Micronesica Supplement No. 3 June, 1991 by MARC

MICRONESICA

A Journal of the University of Guam DEVOTED TO THE NATURAL SCIENCES OF MICRONESIA AND RELATED AREAS

ISSN 0026-279X

Supplement No.3, June 1991

Exotic Pests in the Pacific-Problems and Solutions I

Proceedings of a Workshop May 31-June 1, 1990

Edited by

R. Muniappan M. Marutani

Agricultural Experiment Station

G. R. W. Denton

Water and Energy Resources Institute University of Guam

Sponsored by Pacific Science Association Scientific Committee on Entomology & Agricultural Experiment Station University of Guam

MICRONESICA A Journal of the University of Guam Devoted to the Natural Sciences in Micronesia Founded by Benjamin C. Stone in 1964

EDITOR: CHRISTOPHER S. LOBBAN The Marine Laboratory, University of Guam UOG Station, Mangilao, Guam 96923, U.S.A.

Editorial Board: C. E. BIRKELAND, The Marine Laboratory, University of Guam L. G. ELDREDGE, Bernice P. Bishop Museum, Honolulu, Hawaii

F. R. FosBERG, National Museum of Natural History, Smithsonian Institution, Washington, D.C. M. J. LEVIN, Population Division, U.S. Bureau of the Census, Washington, D.C. M. MARSHALL, Department of Anthropology, University of Iowa, Iowa City J. E. RANDALL, Bernice P. Bishop Museum, Honolulu, Hawaii R. H. RICHMOND, The Marine Laboratory, University of Guam D. H. RUBINSTEIN, Micronesian Area Research Center, University of Guam B. C. STONE, Bernice P. Bishop Museum, Honolulu, Hawaii J. H. UNDERWOOD, Dept. Anthropology, University of Arizona, Tucson MICRONESICA is a forum for original research in the fields of physical and social anthropology, archeology, linguistics, and ethnology; systematic and ecological botany and zoology, marine sciences, and related disciplines concerned primarily with Micronesia and adjacent areas. MICRONESICA is published twice a year (June and December) by the University of Guam. Subscription prices: Individual $15 per volume, Institutional $25 per volume (including any Supplement). Views expressed by the authors are their own and do not necessarily reflect those of the University, the editors or printers. ÂŠ 1991 by the University of Guam Press Micronesica is listed in Biological Abstracts (BIOSIS)

CONTENTS v-vi Vll

1-4 5-13 15-31

33-39 41-45 47-50 51-62 63-69 71-81 83-92 93-98 99-101 103-107 109-116 117-122 123-127 129-133

Introductory Remarks ................................................................. G. G. E. Scudder Resolutions Introduction of Arthropod Pests into the Hawaiian Islands ..... J. W. Beardsley Sources of New Insects Established on Guam in the Post World War II Period ............................................................................................... /. H. Schreiner Movement of New Insects into the Carolines and the Marshalls in Recent Years ........................... ............................................................................... D. Nafus Introduced Vector-Borne Diseases in the Pacific ...:........................ R. C. Russell Plant Diseases of Recent Introduction to Guam ............................... G. C. Wall Introduced Ornamental Plants that Have Become Weeds on Guam ...................... ............................................. J. McConnell & R. Muniappan Spread of Fresh-Water Pomacea Snails (Pilidae, Mollusca) from Argentina to Asia ........................................ ............................................................... 0. Mochida Brown Tree Snake (Boiga irregularis) on Guam: A Worst Case Scenario of an Introduced Predator ........................... ........................................ M. J. McCoid The Distribution and Biological Control of Lantana camara in Micronesia .............................. G. R. W. Denton, R. Muniappan & M. Marutani Biological Control: Mutual Advantages of Interaction Between Australia and the Oceanic Pacific ........................................ ............................. D. F. Waterhouse Banana Skipper, Erionota thrax (L.) (Lepidoptera: Hesperiidae) in Papua New Guinea: A New Pest in the Pacific Region .... D. P. A. Sands, M. C. Sands & M. Arura Biological Control of Some Introduced Pests in the Federated States of Micronesia .......................................................................... .................N. Esguerra Distribution and Control of Chromolaena odorata (Asteraceae) .......................................................... R. Muniappan & M. Marutani Occurrence of the Giant African Snail in the Ogasawara (Bonin) Islands, Japan ................................. .................... ........................ K. Takeuchi & S. Kayano Possibilities for the Biological Control of the Breadfruit Mealybug, Icerya aegyptiaca on Pacific Atolls ........................................... D. F. Waterhouse Cultural Methods of Pest Control on Taro (Colacasia esculenta Schott) in American Samoa ..................................... S. Fatuesi, P. Tauili'ili & A.M. Vargo Automated Identification of Insects in Flight... ..................................... A. Moore

)

MICRONESICA INSTRUCTIONS FOR CONTRIBUTORS Scientific research reports, notes, review papers, bibliographies, and book reviews in anthropology, biology, and related fields are accepted on the basis of their originality and their pertinence to Micronesia and the adjacent Pacific areas. Descriptions of new species will be considered formal papers, no matter how short; information in range extensions will be considered notes, no matter how long. The manuscripts must be written in English, but a summary in another language is acceptable. Each manuscript will be reviewed by at least two members of the Editorial Board or by specialists other than board members in whose field the paper lies. Manuscripts should be sent via airmail to The Editor, The Marine Laboratory, University of Guam, UOG Station, Mangilao, Guam 96923, U.S.A. The original and two clear copies of text and artwork are required; the original will be retained in the editorial office while the copies aresent out for review. Authors must follow the guidelines below. Papers which deviate from the required format may be returned for revision before review. General: Micronesica has a broad readership. Authors are encouraged to write the Abstract and opening paragraphs of the Introduction to be intelligible to that broad public, even though the body of the work may be highly specialist. The manuscript must be typed on one side of 8'hX 11 or A4 paper, with generous margins all around. DOUBLE SPACE EVERTHING, including abstract, references and tables. Clear print is essential-faint type or dot matrix printout is unacceptable. The layout of the text should follow a scientific format suitable to the material (see previous issues of Micronesica). Title page: The first page should give dnly the title (capitals and lower case, please), authors'

names and addresses, and the running head. Present addresses, if different, are to be given in a footnote on the first page. However, contribution numbers should be given in the Acknowledgements. Abstract: For formal papers, the second page should provide an informative abstract of

not more than 300 words. Remember that Micronesica is accessible through electronic retrieval, and give a synopsis that is complete without reference to the text. Do not repeat information given in the title.

Text: Main headings are set in capitals and lower case, centered, subheadings are in small capitals, centered. Underline words to be set in italic (Latin names, foreign terms, emphasis), but do not underline for boldface. Avoid footnotes in the text. Indicate with a marginal note where tables and figures should appear. Cite references by author and date, and follow the punctuation style carefully: "Smith (1987) found . .. "; "as shown by various authors (Cheng et al. 1979, Cruz 1986, 1987, Jones & Jones 1989) . ..". Well-known, standard acronyms such as DNA may be used without definition, but other acronyms and abbreviations should be defined when first used; use them sparingly. Acknowledgements are placed at the end of the text, before the References. . References: Ensure that references are complete and accurate. Please follow style and

punctuation closely! List references in alphabetical order by first author. If citing several papers by one author, list single-authored papers in chronological order (Adams 1976, Adams 1980 . . .), then co-authored papers in alphabetical order of co-authors (Adams & Brown 1989, Adams & Ng 1975), and finally, multiple-author papers in chronological

order (Adams et al. 1985, 1988 ... ). Write out all authors' names in each reference do not use a dash instead-and note that second and subsequent authors' initials are placed before the name. Do not italicize journal or book titles. Do not give issue number unless each issue is separately paginated. Examples of journal articles and books: Underwood, J. A. 1989. Population history of Nauru: a cautionary tale. Micronesica 22: 3-22. Randall, J. E. 1958. A review of the labrid fish genus Labroides, with descriptions of two new species and notes on the ecology. Pac. Sci. 12: 327-347. Trono, G. C., Jr. & E. T. Ganzon-Fortes. 1988. Philippine Seaweeds. National Bookstore, Manila. 330 pp. Example of article in book: Tsuda, R. T. 1985. Gracilaria from Micronesica: key, list and distribution of the species. In I. A. Abbott & J. N. Norris (eds), Taxonomy of Economic Seaweeds, pp. 91-92. Calif. Sea Grant Program, La Jolla. Tables: Tables must be double spaced and collected at the end of the text. Explain any abbreviations in the legend at the top of the table. Illustrations: Plan illustrations to fit the printed page size, 125 mm X 180 mm, allowing enough space for the legend. One set of illustrations for the printer should be mounted on heavy card and covered with protective sheets of paper; two review sets may be unmounted. Be sure to indicate the top of the illustration and your name and figure number on the back of the mounting board. When a scale bar is appropriate its length should be given in the legend, not on the illustration. Do not give magnifications. If an illustration has been published before, due credit must be given, and written permission to reprint obtained from the copyright holder, the author, and the illustrator if necessary. Photographs and line art figures should be numbered in the sequence in which they are used in the text, and the legends grouped onto pages placed at the end ofthe manuscript, after any Tables. Photographs must be clear and be printed on glossy paper with good contrast. Crop photographs to include only essential detail. Cropping for ungrouped photographs may be indicated on the print borders or on an overlay sheet. If several small photographs are used, they should be grouped to fit the page width or made into full page plates; in that case, each photograph should bear its number legibly in one of the lower corners. Color photographs will be printed if the author bears the cost of reproduction. Line art must be presented as original India ink drawings or as photographic reductions or Photomechanical Transfers (PMT's). Photocopies are unacceptable. Send artwork flat and mounted, not rolled. If the originals are larger than 11 X 14 have them professionally reduced, preferably to a size to fit the journal page. Be sure than lettering is large enough that when printed the smallest letters will be at least 1 mm high and that lines are thick enough and spaced widely enough that they will not fade or run together when reduced. Use transfer lettering or mechanical lettering devices, not a typewriter to produce lettering. Computer-generated graphics are acceptable only if they meet professional line art standards. Proofs: Authors will be sent galley proofs for correction. Authors may be chatged for changes other than corrections of typesetting errors. Reprints: Reprints may be ordered on a form sent out with the proof sheets. Rev. 1/91.

Introductory Remarks G. G. E.

SCUDDER

Department of Zoology The University of British Columbia Vancouver, B.C. Canada V6T JZ4

On behalf of the Pacific Science Association, I wish to welcome everyone to this workshop. The PSA is an international, non-governmental, regional scientific organization founded in Hawaii in 1920. It is thus 70 years old this year. The objectives of the Association are to initiate and promote cooperation in the study of scientific problems relating to the Pacific region, more particularly those affecting the prosperity and well-being of the Pacific peoples. Traditionally, principal activities of the PSA have been the running of Congresses every four years. The XVI Congress was held in Seoul in 1987, while the next will be held in Honolulu, Hawaii, May 27-June 2, 1991 with the theme "Towards the Pacific Century: The Challenge of Change." In addition, the PSA organizes Inter-Congresses, which are held at four-year intervals, meeting between Congresses. The 6th Inter-Congress was held in Chile in 1989 with the theme "The Pacific Ocean: Bridge or Barrier?". The next InterCongress will be held in Okinawa in 1993. At the Seoul Congress, the PSA Council decided that the Association should be reorganized, and that Congresses and Inter-Congresses would stress interdisciplinary symposia, and that Scientific Committees should promote study of discipline-based topics between the major PSA meetings. As a result, one of the topics for the Honolulu Congress will be "Global Environmental Change-Pacific Aspects." Many scientists are stressing that governments should be concerned about the changes which could occur in the next few decades. The latest global change models show that an average temperature increase of 2.5-3.0째C can be expected, and that the south and north colder regions of the world could experience an increase of 6- 1ooc. If this is true, there would be melting of glaciers and ice-caps with considerable rise in sea-level as a result. This would result in catastrophic flooding of all low-lying islands in the Pacific, and similar low-lying areas throughout the world. The consequences of such flooding have not received serious consideration to date, but can no longer be ignored. The PSA Scientific Committee for Entomology will take part in these interdisciplinary symposia, but we also have a program of our own that has three major objectives: (1) To promote interaction between Pacific entomologists, and to this end we are finalizing a revision of our list of entomologists working in and on the Pacific.

)

(2) To promote studies on the Pacific insect fauna through various cooperative efforts. In particular, we are now working with the PSA Committees on Botany and Conservation in organizing a biological survey of New Caledonia. (3) To study the insect pest problems in the Pacific. Thus, at the Bali InterCongress in 1977, we held a symposium on the Brown Planthopper, which is a major pest of rice, and at the Manila Inter-Congress we held a symposium on transport of pests. At the PSA reorganization meetings in Seoul, and subsequently in discussions over the last two years at special meetings held in Singapore and Macao, it was decided that we would try to hold Scientific Meetings in the islands of the Pacific. The PSA Scientific Committee on Entomology decided to hold a workshop on "Exotic Pests in the Pacific," and thus this meeting here in Guam is the first of these new endeavors. As Chairman of the PSA Scientific Committee on Entomology, and as a member of the PSA Council and Executive Committee, I am pleased to see that so many have been able to come to Guam to participate in this workshop. We are deeply indebted to Dr. R. Muniappan for organizing this workshop on our behalf, and the University of Guam for hosting this meeting. The program with 25 scheduled papers clearly demonstrates that we selected an appropriate topic. I look forward to the next two days, and hope all will find the discussions interesting and worthwhile.

- l

Resolutions

These Resolutions were passed on June 1, 1990, at the Workshop on "Exotic Pests in the Pacific-Problems and Solutions" held under the auspices ofthe PSA Scientific Committee for Entomology at the University of Guam, Guam. The Scientific Committee for Entomology of the Pacific Science Association recognizes the great importance of ocean barriers in limiting the risks for spread and introduction of exotic pests, weeds, and insect vectors of disease. While recognizing the sovereign authority of nations, their states and territories for all quarantine matters, this workshop believes that the responsibility rests with each of them to minimize these risks and presents the following resolution: Whereas well over 90 percent of the major pests in the Pacific are exotic introductions; and Whereas there is concern for the introduction of mosquito-borne diseases to areas where they do not exist and would constitute a hazard to public health; and Whereas there is a steady increase in the frequency of introductions which is correlated with increased air transportation and commerce; and Whereas it is apparent that these introductions are also associated with transportation in First Class mail, horticultural imports, and intentional and unintentional transport in planes and ships; and Whereas there are ever increasing new air routes being introduced into the Pacific area by various carriers; and Whereas island ecosystems are fragile, easily perturbed, and the biota subject to competition from and extermin'ation by introduced exotics; Be it Resolved that the Pacific Science Association urge all nations, their states and territories to adopt effective means of inspecting all quarantinable packages, including First Class mail; and Be it Resolved that these establish plant quarantine procedures for (i) inspections; (ii) holding in a quarantine area, or (iii) treatment when necessary; and Be it Further Resolved that there be disinsectization of all aircraft traveling to and between islands, and that airports and their surroundings be maintained in a sanitary condition in order to reduce the chances of export and establishment of introduced pests and vectors.

Micronesica Suppl. 3: 1- 4, 1991

Introduction of Arthropod Pests into the Hawaiian Islands JOHN

BEARDSLEY

Department of Entomology College of Tropical Agriculture and Human Resources University of Hawaii, Honolulu, Hawaii 96822

Abstract-Nearly all pest arthropods now present in Hawaii are nonnative species that have been introduced to the island since their discovery by man ca. 1,000 years ago. A few introduced pests, such as houseflies, head and pubic lice, and cockroaches, were present at the time of European discovery (mid to late 18th century), presumably as a result of Polynesian exploration and trade. Accidental introductions of pest arthropods became more common during the 19th century as a result of commerce with the outside world, and this trend has accelerated during the 20th century, despite the application of quarantine regulations during the 1890s. The advent of regular trans-Pacific air travel during the 1940s and of jet aircraft during the 1960s have provided new opportunities for rapid dispersal of pest arthropods, to Hawaii and throughout the Pacific Basin~ These are exemplified by recent epidemic outbreaks of such new pests as the spiraling whitefly, the leucaena psyllid and the melon thrips. Due to the constantly increasing volume of transPacific air traffic to Hawaii, and to the existence there of many underutilized ecological niches, it appears likely that new pests will continue to become established there at a relatively high rate, despite improvements in quarantine enforcement and pest detection techniques. Hawaii is a group of islands isolated near the center of the Pacific Ocean with a land area of about 6,450 square miles. There are about 10,000 species of insects and other terrestrial arthropods present in Hawaii today. Of these, somewhat more than 2,000 species have been introduced into the islands from overseas since man first settled there, around 1,000 years ago. Hawaii has some 500 species of terrestrial arthropods and mollusks which can be classed as pests. Of these, only a very few are native to the islands: 98% of the pest species have been introduced. Therefore, in Hawaii at least, exotic pest problems are almost our only pest problems. The Polynesian voyagers who first populated the Hawaiian islands probably carried relatively few foreign arthropods with them. The first Europeans to visit the islands, in the middle to late 18th century, found the Hawaiians bothered by flies (probably Musca domestica L.), head and pubic lice, and not much else. Their crops (taro, coconut, sweet potato, yams, sugarcane and a few fruits) apparently were largely pestfree. There were no mosquitoes (Illingworth 1923).

Micronesica Suppl. 3, 1991

However, following the European discovery of the islands, commerce with the outside world began, and introductions of foreign arthropods became more frequent. In the era of sailing ships many weeks or months at sea were passed before landfall was made in Hawaii. At first, long-lived, hardy arthropods which could live aboard ship in man's shadow (e.g., cockroaches) were among the few which could make it alive to the islands. Other species which became established during this period could live in building materials or ballast (e.g., centipedes, millipedes, tenebrionid beetles) or were able to reproduce during the voyage (e.g., stored product pests, mosquitos in water casks, fleas on domestic animals). With European settlement and the initiation of European agriculture, the importation of living plant materials for propagation brought with it a multitude of foreign plant parasites, like scale insects, mealybugs, and aphids. Uncontrolled importations of soil, lumber and furniture brought such pests as ants, boring beetles and termites. The implementation of plant quarantine regulations by the Hawaiian government, near the end of the nineteenth century slowed, but did not halt the influx of new pests. Regular steamship traffic between Hawaii and the continents of North America, Asia, and the South Pacific Islands, beginning in the late nineteenth century, reduced transit time for both humans and arthropod stowaways, increasing chances for new pests to survive. The volume of overseas traffic to Hawaii increased as the nonindigenous human population of the islands grew, and as plantation agriculture, bas~d largely on imported labor, became the economic mainstay of the island economy. More and more new exotic arthropods arrived and became established, as may be seen from records published since 1905 in the Proceedings of the Hawaiian Entomological Society. Published records of accidental introductions of arthropods into the islands prior to that year are generally sketchy and incomplete. Much of the available information was summarized by Illingworth (1923). The rate of arthropod introduction into Hawaii took a quantum leap with the arrival of the airplane. Regular transpacific air transport was initiated during the late 1930s, and has been increasing exponentially ever since. In the 1960s, jet aircraft, each capable of transporting several hundred people as well as innumerable hitch-hiking arthropods, were introduced in the Pacific Basin. These aircraft now reach Hawaii hourly from North America, Asia and the South Pacific. For the past 25 years or so, new, accidentally introduced, foreign arthropods have been found established 路in Hawaii at the rate of about 20 species per year (Beardsley 1979). That is, since 1965, about 500 new terrestrial arthropod species have been accidentally introduced into and become established in the Hawaiian Islands. Many of these immigrants have been of little or no economic "consequence. However, hardly a year has gone by during this period that we were not faced with one or more new arthropod pest problems of serious economic significance. In 1984, I put together a list of new arthropod pests of minor to major importance that were found for the first time in Hawaii since 1950. That year

Beardsley: Arthropod Pests in Hawaii

marked the midpoint of the 20th century, as well as the beginning of major increases in transPacific air traffic related to military activities and tourism. The list contained the names of 105 species (including three mollusks). It included such major pests as: spiraling whitefly (Aluerodicus dispersus Russell), sweetpotato whitefly [Bemesia !abaci (Gennadius)], taro root aphid [Patchiella reaumuri (Kaltenbach)], Eurasian pine adelgid [Pineus pini (Macquart)], coconut scale (Aspidiotus destructor Signoret), Egyptian hibiscus mealybug [Maconellicoccus hirsutus (Green)], leucaena psyllid (Heteropsylla cubana Crawford), southern green stink bug [Nezara viridula (L.)], melon thrips (Thrips palmi Karny), Western flower thrips [Frankliniella occidentalis (Pergande)], banana root borer [Cosmopolites sordidus (Germar)], black twig borer (Xlosandrus compactus Eichhoff), banana skipper [Erionota thrax (L.)], lawn armyworm [Spodoptera mauritia (Boisduval)], litchi fruit moth [Cryptophlebia ombrodelta (Lower)], celery leafminer [Liriomyza trifolii (Burgess)], Malaysian fruit fly [Dacus latifrons (Hendel)], long-legged ant [Anaplolepis longipes (Jerdon)], and brown garden snail [Helix aspersa (Muller)]. In preparing this paper I updated my 1984list to the end of 1989, and added 36 additional pest species new to Hawaii, including the following important ones: fruit-piercing moth [Othreisfullonica (Clerck)], sugarcane tingid [Leptodictya tabida (Herrick-Schaeffer)], annona seed wasp [Bephratelloides cubensis (Ashmead)], lesser cornstalk borer [Elasmopalpus lignoscellus (Zeller)], blue alfalfa aphid (Acyrtosiphon kondoi Shinji), yellow sugarcane aphid [Siphajlava (Forbes)], and tropical nut borer [Hypothenemus obscurus (Fabricius)]. Thus, by my reckoning, 140 ~ew pests have become established in Hawaii during the past 40 years, or, on average 3.5 new pests per year. Comparable data for other parts of the world are not readily available, but it would be instructive to determine if other Pacific island areas have experienced similar rates of new pest introduction. The advent of regular jet flights between widely separated localities throughout the Pacific Basin by aircraft which are rarely subject to inspection or disinsectization, has resulted in the extremely rapid spread of some pest species. The leucaena psyllid is a case in point. This pest was unknown outside of the Caribbean region and adjacent Central America until it turned up in Florida in 1983. It made the big jump to Honolulu the following year where it soon devastated leucaena (tangantangan) throughout the Hawaiian Islands. Some people thought that this was a good thing, but others, mostly ranchers who relied upon leucaena for forage, felt otherwise. In the six years since its arrival in Hawaii, the leucaena psyllid spread completely throughout the Pacific Basin and into southeast Asia, as far west as India and Sri Lanka. It is certain to spread circumglobally wherever leucaena grows, within a few more years at most. I think it is likely that the leucaena psyllid has been spread primarily as hitch-hiking adults on aircraft. Adults of this psyllid are strongly attracted to lights, which are used for night loading aircraft. Its host, Leucaena leucocephala, is one of the most common elements of the disturbed area vegetation that is usually found growing around airports in the tropics. The spiraling whitefly, the melon thrips, and the celery leafminer, although they spread

Micronesica Suppl. 3, 1991

with less spectacular rapidity, are also pests of recent arrival in Hawaii which have become widely distributed in the Pacific Basin via aircraft during the past decade or so. Some kinds of pests, for example scales, mealybugs and whiteflies, can only be transported long distances on living plants. In Hawaii, 12 new whitefly species have been found established during the past 25 years. Many of these species, like the spiraling whitefly, come from Florida and the Caribbean region. How can such pests get into Hawaii when all living plants imported there are subject to inspection? The answer, I believe, is largely via the U.S. Mail. Air mail packages are first class mail, and within the U .S. they can not be opened for inspection unless the recipient agrees. If an insect-infested plant is air-mailed from Florida or another U.S. state to a recipient in Hawaii there is no way to prevent the importation and possible establishment of any pests which it harbors. Many important Neotropical arthropod pests are now established in Florida, and others probably are being imported there, frequently on smuggled plant material. This source, which might be called the "Florida Connection", may be responsible for the increased number of Neotropical pests which have become established in the Pacific Basin during recent years. Such new pests in the Pacific as the spiraling whitefly, the annona seed wasp, the sugarcane tingid, the yellow sugarcane aphid and the tropical nut borer may have reached Hawaii by this means. Two measures which coukJ be implemented to reduce the spread of new arthropod pests into and within the Pacific Basin are: 1) the regular disinsectization of all aircraft arriving at, and flying between, Pacific destinations, and 2) the inspection of all packages shipped by mail, first class or otherwise. Unless such measures ca路n be initiated and effectively carried out, it seems reasonable to assume that new pests will continue to become established in Hawaii and in other Pacific areas at the relatively high rates which presently prevail. Journal Series No. 3538, Hawaii Institute ofTropical Agriculture and Human Resources. References Beardsley, J. W. 1979. New immigrant insects in Hawaii: 1962 through 1976. Proc. Hawaii. Entomol. Soc. 23: 35-44. Illingworth, J. F. 1923. Early References to Hawaiian Entomology. B.P. Bishop Museum Bull. 2. 63p.

