONTROL 145 factor distinctly in favor of micro-organisms is their characteristically high host specificity. But at least in one outstanding respect, microorganisms are potentially more dangerous than the metazoan predators and parasites. It is agreed that their phenomenal biotic potential insures the capacity under favorable conditions to build up rapidly a population capable of causing an epizootic. Ironically, however, it is this same biotic potential which increases the chances of there appearing a mutant type whose metabolism is just different enough to produce undesirable results. The story behind the use of the myxomatosis virus in the biological control of rabbits in Australia is a case in point, particularly with the subsequent appearance of nonvirulent, but immunizing, strains in the field (Ratcliffe et al. 1952, Fenner and Day 1953, Mykytowycz 1953). It is felt, however, that in the final analysis it is truly the disease-producing agents in an environment which determine both qualitatively and quantitatively the biota the environment will maintain (Mead 1955b). In other words, when organisms are in the presence of, and in harmony with, the agents that produce disease, their numbers will not be excessive ecologically and there will be at least a semblance of endemicity. The mission of this phase of biological control, then, is either to bring the foreign species into reasonable balance with the other organisms in the environment through the introduction of a natural diseaseproducing agent, or to attempt to eradicate the foreign species by introducing an unnaturally virulent disease-producing agent. This mission is a big one and it will not be fulfilled until the biological control man, the malacologist, the microbiologist, and the ecologist all work together on a closely integrated program. Together, they should be able to speed up and intensify the natural process of population decline which sooner or later overtakes an invading species. This decline, it is felt, is nothing more than an indication that the disease factor has finally caught up with the invading species. Disease is the ultimate controlling factor; and it is one that must be recognized and properly evaluated in the problem of the giant African snail. CHAPTER 8 CONTROL THROUGH HUMAN USE Human Consumption The human consumption of land snails has been practiced since the very earliest times, and even during the height of the Roman Empire it was of such common practice that special utensils were devised so that the soft parts of the snail might be more easily extracted and eaten (Taylor 1900). The demand for the edible helicine snails has been so great that "heliculture" or snail farming, especially in parts of Europe, is today a very sizable industry. Not only have snails been raised in almost unbelievable quantities for shipment alive to domestic and foreign urban areas, but with the modern facilities for refrigeration and quick shipment, snails are prepared with butter and spices ahead of time at the snail farms so that all the consumer has to do is to warm the "instant snails'' for a few minutes in the oven before eating them. Their very high nutritive value has been expounded upon by Leger (1925) and Moretti (1934). Simon (1940) has pointed out that they are neither fish nor flesh and their consumption is permitted on meatless days. As might be expected, there has been much written, in several languages, on every conceivable phase of raising, preparing, and marketing edible snails. Among the most notable contributions are those of Locard (1890), Rust (1915), Boisseau and Lanorville (1931), Arnould (1933), Kaibo (1935), Ri (1935), Maubert (1943), Pardo (1943), Metteo (1946), Ben them Jutting (1952a), and especially Cadart (1955). Rust urged the American people during the trying periods of World War I to become more adventurous in their eating and, 146 CONTROL THROUGH HUMAN USE 147 as a substitute for meat, partake of the edible helicine snails already introduced into parts of this country. Even the canning of pickled snails was encouraged (Anon. 1914). In France, Remlinger (1916) made a more practical suggestion by recommending that snails be used instead of precious beef in preparing bacteriological culture plates. The Second World War brought a return of the suggestion that they be used for human consumption (Aguayo 1944b). The fact that native gastropods were already being eaten by people in this country was pointed out by Allen (1916) in this reference to the consumption of the giant slug Ariolimax in parts of Oregon and Washington and Triodopsis albolabris in Massachusetts. Earlier, de Stefani (1913) reported that great quantities of Helix pisana (Theba) were being sent from Sicily to the United States. (It is significant that this species later became a serious pest in California.) Some idea of the amount of helicine snails consumed in this country is indicated in figures released by the U.S. Bureau of Entomology and Plant Quarantine wherein 721 cases and 24,969 baskets of living snails were imported into New York City alone between May, 1947, and April, 1948 (Mead 1951b). It has been stated that for a number of years almost exactly 1,000,000 pounds of snails has been imported annually into this country just from Morocco (Heifer 1949). With approximately fifty snails to the pound, the total number of snails would figure close to 50,000,000; or, in linear measurement, close to a thousand miles of snails. Who would have guessed that such an appetite for snails existed in this country! In France, however, the amount of snails consumed is understandably about fifteen times as great (Cadart 1955). In British West Africa, various species of Achatina and Archachatina are eaten to a very great extent. In Ghana, these snails actually form the largest single item of animal protein in the diet of the common people. It was in that country that the author saw the natives eating giant snails which had been put in boiling water or on hot coals, removed from the shell, freed of the soft visceral mass, chopped, and combined with some starchy dish, such as cassava or cocoyam, and palm oil or peanut oil to form their standard meal of " f u f u . " Lang (1919) similarly describes the methods used by the natives in the former Belgian Congo and adds further that the snails are occasionally served at the tables of Europeans. He did not indicate, however, how well they were received in the latter case. During the dry season, the snails tend to go into estivation and they become very difficult to find. They are therefore collected in great quantities during the rainy season. To assure a supply throughout much of 148 CONTROL THROUGH HUMAN USE the year, a large portion of the catch is dehydrated. The shells are broken to remove the animals, which are eviscerated, washed in water but a single time, drained, impaled on split bamboo sticks, smoked and partially dried for six hours over a slow fire of wood embers, and, finally, completely dried in the sun before they are stored (Martinson 1929, cf. Dartevelle 1953). For many years now, such inroads have been made on the native achatinas of West Africa that their collections have had to be regulated by the colonial forest conservation laws, through the co-operation of the local native chiefs. To maintain jurisdiction over the collecting areas is a sovereign right of the chiefs; they set the collecting dates and determine what percentage of the catch is to be given to them as tribute. The increasing scarcity of achatinas has forced a greater and greater limitation of the collecting period. This in turn has put a high premium upon making use of every available moment during the short collecting period. The seriousness with which the natives regard this matter is reflected in the words of G. Saunders of Kintampo, Ghana, who writes (in litt. Feb. 6, 1952) that "the date of a baby-show had to be postponed because it coincided with the opening of the snail-collecting season, and all the mothers go snail hunting!" Conolly (1939) stated that A. fulica in its native heath of East Africa is regarded as a ''culinary delicacy." In the region of Kenya and Zanzibar, however, Williams (1951) was unable to confirm this in the inquiries he made. This may help explain why Williams, Abbott, and Krauss found the native species of Achatina considerably more abundant than Bequaert and the author found in certain comparable regions of central West Africa. Torres (1950) has indicated that the native South American giant land snail, Strophocheilus oblongus, known as "Arua," is consumed by the aborigines in that region. Ernest G. Holt, of the U.S. Soil Conservation Service, stated to the author that years ago he actually saw indisputable evidence of the fact that the natives eat this snail. As far as its use for human food is concerned, it quite likely is comparable in many ways to the giant African snail. Monte (1944) and Anon. (1945) report on other snails used in South America as food and medicine. It was known before World War II that the Malays and Chinese would eat A. fulica (Jarrett 1923, 1931). This knowledge, coupled with the very rapid spread of the snail during the Japanese occupation and the persistent rumor that the Japanese ate the snails, apparently was responsible for the appearance in the literature of the assumption that the Japanese people considered the giant African CONTROL THROUGH HUMAN USE 149 snail a "great delicacy," for example, Abbott (1948), Anon. (1948b, 1949f, g), Herklots (1948), Rees (1951), Somanader (1951), and Zuk (1949). As doubt about this increased, efforts were made to soften it with the suggestion that the Japanese ate them, but with wry faces (Anon. 1949j). Rumors from other quarters carried the story that the Japanese people actually disliked the snails and would not touch them at all. This was elaborated with the suggestion that the Japanese soldiers introduced the snails into the islands for sheer spite. It was perfectly obvious at the time that the problem of the giant African snail was surveyed in Micronesia that this whole question had to be elucidated with as much supporting evidence as possible. The information which was gathered was almost entirely hearsay; but the story from the most reliable and probably least biased sources was always the same, viz., that the Japanese very definitely did consume the giant snail. It was in Chichi Jima that the bulk of the information was obtained. Moses Savory of that island stated that he had not only seen the Japanese soldiers eat the snails, but he himself had eaten them. He described how the achatinas were roasted in their shells, removed, washed in two changes of water to clear them of most of the slime, eviscerated, and eaten with soya sauce; that is, essentially only the muscular foot was eaten. He said that they were something like limpets and were "not bad," though his lack of enthusiasm in relating the details and the expression on his face seemed to tell a somewhat different story. His brother Hendrick added that it was his belief that the Japanese soldiers "liked" the achatinas, that they were especially fond of soup for breakfast and that they would often prepare "snail soup" for that first meal of the day. Wilson Savory and Mrs. Oste Webb, also of that island, concurred with details of their own observations. The native skipper of a boat that took the author from Moen to Dublon, in the Truk Islands, stated through an interpreter that the Japanese people had eaten the giant snails on Dublon. According to him, they were heated in the shell, removed, cooked again, eviscerated, and eaten with soya sauce. The people of Moen gave a similar story. In Guam, Mr. Joaquin Guerrero of the Naval Government Agricultural Experiment Station turned over to the author a copy of a letter which he had addressed to the head of the Commerce and Industry Department in Guam on December 26, 1945. The following quotation from the letter, made with Mr. Guerrero's kind permission, presents earlier convincing evidence: "Mr. Jose L. Shimizu further stated that while he was imprisoned at the stockade, Island 150 CONTROL THROUGH HUMAN USE Command, he overheard statements made by the Japanese prisoners to the effect that while they were hiding in the woods, they sometimes ate these snails." While in Koror, Palau Islands, Kondo (1950c) obtained the following information on this subject: "The African snail . . . was introduced by an enterprising Japanese just prior to the war to be raised for canning. A small factory or 'koba' was constructed and Okinawans were employed to gather and to process the snails. The informant stated that the canned product was excellent eating but he himself would not take the trouble of gathering the living snails to prepare them for the table. With the advent of the war the . . . factory closed down. It is not known whether the canned snails were exported to Japan but presumably they were since the objective of most of these enterprises was to send the finished product back to the mother country. It is also not known how the snails were prepared for canning but because most of the shellfish were cooked in soysugar sauce the same method was probably applied in this instance." Additional information from Saipan is reported by Jones (1949). He states, "Japanese soldiers consumed them in quantity when normal meat supplies were short. It is reported by reliable native sources that these snails were dried and shipped in large quantities to the Japanese homeland during the war. The price was equivalent to five United States cents per pound." As one seeks deeper into the reasons behind the importation and consumption of the giant African snails by the Japanese, one frequently finds that the presumed great medicinal properties of the snails have provided the motivation. But this is not something invented by the Japanese. Taylor (1900:428) not only discusses the possible curative powers of snails in ridding the body of tuberculosis but recommends a method of puncturing the snail shell "to enable the patient to suck the oozing liquor." Clapp (1902) similarly refers to slugs in coconut milk as a cure for asthma. The use of helicine snails in medicine has been traced by Cadart (1955) from the earliest times right up to the present where it has been reported that the mucus of Helix pomatia favors the action of penicillin. Hooper (1910) and Read (1931) have given rather exhaustive and certainly most interesting treatments of the use, by the Orientals, of snails and other animals in materia medica. An article in the April 19, 1936, issue of the Osaka Mainichi Shinbun, translated by R. Urata and E. I. Naito (Anon. 1936), discusses at length, with testimonials, the special curative powers with which snails are supposed to be endowed. The curative substance extracted from the snails is stated to be "Ishimoto CONTROL THROUGH HUMAN USE 151 Negligin" which proves to be orthocalcium phosphate. This is perhaps the "medicinal substance" which the Institute of Medicine in Japan was reportedly extracting from the giant snails (Pangga 1949:342). This chemical is claimed to cure kidney disease, tuberculosis, anemia, diabetes, asthma, urticaria, circulatory disorders; to improve constipation and hemorrhoids; to prevent influenza; to restore virility and vitality; to perpetuate beauty and clear the skin; and to be recommended especially for those who sing a lot and those in need of hormonal injections! The keynote for the entire article is struck in the first paragraph with this sentence: ''Buying meat and fish for your table is unnecessary, there will be no sickly person in the house, doctors are kept away and it brings smiles to the home." Tazawa (1934 et seq.}, especially, urged the cultivation of A. fulica and made numerous recommendations for increasing production and simplifying the task for raising the snails. Ri (1935), Kagiya (1936), and Takamura (1936) made further suggestions in a whole series of articles appearing in Nogyo Sekai ("Agricultural World"). The series (in which the giant snail was variously referred to as the "edible snail," "land ear-snail," and "Shirafuji snail") was obviously designed to promote the development of a "new industry" and covered everything from a strictly zoological treatment (Taki 1935) to an attempt to establish a convincing parallel with the edible helicines of France (Kaibo 1935, Sato 1935). With propaganda of this sort being turned out in agricultural journals, newspapers, magazines, and the pamphlets of those promoting snail sales, it is little wonder that within a few years A. fulica became established in a number of new localities. The infestation in the island of Maui in the Hawaiian Islands is unquestionably a result of this advertising campaign. The Japanese importer there had not only introduced A. fulica for "food or medicine," and had advertised them in the papers in that manner, but had proved to be his own best customer by consuming 425 of them himself (Pemberton 1938). (Since he had reportedly sold only six other specimens since he had obtained the original stock over a year previous to that time, it is possible that he was eating them out of sheer necessity!) The infestation in Chichi Jima is another case in point. It was determined from the older residents that sometime in 1937-38 a store owner in Chichi Jima, by the name of Katsu Nishimura, brought back from Japan some of these snails in the hopes that their broth would clear up the tuberculosis he and one or both of his two children had contracted. It might be recalled at this point that it was for precisely this same reason that they were introduced into the island of Mauritius (Bosc 152 CONTROL THROUGH HUMAN USE 1803). Mr. Nishimura jealously guarded these snails and fed them only the most choice food items. Only his closest friends received some of the first batch of eggs. As expected, the containers were often inadequate and there were early escapes. All too soon the population build-up of the escapees took on the familiar pattern, and the people on Chichi Jima found themselves plagued with a new pest. Charles Washington, the oldest resident on that island, said that Nishimura died of tuberculosis before the war and his two children were evacuated to Japan. The story of the Pagan infestation runs essentially in this same vein (Kondo 1950a, c). In a strange sort of way, the giant snail, at first the apparent precious bearer of a medical panacea and then a cursed plague of the vegetable and flower garden, turned out to be a salvation to the besieged Japanese soldiers. Fishing could not be indulged in because of a constant state of alert. Supplies were cut off, although often enough there were sufficient stores of rice to last for a considerable period of time. The only source of animal protein on the islands was the giant snails, which in the meantime had become abundant. These then provided that critical food item. In fact, it was stated that the commanding officer of the Japanese contingent on Chichi Jima insisted that his men spend their rest period collecting these snails so that they might be added to their store of food. There was found much evidence in support of the assumption that these men "seeded" the giant snails near their hundreds of small fortifications so that they would have an easily obtainable protein item in the event of siege (cf. Benthem Jutting 1952b:395). It seems perfectly clear from the foregoing that there should exist no reasonable doubt about the facts that the Japanese people actually have eaten the giant African snails, that the original purpose in many cases for eating them was their supposed curative powers, that the snails were introduced into uninfested areas with the intent that they were to be eaten, and that not infrequently the exigencies of the war made eating them a necessity. There still is uncertainty regarding the extent to which the Japanese would under normal conditions voluntarily select the giant snail as a food item of choice. In spite of the praises sung in their "flavor," there is a great deal of evidence to the effect that such claims were only to promote additional sales; that is, they were apparently raised to sell and not raised to eat. Buyers in turn sold them to still others. According to Esaki and Takahashi (1942), some of these who got in early on the rapidly expanding market, amassed a net profit of 2,000 yen in six months. These authors have very beautifully summed up their opinions in this whole CONTROL THROUGH HUMAN USE 153 matter in the following (translated): "It will first be necessary to erase the name 'edible snail' to correct people's misbelief . . . it is not so tasteful, as can be ascertained if one tastes it. Even though it be tasteful and rich in nourishment, Japanese are not used to eat snails, demand will be very scarce, and it has no commercial value as food material or canned food.... If difficulties in cooking and taste problem are ignored, it is certain that the snail will not be popularized among Japanese, judging from the actual affairs in various places. . . . Canning and other commercial enterprises are totally out of the question." It is more than probable that today these statements reflect accurately a consensus among the Japanese people. It is significant that Esaki had been asked to visit Koror Island and offer suggestions for the control of the giant snail; he therefore was familiar with the attempts to can the snails on that island. The question immediately comes up as to whether or not other peoples react in the same way. It has already been mentioned that the Chinese in Singapore have been observed eating A. fulica. M. W. F. Tweedie, director of the Raffles Museum and Library in Singapore, corroborated this with the following statement (in litt. Dec. 19, 1949): "In the Japanese Internment Camp in Singapore the proteinstarved internees used them for food; they are said to be palatable, suitably spiced, but rather tough. The trouble was that there were not enough snails to go around." A. F. Caldwell of the University of Malaya adds the following firsthand information (in litt. April 23, 1953): "During the Japanese occupation these snails were eaten by the civilian internees after allowing the snails to empty the digestive tract for a day or two. They taste somewhat like a piece of tough kidney but the flavor is not so pronounced." H. A. A. M. Wirtz stated to the author that while he was a prisoner of war in Singapore, he had eaten many achatinas. He prepared them by breaking the shell, removing the viscera, and cooking them for seven to ten minutes. Cooking them longer made them tougher. If fat of any sort was available, it was added to them. He claimed that he did not object to eating them, although it was apparent that they were not relished. Garnadi (1951) reports that because of the slime in the freshly prepared snails, he found them less appetizing than dishes containing the dried snails. According to him, Franssen (1936) considered the snails "very tasty." Thomas (1949), in an apparently somewhat facetious vein, makes the following comment, 'Tour photographs of the giant Achatina snail made my mouth water, for here, undoubtedly, was my 1944 Christmas dinner in Singapore, looking every bit as succulent as one of those specially fattened snails. . . . Three friends and I con- 154 CONTROL THROUGH HUMAN USE sumed 4 0 . . . . " From the Hong Kong area, Herklots (1948) makes a similar report: ''During the occupation these snails were certainly eaten by the hungry Chinese and even after the war large specimens could be seen for sale in the market. They are made into soup which is said to be sweet and nutritious. A recipe worth trying is this one: Remove snails from shells, stew in water, rub hard with ashes to remove the slime, wash and boil with lean pork. One of ray gardeners who has been suffering from a chronic cough swears that he has been permanently cured by eating these snails regularly." Additional firsthand information on eating the giant snails in concentration camps is offered by Kalshoven (1950). A letter from R. E. Dean of HongKong (Dec. 15, 1949), however, reports that the Chinese in that area are no longer collecting the giant snail for consumption. According to Vosburgh (1950), the Chinese in the coastal regions of Formosa similarly do not eat these snails, although the aborigines in the interior eat them regularly with evident relish. These latter people are the only ones other than the native West Africans known to eat the achatinas from choice. It would be interesting to know whether developing an appetite for these snails is producing any appreciable effect upon the snail population in that region. The people of the Philippine Islands were also forced during the war years to collect and eat the giant snails (Allan 1949) and they are reported as being offered for sale in the Manila markets (Pangga 1950). There is no evidence that this practice is being continued now although Talavera and Faustino (1933) indicate that the freshwater snails Pila luzonica Reeve and Vivipara burroughiana Lea are regularly consumed by the poorer class of people in Manila. It is therefore possible that the same class of people has more recently turned to eating A. fulica. The Indonesians are known to eat aquatic snails; but they are prejudiced against eating the giant snail, apparently because of its known coprophagous habits (Garnadi 1951). The natives of New Guinea have not so far shown "any great enthusiasm for the snail as a supplement to other diet" (Harrison 1951, Allan 1949). But the Micronesians seem to have gone one step further. It was not a matter of lack of enthusiasm. It was downright antipathy which often approached abhorrence, in every sense of the word. By far the majority of the Micronesians, when asked if they had eaten or would eat the giant snails, were most emphatic in their denial; in fact, it seemed many times that the question had offended them, for they displayed an obvious resentment. One instance in particular stands out in the mind of this investigator. The native skipper in the Truk Islands was asked through an interpreter if he ate the giant snails. He CONTROL THROUGH HUMAN USE 155 registered shock and surprise and then indulged in gesticulations and denunciations which sounded like a mouth-filling Trukees oath. Translation was not necessary. The question however was pressed further with a ''Why not?" To this, no verbal answer was given. Instead, he indulged in a loud, protracted, agonizing demonstration of intense reverse peristalsis that left the interrogator weak, convinced, and without appetite. Even though his feelings were very negative and very intense, his reactions