. l

Micronesica Suppl. 3: 5-13, 1991

Sources of New Insects Established on Guam in the Post World War II Period ILsE

H. ScHREINER

College of Agriculture and Life Sciences. University of Guam, Mangi/ao, Guam 96923

Abstract- The introduction of exotic species into Guam is a continuing problem, with the many serious insect pests being post World War II introductions. In the 1980s at least 25 new species were accidentally introduced. New pests include a grasshopper, a scarab beetle, several noctuid moths and a variety of small beetles, scales, thrips, whiteflies. In the last decade about one-third ofthe new pests on Guam apparently originated from the New World via Hawaii. Other pests arrived from the Oriental region and from other islands in the Micronesian area. Although the mode of introduction of most insects cannot be determined with precision, it can be deduced in at least three recent cases, one having arrived via an Air Force flight, and two having arrived with commercial shipments of ornamentals from Hawaii. The movement of ornamental plants and flowe~s is probably the most important means by which insects are currently moved around the Pacific region. Introduction

The introduction of new insect pests is a serious problem in many Pacific islands. In the South Pacific, past movement of pests has been often linked to shipping lines (Dale & Maddison 1984). Islands with political and trade links often shared pests not found in other intervening islands. Pests moved in the early part of the century were often those associated with the transportable portions of their hosts, such as roots and tubers or baskets made of vegetable material. In recent history, air travel has increased the rate at which pests move about the Pacific. Air transit has speeded up travel between the Pacific islands, and provided new opportunities for insect movement on materials such as fresh flowers and foliage or in the holds and passenger compartments of aircraft. The pace of new introductions has been shown to have accelerated recently in various locations such as Hawaii (Beardsley 1979) and Guam (Schreiner & Nafus 1986). Data from earlier in the century is no longer helpful in identifying areas of greatest quarantine threat. In this paper I will briefly examine the biology of the insects which were first found on Guam during the 1980s, speculate as to the origins and mode of entry, and then summarize the information obtained. At least 24 species of insects and 1 mite have been recorded as new arrivals on Guam during the 1980s. The species recorded as new are either pests or large

Micronesica Suppl. 3, 1991

and showy species for which it is easy to be certain that they are relatively new introductions. Other less conspicuous species have also been identified as new in the last decade, but because they are not of great economic significance, it is difficult to be certain whether they are new introductions or have been here for decades and were simply not noted. No mosquito surveys were done in the 1980s. In previous years mosquito surveys added a few species every decade to the list of insects known to occur on Guam (Ward 1984), and it seems likely the 1980s were no different. New Introductions ORTHOPTERA

Stenocatantops splendens (Thunberg) is a grasshopper whose native distribution is Asia. It was first collected in 1984 at which time it was enormously abundant all over the island and did considerable damage to various vegetable crops. Since then it has been less abundant, though it is the most commonly encountered grasshopper in cultivated fields, and still sometimes damages plantings. Recently, a homeowner brought in a potted ornamental from his yard. The pot contained a large number of grasshopper eggs in the soil, most probably this species. Potted plants might have served as a mode of entry for this pest on Guam. THYSANOPTERA

Thrips palmi Karny, the melon thrips, was first noted in Guam in 1983, though it did not reach outbreak status until 1984 (Schreiner & Nafus 1986). I cannot be certain of the origin of this pest as it is widely distributed in Asia also, but it did not become a problem in Guam until after it had been introduced to Hawaii and had become a pest there. Although it is a serious pest of the foliage of several vegetable crops, these are rarely brought to Guam. I suspect that the more likely mode of entry was in orchid flowers, which are known to be occasionally infested by this species (Nakahara et al. 1986). There is no restriction in the importion of orchids from Hawaii to Guam, and it is known that, as well as individual plants imported by homeowners, several large boxes offlowering plants are imported weekly for sale in retail outlets and the fleamarket. In contrast only limited numbers of orchids were being imported from Asia. HOMOPTERA

Aleurodicus dispersus Russell, the spiralling whitefly, is a Neotropical species which was first noted in Hawaii in 1978. It has a wide host range and ipfests a wide variety of ornamental species. It arrived on Guam in 1981 , most likely with a commercial shipment of ornamentals obtained from Hawaii (Schreiner & Nafus 1986). Since then the spiraling whitefly has spread to a number of other locations in Micronesia and to other islands in the Pacific. Aleurothrixus jloccosus (Maskell), the woolly whitefly, is a pest of citrus and guava trees. During the last 20 years it has spread from its original distribution

Schreiner: New insects in Guam

in the N eo tropics to California, Europe and to islands in the Indian Ocean. It is a recent immigrant to Hawaii, where a species of Eretmocerus already present on Hawaii switched from another host to provide good control of the whitefly (Paulson 1983). An unidentified Eretmocerus sp. appeared in Guam concurrently with the whitefly in 1984, which makes it probable that Hawaii was the source of this whitefly. Ceroplastes ceriferus (F.), the Indian wax scale, was first noted on Guam by Dr. J. Beardsley in 1984 during a brief visit (Beardsley 1986). The source of this insect is not known. It is widely distributed in the tropics and was previously known from Palau (Beardsley 1966). Steatococcus samaraius Morrison is a magarodid species found previously in New Guinea, Palau and Yap (Beardsley 1966). This insect was briefly abundant on monkeypod trees on Guam around 1985, but the ladybeetle Rodolia pumila, introduced to control other magarodid species, feeds on this species, and S. samaraius has not become a pest on Guam. S. samaraius has a wide host range and could have been introduced on a number of ornamental plants or young trees. A Palauan population of several thousand people residing permanently on Guam is known to import ornamental plants from their home island for their gardens on Guam . . LJ . ~:::ropsylla cubana Crawford, the leucaena psyllid, is native to Central America. It spread to Florida in 1983 and to Hawaii in 1984 (Nakahara & Lai 1984) where it became enormously abundant. It appeared almost simultaneously in a number of islands in the western edge of the Pacific in 1985 and then spread westward in the following year, making it possible that this insect was spread in the wind (Waterhouse & Norris 1987). As it is attracted to lights, it might have also spread in the holds of airplanes. This is the most likely method by which the insect reached Hawaii. A palm aphid Cerataphis sp. was found on Guam in 1988 but has not yet been identified. C. lataniae (Boisduval) has been present on other islands in the Marianas and Micronesia for many years (Essig 1956). Cerataphis species are also found in Hawaii and the Orient. On Guam, the palm aphids are not particularly common, and may have here for years prior to being detected. It was found on coconuts and betelnuts. Flaccia diane Fennah is a species of derbid first found on Guam in 1985. Adults occasionally become very numerous on bananas, although they do little damage. This insect was previously known from the Marshall Islands, several Caroline atolls and Kosrae (Fennah 1956), where it also periodically becomes extremely abundant. Melormenis basalis (Walker) is a flatid bug first noted in Guam in 1985. It is a probable import from Hawaii where it was first noted in 1967 (Shiroma 1968). The species originated in the Neotropics. The flatid attacks a number of host plants but is often found on guava. The adults move readily, but the eggs are placed in the stems of plants and the insect could easily have been introduced by this route. M. basalis is quite rare on Guam and may have been present for a number of years prior to being noted.

Micronesica Suppl. 3, 1991

Several armored scales were first noted during the 1980s. Two of these, Furcaspis biformis (Cockerell) (McConnell & Muniappan 1988a) and Genaparlatoria pseudaspidiotus (Lindinger) (McConnell & Muniappan 1988b) are species most commonly found on orchid plants. The former is found primarily in South America but is also known from Hawaii and several other islands of the south Pacific. The latter is tropicopolitan (Williams & Watson 1988). I do not know the origin of the Guam immigrants of these species, but it is almost certain they were imported on orchid plants. Two other species of armored scale were also noted as arrivals in the second half of the decade. Lindingaspis tingi McKenzie was first noted in 1988. Its previous known distribution was the Philippine islands and the islands eastward (Williams 1963, D. Miller pers. comm.). It attacks a variety of plants including orchids, hoya, ferns and cycads. It is not very common on Guam, where it is heavily parasitized. Pseudaonidia trilobitiformis (Green) is another recent arrival. The species originated in southern Asia, but now occurs in some areas of Melanesia as well as in Africa and South America (Williams & Watson 1988). It is not yet known from Hawaii. It attacks a variety of trees, but on Guam it is most damaging to citrus. The citrus types vary in their susceptibility with pummello being most affected on Guam. The scale causes leaf distortion and also attacks the fruit. It is not parasitized on Guam, although Chilocorus nigritus feeds on it. COLEOPTERA

Rhyparida sp. is a small brown chrysomelid which appears identical to a species frequently intercepted in the U.S.A. on aircraft and military cargo originating in East and South-East Asia (R. E. White pers. comm.) This beetle was first collected on Guam in 1985. Its hosts are not known, but adults are found commonly in lawns, and it seems likely the larvae may be root feeders on lawn plants. Popillia lewisii Arrow is a close relative of the Japanese beetle but was found previously only in the Ryukyu islands. It was first observed in 1985 at Anderson Air Force base in an area close to the flight line. A trapping program run for the last five years has not shown it to be spreading and it has been quite rare except in the first year when a spray program with carbaryl was being undertaken to keep the numbers down. That situation may have changed in 1990, however, as trapping which has just resumed after a hiatus of some months shows it to be quite abundant this year. Epitrix hirtipennis (Melsheimer), the tobacco flea beetle, is a North American species which has been present in Hawaii for at least a century (Samuelson 1973). On Guam, it was first identified from specimens collected in 1983. It is a minor pest of eggplants and occasionally young tomato plants may be damaged. Chaetocnema confinis Crotch, the sweet potato flea beetle, was first noticed in 1986 when it was seriously damaging sweet potatoes in a farmer's field, although since that time little damage has been observed. This is also a North American species which had immigrated to Hawaii in 1983 (Lai 1985). The adults make characteristic long narrow feeding tracks on the leaves, and the larvae feed on

Schreiner: New insects in Guam

the roots and surface of the tubers. In the U. S. A. they are considered to be pest of sweet potatoes and are one of a number of species for which resistance is being bred (Chalfant et al. 1990). There appears to be variation in resistance even in local varieties. This flea beetle has since been found in Saipan and in Majuro. LEPIDOPTERA

The noctuid moths of the Micronesian region have not been well studied, but lists of pests and some other species are available for Guam. Several species which appear to be new to the island have been noted recently. These include Helicoperva assulta (Guenee), one of the moths of the corn earworm complex. It was first collected in 1987 both at a blacklight set up in the forest ofNorthwest Field and in a sweet corn field. It is widely distributed in the Old World tropics. The moth is a minor pest compared to H. armigera (Hiibner) which has been present on Guam for many years. Platyja umminia (Cramer) was first noted in 1988 (Denton et al. 1989). In 1989 it had become sufficiently abundant to show up in a student collection. This fruit piercing moth is a South-East Asian species. Another new moth, which is attracted to fruit but which is not a primary piercer, is Parallelia palumba (Quenee), first noted in 1989 both in fruit-baited traps and in a student collection. Hulodes caranea Cramer is a large moth which was first collected in 1986. At that time it was quite numerous at lights. Later in the year it was also found for the first time in a student collection. It is not a pest but, being one of the largest noctuid moths we have on Guam, it is quite conspicuous at the times of the year when it is common. It is known to have been present on Palau for a number of years. DIPTERA

Melanagromyza splendida Frick was only recently identified, though specimens had been collected in 1986. This agromyzid fly is an internal borer in the stem or midribs oflarge leaves (Spencer 1973). Previously, it was known to attack Compositae and Umbelliferae, but on Guam it has been reared from tomato leaves and from midribs of mizuna, an Asian Brassica variety. The native distribution of this species is the tropical and subtropical Americas, but it has been present in Hawaii for many years. HYMENOPTERA

Another relatively recent import is the big headed ant, Pheidole megacephala (F). According to pest control operators it has been present in at least one housing subdivision in northern Guam for a number of years, but it appears to be spreading, having reached the University only in the last year. The big headed ant has a widespread distribution in the tropics. ACARI

Eotetranychus cendanai Rimando is a spider mite known previously from South East Asia. I first noted the species in 1983, but it had probably been on

Micronesica Suppl. 3, 1991

Guam for some years previously. In S. E. Asia it is said to only damage citrus when DDT is used, but on Guam certain varieties of lemons are very susceptible and die back due to the continuing damage to the new leaves and shoots. MOST RECENT INTRODUCTIONS

Halfway through 1990, we already have one possible new introduction for the coming decade. One specimen of Pyropotasia pryeria (Jason), a scarab beetle, has been found on Guam. The beetle arrived on Midway some years ago where it has become enormously abundant on palm blossoms (L. Pinter pers. comm). It is also a horticultural pest. I am not certain this species has been really introduced to Guam, however, because to date the only specimen found was lying dead in a building. Discussion

Information from previous papers (Schreiner & Nafus 1986, Ward 1984) was updated and used to determine the origin of pest species arriving on Guam at various times in the Post World War II period (Table 1). Immediately post war there was considerable movement of men and materials around the Pacific region. Fifteen new insect pests were noted between 1945 and 1954, most ofthem originating in Asia. Thirteen more were introduced in the next 19 years between 1955 and 1970. Most of this time Guam was under U.S. Naval administration, and a very limited amount of travel was taking place between Guam and other parts of the world. Most of the species introduced during this period have tropicopolitan distributions. Two species each clearly originated from Asia and from Hawaii. Six of the species were mosquitos introduced possibly via military flights, and several insects may have been introduced on fresh produce. Around 1970, tourists began coming to Guam and immigration, emigration, and general travel by the local population to and from Guam increased sharply. Of the 21 new species which were first noted in the 1970s, 15 were found between 1970 and 1975. These insects were mostly of Asian origin, and it is highly probable that they were introduced in connection with military operations in Vietnam. Two North American species were also introduced during the decade, one possibly via Hawaii and the other possibly directly from the mainland U.S.A. in produce or ornamentals. Table 1. Source Asia Micronesia Hawaii Unknown Total

Number of insects established on Guam by area of origin.

1945-54

1955-69

1970-79

1980-89

7 1 0 7 15

9 4

8 2 8 7 25

1 2 8 13

6 21

Schreiner: New insects in Guam

In the 1980s at leat 25 new pests or conspicuous species were noted. The available evidence suggests that one-third of them came to Guam from Hawaii. Many were species that had immigrated to Hawaii only one or a few years prior to their introduction to Guam. A number of these species have also been recently found on islands in the South Pacific region (Waterhouse & Norris 1987), and a few have reached South-East Asia. Whereas Hawaii used to be the most isolated island chain in the world, it now appears to be a major transshipment point for introducing the Neotropical fauna to the Old World biota. Hawaii has already been noted as a staging area for the introduction of Old World species into California (Dowell & Gill, 1989). The years immediately following a new introduction, when the insect populations are at a very high level, seem to pose the greatest risk in terms of further emigration of the pest population. Quarantine advisories are only rarely made available to the islands of the Pacific, but it appears these would need to be updated with great frequency to be most useful in helping to prevent the further movement of pests. The Hawaii Pest Report published by the Hawaii Department of Agriculture Plant Pest Control Branch was a very useful early warning beacon for the islands which have flight connections with Hawaii, but unfortunately is no longer published. In addition to Hawaii, Asia and the other islands of Micronesia continue to be important sources of new insects arriving on Guam. Although Guam has obtained direct flight connections to Australia and New Guinea within the last two years, these areas have not yet become obvious as sources of new insects. Guam is not only a recipient of new pest insects, but also an exporter. Guam serves a major role in redistributing insects in Micronesia, as many pests which are first noted on Guam are observed on various islands in the Micronesian region within a few years. On at least one or two occasions Guam has served as a jumping point to transport Asian fauna to Hawaii, as in the case of a bagworm, Brachycyttarus sp., which was first found around the household of a family which had recently moved from Guam to Hawaii. Many of the introductions to Guam in the last decade appear to be associated with the movement of ornamental plants. Half of the introductions were Homoptera, and all but two of these almost certainly immigrated to Guam on host material. Species in other orders may have also traveled by this route. Other species of pests may have traveled in flowers or in potting mixture. Unfortunately there is not a big local industry to produce material for the commercial plant nurseries on Guam, so these import large volumes ofliving material on a frequent basis, primarily from Hawaii. Many local residents also buy small quantities of ornamentals while traveling. In 1990 we had a case where the origin of a plant-infesting insect could be pinpointed with some accuracy. A nursery owner noted serious problems with hibiscus imported from Hawaii. The Guam quarantine service failed to inspect the particular box of hibiscus plants in which the whiteflies traveled, and the whitefly adults were noted by the nursery owner when the box was opened. The insects spread to other hibiscus plants in the nursery, and numerous hibiscus

Micronesica Suppl. 3, 1991

plants had already been sold before the owner realized that the whiteflies were damaging the plants. Fortunately this whitefly proved to be Bemisia tabaci (Genn.), a species which was already present on Guam. It arrived only recently in Hawaii and is currently at very high levels there. Although Guam quarantine personnel in recent years have become better at detecting insects on plant material from the USA, it is clear that they are still missing some. Only in the last year have they started inspecting material which arrives via first class mail, and many of the ornamentals are shipped that way. They do not check all the material that arrives, even that which is clearly marked as being plant material, but in any case, the best inspections could not expect to discover all scale insects or insect eggs in large boxes of potted plants. Because there is not a large agriculture industry on Guam, the farm lobby is not a sufficiently potent force to insist that post-entry quarantine be performed on all living plant material, even though many of the pests introduced on ornamentals have wide host ranges and become crop pests after their arrival. If the number of new pest introductions is to be reduced, post-entry quarantine of all plants, in a compartmentalized facility where insect movement is limited, would appear to be necessary. The moths which arrived during the last decade, several of the beetle species, and perhaps a couple of the Homoptera species most likely traveled to Guam in the hold or passenger compartments of airplanes and were probably not associated with plant material in flight. The scheduling of flights to Guam is such that many travel during the night, leaving and arriving at times when lights are needed. This provides a good opportunity for attracting many species of insects to the holds. Residual insecticide sprays in the holds could do much to decrease this problem, without resulting in any of the liability problems that spraying the passenger compartment might cause. Acknowledgement I thank J. Beardsley of the University of Hawaii, K. Kevan of McGill University, L. Printer of U. S. Navy, J. Medler and G. A. Samuelson of the B. P. Bishop Museum, S. Nakahara, R. E. Poole, R. Smiley, and T. E. White of the USDA Systematic Entomology Laboratory and other taxonomists who have identified Guam material for us over the years. References Beardsley, J. W., Jr. 1966. Homoptera: Coccoidea. Insects of Micronesia. 6(7): 376-562. Beardsley, J. W., Jr. 1986. Notes and Exhibitions. Proc. Hawaii. Entomol. Soc. 26: 9. Beardsley, J. W., Jr. 1979. New immigrant insects to Hawaii: 1962 through 1976. Proc. Hawaii. Entomol. Soc. 23: 35-44. Chalfant, R. B., R. K. Jansson, D. R. Seal & J. M. Schalk. 1990. Ecology and management of sweet potato insects. Annu. Rev. Entomol. 35: 157-180.

Schreiner: New insects in Guam

Denton, G. R. W., R. Muniappan, M. Marutani, J. McConnell & T. S. Lali. 1989. Biology and natural enemies of the fruit piercing moth, Othreis fullonia (Lepidoptera: Noctuidae) from Guam. Proc. ADAP Plant Protection Conf., Univ. of Hawaii. In press. Dowell, R. V. & R. Gill. 1989. Exotic invertebrates and their effect on California. PanPacific Entomol. 65: 132-145. Essig, E. 0. 1956. Homoptera: Aphididae. Insects of Micronesia. 6(2): 15-37. Gressitt, J. L. 1954. Introduction. Insects of Micronesia. Vol. 1. 257 pp. Lai, P. Y. 1985. Notes and Exhibitions. Proc. Hawaii. Entomol. Soc. 25: 17. McConnell, J. & R. Muniappan. 1988a. Red orchid scale, Furcaspis biformis (Cockerell), Diaspididae. Guam Pest Series. Cooperative Extension Service, Univ. of Guam. 1 p. McConnell, J. & R. Muniappan. 1988b. Yanda orchid scale, Genaparlatoria pseudaspidiotus (Lindinger), Diaspididae. Guam Pest Series. Cooperative Extension, Univ. of Guam. 1 p. Nakahara, L. M. & P. Y. Lai. 1984. Hawaii Department of Agriculture Plant Pest Control Branch. Hawaii Pest Report 4(2): 2-8. Nakahara, L. M., K. Sakimura & R. A. Heu. 1986. Notes and Exhibitions. Proc. Hawaii. Entomol. Soc. 26: 10. Paulson, G. S. 1983. The biology and natural enemies of the wooly whitefly, Aleurothrixus floccosus (Maskell) in Hawaii. M.S. Thesis. University of Hawaii, 71 pp. Samuelson, G. A. 1973. Alticinae ofOceania (Coleoptera: Chrysomelidae). Pacific Insect Monograph 30. Shiroma, E. 1968. Notes and Exhibitions. Proc. Hawaii. Entomol. Soc. 20: 264. Schreiner, I. & D. Nafus. Accidental introductions of insect pests to Guam, 19451985. Proc. Hawaii. Entomol. Soc. 27: 45-52. Spencer, K. A. 1973. Agromyzidae (Diptera) of economic importance. W. Junk. The Hague. 418 pp. Ward, R. A. 1984. Mosquito fauna ofGuam: Case history of an introduced fauna. In M. Laird (ed.), Commerce and the spread of pests and disease vectors, pp. 143-161. Praeger, New York. Waterhouse, D. F. & K. R. Norris. 1987. Biological control: Pacific prospects. Inkata Press Pty Ltd. Melbourne. 454 pp. Williams, D. J. 1963. Synoptic revisions of I. Lindingaspis and II. Andaspis with two new allied genera (Hemiptera: Coccoidea). The Bulletin of the British Museum. Entomological Series 15( 1). Williams, D. J. & G. W. Watson. 1988. The scale insects of the tropical South Pacific region. Part 1. The armoured scales (Diaspididae ). C.A.B. International Institute of Entomology.

Micronesica Suppl. 3: 15-31, 1991

Movement of New Insects into the Carolines and Marshalls in Recent Years DONALD NAFUS

College of Agriculture and Life Sciences Agricultural Experiment Station, Mangilao, Guam 96923

Abstract-In 1986, surveys of crop insects were conducted in Palau, Yap, Chuuk, Pohnpei, Kosrae, and in 1989, in Jaluit and Majuro. In the Marshalls, 20 species of insects were found to have entered Majuro or Jaluit for the first time. Of these, ten species were new to the Marshalls, and the rest were already present on other atolls. Using 1975 as the base year, Majuro acquired 16 new records, corresponding to a rate of 1.1 new pests per year, and Jaluit had nine new records, or 0.6 new insects per year. These rates are lower than those for Guam, which got 2.5 new introductions per year during the 1980s. In the Carolines, between 1976 and 1986, Pohnpei acquired about 1.4 new insects per year, Yap 1.1, Palau 0.7, Kosrae 0.7, and Chuuk 0.6. Many ofthe new records in the Carolines were range extensions of insects already present. In both the Carolines and the Marshalls, scales, whiteflies, and other Homoptera accounted for most of the new introductions, but new species of moths, beetles, flies, and thrips were represented as well. The movement of new insects into islands is a serious problem. In Hawaii between 20 and 30 new immigrants arrive every year (Beardsley 1979), and Guam may receive as many as 12-15 (Schreiner & Nafus 1986). In recent years new introductions to the Marshalls and Carolines have not been documented, particularly in the atolls, many of which have not been visited by an entomologist since the 1950s. In the early part of the 20th century, collecting in Micronesia was sporadic, but between 1936 and 1955 an intensive effort was made to determine the insect fauna. In 1936-40, the Japanese entomologist Esaki (1940, 1952) collected economically important insects and recorded many new pests. In the late 40s and early 50s, the United States Navy sponsored a survey of the invertebrate fauna. This and Esaki's surveys provided a baseline against which changes in the insect fauna of Micronesia can be measured. Unfortunately, the systematics of many groups has never been finished and published, so several groups remain unknown. During the 1950s through the late 1970s, the United States Trust Territory government maintained an entomology office, which was responsible for pest control and which kept track of changes in the fauna. In the 1980s, monitoring was sporadic and largely up to individual countries. In 1986, under the auspices of the South Pacific Commission, I surveyed the insects of important crops in Palau, Yap, Chuuk, Pohnpei, and Kosrae. On

Micronesica Suppl. 3, 1991

each of these islands I spent five to ten days visiting selected crops with local extension agents, and collecting associated insects. In 1989, I visted Jaluit and Majuro in the Marshalls on June 20-30, 1989. During this period, I also spent three hours on Likiep. All of these visits were too short to conduct thorough surveys, but I did find a large number of new records and am reporting rates of movements of crop pests into these islands based on these records. These rates are very conservative, and I am certain that there is much that I have not accounted for. In particular, despite the large number of scales found, there are undoubtedly many more, as I only sampled a few, agriculturally important hosts. Most of my sampling was restricted to the plants that the state or country wanted me to look at. From Esaki's work, the Naval Survey, and other activities, the distribution of various economic insects on a variety of crops was determined (Bryan 1949, Pemberton 1954, Gressitt 1954). Following this through the late 1970s, various entomologists including J. Beardsley, R. Owen, J. Tenorio, 0. Demei continued to catalogue pests on selected crops in the various islands in Micronesia, and several systematists published papers on their identification and distributions. A list of these insects, presumably compiled by 0. Demei, was kept by The Office of the Chief Entomologist of the Trust Territories. It is dated 1973, but a few entries were made after that time. For the purposes of this paper, I am assuming that any insects not reported in the published literature, not listed by the Trust Territory Entomologists, or included in the collection of the Bishop Museum in Honolulu are new introductions. This may not always be correct as some species may have been present earlier but not collected, or were misidentified or otherwise overlooked. In addition, I have not been able to check all groups in the Bishop Museum in Honolulu, so I may have improperly included some species as new records when they are not. I apologize for any of these unintentional errors. In have somewhat arbitrarily chosen 1975 as a base year for my calculations of the number of new species entering the various islands each year. 1975 corresponds to the year of publication of several key supplements to the Insects of Micronesia series, and is close to the last date entered on the copy of the Trust Territory list graciously provided to me by H. Adelbai. For convenience, I have organized many of the new records around the main host plants. All new records are given in Table 1, along with their previous distribution records. Coconut

One of the first new insect pests to be recorded in Micronesia was the coconut scale Aspidiotus destructor Signoret (Esaki 1940, 1952). This scale was first introduced to Yap from the Philippines in 1892. From there it spread to Palau in 1898 or 1899, and the Mariana Islands ofSaipan in 1910, Guam 1911 (Vandenberg 1926), and later Rota and Tinian (Esaki 1952). Outbreaks in Yap in 19041907 and Saipan 1914-21, killed 70-80 percent of the coconuts and led to extensive biological control efforts (Nafus & Schreiner 1989).

Table 1.

Movements of insects in the Carolines and Marshalls. Symbols are Pa (Palau or Belau), Y (Yap), C (Chuuk=Truk), Po (Pohnpei= Ponape), K (Kosrae=Kusiae), M (Majuro), J (Jaluit), L (Likiep), x (present before 1975), n (new since 1975).

Scientific Name Homoptera Aleurocanthus spiniferus (Quaintance) (Homoptera: Aleyrodidae) Aleurodicus dispersus Russell (Homoptera: Aleyrodidae) Aspidiotus destructor Signoret (Homoptera: Diaspididae) Asterolecanium sp. (Homoptera: Asterolecaniidae) Ceroplastes rubens Maskell (Homoptera: Coccidae) Coccus hesperidum L. (Homoptera: Coccidae) Coccus viridis (Green) (Homoptera: Coccidae) Dialeurodes citrifolii (Morgan) (Homoptera: Aleyrodidae) Empoasca spp. (Homoptera: Cicadellidae) Eucalymnatus tessellatus (Signoret) (Homoptera: Coccidae) Ferrisia virgata (Cockerell) (Homoptera: Pseudooccidae) H eteropsylla cubana (Homoptera: Psyllidae) Idiocerini (unidentified species) (Homoptera: Cicadellidae) Lepidosaphes beckii (Newman) (Homoptera: Diaspididae) Lepidosaphes gloverii (Packard) (Homoptera: Diaspididae)

Common Name

orange spiny whitefly spiralling whitefly

coconut scale

bamboo soft scale

red wax scale

References

Takahashi 1956, Nafus 1988 Schreiner & Nafus 1986 Schreiner & Nafus 1986

Beardsley 1966

~ ('D

brown soft scale

green scale

X X

Beardsley 1966, 1975 Beardsley 1966 Beardsley 1966

Cll

('D

$l Cll

s路

s('D

() ~

cloudywinged whitefly green leafhopper

"'1

s路 n

tessellated scale

striped mealybug

leucaena psyllid

leafhopper

purple scale

glover scale

X X

Linnavuori 1960, 197 5

X X

Beardsley 1966

i:l 0..

;. "'1

Beardsley 1966 Beardsley 1966

::::;

Table 1. Continued. Scientific Name Homoptera Lepidosaphes laterochitinosa Green (Homoptera: Diaspididae) Parlatoria zizyphus (Lucas) (Homoptera: Diaspididae) Pentalonia nigronervosa Coquerel (Homoptera: Aphididae) Phenacoccus madeirensis Green (Homoptera: Pseudococcidae) Pinnaspis strachani (Cooley) (Homoptera: Diaspididae) Planococcus sp. (Homoptera: Pseudococcidae) Proutista moesta (Westwood) (Homoptera: Derbidae) Pseudaulacaspis pentagona (T&T) (Homoptera: Diaspididae) Pulvinaria psidii Maskell (Homoptera: Coccidae) Pulvinaria urbicola Cockerell (Homoptera: Coccidae) Saissetia neglecta DeLotto (Homoptera: Coccidae) Vinsonia stellifera Westwood (Homoptera: Coccidae) Hemiptera Brachyplatys insularis Ruckes or B. pacificus Dallas (Hemiptera: Plataspidae) Coptosoma xanthogramma (White) (Hemiptera: Plataspidae) Physomerus grossipes (F.) (Hemiptera: Coreidae) Piezodorus hybneri (Gmelin) (Hemiptera: Pentatomidae)

armored scale

black parlatoria scale

banana aphid

Common Name

mealybug lesser snow scale

References Beardsley 1966

Beardsley 1966

Beardsley 1966 Beardsley 1966, 1975

Fennah 1956 n

;;

Essig 1956

citrus mealybug erect-winged blue planthopper white peach scale

Beardsley 1966

() " '"1

::s

&\l

()" s:.>

C/J ~

green shield scale

urbicola soft scale

Caribbean black scale

stellate scale

black island stink bug

Beardsley 1966

black stink bug

Beardsley 1966, 1975

Beardsley 1966

Ruckes 1963, O.C.E.T.T. 1

large spined-footed bug

shield bug

Schreiner & Nafus 1986 n

Ruckes 1963

::::

Table 1. Scientific Name Coleoptera Chaetocnema confinis Crotch (Coleoptera: Chrysomelidae) Cylas formicarius (Coleoptera: Curculionidae) Metriona circumdata (Herbst) (Coleoptera: Chrysomelidae) Lepidoptera Agonoxena pyrogramma Meyrick (Lepidoptera: Agonoxenidae) Badamia exclamationis F. (Lepidoptera: Hesperiidae) ?Brachycyttarus sp. (Lepidoptera: Psychidae) Chrysodeixis chalcites (Esper) (Lepidoptera: N octuidae) Lamprosema diemenalis (Guenee) (Lepidoptera: Pyralidae) Maruca testulalis (Geyer) (Lepidoptera: Pyralidae) Othreis fullonia (Clerck) (Lepidoptera: N octuidae) Penicillaria jocosatrix (Guenee) (Lepidoptera: Noctuidae) Plutelfa xylostella (L.) (Lepidoptera: Yponomeutidae)

Common Name

Continued. Pa

sweetpotato flea beetle

References

sweetpotato weevil

green tortoise beetle

O.C.E.T.T. 1

Gressitt 1955 O.C.E.T.T. 1

z PJ

z (!)

s rJl

coconut flat moth migratory skipper

bagworm green garden looper

Clark 1984

Esaki 1940, Gressitt 1954

bean leaf-roller

$4.

rJl

;. (!)

()

(!)

Gressitt 1954, Esguerra 1990 Clarke 1976

...,PJ g. 5. ~

::s

0..

bean pod borer fruit piercing moth

n? n

mango shoot caterpillar diamondback moth

;;!

::r ~ rJl

Gressitt 1954, Denton et al. 1990

::;;

Table 1. Continued. Scientific Name Thysanoptera Selenothrips rubrocinctus (Giard) (Thysanoptera: Thripidae) Thrips palmi Karny (Thysanoptera: Thripidae) Diptera Liriomyza trifolii (Burgess) (Diptera: Agromyzidae)

Common Name

References

(=i ...., "

redbanded thrips

::s

(=i"

melon thrips

Pol

r./J

serpentine leafminer

Schreiner & Nafus 1986

1 The Office of the Chief Entomologist of the Trust Territories prepared a list dated 1973 which reports the pests of various crops on the islands groups of Chuuk, Palau, Yap, Pohnpei, Kosrae and the Marshalls.

::::

Nafus: New Insects in the Carolines and Marshalls

The scale appeared in Pohnpei prior to 1938, probably from a separate introduction (Esaki 1940). It also appeared in Woleai Atoll sometime before 1940, the high islands in Chuuk (formerly Truk) by 1946 (Beardsley 1966), and the Mortlocks prior to 1960. The coconut scale has continued to spread, arriving in Kapingi-Marangi in 1985, and Majuro in the Marshalls around 1987. From Majuro it is being redistributed to other atolls in the Marshalls including Likiep in 1988, and possibly Wolei in 1989. The infestation on these atolls was severe. On Majuro in June, 1989, the coconut scale was distributed throughout the main atoll island except for a few isolated points adjacent to the beaches. At Laura, the undersides of breadfruit leaves were often completely covered with coconut scale. Estimated populations of scales averaged about 50,000 scales per leaf. Many trees were in poor health and had dropped leaves, and some trees had died. In between the airport and Laura, populations were lower, averaging around 30,000 scales per leaf. This dropped to 2, 700 in areas where the coccinellid Chilocoris nigritus was well established. C. nigritus, along with a second lady beetle, had been introduced earlier by Dr. Nelson Esguerra of the College of Micronesia in Pohnpei. A second lady beetle, possibly a species of Pharellus, was present and widely distributed, but it was unclear if it was having much impact on scale populations, or where it originated from. On one tree, the lady beetle averaged 5.8 beetles per leaf, but scale populations were still over 34,000 per leaf. In all areas parasitoid larvae and pupae were found inside the coconut scale. Parasitization rates were low, however, averaging less than two percent. Infestations on coconut were generally less severe than on breadfruit. Populations ranged from over 2,000 scales per leaflet to slightly over 300 scales per leaflet. Both Chilocorus nigritus and Pharellus were present on coconut, and I suspect that C. nigritus prefers coconut to breadfruit. Again it was most abundant where scale populations were highest. On coconut the levels of parasitization were higher. An average of approximately 13 percent of the scales were parasitized and individual samples ranged up to 26 percent. In addition, a predatory mite was found under the scale cover. I was unable to rear the parasitoids during the visit, so I cannot provide identification. I suspect the parasitoids are contributing to the lower incidence of scales on coconut compared to the breadfruit, although host preferences of the scale or other factors may also be contributing. Another coconut pest accidently introduced to Micronesia is the coconut rhinoceros beetle, Oryctes rhinoceros L., which was first reported in Palau in 1942 (Bryan 1949). It was thought to have come from Indonesia on Japanese shipping. The beetle caused serious damage to the coconuts, particularly right after WWII when there was abundant breeding material present. Extensive efforts were made to remove dead coconuts, and a biological control program was initiated (Gardner 1958, Schreiner 1989). The weevils Rhabdoscelus asperipennis (Fairmaire) and R. obscurus (Boisduval) are both present in Palau, but when they entered and how widely they

Micronesica Suppl. 3, 1991

are distributed in the Carolines is problematic. These species are closely related and difficult to distinguish. R. obscurus was first found on Guam in 1911 and later spread to other islands in the Marianas. Gressitt (1954) indicated that it was widespread but did not list any localities outside the Marianas. Julia (1983) states that R. asperipennis was widespread in the Carolines, but the 1973 Trust Territory list reported it only as on Palau. According to Julia, a thorough search in 1964 failed to uncover R. obscurus in Palau, whereas R. asperipennis was abundant and widespread. In 1983, R. asperipennis was uncommon and R. obscurus was found. Since the T.T.list records both species in Palau, I am assuming that R. obscurus was introduced sometime between 1966 and 1973, and I am not including it in the recent introductions, although problems with identification of these two species may invalidate this assumption. Agonoxena pyrogramma Meyrick was reported from Guam by Fullaway (1912), but was not listed in Micronesia by Bryan (1949) or Pemberton (1954). Clarke (1984) looked for the moth in Micronesia in 1953, but only found it on Kosrae on nipa palm. In 1986, I reared it from coconut leaves on Pohnpei and found larvae which appeared to be this moth on Chuuk. In both cases the moths were abundant and widespread on the islands. It is possible they may have been overlooked in earlier surveys, but if it is as consistently conspicuous as on Guam, I doubt it. I suspect they are relatively new introductions. Hemiberlesia palmae (Cockerell) was found on Chuuk. This scale was reported by Esaki ( 1940), but listed as questionable by Beardsley ( 1966) as specimens were not available for verification. D. Williams verified the record for Chuuk. Citrus

Several pests, primarily scales, mealybugs, and whiteflies, associated with citrus have been moving about the Carolines and Marshalls. The orange spiny whitefly, Aleurocanthus spiniferus (Quaintance), a major pest of citrus in subtropical and tropical regions (Nakao & Funasaki 1979), has recently extended its range to several islands in Micronesia. It is native to Indomalay region, south China, and the Philippines. In 1919 it was found in southern Japan (Clausen 1978), and subsequently spread to the Chuuk and Guam by the late 1940s and early 1950s (Peterson 1955). Sometime before 1982, it became a major pest of lime and orange on Kosrae (Nafus 1988), Pohnpei (Schreiner & Nafus 1986), and Yap in the Caroline Islands. Another whitefly, Dialeurodes citrifolii (Morgan), the cloudywinged whitefly, has recently entered Micronesia, probably from Hawaii, where it has been established since 1966 (Paulson and Kumashiro 1985). I found it on the undersides of citrus leaves around Kolonia, Pohnpei, in 1986. It was not present elsewhere on Pohnpei or in Micronesia. The whitefly was intermixed with populations of the orange spiny whitefly and several species of scales. Because it is transparent, it is easily overlooked and could easily be moved on plant material from island

Nafus: New Insects in the Carolines and Marshalls

to island. On Palau, I did find a few nymphs of a similar whitefly on one lime tree outside Koror, but I could not identify this species as no pupae were found. Three scales, Lepidosaphes gloverii (Packard), Parlatoria zizyphus (Lucas), and Ceroplastes rubens Maskell, and one mealybug Planococcus sp. have extended their Micronesian ranges. Beardsley ( 1966) reported L. gloverii from Chuuk, Pohnpei, and the Marianas. I found it in Yap as well. The black parlatoria scale was previously found only on Pohnpei, but I now report it from Chuuk as well. On Pohnpei it is abundant on various types of citrus in the Kolonia area. On Chuuk it was infesting two citrus trees on Moen near the Governor's house. Beardsley ( 1966, 197 5) originally listed the pink wax scale from the Marianas, Palau, and Kwajalein. I found it on citrus in northern Yap and on breadfruit on Majuro and Likiep. It was parasitized, but I did not have time to rear any of the species. Planococcus citri (Risso) was reported as widely distributed in Micronesia by Beardsley (1966), but recent studies (Cox 1981) have shown this is a species complex rather than a single species. In most cases, P. pacificus Cox is likely to be the species of mealybug present (Williams & Watson 1988), although P. citri is known to be on Guam. In the Marshalls, a species of Planococcus (as citri) was reported from Kwajalein on Aralia leaves. I found a Planococcus species on citrus on Jaluit. It appeared to be pacificus, but this needs verification. Phyllocnistis citrella Stainton, the citrus leaf miner, was found in Guam and Saipan about 1927 (Gressitt 1954). Bryan (1949) recorded it from Palau but not elsewhere in the Carolines or Marshalls. In 1986, I found it on lime in northern Yap. It was abundant and causing partial defoliation. Othreis fullonia (Clerk) is also an important pest of citrus in Micronesia. It has been in the Marianas for a long time, and has an endemic parasitoid fauna associated with the eggs. In the Carolines and Marshalls, it was either uncommon or absent. Gressitt (1954) reported it from Palau, where it was evidently uncommon, but it was not reported elsewhere in Micronesia until relatively recently. In Kosrae, Muniappan found it damaging citrus in 1982, and he and his coworkers recently found it in Pohnpei as well (Denton et al. 1990). These probably represent relatively new introductions as native Telenomus egg parasites are not known to be associated with the moth on these islands. Sweet Potato

Several insects attacking sweet potato have extended their distributions within or entered the region. Many of the insects attacking sweet potato also eat other species of Ipomoea including beach morning glory, and various weedy species. These alternate hosts are common around airports and often residences on most islands. The sweetpotato weevil, Cylas formicarius (F.), is a widespread pest in Micronesia but was not known from the Marshall Islands. On Majuro I found it infesting beach morning glory in Rita. Metriona circumdata (Herbst), the green tortoise beetle, was also present along with the sweetpotato flea beetle Chaetocnema confinis Crotch. The green tortoise beetle is an Asian species, which feeds on the leaves of sweet potato in

Micronesica Suppl. 3, 1991

both the adult and larval stages. It is thought to have entered Guam from the Philippines in 1945 (Gressitt 1955), and from there it spread to Chuuk and Palau soon after. I found it in Yap in 1984, Pohnpei in 1986, and Majuro in the Marshalls in 1989. I did not see it in Jaluit or Likiep. It is obviously moving eastwards towards Hawaii and the U.S. The sweet potato flea beetle has been moving in the other direction, outwards from Hawaii. I found it in Majuro, but not Jaluit or Likiep. It was found on Guam in 1986, and by 1988-9 was present on Saipan, Tinian and Rota. The adults feed on the leaves, chewing tunnel-like tracks, and the larvae tunnel in the roots. It is a serious pest of sweet potatoes in the southern United States. Physomerus grossipes (F.), the large spined-footed bug, was first found in Guam in the 1960s (Schreiner & Nafus 1986). In 1986, I collected it in Palau. This bug feeds on the vines of sweet potato and morning glory. It can occur in great numbers on vines growing up trellises or other supports, but does not occur to any great extent on vines on the ground. Thus, it is not a serious problem. Beans

On beans several insects had new distribution records. Brachyplatys insularis Ruckes is a species endemic to the Marianas islands, and neither it nor any other species of Brachyplatys were found elsewhere at the time of the Naval faunal survey (Ruckes 1963). Between the mid 1950s and 1973, a Brachyplatys species listed as B. pacificus on the 1973 T.T. list appeared in Palau and Chuuk. I found aBrachyplatys species on all ofthe high islands ofMicronesia and in the Marshalls in Jaluit and Majuro. B. paci.ficus and B. insularis are closely related and difficult to distinguish. Ruckes felt the Marianas records of B. paci.ficus were probably B. insularis, and it is possible that the other Micronesian records are all one species. T. J. Henry examined specimens of Brachyplatys from the Marianas, Kosrae, Pohnpei, Yap, and Palau. He felt these were all one species, but tentatively identified them as B. subaenus (Westwood). He is re-examining the identification of the species, and a final determination should be available shortly. Another platispid which recently entered Micronesia and is extending its range is Coptosoma xanthogramma (White). It is frequently found on both long beans (Vigna sp.) and pole beans (Phaseolus sp.), although it is not abundant. It was first recorded from the Marianas in 1968, and by 1984, was present in Kosrae. I suspect it is much more widely distributed. Piezodorus hybneri (Gmelin), originally reported from Palau and the Marianas (Guam, Saipan), was also found in Yap, and Bjork collected it from Tinian in the Marianas in 1985. The green garden looper Chrysodeixis chalcites (Esper) may also be extending its distribution. Gressitt (1954) originally listed it from the Marianas and Palau, but Esguerra ( 1990) also indicates that it is now in Pohnpei. This species has a very wide host range. It is common on various cucurbits in addition to beans. Other new records of species on beans includes Empoasca sp. in the Marshalls, Maruca testulalis (Geyer) in Palau, Liriomyza trifolii (Burgess) in Yap and Pohnpei, and Lamprosema diemenalis (Gue-

Nafus: New Insects in the Carolines and Marshalls

nee) on Pohnpei, Jaluit, and Majuro. L. diemenalis was defoliating soybeans on Majuro and Pohnpei, but was rarely present on other species of beans.

Mango Coccus viridis (Green), Penicillaria jocosatrix Guenee, Selenothrips rubrocinctus (Giard), Vinsonia stellifera Westwood all showed range extensions within or into Micronesia. V. stellifera was extremely abundant on mango in Kosrae, but only on certain trees. In some cases, on trees growing next to each other, one tree would have several scales per leaf and the other tree none. On Palau an unidentified species of leafhopper in the subfamily Idocerini was abundant on mango at the Agricultural Experiment Station. This species is not listed by Linnavuori ( 1960, 197 5), and is assumed to be a new introduction although it is still unidentified.

Banana Cosmopolites sordidus (Germar) was common in Guam in 1939, but Esaki ( 1940) did not find it in any of the other islands at that time. Since then it has spread to Palau, Yap, Chuuk, and Pohnpei. On Pohnpei it is localized. I found extensive damage to one banana plantation. Pentalonia nigronervosa Coquerel has been widely distributed in the Carolines for a long time, and in the Marshalls on Lib and Arno, but it was not reported from Kosrae or Majuro (Essig 1956). I found it in low numbers in both locations. In neither case were numbers high enough to be damaging, but the main threat associated with this aphid is that it is a vector of bunchy top. Bunch top is a fatal viral disease which is not present in either the Carolines or the Marshalls, but is present in the Marianas, Hawaii and Philippines. Because of the presence of this aphid and the current cultural practices, this disease would spread rapidly and be seriously damaging in both island groups. Other Hosts

The spiralling whitefly, Aleurodicus dispersus Russell, originated in Central America, and was accidentally introduced into Hawaii in the late 1970s. Since then, it has been spreading westward into the Pacific and southeast Asia. In 1981 the whitefly reached Guam and became a problem on several species of plants. By 1985 it was established in Palau and Pohnpei. It also established in the Marshalls in Kwajalein and Majuro. In all of these areas, natural enemies have been deliberately or accidentally introduced, and have at least partially controlled the whitefly. Several other Homoptera, including Lepidosaphes laterochitinosa Green on Guava, Pulvinaria urbicola Cockerell on peppers, breadfruit, and plumeria, Phenacoccus madeirensis Green on tomato, Empoasca sp. on eggplant, and Asterolecanium bambusae (Boisduval) on bamboo have new distribution records (Table 1). Proutista moesta (Westwood) was collected on Pohnpei. This species

Micronesica Suppl. 3, 1991

was reported on Palau and in the Marianas previously (Fennah 1956, Gressitt 1954). In Palau, I found populations of Thrips palmi Karny, Badamia exclamationis F. and Plutella xylostella (L.). B. exclamationis probably has been there for many years and is not a new introduction, as Esaki (1940) says it is widespread in Micronesia, but it was not recorded on the 1973 Trust Territory list of pests on Terminalia on Palau. I also found it on Majuro. Gressitt (1954) indicates that it was present in the Marshalls on Likiep, but does not report it from other atolls. I suspect this is a new record for Majuro, but it is natural dispersal. This butterfly is a strong flier and is noted for its migratory habits (Common & Waterhouse 1982). Taxonomic Affinity of Species Moving The majority of new records were Homoptera, principally scales and mealybugs and whiteflies (Table 2). These groups are sedentary, and would not be easily recognized as an insect by most Micronesians. They could also be easily overlooked by quarantine officials if they have not had sufficient training in entomology, a common problem in Micronesia. For the more active Homoptera, there were far fewer new distribution records. Only six new distributions, two for the aphid Pentalonia nigronervosa, one for the psyllid H eteropsylla cubana, two for species of Empoasca, one for P. moesta, and one for an unidentified species of leafhopper were found in these groups. The second most common group was the Lepidoptera with a total of 15 new records involving nine species. Most of these are medium-sized to small nocturnal moths. I suspect movement has been on aircraft, except for 0. fullonia, P. xylostella, and B. exclamationis. 0. fullonia and B. exclamationis are strong fliers and have probably dispersed naturally. P. xylostella probably entered through commerce or was brought in on illegally imported plants from Asia. In Micronesia, aircraft are no longer treated to prevent insect movements. I am certain this has enhanced the movements of several species in recent times. Hemiptera, Coleoptera, Thysanoptera, and Diptera had fewer new records. In the Hemiptera, two species extended their ranges and two other species were recorded from the Carolines for the first time. P. grossipes is an Asian species which has recently been recorded from Guam as well. It may have entered Palau Table 2.

Number of new records in the Carolines and Marshalls by order of insect.