probably reflect the general type of response that most Micronesians, and incidentally most other peoples of the world, would give to such an inquiry. An investigation into the reasons for this almost automatic and seemingly inherent aversion brought to light the following: (1) there is a natural abundance of fish; (2) the Micronesians consider anything coming from the ocean as being basically "clean" and conversely that land invertebrates (worms, slugs, snails, insects, etc.) are basically ''dirty''; (3) the giant snails are extremely slimy and the slime is very tenacious and repulsive; (4) the snails have been observed to feed on many unappetizing things, such as rotting vegetation, human feces, feces of livestock, animal carcasses, including those of their own kind, etc.; (5) when the snails are cooked, the so-called liver mass gives off an acrid, offensive odor; (6) the natives are uncertain as to how to prepare them even if they should become curious about eating them; (7) and, finally, some have eaten them under semistarvation conditions and have subsequently become sick--they reason from this that the snails are probably poisonous. Reputed Poisonous Properties There has been much talk but little written on the presumed toxic nature of the giant snails as a food item. It has been pointed out above that the death of Guaminian dogs and cats has been blamed on the giant snails. Kondo (1950c) writes that he heard in Guam, Saipan, and Pagan reports of deaths following a snail dinner. His informant from Pagan offered the suggestion that death of the Japanese soldiers was caused by insufficient cooking of the snails. He said that they merely threw the snails upon hot coals to broil and ate them without further preparation when they seemed to be done. It was suggested that the copious slime was the toxic factor and that they should have removed this through frequent washings before they attempted to eat them. This conclusion does not appear to be a safe one. Both van Weel (1948) and the author have sampled the slime with no untoward effects, although the former did report that "one feels in the throat an itching, but no more." This same sensation was experienced by the author only when the raw "liver" mass was eaten, probably because of the power- 156 CONTROL THROUGH HUMAN USE f u l digestive juices it contains. It is likely that van Weel had some contamination from that organ. Possibilities of an urticarial property of the slime seem to have been dispelled by van Weel. G. H. Halden of Millbrae, California, writes (in litt. Nov. 1, 1949) that during his internment in the Philippine Islands, he and his wife cooked and attempted to eat some of the giant African snails but a physician companion warned against it. It was said that some of the internees had eaten them and had become sick. The evidence that the snails are poisonous is of course entirely circumstantial. Nausea and regurgitation can be stimulated purely through auto-suggestion; and with a natural repugnance for such things as snails, this would not seem to be difficult, especially with plenty of time afterward to contemplate what had been done in a fit of hunger. In times of stress, the food that is eaten may be contaminated, partially spoiled, of poor quality, or of insufficient and irregular quantity--any one of which can precipitate a digestive upset. In attempting to find a reason for such an upset, it is altogether easy to blame it on the most novel and unusual thing that has been consumed. If snails had been eaten, they would be elected on the spot. But there is another factor involved. It is a well-known fact that rich food taken in any quantity, especially after prolonged deprivation, may cause illness; and, as indicated below, snail meat is rich indeed. Halden himself pointed toward an explanation of this sort when he related that after a great many months in a prison camp, he became ill upon drinking his first glass of milk. Herklots (1948) supplies a bit of information which by comparison is considerably more mild in its censure of the giant snail. He writes, "... a friend who was interned in Borneo tells me that those prisoners who ate these snails did not improve their physique in comparison with those that left this free food alone." It might be facetiously added that worry over dire consequences could explain the difference. Quite aside from all these comments, however, there remains the indisputable fact that other species of Achatina are eaten by choice in great quantity and with impunity in West Africa. But again, these people know how to prepare them. An insufficiently cooked snail which had recently fed on contaminated or infectious material could obviously provide a real threat. It has been suggested that cases of helicine snail poisoning have resulted when the noxious contents of the intestinal tract had not been allowed to pass before the snails were prepared for consumption (Cadart 1955). Experiments in Eating the Giant Snail While in Chichi Jima, it was decided by Kondo and the author that the gustatory qualities CONTROL THROUGH HUMAN USE 157 of A. fulica should be determined through firsthand experience. The shells of a dozen large achatinas were cracked with a hammer and the soft parts were removed. The snails produced very little slime with this treatment and seemed to remain surprisingly placid. The visceral mass was cut off, leaving the muscular foot and mantle. These latter were dropped in salt water as they were prepared. They were washed in several changes of fresh water (at least eight) and a half gallon was used each time. Even the water used in the last washing soon became as viscous as thin egg white; there seemed to be no end to the slime. Finally, the foot of each was bisected and they were placed, slime and all, in a pan of water seasoned with salt and chopped onions. After boiling one and a half hours, they were still tough and rubbery--like pickled limpets or clams. Another forty-five minutes saw some improvement, for they could be cut with the edge of a spoon with difficulty. The black pigment of the foot had boiled out sufficiently to produce an unappetizing black broth. The aroma issuing from the pan was reminiscent of a chicken freshly scalded for plucking and it was not in the least associated with anything for consumption. The gingerly initial taste of the broth convincingly demonstrated the fact that one cannot determine by sniffing how a thing will taste. Instead of hot feathers, it tasted more like a handful of rich forest humus had been added to it. The snails had the consistency of "tough mushrooms," if that can be imagined, and the earthy flavor seemed to help this illusion. Mastication did not crush and render soft the pieces of snail flesh, instead it merely split them into smaller and smaller discreet, firm particles. The obvious "flatness" in flavor seemed to be due to the lack of fats or oils, the addition of which might have helped considerably, as would other seasonings, which unfortunately were not available for this experiment. A number of recipes, originally designed for the Mediterranean helicines, have recently been recommended by those who feel that the giant snail will prove to be good eating (e.g., Garnadi 1951). They all seem to have two things in common, viz., a pound of butter and half the spice cupboard. Almost anything would be acceptable with its own flavor so effectively disguised. It might be of interest to note here that the Savorys' omnivorous dog, Patty, dashed to her dish in great anticipation when the rest of the achatina stew was put there; with her right paw cautiously curled under her, she took one sniff, backed up, and retreated to the far corner to sulk. Subsequent experimentation has shown that boiling the snails in the shell for approximately five minutes and then dressing them produces a much less contracted and more tender morsel. The disadvan- 158 CONTROL THROUGH HUMAN USE tage in this type of preparation, however, rests in the fact that the hot hepatopancreas or "liver" imparts a pungent, acrid odor which to the novice might prove more than discouraging. As will be shown below, the hepatopancreas is the most nutritious part of the snail. To solve the problem of an excessive amount of slime in the freshly prepared snails, van Weel (1948) suggests soaking them for one to two hours in a weak solution of vinegar, lemon juice, or grapefruit juice. This he says "renders the snail hard at the same time, thus making it more palatable." He does not recommend cooking them in brine. The copious slime, on the other hand, was ingeniously and successfully utilized by H. A. A. M. Wirtz as a "binder" in the process of making bread out of ground soybeans in a Japanese internment camp in Singapore. One correspondent in all seriousness volunteered the suggestion that the snails might be cleaned in a washing machine! As an experiment in Saipan, Allie Jones pickled the African snails in brine. Like the cooked snails, they proved edible but, by most standards, not palatable. It should be mentioned at this point that Jones was sent to Saipan, under the sponsorship of the Pacific Science Board in 1949, to investigate the possibilities of making use of the giant snail for human and poultry consumption. Before his experiments were hardly more than under way, he sustained severe burns in a most unfortunate accident. It is hoped that the projected research will ultimately be completed. Apparently extensive experiments of this sort were carried on in Japan by Shingo Tazawa; and according to Esaki and Takahashi (1942), he "insisted that the snail is rich in nourishment, good as a tonic, will give excellent dishes in vinegar mixture, sesame bean-mash, bean-mash mixture, broiled with soy, coquille or stew" and that "these dishes gained high praise at experimental tasting eating parties." However, since Tazawa was doing everything possible to promote the sales of the giant snails, his reports were doubtless biased. As an interesting contrast, but without qualifying evidence, Mahony (1955) states, "The snail is edible, but when cooked and eaten it has a highly offensive odor and taste." If commercial use of the giant African snail is to be made in the future, it would seem at least at this time that it is as impractical as it is unrealistic to presume there will be any extensive use of them as an item of food for humans. The Micronesians will not eat them. And it is sheer folly to assume that they can be "educated to eat them." The same probably obtains for the other peoples of the world who do not already customarily eat snails. Those who do eat snails will not find the giant snails as acceptable as the edible helicines unless some wholly new method of preparation can be devised. CONTROL THROUGH HUMAN USE 159 Although there is need for exploring the possibilities of their use directly in human consumption, it is not at all likely that anything practicable will be discovered. Their use in curries or in the chow mein type of dish is the most promising suggestion made so far. The Japanese of course did eat them, but the reasons for it have been shown to be temporary. The Chinese have eaten them and it is conceivable that because of their naturally greater versatility in eating and because meat of any sort is perennially a particularly scarce item in China, there may be in the future some possibility of marketing processed achatinas in that country. This needs and warrants exploration. It would seem, then, that any use to which the snails can be put must be found in something other than human consumption. Food for Poultry The fact that chickens and ducks will consume achatinas has already been mentioned. This suggests in the light of the present discussion that perhaps the giant snails can be used for poultry feed. Van Weel (1948), Jones (1949), Pangga (1949), and others urge this consideration. The possibility was explored to a small extent by the author in Chichi Jima and inquiries were made elsewhere. Small living snails were given to chickens with no success. They occasionally would unenthusiastically pick at them but would give up almost as soon as they started. Other observers in Micronesia reported the same results. According to Garnadi (1951), Djaenoedin (1942) had no better results even when the snail flesh was mixed with coarse corn. On the other hand, van Weel (1948) found that chickens would not reject well-crushed snails but that they would not take to them as avidly as ducks invariably do. In Guam, Duane H. Kipp (in litt. Aug. 4, 1956) reported that his chickens ate the giant snail "raw, boiled, and roasted'' but "seemed to prefer them crushed and boiled." Steven Haweis (in litt. Oct. 26, 1949) has recommended the method used in the Dominican Republic; viz., the freshly killed endemic snails are pounded with corn meal before they are fed to the chickens. The Formosan Chinese crush the large achatinas before feeding them to their ducks (Vosburgh 1950). South (1926b), however, indicates that in Singapore the Chinese cooked the snails before using them as duck feed. It was decided to cook some of the snails before feeding them to the chickens. The freshly boiled specimens were quite slimy and the chickens were observed to be very slow to peck at them. The heavy slime often discouraged them after an initial examining peck. As the snails became somewhat dried out, the chickens would peck at them more frequently though never enthusiastically. After many hours, the cooked snails would be consumed. Kondo (1950c) observed 160 CONTROL THROUGH HUMAN USE chickens in Guam feeding on the burned carcasses of large specimens of A. fulica. Next, it was decided that boiled and dehydrated snails should be tried. The brown, sun-dried pieces of snail did not elicit even the slightest interest from the chickens, even at feeding time. Nor did remoistening to form a soft, rich meal make any improvement in its attractiveness. In the Seychelles Islands, the dehydrated achatinas are reportedly used for poultry food; but there is no indication as to how well they are received (Rees 1950). It is possible, however, that a reasonably fine meal is made of the dehydrated snails and that this is incorporated into the regular feed. Since chickens apparently do not readily eat the giant snails, a method of this sort would seem to be the only way in which they could be fed to chickens. A portion of the achatina shell could also be ground up and combined with the snail meal as an added source of calcium carbonate. In the proper proportion, it would obviate the necessity of using oyster shell or other calcareous substitutes (e.g., cf. Scholes 1945). But using snail meal in chicken feed brings up some questions. Will it taint the eggs or the flesh of the chickens? What does it contain in the way of essential nutrients? In what way will the chickens react to it? How does it compare with other supplemental feeds? How dependable is the source of supply? As to the possibility of a snail diet imparting an off taste to eggs, van Weel (1948) offers a negative answer with this statement: "No ill flavour was observed in eggs of fowls, fed with Achatina." He has not elaborated, however, on the extent of his experiments. It seems reasonable that further experimentation will prove verifying on this point. As an intresting sidelight, Taylor (1900:430) presents the belief that the flavor of the flesh of sheep is improved when great numbers of snails are consumed in grazing. In contrast, some Chamorros on Guam believe that when pigs are fed on a diet of giant snails, the flesh acquires "a different and undesirable taste" (Kondo 1952). Chemical Assay of Snail Meal Determining the essential nutrients is going to be a more difficult problem. A start in this direction has been made with the reports of chemical analyses of van Weel (1948), Mead (1950b) and Garnadi (1951). There is shown in Table 1 a comparison of the percentages that were found. Because the tests were not equally extensive, the constituent chemicals compare in only four instances. It seems clear from the figures that approximately three-fourths of the weight of the snail, exclusive of its shell, consists of water (cf. Pelseneer 1935:15; Fischer 1941). The variation in the amount of ash is not serious. But as gastropods are notoriously lacking in fats, as such, it is not understood how van Weel and Gar- CONTROL THROUGH HUMAN USE 161 nadi arrived at such comparatively high figures. The presence of small amounts of fats in marine forms has been demonstrated by Albrecht (1921 et seq.) and Struve and Kairies (1930). Biedermann and Moritz (1899) actually showed that ingested fats were deposited in the digestive gland. The "ether solutes" listed by Mead, however, amount to scarcely more than a trace, as was earlier predicted. In fact, it is this paucity of fats which makes it even more practical to consider making use of snail meal; for it should not be necessary to run the meal through a fat-removing process in order to insure against its TABLE 1 CHEMICAL ANALYSES OF DEHYDRATED Achatina fulica, EXCLUSIVE OF SHELL, WITH PERCENTAGES BASED ON DRY WEIGHT Mead (1905b)* Parts assayed Per cent H2O before dehydration Ash Calcium Ether solutes ("fats") Glycogen Nitrogen. . . . .... "Nitrogen -free rest" Phosphate (P2O5) Phosphorus Potassium Protein Foot and mantle 76.51 11.18 0.257 0.75 9 835 Lot At Visceral mass van Weel (1948) All soft parts 80.2 17.7 8.0 60 Garnadi (1951) All soft parts ca. 7% 10.0 65 76.51 10.72 0.265 1.19 9.848 30.5 11.78 3.67 2.75 60.95 35.34 10.7 3.35 60.99 55.6 53.0 * Analyses made by Edwart L. Breazeale of the University of Arizona Agricultural Experiment Station, Tucson, Arizona. t Collected by Mr. and Mrs. Peter J. R. Hill, Koror, Palau Islands, and dehydrated there in a plant drier at about 160° F. for 26 hrs. becoming rancid in storage. The percentages of protein are all very close, and together they vouch for the great value of snail meal as a source of animal protein. The extremely low percentage of calcium suggests that this important constituent is not accumulated to any great extent in the mantle or elsewhere in the body, but is speedily deposited in the shell. The whole problem of calcium metabolism in Helix pomatia and H. aspersa is considerably elucidated by the detailed work of Dexheimer (1951) and Wagge (1952), respectively. W. Fernando (1946) first discovered galactogen in the albumen gland and glycogeri in the albumen gland and reproductive cells of A. fulica. Van Weel found a significant amount of glycogen in his analysis of the soft parts of this species. This carbohydrate has been demonstrated in other gastropods (Creighton 1899, Levy 1890, Erhard 1912, 162 CONTROL THROUGH HUMAN USE Leger 1924). The very high percentage of phosphate (and therefore phosphorus), especially in the visceral mass, which consists largely of the hepatopancreas or "liver," greatly increases the value of the snail flesh as both a feed and a fertilizer. This finding lends support to Leger's statement that the visceral mass of H. pomatia is its most nutritious part and that therefore the snail should be eaten "in to to" (cf. Table 3 and Moretti 1934). In no other constituent in A. fulica is there any significant difference between the values found for the visceral mass and the muscular mass. TABLE 2 COMPARATIVE CHEMICAL ANALYSES OF DEHYDRATED Achatina fulica FROM VOLCANIC AND CORALLINE ISLANDS WITH PERCENTAGES BASED ON DRY WEIGHT* LOT B Origin (island) Type of soil Date collected No. of specimens Ash Calcium Ether solutes Nitrogen Phosphate (P2O6) Phosphorus Potassium Protein Shell: CaCO3 SiOj Koror Volcanic 3 Apr. 1950 29 C Ngannalk Coralline 25 Apr. 1950 12 D Koror Volcanic 27 Apr. 1950 14 .33 Trace 9.7 12.32 5.38 2.93 60.6 0.26 0.75 9.8 12.0 4.0 2.75 61.0 98.7 1.2 0.25 Trace 9.2 13.50 5.36 2.18 57.5 99.0 1.0 98.9 1.1 * Collectors and analyses as for Lot A, Table 1, except dehydration was for 72 hrs. In subsequent analyses, efforts were made to determine whether major environmental differences produce any appreciable corresponding differences in the values of the basic chemical constituents. Specimens collected in areas of volcanic soil were compared with those collected in areas of coralline soil. An examination of Table 2 will show that unless many other tests are run with essentially the same results, the slight differences shown cannot be considered significant. It is probable that at least a portion of the SiO2 in the pulverized shell analyses came from contamination with soil. But, on the subject of variations, Brand (1932) made interesting discoveries in checking the chemical composition of H. pomatia at various seasons of the year. The variability which he found may help explain CONTROL THROUGH HUMAN USE 163 why there is not closer agreement in the analyses of different investigators. Table 3 has been drawn up to emphasize this point. Snails often consume a fair amount of soil. Since the digestive tract is left intact, and since the soil which it may contain will vary from place to place in its chemical makeup, it is more than apparent that still another variable enters the picture in this way. Further contributions along this line have been made by Boycott (1921), Vincent and Jullien (1941), and Janda (1951), who report on TABLE 3 CHEMICAL ANALYSES OF Achatina fulica AND Helix pomatia WITH PERCENTAGES BASED ON LIVE WEIGHT Balland (1898) Species Brand (1932) Leger (1932) H . pomatia Foot 1.64 Garnadi (1951)f Mead (19506)* A.fulica Foot 2.61 .060 0.176 2.31 van Weel (1948) A. fulica All soft parts 3.51 H. po- H.pomatia matia Parts assayed. . . All soft All soft parts parts Ash 2-3.4 1.55 Calcium Ether solutes ("fats") 1.08 0.5 "Extractives" . . . 1 97 1.5-2.0 Nitrogen "Nitrogen-free rest" Glycogen Phosphate (P2O6) Phosphorus Polysaccharides. . 1.5-2.8 Potassium Protein 16.10 Water 79.30 80-84 A. fulica "Liver" All soft parts 2.29 1.52 Visc, mass 2.52 .062 0.279 2.31 0.42 1.91 1.78 5.98 0.988 1.58 4.63 0.842H 2.77 0.862 8.30 2.51 1.19 15.87 80.16 17.31 72.64 8.06 84.7 0.646 14.32 76.51 0.787 14.32 76.51 1 1 80.2 * Computed on basis of dry weight percentages, t As a part of "extractives." the presence of manganese in land mollusks (cf. Andre 1923), and Fischer (1941), who states that under varying conditions the water content of aquatic snails will go from 66 per cent, after prolonged exposure to air,to an apparent optimum of 86-93 per cent. Chemical studies of marine forms have been made by Albrecht (1921 et seq.} and McCance and Shipp (1933), B. Bergeret of Yaounde, French Cameroun, has announced (in litt. Dec. 25, 1953) that he has established the nutritional value of Archachatina marginata and Archachatina sp. Dr. Y. W. Hes very kindly turned over to the author an unpublished manuscript, dated August, 1951, containing a chemical analysis of the "big shells" (>15 mm.), "small shells" (<15 mm.), and 164 CONTROL THROUGH HUMAN USE "operculae" (i.e., epiphragms) of eastern Javanese specimens of A. fulica. This analysis is probably the only one extant for the giant snail; and it is for this reason that it has been reproduced, with permission, in Table 4. A somewhat comparable analysis was made by Turek (1933) on the shells of Cepaea nemoralis in which there were found the following values: water 0.48 per cent, N 0.201 per cent, SiO2 0.217 per cent, Fe 0.0014 per cent, Ca 38.88 per cent, Mg 0.0025 per cent, organic matter 1.36 per cent (of which 14.8 per cent was N), P 0.0040 per cent, and conchiolin 1.21 per cent. In discussing the shell of A. fulica, Garnadi (1951) states merely that it is made up largely of CaCO3 with small amounts of CaP2O5 and MgCO3. Other analyses of gastropod shells were reported upon by Clarke and TABLE 4 CHEMICAL ANALYSIS OF SHELLS AND EPIPHRAGMS OF Achatina fulica (Ex MS W. HES) Large Shells Small Shells Epiphragms Water (dried at 105° C.) CO2 N SiO2 Fe2O2+Al2O23 C2O 0.2 39.6 0.1 Traces 54.6 0.3 5.0 0.3 39.8 0.1 0.1 53.7 0.2 4.9 MgO Organic matter 1.4 36.7 0.4 0.3 0.2 50.1 0.8 9.7 Wheeler (1917, 1922), Boring (1872), Steel (1922), and Struve and Kairies (1930). Stolkowski's more recent work (1951) especially should be mentioned because of its detailed treatment and exhaustive bibliography. Vitamin Assay of Snail Meal Even though vitamin A is quite susceptible to oxidation, especially in dehydration, a preliminary assay for this vitamin was conducted in the nutrition laboratory of the Univeristy of Arizona. The tests run by Mitchell G. Vavich of that laboratory on the dehydrated material sent from Koror revealed the presence of vitamin A in the dehydrated visceral mass to the extent of 5.4 Fg/g. For comparison, beef liver has 300-400 Fg/g. The pro-vitamin carotene, not heat-labile at moderately high temperatures, appeared in 10.3 Fg/g. These figures are low for animal tissue. Analyses run on the dehydrated foot and mantle tissue brought still lower values. These tests establish the fact that snail meal does not provide a good source of vitamin A or carotene; and if it is to be used as a food supplement, other sources must be depended upon for CONTROL THROUGH HUMAN USE 165 these essential factors. It is of at least academic interest, however, to report the presence of these factors even in small quantity. Projected tests will determine the relative amounts in the live and freshly killed snails. Vitamin B1 (thiamine) was found in only small amounts by Roller (1941) who discovered it in the liver and whole animal of active Helix pomatia in 0.3 Fg/g and 0.9 Fg/g, respectively. In the dormant animal, the assays were negative. In contrast, vitamin B2 (riboflavin), in this species and in H. aspersa, was not only found in considerably higher amounts, but the values actually increased during hibernation. The amounts, expressed in terms of micrograms per gram of the fresh tissue, are: hepatopancreas 16.3, kidney 7.0, muscle 1.4, and genital system 2.3 (Rafly and Ricart 1943). Hoar and Barberie (1945) have shown that marine mollusk tissue is higher in vitamin B2 than mammalian muscle and fish; and that drying, canning, or salting the tissue may produce great losses of this vitamin, whereas freezing and smoking do not. Destruction of "vitamin B" was earlier demonstrated by Jones and Murphy (1926) in the dehydration of oysters at temperatures as low as 40° C. In contrast, preliminary chick feeding tests at the University of Arizona indicate that there is no vitamin B deficiency when dehydrated giant African snail meal is the only possible source of this vitamin complex (Mead 1959b). The presence of vitamin C (ascorbic acid) in H. pomatia, in considerably less quantity than in beef, was reported by Wenig and Halacka (1948). Earlier, Marchi (1928a, b, 1929) and Random and Portier (1923) found