Order Homoptera Lepidoptera Coleoptera Hemiptera Thysanoptera Diptera

Marshalls

Carolines

Total

23 12 3

42 18

4 2

0 0 0

6 6

Nafus: New Insects in the Carolines and Marshalls

from either the Philippines or from Guam. The other species all attack legumes, which are common plants in beach or disturbed areas, including near airport runways. Three species of Coleoptera had new distribution records. One ofthese, C. confinis, is new to the region. It is an American species which established in Hawaii in 1983 (Lai 1985) and has recently spread to Guam. I suspect that it is present in the Carolines as well, but it entered the region after my survey in 1986. The other two species have long been established in Micronesia and are gradually extending their range. In all three cases, the beetles feed on hosts which are common near beaches or disturbed areas and are present near airport runways. I suspect they are moving in aircraft cargo holds. Only two species of thrips and one of Diptera have extended their ranges. I suspect these moved on plant material, either imported foods or flowers. Thrips palmi is present in orchid flowers (Waterhouse & Norris 1987) and probably moved that way. Rates of Movement

Of the 47 species with new distributions, 16 of these are new to both the Carolines and the Marshalls. Within the Carolines I found 48 new distribution records involving 33 species. I am sure this is well under the amount that actually took place. Of these 33 species, 14 were new to the Carolines, and the rest were range extensions of insects already present. Pohnpei had the largest number of new insects, acquiring about 1.4 new insects per year on agriculturally important crops (Table 3). Yap had fewer new introductions, averaging around one species per year, and Chuuk, Palau, and Kosrae had the fewest. In Majuro and Jaluit in the Marshalls, we see a similar picture. In the Marshalls there were 25 new distribution records involving 20 species. Half of these were new to the Marshalls, and the remainder already present within the Marshalls on other atolls. Majuro had 16 new records, of which nine were new to the Marshalls and seven were species present on at least one other atoll. Jaluit had nine new insects, four of which were new to the Marshalls, but three of these were also found in Majuro. Thus, of the nine, only one was not recorded elsewhere Table 3. Number and rate of new introductions of insects into the Caroline and Marshall Islands after 1975. Rates are based on surveys in 1986 in the Carolines and 1989 in the Marshalls. Location Pohnpei (Carolines) Majuro (Marshalls) Yap (Carolines) Palau (Carolines) Kosrae (Carolines) Chuuk (Carolines) Jaluit (Marshalls)

Number of new introductions 15 16 11

8 8 7

Number per Year 1.4 1.1 1.0 0.7 0.7 0.6 0.6

Micronesica Suppl. 3, 1991

in the Marshalls. The other eight were present in either Majuro, Kwajalein or both. Majuro acquired about 1.1 insect pests per year, and Jaluit had 0.6 per year. Possible Origins

Air traffic in Micronesia has grown steadily since 1950, and is probably the most important link in the movement of new pests into the region. Kwajalein and Majuro have air connections from Hawaii, Guam, and other Pacific destinations such as Nauru. Majuro serves as the hub for connections with other Marshallese atolls. Jaluit has air connections only within the Marshalls. All of the high islands in the Carolines have jet aircraft landing daily, although Kosrae was only served by small planes until 1986. In the eastern Carolines, air traffic island hops between Hawaii and Guam with stops in Majuro, Kwajalein, Kosrae, Pohnpei, and Chuuk. In the western Carolines, flights originate and terminate in Guam and the Philippines, with stops in Palau and Yap. There are no direct flights between the eastern and western Carolines, although passenger traffic between them without quarantine inspection in Guam is possible. Without detailed studies, it is not possible to determine origin of the majority of the new introductions with certainty, so the following comments are speculative and largely based on prior distributions. I am discussing the Marshalls, the eastern Carolines, and the western Carolines as separate groups. In part this is done because of the air traffic patterns. Recent distribution records suggest Kwajalein and Majuro are the focal point for most new introductions in the Marshalls. From there, such as in the case of the coconut scale, they spread to other atolls. Judging from the case in Jaluit, this is the pattern for the majority of the insects, although a few insects such as Eucalymnatus tessellatus (Signoret) have made it to atolls without direct air service to points outside the region. Within the Marshalls, there is no quarantine or aircraft disinsection. In the eastern Carolines the general pattern of movement is influx from Guam or Hawaii, followed by inter-island redistribution. Most introductions seem to be on plant material moved between islands, as the majority of new introductions are scales and mealybugs. Quarantine officials from the islands indicated that they knew there was considerable smuggling of plant materials. At least one-third of the new introductions are from Guam or Hawaii. The remainder are range expansions of prior introductions within the Carolines, or additional re-introductions from Guam or Hawaii. In the western Carolines, the flow of insects is from Guam and the Philippines. Palau is probably receiving about 25 to 40% of its new insects from the Philippines or other Asian points, and the remainder from Guam or other Caroline Islands. After Hawaii and Guam, Palau is the most important entry point into Micronesia for new species. Its diversity of agricultural pests is similar to that of Guam, and is much higher than any of the other islands in the Carolines. A number of important pests such Oryctes rhinoceros (L.) and Segestes unicolor Redtenbacher occur in Palau, but are absent

Nafus: New Insects in the Carolines and Marshalls

from the rest of Micronesia. Yap has had a rather high rate of new introductions in recent years. Most of these insects are scales or other Homoptera. Guam and Palau are probably the main source, although a few species may be coming from the eastern Carolines. The movement of new pests into the newly emerging countries in Micronesia is a serious problem. In the past, most of these islands have enjoyed a relatively pest free environment, but this is changing rapidly. The new pests reported in this paper, and others which will arrive in the future, will exact a heavy toll on these nations by forcing new expenditures for pesticides or other methods of pest control, increasing pollution, and increasing crop losses. In some cases it may no longer be possible to grow certain crops economically. I feel that the two main problems are the undetected movement of live plant material, and movement of insects in the cargo holds of aircraft. Of the two, movement of plant material is probably the more important. Quarantine cannot guarantee that new pests will be prevented from establishing, but it can significantly slow their movement to new areas. Efforts to strengthen quarantine on each island or island group need to be made. This should include quarantine within islands in a particular country. Acknowledgements I thank the Heads of the Departments or other officials in the Federated States of Micronesia and Palau, and the Republic of the Marshalls for coordinating my activities while visiting. This includes: Sam Falanruw, Yap; Sailas Henry, Adelino Lorens, Pohnpei; Gerson Jackson, Kosrae; Arthur Ansin, Chuuk; and Herman Francisco, Palau; Jimmy Joseph and Douglas Garrott, Marshall Islands. I also thank Albert Arbedul and Haruo Adelbai, Palau; Patrick Sogaw and other extension personnel, Yap; Aluis Ehpel, Kadalihno Lorens, and Dr. Nelson Esguerra, Pohnpei; Takumi George and other Department personnel, Kosrae; Hermes Refit and the entire extension and quarantine service staff on Chuuk. I also thank the following taxonomists for their help: D. J. Williams, British Museum of Natural History, J. Beardsley, Univ. of Hawaii, P. Maddison, D.S.I.R. Auckland, T. J. Henry, U.S.D.A., and S. Nakahara, U.S.D.A. Systematics Laboratory. Funding was provided by the South Pacific Commission and the United Nations F.A.O. References Beardsley, J. W. 1966. Homoptera: Coccoidea. Insects ofMicronesia 6: 377-562. Beardsley, J. W. 1975. Homoptera: Coccoidea, suppl. Insects of Micronesia 9: 657-681. Beardsley, J. W. 1979. New immigrant insects to Hawaii: 1962 through 1976. Proc. Hawaii. Entomol. Soc. 23: 35-44. Bryan, E. H. 1949. Economic Insects of Micronesia. Report of the Insect Control Committee for Micronesia. 1947-8. Washington, D.C., Natl. Res. Counc., 29 p. Clarke, J. F. 1976. Microlepidoptera: Tortricoidea. Insects of Micronesia 9: 1144.

Micronesica Suppl. 3, 1991

Clarke, J. F. 1984. Microlepidoptera: Gelechioidea. Insects ofMicronesia 9: 145155. Common, I. F. & D. Waterhouse. 1982. Butterflies of Australia. Angus and Robertson Publishers, London. 434 pp. Cox, J. 1981. Identification of Planococcus citri (Homoptera: Pseudococcidae) and a description of a new species. Systematic Entomology 6: 47-53. Clausen, C. P. 1978. Introduced Parasites and Predators of Arthropod Pests and Weeds: a World Review. U.S. Dep. Agric. Handb. 480. Denton, G., R. Muniappan, M. Maurtani & T. S. Lali. 1990. The distribution and biological control of the fruit piercing moth, Othreis fullonia in Micronesia. Abstract 5, 2nd ADAP Crop Protection Conf. May 29-30, Univ. Guam, Guam. Esguerra, N. 1990. Control of the leaf-footed plant bug on cucumber by nontrellising. Abstract 16, 2nd ADAP Crop Protection Con( May 29-30, Univ. Guam, Guam. Esaki, T. 1940. A preliminary report on the entomological survey of the Micronesian Islands under the Japanese mandate, with special reference to the insects of economic importance. Proc. 6th Int. Cong. of Entomol. 6: 407415. Esaki, T. 1952. Notes and records on some important pests of Micronesia mostly introduced during the period under Japanese mandate. Proc. 9th Int. Cong. Entomol. 1: 813-818. Essig, E. 0. 1956. Homoptera: Aphididae. Insects of Micronesia 6: 15-37. Fennah, R. G. 1956. Homoptera: Fulgoroidea. Insects of Micronesia 6: 39-176. Fullaway, D. T. 1912. Entomological notes. Guam Agr. Exp. Sta. Rpt. 1911: 2635. Gardner, T. R. 1958. Biological control of insect and plant pests in the Trust Territory and Guam. Proc. lOth Int. Cong. Entomol. (1956) 4: 465-469. Gressitt, J. 1954. Introduction. Insects ofMicronesia 1: 1-257. Gressitt, J. 1955. Coleoptera: Chrysomelidae. Insects of Micronesia 17: 1-60. Julia, J. F. 1983. Oryctes and Rhabdoscelus problems in the Republic of Palau. South Pacific Comm. Doc. 1806. 19 p. Lai, P. Y. 1985. Notes and exhibitions. Proc. Hawaiian Entomol. Soc. 25: 16. Linnavuori, R. 1960. Homoptera: Cicadellidae. Insects of Microinesia 6: 231344. Linnavuori, R. 1975. Homoptera: Cicadellidae, Suppl. Insects of Micronesia 6: 611-632. Nafus, D. 1988. Establishment of Encarsia smithi on Kosrae for control of orange spiny whitefly, Aleurocanthus spiniferus. Proc. Hawaii Entomol. Soc. 28: 229-231. Nafus, D. & I. Schreiner. 1989. History of biological control in the Mariana Islands. Micronesica 22: 65-106. Nako, H. K. & G. Y. Funasaki. 1979. Introductions for biological control in Hawaii: 1975 & 1976. Proc. Hawaii Entomol. Soc. 13: 125-128.

Nafus: New Insects in the Carolines and Marshalls

Paulson, G. & B. Kumashiro. 1985. Hawaiian Aleyrodidae. Proc. Hawaii. Entarnal. Soc. 25: 103-124. Pemberton, C. E. 1954. Invertebrate consultants committee for Pacific. Report for 1949-54. Pacif. Sci. Bd. Natl. Acad. Sci., Natl. Res. Counc., 56 p. Peterson, G. D. 1955. Biological control of the orange spiny whitefly in Guam. J. Econ. Entomol. 48: 681-683. Ruckes, H. 1963. Heteroptera: Pentatomoidea. Insects ofMicronesia 7: 307-356. Schreiner, I. 1989. Biological control introductions in the Caroline and Marshall Islands. Proc. Hawaii. Entomol. Soc. 29: 57-69. Schreiner, I. & D. Nafus. 1986. Accidental introductions of insect pests to Guam, 1945-1985. Proc. Hawaii. Entomol. Soc. 27: 45-52. Takahashi, R. 1956. Homoptera: Aleyrodidae. Insects of Micronesia 6: 1-13. Vandenberg, S.R. 1926. Report of the Entomologist. Guam Agr. Exp. Sta. Rpt. 1925, 17-20. Waterhouse, D. & K. Norris. 1987. Biological Control: Pacific Prospects. Inkata Press, Melbourne. 454 pp. William, D. & G. Watson. 1988. The Scale Insects of the Tropical South Pacific Region, Part 2: The Mealybugs (Pseudococcidae). C.A.B. International Institute of Entomology, London. 260 pp.

Micronesica Suppl. 3: 33-39, 1991

Introduced Vector-Borne Diseases in the Pacific RICHARD

RUSSELL

University of Sydney, Medical Entomology Unit, Department of Infectious Diseases and Microbiology Westmead Hospital, Westmead, NSW 2145, Australia

Abstract-The major vector borne diseases relate to infections with malarial, filarial and arboviral pathogens. Within recent decades the actual or potential movement of these pathogens and/or their vectors into or within the Pacific Region has concerned national and international health authorities. The introduction of vectors of malaria into areas where they do not exist is a major concern; the appearance of exotic Anopheles species and instances of local malaria transmission in Guam illustrate this concern, and islands of the southwest Pacific remain vulnerable. Dengue fever has been active in recent years and, although some indigenous mosquitoes are vectors, further dispersal of the efficient exotic vectors Aedes aegypti and Ae. albopictus in the region should be prevented because of their domestic habits. Other arboviruses also constitute a threat; the introduction of Ross River virus 10 years ago was an instance where local mosquitoes were competent vectors of an exotic pathogen. Filarisasis is not considered an acute problem but efficient Aedes vectors (which may also be arbovirus vectors) should not be allowed to spread to areas with less competent indigenous filarial vectors. Vector control can be difficult once exotic pests are established; quarantine provisions, and efficient routine surveillance and control operations can be used to help prevent many of the above concerns.

Increasing air travel within the Pacific is posing an increasing threat of introducing/dispersing mosquito vectors and/or vector-borne disease within the region, particularly that transmitted by mosquitoes. There are two inherent concerns: transport of infectious human carriers of disease pathogens, and transport of insect (mosquito) vectors (with or without the pathogens). There have been many instances of the former, where travellers have arrived in a country of the region 'incubating' causative agents of exotic disease such as malaria plasmodia, dengue fever and Japanese encephalitis viruses, as well as various filaria. For a vector-borne disease (as distinct from the pathogen causing such disease) to be introduced to an island in the region, the presence of a local susceptible vector is required to achieve transmission from the infected carrier. This vector may be an indigenous species which is competent to maintain and transmit the pathogen, or it may be an introduced exotic species which is an efficient natural vector and has been transported and become established locally; but a suitable

Micronesica Suppl. 3, 1991

vector available locally when the pathogen is imported with a human carrier is prerequisite for introduction of a disease. Insect Vectors of Disease The carriage of insects on international aircraft has been documented over many decades and has been reviewed by Russell et al. (1984). There has been a number of survey reports of exotic insect vectors being found on board aircraft entering countries in the Pacific/Southeast Asian regions, e.g. Japan (Takahashi 1984), Philippines (Basio et al. 1970, Basio 1973), Singapore (Goh et al. 1985) and Australia (Russell et al. 1984). Vector and pest species from the genera Aedes, Anopheles, Culex and Mansonia have been recorded in the above-mentioned collections, particularly Aedes aegypti and Ae. albopictus, important vectors of dengue; Anopheles sundaicus and An. subpictus, malaria vectors of concern for coastal environments; Culex tritaeniorhynchus and Cx. gelidus, important vectors of Japanese encephalitis. Not all (e.g. Cx. quinquefasciatus) are exotic species for the country of import but may be of concern even so if they are carrying pathogens or introducing genes for insecticide resistance. It is not only the passenger cabin and cargo holds of aircraft that may be important: mosquitoes and other insects have been shown to be able to survive international travel in the wheel bays of large modern passenger aircraft (Russell 1987). There have been some introductions of vector mosquitoes of major consequence into islands of the Pacific region. Ae. aegypti is now widely distributed in the Pacific, where dengue has become virtually endemic, but dispersal is still occurring; it appears to have been only recently introduced to some islands, e.g. Tokelau, and there are still some islands where it has not yet been introduced (Pillai & Ramalingam 1984). There are also some islands from which it is thought to have been eradicated (e.g. Rarotonga and Guam), so its movement into and through the region is still of concern and should be prevented. Ae. albopictus was introduced to the Carolines and Marshalls during the 1960s or 1970s, probably from Guam or Saipan, and the species was unknown east of PNG until 1980 when it was recorded from the Solomons after introduction probably from PNG; Ae. vigilax has been introduced to Fiji and Tonga, Ae. australis and Ae. notoscriptus into New Zealand, and Cx. quinquefasciatus was relatively recently introduced to Tokelau (Pillai & Ramalingam 1984). There is a valid major concern for the introduction of malaria vectors of the Anopheles punctulatus group into the central and eastern parts of the south Pacific where ground pool habitats appear to be eminently suitable for the establishment of An. farauti or An. punctulatus (Self & Smith 1984). Concerns remain also for intraregional transportation and establishment of vectors. For instance, if Ae. aegypti is currently absent from Guam there is considerable risk of its reintroduction from Asian ports (particularly by air from Manila) but also from neighboring islands in Micronesia. There are risks of interisland transfers elsewhere, e.g. Aedes aegypti occurs in Tonga but not in nearby Rarotonga (Cook Islands), while Aedes polynesiensis (a vector of filariasis, dengue

Russell: Vector-Borne Diseases

and Ross River virus) occurs on Rarotonga but not in Tonga; the night-biting mosquitoes of the Ae. kochi group, which are important vectors of filariasis in Samoa and Tonga are absent from neighboring French Polynesia which has daybiting vectors. The situation on Guam, with its heavy influx of international traffic and consequent establishment of several malaria, dengue and arbo-encephalitis vectors over the years, is a prime illustration of the reality of such transport of insects. At least 18 species of mosquitoes have become established on the island of Guam, 15 through international (primarily aerial) transportation, mostly since WW II; these include 5 Anopheles species on a previously anopheline-free island (Ward 1984). Of course, there is also the potential (and indeed the reality) of sea-transport of mosquitoes, especially for Aedes species which can colonize ships' receptacles such as water barrels and water tanks, as well as some cargo such as tires, and can thus survive long sea journeys and be introduced to new ports. Aedes togoi, a potential filariasis vector, was introduced from Japan to Malaysia about 1960 by ship; and Anopheles litoralis, a malaria vector, was exported from the Philippines to Malaysia by ship in 1970 (Pillai & Ramalingam 1984). The introduction of Ae. albopictus to the USA through the agency of tyre imports (Francy et al. 1990) is a more recent illustration. Aedes aegypti and Aedes albopictus are the species of greatest concern potentially most likely to be dispersed and introduced within the region by ship. Prior to this century and the proliferation of air travel, the transport of A e. aegypti via sea trading routes over centuries enabled its dispersal and establishment in many ports of the tropical and subtropical regions of the world, and no doubt this sea-transport continues as illustrated by the occasional recording of Aedes eggs and larvae on vessels arriving in northern Australia from Asian ports. Vector-Borne Disease

From a disease perspective, a major concern in the region is that malaria vectors will be introduced and become established in islands of the non-malarious Pacific, but the potential for viraemic humans to introduce dengue and other viruses into countries of the Pacific where efficient vectors exist is also important. The potential for such introductions and the implications should be reinforced to health authorities, the medical profession and the general public. Malaria infections acquired during flight and on the ground at European airports (Curtis & White 1984, Isaacson 1989) attest to the potential for movement of pathogens with vectors in international air traffic. Considerable malaria activity continues in Melanesia (PNG, Solomons, Vanuatu) and eastern Asia from whence parasite importations to Pacific area could (and perhaps do) occur-but currently there are no vectors east of Vanuatu in the south Pacific and none in Micronesia except for Guam where all local potential vectors have been introduced. Malaria infections recorded in Guam have been attributed to these introduced Anopheles vectors (Smith & Carter 1984).

Micronesica Suppl. 3, 1991

Dengue virus activity has been widespread in recent decades and almost continuous in the Pacific region through 1988-90, with various serotypes reported, and haemorrhagic manifestations and deaths in some countries; dissemination of the viruses can be presumed to be via inter-island travellers, although dispersal of infected vectors cannot be excluded for those countries not insisting on disinsection of arriving aircraft. A signal example of the potential for disease caused by the introduction of an exotic disease agent able to be transmitted by local vectors is that related to the introduction of Ross River (RR) virus to the region in 1979. In April 1979, cases resulting from RR virus infection appeared in N adi, Fiji and were subsequently reported elsewhere in the country to eventually total an estimated 50,000 clinical cases (possibly more than 300,000 infections among the 630,000 population); by August 1979 the outbreak spread to American Samoa (with an estimate ofhalfthe population affected), by November 1979 to Wallis and Futuna Islands and thence to New Caledonia; by February 1980 cases were appearing in the Cook Islands. The likelihood is that the virus was introduced to Fiji from Australia in 1979 in a human incubating the disease or in a mosquito that escaped aircraft disinsection, probably the former who infected local mosquitoes at a time when the introduced species, and known vector in Australia, Ae. vigilax was very abundant in the area of the Nadi airport. Human transportation of the virus to the other island groups of the region, initiating local outbreaks, can explain the rest of the story. As far as local vectors are concerned, introduced species such as Ae. vigilax and Ae. aegypti, and the local Culex annulirostris, could have been involved as vectors in Fiji but, in all the island groups involved other than New Caledonia, a local species, Ae. polynesiensis, appeared to have been most commonly involved as the vector for the exotic virus (Miles 1984). Although RR virus may cause considerable morbidity within communities, that it does not cause human mortality is something for which the region can be thankful considering the extent and intensity of the 1979-80 outbreak. The implications of a similar introduction and epidemic of Yellow Fever virus in the region should be considered by health authorities. Minimizing the Risks of Introduction and Establishment of Vectors and Vector-Borne Disease There is a requirement for an improvement in collection and dissemination of technical information relating to potential risk of introductions and actual introductions, and to national measures designed to prevent or inhibit introduction and establishment of exotic pests. Education of the travelling public, travel agents, and companies involved in international transport, trade and tourism, as well as health authorities should be improved to complement 'official' quarantine measures. Aircraft introductions probably present the major problem for the region, and air routes presenting the greatest risks for importing vectors or human carriers should be identified by each national authority and local quarantine measures

Russell: Vector-Borne Diseases

instituted where required. For instance, flights from malarious to non-malarious regions, particularly where nighttime departures that may attract Anopheles vectors are involved, should command quarantine attention. And humans recently visiting yellow fever endemic regions of Africa or South America, and wishing to enter islands of the region with competent vectors, should command quarantine attention. Aircraft disinsection is often employed as a first line of defence in many countries of the region to prevent introduction of exotic insect vectors, and whether done 'blocks-away', 'in-flight' or 'on-arrival', it can be an effective quarantine procedure (Russell et al. 1984, Russell 1989, Russell & Paton 1989). However, airport sanitation is an important complementary measure and assists in preventing the establishment of introduced species. Regular and frequent surveys should be undertaken in the vicinity of all major air- and sea-ports in the region to monitor abundance of indigenous insect species and detect the presence of exotic species. International vessels and aircraft should be examined periodically, and insects collected and identified to assess relative risks associated with particular routes. Seaport inspections and consideration of shipping routes should not be ignored, and routine ship inspections should include examination of water containers for mosquito larvae and eggs. Sanitation of the port vicinity should be maintained and improved where necessary to prevent establishment of introduced species and minimize breeding of local vectors that may contact passengers and infiltrate departing vessels or aircraft. Measures should include appropriate engineering and chemical treatment, as well as collection and disposal of non-essential containers that serve as larval habitats. Airport lounges need to be screened and air-conditioned if possible to protect transit passengers from passing on or picking up mosquito-borne infections. In an international health context, vector surveillance and control is as relevant for points of departure as for points of entry, and international cooperation will go a long way to reducing the risk faced by many countries in the region. There are many considerations within the overall context of this issue. However, the establishment of malaria vectors in the malaria-free zone of the Pacific would greatly threaten the public health, the introduction of supplementary efficient domestic dengue vectors into the eastern Pacific and/or the continuing introduction of various dengue virus serotypes to the Pacific would seriously increase the incidence and severity of the disease, and the introduction of potentially calamitous exotic virus diseases such as yellow fever is something that cannot be excluded with modern travel. With the current situation of decreasing quarantine barriers world-wide, health authorities of the Pacific basin and its rim should carefully consider the implications for the introduction of vectors and vector-borne diseases, and institute guidelines and provisions to reduce the risk of such introductions, in cooperation with fellow nations of the region. References Basio, R. G. 1973. The mosquito control program at the Manila International Airport and vicinity (Philippines) with comments on problems encountered

Micronesica Suppl. 3, 1991

on the aerial transportation of mosquitoes. In Y. C. Chan, K. L. Chan & B. C. Ho (eds.), Vector Control in Southeast Asia, pp. 78-84. Proc. First SEAMEO workshop. Basio, R. G., M. J. Prudencio & I. E. Chanco. 1970. Notes on the aerial transportation of mosquitoes and other insects at the Manila International Airport. Philipp. Entomol. 1: 407-408. Curtis, C. F. & G. B. White. 1984. Plasmodium falciparum transmission in England: entomological and epidemiological data relative to cases in 1983. J. Trop. Med. & Hyg. 87: 101-114. Francy, D. B., C. G. Moore & D. A. Eliason. 1990. Past, present and future of Aedes albopictus in the United States. J. Am. Mosq. Cont. Assoc. 6: 127132. Goh, K. T., S. K. Ng & S. Kumarapathy. 1985. Disease-bearing insects brought in by international aircraft into Singapore. Southeast Asian J. Trop. Pub. Hlth. 16: 49-53. Isaacson, M. 1989. Airport malaria: a review, Bull. Wld. Hlth. Org. 67: 737-743. Miles, J. A. R. 1984. On the spread of Ross River virus in the islands ofthe south Pacific, In M. Laird (ed. ), Commerce and the spread of pests and diseases, pp., 209-224. Praeger, New York. Pillai, J. S. & S. Ramalingam. 1984. Recent introductions of some medically important Diptera in the northwest, central and south Pacific (including New Zealand), In M. Laird (ed.), Commerce and the Spread ofPests and Diseases, pp. 81-101. Praeger, New York. Russell, R. C. 1987. Survival of insects in the wheel bays of a Boeing 747B aircraft on flights between tropical and temperate airports. Bull. Wld. Hlth. Org. 65: 659-662. Russell, R. C. 1989. Transport of insects of public health importance on international aircraft. Travel Medicine Int. 7: 26-31. Russell, R. C. & R. Paton. 1989. In-flight disinsection as an efficacious procedure for preventing international transport of insects of public health importance. Bull. Wld. Hlth. Org. 67: 543-547. Russell, R. C., N. Rajapaksa, P. I. Whelan & W. A. Langsford. 1984. Mosquito and other insect introductions to Australia aboard international aircraft, and the monitoring of disinsection procedures, In M. Laird (ed.), Commerce and the Spread of Pests and Diseases, pp. 109-41. Praeger, New York. Self, L. S. & A. Smith. 1984. Preventive measures against importing malaria vectors into Pacific islands, In M. Laird (ed.), Commerce and the Spread of Pests and Diseases, pp. 163-175. Praeger, New York. Smith, A. & I. D. Carter. 1984. International transportation of mosquitoes of public health importance. In M. Laird (ed.), Commerce and the Spread of Pests and Diseases, pp. 1-21. Praeger, New York. Takahashi, S. 1984. Survey on accidental introductions of insects entering Japan via aircraft, In M. Laird (ed. ), Commerce and the Spread of Pests and Diseases, pp. 65-79. Praeger, New York.

Russell: Vector-Borne Diseases

Ward, R. A. 1984. Mosquito fauna of Guam: case history of an introduced fauna, In M. Laird (ed.), Commerce and the Spread of Pests and Diseases, pp. 143162. Praeger, New York.

Micronesica Suppl. 3: 41-45, 1991

Plant Diseases of Recent Introduction to Guam GEORGE

WALL

College of Agriculture & Life Sciences Agricultural Experiment Station University of Guam, Mangilao, Guam, 96923

Abstract-Over 50 new reports of plant diseases were included in the Guam Agricultural Experiment Station Annual report of 1988. Some have evidently been here for some time but have escaped detection previously; others are recent introductions. Bacterial blight of cassava (Xanthomonas campestris pv. manihotis) was first detected in 1986. Fruit blotch of watermelon (Pseudomonas pseudoalcaligenes subsp. citrulli) was first seen in 1987; several damaging blotch epidemics have resulted in serious economic losses to watermelon growers. A sample of a lanzones tree (Lansium domesticum) with stem galls was found to have plant parasitic nematodes. Leaf scald of cabbage was found in 1988; it is caused by a strain of Xanthomonas campestris pv. campestris, the same bacterium causing black rot. More recently, Bacterialleafblight (Pseudomonas avenae) and leaf blight (Bipolaris maydis) have been found on corn. Some control measures are suggested that can slow the rate of pest importation. Introduction

New plant diseases appear on Guam with relative frequency. Although islands are blessed with a large degree of isolation from other land masses, and therefore sources of pest problems, modern transportation and our modem lifestyles are closing the gap more quickly than our agriculture can adapt to new pest programs. Our quarantine regulations, both federal and local, and the immense efforts made by our quarantine authorities, are successful in protecting us from accidental introductions of new pests, to a certain degree. Our own human . population, however, is quite able to find ways of overcoming the very quarantine regulations that are supposed to protect us. Quarantine authorities often discover individuals attempting to "smuggle" plant propagules, fruit, and produce. The reasons why we may want to introduce these materials are numerous and varied, and the reasons why some people attempt to go around legal procedures are also many. I doubt that anyone would want to bring in a new plant pathogen, yet from time to time individuals unaware of the pathogens in their plant material smuggle in their favorite varieties of this or that plant. Smuggling in plant material is not the only way we can get new diseases introduced. Severe cases of epidemic developments can and have been started

Micronesica Suppl. 3, 1991

by introduction of planting materials through legal channels. Anyone can order seed through catalogs. There are many seedborne diseases that are not present on island, but may one day appear in our fields. This happens more frequently than we may care to admit. A few potentially devastating diseases not occurring here, but which may be introduced with seed, are various downy mildews of grass crops, and late blight, bacterial canker, spot, and speck of tomato. We can also get seed ofwitchweed mixed in with seed of other plants, coming from places where that parasitic plant is established. Various weed seeds are thought to find their way into new areas in this way as well. Infective vectors of plant pathogens can also reach our island, and potentially introduce a new pathogen. Consider, for instance, how often big airplanes are loaded at night, under strong floodlights that surely attract a vast number of nearby insects. One way or another, many plant pathogens have found their way into Guam in recent years. In the 1988 Guam Agricultural Experiment Station Annual Report, there are more than 50 accounts of plant diseases not known three years earlier (Russo et al. 1985). A number of these plant diseases have evidently been present for some time, while others are recent introductions. Some recent cases of actual interceptions, limited establishments, and economically damaging epidemics of newly introduced plant diseases on Guam will be discussed. Interceptions

Numerous plant introductions have been intercepted because samples have been found to contain live plant-parasitic nematodes. Many ornamentals and root crops have been confiscated for this reason. Plant propagules of various sorts have been intercepted and found to be infected with fungal and bacterial pathogens. Produce is sometimes found with potentially infective exotic pathogens. Cabbage heads with black specking, which could be caused by either Turnip Mosaic Virus or Cauliflower Mosaic Virus, or both, have been found several times in produce shipments from Hawaii and from mainland ports. Chinese cabbage from Hawaii has also been confiscated for the same reason. Although specking in chinese cabbage can be caused by a physiologic disorder, it can also be due to Cauliflower Mosiac Virus infection (Sherf & MacNab 1986). Established Pathogens

Individual trees have been found infected with exotic diseases. One mango tree (Mangifera indica) was found in Mangilao in 1988, infected by Woody Gall Virus (see Cook 1975). The tree came from a Haden seed; the fruit was bought at a local supermarket (imported most likely from Mexico via USA). Symptoms were obvious after the tree was 8-10 years old. A lanzones tree (Lansium domesticum) was found in Sinajana in 1987 infected by a nematode causing galls on its branches. The pathogen is pending identification.

Wall: New Plant Diseases of Guam

Isolated fields have been found infected with exotic diseases as well. A cabbage field in Barrigade (in 1987) had many plants infected by bacterial leaf scald, caused by a strain of Xanthomonas campestris pv. campestris (Queensland Dept. of Primary Industries 1982). While the more common strain causing black rot is reported here (Russo et al. 1985) and can be frequently found in cabbage fields, the leaf scald strain is previously unreported. A corn field in Dededo (in 1990) had northern leaf blight, a disease caused by the fungus Bipolaris maydis (Shurtleff 1980). During the last decade, this exotic disease created disastrous epidemics in the corn belt region of the United States. The fungus is capable of infecting the corn ears. The same corn field in Dededo was found to have several plants infected with bacterial leaf blight, a disease caused by Pseudomonas avenae (Shurtleff 1980). Again, this corn pathogen was found infecting the corn ears. Epidemics

Cassava blight (Xanthomonas campestris pv. manihotis) was recently introduced. It was first found on Guam in 1986 by G. V. H. Jackson (SPC, Suva, Fiji). It was very likely introduced through infected cuttings, and spread throughout Mangilao, Barrigada, and Inarajan (Wall & Santos 1987). The watermelon fruit blotch (Pseudomonas pseudoalcaligenes subsp. citrulli) was almost certainly introduced through commercial seed, and developed into an epidemic in the Dandan area (Wall & Santos 1988, Wall 1989). It has also occurred in various other locations throughout Southern Guam and in Barrigada. Outbreaks still occur. Fruit blotch may have even found its way to Guam in more ways than one, since fruit shipments coming in from Tinian were intercepted and infected fruit were found (Wallet al. 1990). Solutions

There is no easy solution available, only hard work to enforce quarantine regulations, educate the public, and keep abreast of disease development and identification. The bacterial blight epidemic of cassava has been controlled in Inarajan Experiment Station so far, for instance, by a series of control measures. These included eradication from the station and surrounding area, strict sanitation to destroy all infected plant residues, and quarantine (no cassava planted) for 3 years (Wall & Santos 1987). On a more general basis, and dealing with current and future dangers of important exotic pests, the following points need consideration: 1. Our plant quarantine authorities need to hire and train personnel to handle the specimens we have traditionally handled at the University, because sample numbers are constantly increasing and so is our work load. 2. Plant quarantine authorities can now make use of computers and databases for quick references on pests reported from different parts of the world. (FAO database, T. Putter, FAO, Rome, Personal communication).

Micronesica Suppl. 3, 1991

3. Publications such as the SPC book on exotic pests (O'Connor 1967) make the job easier for persons in charge of inspections and identifications. This publication is outdated, however. Such a list should be updated periodically, at the very least every two years. 4. Techniques for identifying plant pathogens are making the job easier every day, but these techniques are often expensive, and otherwise time-consuming. Already there are serological (ELISA) kits available commercially to identify many viruses, some fungi, and bacteria. The technique could be used either at the point of origin to insure healthy stock is being exported, or at the port of entry to screen shipments for specific exotic pests. 5. Individuals ordering their own seed through catalogs should surface-disinfest their seed before planting. This would not necessarily eliminate systemic pathogens, but would eliminate many pathogens present as contaminants. Seed companies often have a seed treatment service available, upon request by the customer. The implementation of this recommendation requires public awareness. 6. We need to keep in mind that it is easier to exclude than to eradicate. Eradication campaigns have seldom worked. Every dollar spent in efforts to exclude important exotic pests is potentially saving a tenfold amount that would be required for eradication or a hundredfold for continuous control. Funds should be made available generously to our plant quarantine officials, to ensure they can operate adequately and effectively. 7. We need propaganda campaigns to constantly keep the public aware of the exotic pest problem. We can be our own worst enemies. Conclusion

In spite of the alarming rate of new pest reports, Guam is still free of a vast number of important plant diseases. There are several bacterial diseases of solanaceous crops that are still not found here, for instance, although they can be seedborne (Sherf & MacNab, 1986), and there is a large number of cucurbit and solanaceous viruses not found here as yet. Moko disease of banana is fortunately absent, as is the lethal yellows disease of palms, and the downy mildews of grasses. Historically, however, plant quarantine regulations have had only limited success in excluding pests from a region. Given enough time, these exotic pests can and do find their way in. In the case of important commodities, it is economically and environmentally advantageous to prolong this pest-free period as much as possible. Therefore, existing plant quarantine regulations need to be supplemented in a number of ways, as discussed previously. Most of the recommendations made here have a price tag that would have to be paid by our local quarantine authorities. Some require international funds and cooperation. Perhaps these recommendations will never be implemented. To do so would require that they survive the political battlefields of government policy-making, local and international. Nevertheless, they need to be pointed out and considered.

Wall: New Plant Diseases of Guam

References Cook. A. A. 1975. Diseases of Tropical and Subtropical Fruits and Nuts. Hafner Press, NY. pp. 238-240. Guam Agricultural Experiment Station. 1988. Annual Report. CALS Media, University of Guam. pp. 18-20. O'Connor, B. A. 1967. Exotic Plant Pests and Diseases. South Pacific Commission, Noumea, New Caledonia. Queensland Department of Primary Industries. 1982. A Handbook of Plant Diseases in Colour, Vol. 1, Fruit & Vegetables. S. R. Hampson, Government Printer, Queensland. Russo, V. M., G. Beaver, F. Cruz, and H. Rubin. (1985). Plant pathogens and associated hosts on Guam. Technical Report, AES Publication 46, CALS, University of Guam. Sherf, A. F., A. A. McNab. 1986. Vegetable Diseases and Their Control, 2nd ed. Wiley-Interscience, New York. 728 pp. Shurtleff, M. C. (ed.) 1980. Compendium of Corn Diseases, Second edition. APS Press, St. Paul. 105 pp. Wall, G. C. 1989. Control of watermelon fruit blotch by seed heat-treatment. Phytopathology 79: 1191. Wall, G. C. & V. M. Santos. 1987. Bacterial blight of mendioca (cassava). Agricultural Experiment Station, GCE Publication No. PP 88-1, University of Guam. Wall, G. C. & V. M. Santos. 1988. A new bacterial disease of watermelon in the Mariana Islands. Phytopathology 78: 1605. Wall, G. C., V. M. Santos, F. J. Cruz, D. A. Nelson, & I. Cabrera. 1990. Outbreak of watermelon fruit blotch in the Mariana Islands. Plant Disease 74: 80.

Micronesica Suppl. 3: 47-49, 1991

Introduced Ornamental Plants that Have Become Weeds on Guam J. McCONNELL and R. MUNIAPPAN Agricultural Experiment Station, University of Guam Mangilao, Guam 96923 USA

Abstract-Introduced ornamental plants that are considered weeds on Guam include Antigonon leptopus Hooker & Arnott, Spathodea campanulata Beauv., Coccinea grandis, Bauhinia monandra Kurz., Clerodendrum quadriloculare (Blanco) Merrill, Lantana camara L., Ficus spp., Asystasia gangetica (L.), Pilea microphylla (L.) Liebm., Wedelia trilobata (L.) Hitche., Mikania scandens and Mimosa pudica L. Characteristics of ornamental plants that could become weeds on Guam include rapid growth/regrowth and prolific seed production. Introduction Many introduced perennial ornamental herbs, shrubs, vines and trees have become weeds in their new habitats in the absence of their natural enemies that kept them under check in their native habitats. Neel & Will (1978), Hardt (1986), and Patterson (1976) listed a number of introduced ornamental plants that became weeds in the United States. Hazard (1988) has drawn attention to the risk of introducing weeds in the process of introducing pasture, crop and ornamental plants into Australia. He has also given examples of such weeds. Several plants introduced as ornamentals have become serious weeds on Guam. This paper presents a list of such exotic weeds established on Guam with some notes on them.

Antigonon leptapus Hooker & Arnott (Polygonaceae) - Chain of Love, Mexican Creeper This aggressive spreading vine is a native of Mexico and has been introduced to many tropical and sub-tropical regions as an ornamental plant. The heart shaped leaves are 3 to 5 inches long and are distributed alternately along vining stems. It flowers throughout the year and is prolific in producing seeds. Small edible underground tubers are produced. This has become a very aggressive weed and has overgrown most vegetation in many areas in Guam. Spathodea campanulata Beauv. (Bigoniaceae) - African Tulip Tree This is a large showy tree growing to 70 feet. It is quite common in Guam. The leaves are dark green, 1 to 2 feet long. The flowers are orange-scarlet lined

Micronesica Suppl. 3, 1991

with yellow. The flowers are found in abundance through much of the year. The wood of this tree is soft and is easily broken during storms. The fallen pieces are capable of rooting and develop into trees. It is also a very prolific seed producer and easily propagates by seed. It was introduced as an ornamental but it has escaped cultivation to many of the roadsides in Guam. Coccinea grandis (L.) Voigt (Cucurbitaceae) This plant is a recent introduction to Guam and is just becoming established. It is not really used as an ornamental but is used medicinally and as a food plant. It has not yet been observed producing fruits in Guam and has apparently spread only vegetatively. It is capable of covering the canopy of other plants including large trees. In Hawaii and the Caribbean, birds eat the dark pink ripened fruits. Bauhinia monandra Kurz. (Caesalpiniaceae) Orchid Tree The orchid tree has leaves shaped like butterfly wings and attractive orchidlike flower. It is often used as a landscape plant. It is generally propagated by cutting or seed. It is considered a fast growing tree and heavy seed producer. It has escaped from cultivation and has become established in many wooded areas in Guam. Clerodendrum quadriloculare (Blanco) Merrill. (Verbenaceae) This shrub has an attractive dark foliage and is used as a landscape plant. It produces many suckers from the roots and is a prolific seed producer. It readily establishes along the foundations of houses and is difficult to remove. Clerodendrum has escaped cultivation and can be observed along roadsides in several areas around Guam. Lantana camera L. (Verbenaceae) Lantana is a popular ornamental shrub that has been introduced to many areas because of its abundant and colorful flowers. In tropical and subtropical regions it has become a serious weed. Birds feed on the berries oflantana spreading it from cultivation to pastures and forests. With thorny woody stems lantana is difficult to remove once established. Ficus spp. (Moraceae) Many species of Ficus are used as landscape plants or as house plants. They have a great potential to become weeds if allowed to escape cultivation. They can become a weed of other ornamental plants and are difficult to eradicate once established. Seeds, spread by birds, can germinate in the crotch of other tress and eventually overgrow the host tree. Cut trunks will readily sprout new vegetation. Many species are prolific seed producers. In addition, the surface root system can damage sidewalks and curbs. Asystasia gangetica (L.) (Acanthaceae) Asystasia is a trailing perennial herb, which can climb several feet high on fences and taller plants. It is cultivated as a cover crop but has escaped cultivation

McConnell& Muniappan: Ornamentals as Weeds

and became established in many areas on Guam. Asystasia is a prolific seed producer and rapidly fills an area with seedlings.

Pilea microphylla (L.) Liebm. (Urticaceae) Artillery Plant Pilea is often grown as an ornamental. It is a small plant but is a prolific seed producer. It is commonly found on the surface of potted plants and in lawns. It is an aggressive grower and quickly makes a dense cover in shade. It is even capable of overgrowing small orchid seedlings growing in crushed limestone. Nurseries on Guam import various forms of this plant from Hawaii. Wedelia trilobata (L.) Hitche. (Asteraceae) This plant is commonly used as a ground cover. It has showy yellow flowers and quickly establishes a dense cover that competes well with other plants. It is aggressive enough to become a weed. There are many locations around Guam where it has escaped cultivation and has covered entire clearings. Mikania scandens (L.) Willd. (Asteraceae) This is a member of the Eupatorium tribe. It is an aggressive vine with white flowers. It is suited to colonizing in wild places with its aggressive vining habit. It can be observed growing in large patches around Guam. Mimosa pudica L. (Mimosoideae) Sensitive Plant This is a low growing thorny plant with reddish-brown stems that are sensitive to touch. It is a common curiosity plant in temperate regions and escaped from cultivation in many tropical and sub-tropical regions. It is a serious weed in lawns and roadsides on Guam. Because it has woody stems, it is difficult to weed by hand. Plant Characteristics for Identifying Potential Weeds It is important that we consider the potential of new ornamental plants for becoming weeds in Guam. Australia has prohibited the entry of 66 species and 21 genera that have weed potential (Hazard 1988). One of the difficulties in deciding to introduce a new species is that some of the characteristics of a weed are desirable characteristics for some intended uses. For example, ground covers should be aggressive growers so that they will quickly establish in a planted area. Another desirable characteristic for an ornamental is prolific flowering. Individuals in tropical and sub-tropical regions must become more aware of the risks in freely introducing new ornamentals. Knowledge of potential weeds and potential characteristics to watch for would help lessen the introduction of new weeds. Once a plant becomes a weed in a region, it has the potential to be a weed elsewhere with similar bioclimatic regimes, and neighboring areas should be alerted to the potential threat.

Micronesica Suppl. 3, 1991

References Hardt, R. A. 1986. Japanese honeysuckle: from "one ofthe best" to ruthless pest. Arnoldia 27-34. Hazard, W. H. L. 1988. Introducing crop, pasture and ornamental species into Australia. The risk of introducing new weeds. Australian Plant Introduction Review 19-36. Neel, P. L. & A. A. Will. 1978. Grevillea chrysodendron R. Br.: Potential weed in South Florida. HortScience 13: 18-21. Patterson, D. T. 1976. The history and distribution of five exotic weeds in North Carolina. Castanea 41: 177-180.

Micronesica Suppl. 3: 51-62, 1991

Spread of Freshwater Pomacea Snails (Pilidae, Mollusca) from Argentina to Asia 0SAMU MOCHIDA National Agriculture Research Centre, Ibaraki-ken, Tukuba-si, Kan 'nondai, 305 Japan

Abstract-Pomacea canaliculata (Lamarck) was introduced as a human food from Argentina into Taiwan (China) in 1979-80. It was introduced from Taiwan to Japan in 1981, the Philippines in 1982, China (Guangdong, Fuzhou, & Hangzhou) in March 1985, Korea (Suwon) in 1987 or earlier, Malaysia (Sarawak) in 1987 or earlier, and Indonesia and Thailand by 1989. The snail has escaped from aquaculture and established in open fields affecting aquatic plants. Estimation of areas infested by this snail was 171,425 ha in Taiwan in 1986, 16,195 ha in Japan in 1989, and ca. 400,000 ha in the Philippines in 1989. In Japan ethylthimetone + thiocyclam G, cartap G, bensultap G, and IBP G are registered as snail repellents, and IBP G shows some toxicity. Calcium cyanamide is registered as a fertilizer with toxicity to snails. Metaldehyde, triphenyl-tin acetate, and niclosamide are not registered for control purposes because of high toxicity to fish. In Taiwan, however, metaldehyde and triphenyl-tin acetate are registered. In addition, the predatory black carp (Mylopharyngodon piceus (Richardson)) and common carp (Cyprinus carpio L.) are recommended as control agents. In the Philippines, 5 chemicals including triphenyl tin-acetate, niclosamide, and endosulfan are in the market. Hand-picking, ducks, setting of metal screens, etc. are also recommended. More effective methods of control have yet to be developed. Introduction

Pomacea canaliculata (Lamarck) commonly known as golden snail, golden apple snail, apple snail, jumbo snail, and golden miracle snail, is indigenous to South America and was introduced to Taiwan from Argentina in 1979-1980 and into Kyuusyuu and Wakayama, Japan, from Taiwan in 1981. The snail was cultured in ponds and sold as fresh, canned or bottled human food. However, currently it has little commercial value in either Taiwan or Japan. There are at least three species of Pomacea that occur in the Philippines: P. canaliculata was introduced from Taiwan into the Rafael Atayde Hatchery, Lemary, Batangas, Luzon in 1982. P. gigas (Spix) was imported by the Bio-Research Institute, Metro Manila from Florida, USA in 1983. P. cuprina (Reeve) was also

Micronesica Suppl. 3, 1991

introduced to Manila around 1983. P. canaliculata was introduced from Luzon around 1983 and directly from Argentina or Taiwan into the Asturias Farm in Cebu in 1984 for commercial production (Mochida 1987, 1988a, b). Pomacea culture was recommended to farmers through a livelihood project of the Philippine Government up to 1988. Pomacea snails were brought to Fuzhou from Guangzhou, China in March 1985. They were also found in aquaculture at Suwon, Korea in 1986, in Sarawak, East Malaysia in 1987, and Bangkok, Thailand in 1990. The spread of the snails in East Asia is shown in Fig. 1. Taxonomy The scientific name -of the freshwater snails introduced from Argentina to Japan via Taiwan around 1981 was frequently confused. According to Habe (1986), the valid and invalid names are as follows: Phylum Mollusca Class Gastropoda Order Mesogastropoda Pilidae (Synonym; Ampullariidae) Family Pomacea canaliculata (Lamarck) (Plate I A-E) Species Ampul/aria canaliculata Lamarck A. canaliculata Lamarck A. insularum Hamada & Matsumoto 1985 (nee d'Obigny 1839) A. insularus Chang 1985 (nee d'Orbigny 1839) A. insularus Miyazaki 1985 (nee d'Orbigny 1839) A. insularus Hirai et al. 1986 (nee d'Orbigny 1839) A. insularus Kaneshima et al. 1986 (nee d'Orbigny 1839) A. insularus Miyahara et al. 1986a, b (nee d'Orbigny 1839) A. insularus Oya et al. 1986 (nee d'Orbigny 1839) Pest status of Pomacea canaliculata Aquatic plants affected by this snail are: young rice seedlings ( Oryza sativa L.), taro (Colocasia esculenta (L.) Schott), swamp cabbage (Ipomoea aquatica Forskal), lotus (Nelumbo nucifera Gaertner), mat rush (Juncus decipiens Buchenau), Chinese mat grass (Cyperus monophyllus Vahl), wild rice (Zizania latifolia Turczaninov), Japanese parsley/dropwort (Oenanthe stolonifera Wall), water chestnuts (Trapa bicornis Osbeck), and azolla (Azolla spp.). Also, the snail fed on the terrestrial plants such as cabbage placed in their culture ponds. In Taiwan, Pomacea snails infested 17,000 ha in 1982 and 171,425 ha in 1986. In 1986, the area treated with molluscicides was 103,350 ha, and the estimated loss in rice fields alone was US $30.9 million. Over half a million fingerlings of black carp (Mylopharyngodon piceus (Richardson)) and even more of common carp (Cyprinus carpio Linne) were released to control the snails (Table 1).

Moeh·da· 1 • Freshwa t er Snails

Figure 1.

. m . East Asia. Spread of Pomacea snails

Micronesica Suppl. 3, 1991

Table 1.