that this vitamin was present in a number of marine mollusks; but Nespor and Wenig (1939), after finding it in H. pomatia, concluded that it did not figure importantly in invertebrate metabolism. According to Meenakshi (1951), vitamin C in A. fulica, as in other gastropods examined, is strangely concentrated intracellularly toward the distal ends of the hepatopancreatic cells. These several reports, plus such more comprehensive works as that of Waisman and Elvehjem (1941), should stimulate in the near f u t u r e fairly exhaustive vitamin assays of A. fulica. Amino Acid Assay of Snail Meal The discovery of a very high protein content in the snail meal in turn has introduced the question of the amino acid makeup of this protein. Preliminary amino acid assays revealed the very significant fact that the essential amino acid lysine was present in the dehydrated, pulverized snails to the extent of 11/3times that of whole chicken egg (Mead and Kemmerer 1953; Anon. 1953d). The consistent deficiency of lysine in the common sources of plant protein used as food supplements, such as cotton seed meal, makes this discovery all the more important. In these 166 CONTROL THROUGH HUMAN USE same preliminary tests, arginine was shown to be 21/3times the relative amount found in whole egg; this amino acid, however, has not been established as essential in normal nutrition. Other amino acids were sufficiently high in concentration, as compared with whole egg, to increase considerably the estimates of the value of snail meal as a poultry and livestock supplemental feed. The variation in the assays of snails collected in different environments suggests that a series of experiments should be set up to determine, among other things, what environmental conditions predispose the maximum production of lysine. Sekine (1926) has not only demonstrated the presence of lysine in the marine bivalve Meretrix meretrix but has determined that it is almost 1f times the amount in meal made of the salmon, Oncorhynchus mason. It is not enough to demonstrate that certain amino acids are present in desirable amounts. There still remains the question as to whether or not growing animals will actually respond properly to a diet in which the protein item consists largely or entirely of snail meal. This properly implies that other growth-promoting factors, not yet determined, would be allowed to come into effect under such conditions. Just such investigations as these are currently in progress at the University of Arizona, with day-old chicks as the experimental animals. Attempts will be made to determine the most effective way in which to incorporate the meal into chicken feed and to detect any possible idiosyncrasy that the chickens might have to the snail meal. In the chick feeding tests, a comparison will be made between snail meal and other sources of animal protein, such as blood meal and fish meal. A. F. Caldwell of the University of Malaya gives some interesting information (in. litt. April 23, 1953) regarding the response of ducks to a diet of snails. He states, "I did notice that the addition of snails to the diet had a bearing on egg production; in dry weather when the snails retreated to hide outs egg production fell, but increased again as soon as they were given snails." H. A. A. M. Wirtz of Java independently offered essentially identical information. Along this same line, Fronda (1919) and Frigillana (1923) in the Philippine Islands reported that when their experimental chickens were fed ground and boiled snails (probably the freshwater snails Vivipara and Pila according to Talavera and Faustino 1933), there was a marked increase in the number and size of the eggs and the weight of the chickens as compared to chickens on substitute diets of copra meal, "palay" (unhulled rice), or cowpeas. Geseco (1921) obtained similar results with ducks. Because the snails were periodically difficult to obtain, Cruz (1932) attempted to find a sub- CONTROL THROUGH HUMAN USE 167 stitute. On the basis of cost, shrimp meal was recommended. But now that the people in the Philippines have a free source of A. fulica, it is quite likely that giant African snail meal is currently being used. Food for Livestock According to Kalshoven (1950), Djaenoedin (1942) has recommended that the giant snails be incorporated into the daily rations of pigs. The value of fortifying their diet with snail meal can be predicted on the basis of the present experiments with chicks. The same goes for other livestock and pet feeds. Wilson Savory of Chichi Jima stated that his pigs would take to the live snails quite as readily as they would to the boiled and chopped ones. In contrast, Frank L. Brown of Rota failed to persuade his pigs to eat the live snails, even after considerable fasting, although they would eat them if they were cooked. Father H. F. Costigan of Kolonia, Ponape, regularly fed his pigs on small cooked achatinas left in the shell and mixed with other food. In Tinian, a farmer was seen to feed his pigs boiled achatinas which had been removed from the shells (Chamberlin 1952a). In no case were the snails eaten avidly nor were the larger specimens consumed unless they were crushed or removed from the shells. Garnadi (1951) states that the surgeon of the Veterinary Institute of Bogor determined that pigs put on a diet entirely of giant snails lost weight but showed no signs of sickness. Fish Bait Using the giant snails as salt-water fish bait in preliminary experiments in Saipan did not prove encouraging (Jones 1949). Living edible European snails, however, are being used as freshwater fish bait in the United States (Ingrain 1952). If the numbers sold for this purpose are any criterion, they must be good. The raw flesh of the Philippine pond snail Pila luzonica Reeve is reportedly used as a bait for line fishing (Talavera and Faustino 1933). Fertilizer In spite of the fact that primitive peoples all over the world have made use of disintegrating animal flesh to replenish the plant-growth-promoting constituents in depleted soils, the survey of the giant African snail problem in Micronesia failed to reveal any extensive use of A. fulica for this purpose. The people of Chichi Jima were the only ones observed at that time to be making direct use of the giant snail in this manner. There they collected mostly the small, thin-shelled snails in metal oil drums. These were allowed to stand in the hot sun until the snails died and had reached a high degree of putrefaction. These rotting, maggot-infested snails were then scooped out, the shells were crushed, and the putrid, slimy, odoriferous mess was added as a fertilizer to young vegetable plants. The reason for the generally better quality of the crops in Chichi 168 CONTROL THROUGH HUMAN USE Jima, as compared with other areas, rests at least in part in the use of snail fertilizer. Its use in this case, however, introduced two pronounced disadvantages. The inadequately covered oil drums permitted escape of the fly maggots as a threat to public health, and the addition of the crushed shells to the soil moved the soil pH even more strongly in a basic direction. More recently, the latter disadvantage has been eliminated in the similar use of snail fertilizer in Guam in that only the "liquor" is drawn off and diluted with ten parts of water before being added to the soil (Peterson 1957b). In areas where the soil is excessively acid, as is the case in humus soils of forests, the alkalizing effect of adding crushed shells would of course be desirable as far as most crops are concerned. But conversely, coralline soils are already excessively basic and should not have more calcareous material added to them. This suggests immediately that if there is ever commercial use of the giant snails in fertilizers, two different kinds should be made available--one with, and one without shell fragments. One thing is certain, the thin, quickly leached soils of tropical areas need enriching of some sort if they are to be used for much more than two or three years. Where this is not possible, the natives abandon the plot, burn off another portion of the forested area and use that until the soil nutrients in turn are depleted. All too often the abandoned land is soon taken over by sword grass (Miscanthus) or some other noxious plant with little chance, for a long time, of the land being recovered again for crops. Production of Snail Meal The practicability of producing fertilizer from the giant African snail is in most respects on a par with that of producing the supplemental feeds. In both cases, it would be desirable to dehydrate the snails and reduce them to a powder or meal. In this form it could be stored and used when needed either as a fertilizer or a feed supplement. The big question at this point is not whether the snail meal has value in either of the categories. The relatively high percentages of phosphate and lysine quickly answer that question. The real question is: "How dependable is the source of supply?" An examination of the reports in the literature and an examination of an invading population would tend to convince one that the snails are present in inexhaustible numbers. This is misleading. In any given area, irrespective of the abundance of the snails, an intensive collecting campaign will remove in a relatively short time so many snails that no further specimens can be found even with diligent hunting. Experience has demonstrated over and over again, however, that eradication is not thereby effected. In a year or two the residual snail population builds up in appreciable numbers once CONTROL THROUGH HUMAN USE 169 again (unless the phenomenon of population decline comes into the picture, vide infra). This almost automatically suggests that perhaps a system of ''harvesting" the snails could be set up wherein the collecting could be rotated from one area to another on a long enough program so that population recovery could take place before reharvesting. Let us assume that the island, or region of the mainland, and the snail population are large enough for such a system of harvesting. There still remains the problem of labor. Native help could be enlisted. With the coconut tree actually or virtually wiped out on a number of the Micronesian islands, and with no chance of reviving trade in the Japanese-introduced sugar cane, the Micronesians in many cases are hard put to find some means of providing for their needs. In such areas, collecting and selling the ubiquitous giant African snail could be a welcome source of native income. In some other areas, however, attempting to enlist the help of the native people would not be altogether easy, for the pay might not be sufficiently high to arouse their interest. But in past times in a number of areas where only a very modest bounty could be offered for the snails during their initial build-up, considerable native help was obtained. The "emergency" nature of the situation at that time and possible elements of conscription may help explain the co-operation that was obtained. In a regular program of harvesting the snails, however, these persuasive factors would not be present. Nonetheless, since the native people will be collecting the snails in their garden plots as the most effective and most inexpensive means of controlling them, for a modest fee they could be persuaded to bring them into a common collecting place instead of trying to destroy them at an often even greater expenditure of energy. Let us suppose further that, through the means indicated, sufficient snails are brought into a centrally located processing plant to keep it in indefinite operation. There still must be worked out an economically feasible processing program in which at least some of the snails would be removed from their shells before they are dehydrated. Developing an economical method of dehydration, avoiding chemical decomposition with excessively high temperatures, avoiding maceration with insufficiently high temperatures, and maintaining dehydration in storage, are only a few of the problems which would have to be met. And even with these out of the way, there still remains the problem of marketing the snail meal on a commercial basis. This problem is sufficiently great in the minds of some investigators to cause them to reject summarily the whole idea of 170 CONTROL THROUGH HUMAN USE attempting to make use of the snail meal. They state: "With shipping rates as high as they are, how can you ever hope to market snail meal in this country at a price which will permit it to compete successfully with other animal protein sources?" But it is not in this country that the greatest critical shortage of animal protein food supplements exists. Ironically enough, it is right in the areas where the snails abound that there is the greatest need! As a case in point, in Guam in 1949 eggs sold for $2.50 per dozen and chickens sold for $3.00-4.00. The high cost and scarcity of imported chicken feeds were given as two of the major reasons for these exorbitant prices. Yet right on that island there are tons of giant snails, living, dying, and going to waste. Marketing should be no problem there or in any comparable area. The possibility of marketing the meal on a large scale commercially not only seems remote, but the present need for it is questionable. Local demand alone should take care of the supply. In regions where lime is in demand, the achatina shells could be reduced in lime kilns as a by-product in the preparation of the snail meal. The shells of Melania and Corbicula are apparently being used in this manner in Malaysia (Benthem Jutting 1948:137). It is a paradox indeed that these snails are forming from unwanted rotting and decaying materials a very rich source of protein for which there is a continually growing need both as a poultry feed supplement and a fertilizer, and yet next to no direct use is being made of them. More than that, current methods of control call for their destruction with no apparent thought given to their utilization. There has been a definite hesitancy, on the part of many who are concerned with the control of the giant snail, to prescribe any measure which would in any way make, or even suggest, a use for the snails. The reason for this undoubtedly rests in the fears reflected in the words of the first major survey report coming out of the Pacific after World War II, viz., " . . . i f any persons found the snail a benefit, others would be tempted to introduce it elsewhere, regardless of the common good. It might, therefore, be unwise to try to develop uses for it." (Townes 1946; cf. Rees 1950:586). It cannot be denied that the warning is a sound one; but it has had an inordinate effect upon subsequent control programs. There is considerable doubt in the minds of some that any really appreciable spread of A. fulica would result from finding a use for it. On the contrary, finding a use for it could be responsible for reducing the population to the point where the snails would practically cease to be a problem. It is the great multiplicity factor which largely causes them to be branded as a pest. But, whether or not any use is found for them, the snails will con- CONTROL THROUGH HUMAN USE 171 t i n u e to spread into new areas, in spite of all efforts to the contrary. If they are going to spread, why not enlist the interest of man to help keep their numbers down to the point where they are not a serious problem? Man always has been at the head of the list of known enemies of the giant snail. It is almost completely impossible to try to teach man, in the recently invaded areas, to eat the snails; but at least he can be taught to make indirect use of them toward his own benefit. In this manner, the snail could be a blessing rather than a curse (Mead 1953b, 1955b). Apropos of this, much attention is currently being given to various invertebrates as possible natural sources of highly complex organic compounds, a development which holds great promise for the fields of medicine and organic chemistry. CHAPTER 9 CONTROL THROUGH LEGISLATIVE ACTION Ever since the very earliest stages in the spread of Achatina fulica from its East African homeland, there have been various edicts and decrees urging or demanding that the local people co-operate in collecting the snail specimens and assist in preventing their spread into uninfested areas. Some investigators (e.g., Corbett 1933) in addition suggested that all foreign plants be subject to inspection before entering the country. By and large, these early measures were only of transitory, if any, effect, largely through the fact that it requires much more than regulations to stem the tide of this giant snail pest. Forcible and effective quarantine regulations against A. fulica were not promulgated until 1936. Early in May of that year, it was decided by the Ministry of Agriculture and Forestry in Japan that this snail had clearly earned for itself in Formosa the reputation of being a serious pest. Forthwith there were set up emergency regulations which permitted the confiscation not only of all specimens entering Japan, but of all living specimens in the country at that time (Pemberton et al. 1939, Esaki and Takahashi 1942). A concerted program of propagandizing the imagined medicinal properties of these snails had succeeded, in the years previous to that, in spreading them into many areas. The thoroughness with which the regulations were carried out, and probably to a lesser extent the severity of the winters in Japan, together stamped out all sign of the giant snail. There is today no evidence of its having become established in that country. This is the 172 CONTROL THROUGH LEGISLATIVE ACTION 173 only instance known where A. fulica has gained entrance to an area and has been completely eliminated. It should be kept in mind, however, that because these snails were cherished and coveted right up to the time they were banned, they never actually became "established" in the real sense of the word. Such, unfortunately, is not the history of the Hawaiian infestation. Two-year-old populations of the giant snail were discovered in both Maui and Oahu in November, 1938 (Pemberton 1938, Mead 1949b). Every effort was made to stamp out these infestations and, for a while, it seemed that the snail had been eradicated. In fact, eradication was announced by several prominent papers dealing with this subject. But it was the same old story all over again. Live individuals were continually found in the areas of the original infestation. Persistent control measures (costing in excess of $200,000) coupled with stringent quarantine regulations, with heavy penalties for people caught transporting the snails or even having them in their possession, did succeed in keeping the infestations pretty well corralled until 1951 (Fullaway 1943). In that year, a number of new loci were discovered, in both Oahu and Maui, in some cases miles from the original site of infestation (Wong 1951). These discoveries have made intra-island quarantines essentially impractical and have moved one government official to state, "We simply do not have the money to resort to measures approaching eradication. We have the snail in such quantities on Oahu today and it is in so many places that it now becomes a matter of individual control and not a possible function of the Government --just like exotic weeds and insect pests in wide variety." Inter-island quarantines, on the other hand, were successful in preventing the spread of A. fulica to any of the other islands in the Hawaiian chain for a period of twenty years. The discovery of established infestations on the islands of Hawaii and Kauai in 1958, however, clearly demonstrated once again that in spite of the best quarantine regulations this pest may be able eventually to effect a breakthrough. Shortly after the inception of the Trust Territory of the Pacific Islands, a set of basic quarantine regulations was drawn up to limit as much as possible any further spread of plant pests, including the giant African snail (U.S. Navy 1950, Pemberton 1954). These have provided the authority for the promulgation of numerous specific quarantine regulations and controls, a general policy for which is discussed by Bryan (1949) and Cooley (1950) (cf. 1CCM 1947, 1948). Subsequently, the Philippine government set up comparable regulations imposing fines and/or imprisonment for offenders (Pangga 1947). 174 CONTROL THROUGH LEGISLATIVE ACTION In the United States, California has had for a number of years specific quarantine regulations covering the threat of introducing snail pests. For over thirty years, that state has been battling foreign snails at a cost of over $500,000. During this time, only one of four introduced helicines, Theba pisana, is believed finally to be eradicated (Armitage 1949) in spite of notes in the literature to the contrary (USBEPQ 1953). Finding as many as 3,000 specimens on a single orange tree is a good index of the seriousness of the problem that this species has presented (Gammon 1943). Attempts by the state to control two of the other helicines, Helix aspersa and Otala lac tea., have been considered impractical because they have become too thoroughly established in many different areas (Messenger 1950). The fourth helicine, Helix aperta, has a narrow but firm toehold in the San Diego area. Concerted efforts are still being made to eradicate this snail. Its habit of retreating deep underground for long periods of time, however, has so far frustrated every attempt. California border and port quarantine inspectors have for years periodically intercepted shipments of these and other species of potentially harmful snails. For example, in July, 1951, 122 cases of living H. aperta from Tunisia were intercepted at one of California's border stations after passing through New York as a port of entry (Messenger in litt. July 13, 1959). Experience of this sort, coupled with considerably more experience with introduced, pestiferous insects, mammals, and birds, has moved such investigators in California as Storer (1931, 1934, 1949) and Hanna (1939, 1948) to take an unequivocal stand against introduction of foreign animals. In fact, it was these two investigators who were responsible for bringing to the attention of the proper authorities the fact that two specimens of the largest of all land snails, Achatina achatina were to be found alive in California (Dickson 1946, Hanna 1948, Mead 1949b). These two specimens and all their eggs were quickly destroyed. But this incident in addition to the sudden interception of about fifty live specimens of A. fulica during California port inspection of war salvage material in 1947 (Messenger 1947) speedily put California's well-organized quarantine service into high gear. And in spite of the fact that, since 1947, over ninety live specimens of A. fulica have been intercepted in California ports, this snail pest has never become established in the slightest, reports in the literature to the contrary notwithstanding. Unfortunately, the campaign in Southern California to eradicate O. lac tea in 1951 caused many to believe that A. fulica had at last become established in California, as newspapers referred to O. lac tea as the "African Snail" and the "striped African Snail." CONTROL THROUGH LEGISLATIVE ACTION 175 Not only has there been a redoubling of effort in California to detect the possible presence of the giant snails in cargoes, but rigorous steps are taken when the snails are found. The holds are fumigated with methyl bromide or hydrocyanic gas. Infested war salvage equipment, and such other equipment that can stand it, are treated with KOH or NaOH in live steam (Messenger 1952). During the years 1949, 1950, and 1951, six, eight, and twelve interceptions, respectively, were made in California ports (Messenger 1949-51). An explanation for this increase, according to Messenger (in litt. July 8, 1952), is found possibly in the fact that war salvage material from the Mariana Islands was being obtained from areas deeper and deeper into the bush where prolonged overgrowth with vegetation provided more suitable retreats for the giant snails. In the years 1952 through 1958, the interceptions were 5, 8, 0, 0, 0, 1, and 0, respectively (Messenger 1952-54, Messenger and Breech 1958). At this writing, there have been reported no interceptions of A. fulica in California in 1959 and 1960. The reduction in interceptions undoubtedly finds its explanation in the fact that there has been a commensurate reduction in, and finally a virtual discontinuance of, the process of bringing war salvage material into United States ports. To a much lesser extent, the marked decline in the snail populations in the infested areas may be having its effect. Chapter 49 of the Arizona Code amply provides for measures to meet the threat of introducing foreign snails. California and Arizona, then, are the only two states which have quarantines so designed that snails can be excluded. In contrast, before 1951, the federal quarantine regulations lacked any provision for prohibiting the entry of snails. These facts were pointed out by Mead (I949b, c, e) and the danger inherent in such a setup was emphasized. With this, the U.S. Public Health Service issued orders for its inspectors to assist the USDA plant quarantine inspectors in their attempts to intercept the giant snail. Legally, however, this could not be enforced as the USPHS Regulation 71.156 provides only for an "animal . . . vector of human disease or any . . . animal . . . capable of being a vector of human disease." In addition to all the depredations of the giant snails, they cannot so far be justly accused of being vectors of human disease. Morgenstern (1949) most unfortunately was confused on this point. The fact that the dying snails form potent sources for the breeding of disease-carrying flies provided the only tangible connection with public health, but this was so tenuous that it could not conceivably be embraced by the existing regulations. Photostat copies of Mead's article (1949b) were sent by the U.S. 176 CONTROL THROUGH LEGISLATIVE ACTION Army to personnel in snail-infested areas in the Pacific islands in an effort to reduce the spread of the snail in those areas and minimize the chances of the snail's showing up on army equipment returned to the States. Such steps as these could at best be only temporary expedients. Interceptions of the giant African snails in other ports of the United States added further emphasis to this point. After a single live giant snail specimen was found in San Diego on a ship carrying war surplus from Manila, a message of warning was sent ahead of the north-bound ship to San Francisco, Portland, Seattle, and