Area infested by Pomacea snails, area treated with molluscides, and fingerlings released for controlling the snails in Taiwan, China (MAF, Taiwan, 1985b, 1986). Year

Area Infested (ha) Rice fields Others (Total)

1982

1983

1984

13,000 4,000 (17,000)

40,574 11,071 (51,645)

72,780 16,500 (89,280)

147,311 19,382 (166,693)

151,444 19,980 (171,425)

32,000 14,000 (46,000)

15,000 5,000 (20,000)

35,560 12,135 (47,695)

90,000 13,350 (103,350)

8.3

14.9

30.1

30.9

Area Treated with Molluscides (ha) Rice fields Others (Total) Estimated Loss in Rice Fields US$ (X million) No. Fingerlings Released Black Carp* Common Carp**

2.7

85,000 500,000

1985

1986

592,000 650,000

*Mylopharyngodon piceus. **Cyperinus carpio.

Since 1981 the snails were introduced for aquaculture from Taiwan into some locations like Nagasaki (Kyuusyuu), and Wakayama (Honsyuu) in Japan. Eventually snails escaped from the culture ponds into open fields, streams, ponds, and rice fields. In 1986 the Federal government spent 9 million yen (US $64,385) to control them in 176 ha of rice fields. As of the end of 1989 the snails have been observed in 35 out of 47 prefectures (Fig. 2). The areas of different crops attacked by the snails are given in Table 2. In the Philippines, the infestation of rice fields by the snails was estimated 13,945 ha in February 1988 (Rejesus et al. 1989) and about 400,000 ha in 1989 (FAO 1989). Further FAO ( 1989) also estimated 1 to 40% yield loss of rice due r to this snail. P. canaliculata was confirmed as an intermediate host of the rat lungworm, Angiostrongylus cantonensis Chen (Namatoda, Metastrongylidae), causing the disease eosinophilic meningoencephalitis in humans in Taiwan (Chen 1985) and in Japan (Nishimura et al. 1986, Nishimura & Sato 1986).

Plates I, II. Pomacea snails and damage to rice plants in the Philippines. A, Egg masses on a concrete wall of water path. B, An egg-mass on rice plants. C, Copulation, f = female; m = male. D, A big female. E, 2 snails. F, Rice field attacked by Pomacea snails. G, Potted 2- and 4-wk old rice seedlings damaged under experimental condition by Pomacea. H, Transplanted 2- and 4-wk old rice seedlings damaged under experimental condition by Pomacea.

lOCffi 1

MJ~~IIrs~~~~s (Golden a~~

2 wk - ofd rice

Mochida: Freshwater Snails Table 2.

1985 1986 1987 1988 1989

Occurrence and crops attacked by Pomacea snails in Japan (ha).

Rice

Lotus

Taro

51 172 9,786 10,212 16,122

0 3 68 56 59

* 7 10 10

Matrush Wild Rice 1 0 5 5 4

0 0 * * *

Jpn Parsley

Others

Occurrence

0 0 0 0 *

0 0 30 * 0

3,774 6,168 9,896 10,283 16,195

*Damaged but less than one ha.

Biology

Observations on the biology of the snails and crop damage by them were reported in Taiwan by K. M. Chang (1985), W. C. Chang (1985), MAF (1985a b, 1986); in Japan by Nishiuchi (1984), Hamada & Matsumoto (1985), Hirai et al. (1986), Kaneshima et al. (1986), Miyahara et al. (1986a, b), Oya et al. (1986); and in the Philippines by IRRI (1987), Mochida (1987), Saxena et al. (1987), Adalia & Morallo-Rejesus (1989). Guidelines for culture of Pomacea spp. in the Philippines were given by Manacop ( 1986), whereas general information on the snail and crop damage were shown by Miyahara (1987) and Mochida (1988a, b). Eggs hatch in about 3 weeks after oviposition depending on temperature. Young snails fall into water from the egg clusters deposited above the water surface. Hatchability is 7-90%. Snails reach maturity in 2 months and are about 3 em in height. Copulation (Plate IC) takes place in water and females oviposit an average of 321 eggsjeggmass on plants, concrete walls or stones above the water surface at night (Plate I A, B). A female deposits an average of 4,375 (range 2,410-8,680) eggs per year. The snails are estimated to survive for 2-3 years in Japan (Miyahara et al. 1986a) and 3-5 years in Taiwan (K. M. Chang 1985). Mortality is high at water temperatures above 32째C. The snails can survive for 15-20 days at 0째C, 2 days at -3째C, but only 6 hours at -6째C. In Okinawa, Japan, it has been confirmed that the snails can survive 234 days without water. Snails less than 1.3 em in height are not capable of damaging rice seedlings, whereas those more than 1.5 em can feed on seedlings in water. At water depths less than 1 em, rice seedlings are hardly damaged (Miyahara 1987). It was confirmed at IRRI that two-week-old TN1 rice seedlings were eaten out by 20 snails of 3.8-4.3 em in height whereas four week old plants were less damaged, and six week old plants were hardly damaged at all (Plate II G, H). Current Control Methods and Further Directions

The aquaculture of Pomacea for commercial food purposes eventually resulted in snails escaping into the surrounding environment and establishing in rice and other crop fields, ponds and streams. At present, large scale control of Pomacea is too expensive although their temporary eradication from small fields

Micronesica Suppl. 3, 1991

Figure 2.

Spread of Pomacea snails in Japan.

1000

KM,

Mochida: Freshwater Snails

of rice, lotus, swamp cabbage and fish ponds may be achieved with chemicals. However, specimens get reintroduced from adjacent areas in irrigation waters the following season. In Japan and Taiwan many trials were conducted involving hand picking, chemicals, and sand pumps without much success. Plant quarantine regulation seems to be the best method to prevent introduction of Pomacea snails into rice growing countries. The methods recommended in Taiwan are: 1. Pick up and crush egg masses and snails. If possible use them as feed for ducks, chickens, and fish. 2. Burn the straw after harvest in the rice fields where snails are a serious problem. 3. Place 5-mm mesh metal screens at the irrigation water inlets of rice and other fields with aquatic plants. 4. Application of Molluscicides: a. in rice fields: i) triphenyl-tin acetate 45% WP at 0.6 kgjha at water temperatures higher than 20°C and at 1.2 kgjha at lower temperatures; do not drain water for 3 days and keep water at 1 to 3-cm depth for 7 days; do not use this chemical close to fish ponds and streams; protect skin from the chemical to avoid rubefaction. ii) metaldehyde 80% WP at 1.2 kgjha at water temperatures higher than 20°C. b. in waterways, streams, and ponds: i) metaldehyde 80% WP at 2.4 kgjha. 5. Release fingerlings ofblack carp M. piceus and common carp C. carpio in ponds, rivers, and streams. 6. Place 2 kinds of metal screens (6-10 and 16-32 mm mesh) at each water inlet and outlet of Pomacea infested fish ponds, aquatic field crops like swamp cabbage and taro. In Japan, triphenyl-tin acetate/chloride/hydroxide, niclosamide (Bayluscide®), and other chemicals with high fish toxicity are prohibited for use in rice fields. Of the 32 chemicals tested in 1986 (MAFF, 1987), as of June 30, 1990, only ethylthiometone + thiocyclam (Ekamat® 5G), cartap (Padan ®G), bensultap (Ruban® 4G), and IBP (Kitazin P ® 17G) were registered as repellants. IBP G was toxic. Calcium cyanamide is registered as a fertilizer with toxicity. Even though application of 200-400 kgjha is recommended 3 days after the first plowing, calcium cyanamide 200 kgjha is preferred in practice as 400 kg/ha provides too much N. A metal screen ( 5 mm mesh) should be placed over water inlets to prevent snail introduction. Picking snails by hand during land preparation and 5-7 days after mechanical transplanting, and destroying pink eggmasses are also recommended. The mortality in winter was 81% in fields plowed by rototiller from January to February and 57% in fields unplowed in Miyazaki prefecture (MAFF 1987). Metaldehyde 6% is effective when broadcasted at 40-50 kgjha ( = 2.4-3.0 kg aijha) or baited at 10 kg (=0.6 kg aijha) by mixing at 2:2:1 by weight with Irish potato and wheatflour. Keeping 1 em water in the field is effective in protecting rice seedlings from snail damage but this method has problems in practice under monsoon conditions in Asia.

Table 3.

LC50/95 values (ppm, ai) of 4 chemicals for golden snail (P. canaliculata) and nile telapia (Oreochromis niloticus), IRRI, 1986-1987 (Mochida, unpublished). Telapia 1

Snail Male LC50

Chemical

Lab 4 ethoprop metaldehyde niclosamide triphenyl-tin acetate

10%0 6%0 0.5 70%WP 0.6 60%WP 0.1

Field 5 ethoprop metaldehyde niclosamide triphenyl-tin acetate

10%0 6%0 70%WP 60%WP

1 2

3 4

Female LC50 48

Male LC95 48

Female LC95 48

LC50 24

96 hrs

(")

'"I

0.4 0.3 0.1>

17.9 0.3 0.3 0.1

15.2 0.4 0.2 0.1>

0.6 1.0 0.2

0.5 0.6 0.2>

75.8 1.3 0.6 0.5

43.8 0.9 0.6> 0.5>

::s

(1)

1.2-1.5 0.8 0.1>

0.6-0.8 0.7 0.06

0.6-0.8 0.5 0.007

0.6-0.8 0.4 0.002

(i " ~

en s::

~ ~Vol

0.8 0.2 0.2

0.8 0.2 0.1

1.8 0.7 0.3

1.5 0.7> 0.2

Avg size was 5.2 em in body length and 3.0 gin fresh weight. 10 fish X 3 repl. Avg size was 4.2 em in altitude and 13.4 g in fresh weight. 20 snails X 3 repl. Avg size was 4.1 em in altitude and 13.0 g in fresh weight. 20 snails X 3 repl. Tested in plastic container with clean water and air pump at 10 fish/1 0 lit and 20 snails/20 lit water. Tested in 1 m X 1 m iron-framed plots with 21-day-old TNl seedlings with irrigated water of 5 em depth.

>0.8 0.4> 0.6

>0.8 0.4> 0.5

>0.8 0.4> 0.1

::0 ~

Mochida: Freshwater Snails Table 4.

Chemicals sold for snail control in the Philippines (FAO, 1989) kg aijha

Common Name

Brand Name

TPTA*

Brestan C

0.2

TPTC* TPTH* niclosamide endosulfan

Aqua tin Telustan Bayluscide Thiodan Endox Endosulfan

0.2 0.18 0.2 0.75

Immediately after transp. or seeding or needed

*Banned in Jan. 1990. Table 5.

Laboratory test of pyridaphenthion (Ofunack速) for Pomacea snail control in clean water (Mitsui Toatsu Co., unpublished)

Formul. (%,g)

Dosage (kg/ha)

No. snails tested

0.5 1.0 2.0 4.0 6.0 control or untreated

30 30 30 30 30

15 15 15 15 15

Table 6.

Mortality 7 DT (%) 0.0 26.7 100 100 100 0.0

Fish toxicity of LC50 (ppm, 48 hrs) of 6 chemicals with killing effect on Pomacea canaliculata. Common carp C. carpio

pyridaphenthion (Ofunack) endosulfan (Thiodan, etc.) niclosamide* (Bayluscide) TPTA (fentin acetate) (Brestan) TPTC (fentin chloride) (Aquatin) TPTH (fentin hydroxide) (Telustan)

12.0 0.0072 0.235 0.064 0.055 0.050

Killifish Oryzias latipes (Temminch et Schlegel)

Goldfish Carassius auratus (Linne)

10.0

10.0 0.279

0.056

*Formerly known as clonitralid.

In the Philippines 4 chemicals other than 32 tested in Japan were tested at IRRI in 1986 (IRRI 1987). LC50 (48 hrs) for females was 0.1, 0.3, and 0.3 ppm in the laboratory tests and 0.2, 0.2, and 0.8 ppm in the field tests for triphenyltin acetate (Brestan速), niclosamide (Bayluscide速), and metaldehyde (Namekill速),

Micronesica Suppl. 3, 1991

respectively. LC50 (24 hrs) for nile tilapia (Saratherodon niloticus (Linne)) was 0.1, 0.8, and 1.2-1.5 ppm in the laboratory tests and 0.6, 0.4, and 0.8 ppm in the field tests, respectively. Ethoprop (Mocap速) was not effective on snails (Table 3). Tetraphenyl-tin acetate/chloride/hydroxide, niclosamide, and endosulfan were the 5 chemicals sold for Pomacea snail control in the Philippines (Table 4). However, chemicals containing tin have been banned since 1990. Pyridaphenthion (Ofunack速) showed promising results (Table 5). The toxicity of pyridaphenthion to fish was lower than the 5 other chemicals sold in the Philippines (Table 6). Morallo-Rejesus et al. (1989) conducted cultural, environmental, chemical and biological (using ducks) control methods for the control of Pomacea snails in the Philippines, but did not find any definitively effective control method. Van Dinther (1973), Van Dinther & Stubbs (1963) reported that 2 kinds of snails, Pomacea dolioides (Reeve) and P. glauca (Linne), are pests of rice, especially in seedbeds and direct-sown fields in Surinam, South America. They mentioned that copper sulphate, BHC, niclasamide, sodium pentachlorophenate (Na-PCP), triphenyl-tin acetate, and N-trityl morpholine (Frescon 速), were effective on the snails but all of them were not practically used for several reasons including their high toxicity to fish and other fauna. Na-PCP is recommended in rice fields where Pomacea snails are serious pests along the Sao Francisco in Brazil (de FreidasMachado 1953). Effective pesticides, either oil-based or botanical, or both with less toxicity to non-target organisms should be evaluated. References Adalla, C. B. & B. Morallo-Rejesus. 1989. The golden apple snail, Pomacea sp., a serious pest of lowland rice in the Philippines. BCPC Monograph 41: 41 7422. Chang, K. M. 1985. Agricultural pest of apple snails in Taiwan [in Japanese]. Newsl. Malacol. Soc. Jpn. 16: 1-7. Chang, W. C. 1985. The ecological studies on the Ampul/aria snail (Cyclophoracea: Ampullaridae) [in Chinese, English summary]. Bull. Malacol. [Taiwan] 11: 43-51. Chen, E. R. 1985. Zoonotic importance ofAngiostrongylus cantonensis in Taiwan. [Abstr.] First Seminar on parasitic diseases, Sept. 30 - Oct. 4, Taipei. p. 29. de Freitas-Machado, 0. 1953. Estranha praga do arroz no baixo Sao Francisco. Lavoura Cria~ao 8: 20. FAO. 1989. Integrated "Golden" kuhol management. 41 pp. Habe, T. 1986. Japanese and scientific names of the apple snail introduced from South America [in Japanese]. Newsl. Malacol. Soc. Jpn. 17: 27-28. Hamada. Y. & T. Matsumoto. 1985. [Jumbo snails in Kumamoto prefecture, Japan] [in Japanese]. Kyuusyuu no Kai 24: 5-12. Hirai, Y., S. Oya & Y. Miyahara. 1986. Population census of the apple snail, Ampullarius insularus, in a paddy field [in Japanese, English summary]. Proc. Assoc. Pl. Prot. Kyushu. 32: 88-91.

Mochida: Freshwater Snails

IRRI. 1987. Annual Report for 1986, pp. 337-340. International Rice Research Inst., Manila. Kaneshima, M., S. Yamauchi & K. Riga. 1986. Sexual maturity ofthe apple snail, Apullarius insularus [in Japanese]. Proc. Assoc. Pl. Prot. Kyushu. 32: 101103. MAF [Ministry of Agriculture and Forestry], Taiwan. 1985a. [Control of apple snails] [in Chinese]. Taiwan Prov. Government, Taipei, Taiwan, China, 2 pp. MAF [Ministry of Agriculture and Forestry], Taiwan, 1985b. [Report on Pomacea snails; their occurrence, damage, and control for 1985] [in Chinese]. Taiwan Prov. Government, Taipei, Taiwan, China. 6 pp. MAF [Ministry of Agriculture and Forestry], Taiwan. 1986. [Comprehensive report on the control of Pomacea snails for 1986] [in Chinese]. Taiwan Prov. Government, Taipei, Taiwan, China. 9 pp + 8 tables. MAFF [Ministry of Agriculture, Forestry, and Fisheries], Japan. 1987. [Results on the experiments of the control of the apple snail, Pomacea canaliculata (Lamarck), with chemicals.] [in Japanese]. Japan Plant Protection Association, Tokyo, Japan. 187 pp. Manacop, P. R., Sr. 1986. The breeding and culture of the golden miracle snail. Los Banos, Philippines. 15 pp. Miyahara, Y., Y. Hirai & S. Oya. 1986a. Occurrence of Ampullarius insularus d'Orbigny injuring lowland crops [in Japanese]. Syokubutu-Booeki. 40: 3135. Miyahara, Y., Y. Hirai & S. Oya. 1986b. Oviposition and hatching rate of Ampullarius insularus d'Orbigny in Kyushu [in Japanese, English summary]. Proc. Assoc. Pl. Prot. Kyushu. 32: 96-100. Miyahara, Y. 1987. Biology and damage ofthe apple snail, Pomacea canaliculata (Lamarck) [in Japanese]. Takeda Yakuhin Koogyoo Co., Tokyo, Japan. 22 pp. Miyazaki, J. 1985. On Japanese snail of Ampullarius insularus d'Orbigny [in Japanese]. Newl. Malacol. Soc. Jpn. 16: 28-29. Mochida, 0. 1987. Pomacea snails in the Philippines. Inti. Rice Res. News!. 12(4): 48-49. Mochida, 0. 1988a. Nonseedborne rice pests of quarantine importance. In IRRI, Rice seed health, pp. 11 7-129, Los Banos, Philippines. 362 pp. Mochida, 0. 1988b. The rice water weevil (Lessorhoptrus oryzophilus Kuschel) and the freshwater snail (Pomacea canaliculata (Lamarck)) as important pests of crops for plant quarantine in Asia. In ASEAN PLANTI, Movement of pests and control strategies, pp. 71-80, Selangor, Malaysia. 423 pp. Morallo-Rejesus, B. M, A. S. Sayabac & R. C. Joshi. 1989. The distribution and control of the introduced golden snail (Pomacea sp.) in the Philippines. In ASEAN PLANTI, Introduction of germplasm and plant quarantine procedures, pp. 213-224, Selangor, Malaysia. 368 pp. Nishimura, K., M. Mogi, T. Okazawa, Y. Sato, H. Toma & H. Wakibe. 1986. Angiostrongylus cantonensis infection in Ampullarius canaliculatus (La-

Micronesica Suppl. 3, 1991

marck) in Kyushu, Japan. Southeast Asian J. Trop. Med. Pub. Hlth. 17: 595600. Nishimura, K. & Y. Satao. 1986. Natural infection with Angiostrongylus cantonensis in Ampullarius canaliculatus (Lamarck) in the Ryukyu, Japan. Jpn. J. Parasitol., 35: 469-470. Nishiuchi, Y. 1984. Topics of "Jumbo Snail" [in Japanese]. Agchem Age. 150: 47-51. Oya, S., Y. Hirai & Y. Miyahara. 1986. Injuring habits of the apple snail, Ampullarius insularus d'Orbigny, to the young rice seedlings [in Japanese, English summary]. Proc. Assoc. Pl. Prot. Kyushu. 32: 92-95. Saxena, R. C., A. V. de Lara & H. D. Justo, Jr. 1987. Golden snail, a pest of rice. Intl. Rice Res. Newsl. 12: 24-25. van Dinther, J. B. M. 1973. Molluscs in agriculture and their control. World Crops (1973): 282-286. van Dinther, J. B. M. & R. W. Stubbs. 1963. Summary of research on the control of rice snails in Surinam. Surinam Agric. Exp. Sta. Paramaribo Bull. 82: 415420.

Micronesica Suppl. 3: 63-69, 1991

Brown tree snake (Boiga irregularis) on Guam: a Worst Case Scenario of an Introduced Predator MICHAEL JAMES McCoin

Department of Agriculture, Division of Aquatic and Wildlife Resources, P. 0. Box 2950, Agana, Guam 96910 U.S.A .

Abstract-Boiga irregularis was probably introduced on Guam a little over forty years ago. Since then, the snake population has become very dense and has been responsible for enormous biological, economic, and cultural damages. Because of severe reductions in vertebrate biodiversity, changes in insect densities and diversity and vegetational shifts should occur. Introduction The brown tree snake (Boiga irregularis) was probably introduced to Guam, Mariana Islands, in the late 1940's (Fritts 1988) via surface cargo movements of surplus U. S. military equipment. The most likely source population is Manus Island (Admiralty Islands), north of Papua New Guinea (T. Fritts, pers. comm.). Reports of B. irregularis four to eight feet in length began in the early and mid 1950's (Elvidge 1955). These sightings were from the vicinity of the village of Santa Rita, adjacent to the U. S. Naval port facility, and probably represented several generations of snakes (Fritts 1988). Until relatively recently, there was little documentation of the spread of B. irregularis on Guam. This may be due to the secretive behavioral characteristics of the snake (nocturnal and arboreal). However, during the 1960's and 1970's, the avifauna underwent significant spatial and density changes. Two causative agents were explored: disease and pesticides, but both failed to explain the observed avifauna! changes (Grue 1985, Savidge 1986). Savidge (1987a) demonstrated that B. irregularis was the primary culprit in the demise of virtually the entire native forest-dwelling avifauna. This research spawned numerous popular articles (Anonymous 1989, Carey 1988, McCoid 1989a, Montgomery 1988, Pimm 1987, Quammen 1985, Rauzon 1989, Savidge 1987b, and Teodosio 1987) and opened the door for continued research into the remarkable role which the introduction of a single species of vertebrate has had on Guam. The impact of the brown tree snake on Guam extends beyond that of the conspicuous biological impact. Significant negative effects have also been observed in virtually all aspects of life on Guam (cultural, economic, and health). All these negative impacts can be examined on two levels: primary and secondary effects.

Micronesica Suppl. 3, 1991

Primary Effects BIOLOGICAL

Native vertebrate biodiversity has been heavily impacted. Eight species of forest-dwelling birds have been extirpated from Guam due primarily to B. irregularis predation (Table 1) (Aguon 1991). Three additional species of native forest-dwelling birds persist on Guam, but have suffered serious population and range reduction (Table 1). Five species of introduced birds have shown either declining or stationary population levels (Table 1) (Beck et al. 1988). Wiles (1987) indicated that B. irregularis may be a factor in the decline of the native frugivorous, nectivorous Mariana fruit bat (Pteropus m. mariannus). Five species of introduced small mammals occur on Guam (three rats, Rattus rattus, R. norvegicus, and R. exulans; a mouse, Mus musculus; and a shrew, Suncus murinus) and were formerly widespread. While no empirical data have been published, these introduced mammals currently appear to be relatively restricted to urban areas and may be at lower population densities than were formerly exhibited (pers. obs.). Savidge (1988) showed that B. irregularis regularly consumed these introduced mammals. The snake may be a factor in the spatial restriction of these species on Guam. Larger snakes now occur in urban as compared to non-urban locations and snakes greater than 125 em snout-vent length are almost always urban (M. McCoid, unpub. data). These phenomena may be related to prey availability. Besides the avian and mammalian reductions in population and diversity, evidence suggests that B. irregularis has also negatively impacted at least two Table 1. List of native and introduced urban and forest-dwelling birds that have been impacted by B. irregularis. 1 =extinct, 2=extirpated on Guam, other populations exist, 3=extinct in the wild, captive populations exist, 4=rare to uncommon on Guam, other populations exist, and 5= introduced. Species Rallus owstoni Ptilinopus roseicapillus Gallicolumba x. xanthonura Halcyon c. cinnamomina Rhipidura rufifrons uraniae Myiagra freycineti Zosterops c. conspicillata Myzomela cardinalis Corvus kubaryi Aplonis o. opaca Aerodromus vanikorensis bartschi Columba Iivia Streptopelia bitorquata Passer montanus Lonchura malacca Dicrurus macrocerus

Common name

Status

Trophic Level

Guam Rail Mariana fruit dove white-throated ground dove Micronesian kingfisher rufous-fronted fantail Guam flycatcher bridled white-eye cardinal honeyeater Mariana crow Micronesian starling island swiftlet

2 4 4 4

omnivore frugivore frugivore insectivore/carnivore insectivore insectivore insectivore nectivore/insectivore omnivore omnivore insectivore

rock dove Philippine turtle dove Eurasian tree sparrow chestnut mannikin black drongo

5 5 5 5 5

granivore granivore granivore granivore insectivore

3 2 2 3 1 1 1

McCoid: Brown Snake on Guam

species of native reptiles. Sabath (1981) recorded both Perochirus ateles and Gehyra oceanica as conspicuous members of the herpetofauna of Guam in 196869. Since then, both species have been extirpated from Guam. Both geckos persist in relatively large numbers on Cocos Island, a barrier reef island 2.5 km SW of Guam. Cocos Island remains free of B. irregularis at this time. CULTURAL

Until the introduction of B. irregularis, the only other snake in the Mariana Islands was the blind snake, Ramphotyphlops braminus. This latter species is a small, fossorial, rarely encountered form that is commonly mistaken for an earthworm. For this reason, the Mariana Islands were characterized by the indigenous peoples as snake-free. The infestation and proliferation of B. irregularis has resulted in concerns about many aspects of life on Guam. Accommodations in human life styles are having to be made to prevent bites (see below), provide a snake-free habitation and workplace, and protect pets and livestock (see below), all of which, until forty years ago, were not necessary. ECONOMIC

These nocturnal and highly arboreal snakes routinely climb guy-wires on power poles, cross transmission lines, and enter transformers. Because of these behaviors, the brown tree snake has been responsible for enormous economic damage due to power outages (Fritts et al. 1987). These problems continue to the present. Fritts & McCoid ( 1991) showed that almost 50% of the people that raised domesticated fowl were aware of some level of predation by brown tree snakes. It is difficult to assign a monetary value to this type of predatory pressure, but if a subsistence farmer is relying on chickens for a portion of his family diet, then the effects could be economically detrimental. Both Fritts ( 1988) and Fritts & McCoid ( 1991) also found that pets and cage birds suffered mortality due to B. irregularis. HEALTH

The genus Boiga is recognized as rear-fanged and venomous (Cogger 1988). While the potential for serious problems associated with a bite from B. irregularis have been recognized (Fritts 1988), it wasn't until very recently that the effects of bites were documented (Fritts et al. 1990). The responses to bites appear to be anaphylactoid and in two of the cases involving human infants, had they not been at a hospital, the likelihood of death existed (Fritts et al. 1990). Since the reaction to the bite is anaphylactoid, the possibility exists that certain adults may also be more susceptible to severe reactions. Secondary Effects At this time, secondary effects are difficult to discern. This is due to a number of confounding factors. For example, few historical comparative or baseline data

Micronesica Suppl. 3, 1991

exist on insect and plant abundances and distributions on Guam. Moreover, Guam is currently experiencing enormous economic development. Large tracts of land are undergoing conversion to urban situations, further clouding the documentation of the distribution and biodiversity of the existing fauna and flora. Despite these factors, some predictions can be made regarding the long-term effects of altering the biodiversity of an island ecosystem. FAUNAL CHANGES

Table 1 lists the avifauna that have been impacted by the introduction of B. irregularis and their trophic roles. Six of the eight extirpated species of birds were either wholly or partially classified as insectivores. The only remaining avian insectivore (A. vanikorensis) has a population estimate of approximately 450 (G. Wiles, pers. comm.). The two native omnivorous birds that remain on Guam have low (A. opaca, n = 500) to very low (C. kubaryi, n =50) populations and are distributionally localized (G. Wiles and J. Guerrero, pers. comm.). The introduced birds include only one insectivore (D. macrocerus) and this species currently exhibits only localized distribution and low population levels (G. Wiles, pers. comm.). Based on population estimates supplied by Aguon ( 1991) (and references therein) for the native insectivorous or omnivorous birds that are extinct, extirpated, or at low levels, and assuming that Guam harbored at least as many birds as islands of smaller surface area in the northern Mariana Islands, a conservative estimate of the total number of insectivores/omnivores that are missing from Guam is 300,000 individuals. Additionally, an unknown number of insectivorous geckos (see above) are missing from Guam. Ifthis scenario is correct, then considerable increases or changes in insect abundance and diversity should be expected. FLORAL CHANGES

If a plant community is dependent upon frugivorous, nectivorous, or granivorous vertebrates to enhance the continuity of that community by providing mechanisms for seed dispersal (postzygotic) or pollination (prezygotic), then the prediction on Guam would be that floral changes should also be expected. Table 1 shows that the two species of native birds that were frugivores no longer occur on Guam. Additionally, four species of introduced granivores have restricted ranges on Guam and occur in low densities (Becket al. 1988). Potential pollinators were M . cardinalis, Z. conspicillata, and P. m. mariannus remains (n =400-600; G. Wiles, pers. comm.). While Z. conspicillata is classified as an insectivore, this species frequently foraged in flowers and may have acted as a pollinator. These facilitators in the maintenance of floral diversity and abundance are no longer functioning at the levels that they were. Because of this, species of plants that utilized these dispersal mechanisms on Guam should exhibit contractions in distribution. This may be difficult to test because of herbivory by introduced vertebrates, such as deer, pigs, and goats. However, both Muniappan (1988) and Denton et al. (1991) postulated that the range restriction of Lantana

McCoid: Brown Snake on Guam

camara on Guam was due to the absence of frugivorous birds. This might be the first demonstrable vegetational effect of the removal of a facilitator of dispersal by B. irregularis predation.

Future of Boiga on Guam It is likely that B. irregularis will persist on Guam indefinitely. This is despite the facts that almost all native birds have been extirpated and introduced small mammals have been restricted to primarily urban areas. The majority of B. irregularis on Guam are three to four feet (90-120 em) in total length (Fritts 1988). Savidge ( 1988) demonstrated that this size and smaller were consuming primarily lizards, which McCoid ( 1991) found belonged primarily to three introduced species. The reproductive characteristics of lizards on Guam, both native and introduced, indicate that these species have sustainable yield as prey for B. irregularis (McCoid 1989b). Because of this, Guam will remain a source population for further infestations in the Pacific Basin.

Inter-island Movements Boiga irregularis has proven adept at moving from Guam to other islands. Fritts (1987, 1988) reported that B. irregularis has shown up on Oahu, Wake Island, Kwajalein Island, Diego Garcia Atoll, and Saipan (M. McCoid & D. Stinson, unpub. data). Sightings (but no specimens) of snakes are also reported to Tinian, Kauai (Telfer 1989), and Pohnpei. These inter-island movements are occurring via civilian and military aircraft and surface cargo shipments because Guam is a commercial and military shipping hub. The economic development that the Pacific Basin is currently enjoying will probably continue into the 21st century and will generate continued high levels of commercial and military intercourse. Because of this, off-island movements of B. irregularis will continue to occur and will increase the likelihood of infestation on islands that currently engage in military and economic traffic with Guam.

Control and Containment At this time, no prescribed chemical or mechanical methods exist to prevent infestation of cargo or aircraft by B. irregularis. Currently, the most powerful weapon in the arsenal against the brown tree snake is education. The Government of Guam (Division of Aquatic and Wildlife Resources), U.S. Department ofthe Interior (Fish and Wildlife Service), and the U. S. Department of Defense are actively engaged in training of personnel to intercept B. irregularis in high-risk situations (port and airport facilities) on Guam and other islands in the Pacific community judged to be at risk for infestation. Recent evidence suggests that these efforts may be successful. The B. irregularis from Saipan (M. McCoid and D. Stinson, unpub. data) was found by an airline employee who was aware of the threat because of educational materials funded by the U. S. federal monies.

Micronesica Suppl. 3, 1991

Within the past two years, 15 snakes have been intercepted by U. S. military security or customs personnel in high-risk situations on Guam (in or on aircraft, in hangers, on runways, or on cargo) (Petty Officer A. Hill, U.S. Military Customs, COMNAVMAR, personal communication). All military customs personnel assigned to Guam are required to include a seminar on B. irregularis in their training regimen. It is clear though, that these efforts will not be enough to prevent the spread of the brown tree snake to other islands in the Pacific. Population densities of B. irregularis are high (T. Fritts, pers. comm.) and it is unlikely that all snakes will be intercepted. Because of this, the next phase of brown tree snake control should involve the development of chemical and mechanical means for more effective control and local eradication in specific high-risk sites. Acknowledgements I thank A. Hill for sharing information on U. S. military efforts in snake control. Both R. Hensley and C. Aguon read and commented on this manuscript. Co-workers R. Beck, G. Wiles, and J. Guerrero shared their thoughts. Support for this paper was provided by the Endangered Species Conservation Program, Project E-4. References Anonymous. 1989. Tree snake could scourge all islands. Pac. Mag. 14(2):49. Aguon, C. F. 1991. The birds of the Mariana Archipelago: factors in their demise. Proc. Natl. Park Serv. Workshop. In press. Beck, R. E., Jr., G. J. Wiles, & M. J. Ritter. 1988. Native bird status surveys and natural history. Div. Aquat. Wildl. Res. FY 1988 Ann. Rep. pp. 105-116. Carey, J. 1988. Massacre on Guam. Natl. Wildl. August-September 1988: 13-15. Cogger, H. G. 1988. Reptiles and Amphibians of Australia. 4th ed. Reed Books Pty. Ltd. NSW, Australia. 688p. Denton, G. R. W., R. Muniappan & M. Marutani. 1991. The distribution and biological control of Lantana camara in Micronesia. Micronesica Suppl. 3: 71-81. Elvidge, F. Q. 1955. Annual report of the Governor of Guam. 58p. Fritts, T. H. 1987. Movements of snakes via cargo in the Pacific region. 'Elepaio 47: 17-18. Fritts, T. H. 1988. The brown tree snake, Boiga irregularis, a threat to the Pacific islands. U. S. Fish Wildlife Serv., Biol. Rep. 88(31). 36p. Fritts, T. H. & M. J. McCoid. 1991. Predation by the brown tree snake on poultry and other domesticated animals on Guam. Pies. in press. Fritts, T. H., N.J. Scott, Jr. & J. A. Savidge. 1987. Activity ofthe arboreal brown tree snake (Boiga irregularis) on Guam as determined by electrical outages. The Snake 19: 51-58. Fritts, T. H., M. J. McCoid & R. L. Haddock. 1990. Risks to infants on Guam from bites of the brown tree snake (Boiga irregularis). Amer. J. Trop. Med. Hygen. 42: 607-611.

McCoid: Brown Snake on Guam

Grue, C. E. 1985. Pesticides and the decline of Guam's native birds. Nature 316: 301. McCoid, M. J. 1989a. Snake in the grass. R&R Pac. Mag. 4(6): 14,27,30. McCoid, M. J. 1989b. Biology of the brown tree snake. Div. Aquat. Wildl. Res. FY 1989 Ann. Rep. in press. McCoid, M. J. 1991. Exotics in paradise: an island in transition. Proc. Natl. Park Serv. Workshop. in press. Montgomery, S. 1988. Guam's serpentine interlopers. Massachusetts SPCA J. September-October 1988: 28-31. Muniappan, R. 1988. Biological control of the weed, Lantana camara in Guam. J. Pl. Prot. Tropics 5: 99-101. Pimm, S. L. 1987. The snake that ate Guam. Trends Ecol. Evol. 2: 293-295. Quammen, D. 1985. Island Getaway. Outside Mag. October 1985: 19,21,23,25. Rauzon, M. 1989. Year of the serpent. Islands 9(6): 29-30. Sabath, M. D. 1981. Gekkonid lizards of Guam, Mariana Islands: reproduction and habitat preference. J. Herp. 15: 71-75. Savidge, J. A. 1986. The role of disease and predation in the decline of Guam's avifauna. Ph.D. dissertation, Univ. Illinois, Champaign-Urbana. Savidge, J. A. 1987a. Extinction of an island forest avifauna by an introduced snake. Ecology 68: 660-668. Savidge, J. A. 1987b. Death on an island. Living Bird Quart. Winter 1987: 6-10. Savidge, J. A. 1988. Food habits of Boiga irregularis, an introduced predator on Guam. J. Herp. 22: 275-282. Telfer, T. 1989. Snake(s) on Kauai? Hawaii Wildl. Newsl. 4(3): 8. Teodosio, R. 1987. Tree snake brings Guam blackouts. Pac. Mag. 12(6): 42. Wiles, G. J. 1987(1988). The status of fruit bats on Guam. Pac. Sci. 41: 148-157.

Micronesica Suppl. 3: 71-81, 1991

The Distribution and Biological Control of Lantana camara in Micronesia G. R. W. DENTON,* R. MuNIAPPAN and M. MARUTANI College of Agriculture and Life Sciences, University of Guam, UOG Station, Mangilao, Guam 96923

Abstract-Lantana camara L. was introduced into Micronesia as an ornamental plant at the turn of the century and quickly established as a significant weed pest throughout the region. Over the past forty years various natural enemies of lantana have been introduced into Micronesia in an attempt to curb the spread of the lantana. However, little information has come out of the area with respect to the establishment and effectiveness of these species. For this reason, surveys were undertaken to assess the status and natural insect enemies oflantana on major islands within the Caroline and Mariana island groups. Lantana was found to be widespread throughout the region although its abundance and pest status varied between islands. At least four different color varieties were noted, the most common of which resembled the Hawaiian Pink, L. camara var. aculeata. Seven of the thirteen exotic insect species introduced into the area were found to have established, although not all were found on all islands where lantana was present. Similarly, certain insects were more effective bio-control agents than others although once again inter-island variability was evident. The five most important insect species, in terms of their distribution and effectiveness, were the tingid bug, Teleonemia scrupulosa Stal; the hispine beetle, Uroplata girardi Pic; the pterophorid moth, Lantanophaga pusillidactyla (Walker); the tortricid moth, Epinotia lantana (Busck), and the pod-fly Ophiomyia lantanae (Froggat). Introduction

The perennial shrub, Lantana camara L., is a member of the Verbenaceae family and a native of the tropical and sub-tropical Americas. Several highly colored and attractive varieties were first introduced as ornamental plants to other parts of the world during the late seventeen hundreds. Certain vigorous growing varieties eventually escaped their garden confines and, in the absence of their natural enemies, flourished in their new environments. Unfortunately their *Present Address: Water & Energy Research Institute, University of Guam, UOG Station, Mangilao, Guam 96923

Micronesica Suppl. 3, 1991

tendency to form impenetrable thorny thickets, invade pasture land, and compete with agricultural crops for water, nutrients and light, quickly established them as a serious weed pest in many areas. Indeed, lantana has been reported in 47 countries (Holm et al. 1977) and is considered to be among the ten worst weeds in the world (Holm & Herberger, 1969). Lantana is widely distributed throughout the Pacific (Thaman 1974) although relatively few publications have emerged regarding its current distribution, pest status and control within Micronesia. For this reason, surveys were conducted, between September 1988 and October 1989, on Pohnpei, Chuuk and Kosrae in the Eastern Caroline Islands; Belau and Yap in the Western Caroline Islands; and Guam, Rota, Aguijan, Tinian and Saipan in the archipelago of the Marianas (Fig. 1). The findings of the study are outlined below. Distribution Within Micronesia

Lantana camara was probably introduced into Micronesia by European settlers during the late eighteen and early nineteen hundreds although records are incomplete. Most likely, Guam, the Marshall Islands, Pohnpei and Belau were important entry points for the weed into the region. By the mid 1940's it was considered to be problem weed on several Micronesian islands (Schreiner 1989). During the current study, lantana was found on all islands visited except Rota and Kosrae. For how long these islands can remain free of lantana is uncertain. Given the close proximity of neighboring islands, where lantana is found, it would seem only a matter of time before the weed is introduced unless the appropriate regulatory authorities act to ensure otherwise. Pest Status

Where present, lantana was found to be more abundant on some islands than others (Table 1). For example, it was extremely common and widespread on the uninhabited island of Aguijan. Likewise, on the neighboring islands of Tinian and Saipan, large expanses of lantana were found in the southern portion of each island and were of concern to local graziers. On all other islands visited, lantana was not considered to be a major weed pest. On the contrary, the general feeling among the local peoples was that the pest status of lantana had declined in recent years following the introduction of some of its natural enemies into the area. Such inter-island differences in the pest status of lantana are due, at least in part, to differences in agricultural and land use practices in addition to variations in soil type, climate, and the abundance and effectiveness of the various natural enemies introduced into the area. Equally important are the many varied and different interactions between lantana and other plant and animal species sharing the same environment. On Aguijan, for example, the spread of lantana had been greatly facilitated by the grazing influence of the island's large, endemic goat population. At the same time, however, its growth and vigor was to some extent

140°

(6)

( " ) TINIAN.'SAIPAN 5 (4),'AGUIJAN , ROTA (3 ) GUAM

ENEWETAK ATOLL'<

(2) YAP

• ISLANDS

CHUUK ( 7 ) ISLANDs '.''

( 1 ) /BELAU

• ISLANDS

(S) POHNPEI ., ISLANDS

KOSRAE •

200

400

BOO km

GOO

(2) Cape York (AUSTRALIA) 10

40 km

Common scale for insets

15°15'N

fJ~~'E.

(6)

SAIPAN

(S)

\.f)

~ e

•

f1TINIAN 15°00'N

(4)

~AGUIJAN

·~ (}

W·

ANGAUR

Figure 1. Micronesian Islands surveyed for Lantana camara and associated natural enemies (sampling locations are indicated by closed circles).

Micronesica Suppl. 3, 1991 Table 1. The status of Lantana camara in Micronesia. Island

Status

Western Caroline Islands Belau Yap

++ +

Eastern Caroline Islands Pohnpei Truk Kosrae

+ + A

Archipelago of the Marianas Guam Rota Aguijan Tinian Saipan

+ A ++ ++ +

++=Abundant and widespread and/or of local importance. +=Relatively common but restricted distribution. Generally not of local importance. A=Absent.

hampered, here, by some fierce competition for space between itself and another recent invader, the Siam weed (locally referred to as "masiksik"), Chromolaea odorata (L.) R. M. King and H. Robinson (Asteraceae). The latter weed species was found to varying degrees on all islands visited except Chuuk where it was entirely absent. Both the above examples illustrate direct interactions with other living organisms. Indirect interactions are, of course, less obvious although a noteworthy example tentatively links the relatively recent decline of lantana, on Guam (Fig. 2), to the brown tree snake, Boiga irregularis which was accidentally introduced onto the island sometime shortly after the last world war (Savidge 1987, Muniappan 1988). This species has dramatically reduced the forest avifauna of Guam among which are several species suspected of playing a key role with respect to lantana seed dispersal on the island (Muniappan 1988, Denton et al. in press). Interestingly, this is the first time that evidence (albeit circumstantial) has been presented to illustrate the negative impact of the brown tree snake on Guam flora. Taxonomic Variations

Our surveys revealed at least four distinct color-morphs of lantana, although not all were found together on the same island. The most commonly encountered color-morph was the pink and yellow flowering variety believed to be the Hawaiian Pink taxon, L. camara aculeata. It was found on all lantana-infested islands except Yap. An attractive Orange-red and yellow variety, probably the

Denton et al.: Biological Control of Lantana

1987

1971

Figure 2.

Decline of Lantana camara on Guam between 1971 and 1987.

Hawaiian orange-red taxon, L. camara mista, was found in Palau (Koror), Yap (largest most southerly island only), and Saipan (a single, small bush amidst 2 hectares of established L. camara aculeata located at Koblerville in the south of the island). On Angaur, a tiny island in the south of the Palau island group, two additional color-morphs were found. One was very similar to L. camara aculeata although the corolla of each floret was bordered with orange. The other was a relatively uncommon, light orange and yellow flowering variety similar to L. camara mista although somewhat paler in appearance.

Introduced Natural Enemies

Attempts to control lantana biologically were first pioneered in Hawaii at the turn of the century (Perkins & Swezey 1924) although it was not until the late forties that candidate species of insects were first introduced into Micronesia for this purpose. Of the 35 insects that have been used for the biological control of lantana, thirteen species are known to have been introduced into Micronesia (Table 2). However, relatively little information has subsequently emerged from this area, regarding the fate, distribution and effectiveness of these biological control agents. Our surveys indicated that only seven of the thirteen species established (Table 2). However, not all were represented on all islands where lantana was present and, typically, some species were more effective than others. In addition, the effectiveness of certain species varied between islands and between different colored varieties of lantana. Further details of these findings are outlined below.

Micronesica Suppl. 3, 1991 Table 2.

Natural enemies of Lantana carnara introduced into Micronesia.

Natural Enemy HEMIPTERA Tingidae Leptobyrsa decora Teleonernia scrupulosa

Date First Introduced

Established

1971 1948

No Yes

COLEOPTERA Chrysomelidae Octotorna scabripennis Uroplata girardi Cerambycidae Plagioharnrnus spinipennis

1971 1963

No Yes

1973

DIPTERA Agromyzidae Ophiornyia lantanae

1948

Yes

1955 1958 1955

No Yes No

1948

Yes

1948

Yes

1955 1955

No Yes

LEPIDOPTERA Noctuidae Diastema tigris Hypena strigata Noegalea esula T ortricidae Epinotia lantana Pterophoridae Lantanophaga pusillidactyla Pyralidae Pseudopyrausta acutangulalis Salbia haernorrhoidalis

Teleonemia scrupulosa STAL (TINGIDAE) Both adults and developing nymphs of this bug feed on vascular fluids drawn from the flower buds, growing tips and young leaves of lantana. Damaged florets blacken and eventually wither whilst damaged leaves display chlorosis, progressive necrosis and mild to severe distortion. These symptoms may manifest themselves for some distance from the actual point of attack due to the systemic action of the salivary toxins. Originally from Central and South America, T. scrupulosa was one of four insects shipped to Pohnpei from Hawaii in 1948. It was reported to have established on Pohnpei in 1951 (Pemberton 1954) and in 1956 seemed to be exerting some control (Gardner 1958). It was considered to be only partially effective by 1967 (Schreiner 1989) and during the present survey it was not found anywhere on the island. However, on a visit in 1989 one of us (RM) found a small population of T. scrupulosa feeding on lantana near Ponape Agriculture and Trade School. The reason for the decline in effectiveness of T. scrupulosa on Pohnpei is not clear although we suspect the high annual rainfall (200" spread fairly evenly

Denton et al.: Biological Control of Lantana

throughout the year) is a major factor since the bug does not do well in unusually wet or cold conditions (Harley 1971 ). T. scrupulosa was introduced to Belau from Pohnpei in 1960 but has only established on the pink and yellow lantana found growing there. Both pale and dark orange/yellow flowering varieties (in Angaur and Koror respectively) were completely free of tingid attack indicating host incompatibility-a phenomenon previously reported by Harley & Kassulke ( 1971) for certain color-morphs of lantana from Australia. Host incompatibility also seems to be the most likely reason why a shipment oftingids sent to Yap from Belau in 1962 failed to survive on the orange/yellow variety of lantana found growing there. In 1963 the tingid was introduced to Saipan from Belau and quickly established. From Saipan it was introduced into Guam in 1969 and again in 1971. On both islands it has proved to be a very effective bio-control agent, particularly in open, sunny areas during the dryer months. The bug was also found to be exerting a good degree of control on lantana on Chuuk, Aguijan and Tinian although no records exist of it ever being introduced to these islands. Uroplata girardi PIC (CHRYSOMELIDAE) This beetle feeds exclusively on lantana leaves reducing the plants effective photosynthetic area and upsetting its water balance. Both adults and larvae attack the leaves; the former scarifying the leaf surface and the latter mining the mesophyll layer protected between the upper and lower epidermis. Specimens of U. giarardi were first introduced from Hawaii into Micronesia, via Pohnpei and Saipan, in 1963. Shipments were subsequently transported to Belau from Pohnpei in 1974 and to Guam from Saipan in 1967. The beetle was found on all of these islands at the time of our survey and was found to be exerting light to moderate leaf damage to plants. It was most effective in shaded areas at relatively high humidity. We also found U. girardi on Chuuk where it had, presumably, been introduced from Pohnpei. However, we were unable to locate records documenting the introduction of any natural enemy of lantana to this island. On Belau, the beetle was observed to be equally effective on both the orange/yellow and pink/yellow flowering varieties, unlike the tingid. For this reason, it could prove a useful addition to the suit of natural enemies found on Yap should the need ever arise. The absence of U. girardi from Tinian coupled with the need for additional bio-control measures here prompted us to introduce approximately 700 specimens of the beetle to the island in August 1989. Surveys, made 10 months later, revealed that although the beetle had established, the population was small and ineffective and had spread little from its point of release. Further releases, at more favorable sites on Tinian and on the neighboring island of Aguijan, are scheduled for the near future. Ophiomyia lantanae (FROGGAT) (AGROMYZIDAE) Commonly known as the pod-fly, the larvae of this species excavate tunnels in the developing fruits causing them to shrivel and drop prematurely. 0. lantanae is a native of Central America and was one of the four insects initially introduced

Micronesica Suppl. 3, 1991

into Pohnpei from Hawaii in 1948. It was reported to have established in 1949 (Bryan 1949) although our survey failed to find it anywhere on the island. The deliberate introduction of the fly to Guam from Hawaii, in 1971, was the only other record found of its entry into Micronesia. Despite this, it was also located on Tinian, Saipan and the Belauan island of Angaur during the present survey. On all these islands it was observed to be well established and widely distributed. The numbers of fly infested pods frequently exceeded 50% of total pods produced per flowering head. According to Waterhouse & Norris (1987), the impact of 0. lantanae in reducing the abundance and distribution of lantana remains in question since the larvae only feed on the endosperm and do not affect the seed embryo. However, we have experimental evidence to indicate that fly infested fruits are rendered less attractive to birds (Table 3) which, in turn, must reflect upon lantana seed dispersal and distribution by avifauna. Interestingly, this was originally suggested by Beeson & Chatterjee (1939) over half a century ago.

Lantanophaga pusillidactyla (WALKER) (PTEROPHORIDAE) L. pusillidactyla is one of four surviving lepidopteran species introduced into Micronesia to control lantana. The larvae of this species feed predominantly on the florets of lantana and have, on occasions, been observed to attack the fruits. Thus, they are instrumental in reducing seed production. L. pusillidactyla was among the first insects to be introduced into Pohnpei in 1948 and from there was sent to Belau in 1960. Despite there being no other records of its intentional distribution throughout Micronesia, it was found to be widespread during the present study and was recorded on all islands where lantana was present except Aguijan. Epinotia lantana (BuscK) (TORTRICIDAE) The larvae of this Central American moth aggressively attacks the florets receptacles and developing pods of lantana significantly influencing seed production. It was among the first shipment of insects sent Pohnpei in 1948 and from there would appear to have spread throughout Micronesia in a fortuitous Table 3. The acceptability of Ophiomyia lantanae infested lantana fruits to the Philippine turtle dove, Streptopelia bitorquata (Temmink). % Lantana Fruits Eaten Over 6 Hour Period*

Succulent Fruit

Dry Fruits

Feeding Trial

Non-infested

Infested

Non-infested

Infested

1 2 3

100 100 100 100

42 53 56

34 10 35 26

4 0 4

Average

*Between 50-100 berries/category used in each trial.