Vancouver. The alerted inspectors in Vancouver found eight more snails in the cargo; fortunately, all of the specimens were dead (Gardiner 1949, Zuk 1949). Another ship carrying 8,000 tons of war salvage material from the Pacific was found in Baltimore, Maryland, to be snail infested. The inspectors sought and received the co-operation of the importers, and recommendations to fumigate the entire cargo with HCN gas were carried out (Brubaker 1950). But it cost the importers $22,500! The giant snails were found on still another ship in that same year of 1949 in Newark, New Jersey. An accidental fire brought a quick solution to the problem by destroying the cargo--and the snails. In 1950, snails were found in the cargo of a ship landing at New Orleans, Louisiana, (cf. McCrory 1950). The following year, a shipload of scrap metal from Guam was similarly found infested with the giant snail in Portland, Oregon, and was fumigated before it was discharged (Burch 1951). With the American public now informed (and in some cases, unfortunately, misinformed) by the publicity given in scientific journals, magazines, and newspapers to the problem of the giant African snail, definite steps were taken to set up legislation empowering the Department of Agriculture to prohibit entry of this snail. Representative Wingate Lucas, in response to the urging of the people of Texas, was the first to start such action. On the basis of an original inquiry directed to the author, it was decided by Lucas to introduce bill HR 6242 in the 81st Congress on September 27, 1949 and it was immediately referred to the Committee on Agriculture. The wording of the bill, however, was unfortunately such that only A. fulica would be excluded. Any person acquainted in the slightest with the giant African snails knows that any one of the many species in that big group probably has the potentiality of becoming as serious a pest as A. fulica. Immediate but unsuccessful attempts were made to obtain a rewording of the bill (Mead 1950b:44). It was agreed by many that the bill as it stood would at least meet the immediate problem at hand. Others frankly feared the bill was a hasty outgrowth of a ''snail panic" and that it would bring undesirable legislation (cf. Anon. CONTROL THROUGH LEGISLATIVE ACTION 177 1949b). Nine months later (June 28, 1950), Representative Harold D. Cooley, chairman of the Committee on Agriculture, released HR 6242 from committee without amendment. His Report No. 2363 carried an indorsement of approval both from his committee and K. T. Hutchinson, assistant secretary of agriculture. After passing to the Committee of the Whole House on the State of the Union, it was read before the House and passed without amendment on July 27, 1950. The following day it was read before the Senate and referred to the Committee on Agriculture and Forestry. Four months later (November 29, 1950), Senator Elmer Thomas reported back to the Senate from this committee and recommended (Report No. 2583) that HR 6242 be passed without amendment. In spite of this recommendation, Senator B. R. Maybank proposed an amendment (cf. Cong. Rec. 96[12]: 16621) to this bill and it was passed by the Senate. The title of the bill was changed to "An Act to Amend the Agricultural Adjustment Act of 1938, as Amended, and to Prevent the Entry of Giant Snails into the United States, and for Other Purposes." The bill, in this hybridized form, quite understandably failed to receive favorable action and died in the House during the last-minute rush of the 81st Congress. This meant starting all over again. This time, however, the author solicited the help of senate majority leader E. W. McFarland as well as that of Representative Lucas. The wording of the old bill was revised to include "any terrestrial or fresh-water mollusk." It was introduced as S 1489 to the Senate on May 15, 1951 by Senator McFarland and referred immediately to the Committee on Agriculture and Forestry. On August 9, 1951, Senator Ellender of this committee submitted a report (No. 628) indicating that this bill had the approval of Secretary of Agriculture C. F. Brannan and recommending that it be passed without amendment. An identically worded bill was introduced as HR 4443 to the House on June 13, 1951 by Representative Lucas and referred to the Committee on Agriculture. Once again, Representative Cooley of that committee reported back on the bill, recommending on August 7, 1951 that it be passed without amendment. His report (No. 800) carried a message of approval from Acting Secretary of Agriculture C. J. McCormick. The bill was read before the House on August 20, 1951 and passed without amendment. A week later, S 1489 came up for reading in the Senate and it was agreed to substitute HR 4443 and to pass it without amendment. On September 12, 1951, HR 4443 was signed by both the Vice-President and the Speaker of the House. Ten days later, it was signed by President Truman and became Public Law 152 of the 82d Congress (cf. Smith 1951, Burch 1952b). Under this new authority, the Secretary of Agriculture was em- 178 CONTROL THROUGH LEGISLATIVE ACTION powered to draw up regulations controlling the entry of mollusks in this country. Advice was sought from many persons, agencies, and institutions during their preparation. Acting Secretary of Agriculture K. T. Hutchinson published on July 25, 1952 a notice of the proposed regulations and copies were sent to the sources of advice for criticism and suggestions. The revised regulations were published by Acting Secretary of Agriculture C. J. McCormick on October 22, 1952 and became effective that date (Anon. 1953b). The following quotation from McCormick (Cooley and McCormick 1951) explains how this "snail legislation" will fit into the long-range plans of the Department of Agriculture: "A study is now being made by the Interdepartmental Committee on Pest Control of proposed legislation which would encompass a much wider field than is covered in HR 4448. This proposed legislation will probably include provisions for control of such plant pests as worms, insects, nematodes, slugs, and snails, any form of protozoa, fungi, bacteria, or other living parasitic plants, any living viruses, and similar or allied organisms, which can directly or indirectly injure or cause disease in plants or parts thereof. It will require considerable time to complete the studies which are being made with respect to such legislation. . . ." The existing regulations, however, will not just function in an interim fashion but will reveal through actual application just what is really needed in the proposed more extensive legislation. With the assistance of the publications of Ling (1952, 1954) and the FAO Plant Protection Bulletin, a cursory check was made of abstracts of the plant quarantine regulations of somewhat over one hundred governments. Slightly less than 3 per cent, that is, only three governments, viz., the Republic of the Philippines, Union of South Africa, and the United States of America (Anon. 1953b) were listed as making specific mention of "mollusks" among the animals prohibited or restricted. Less than 8 per cent of the total were listed as having an "invertebrate" clause to cover agricultural pests other than insects, thus technically embracing the mollusks. The countries included were: Canada, Ceylon, Chile, Kenya, New Zealand, Sudan, Trinidad and Tobago, and Uganda Protectorate. Nine other countries, the greater share of which are small island governments, had vague, all-inclusive clauses which presumably would give them the authority for excluding harmful mollusks (e.g., "and other small animals," "organisms or other agents," 'living pests in any stage," "any object carrying an injurious pest," and so forth). But over 80 per cent of the governments listed had little more in their regulations than the restriction and prohibition of certain specific pests and diseases without any clause worded broadly enough, apparently, to pro- CONTROL THROUGH LEGISLATIVE ACTION 179 vide authority for the exclusion of harmful mollusks. It can be seen from this survey that the vast majority of the governments have neither given consideration to the problem of excluding molluscan pests, actual or potential, nor prepared themselves for such an eventuality. With respect to A. fulica, the quarantine picture is much less clear in other countries. In England, Sir W. Wakefield especially has been concerned about the threat of the giant snail in the colonies and has brought the matter to the attention of Parliament (Anon. 1949d). In Australia, T. H. Harrison (1951) reported that two interceptions of A. fulica in Sidney taught them that their regulations were inadequate; hence there was passed Statutory Rule 1948, No. 92, which authorizes the inspectors to follow produce to the warehouses if necessary to intercept the snails. Quarantine inspection of imports is also effective in Sarawak, according to Tom Harrisson of the Sarawak Museum (in litt. Aug. 25, 1952). Internal quarantines on the other hand present quite a different problem as indicated in Harrisson's words, "Control within the country is quite impracticable as communications are by water in thousands of small craft often carrying leaf-thatch, vegetables, etc." In reports and correspondence, similar reasons are given for the lack of specific quarantine regulations for this snail in India, Indonesia, Malaya, and the Seychelles Islands. Some (e.g., Anon. 1948c) have insisted that definite steps should be taken in the infested areas to force the people to collect the snails by hand and destroy them. Although large-scale collections of this sort produce a discernible effect upon the snail population, they are only ameliorative and transient, and most certainly do not warrant the sizable expenditures of governmental funds that have been made in past times--especially during snail population build-up. As the only inexpensive means available of controlling the snails, the people should be ''urged" to collect and destroy them in their own areas; but attempting to exert force along this line through legislation would be advisable in only the most unusual cases. One thing is certain about quarantine regulations: Snails cannot be legislated out of existence. At best, quarantines may delay indefinitely the entry of the snail pest. The more carefully the quarantines are thought through and carried out, the more effective will be the barrier and delaying action. As in all good quarantines, even though there is no guarantee of permanent exclusion of the pest, there is provided more time--"borrowed time"--in which to develop more effective controls and perhaps even eradicative measures that do not now exist. In this way, external quarantines, and in some cases internal quarantines, can be effective and practicable. Thus they warrant serious consideration in any long-range planning. CHAPTER 10 POPULATION DECLINE The phenomenon of population decline in Achatina fulica was first reported upon by Green (1910c, 1911b) in Ceylon. At the present stage of our investigations, the following words of Green seem to be almost prognosticative: "A newly introduced pest usually increases out of all due proportion during the first few years of its existence in a new country, after which it gradually reverts to more normal conditions, There are already indications that this process is taking place locally. The particular village [Halawegoda] in which the pest is said to have originated, is at the present time not nearly so thickly infested as are some of the neighbouring villages to which the pest has since spread. A similar levelling process has apparently occurred in other countries into which this snail has been introduced." In spite of the early date in which these ideas were advanced, and the clarity with which they were presented, the literature has been essentially devoid of any reference to the phenomenon of population decline until the past few years. In discussing the introduction of the giant toad Bufo marinus, Townes (1946) significantly pointed out that the populations of this predator dwindled after an initial build-up and that the average size of the animal diminished. Williams (1951) learned in Zanzibar that the several species of Achatina on that island years ago caused damage to crops but that now the farmers pay little heed to them. It was the words of Green that persuaded the author in 1949 to write to a number of investigators to inquire about the status quo of the problem in the areas where A. fulica had been established the longest. Although there were differences of opinion expressed in the 180 POPULATION DECLINE 181 response, even from different people in the same area, by far the majority indicated that there has been noticed in recent years a definite reduction in the number of the snails and, therefore, in the problem itself. Several people, apparently independently, observed that the decline was noticeable in the older populations, but that the newer populations were just as vigorous and damaging as the older populations had been in their early stages. A few reported conditions suggesting an essential decimation of the snail population apparently through the effect of some grossly adverse factor. Most of these reports came from correspondents in Ceylon. In New Britain, however, G. S. Dun (in litt. Nov. 5, 1951) stated that in Kavieng the giant snails used to be very abundant, that recently they have decreased so in numbers that vegetables can once again be grown, and that "high piles of moribund snails" have been encountered. Two main facts, then, came to the fore in these reports, viz., that population decline in A. fulica is actually in process at least in some areas, and that an unfavorable factor or combination of factors is responsible for a catastrophic effect upon certain populations. It was eminently apparent that a thorough investigation of the phenomenon of population decline was needed. Because of the preponderance of positive reports from Ceylon, because the infestation in that country was over fifty years old, and because the majority of known predators of this snail was in that country, a proposal was made to set up in that area an extensive research project. This project was conducted by the author under a grant (NSF-G519) from the National Science Foundation, Washington, D.C., for a period of nine months in 1954. A preliminary survey in Ceylon quickly verified the earlier reports of population decline. In fact, populations were found in surprisingly great variety. Some, such as those in Dambana and Balangoda, were in the early fulminating stage. Others had been in progress for thirty to fifty years and were either stabilized, declining, or apparently extinct. When all information was pieced together, a generalized picture of the population "life cycle" took shape. After firm, unobtrusive, insidious establishment, the giant snail appears almost explosively in great numbers. This is revealed by the fact that first announcements of newly discovered populations are usually made in an urgent, almost frantic manner with descriptions depicting snails literally all over the place. The numbers continue to build up despite control measures in various combinations. The population then seems to stabilize at a high level for an indefinite period of time. At this time, poisoning campaigns in some cases paradoxically seem to favor further build-up of the snail. Ultimately a gradual population decline takes 182 POPULATION DECLINE place. Although there are subsequent periods of recrudescence, after a certain stage the population rarely or perhaps never attains the level of the first peak. With this progressive decline in numbers, there is a concomitant decline in the average size of the individual snails so that instead of a population of true "giant" forms from 4-6 inches, they may be quite modest-appearing forms, ranging from 3-41/2inches. In some cases, however, the population may go into a sudden, unpredictable, and pronounced slump with the result that live specimens can be found only after prolonged hunting, or not at all. More than a simple decline is obviously involved in these latter cases. Somehow the decline has been augmented and intensified until extinction or near-extinction has occurred. Herein lies the very heart of the problem of population control. The problem of attempting to determine the causal factors in the phenomenon of decline is not going to be an easy one; for it is an ecological problem, and ecological problems are notoriously complex. Several different areas of investigation have been suggested as an approach to the problem. These are treated under separate headings below. Population Senility Some investigators have attributed decline in the populations of the giant snail to a mythical "population senility." In so doing, they merely beg the question; for in effect they have done nothing more than give an ambiguous term to the phenomenon (Mead 1955b). To assume that all populations will eventually go into a state of senility and reduced vigor does not get to the matter of causation. Sterility On some occasions in the past, investigators have encountered situations wherein, even with diligent hunting, not a single mature snail specimen could be found in the gravid state. Both Kondo and the author have examined snail populations in the Pacific reflecting this characteristic (e.g., the Auluptagel population in the Palau Islands). In attempting to find an explanation for this, some have assumed that a physiological process of sterility had entered the picture and was effecting a population decline through reduced fecundity. The apparently favorable nature of the environmental conditions, however, may have been misleading; for unsuspected factors may have been responsible for precipitating a perfectly normal, temporary period of generalized reproductive inactivity. There probably is a decline in the reproductive capacity of older individuals in any population; but this, by itself, cannot explain population decline. In the gigantic individuals, which may not necessarily be gerontic, there is fair evidence of sterility. A correlation be- POPULATION DECLINE 183 tween gigantism and sterility in mollusks has been emphasized by Pfeffer (1928). This same correlation has been offered to explain the fact that gigantic forms of A. fulica not infrequently appear in the early sigmoid growth stage of the population, but quickly disappear, never to be seen again. It perhaps is therefore relevant that Kondo (1952) found gigantic individuals most prominently at the "forefront" of the Laguna population in Guam. There has been found in neither the field nor the laboratory any evidence of sterility through metazoan parasitism (cf. Szidat 1941). Starvation Chamberlin (1952a) concluded in his ecological study of A. fulica on Tinian that starvation was the main cause for the great number of dead shells found on that island. Although this same idea previously had been less positively expressed (Mead 1950b, c), it would seem now, in view of the catholic appetite of this snail and its faculty for withstanding for a long time complete lack of food, that starvation could not possibly explain population decline except in limited cases under most unusual circumstances. The same reasoning would apply to the matter of nutritional deficiences, including the very important calcium. As a contributory factor, starvation could have its effect; but alone, it would not be generally decisive. Exposure Practically without exception, the investigators who have made and reported upon a survey of the problem of the giant African snail have emphasized the apparently high incidence of death by exposure to direct sunlight. This point has been elaborated upon above. Under the prevailing conditions in most infested areas, it is not understandable how this factor could have anything more than an incidental effect except in cases of severe overcrowding in marginal areas. And even in the latter cases, the factor would be selflimiting. The fact remains, however, that there appears to be an inordinate amount of death by this means in some areas. As a possible explanation, an unknown second factor may be contributing to their general unthriftiness causing them to make a less vigorous, and therefore less effective, escape from adverse conditions. Traumatic Breaks Snails inhabiting rough, rocky terrain are found in a high percentage of the cases to have shells which have been broken and mended and broken and remended. The unsightly appearance of these snails has encouraged some observers to conclude that they were unthrifty and hence were giving way to a population decline. Later, it was suggested that some "other factor," such as a genetic, a pathological, or a nutritional factor, was present to produce the thin, brittle shells which in turn were more prone to fatal or traumatic breaks. But whatever the explanation, the manifest ability 184 POPULATION DECLINE to withstand even severe breaks and the high reproductive potential in this species surely prevent traumatic breaks from producing a decline effect upon the population. In fact, some of the most vigorous populations examined in the Pacific islands were of this fragile, broken-shell type. Predators Because the predators are the most tangible of the unfavorable factors in the environment of the giant snail, they have been seized upon by both the casual observer and the professional investigator as the most likely factors in effecting the observed decline. The Indian glowworm, for example, is believed by many to be largely if not entirely responsible for the fact that A. fulica is now not nearly the pest that it used to be in the areas where this predator is endemic. It is true that the glowworm is obviously more common in some infested endemic areas than it ever could have been without the giant snail. A distinction between cause and effect, however, has not been made. Nonetheless, this predator and other predators believed to be capable of holding A. fulica into at least a simulated state of population decline have been used, and will continue to be used, as potential agents of biological control in non-endemic areas. That they will accomplish their mission remains today a moot question indeed. But irrespective of this point, there is still the fact that population decline is manifesting itself in areas where predators are either unknown or nonexistent. Undoubtedly the majority of investigators feel that predators could have an appreciable augmenting effect in the phenomenon of decline, but that predators, by themselves, are not responsible for the current widespread decline in the populations of the giant snail. Genetics The subject of genetics and its possible relation to the unthrifty nature of some giant snail populations, and therefore to population decline, was first introduced by Mead and Kondo (1949). Subsequent reports by these authors presented further evidence from the field in support of a genetic rather than ecologic explanation for some of the observed phenomena and emphasized the need for a genetic study of this snail species. Variations in individuals, and in populations in particular, were explained on the basis of possible genetic origin. Some of these variations, such as the malformed specimens (roughly in a 1:3 ratio!) in the Army Hill population in Saipan, have been interpreted by some observers as indubitable signs of population degeneration and decline. If the explanation is a genetic one, as suggested above in the discussion of variations, then the variations would contribute to population decline only if they interfered with the normal development of the population. As pointed out POPULATION DECLINE 185 above, gigantic forms show signs of being reproductively sterile. A genetic linkage between gigantism and sterility would have a qualitatively modifying effect upon the development of the population. On the other hand, if the "climax type" in a normal-appearing population possessed a reproductive inadequacy of genetic origin, there would follow a quantitatively modifying effect upon the subsequent development of the population. A population decline would be inevitable and proportional to the severity of the inadequacy. The inferences are tempting in the extreme; for all observed phenomena in the field can be explained plausibly through the medium of established genetic principles. Unfortunately the entire subject, as it pertains to A. fulica, has proceeded scarcely beyond the purely speculative stage. Disease Annandale (1919) in India was the first investigator to report the belief that A. fulica was "subject to some kind of fatal epidemic"; but he did not suggest that there was any connection between it and a general population decline. South (1926b) accepted this report and listed the "fatal epidemic" as one of the controlling factors in the biology of the giant snail. Early in 1947, Daniel B. Langford, assistant entomologist of the Trust Territory of the Pacific Islands, reported that many achatina populations in the Pacific area seemed to be changing character; that many of the individuals had thinner, more fragile, often badly distorted, lighter-colored shells; and that reproduction and numbers of individuals were noticeably reduced. This syndrome suggested to him that a disease was present. ''Diseased'' specimens from Koror, Palau Islands, were sent to investigators at the University of Hawaii in the hopes that disease-producing agents would be found. The findings were inconclusive. According to Langford, "diseased" specimens were then introduced in the achatina populations in Oahu. When these populations were examined by the author in January, 1955, there was no tangible evidence of the described syndrome in any of the specimens. In January, 1949, Langford introduced from Koror "diseased" specimens in the Anigua population, near Agafia, Guam. Kondo (1950c) examined this population and determined that the typical syndrome was present. The shells of specimens that had ben dead for at least several months were compared with the specimens showing the syndrome. There were striking differences. Kondo pointed out that one might be tempted to conclude that the ''disease" had been effectively introduced; but he emphasized that there was just as convincing evidence that two genetically different types had interbred. 186 POPULATION DECLINE It should be pointed out, however, that elsewhere in the Pacific islands there were noted just as pronounced differences between the ancestral shells and the living specimens. This suggests that perhaps the so-called "diseased" specimens are nothing more than representatives of a stage in the genetic evolution of the population. If this reasoning is correct, and if the populations are genetically similar, then probably the Anigua population would have appeared just the same had Langford not introduced the Koror specimens. This also would make more understandable the failure of the syndrome to make its appearance in the Oahu populations. The reports of Langford created the first wave of optimism in the problem of the giant African snail. But this optimism was short-lived. In addition to the failure of investigators to find any pathogens in Hawaii, Lange (1947, 1950) early reported in his survey of the problem that he saw no evidence of an epizootic. He sent preserved specimens to the author for examination; but the gross anatomy showed no signs of pathology (Mead 19500:237). An examination of fifteen of Langford's original "diseased" specimens from Koror revealed the fact that, in spite of their nine months in Guam in open bell jars without food or water, all but one was very much alive and apparently none the worse for the ordeal (Mead 1950b, c). During the summer of 1949, many specimens from a number of different populations were examined in the Pacific Islands. Instead of pathology, genetics appeared to form a more logical basis for explaining the common syndrome (Mead and Kondo 1949). With the announcement of these findings, the hope of using a parasitic disease to control A. fulica was virtually abandoned and renewed efforts were made to explore further the possibilities of using predators as biological control agents. That there actually exists a disease of A. fulica continued to remain very much a possibility. The reports of Annandale and South, and the frequent reports from Ceylon reasserted that there was still more to the story. An exhaustive check of the literature brought to light the fact that practically nothing is known about gastropod pathology. The works of Szabo and Szabo (1930 et seq.), despite their ambitious titles, concern almost entirely anatomical anomalies and gerontic changes. Fromming (1954a) makes the same type of contribution. A presumed "contagious disease" of H. aspersa is reported upon at length by Boycott and Oldham (1938); but although their discussion is convincing, it is not conclusive. Aside from the fact that no microbiological investigation was made, the factors of genetics and nutrition were not adequately controlled. Muma (1954) states that the tree snail Drymaeus dormani in Florida is subject to a dis- POPULATION DECLINE 187 eased condition of possible bacterial origin which causes the snails to turn a greenish color and die. Further work on this problem is currently in progress. Dr. C. Vago of the Station de Recherches Sericicoles in Ales, France, writes (in litt. Mar. 23, 1955) that in the past few years he has studied over ten different diseases of several species of snails. His research so far has convinced him that a paracoli occasionally goes into a highly virulent form and produces an epizootic among the commercially raised helicines. Recently, one such epizootic killed over 50,000 snails in "snail parks" where the snails are raised in close confinement. It was obvious at this point that investigating the possibility of disease being a causal agent in the decline of the giant snail population in Ceylon would have to be essentially a pioneering work. First of all, the extent and effects of predation had to be determined. This factor was announced early in the investigation as not being decisive (Mead 1955b). By far the most effective predator was the Indian glowworm. But this attacked only the smaller specimens of A. fulica. It would be expected that any snails that escaped the ravages of the glowworm during the first two months of their lives would be too large to be attacked and, in the absence of other killing factors, would eventually become gerontic individuals in a population manifesting heterogeneity with respect to the size and age range of its individuals. Yet in some areas where this predator was commonly encountered, the giant snail population was in a truly remarkable state of homogeneity in that all individuals were clearly under one and one-half years of age. In others, the snails were not over two to three years of age. Older and larger individuals, which would be immune to the attacks of the glowworm, were completely absent, although their old, worn, dead shells were all too common. It was clear, then, that something more than predation was operative. An epizootic of decimating proportions, for example, or a chronic disease that is aggravated with the increased age of the host, could definitely produce the observed homogeneity in the population. Fitting into this suggestion is the fact that in these same areas in Ceylon there was a serious build-up of the snails in 1951 and such a sharp decline in 1952 that control measures were not undertaken. These populations then had had one and one-half to two years to recover from the decline--a period of time approximately equivalent to the age of the average individual. As a further pertinent point, it should be borne in mind that micro-organisms of disease, unlike predators, have the reproductive capacity to produce quickly a catastrophic effect upon a population (Mead 1955b). 188 POPULATION DECLINE With epizootiological evidence building up in favor of the disease hypothesis, a great many specimens were brought into the laboratory to be dissected in the living state for a critical anatomical examination. The results were discouraging, for the viscera and the tissues in general appeared within the range of normal in practically every case. A break finally came with the advent of the summer rainy season. At that time, snails were seen in great numbers during the day crawling about in the rain. As they were being observed in the field, it was noticed that one out of every two to three specimens had leukodermic lesions on the tentacles and, occasionally, on the face and neck. Surely these were symptoms of disease. The gross anatomy of these specimens was examined in the laboratory, but the findings were negative as before. This suggested on the face of it that there was no visceral phase of the disease. At this stage of the investigation, the syndrome was described at the meetings of the Second Pan-Indian Ocean Science Congress in Perth, Australia, in August, 1954. Subsequent histopathological examinations of the hepatopancreas and the kidney, however, revealed pronounced differences between specimens manifesting symptoms and specimens without discernible symptoms. Corresponding changes in other elements of the viscera were suspected. As more and more specimens were examined, a definite sequence of progressive stages in the disease began to take shape. Leukodermia is caused by the systematic destruction of melanophores in a small locus in the dermis. The melanophores first appear fractured, then reduced to a mass of granules, and finally disappear entirely. Because of concomitant tissue destruction and weakening in the dermis, the lesion may become elevated into ridges, tuberculations, or horns; and as the lesion enlarges, it may coalesce with adjacent lesions. Eventually the tentacle becomes shortened and distorted and, finally, it may remain partially or completely invaginated. Occasionally the initial lesions may diminish in size, but other lesions appear elsewhere on the forepart of the body. It should be remembered that the severity of the "liebespiel" is responsible for leaving pigmentless scars on the surface of the body. Although they are clearly distinguishable from leukodermic lesions, the lesions might easily be confused by the casual observer. As announced in the preliminary report of this disease (Mead 1956a), the whole epizootiologic picture is that of a chronic enzootic disease of uncertain etiology, high incidence (35-68 per cent), and low gastropod host specificity. Observations in the field and laboratory suggest that the disease is spread through contact, that it is highly contagious, and that the immunity is low. A great many natural factors in POPULATION DECLINE 189 the environment favor the development of an epizootic. There characteristically is a high moisture factor during the periods when the snails are active; the traumatic breaks in the shells would allow additional portals of entry for disease agents; the feeding habits of the snails include eating the substrate, consuming dying or dead individuals, and mutual rasping of slime from the surface of the body; and their habit of aggregating in moist, protected niches increases considerably the element of physical contact. In the areas in Hawaii where A. fulica abounds, certain terrestrial arthropods, for example, the amphipod Orchestia platensis, the isopods Porcellio laevis and Metoponorthus pruinosus and unidentified collembolans, frequently have been seen at night crawling over the extended achatina and apparently feeding on mucus on the surface of the body. Their presence implicates them as possible agents of contamination if not actual vectors. A transovarian mode of transmission, however, still remains a strong possibility. The various stress factors in the environment surely must provide an aggravating influence, perhaps even to the extent of precipitating an acute phase of the disease. Specimens kept for six months under conditions demanding estivation showed a much slower rate of progress of the disease; and, although there were new lesions, in some instances the old lesions appeared to be undergoing tissue repair with irregular areas of excessive melanosis. Apropos of this, Ghose (1959) has reported that tentacular regeneration in this species proceeds much slower during estivation. The course of the symptoms moved noticeably faster in specimens kept under conditions of high humidity and abundant food, which encouraged prolonged periods of activity. The progress of the disease is clearly linked with metabolic rate. From this point, the complications, implications, alternatives, and speculative possibilities literally run out in all directions. Obviously, a great deal of pioneering research is required to produce any sort of conceptual perspective in this complex problem. As a good start in this direction, a United States Public Health grant (E-1245 [C3]) has been made in support of a projected investigation of the etiology and pathology of this disease. The etiological research completed in Ceylon has essentially eliminated from further consideration protozoans, malacogenous fungi, yeasts, and spirochetes. Hence, attention is being focused on the bacteria and viruses. Similar leukodermic lesions were found in specimens of A. fulica in Singapore, Hong Kong, and Hawaii (cf. van Zwaluwenburg 1955). Because of the high rate of infectivity in the Ceylon populations, and because the snail infestations to the east of Ceylon undoubtedly were 190 POPULATION DECLINE started by specimens from Ceylon, it has been provisionally assumed that the disease has been carried with the snails and that similar symptoms mean the same etiology. There is evidence, however, suggesting that different "strains" of the pathogen exist. For example, the infected specimens in Hawaii show a conspicuous paucity of tuberculations and an inordinately high incidence of large lesions, 3-6 mm. in diameter, located at the base of the ocular tentacles (67 per cent of positive individuals). Even in Ceylon, groups of infected specimens from different populations displayed noticeably different tendencies in their symptoms; in some, facial lesions were common, in others, they were almost completely absent; in still others, the lesions were conspicuously absent from the ventral tentacles. An explanation for these differences conceivably cannot rest entirely with the pathogen; for there must be differences in host response stemming from the different genetic make-up of the various populations. The difficulty in distinguishing between possible strain difference and difference in host response was most dramatically brought out in an experiment wherein the pathogen was transferred from A. fulica to Limax flavus to Helix aspersa. Out of 106 inoculated specimens of H. aspersa, 14.15 per cent were unmistakably positive; but the heretofore unobserved symptoms in H. aspersa were vastly different from those of A. fulica inasmuch as the leukodermia was generalized rather than localized and the tentacles were atrophied rather than distorted. From what was known of the distribution of the disease, it was assumed that it would probably be found in all of the infestations east of Ceylon and India (Mead 1956a). More recent support for this assumption was obtained when specimens of A. fulica were collected in Bangkok, Thailand, in December, 1957, and found to be diseased (Mead 1958a, b). In contrast to the syndrome in the Hawaiian specimens, however, leukodermia was much less pronounced and tuberculation was severe. At that time, the only inkling we had regarding the possible existence of the disease in the Pacific island specimens was a casual remark of Chamberlin (1952b: 12) tucked away in a footnote, viz., "Occasional individuals had pigmentless blotches on the skin but examination of these showed no coincident anomalies." As good fortune would have it, the return trip from Thailand was routed through Guam and it was possible to examine a number of giant snail specimens in the vicinity of the military airport. These were found to have the disease syndrome more nearly like that of the Thailand specimens than the Hawaiian specimens. This information suggests strongly that POPULATION DECLINE 191 all other Pacific island populations will be found to have the disease in one form or another. It is of interest to note that A. J. Kohn, during the 1957 Yale Seychelles expedition, found typical symptoms on specimens collected on Hitadu Island of the Addu Atoll in the Maldive Islands. On the other hand, Ghose (1960) failed to find any specimens in India with discernible symptoms; but his field work seems to have been quite limited. It is not known whether the disease occurs in the East African achatinas; but it is felt that more than likely it does. Finding a strikingly similar syndrome in the introduced achatinid Rumina decollata in central Arizona (Mead 1959b) has precipitated the feeling that the disease of the giant African snail is not unique, but one of a whole class of diseases of varying pathogenicity found in many species and in many parts of the world. It is a strong temptation to extrapolate further to assume that it is these diseases which successfully hold most snail populations in check and produce a major effect in population fluctuations. Actually, it would be a paradox indeed if snails did not have their own diseases. It is easy to predict that a number of snail diseases will be found and reported upon in the next relatively few years. When the effects of these are known, infinitely more will be understood about establishing balance in invading populations. After the etiological agent has been discovered, it will be important to determine its affinity for forming different strains of varying pathogenicity. An understanding of the pathology must be sought at the histopathological level; but as Pan (1958) has emphasized, this has as its prerequisite an understanding of the normal histology. Thanks to the meticulous and exhaustive work of Ghose (1960), we know now much about the gross anatomy, developmental anatomy, and histology of A. fulica. Similarly, an understanding of the epizootiology depends upon a rather considerable knowledge of the ecology, and particularly of the population dynamics, of this important species. We are less fortunate here; for although there have been recorded many observations made at the time the invading populations were at their fulminating stage, practically nothing is known about the normal course of development of a population. The field investigations of Mead and G. D. Butler, Jr., in Hawaii are contributing to our knowledge in this respect; but considerably more basic work is urgently needed. So far, the epizootiological data recorded from examinations of the unusually heavy population of snails in the Mahinui area of Oahu since 1955 have given us our best, but by no means clear, conception 192 POPULATION DECLINE of the course and effects of the disease on a relatively recently established giant snail population. In January, 1955, the incidence of the disease was 14.5 per cent and not a single specimen could be classed as having "severe" symptoms. One hundred adult specimens collected at random in August, 1957, indicated that the incidence had increased to 57.0 per cent and specimens with advanced symptoms were common. Exactly one year later, the incidence had raised to 68.6 per cent and the majority of the positive specimens had prominent lesions. Throughout this period of time, the snail population level remained essentially stable. In December, 1957, 400 specimens collected at random were measured, weighed, examined to determine and record the presence and extent of disease symptoms, and released in the area where they were collected. In the following August, 132 (33 per cent) of these specimens were recovered and reexamined. The new data demonstrated that in the intervening months negative specimens had become positive and positive specimens had either become more severe or had disappeared. In essence, a far bigger population turnover had been taking place than was ever suspected; and the high reproductive potential had apparently insured the maintenance of the population load (Mead 1959b). This information is compatible with the earlier conclusion that the disease reduces life expectancy or, conversely, that age acts as a decisive stress factor in the presence of the disease. The disease, however, has not just been spreading from one achatina to another; experimental evidence supports the assumption that natural transmission has taken place between A. fulica and Bradybaena similaris and Subulina octona in this area. These latter two species were completely negative for symptoms in other areas in Oahu which had not been invaded by Achatina. Identical sampling techniques in the Mahinui population in September, 1959, revealed the following facts: the incidence had increased to an amazing 83.0 per cent; the vast majority of diseased specimens were of the "severe" type with one or more large (5 mm.+) leukodermic lesions; for the first time several specimens were found to have multiple lesions (12+) on the exposed parts of the body; and, also for the first time, there was a marked population decline. There is good evidence that the disease becomes an increasingly more effective limiting factor only after the snail population has begun to level off following the sigmoid growth stage. Research is being continued in the Mahinui area, but unfortunately the predatory snail Euglandina has recently invaded the experimental site as an incalculable variable. POPULATION DECLINE 193 M u l t i p l e Factors It is obvious that under natural conditions, unfavorable environmental factors are mutually augmentative but operate wholly independently. At any time in any one place, some may be producing the maximum unfavorable effect while others are temporarily ebbing or remaining neutral in their effect. But, as in any other system of independently operating variables, there are inevitable periods of synchronization. If a period of synchronization is at a time when the unfavorable factors have reached their maximum intensity, then the effect upon the population will be most severe. In some animal groups, this could mean local extermination, particularly when man-made unfavorable factors are added to the natural compliment. Possible instances of this nature have been observed in Ceylon and in the Pacific islands. A case in point is the population of A. fulica at the Dodangoda post of the Godahene estate near Neboda, Ceylon. In 1951, the giant snail was so abundant that the leguminous cover crop Pueraria was almost completely denuded. A program of poisoning the snails with metaldehyde bait was immediately initiated. By 1952, the population of the snails had diminished to the point where f u r t h e r poisoning was not considered necessary. By 1953, the snails were scarcely in evidence and the local people literally ''forgot about them" in that area. In 1954, the author made a lengthy and exhaustive search of the area during the rainy season and in the rain. Not a single live specimen could be found. Nor could any other snail be found alive, including the ubiquitous and common Subulina octona. Empty shells of A. fulica were there in abundance, but their weathered condition indicated in every case that death had occurred many months previous to that time. No predators were found; but a large number of weathered but characteristically broken shells near bandicoot burrows indicated that this mammalian predator had been active in the past. More recently, the bandicoots had turned to the large, rich seeds of the rubber plant for their food. With no live snail specimens present, positive evidence of the disease was not discoverable. However, it seemed more than significant that live snails in the nearby Pettigola section of this estate had the highest infectivity rate (64 per cent) that was found in Ceylon. Surely, then, the chief unfavorable factors were poisoning, predation, and disease. Possible unfavorable meteorological conditions could not be determined beyond the fact that rain had been unusually abundant during the preceding months. There appears to have occurred in that area virtual or actual extinction not only of A. fulica and other introduced snails but of 194 POPULATION DECLINE the endemic snails as well. Cause for this cannot be laid wholly at the feet of the poisoning program as their program was of such a scale that it could not possibly have been eradicative. In the absence of any further information, one can scarcely escape the plausibility of the assumption that the disappearance of the snails was caused by the compounding of unfavorable factors. But, more than a simple summation of the unfavorable effects, there actually may be a synergistic action in certain combinations of factors. If there is a formula for producing an eradication of this snail pest, it probably will be found in the realm of multiple unfavorable environmental factors. CHAPTER 11 OUTLOOK In this work so far, definite attempts have been made to determine and evaluate the status of the several aspects of economic malacology and the problem of the giant African snail in particular. At this point, it will be well to look ahead and, on the basis of what is known, to make prognostications which should indicate not only where we are going but what we should be doing in the host of problems presented by economically important snails. Dispersal A question of great concern in the outlook for the problem of the giant African snail is, "Just how far can and will this snail spread into the uninfested areas of the world?" First of all, we know full well that we have in Achatina fulica an exceedingly hardy, tenacious, variable, and adaptable molluscan pest with a high reproductive potential and remarkably few natural enemies. Since the limits of these faculties are not known in any case, the question at hand, unfortunately, can be answered only in speculative terms. Even so, information is available which forms a firm basis for speculation. Finding A. fulica firmly entrenched in Ani Jima (lat. 27° 08' N.) in the Bonin Islands is of real significance; for it is situated farther n o r t h than any other known established population of this species, with a possible exception suggested by the recent discovery of specimens on Amami Oshima (lat. 28° 18' N.) in the Ryukyu Islands. The well-established populations in Okinawa Jima of the Ryukyu Islands are almost as far north (lat. 26° 03'-51' N.). Nuttonson (1952) has established for the Ryukyu Islands the agro-climatic analogues in the Northern Hemisphere. From these it can be stated flatly that A. fulica could become established in the comparable areas of Florida, Alabama, Mississippi, and Louisiana. Because island climates 195 196 OUTLOOK are characteristically more mild than continental climates at the same latitude, it is to be expected that the northern and southern frontiers of this species will be populations in islands which have a subtropical-temperate climate. If, however, the giant snail adapts itself as a greenhouse pest, it may become established in the more temperate continental areas. Significantly, the populations in the Bonin and Ryukyu Islands are vigorous and thriving. Further, the winter climates in these islands and in the infested areas of Kauloon in Hong Kong Colony, Amoy in China, and interior Formosa, are considerably more vigorous than in some of the areas in East Africa where A. fulica is endemic. This suggests that the snails have not yet reached the threshold of minimum tolerance to cold. From what we know, it is reasonable to assume that the giant snail has the capacity to establish itself in all oceanic islands which have an adequate cover of vegetation and fall between 30° north and 30° south latitude. This embraces the greater share of Oceania and indicates that such uninfested island groups as the Marshall, Gilbert, Ellice, and Line, are vulnerable. Uninfested continental areas which are tropical, subtropical, and, in some cases, subtropical-temperate are similarly vulnerable. This includes northern and much of eastern coastal Australia, as Harrison (1951) agrees. The most extensive of the suitable but uninfested areas, however, are in the New World where they form an essentially continuous belt from the Gulf States of the United States south, through the West Indies and the lower altitudes of Mexico, to the greater share of South America. That this snail, so far as we know, has not yet become established in these vast areas is almost a miracle, especially when one considers the general inadequacy of plant quarantine regulations in the many countries included. In fact, some investigators have been moved to suggest that there is some basically intolerable factor in the tropics of the New World which has kept this snail form becoming established. Under the circumstances, however, it appears that it is merely a matter of time until these areas are invaded. In the Pan-Indian Ocean area and in Indonesia, the giant snail has already made considerable progress in its spread. It will continue this spread until it is essentially ubiquitous. Australia, northern New Zealand, southern Japan, and continental United States are peripheral areas about which there is considerable doubt because it is not known to what extent A. fulica has the genetic capacity to form populations of more hardy individuals. Masahiko Nakada of the blight prevention section of the agriculture-forestry ministry in Tokyo, however, predicted recently that the giant snails will make OUTLOOK 197 their appearance in Kyushu, the south island of Japan, in the near future. In regions where there frequently occurs sudden lowering of the temperature below 50° F., this snail probably will not permanently establish itself except possibly in greenhouses and nurseries. On the other hand, where temperatures are reduced gradually and are not excessively low, it is not unlikely that the more hardy individuals will bury themselves in the ground and survive, as they do in the Bonin Islands. The inordinate sensitivity of this snail to higher altitudes seems to stem not from the factor of cold alone but quite likely from intolerance to a diurnal temperature fluctuation beyond a point where, to survive, it normally would need a period of several days to become conditioned physiologically. Of course, there may be a barometric factor; but it is significant that snails are surviving colder winter temperatures at sea level in Ani Jima than they can at 5,000 feet altitude in Ceylon (Mead 1955b, 1957b). Green (1910b), for example, reported that achatina specimens released at an altitude of 4,000 feet in Ceylon died one by one without reproducing. Other specimens released at an altitude of 6,600 feet in Masuri at the foot of the Himalayas in central India similarly failed to establish themselves (Benson 1858). Reliable correspondents have indicated that specimens have not been encountered much above 2,000 feet in Mauritius and 3,300 feet in Java. This information correlates with the findings of Pilsbry (1919:60) wherein achatinas were not found above 1500 meters in Africa. Although deserts would appear to be completely forbidding to the establishment of this and other snail pests, it has been emphasized that in the cultivated areas of deserts conditions are surprisingly suitable, as evidenced by the fact that several exotic terrestrial gastropods have become established in the Sonoran Desert of southern Arizona within the past few years (Mead 1952a, c, 1953, 1959a). During the past ten years, the threat of establishing the snail in the United States has come chiefly from ships carrying war salvage from the Trust Territory of the Pacific Islands (Messenger 1952:252). The interceptions of the snails have been discussed above under the topic of quarantines. There is essentially no war salvage arriving today in this country as remaining shipments are reportedly being directed to Asia. Concurrent with the tapering-off of these shipments, there has been a greater increase in the spread of the giant snail in Hawaii. The threat, therefore, is still coming in from the west, with the difference that it is now much closer to the United States, and to California in particular. Fortunately, the recently enacted federal 198 OUTLOOK quarantine regulations will provide a considerable buffer effect to this threat. But there are no prospects of the threat becoming less in the near future. The actual threat that this snail presents to the agricultural state of California has been much debated. Bequaert (1950b:74) has stated, "It is, in my opinion, extremely doubtful that A. fulica could ever become established in California or elsewhere in the continental United States, to judge from the present distribution and the ecology of this snail." In direct contrast, Mead (1949b, c] has stated, "It would be especially bad in places like California, the Gulf states, and the Southeastern states." On the other hand, Abbott (1949) has stated, "Fortunately, the climate of that state [California] is not considered to be entirely ideal for the snail." The following year, he modified his stand with, ". . . it is my opinion . . . that the southern third of Florida is the only suitable area for its survival for any great length of time." These differences of opinion are the natural outcome of having to base predictions on woefully inadequate information; and they underscore the advice Bequaert gives in the following words, ". . . controlled experiments, under strict supervision, but under natural conditions in a presumably favorable area, are called for in order to determine once and for all whether or not the snail could survive the winter and reproduce freely in the United States." Let us assume for the moment that the only place in the United States A. fulica could become established is the southern tip of Florida. Let us assume further that at best only an unthrifty outpost population of the giant snail, causing little local damage, could become established. Nothing but a dim view could be taken of such a situation; for the snail pest could use that beachhead as a springboard to start populations elsewhere in more favorable areas in the western hemisphere. Cuba, being nearby, and having vast agricultural areas and a fabulously rich mollusk fauna, would be particularly vulnerable. Aguayo (1950) has already sounded a word of warning in that country. Since man is the principal agent of dispersal of this snail pest, any measure to check the spread will have to be aimed at man himself. Quarantine regulations have been set up for this purpose; and for this purpose they should be maintained indefinitely. They are far from being universally practicable; but in certain areas they have been eminently successful. Keeping this snail out of California in spite of the frequent interception of live specimens, and restrict- OUTLOOK 199 ing it to the two originally infested islands of Oahu and Maui in the Hawaiian Islands for a period of twenty years, are cases in point. Bionomics We know amazingly little about the biology and ecology of the giant snails--far less than for any other agricultural pest of comparable economic importance. We do not know how long they live, how far they travel, how much heat or cold they can tolerate, why they are limited by altitude, whether they can go into true hibernation as well as estivation, why they move en masse out of a favorable area into an apparently unfavorable one, how important their own diseases are in regulating their numbers, what the biological factor is that accounts for the tremendous vigor early in the invasion of a new territory, or very much about literally a score of other equally significant but unknown factors operative in their existence. These thoughts bring out the obvious realization that just as soon as comprehensive behavioral study can be made of these snails, there will be not only an increase in the efficacy of the existing control measures but new and more effective means of control can be conceived. The unpublished thesis of Ghose (1960) gives us our most valuable contribution on the anatomy, histology, and embryology of A. fulica. Hatai and Kato (1943) made detailed studies of the growth and development of this species. Their ecological observations are valuable, but as is the case in so many other reports based on field observations of A. fulica, the bits of information amount to such a pitifully small portion of the needed information that not even speculation will pull the parts together into any sort of even vague picture of the over-all ecology. The work of Takahashi (1942) gives us more specific information on the movement of this snail within a fairly limited area of a recently invaded compound; but like Matsui's report (1942) on the total amount of movement of the giant snail in a diurnal cycle, there apparently has been no translation from the Japanese. Edelstam and Palmer (1950) and Palmer (1951) of Sweden have made significant parallel studies with Helix pomatia; but like all parallel studies, they are scarcely more than suggestive. Control So far as is known, for over one hundred years eradicative measures of all types have been uniformly unsuccessful throughout the areas of the world invaded by A. fulica. Even the most intensive control programs, such as those in Oahu, Guam, Koror, and Hong Kong, have proven in the long run only to have impeded the spread of that snail pest. There has been, in fact, only one case where a major land snail pest has been eradicated by man, viz., Theba pisana in La Jolla, California. Basinger, who was in 200 OUTLOOK charge of the control program, spared nothing in his efforts to eradicate this pest. After a few years, he stated (1923a), ". . . the vast majority . . . have long ago been destroyed, but, when it conies to extermination, it is much harder to get the remaining few than the first hordes." Despite the fact that he thought he had virtually accomplished an eradication, as betrayed by the title of his article, it took almost exactly 27 years more actually to accomplish it! And during this time, almost unlimited manpower was made available to the program through federal and state relief administrations. The total cost of eradicating T. pisana and attempting to eradicate two other helicine species came to over $500,000. To date, the other two species are still very much at large (Messenger 1950). The factor of "the irreducible minimum," which is the difference between control and eradication, has plagued the program right from the beginning. With many examples of expensive failures in the past, it is little wonder that Green (1910c), South (1926b), Garnadi (1951), and others have taken an unequivocal stand against "all out" measures to eradicate A. fulica. Certainly their advice warrants serious consideration if the infestation has become firmly established in a favorable environment. If, on the other hand, the infestation is only in its initial phase, or is in a marginally suitable environment, eradicative measures should be undertaken. But in nearly every infestation, at least some measure of control should be exercised. The big problem is: "What control measure, or combination of control measures, should be used?" Naturally, the answer would vary according to the circumstances involved; but in general, all practicable control methods should be used where they are not mutually counteractive (e.g., the introduction of predatory snails and the use of molluscicides). Unfortunately, many of the control measures are relatively expensive and they therefore are completely beyond the consideration of most native peoples in the infested and potentially infested areas. Under such circumstances, hand picking, the construction of barriers, and other mechanical devices would have to be resorted to. But these, like any other control measure, must be maintained indefinitely. It is at this point that new difficulties arise. If after a few weeks or months of attempting to control the snail there is little or nothing in the way of tangible results, even the more fastidious people tend to give up and let the snails take over. But what is worse, peoples in many areas show a marked apathy, and even antipathy, for making any attempt to help themselves in this problem. For instance, R. E. Dean writes (in litt. Jan. 17, 1952) that in Hong Kong the government offered snail bait free to the people, "but the re- OUTLOOK 201 sponse was only lukewarm." The prevailing attitude among the people in the Pacific islands is that all responsibility for controlling the giant snail rests squarely upon the shoulders of the government. If anything is done, then, the government must resort to force, inducements through bounties, or government-supported large-scale control programs. All of these measures in the past have proven at best to be of only transient success. Chemical control has been demonstrated many, many times to be expensive and therefore practicable only on a small scale. In addition to the expense, tropical weather conditions in general are antagonistic to the use of molluscicides. Weather that is particularly favorable for snails is automatically unfavorable for chemical control measures as the rain washes away the molluscicide and encourages the growth of mold on snail baits; the high humidity favors recovery from the paralyzing effect of metaldehyde; and the lack of sun eliminates a killing factor that would ordinarily follow the paralyzing effect of the chemical. Dry weather is unfavorable to snails; but they escape it, and any molluscicides put out at that time, by going into estivation. But even beyond the expense and the weather, the fact remains that the available molluscicides are far from satisfactory in their effects. The tools of analysis devised by Basinger (1935), Thomas (1944), and others; the increasing interest of the bigger chemical companies in the development of new molluscicides; and the continued research by a larger and larger group of investigators, give us new hope that before long a truly effective, inexpensive, safe, and easily used molluscicide will be developed. As it is now, in the vast majority of the cases, after repeated chemical control measures are finally discontinued, the giant snail population builds up to the point where it is just as bad if not worse than it was before. In some cases where population decline is taking place or where there has been a fortuitous concomitancy of several control factors, chemical control measures appear to be decisive. This suggests immediately that there must be a better understanding of both the phenomenon of decline and the effects of combining unfavorable factors. In the broadest sense, man is an agent of the biological control of A. fulica because he is a natural controlling factor. Any hope of making use of the snails on a commercial basis for human consumption rests almost entirely with the native peoples who will readily eat them, viz., the East Africans, the Chinese, and the Formosan aborigines. These people have diets notoriously deficient in protein; and it would seem on the face of it that the giant snails could provide 202 OUTLOOK at least a partial answer to the problem. But in spite of the logic involved, the chance is remote indeed that there will be even a modest amount of commercial use of the snail for human consumption in the foreseeable future. The possibility of there being more use of the dried snail meal for poultry and livestock supplemental feeds is becoming steadily greater with continued research. It is only through the use of the snails in this manner that man will become a steady, dependable source of "natural" control. Control of this sort would be on an "individual" basis; and the program as a whole, as in other types of biological control, would be self-sustaining. The single greatest obstacle to this whole approach is the fact that it takes additional time and energy to prepare the snail meal. The native peoples, first of all, have to be convinced that this is worthwhile; and, second, they have to overcome their mutual loathing for handling the slimy snails. Progress is bound to be slow in this aspect of control. The other aspects of biological control are also in their infancy. The prospects of using radiation-induced sterility of males as a successful biological control weapon in insect control raises the highly provocative question of its possible use in snail control (cf. Knipling 1955, Baumhover et al. 1959). Whatever the results of introducing multiple predators of the giant snail in the Hawaiian Islands, they are bound to be significant, particularly since in a number of instances more than one species of predator has been released in the same experimental area. Euglandina, and possibly the smaller or both species of Gonaxis, will eventually become ubiquitous, spreading in many cases as far as Achatina has, but in a shorter period of time. The experimental island of Agiguan is only beginning to reveal its complex ecological story (Mead 1956b). With both Gonaxis and the phenomenon of population decline being variables in the current experiments on that island, the results are almost certain to be misleading one way or the other. Quite unfortunately, however, the research program on Agiguan Island was scheduled to be brought to a close in 1958 (Coolidge 1955). Disease at last has entered the picture. Its presence provides the most plausible explanation for the almost universally occurring decline in the older populations. The nature of the pathology and the etiology of the disease should be known in the near future. It is almost a certainty that the predatory snails Gonaxis, Euglandina, and other predatory snails used in molluscan biological control, will be found to be susceptible to the disease that has been discovered in A. fulica. Ultimately, not one, but many diseases will be found and considered in the light of biological control. Because of their affinity OUTLOOK 203 for developing strains of variable pathogenicity, disease agents will always have to be handled with respect and understanding in any biological control program. Decline When A. fulica is in a pronounced state of population decline, it ceases to be any more of a pest than some of the endemic snails. Determining accurately the causes for population decline in A. fulica will provide a whole new perspective in the control of this pest. When the entire story is known, it will undoubtedly be f o u n d that there are many factors involved, the sum of several of which will produce a decline. Probably the disease factor is the domin a n t one. Long ago, McCreery (1890) concluded that extinction was usually brought about by a large number of small conditions. If man is ever to eradicate the giant African snail on anything but the smallest scale, he will do it through the use, augmentation, acceleration, and intensification of the several factors, including disease, operative in producing the decline. The discovery of the secret of population decline in this and other species is of the very first importance, and research in the future should be directed toward this end. BIBLIOGRAPHY ABBOTT, R. T. 1948. The spread and destructiveness of the giant African snail, Achatina fulica. Nautilis, 62(1):31-34. 1949. March of the giant African snail. Nat. Hist., 58(2):68-71. 1950. Snail invaders. Nat. Hist., 59(2):80-85. 1951a. Battle of the snails. Science Digest, 30:15-19. 19516. Operation snailfolk: Pacific island biological control experiment. Nat. Hist., 60:280-85. 1951c. Report on the introduction of natural enemies of the giant African snail to Agiguan Island, Marianas Is. Invert. Consultants Comm. for Micronesia, Pac. Sci. Board, Nat. Res. Council. 11 pp. (Mimeographed.) 1952. New federal regulations on importing mollusks. Amer. Malacol. Union, Ann. Kept., 1952:9. ADAM, W., AND LELOTJP, E. 1943. Recherches sur les parasites des mollusques terrestres de Belgique. Memoires Musee Royal d'Hist. Nat. Belgique, 33(62): 1-40. ADAMS, L. E. 1938. Notes on enemies of mollusca. Jour. Conch., 21(2) :49. AGUAYO, C. G. 1944a. Leptinaria lamellata y otros moluscos introducidos en Cuba. Rev. Soc. Malacol. Carlos de la Torre, Habana, 2(2):51-58. 19446. Los moluscos comestibles de Cuba. Ibid., 2(1): 17-20. 1950. Caracoles trotamundos. Bol. Historia Natural, Sociedad Filipe Poey, 1(2): 75-82. ALBRECHT, P. G. 1921. Chemical study of several marine mollusks of the Pacific Coast. Jour. Biol. Chem., 45:395-405. 1923a. Chemical study of several marine mollusks of the Pacific Coast. The liver. Ibid., 57:789-94. 1923&. The reproductive system. Ibid., 56:483-87. ALICATA, J. E. 1938. The land-snail an intermediate host of the cecae [sic] fluke of poultry. Science, 88(2275): 129. 1950. Observations on the biology and control of a garden slug injurious to orchids in Hawaii. Bull. Pacific Orchid Soc. Hawaii, 8(3): 279-85. ALICATA, J. E., AND BESS, H. A. 1952. Fireflies may control flukes. Hawaii Farm Science, 1(3) :3, 6. ALLAN, J. 1949. The giant African snail--a crop-destroying traveller. Australian Mus. Mag., 9(11) :374-77. ALLEE, W. C., EMEBSON, A. E., PARK, O., PARK, T., AND SCHMIDT, K. P. 1949. Principles of animal ecology. Philadelphia, W. B. Saunders Co., 837 pp. 205 206 BIBLIOGRAPHY ALLEN, J. A. 1916. Edibility of Polygyra and Ariolimax. Nautilus, 29:143. ALPHEN DE VEER, E. J. VAN 1954. De agaatslak (Achatina fulica Fer.), een gevaar voor jonge bosculturen. Penggemar Alam, 34(1/2) :36. ALTSON, R. A. 1950a. Giant snail. Rubber Res. Inst., Malaya, Report for the Period Jan., 1941, to Aug., 1945, p. 67. 19506. Other animal pests. Ibid., Report for the Period Sept., 1945, to Dec., 1948, pp. 187-88. ANDERSON, A, W., AND TAYLOR, T. H. 1926. The slug pest. Leeds Univ. Bull., No. 143, pp. 3-14. ANDERSON, R. M. 1934. Effect of the introduction of exotic animal forms. Proc. Fifth Pacific Sci. Cong., 1:769-78. ANDRE, E. 1923. Le manganese chez les animaux lacustres abyssaux. Ann. Biol. Lacustre Bruxelles, 12:11-14. ANDRE, M., AND LAMY, E. 1930. Les acariens parasites des mollusques. Jour, de Conchy!., 74(3): 199221. 1931. Les acariens parasites des mollusques (notes complementaires). Ibid., 75(4): 322-27, ANDREWABTHA, H. G., AND BIRCH, L. C. 1954. The distribution and abundance of animals. Chicago, Univ. Chicago Press, 782 pp. ANDREWS, E. A. 1948. Achatina achatina (L.) lays its eggs. Nautilus, 61(3): 95-96. ANNANDALE, N. 1907. The distribution in India of the African snail Achatina fulica Fer. Records Indian Museum, Calcutta, 1:176-77. 1919. Mortality among snails and the appearance of bluebottle flies. Nature, 104:412-13. ANONYMOUS 1913. Cannibal snail, Nautilus, 26:132. 1914. Canned snails. Pure Products (New York), 10(5) :244. 1925. Jaarboek Dept. Landbouw Nijverheid Handel, Batavia, p. 61. 1927a. Une achatine vivante. Bull. Cercle Zoologique Congolais. Revue de Zoologie et Botanique Africaine, 4(3) :74. 19276. Control of snails in orchards. N. Zealand Jour. Sci. Tech., 9:96. 1929. Control of snail infestation in orchards definitely outlined. Calif. Citrograph, 14(9): 270. 1930. Slugs and snails. Queensland Agric. Jour., 34(1): 115. 1932. Further spread of the land-snail Achatina fulica. Nature, 129:692. 1936. A discovery of a new cure for consumption and kidney trouble: the use of giant snails. Translated by Rokuro Urata and Earl I. Naito. Osaka Mainichi Shinbun, April 19, 1936. 1939. Mauritius, Dept. of Agric., Ann. Kept., 1938:42. 1941. De agaatslak in Atjeh. Tropische Natuur, 30:173. 1942. Notes sur la preparation et le mode d'emploi (Tun app&t empoisonne centre les colimagons. Revue Agricole de Tile Maurice, 21(6): 271. 1945. Ainda o arua. Chacaras e Quintals, 71:362. 19460. Beware of the giant African snail. News Bull. Dept. Agric. Commerce, Manila, 1(6/7): 65. BIBLIOGRAPHY 207 19466. Egg laying snails menace three Pacific isles. The Sphere (London), 187: 400. 1947a. Giant snails numerous in Kokopo District, N. G. Pac. Islands Monthly, 18(2): 33. 19476. [Note] News Bull. Dept. Agric. Commerce, Manila, 1(11): 114. 1947c. Millions of snails. Pac. Islands Monthly, 18(2) :75. 1947d. Pests left by Japs: Giant snails in the Marshalls. Ibid., 18(3):22. 19470. Snail stew. Newsweek, 29:64. May 26, 1947. 1948a. Concerning snails. Pac. Islands Monthly, 19(3) :49. 19486. Fight snail with beetles. Sci. News Letter, 54(3) :44. 1948c. Giant snails. Pac. Islands Monthly, 18(7) :24. 1948d. Giant snails: Jap infestation not yet under control. Ibid., 18(12) :40. 19480. Giant snails: Scientists at work on most difficult problem in New Guinea. Ibid., 18(6) :71. 1949a. Cannibal snail, fearsome enemy of giant land snails. Hobbies, 54:155. 19496. Danger in curly shells. Pathfinder, 56(16): 24-25. Aug. 10, 1949. 1949c. Excerpt from the "Crown Colonist." Pac. Islands Monthly, 19(10) :48. 1949d. The giant snail. The Times (London), No. 51,517, p. 6. Oct. 20, 1949. 1949e. Giant snails repelled. Sci. News Letter, 56(18): 279. 1949/. Has the giant snail a food value? Pac. Islands Monthly, 19(9): 6. 19490. Ravenous snail. Time, 53(13): 74. Mar. 28, 1949. 1949A. Snails and slugs (Helix aspersa and Limax sp.). Agric. Gaz. N. S. Wales, 60(10): 539-41. 1949z\ War of the snails. Pathfinder, 56(6) :20. Mar. 23, 1949. 1949;". War on African snail. Sci. News Letter, 56(6) :91. 1953a. Kill giant African snails. Ibid., 64(4): 55. 1953&. Plant quarantine announcements: United States of America. FAO Plant Protection Bull, 1(4) :63. 1953c. Snail snoop. Amer. Mag., 156(5) :56. 1953d. Snails useful feed. Sci. News Letter, 65(13): 207. 1954. Colony of Singapore annual report, 1953. Singapore, Gov. Printer, 265 pp. 1956a. Biological control of Achatina by Gonaxis. Pac. Sci. Assoc. Information Bull., 7(6): 11-12. 1956&. Cannibal snail gets a job. Life, 40(16): 159-60. April 16, 1956. 1956c. The hunter snail. Time, 67(2) :44, 46. Jan. 9, 1956. 1956d. Proceedings of the third annual meeting of the Natural Science Foundation. Philadelphia, 19 pp. (Unnumbered.) 1959. All about garden snails . . . especially how to get rid of them. Sunset, 122 (3): 174-78. 1960a. Biological control, giant African snail, Hawaii. Pac. Sci. Assoc. Information Bull., 12(2) :9-10. 19606. Transactions of the first international conference of insect pathology and biological control. Praha. August 13-18, 1958. 653 pp. (Sep. printed.) ARAUJO, H. B. DB 1952. Controle as lesmas. Revista Agronomica (Porto Alegre), 16:363-66. ARIZONA CODE. Chapter xlix. Agriculture and Horticulture. 1939. ARMITAGE, H. M. 1949. Bureau of entomology. Thirtieth Annual Report. Calif. Dept. Agric. Bull., 38(4): 176-77. ARNOULD, C. 1933. Un elevage et un parcage pratique des escargots. I/Agriculture Nouvelle, 43(1673) :313-14. 208 BIBLIOGRAPHY ASHFORD, C. 1881. Notes from the Isle of Wight. Jour. Conch., 3:132-35. ATKINS, W. R. G., AND LEBOTJR, M. V. 1923a. The H-ion concentration of the soil and of natural waters in relation to the distribution of snails. Roy. Dublin Soc. Sci., Proc., N.S., 17(2531):233-40. 1923&. Soil reaction, water snails and liver flukes. Nature, 111:83. AUSTIN, G. D. 1924. The Indian glow-worm (Lamprophorus tenebrosus Wlk.) Yearbook, Dept. Agric. Ceylon, Peradeniya, 1924:68-69. BAER, J. G. 1952. Le male (Tun phoride commensal des achatines de la foret tropical africaine. Revue Suisse Zool., 59 :239. BAIRD, W. 1850. Long-suspended vitality of a snail. Ann. & Mag. Nat. Hist. 2d ser., 6:68-69. BAKER, E. W., AND WHAKTON, G. W. 1952. An introduction to acarology. New York, Macmillan Co., 465 pp. BAKER, F. 1934. A conchological Rip Van Winkle. Nautilus, 48(1) :5-6. BAKEK, R. H. 1946. Some effects of the war on the wildlife of Micronesia. North Amer. Wadlife Conference, Trans., 1946:205-12. BALACHOWSKY, A. S. 1937. La destruction des limaces et autres gasteropodes phytophages, par le metaldehyde. Chimie & Industrie, 38(4): 213-15. BALLAND, A. 1898. Sur la composition des poissons, des crustaces et des mollusques. C. R. Acad. Sc., Paris, 126:1728-31. BALOCK, J. W. 1951. Ethylene dibromide for destroying fruit fly infestations in fruits and vegetables. Science, 114(2953): 122. BANKS, N. 1915. The ocarina or mites: A review of the group for the use of economic entomologists, USDA Report No. 108, 153 pp. BARNES, H. F. 1949a. The control of slugs on allotments and gardens. The Northern Allotment & Garden, May-July, 1949. 19496. The slugs in our gardens. New Biology (Penguin Books), 6:29-49. BARNES, H. F., AND WEIL, J. W. 1942. Baiting slugs using metaldehyde mixed with various substances. Ann. Appl. Biol., 29(1): 56-68. BARRY, D. H. 1959. Observations on the giant African snail. West African Pharmacist, 1959:92-93. BASINGER, A. J. 1923a. Eradication of the white snail, Helix pisana at La Jolla, California. Calif. Dept. Agric. Bull., 12 (1/2): 7-11. 19236. A valuable snail poison. Jour. Econ. Ent., 16:456-58. 1923c. The white snail (Helix pisana) at La Jolla, California. Ibid., p. 522-26. BIBLIOGRAPHY 209 1927. The eradication campaign against the white snail (Helix pisana) at La Jolla, California. Calif. Dept. Agric. Bull., 16(2):51-77. 1931. The European brown snail in California. Univ. Calif. Agric. Expt. Sta., Berkeley, Bull. No. 515, pp. 1-22. 1935. Measuring the efficiency of materials used for snail control. Jour. Econ. Ent., 28:903-5. 1940. Brown snail infestations. Calif. Citrograph, 25(7): 226. BAUMHOVER, A. EL, HUSMAN, C. N., SKIPPER, C. C., AND NEW, W. D. Field observations on the effects of releasing sterile screw-worms in Florida. Jour. Econ. Ent., 52(6): 1202-6. BAYFOBD, E. G. 1906. DrilusflavescensRossi, 9 , and its larva. Ent. Monthly Mag., 42:26768. 1959. BEEKLEK, J. G. 1944. U.S. Patent 2,363,852, Nov. 28, 1944. BEELEY, F. 1935. SnaUs. Rubber Res. Inst., Malaya, Ann. Kept., 1934:109. 1938a. The giant snail Achatina fulica (Fer.): Suggestions for control. Jour. Rubber Res. Inst. Malaya, 8(2): 130-39. Reprinted: Trop. Agriculturist, Ceylon, 90:247-54. 1938&. Snails and slugs (Mollusca). Rubber Res. Inst., Malaya, Ann. Rept., 1937:152. BEHUBA, B. K. Depredations of the giant African land snail, Achatina fulica (Ferussac) in Balasore (Orissa). Jour. Bombay Nat. Hist. Soc., 54(2): 287. BELLEVOYE [sic] 1870. Ann. Soc. Ent. France, 10(4) rxxxv-xxxvi. 1955. BENSON, W. H. Note sur la transportation et la naturalisation au BengaJe de FAchatina fulica de Lamarck. Jour, de Conchy!., 7:26&-68. BENTHEM JUTTING, W. S. S. VAN 1934. Achatina fulica (Fer.) in the Netherlands East Indies. Jour. Conch., 20(2): 43-44. 1948. On the present state of the malacological research in the Malay Archipelago. Chronica Naturae, 104(5): 129-38. 1952a. A snail farm in the Netherlands. Basteria, 16(1/2), 25-30. 1952&. Systematic studies on the non-marine mollusca of the Indo-Australian Archipelago. III. Critical revision of the Javanese pulmonate landsnails of the families Ellobiidae to Limacidae, with an appendix on Helicarionidae. Treubia, 21(2): 291-435. BEQUAERT, J, C. 1919. Parasites of Achatina. In: H. A. Pilsbry, A review of the land mollusks of the Belgian Congo. Amer. Mus. Nat. Hist., Bull., 40:61-63. 1925. The arthropod enemies of mollusks with description of a new dipterous parasite from Brazil. Jour. Parsit., 11:201-12. 1926. A dipterous parasite of a snail from Brazil, with an account of the arthropod enemies of mollusks. In: Medical report of the Hamilton Rice seventh expedition to the Amazon. Harvard Inst. for Trop. Biol. and Med., Contr., 4:292-303. 1950a. Enemies of Achatina. Proc. Hawaiian Ent. Soc., 14(1) :5-6. 1950&. Studies in the Achatininae, a group of African land snails. Bull. Mus. Comp. Zool., Harvard, 105(1): 1-216. 1858. 210 BIBLIOGRAPHY BEQUAERT, J. C., AND CLENCH, W. J. 1934. Studies of African land and fresh-water mollusks. 3. West African Achatinae related to Achatina baleata Reeve. Amer. Mus. Novitates, 705:1-16. 1937. The supposed introduction of an African Archachatina into the West Indies. Nautilus, 51:33-34. BEREZINA, V. M. 1948. ("Development of bait method for the control of slugs"; in Russian.) Leningrad. Inst. Zashch. East. Shorn. Trud., 1:176-77. BERG, C. O. 1953. Sciomyzid larvae (Diptera) that feed on snails. Jour. Parasit., 39:63036. BERG, C. O., KARLIN, E. J., AND MACKIEWICZ, J. C. 1955. Fly larvae that kill snails. Amer. Malacol. Union, Ann. Rept., 1955: 9-10. BERTRAND, H. W. R. 1928. Control of the Kalutara snail. Trop. Agriculturist, Ceylon, 71(3) : 15152. 1941. Prevention of damage to young rubber by snails and slugs. Ibid., 97(6): 327. BESS, H. A. 1956. Ecological notes on Lamprophorus tenebrosus (Walker) (Coleoptera: Lampyridae), an enemy of the giant African snail. Proc. Hawaiian Ent. Soc., 16(1): 24-29. BESS, H. A., AND ALICATA, J. E. 1953. Japanese lampyrids introduced. Proc. Hawaiian Ent. Soc., 15(1) :12. BHATIA, M. L., AND KEILIN, D. 1937. On a new case of parasitism of snail (Vertigo genesii Gredl.) by a dipterous larva. Parasitology, 29(3): 399-408. BlEDERMANN, W., AND MoKITZ, P. 1899. Beitrage zur vergleichenden Physiologic der Verdauung. III. Ueber die Function der sogenannten "Leber" der Mollusken. Pflligers Archiv. f. d. gesamte Physiol. d. Menschen u. d. Tiere, 75 :l-86. BINGLEY, W. 1829. Animal biography, or, popular zoology; illustrated by authentic anecdotes of the economy, habits of life, instincts and sagacity, of the animal creation (7th ed.). London, Longmans, Green & Co., 4:333. BIKKINSHAW, F. 1928. The giant snail (Achatina fulica Fer.). Malayan Agric. Jour., 16(4): 154. BLAISDELL, K. M. 1950. Arion circumscriptus, a new intermediate host of the cat hmgworm, Aelurostrongylus abstrasus. Cornell Vet., 40(2): 197-200. BLATJVELT, W. E. 1952. Pest control 1951--Outlook for 1952. N.Y. State Flower Growers' Bull., 77:9-11. BLOWERS, J. W. 1954. A new method of controlling slugs and snails. Orchid Review (London), 62 (730): 56. BODENHEIMER, F. S., AND ScHIFFER, M. 1952. Mathematical studies in animal populations. I. A mathematical study of insect parasitism. Acta Biotheoretica, 10(1/2): 23-56. BIBLIOGRAPHY 211 BOETTGER, C. R. 1937. Eine Landschnecke des tropischen Afrikas in Berlin. Sitzungsber. Gesell. Natur. Freunde Berlin, 1937(10/12):314-20. 1938. Schnecken in Sendungen von Kamerunbananen. Ibid., 1938(1/3): 8384. 1947. Einschleppung von Schnecken mil Bananensendungen nach Deutschlaiid. Arch, f. Mollusk., 76(1/3): 65-67. 1951. Eine afrikanische Schnecke als ein sich ausbreitender Schadling in Pflanzungen warmer Lander. Abhandl. Braunschweigischen Wissenschaft. Gesell., 3:10-18. BOISSEAU, G., AND LANORVILLE, G. L'escargot; elevage et parcage lucratifs; preparation culinaire et vente (2d ed.; 1931). Paris, Librairie Hachette et Cie. 92 pp. BOOT'S PUKE DRUG Co., LTD., POWELL, A. D., TAIT, J. C. E., AND LORD WOLMER. 1939. British Patent 508,125, June 27, 1939. Bosc, L. A. G. 1803. Agatine, Achatina. In: C. Deterville, Nouveau Dictionnaire d'Histoire Naturelle Appliquee aux Arts ... par une Societe de Naturalistes et d'Agriculteurs. Paris. 1:134. Bo WELL, E. W. 1917. Larva of a dipterous fly feeding on Helicella itala. Proc. Malacol. Soc. London, 12:308. BOYCOTT, A. E. 1920. On the size variation of Clausilia bidentata and Ena obscura within a "locality." Proc. Malacol. Soc. London, 14:34-42. 1921. Further observations on the occurrence of manganese in land and freshwater mollusca. Naturalist (London), 771:135-38, 209-11. BOYCOTT, A. E., AND OLDHAM, C. 1938. A contagious disease of Helix aspersa. Proc. Malacol. Soc. London, 23(2): 92-96. BRAND, T. F. VON 1931. Der Jahreszyklus irn Stoffbestand der Weinbergschnecke (Helix pomatia). Zeit. Vergl. Physiol., 14:200-264. 1932. Die chemische Zusammensetzung von Helix pomatia wahrend verschiedener Jahreszeiten. Archivio Zool. Italiano Torino, 16:871-78. BRERO, J. VAN 1933. Slakkenplaag. De Orchidee (Bandoeng), 2(7): 154-57. BROWN, A. C. 1951. Plant inspection and quarantine and their relationship to the horticultural interests of Florida. Citrus Indus., 32(3):5-9, 18. BRUBAKER, S. 1950. Giant African snails. Foreign Agric., 14(2):39-40. BRUES, C. T. 1907. Some new exotic phoridae. Ann. Historico-Naturales Musei Naturalis Hungarici, Budapest, 5:400-413. 1942. A species of Phoridae bred in Hawaii from the immigrant African land snail (Achatina fulica). Proc. Hawaiian Ent. Soc., 11(2): 154-56. BRYAN, E. H., JK. 1949. Economic insects of Micronesia. Report of the insect control committee for Micronesia 1947-48. Washington. D.C., Nat. Res. Council. 29 pp. BUCKHTJRST, A. S. 1947. Slug control with D.D.T. Fruit Grower (London), 103:146, 149. 1911. 212 BIBLIOGRAPHY BUKCH, J. B. 1960. Some snails and slugs of quarantine significance in the United States. USDA, Agric. Res. Serv., PI. Quar. Div., ARS 82-1. 73 pp. BURCH, J. Q. 1951. Ship fumigated. Conch. Club So. Calif., Minutes, 107:5. 1952a. Fighting the African snails. Ibid., 120:3. 19526. Snail bills in Congress. Ibid., 117:4. BUKTON, M. Slugs and snails, and other tales. Illus. London News, 209:564, Nov. 16, 1946. 1949. Giant land snails (of the family Achatinidae). Ibid., 214:120, Jan. 22, 1949. BUTOT, L. J. M. 1952a. Ons eerste gezamelijk onderzoek. (Also in Indonesian. With English summary.) Tropische Natuur, 32(2/8): 56-58. 19526. Mieren als slakkenverzamelaars. (With English summary.) Ibid., p. 78. CADART, J. 1955. Les escargots (Helix pomatia L. et Helix aspersa M.), biologic, elevage, parcage, histoire, gastronomic, commerce. Savoir en Histoire Naturalle, Vol. 24. Paris, Paul Lechevalier. 420 pp. CALKINS, W. W. 1877. Tenacity of life shown by some marine mollusks. Amer. Nat. 11:687. CALLAN, E. McC. 1946. 1941. The control of slugs by meta bait in Trinidad, B.W.I. Trop. Agric. (Trinidad), 18:211-13. CAMERON, A. E. 1939. The control of potato slugs. Jour. Ministry Agric. Great Britain, 46(5) : 454-62. 1951. White fly spraying trials to reduce insect plague in Mildura district: Control of snails. Citrus News (Australia), 27(9): 138. CAMPBELL, W. S. 1897. How the common garden snail (Helix aspersa) is spread about. Agric. Gaz. N. S. Wales, 8(2): 115. CANSDALE, G. S. 1940. Giant forest snails of the Gold Coast. The Field (London), 175(4554) : 546. CARMICHAEL, E. B. 1928. Action of ultraviolet rays on Limax flavus Linn. Amer. Jour. PhysioL, 85:358. 1931. The action of ultra-violet radiation on Limax flavus Linnaeus. I. The varying effects on non-pigmented and pigmented eggs. PhysioL Zool., 4:575-80. CAZIOT, E. 1926. Dispersion geographique de PHelix pomatia. Bull. Soc. Etude Sci. Nat. Musee d'Histoire Nat. Elbeuf, 44 .-132-34. 1928. Mode de dispersion des mollusques terrestres et fluviatiles. Riviera Scientifique (Nice), 15(3):45-51. CHAMBERLIN, J. L. 1952a. Final report on an ecology and population study of the giant African snail on Tinian, Marianas Islands. Invert. Consultants Comm. for Micronesia, Pac. Sci. Board, Nat. Res. Council. 27 pp. (Mimeographed.) 19526. Wind orientation in Achatina. Amer. Malacol. Union, Ann. Kept., 1952:12-14. BIBLIOGRAPHY 213 CHARMOY, D. d'E., AND GEBERT, S. 1921. Insect pests of various minor crops and fruit trees in Mauritius. Bull. Ent. Res., London, 12(2): 181-90. (Reprinted: 1922. Mauritius, Dept. Agric. Scientific Series, Bull. No. 8, p. 14.) CHIPMAN, R. N., AND SEIBERT, F. J. 1939. Canada patent 381,540, May 23, 1939. CHITWOOD, B. G., AND CHITWOOD, M. B. 1934. Daubaylia potomaca n. sp., a nematode parasite of snails, with a note on other nemas associated with molluscs. Helminthol. Soc. Wash., Proc., 1(1):8, 9. CHOCK, Q. C., AND NAKAO, H. S. 1951. Experiments conducted with the giant African snail, Achatina fulica (Fer.) to determine the lethal minimum exposure at low temperatures. Hawaii, Board of Commissioners of Agric. and Forestry, Div. Entom. and Marketing. 4 pp. (Dittoed.) CHONG, M. 1954. Tefflus hacquardi de Chaudoir. Proc. Hawaiian Ent. Soc., 15(2): 270. CLAPP, G. H. 1902. "Slugs" as medicine. Nautilus, 16:84. CLARK, A. H. 1949. The invasion of New Zealand by people, plants and animals: The South Island. New Jersey, Rutgers Univ. Press, 465 pp. CLARK, F. W., AND WHEELER, W. C. 1917. The inorganic constituents of marine invertebrates. U.S. Geol. Surv., Prof. Papers, 102:1-56. 1922. Ibid., 124:1-62. CLARKE, W. J. 1924. Dispersal of molluscs by commerce. Naturalist (London), 1924:363. CLAUSEN, C. P. 1940. Entomophagous insects. New York, McGraw-Hill Book Co., 688 pp. CLAYTON, H. H. 1927. World weather records. Smithsonian Misc. Coll., 79:vii+1199. CLENCH, W. J. 1925. Snails eaten by shrews. Nautilus, 39:28. 1949. Some kodachromes of live Achatina fulica. Amer. Malacol. Union, News Bull. & Ann. Rept., 1949:3. COGHILL, H. E. 1909. Mice devouring snails. Irish Nat. (Dublin), 18:27. COLLINGE, W. E. 1921. The starling: Is it injurious to agriculture? Jour. Ministry Agric. Great Britain, 27(12): 1114-21. COMBER, E. 1906. A list of the marine Mollusca in the Bombay Natural History Society's collection. Jour. Bombay Nat. Hist. Soc., 17(1): 207-18. CONNOLLY, M. 1912. A revised reference list of South African non-marine Mollusca; with descriptions of new species in the South African Museum. Ann. South African Mus., 11(3): 193. 1931. The distribution of the non-marine Mollusca throughout continental Africa. Jour. Conch., 19(4):98-107. 1939. A monographic survey of South African non-marine Mollusca. Ann. South African Mus., 33:1-660. 214 BIBLIOGRAPHY COOK, A. S. 1946. Enemies of snails. Jour. Conch., 22(8) : 186. COOK, O. F. 1897. A new wingless fly from Liberia. Science, N.S., 6:886. COOLEY, C. E. 1950. Plant quarantine activities in the Pacific, with particular reference to the oriental fruit fly. Calif. Dept. Agric., Bull., 39(1): 10-16. COOLEY, H. D., AND HUTCHINSON, K. T. 1950. Prohibit entry of giant African snails. House of Representatives Report No. 2363, 81st Cong., 2d sess. COOLEY, H. D., AND McCoRMicx, C. J. 1951. Prohibiting entry of mollusks. House of Representatives Report No. 800, 82d Cong., 1st sess. COOLIDGE, H. J. 1947. First annual report Pacific Science Board. Washington, D.C., Nat. Res. Council. 50 pp. 1948. Second annual report. Ibid., 73 pp. 1949. Third annual report. Ibid., 154 pp. 1950. Fourth annual report. Ibid., 114 pp. 1951. Fifth annual report. Ibid., 96 pp. 1952. Sixth annual report. Ibid., 87 pp. 1953. Seventh annual report. Ibid., 93 pp. 1954. Eighth annual report. Ibid., 61 pp. 1955. Ninth annual report. Ibid., 57 pp. COOMANS DE RuiTER, L. 1960. CORBETT, 1929. 1933. 1937. 1939. De agaatslak in Nieuw-Guinea. Nederlands Nieuw-Guinea, 8(1): 18-21. G. H. Achatina fulica. Malayan Agric. Jour., 17(8) :275. The giant snail (Achatina fulica Fer.) in Malaya. Ibid., 21(2): 77-79. Achatina fulica Fer. Malaya Dept. Agric. Gen. Series, No. 26, p. 40. The giant snail (Achatina fulica Fer.). Malayan Agric. Jour., 26(9): 376-79. CORBETT, G. H., AND PAGDEN, H. T. 1941. A review of some recent entomological investigations and observations. Malayan Agric. Jour. 39(9):347-75. COTTON, B. C. 1940. The Kalutara snail, Achatina fulica Ferussac, attacks rubber trees in Ceylon. South Australian Naturalist, 20(4) :iii. 1949. These snails may invade your garden. Jour. Dept. Agric. South Australia, 52:561-65. COUKTOIS, A., AND DlJVAL, M. 1927. Sur la mort des escargots immerges dans Peau. C. R. Soc. Biol., Paris, 97:1695-97. CRAGG, J. B., AND VINCENT, M. H. 1952. The action of metaldehyde on the slug Agriolimax reticulatus (Miiller). Ann. Appl. Biol., 39(3):392-406. CRAWSHAY, L. R. 1903. On the life-history of Drilus flavescens Rossi. Trans. Ent. Soc. London, 1903:39-51. CREIGHTON, C. 1899. Microscopic researches on the formative property of glycogen. London, A. & C. Black. 2 parts, 152 pp., 127 pp. BIBLIOGRAPHY 215 CROS, A. 1926. Moeurs et evolution du Drilus mauritanicus Lucas. Bull. Soc. Hist. Nat. Afrique du Nord, 17(7): 181-206. 1930. Malacogaster passerinii Bassi: Mceurs, evolution. Ibid., 21:133-60. CKOSSE, H. 1894. Faune malacologique terrestre et fluviatile de la Nouvelle-Caledonie et de ses dependances. Jour, de Conchy!., 42:161-473. CRUZ, G. B. 1932. Studies on the methods of feeding ducks. Philippine Agric., 20(8) :53548. DALGLIESH, G. 1907. Carnivorous propensities of a slug. Zoologist (London, 4th Ser. 11 (127): 277. DAMMERMAN, K. W. 1929. The agricultural zoology of the Malay Archipelago. Amsterdam, J. H. de Bussy, Ltd. 437 pp. DABTEVELLE, E. 1952a. Achatina fulica hamillei Petit du Congo beige. Basteria, 16(4) :56. 19526. Helix (Cryptomphalus) aspersaO.F.Miiller--Conditions de vie etranges. Ibid., 16(1/2) :4-5. 1953. Les "N'zimbu," monnaie du Royaume du Congo. Bull, et Mem. Soc. Roy. Beige Anthrop. et Prehist., Memoires, N.S., 64(1) :xv+251. 1954. [Report on giant African snail prepared at request of Hawaiian Board of Agriculture and Forestry. Trans, by Philip Weber.] 14 pp. (Mimeographed.) DAVIS, C. J. 1954. Report on the Davis expedition to Agiguan, July-August, 1954. Ecological studies, island of Agiguan, Marianas Islands as related to the African snail, Achatina fulica Bowdich, and its introduced predator, Gonaxis kibweziensis (E. A. Smith). Invert. Consultants Comm. for the Pacific, Pac. Sci. Board, Nat. Res. Council. 24 pp. (Mimeographed). 1958a. Achatina fulica Bowdich. Proc. Hawaiian Ent. Soc., 16(3) :340. 1958&. Recent introductions for biological control in Hawaii, III. Ibid., pp. 356-58. 1959. Ibid., IV. 17(1): 62-66. DEAN, R. E. 1948. Annual report of the superintendent of gardens for the year 1947/48. Appendix N. Hong Kong. 5 pp. 1950. Annual report by the superintendent of gardens for the period ended the 31st March 1950. Hong Kong, 10 pp. DESMAREST, A. G. 1824. Memoire sur une espece d'insectes des environs de Paris, dont le male et la femelle ont servi de types a deux genres differens. Ann. Sci. Nat., 2:257-70. (Also in Bull. Soc. Philomath., Paris., 1824:57-62.) DESMAKEST, E. 1889. [Larvae of Drilus flavescens in Helix nemoralis.] Ann. Soc. Ent. France 9(6):cxv. DETHIER, V. G. 1947. Chemical insect attractants and repellents. Philadelphia, Blakiston Div., McGraw-Hill Book Co. 289 pp. DEUBEL, F. 1913. Die Entwicklung des Drilus concolor Ahr. Verfi. Mitt. Siebenburg. Ver. Naturw., 43:1-8. 216 BIBLIOGRAPHY DEXHEIMER, L. Beitrage zum Kalkstoffwechsel der Weinbergschnecke (Helix pomatia). Zool. Jahrb. Abt. Allg. Zool. u. Physiol. Tiere, 63(1): 129-52. DIAS, S. J. F., AND THAMOTHERAM, T. V. 1939. The control of the giant snail (Achatina fulica Fer.) at the Tabbowa vegetable seed station by the use of "meta-bran" bait traps. Trop. Agriculturist, Ceylon, 92 (4): 222-23. DICKINSON, W. H, 1946. Giant West African land snails in captivity. The Aquarium, 15(8) : 15152. DJAENOEDIN, R. 1942. Pertjobaan dengan sipoet Achatina sebagai Makanan untuk binatang. Balai Pemeriksaan Hewan di Bogor. (Unpub. report.) DORING, A. 1872. Bemerkungen liber die Bedeutung und TJntersuchungen iiber die chemische Zusammensetzung der Pulmonatenschale. (Inaugural Dissertation, Gottingen.) 41 pp. DOLLFUS, G. 1899. In: H. Douville, Sur une coupe de Madagascar, Bull. Soc. Geol. France, 27 (2): 395. DOSCAS, A. E. C. 1929. Giant snail (Achatina fulica). Malayan Agric. Jour., 17(8):287. DOUCETTE, C. F. 1954. An introduced slug, Arion ater (L.), and its control with metaldehyde. Jour. Econ. Ent., 47(2): 370. DOUENCE, A. 1929. Les appats empoisonnes: Leur utilisation dans la lutte centre les ennemies des cultures. Annales des Epiphyties, 15(1/2): 1-96. DRZ, A. 1913. Bacteries symbiotes chez le Cyclostome. Rev. Sci. Paris, 1913:436. DUFO, H. 1840. Observations sur les mollusques marins, terrestres et fluviatiles des lies Sechelles et des Amirantes. Ann. Sci. Nat., 2d ser. Zool., 14:45-80, 116-221. 1951. DUPONT, M. In: A. Morelet, Addition a la faune paleontologique de Tile Maurice. Jour, de Conchyl., 26:171. DUPONT, R. 1935. La faune des Sechelles. 1. Liste des mollusques terrestres et fluviatiles. Soc. Roy. des Arts et des Sci. de Maurice, Trans., Ser. C, 1933/34(3): 138. DTJKHAM, H. E. 1920. Of slugs. Gardeners' Chronicle, 68:46, 85-86, 110. DUSTAN, A. G. 1927. Notes on some barriers used against garden slugs in 1926. Quebec Soc. Protect. Plants, Ann. Kept., 19:22-23. 1878. DUVAL, M. 1930. Concentration moleculaire du sang de Tescargot; ses facteurs, ses variations; influence de Tetat d'activite de Tanimal. Ann. de Physiol. et de Physicochimie Biol, 6(3): 346-64. Division of Entomology and Marketing. Hawaii Board of Commissioners of Agric. and Forestry, Biennial Kept., 1953/54:27. DWIGHT, J. L, 1955. BIBLIOGRAPHY 217 C., AND PALMER, C. Homing behavior in gastropods. Oikos, 2(2): 259-70. A. J., AND BRANNAN, C. F. Preventing the entry of certain mollusks into the United States. Senate Report No. 628, 82d Cong., 1st sess. ELLIOTT, W. T. 1918. Note on the malacophagous propensities of Helix nemoralis Linn. Proc. Malacol. Soc. London, 13:20. ELTON, C. S. 1958. The ecology of invasions by animals and plants. New York, John Wiley & Sons, Inc. 181 pp. ERHARD, H., AND ZIEGLWALLNER, F. 1912. Ueber das Auftreten von Glykogen nach Fattening mit einigen Monosacchariden, einem Disaccharid und Fett, nach morphologischen Beobachtungen an der Weinbergsehnecke. Zeitsch. Biol., 58 :554-57. EDELSTAM, 1950. ELLENDER, 1951. ESAKI, T. ("Giant snail of Ponape Island, in relation with lantana"; in Japanese.) Sangyo no Nanyo ("The Industrial South Seas"), 4(4). ESAKI, T., AND TAKAHASHI, K. 1942. Introduction of the African snail, Achatina fulica Femssac into Japan, esp. Micronesia, and subsequent developments. Trans, from Japanese by Toyohi Okada. Kagaku Nanyo ("Science of the South Seas," Jour. Palau Trop. Biol. Sta.), 4(3): 16-25. ESSLEMONT, J. M. 1941. 1938. Observations on the control of potato slugs. Scottish Jour. Agric., 21 (2): 147-52. EWART, A. J. 1910. Slugs and snails. Jour. Dept. Agric., Victoria (Australia), 8(12):807-8. EYERDAM, W. J. 1952a. Solomon Island megapodes recommended as destroyers of African snail pests. Amer. Malacol. Union, Ann. Kept., 1952:29-30. 19526. Ibid., Conch. Club So. Calif., Minutes, 121:4-5. FAIRWEATHER, J. 1937. Pests and diseases. Malaya Dept. Agric. Gen. Series, No. 27, p. 143. FANTHAM, H. B. 1921. Some parasitic protozoa found in South Africa. IV. So. Africa. Jour. Sci. (Johannesburg), 18:169. 1923. Ibid., VI. 20:495. 1924. Ibid., VII. 21:442-44. 1925. Ibid., VHI. 22:346-49. FANTHAM, H. B., AND ROBEKTSON, K. G. 1927. Ibid., X. 24:441-49. FEIJ, IR. P. J. 1940. Enkele waarnemingen betreffende de agaatslak. De Bergcultures, 4(35):1112-14. FENNEB, F., AND DAY, M. F. 1953. Biological control of the rabbit in Australia. Science News (Penguin Books), 28:7-22. FERGUSON, F. J. 1948. Beetles not bullets. Sunday Mirror Magazine, Sept. 5, 1948, p. 17. 218 BIBLIOGRAPHY FERNANDO, H. E. 1952. The Kalutara snail or the African snail and its control. (Also in Singhalese.) Ceylon Dept. Agric. Propaganda Handbill. 4 pp. FERNANDO, W. 1946. Studies on the pulmonates of Ceylon. I. The storage of galactogen and glycogen in the gonad and albumen gland of Achatina fulica Femssac. Ceylon Jour. Sci. B. Zool. 23(2):97-104. FERUSSAC, A. E. 1821. Tableaux systematiques des animaux mollusques classes en families naturelles, ... les mollusques terrestres ou fluviatiles, vivants ou fossiles Paris, Arthus Bertrand, 110 pp. FlGUEIREDO, E. R. DE, JR., AND LEPAGE, H. S. 1953. Caraeois daninhos as orquideas. Biologico, 19(2): 31-33. FINCHER, F. 1947. Lampyris noctiluca L. (Col. Lampyridae) attacking slug. Ent. Monthly Mag., 83 :149. FISCHER, P.-H. 1939. Escargots europeens acclimates en Australie. Jour, de Conchyl., 83(3) : 245-53. 1941. Teneur en eau des tissus chez quelques Gasteropodes aquatiques. Bull. Mus. National d'Hist. Nat., Paris, 2e ser., 13(2): 112-18. FITZGERALD, L. D. E. F. V. 1947. Giant snail. Rubber Res. Inst., Malaya, Ann. Rept., 1940:111. FLOWER, S. S. 1922. Longevity of molluscs. Cairo Scientific Jour., 10(104/105): 115-17. FONSECA, J. P. DA, AND AlJTUORI, M. 1932. Principaes pragas do cafe no estado de Sao Paulo. Inst. Biol. Defesa Agric. e Animal, Div. Vegetal, Sao Paulo, 87 pp. FORD, J. 1887. Some remarks on the migration of mollusks. Conchologist's Exchange, 2 (5): 71-72. FOSBERG, F. R. 1957. Conservation situation in Oceania. Ninth Pac. Sci. Cong., Thailand. Report of the Standing Comm. on Pac. Conservation. (Processed.) FOSTER, T. D. 1937. Productivity of a land snail, Polygyra thyroides (Say). Ecol., 18:54546. FRANSSEN, C. J. H. 1936. Nieuwe perspectieven bij de bestrijding van de agaatslak. (Unpub. report.) FREEMAN, O. W. 1951. Geography of the Pacific. New York, John Wiley & Sons, Inc. 573 pp. FBENCH, C. 1906. Slugs. Jour. Dept. Agric., Victoria (Australia), 4(7): 445-47. FRIGILLANA, G. R. 1923. A study of the effects of snails as a supplement to a ration for laying hens. Philippine Agriculturist, 12 (6): 239-45. FROMMING, E. 1929. Wie lange kann Agriolimax agrestis im Wasser leben. Arch. Mollusk., 61 (4/5): 196-200. 1949. Untersuchugen iiber aufgenommene Nahrungsmengen und DDT-wirkung bei zwei Nacktschneckenarten. Zeitsch. f. Hygienische Zool. u. Schadlings., 37:320-25. BIBLIOGRAPHY 219 Untersuchungen liber die Wirksamkeit des DDT-praparates Duolit auf Landlungenschnecken. Anz. f. Schadlingskunde, 23:57-60. 1951. Allgemeine Betrachtungen liber die Priifung molluscizider Mittel. Ibid., 24:26-28. 1952a. Neuere Erkenntnisse liber unsere Nacktschnecken als Schadlinge. Angewandte Botanik, 26(2): 94-99. ,, 19526. Unsere Wegschnecke (Arion) als Ubertrager von Krankheiten fiir Mensch und Pflanze. Zeitsch. f. Hygienische Zool. u. Schadlings., 40(11/12): 376-79. 1954a. Der Alterstod bein einer Wegschnecke. (Arion empiricorum Fer.). Naturwissensch. Rundschau, 1954 (8): 340-41. 1954&. Biologic der mitteleuropaisehen Landgastropoden. Berlin-Mlinchen, Duncker & Humblot, 404 pp. 1955. Die Rolle unserer Schnecken bei der Ausbreitung von infektiosen Pflanzenkrankheiten. I. Faule-krankheiten bei Erntegut. Zeitsch. f. Angewandte Zool., 1955 (4): 433-39. FROMMING, E., AND PLATE, H.-P. 1952. Sind die Metaldehyd-koder zuverlassige Schneckenbekampfungsmittel? Zeitsch. f. Hygienische Zool. u. Schadlings., 44:130-31. FROMMING, E., AND BJEMSCHNEIDER, R. 1952. Uber die Wirkung von Insektiziden auf Gastropoden. Mitt. II. Anz. f. Schadlingskunde, 25:186-89. FRONDA, F. M. 1919. A study of the effects of animal and plant proteins in rations for laying hens. Philippine Agriculturist, 7(8): 235-52. FULLAWAY, D. T. 1950. 1939. Report of the territorial entomologist and chief plant inspector. Hawaii Board of Commissioners of Agric. and Forestry, Biennial Kept., 1937/ 38:24. 1941. Ibid., 1939/40:21. 1943. Ibid., 1941/42:29-30. 1945. Ibid., 1943/44:9. 1947. Ibid., 1945/46:38, 40, 46. 1949. Ibid., 1947/48:69-70. 1952. Biological control of insect pests in the Hawaiian Islands since 1925. Hawaii Board of Commissioners of Agric. and Forestry, Kept., 1952: 98-105. GAIN, W. A. 1896. Enemies of our land and fresh-water molluscs. Naturalist's Jour., 1896: 78-79. GALLI-VALERIO, B. 1918. Parasitologische Untersuchungen und Beitrage zur parasitologischen Technik. Zentralblatt f. Bakteriol., Parasit. u. Infektionskr., Originate, 80:264-71. GAMMON, E. T. 1943. Helicid snails in California. Calif. Dept. Agric., Bull, 32(3): 173-87. GARDINER, J. G. 1949. Report of the Minister of Agriculture for the Dominion of Canada for the year ended March 31, 1949. Can. Mm. Agric., Rept., 1948/49:105. GARNADI, P. 1951. Die grosze Aehatschnecke (Achatina fulica Fer.). Hemera Zoa, 58:279301. 220 BIBLIOGRAPHY GASKOIN, J. S. 1850. On the habits of Helix lactea. Proc. Zool. Soc. London, 18:243-44. GEBHARDT-DUNKEL, E. 1953. Die Trockenresistenz bei Gehauseschnecken. Zool. Jahrb., Abt. f. Allg. Zool. u. Physiol. der Tiere, 64(2): 235-66. GERMAIN, L. 1921. Faune malacologique terrestre et fluviatile des lies Mascareignes. Mission Zoologique de M. Paul Carie aux lies Mascareignes. Angers, Gaultier et Thebert. GHOSE, K. C. 1959. Observations on the mating and oviposition of two land pulmonates, Achatina fulica Bowdich and Macrochlamys indica Godwin-Austen. Jour. Bombay Nat. Hist. Soc., 56(2): 183-87. 1960. A monograph on the giant land snail Achatina (Lissachatina) fulica fulica Bowdich. (Unpub. thesis, City College, Calcutta). GIMINGHAM, C. T. 1940. Some recent contributions by English workers to the development of methods of insect control: Slugs. Ann. Appl. Biol., 27(2): 167-68. GIMINGHAM, C. T., AND NEWTON, H. C. F. 1937. A poison bait for slugs. Jour. Ministry Agric. Great Britain, 44(3): 24246. GLENDENNING, R. 1947. Slug control in British Columbia. Canada, Sci. Serv., Dept. Agric., Div. Ent., Publ. No. 70, 4 pp. (Processed.) GODWIN-AUSTEN, H. H. 1908. The dispersal of land shells by the agency of man. Proc. Malacol. Soc. London, 8(3): 146-47. GOOT, P. VAN DER 1939. Rapport over de bestrijding van de agaatslak in de gemeente Buitenzorg van 1 Juli t/m 31 December 1939. [Unpub. report.] GOKTON, G. R. 1919. A dangerous snail introduced in California. Calif. State Comm. of Hort., Monthly Bull., 8(1): 7-9. GOSECO, A. 1921. A study of the effects of animal and plant protein feeds on the egg production of ducks. Philippine Agriculturist, 9(8/9): 197-207. GOUREAU, C. 1843. Note sur un Diptere, dont la larve vit dans PHelix conspurcata (Melanophora helicivora Goureau). Ann. Soc. Ent. France, 2e ser., 1:77-80. GRAVATT, G. F., AND MARSHALL, R. P. 1917. Arthropods and gastropods as carriers of Cronartium ribicola in greenhouses. Phytopathology, 7(5):368-73. GREEN, E. E. 1909. 1910a. 19106. 1910c. Entomological notes. Trop. Agriculturist, 33:136, 431. African snail (Achatina fulica). Ibid., 35(4) :311. Introduction of an African snail. Ibid., 35(2): 120-21. Report on the outbreak of Achatina fulica. Circulars & Agric. Jour. Royal Bot. Gardens, Ceylon, 5(7):55-64. IdlQd. Sudden appearance of an African snail in Ceylon. Spolia Zeylanica, 7(25):56. 191 la. Remedy for slugs and snails. Trop. Agriculturist, 36:131-32. 19116. The wanderings of a gigantic African snail. Zoologist (London), 4th ser., 15(170): 41-45. BIBLIOGRAPHY 221 GRESSITT, J. L. 1952. Description of Kayangel Atoll, Palau Islands. Atoll Res. Bull., 14:1-6. 1954. Insects of Micronesia, Vol. 1. Hawaii, B. P. Bishop Museum, 257 pp. GUDE, G. K. 1914. The fauna of British India including Ceylon and Burma. Mollusca. II. Trochomorphidae--Janellidae. London, Taylor and Francis. 504 pp. HAGEN, B. 1952. Die bestimmenden TJmweltsbedingungen flir die Weichtierwelt eines sliddeutschen Fluszufer-Kiefernwaldes: Mollusca terrestria in Pinetoericae. Veroffentl. Zool. Staatssammlung (Munich), 2(2/3): 161-276. HALL, A. H. 1932. Eradication of slugs and snails. Nature, 130(3274): 170. HAMADA, N. 1935. ("The edible snail which is all the rage in Japan is offensive in the South Seas"; in Japanese.) Taiwan Nichi-nichi Shinpo ("Taiwan Daily News"), Nov. 5, 1935. HAMILTON, W. J., JR. 1930. The food of the Soricidae. Jour. Mammal., 11(1): 26-39. HANNA, G. D. 1939. Exotic mollusca in California. Calif. Dept. Agric., Bull., 28(5): 298-321. 1948. Indiscriminate scattering of exotic mollusks. Nautilus, 62(1): 29-30. HANSON, A. J. 1930. The slug as the intermediate host of the microscopic tapeworm of chickens. Western Wash. Exp. Sta. Bull., 18-W:11. HARDY, D. E. 1952. Notes on Phoridae. Proc. Hawaiian Ent. Soc., 14(3): 355-56. HARRISON, T. H. 1951. The giant snail. Health, N.S., 1(3): 16-18. HARTLEY, A. 1898. Helix nemoralis and H. virgata without food for fourteen months. Jour. Conch., 9(4): 105. HASS, C. P. J. DE 1937. Over een eieren leggend wijf je van Lamprophorus dorsalis. Tropische Natuur, 26(1/2): 13-16. HATAI, S., AND KATO, G. 1943. Observation upon growth of shells and ecology of Achatina fulica Ferussac in Palau. Trans, from Japanese by Toyohi Okada. Kagaku Nanyo ("Science of the South Seas/5 Jour. Palau Trop. Biol. Sta.), 5(2): 1-19. HAVILAND, M. D., AND PITT, F. 1919. The selection of Helix nemoralis by song thrush (Turdus musicus). Ann. Mag. Nat. Hist., 9th ser., 3:525-31. HEATHER, W. A. 1955. The kamarere forests of New Britain. Empire Forest Rev., 34(3): 255-78. HEGNER, R., AND CHTJ, H. J. 1930. A survey of protozoa parasitic in plants and animals of the Philippine Islands. Philippine Jour. Sci., 43(3):451-82. HEIM, R. 1956. House that Jack built. Internal. Union for the Protection of Nature, Bull, 5(2/3):3. 222 BIBLIOGRAPHY HELFER, H. 1949. This business of snails. Nations Business, 37(8) :71. HELY, P. C. 1946. Control of snails on citrus trees. Agric. Gaz. N. S. Wales, 57(5): 239-40. HENDERSON, J. 1936. How some snails survive grass fires. Nautilus, 50(2): 71-72. HERKLOTS, G. A. C. 1948. Giant African snail Achatina fulica Fer. Food & Flowers, 1(1): 1-4. HEKRSTROM, G. 1953. Akersniglar skadar hostvete: nordvastra Skane. Statens Vaxtskyddsanstalt Vaxtskyddsnotiser (Sweden), 4:52-53. HES, J. W. 1949. Agaatslakken in suikerriet. (With English summary.) Chronica Naturae, 105 (9): 226-27. 1950. The African giant snail and sugar cane. Sugar Jour., 12(8): 18. 1951. Some remarks on the "hibernating" of Achatina fulica Fer. (Unpub. MS.) HEUBE, see HEUBEL HEUBEL, G. AD. 1937. Enkele gegevens over de agaatslak (Achatina fulica Fer.) in de Lampongsche Districten. De Bergcultures, 11(47): 1667-70. 1938. (As "Heube.") Ibid., II. 12(14):400-401. HILL, A. F. 1952. Economic botany (2d ed.) New York, McGraw-Hill Book Co., 592 pp. HILL, H. R. 1951. Exotic mollusca in California. Conch. Club So. Calif., Minutes, 109:1-4. HOAR, W. S., AND BARBERIE, M. 1945. Distribution of riboflavin in fresh and processed fish. Canadian Jour. Res., Sect. E., Med. Sci., 23(1):8-18. HODSON, W. E. H. 1924. Preliminary experiments in the control of slugs. Jour. Roy. Hort. Soc. London, 49(2): 191-96. 1925. Some further notes on the control of slugs. Ibid., 50:257-59. HOFFMAN, W. H. 1936. Ratten und Schnecken. Archiv f. Schiffs u. Tropenhygiene, 40:397. HOLMES, C. H. 1954. Seed germination and seedling studies of timber trees of Ceylon. Ceylon Forester, N.S., 1(3): 3-51. HOLTTUM, R. E., AND FlJRTADO, C. X. 1934. The effects of the introduction of exotic plants in the Malay peninsula. Proc. Fifth Pac. Sci. Cong., 1:797-99. HOOPER, D. 1910. Materia medica animalium Indica. Jour. & Proc. Asiatic Soc. of Bengal, 6:507-22. HORA, S. L. 1928. Hibernation and estivation in gastropod molluscs: On the habits of a slug from Dalhousie (Western Himalayas), with remarks on certain other species of gastropod molluscs. Records Indian Museum, Calcutta, 30:357-73. HORA, S. L., AND RAO, H. S. 1927. Hibernation and aestivation in gastropod molluscs. Records Indian Museum, Calcutta, 29 (2): 49-62. BIBLIOGRAPHY 223 HOWES, N. H., AND WELLS, G. P. 1934a. The water relations of snails and slugs. I. Weight rhythms in Helix pomatia L. Jour. Exp. Biol., 11:327-43. 1934&. The water relations in [sic] snails and slugs. II. Weight rhythms in Arion ater L. and Limax flavus L. Ibid., 11:344-51. HUDSON, J. P. 1937. Meta-bran for the destruction of snails. The Gardener's Chronicle, 3d ser., 102 (2659): 436. HUDSON, R. H. 1937. Danger to birds from the use of meta-bran. The Gardener's Chronicle, 102(2644): 153. HUTCHINSON, K. T. 1952. Mollusks. Notice of proposed regulations governing entry. Fed. Register, 17 (145): 6820-27, July 25, 1952. HUTSON, J. C. 1920. The African snail (Achatina fulica). Trop. Agriculturist, 55:217-25. HUTSON, J. C., AND AUSTIN, G. D. 1924. Notes on the habits and life history of the Indian glow-worm (an enemy of the African or Kalutara snail). Ceylon, Dept. Agric. Bull., 69:1-16. HYAMS, E. 1951. Snails and spengler. New Statesman and Nation, 41:127, Feb. 3, 1951. ICCM--INSECT CONTROL COMMITTEE FOR MICRONESIA 1947. First report of the Insect Control Committee for Micronesia, Pac. Sci. Board, Nat. Res. Council. Honolulu, Hawaii. Oct. 3, 1947. 6 pp. (Mimeographed.) 1948. Minutes, meeting of Insect Control Committee for Micronesia, held at the Experiment Station, H.S.P.A. April 16, 1948. 12 pp. (Typescript.) 1950. Insect Control Committee for Micronesia (A committee of the Pac. Sci. Board, Nat. Res. Council), fourth annual meeting March 9-10, 1950, Honolulu, 4 pp. (Mimeographed.) ICCP--INVERTEBRATE CONSULTANTS COMMITTEE FOR THE PACIFIC 1952. Sixth annual meeting of Invertebrate Consultants Committee for the Pacific, Pac. Sci. Board--Nat. Res. Council, February 13 and 14, 1952, Honolulu. 17 pp. (Mimeographed.) 1953. Invertebrate Consultants Committee for the Pacific--Pac. Sci. Board; Minutes of seventh annual meeting, February 18, 1953, Hawaiian Sugar Planters Association Experiment Station, Honolulu, 21 pp. (Mimeographed.) INGRAM, W. M. 1942a. Food habits of Haplotrema minimum Ancey and habits of associated mollusks on the Mills College Campus. Nautilus, 55(3): 98-102. 19426. Snail associates of Blarina brevicauda talpoides (Say). Jour. Mammal., 23 (3): 255-58. 1944. Snails hoarded by Blerina at Ithaca, New York. Nautilus, 57(4): 135-37. 1946. Mollusk food of the beetle Scaphinotus interruptus (Men.). Bull. So. Calif. Acad. Sci., 45(1):34-36. 1950. Feeding of the beetle Calosoma on snails. Nautilus, 63:142, 143. 1952. Thousands of living European snails sold as fish bait in state of Ohio. Ibid., 66(1): 26-29. INO, T. 1949. The effect of food on growth and coloration of the top shell (Turbo cornutus Solander). Jour. Marine Res., Sears Foundation for Marine Res. (Yale Univ.), 8(1): 1-5. 224 BIBLIOGRAPHY JACOBSON, E. 1936. Lamprophorus. Tropische Natuur, 25(6): 101. JANDA, V. 1951. Vyskyt manganu v tkanich nekterych 2ivocichu. ("Manganese in the tissues of certain animals"; with English summary.) Ceskoslovenska Spolecnost Zoologicka, Vestnik, Prague, 14:183-206. JARRETT, V. H. C. 1922. Notes on an acclimatisation experiment with Helix aspersa. Singapore Naturalist, 1(1): 74-75. 1923. The occurrence of the snail Achatina fulica in Malaya. Ibid., 1 (2): 73-76. 1931. The spread of the snail Achatina fulica to South China. Hong Kong Naturalist, 2(4): 262-64. 1932. Notes and comments. (Conchology. Achatina fulica.) Ibid., 3(1) :70. 1949. Achatina's odyssey. The Times (London), No. 51,482, p. 5. Sept. 9, 1949. JAKY, S. G. 1939. 1937. "Meta-fuel" for the control of slugs and snails. Scientific Hort., 7:18085. "Meta fuel" and slug control. Jour. Southeastern Agric. Coll., Wye, Kent., 40:183-86. JARY, S. G., AND AUSTIN, M. D. JASKI, C. J. 1953. Achatina fulica. Tropische Natuur, 33(3): 91-98. JEFFERSON, R. N. 1952. The control of slugs; varying degrees of success achieved with metaldehyde sprays and dusts. Florists' Review, 111(2869): 29-30, 123-27. JOHNSON, C. W. 1929. Diptera destroying snails. Psyche, 36(2): 106. JONES, A. L., JR. 1949. Giant African snail (Achatina fulica). A study, testing and development of its economic uses. Invert. Consultants Comm. for Micronesia, Pac. Sci. Board, Nat. Res. Council. 2 pp. (Typescript.) JONES, D. B., and MURPHY, J. C. 1926. The vitamin content of oysters. Proc. Soc. Exp. Biol. Med., 23:519-20. JOTJBERT, C. J., and WALTEKS, S. S. 1951. The control of snails and slugs. Farming in So. Africa, 26(308): 379-80. JUTTING, see BENTHEM JUTTING KABURAKI, T. 1940. Further notes on the effect of some exotic animals upon the fauna of Japan. Proc. Sixth Pac. Sci. Congr., 4:229-30. KAGIYA, D. 1936. ("Important points in the growth and feeding of edible snails"; in Japanese.) Nogyo Sekai ("Agricultural World"), Jan., 1936, pp. 165-68. KAIBO, K. 1935. ("The breeding of edible snails in France"; in Japanese.) Nogyo Sekai ("Agricultural World"), April, 1935, pp. 67-68. KALSHOVEN, L. G. E. 1950. Mollusca, weekdieren. In his: De plagen van de cultuurgewassen in Indonesie, 2 vols, 'S Gravenhage, Bandoeng, Van Hoeve, 1:46-59. KAMADA, T. 1933. The vapour pressure of the blood of the edible snail. Jour. Exp. Biol., 10(1): 75-78. BIBLIOGRAPHY 225 KANAMARTJ, T. 1938. ("The fate of edible snail, land earsnail"; in Japanese.) Venus, 8(1): 59-60. KARLIN, E. J., AND NAEGELE, J. A. 1958. Slugs and snails in New York greenhouses. N.Y. State Coll. of Agric., Cornell Univ. Ext. Bull. 1004, 15 pp. 1919. On the life-history and larval anatomy of Melinda cognata Meigan (Diptera Calliphorinae) parasitic in the snail Hellicella [sic] (Heliomanes) virgata Da Costa, with an account of the other diptera living upon molluscs. ParasitoL, 11:430-55. 1921. Supplementary account of the dipterous larvae feeding upon molluscs. Ibid., 13:180-83. KEUCHENIUS, P. E. 1914. Een slakkenplaag in de Hevea. Besoekische Proefstation, Mededeelingen No. 10, pp. 7-9. KEW, H.W. 1893. The dispersal of shells, an inquiry into the means of dispersal possessed by fresh-water and land mollusca . . . with a preface by A. R. Wallace. Internat. Sci. Ser., Vol. 75. London, K. Paul, Trench, Trubner & Co., Ltd., 291 pp. KHALIL, M. B., and HELMY, I. S. 1945. The palm-leaf as a snail trap: A preliminary report. Jour. Roy. Egyptian Med. Assoc., 29(1/2): 1-6. KlECKEBUSCH, W. KEILIN, D. 1953. Beitrag zur Physiologic des chemischen Sinnes von Helix pomatia. ZooL Jahrb., Abt. f. Allg. ZooL u. Physiol. der Tiere, 64(2): 153-82. KlRKLAND, A. H. 1904. Usefulness of the American toad. USDA, Farmers' Bull. No. 196,16 pp. KLEINER, A. 1931. Die Conchylien-Aufnahme der Vogel. (In Hungarian; with German summary.) Aquila (Budapest), 36/37:105-16, 116-20. 1936. La consommation des mollusques par les oiseaux. Oiseaux et la Revue Frangaise d'Ornithologie (Paris), 6:233-51. KNIPLING, E. F. 1955. Possibilities of insect control or eradication through the use of sexually sterile males. Jour. Econ. Ent., 48(4): 459-62. KOBELT, W. 1871. Veranderungen in Conchylienfaunen. Nachrichtsblatt der Deutschen Malakozool. Gesell., 1871:9-14. KOCHS, W. 1892. TJeber die Vorgange beim Einfrieren und Austrocknen von Tieren und Pflanzensamen. Biol. Centralblatt, 12:330-39. KOKLER, M. 1941. Distribuzione della vitamina BI (aneurina e cocarbossilasi) nei Metazoa. Boll. Soc. Ital. Biol. Sperimentale, 16(11): 718-19. KONDO, Y. 1949. Report on experiment island. Invert. Consultants Comm. for Micronesia, Pac. Sci. Board, Nat. Res. Council. 5 pp. (Typescript.) 1950a. The giant African snail (Achatina fulica) in Palau, Pagan, and Guam. Ibid., 13 pp. (Mimeographed.) 19506. Laboratory report on carnivorous snails. Ibid., 3 pp. (Mimeographed.) 1950e. Report on the Achatina fulica investigation on Palau, Pagan and Guam. Ibid., 94 pp. (Typescript.) 226 BIBLIOGRAPHY KONDO, Y.--Continued 1951a. Laboratory report on Gonaxis kibweziensis (Sm.). Ibid., 5 pp. (Mimeographed.) 1951&. Report to Invertebrate Consultants Committee for the Pacific on Gonaxis kibweziensis and Edentulina affinis. Ibid., 7 pp. (Mimeographed.) 1952. Report on carnivorous snail experiment on Agiguan Island; primary and secondary Achatina-free areas on Rota; and gigantism among Achatina on Guam. Ibid., 50 pp. (Mimeographed.) 1956. Second Helix aspersa in Hawaii and data on carnivorous snails. Nautilus, 70(2): 71-72. KOZLOFF, E. N. 1946. The morphology and systematic position of a ciliate parasitizing terrestrial pulmonate molluscs. (Unpub. thesis, Univ. Calif., Berkeley.) KRAUSS, N. L. H. 1947. Achatina fulica (Ferussac). Proc. Hawaiian Ent. Soc., 13(1) :9. 1951-52. Monthly report on investigations in East Africa and Madagascar of natural enemies of giant snail and rhinoceros beetle. July 1951-January 1952 [title varies]. Hawaii Board of Commissioners of Agric. and Forestry. (Dittographed.) 1955. Tefflus zanzibaricus alluaudi Sternberg. Proc. Hawaiian Ent. Soc., 16(1) :1. KRULL, W. H. 1933. A note on the toxic effect of copper for snails. Jour. Parasit., 20(2): 109. KUHN, M. 1911. Die Trypanoplasmen und deren Verbreitung in einheimischen und auslandischen Schnecken. Schriften d. Physikalisch-okonomischen Gesell. Konigsberg., 52(1): 63-89. KURODA, T. 1930. Land and fresh water mollusca from Benin Islands. Bull. Biogeog. Soc. Japan, 1(3): 195-204. 1938. Supplementary report on the fate of edible snail. Venus, 8(3/4). 1941. A catalogue of molluscan shells from Taiwan (Formosa), with descriptions of new species. Taihoku Imp. Univ., Mem. Fac. Sci. Agric., 22(4): 65-216. LACK, D. 1954. The natural regulation of animal numbers. New York, Oxford Univ. Press, 351 pp. LAMY, E. 1928. Quelques mots sur Palbinisme et le melanisme chez les mollusques. Jour, de Conchyl., 72:127-48. LANE, F. W. 1950. Nature's record fasters. Nat. Hist., 59:429-31. LANG, H. 1919. In: H. A. Pilsbry, A review of the land mollusks of the Belgian Congo chiefly based on the collections of the American Museum Congo expedition, 1909-1915. Amer. Mus. Nat. Hist., Bull., 40:54-58. LANGE, W. H., JR. 1947. The giant African snail, Achatina fulica. In his: Observations on the life history of the Mariana coconut beetle (Brontispa mariana Spaeth) and the introduction of parasites for its control. Invert. Consultants Comm. for Micronesia, Pac. Sci. Board, Nat. Res. Council., pp. 6-8. (Mimeographed.) BIBLIOGRAPHY 227 1950. Life history and feeding habits of the giant African snail on Saipan. Paf. Sci., 4(4):323-35. LANGE, W. H., JR., AND MACLEOD, G. F. 1941. Metaldehyde and calcium arsenate in slug and snail baits. Jour. Econ. Ent., 34:321-22. LANGE, W. H., JR., AND SCIARONI, R. H. 1952. Metaldehyde dusts for control of slugs affecting Brussels sprouts in central California. Ibid., 45 (5): 896-97. LARAMBERGUE, M. DE 1955. Presentation d'achatines gastropodes exotiques nuisibles aux cultures. C. R. Assoc. Frang. PAvanc. Sci., 73:450-52. LATIF, S. M. 19330. Een gevaarlijke vijand. De Orchidee (Bandoeng), 2(5): 108-10. 19336. Het slakkendrama. Ibid., 2(6): 123-27. 1933c. [Note] Ibid., 2(7): 178. 1937. Algemeen Landbouwweekblad voor Nederlandsch Indie, 21:701-2. LAVIOLETTE, P. 1950. Sur un retard de la maturite genitale observe chez Arion rufus L. Soc. Linneenne de Lyon, Bull. Mensuel, 19(3): 52-56. LAWSON, A. K. 1920a. Limnaea [sic] stagnalis devoured by rats. Lanes. Ches. Nat. 12:272. 19206. Vitrea and Pyramidula destroyed by ants. Jour. Conch., 16(4): 127. 1921a. Helicella virgata (Da Costa) destroyed by moles. Ibid., 16:226. 19216. Helix hortensis on thrush stones. Ibid., 16 :264. 1921c. Limnaea [sic] stagnalis destroyed by rats. Ibid., 16:144. 1929. Mollusca eaten by rabbits. Ibid., 18(11): 327-28. 1930. Ibid., 19(7): 190. LEBOUK, M. V. 1915. Some feeding habits of slugs. Ann. Appl. Biol., 1(3/4): 393-95. LEEFMANS, S. 1933a. Nog eens: het slakkendrama. De Orchidee (Bandoeng), 2(7): 151-53. 19336. De slakken. Ibid., 2(8): 199-200. 1933c. Voorloopige mededeeling inzake de bestrijding van Achatina fulica Per. in Batavia. (With English summary.) Landbouw, Landbouwkundig Tijdschr. voor Nederlandsch-Indie, 9(6):289-98. (Also in: De Bergcultures, 8:3-6.) LEEFMANS, S., AND VAN DER VECHT, J. 1933a. De groote agaatslak (Achatina fulica Fer.) in Nederlandsch-Indie. (With English summary.) Landbouw, Landbouwkundig Tijdschr. voor Nederlandsch-Indie, 8(10): 668-77. 1933&. De groote agaatslak (Achatina fulica Fer.) in Nederlandsch-Indie. De Bergcultures, 7(21):579-84. (Also in: Algemeen Landbouwweekblad voor Nederlandsch-Indie, 17[47]: 878-81.) LEEUWEN, D. VAN. 1932. Notes and comments. Conchology. Achatina fulica. Hong Kong Naturalist, 3(1): 71. LEGEB, L. 1924. Sur la valeur nutritive de Tescargot (Helix pomatia L.). C. R. Acad. Sci. Paris, 179:1622-25. 1925. La valeur alimentaire et gastonomique de Tescargot de Bourgogne (Helix pomatia L.). Travaux du Lab. d'Hydrobiol. Pisciculture, Grenoble, 15/16:73-81. LEMAN, G. C. 1912. Helix cantiana Mont, eaten by birds. Jour. Conch., 13(12) :359. 228 BIBLIOGRAPHY LENNOX, C. G. 1949. "The giant snails" (Repartee) Atlantic, 184(4): 17. 1951. Division of Entomology and Plant Quarantine. In his: Biennial Report, Hawaii Board of Commissioners of Agric. and Forestry, 1949-50 :32-33. 1953. Ibid., 1951-52:27-28, 31, 33, 34. LEONARD-LILLE, ETABLISSEMENTS 1938. France, patent 827,518, April 28, 1938. LEONI, I. 1953a. O earamjo nas lavouras cafeeiras. Colheitas e Mercados (Sao Paulo), 9(10/12): 13-14. 19536. O problema do caramujo na lavoura cafeeira. Sao Paulo, Superintendencia dos Servigos do Cafe, Bol., 28(315): 12-15. LESSON, R.-P. 1830. Zoologie. In: Louis I. Duperrey, Voyage autour du monde ... sur ... La Coquille, pendant les annees 1822, 1823, 1824 et 1825. 2(1):318-20. LEVER, R. J. A. W. 1939. Entomological notes. Agric. Jour. (Fiji), 10(1): 17-20. LEVY, M. 1890. Zoochemische TJntersuchung der Mitteldarmdriise (Leber) von Helix pomatia. Zeitschr. f. Biologic, 27:398-414. LEVY, P. 1938. France, patent 833,353, Oct. 20, 1938. LEWIS, D. R., MADEL, G. A., AND DRTJRY, J. 1939. Fatal poisoning by "meta fuel" tablets. British Med. Jour., 1939(1): 1283-84. LEWIS, H. C., AND LAFOLLETTE, J. R. 1941. Snail control in orchards. Calif. Citrograph, 26(5): 117, 132. 1942a. Control of European brown snails. Ibid., 27(5) :122, 144. 1942&. Control of the brown snail in citrus orchards. Jour. Econ. Ent., 35:35962. LEY W. 'l949. "D.P.V of the animal world. Frontiers (Philadelphia), 13(5): 139-41. Li, S., see Hi, J. LING, L. 1952. Digest of plant quarantine regulations (new ed.), United Nations, FAO, Agric. Div., Plant Protection Br., Development Paper No. 23, 164 pp. 1954. Digest of plant quarantine regulations. First supplement. Ibid.9 100 pp. LOCARD, A. 1890. Les huftres et les mollusques comestibles, moules, prairies, clovisses, escargots, etc.; histoire naturelle, culture industrielle, hygiene alimentaire. Paris, J.-B. Bailliere et Fils, 383 pp. LOCKWOOD, S. 1880. Vitality of Helix aspera [sic]. Amer. Nat., 14:214. LONGSTAFF, J. 1921. Observations on the habits of Cochlitoma zebra var. fulgurata (Pf eiffer), and Cochlitoma zebra var. obesa (Pfeiffer), in confinement. Proc. Zool. Soc. London, 1921:379-87. LOPES, H. DE S. 1940. Contribuigao ao conhicimento do genero Udamopyga Hall e de outros Sarcophigideos que vivem em molluscos no Brazil. Revista de Entomologia (Rio de Janeiro), 11:924-54. LOVERIDGE, A. 1945. Reptiles of the Pacific world. New York, Macmillan Co., 259 pp. BIBLIOGRAPHY LOVETT, A. L., AND BLACK, A. B. 229 The gray garden slug with notes on allied forms. Oregon Agric. Coll. Exp. Sta. Bull., 170:1-43. LUCAS, F. A. 1897. Biological society of Washington--282nd meeting, Saturday, November 20. Science, N.S., 6:886. LUCAS, H. 1842. Sur une nouvelle espece du genre Drilus qui habile le nord de 1'Afrique. C. R. Acad. Sci. Paris, 15:1187-89. 1870. Note sur le Malacogaster bassi, sp. nov., vivant dans les Helix dupoteti, lucasi, etc. Ann. Soc. Ent. Paris, 4e ser., Bull. Seances, 10 :lvii-lviii. 1871. Description des deux sexes (Tune nouvelle espece de Malacogaster (M. bassii), precedee de quelques remarques sur cette coupe generique. Ibid., 5eser., 1:19-28. LUCAS, W. 1949. A bill to prevent the entry of certain giant snails into the United States. House of Representatives Bill 6242, 81st Cong., 1st sess. Sept. 27, 1949. 1951. A bill to prevent the entry of certain mollusks into the United States. House of Representatives Bill 4443, 82d Cong., 1st sess. June 13, 1951. LUTZ, F. E. 1927. Minutes meeting of December 7,1926. Jour. N.Y. Ent. Soc., 35(3): 315. McATEE, W. L. 1920. 1918. Food habits of the mallard ducks of the United States. USDA Bull. No. 720, 35 pp. McCANCE, R. A., AND StflPP, H. L. 1933. The magnesium and other inorganic constituents of some marine invertebrates. Jour. Marine Biol. Assoc. United Kingdom, 19(1): 293-96. McCoRMiCK, C. J. 1952. Title 7 Agriculture Chapter HI--Bureau of Entomology and Plant Quarantine Part 324 Mollusks, Regulations governing entry. Fed. Register, 17 (207): 9582-84, Oct. 22, 1952. McCREERY, J. M. 1890. Note on some of the causes of extinction of species. Amer. Geologist, 5:100-104. McCRORY, S. J. 1950. Division of Entomology. Louisiana Dept. Agric. & Immig., Bien. Kept., 34:35. MACDOTJGALL, R. S. 1931. 1951. Insects and other inveretebrates in 1930. Trans. Highland and Agric. Soc. Scotland, 43 :109-54. McFARLAND, E. W. A bill to prevent the entry of certain mollusks into the United States. Senate of the U.S. Bill 1489, 82d Cong., 1st sess., May 15, 1951. MACKERRAS, M. J., AND SANDARS, D. F. 1954. Life-history of the rat lung-worm and its migration through the brain of its host. Nature, 173(4411):956-57. McLAUCHLAN, C. F. Experiments upon certain snails inhabiting Sydney gardens. Proc. Roy. Soc. N. S. Wales, 1947-48:21-24. MACMILLAN, H. F. 1943. Tropical planting and gardening with special reference to Ceylon (5th ed.). London, Macmillan Co., 560 pp. 1949. 230 BIBLIOGRAPHY MAHONY, F. J. 1955. Truk with snails. Time, 66(4) :6. July 25, 1955. MARCHI, C. 1928a. Sul valore vitaminico del molluschi eduli. I. Presenza di vitamina C in lamellibranchi e gasteropodi. Monitore Zoologico Ital., 39(3/4): 83-90. 19286. Sur valore vitaminico dei molluschi eduli. II. Esperimenti con lamellibranchi e gasteropodi cotti e con cefalopodi. Ibid., 39 (9): 215-33. 1929. Ricerche sulla presenza di vitamina C in alcuni molluschi eduli. Memoria R. Comitato Talassografico Ital. Venezia, 152:1-18. MARTINSON, J. S. The local edible-snail industry. Yearbook 1928, Gold Coast Dept. Agric. Bull., 16:231-35. MASEFIELD, J. R. B. 1913. Helix cantiana eaten by birds. Jour. Conch., 14(1): 10. MASSEE, A. M. 1928. Notes on insect pests for the years 1926-1927. East Mailing Res. Sta., A n n . Kept. H, Suppl., 14/15:157-62. 1929. MATSUI, Y. 1942. ("Diurnal periodic activities of the large African snails"; in Japanese.) Kagaku Nanyo ("Science of the South Seas," Jour. Palau Trop. Biol. Sta.), 5(1): 137-43. MAUBERT, J. 1943. La cria del caraeol y su explotacion. Esnea, 30(1266): 5. MEAD, A. R. 1949a. Advance report re selection of islands for experimental biological control of Achatina fulica in the Pacific. Invert. Consultants Comm. for Micronesia, Pac. Sci. Board, Nat. Res. Council. 4 pp. (Typescript.) 1949&. The giant snails. Atlantic, 184(2): 38-42. 1949c. The giant snails. The Biologist, 32(1/2) :9-15. 1949