Denton et al.: Biological Control of Lantana

manner. Its distribution paralleled that of L. pusillidactyla, with which it ranked among the most effective bio-control agents of lantana in this region. Together, these species have been reported to account for up to 80% decline in berry production in Yap (Muniappan 1989) and Guam (Denton et al. in press).

Hypena strigata (FAB.) {NOCTUIDAE) The larvae of this species are foliar feeders, and have proved to be very effective bio-control agents of lantana in Hawaii (Davis 1959). From here the species was introduced to Pohnpei and Yap in 1959, and to Guam in 1967. However, the promise shown by the species in Hawaii did not manifest itself on these islands. On Pohnpei for -example, the species does not appear to have established and on Yap and Guam its impact is insignificant. We also found it at low levels on Aguijan, Tinian and Saipan where it appears to have been accidentally introduced in some way. It is noteworthy that this species has met with little success in Australia because of high mortalities associated with predation and parasitization (Waterhouse & Norris 1987). Quite possibly, the ineffectiveness of this species in Micronesia is for the same reasons. Sa/bia haemorrhoida/is GUENEE (PYRALIDAE) Also known as the lantana leaf roller, this species was introduced into Pohnpei along with H. strigata in 1958. Like the latter species it also feeds on lantana leaves and, as its name suggests, lives under the protective covering of the leaf margin which it folds and fastens down with silk. The species was very common on Pohnpei although its impact on lantana was minimal. It was also found on the neighboring island Chuuk but once again had little impact on the lantana growing there. We did not find it on any of the other islands visited. Attempts were made to introduce S. haemorrhoidalis to Yap and Guam in 1958 but it failed to establish. Likewise, its introduction to Belau in 1960, and again to Yap in 1962, met with similar results. Indigenous Natural Enemies A number of adult, polyphagous insects, indigenous to the islands of Micronesia, have been observed to feed on lantana but we have, so far, only discovered two lepidopteran species that utilize the plant for egg laying and larval development. These noteworthy adaptions are discussed for each species below.

Zizula hylax F. (LYCAENIDAE) Normally a consumer of small, leguminous, ground covering plants, the rather attractive larvae of this species were found grazing the florets of the orange/ yellow lantana on Yap and Angaur. On the former island, larvae were isolated from 25-30% of all damaged lantana flowers examined. It is pertinent to note here, that other species of lycaenid butterfly are known to feed on lantana and two of these, Strymon (=Thecla) ecion and S. bazochii were among the eight candidate species first established in Hawaii for such purposes (Perkins & Swezey

Micronesica Suppl. 3, 1991

1924). However, this is the first work that shows Z. hylax attacks this plant. Also of interest is the fact that Z. hylax is very common on Guam although the larvae are not found on the pink/yellow variety of lantana found growing there. The fact that the larvae can be successfully reared out on this variety in the laboratory suggests the adults prefer nectar from certain color-morphs over others.

Adoxopheyes melia CLARKE (TORTRICIDAE) Isolated from lantana flowers on Guam and Saipan, larvae of this polyphagous species have, in the past, only been recorded to feed on the leaves of Citrus sp. and Maytenus thompsoni (Celastacae) locally known as "luluhut". Like L. pusillidactyla, it only consumes the lantana florets, webbing their remains together to form a protective covering over itself. It was first recorded on Guam in 1971 and at the present time is considered to play a relatively minor role in controlling the spread of lantana on the island. Concluding Remarks

Clearly, lantana is widespread throughout Micronesia although the introduction of several of the weed's natural enemies has, on the majority of islands visited, resulted in an acceptable level of control. In the event that more effective, long-term control of this weed be required, the introduction of additional insect enemies, may be a worthwhile consideration. One candidate species, in particular, that merits consideration is the leaf mining fly Calycomyza lantanae Frick (Agromyzidae). This species is wide spread throughout the southeast Asian region and has been reported to cause severe defoliation of lantana in Malaysia (Ooi 1987). Several other candidate species are available and these are discussed at length in previous communications (Muniappan & Viraktamath 1986, Muniappan 1988). It should be mentioned here, however, that L. camara is a particularly hardy and resilient plant species, that demonstrates a remarkable capacity to recover from the severest of insect attacks. It, therefore, seems unlikely that the weed will ever be reduced to a level where it is difficult to find unless chemical and/ or mechanical means are additionally used to implement its control. Acknowledgements

We gratefully acknowledge the help and advice ofNelson Esguerra and Anita Suta (College of Micronesia, Pohnpei); Bernard Billimont and William Iteioshy (College of Micronesia, Chuuk), Demei Otobed and Haruo Adelbai (Division of Conservation and Entomology, Belau), Patrick Sogaw and Mathias Kugumagar (Dept. of Resources & Development, Yap), Antonio Santos and George Manglona (Northern Marianas College, Saipan) and Charles Frear (Soil Conservation Service, USDA, Saipan). This research was funded by a grant from Tropical and Subtropical Agriculture Program (88-34135-3642) of CSRS, USDA.

Denton et a!.: Biological Control of Lantana

Bibliography Beeson, C. F. C. & N.C. Chaterjee. 1939. Possibilities of control oflantana (Lantana aculeata Linn.) by indigenous insect pests. Indian Forest Records 6: 41-84. Bryan, E. H. 1949. Economic insects of Micronesia. Report of the Insect Control Committee for Micronesia 1947-1948. National Research Council, Washington, D.C. 29 pp. Davis, C. J. 1959. Recent introduction for biological control in Hawaii- IV. Proc. Hawaiian Entomol. Soc. 17: 62-66. Denton, G. R. W., R. Muniappan & M. Marutani. Status and natural enemies of the weed, Lantana camara, in Micronesia. Tropical Pest Management (in press). Gardner, T. R. 1958. Biological control of insect and plant pests in the Trust Territory of Guam. Proceedings of the Tenth International Congress of Entomology, Montreal, Canada. 17-25 August 1956. Biological Control: Reviews of Work 4: 465-469. Harley, K. L. S. 1971. Biological control of lantana. PANS 17: 433- 437. Harley, K. L. S. & R. C. Kassulke. 1971. Tingidae for biological control of Laniana camara (Verbenaceae). Entomophaga 16: 389-410. Holm, L. & J. Herberger. 1969. The world's worst weeds. Proceedings of the 2nd Asian Pacific Weed Control Conference, University of Wisconsin, Madison, 1-14. Holm, L., D. L. Plucknett, J. V. Pancho & J.P. Herberger. 1977. The world's worst weeds. Distribution and biology. University Press of Hawaii, Honolulu, Hawaii. 909 pp. Muniappan, R. 1988. Biological control of the weed, lantana camara, in Guam. J. Plant Protec. Tropics 5: 99- 101. Muniappan, R. 1989. Biological control of Lantana camara L. in Yap. Proc. Hawaiian Entomol. Soc. 29: 195-196. Muniappan, R. & C. A. Viraktamath. 1986. Status and biological control of the weed, Lantana camara, in India. Tropical Pest Management 32: 40-42. Ooi, P. A. C. 1987. A fortuitous biological control of Lantana in Malaysia. Tropical Pest Management 33: 234-235. Pemberton, C. E. 1954. Invertebrate Consultant Committee for the Pacific. Report for 1949-54. Pacific Science Board. National Academy of Science Research Council. 56 pp. Perkins, R. C. L. & 0. H. Swezey. 1924. The introduction into Hawaii of insects that attack lantana. Bull. Exp. Station, Hawaiian Sugar Planters Association. 83 pp. Savidge, J. A. 1987. Extinction of an island forest avifauna by an introduced snake. Ecology 68: 660-668. Schreiner, I. 1989. Biological control introductions in the Caroline Islands. Proc. Hawaiian Entomol. Soc. 29: 57-69. Waterhouse, D. F. & K. R. Norris. 1987. Biological Control: Pacific Prospects. Inkata Press, Melbourne. 454 pp.

Micronesica Suppl. 3: 83-92, 1991

Biological Control: Mutual Advantages of Interaction Between Australia and the Oceanic Pacific DOUGLAS

WATERHOUSE

ACIAR, GPO Box 1571 Canberra, ACT, Australia 2601

Abstract-Australia and Oceanic Pacific countries share a number of major introduced insect and weed pests and both must endure continuing intrusion of new pests. Some important Pacific pests have already been brought under biological control in Australia and the natural enemies employed will almost certainly be of value also in the Pacific. Conversely, some pests already in the Pacific, but which have not reached Australia (e.g. banana skipper, spiraling whitefly), are targets for biological control programs. If these programs are successful (as they probably will be), the reduction in pest densities should help to delay the further spread of these pests. An additional important benefit is that, if other countries are invaded, tested natural enemies will be readily available. There are clearly mutual advantages in close regional collaboration in biological control activities. Reasons for emphasizing the biological control approach and criteria for selection of target pests are discussed. Introduction

From a biological point of view, for several million years, the Pacific oceanic islands and Australia were substantially-although not completely-isolated from each other and from the rest of the world by the large expanses of water that surround them. Of course, over the millenia the insect faunas that had managed to establish themselves had been changing steadily due to evolutionary forces. Rarely, additional species would be transported in upper air currents, by birds or on flotsam from one country and established in another, sometimes far away. The undoubted trend, however, was for the insect faunas of isolated Pacific islands or island groups to diverge from each other rather than to become progressively more similar. In the Pacific all this was changed, first by the arrival of Polynesian, Melanesian and Micronesian peoples and then dramatically so by the abrupt intrusion of European man in ever increasing numbers into the region in his sailing ships, together with an ever increasing range of plants and animals, many of them with insects as fellow travelers. I shall first touch briefly on the origin and nature of the Australian insect fauna and follow, equally briefly, with that of the Pacific.

Micronesica Suppl. 3, 1991

Human occupation of Australia began when the first Aborigines arrived from the north at least 40,000 years ago. At that time the accumulation of ice at the poles resulting from the most recent ice age had lowered the ocean by as much as 100 m below its present level. This resulted in the emergence of extensive land bridges, not only between Papua New Guinea, mainland Australia and Tasmania, but also connecting many islands now lying between Papua New Guinea and Asia. Nevertheless, even at the time of lowest sea level, the shortest distance across the ocean deeps between Malaysia and Australia still involved 8 sea voyages, the longest of these being nearly 70 km. These first people had, of course, been preceded by many northern .organisms that had evolved earlier and had been steadily extending their ranges southwards over roughly the same route for the previous 3 or 4 million years. In the new land the Aborigines would have recognized examples of the flora and fauna with which they had already become familiar during their wanderings through the island arcs to the north. However, as they progressed southwards, this familiar biota became attenuated, and increasingly they encountered an extensive group of southern organisms that had evolved on the Australian continent during the 50 or so million years of its isolation from other large land masses, especially that known as Gondwana Land. No doubt many of these organisms were then as strange to them as they were to European man when he arrived several tens of thousands of years later. Probably the only insects to travel along with the early Aborigines were human lice. These feature in dreamtime legends and must presumably have long antedated the arrival of Europeans. Furthermore, Bayly in 1777 wrote of Tasmanian Aborigines at Adventure Bay (who at that stage had no known contact with Europeans) that they 'have many lice sticking about the neck and other parts of the body.' The Maoris carried lice before the arrival of Europeans and lousiness may indeed have been a universal human condition, since lice have been recovered from mummies of ancient Egypt, preColumbian America, fifteenth century west Greenland and the Aleutian Islands. There is no evidence from legends that human fleas were fellow travellers. This is not surprising since, because their larval life cycle is spent off their host, fleas would have had extreme difficulty in surviving a long trek to Australia unless a fairly uninterrupted flow of people made the journey. Nakedness would not help fleas either and it is of interest that the dingo which, with its more hospitable hairy body, would .be capable of transporting man-infesting fleas, did not arrive in Australia before about 4,000 years ago. Estimates suggest that there are at least 250,000 species of native insects in Australia, possibly 1,000 or so unwelcome migrants that have gained access since European colonization and a much smaller number than this ofbeneficial species, mainly introduced intentionally for biological control. Some native Australian insects have proved to be important pests of man and his crops, such as the bushfly, the Queensland fruitfly, several large brown blowflies, a number of species of mosquitoes, sand flies, termites, the Australian plague locust and the larvae of a number of moths-but, at most, no more than

Waterhouse: Advantages of Biological Control Interactions

a few hundred species. However, a far larger number of unintentionally introduced species have become serious agricultural pests. That they are agricultural pests is not surprising, since almost all Australian agriculture is based on introduced plants. Indeed, the only native plant to have gained wide acceptance both in Australia and overseas as a source of human food, is the eastern Australian macadamia nut. Of course, the aborigines also ate various grass seeds and the fruits of a handful of shrubs and trees, but none of these is in widespread use today. A number of plants that have been introduced-some intentionally, others unintentionally-have become our most important weeds. These introduced weeds, like the majority of our most important insect pests, have seldom, if ever, been accompanied by the full complex of natural enemies that serve to reduce their abundance in their area of origin. So much, very briefly, for Australia: what of the Pacific oceanic islands before the arrival of European man? Many Pacific islands have never had continental connections and were elevated from the ocean bottom. Thus, animals requiring land connection for dispersal were excluded from ever reaching them. It is possible to generalize that islands not greatly isolated tend to have richer faunas than very isolated islands; and older islands generally richer faunas than younger islands or atolls. Another generalization is that the terrestrial fauna of oceanic islands is characterized by poverty and disharmony (the absence of many groups). Until 2 or 3 thousand years ago when humans first appeared in the Pacific, vertebrates were poorly represented: mammals only by bats; birds largely by wideranging sea birds with generally very few land birds; reptiles by a few skinks and geckos; and amphibia and fresh-water fish were absent. In so far as arthropods are concerned, it appears that most immigration to oceanic islands was by aerial dispersal. In the south west Pacific area this was mostly in storm winds contrary to the direction of prevailing good-weather winds, but sometimes probably by sea birds. There is, indeed, a fairly good correlation between the types of insects trapped on ships at sea and those that have populated the more isolated islands. It seems that relatively few insects were able to colonize islands by surviving on flotsam. Crucial to the establishment of phytophagous insects on islands is the prior establishment of appropriate plants to provide breeding sites. In the case of insects dispersing in moving air masses, another major hazard is that of making a landfall, seeing that only so little of the Pacific is above water. Other serious obstacles include damage in strong winds; desiccation, particulary in clear weather; injury on landing; and adverse effects from salt spray. Insects derived from the Oriental/Southeast Asian region (China, Japan, Malaysia, Philippines, Indonesia) dominate the fauna of the mid Pacific as well as the islands of the western Pacific. The influence of North, Central and South America is quite minor west of the Galapagos. Even in Hawaii, which is nearest to North America, American elements (except for birds) are not dominant and, in the Marquesas and Society Islands, American influence is still less. Hawaiian plants also display a predominantly Oriental influence. In stark contrast, Aus-

Micronesica Suppl. 3, 1991

tralian influence on the oceanic fauna of a few hundred years ago was weak beyond New Zealand, New Caledonia, New Guinea and nearby islands. If an insect survives the journey to an island and manages to reproduce, it may become established in an empty niche, enter into competition with relatives, or compete with unrelated ecological equivalents. It is interesting that, for Hawaii, only about 250 perpetuated establishments over the last 5 million years are necessary to account for the present insect fauna which amounts to 7,000 or so species. This is only one successful establishment every 20,000 years or so. Similarly, there are about 1,800 species of seed-bearing plants in Hawaii and these appear to have been derived from an estimated 272 original introductions. By comparison with the 40,000 years or more that aborigines have been in Australia, human colonization of the oceanic Pacific islands is recent-possibly less than 4,000 years and in many areas less than 1,500 years. The early Polynesians sailed eastwards into the Pacific in long out-rigger canoes bringing with them coconuts, bananas, taro, yams, sweet potato, breadfruit and a few other plants. The Micronesians and Melanesians apparently moved eastwards into their areas of the Pacific somewhat later and they too brought with them the same basic food plants. Of the 30 insect pests rated to be the most important in the Pacific, 4 occur nowhere else (3 on coconut and 1 on taro). A further 15 are native to the Oriental region. Only 2 come from the Americas and several are pests of vegetables introduced with these from Europe or Africa. In contrast, only 3 of the 17 most important weeds are native to the Oriental region, whereas 13 originated in Central or South America-a situation so different from that of insect pests that it points to a quite different means of dispersal, probably man-mediated in the last couple of hundred years. Pest Species More than half of both the major Pacific insect pests and the major Pacific weeds also occur in Australia. For a number of these, moderately to very successful biological control campaigns have already been launched in Australia and others are in progress. Among the moderately to very successful projects I would place, for insect pests, the diamondback cabbage moth (Plutella xylostella (Linnaeus)), the green vegetable bug (Nezara viridula (Linnaeus)), and, most recently, the banana weevil borer (Cosmopolites sordidus (Germar)). For weeds I would list Salvinia molesta, water hyacinth (Eichhornia crassipes) and Lantana camara. It is highly probable that Australian experience with the biological control of these pests could be utilized to great benefit for Pacific nations and I shall illustrate this in a moment. I shall also take one example of a recently arrived pest, the leucaena psyllid (Heteropsylla cubana) and two examples where new pests are spreading steadily in the Pacific, but have not yet reached Australia. These are the banana skipper and the spiraling whitefly. Both are amenable to biological control and indeed have already been controlled in Hawaii. Biological control of introduced pests is

Waterhouse: Advantages of Biological Control Interactions

highly appropriate for oceanic island nations with small human populations and extreme shortage of foreign exchange for purchasing pesticides. The extension of effective biological control of such pests to other Pacific islands where they have become established would not only be of great value to these islands but, by reducing populations of pests that Australia does not yet have to low levels, should diminish the speed with which they reach its shores. Aleurodicus dispersus RUSSELL A pest that has steadily extended its distribution in the Pacific in the direction of Australia is the spiraling whitefly Aleurodicus dispersus which is native to Central America and the Caribbean region. It first gained a toehold in southern Florida in 1957 and then a bridge-head in Hawaii in 1978. In the last 7 years it has invaded several islands in Polynesia and Micronesia and also the Philippines and it seems inevitable that it will reach Australia before long. All stages suck sap from the leaves of their host plants causing unthriftiness and leaf drop, but seldom death. However, copious amounts of waxy white flocculent material are secreted and this creates a very unsightly appearance when dispersed by the wind. Even more importantly much sticky honeydew is produced which serves as a substrate for dense growth of sooty moulds, which interferes seriously with photosynthesis. The spiraling whitefly adult superficially resembles a tiny white moth and the pest drives its common name from the irregularly spiraling deposits of waxy white flocculence which the female deposits when laying eggs. The spiraling whitefly is an important pest of vegetables, fruit trees, ornamentals and shade trees. Its host range includes well over 100 plant species in more than 26 plant families. Among important hosts in the Pacific are coconut, banana, pawpaw, mango, guava, citrus and capsicum. A. dispersus was discovered in Honolulu in September 1978, and spread rapidly throughout the island of Oahu. By 1981 it was established on the other islands. A search was made for natural enemies in the Caribbean in 1979 and 1980. Three species of predatory coccinellid and two species of aphelinid parasitoid were introduced, studied for host-specificity and liberated. Of these the coccinellid Nephasis oculatus (Blatchley) and the parasitoid Encarsia haitiensis Dozier were the most effective. As is typical of coccinellids, Nephasis was effective in reducing high populations of whitefly, but relatively less effective against low populations. On the other hand, Encarsia was effective in low whitefly populations and ultimately, therefore, the more valuable biological control agent. A. dispersus is now regarded as being under successful biological control in most areas of Hawaii. These parasitoids have been introduced in the last 2 or 3 years to Cook Is, Fiji, Kiribati, American and Western Samoa and Tonga and indications are that they will reduce the spiraling whitefly to the category of a minor pest. When it comes to Australia's turn I would strongly recommend that the parasitoid Encarsia haitiensis be introduced first, leaving the less specific coccinellids to a later stage, if indeed they are required. Indeed I hope that the efficacy will be established

Micronesica Suppl. 3, 1991

in the Pacific of this tactic, of introducing the parasitoid alone, long before we have to face a decision in Australia. Erionota thrax (LINNAEUS) A pest butterfly that is an increasing threat to Australia is the banana skipper (Erionota thrax), a hesperiid that is native to Southeast Asia. It has spread in the last 20 years to Hawaii, Guam and Mauritius and quite recently to northern Papua New Guinea, and more recently, throughout the country. There would appear to be no effective barrier to a southwards extension to Australia, and possibly as far as the banana-growing regions of northern New South Wales. The brownish adults, with a wingspan of about 75 mm fly rapidly and apparently erratically among banana plants in the evening and early morning and are occasionally attracted to lights. The yellow eggs are laid at night mainly on the undersurface of the leaves. The newly hatched larva proceeds to the edge of the leaf where it starts feeding and begins to roll and tie the leaf. As the larva grows, the roll is enlarged and moved towards the midrib. All except the first instar larvae are covered with a whitish waxy powder. Rain causes high mortality of young larvae, due to their lack of waxy powder and the poor construction of their leaf rolls. Older larvae close their rolls more securely and produce enough wax to be water repellent. Depending upon temperature there are two to five or more generations a year. Foodplants include banana, in particular, but the skipper is also reported to attack coconut, bamboo, Manila hemp and several palms. Heavy infestations of bananas leave only the midrib with numerous leaf rolls attached to it. At the height of the attack in Hawaii more than 80% of all banana plants were damaged and E. thrax was regarded as a serious threat to the banana industry. Although there are occasional damaging outbreaks in its area of origin, it is attacked there by a large number of parasitic wasps and flies, many of which are themselves heavily parasitised. After the skipper became established in Hawaii, Guam and Mauritius it was brought under biological control in each country by the introduction of an encyrtid egg parasitoid (Ooencyrtus erionotae Ferriere) and particularly, by a braconid larval parasitoid (Apanteles erionotae Wilkinson). The former is already present in Papua New Guinea. In Hawaii and Guam, there are no native Hesperiidae, so there were no problems relating to possible lack of host specificity when introducing the parasitoids. Mauritius lists the occurrence of four hesperiids, some of which may be native, but the question of host specificity of the introduced parasitoids was apparently not considered. On the other hand, both Papua New Guinea and Australia have many scientifically important native hesperiids. The host specificity of the two parasitoids has been examined to ensure that only adequately specific species will be introduced. Tests on selected Papua New Guinea skippers and birdwings using Apanteles erionotae have proved negative and this parasitoid has recently been liberated in several sites in Papua New Guinea. Specificity tests are in progress against

Waterhouse: Advantages of Biological Control Interactions

additional Australian butterflies, so that the way will be cleared for rapid action should the banana skipper be discovered in Australia. H eteropsylla cubana eRA WFORD The most spectacular recent invasion of the Pacific and Australia by a new insect pest has been that of the leucaena psyllid (Heteropsylla cubana). It spread very rapidly and extensively from its area of origin, namely Central America. It was first reported in the Pacific in April 1984 in Hawaii. Before the end of 1985 it had reached most of the central Pacific islands, Taiwan and the Philippines; by 1986 the Solomon Islands, Papua New Guinea, Australia, Indonesia and Christmas Island (in the Indian Ocean); by 1987 Sri Lanka; and by 1988 India. Leucaena species are leguminous shrubs or trees which are being used increasingly in the Asian-Pacific region as shade trees (for cocoa and other crops) for animal forage, firewood, erosion control and a number of other purposes. The damage caused by H. cubana is primarily to new growth, where eggs are deposited. Large numbers of psyllids removing sap rapidly cause new growth to become stunted. In addition, the deposition of honeydew permits the growth of sooty molds which inhibit photosynthesis. When new growth is repeatedly attacked there is severe defoliation and ultimately death of the plant. The most commonly grown species, L. leucocephala, is particularly susceptible to attack, although some other Leucaena species are less so, as are several of the hundreds of crosses and cultivars that have been produced. It is hoped that suitable resistant or partially-resistant cultivars will become available in due course. In Hawaii the psyllid is attacked by a number of predators, particularly nonspecific coccinellids, that had been introduced previously against other pests. Several of these, including Curinus coeruleus (Mulsant), consume large numbers of psyllids when they are abundant, but turn their attention to other prey when psyllid numbers drop and hence are not very efficient in maintaining psyllid numbers at adequately low levels. A search for more specific natural enemies in the Caribbean resulted in the introduction to Hawaii of a eulophid wasp (Tamarixia leucaenae Boucek) and an encyrtid (Psyllaephagus yaseeni). Tamarixia was not established and died out, but P. yaseeni has been mass reared and established, but with some difficulty, in the field. A problem arises for Australia and a number of Pacific countries in deciding whether or not to consider introducing parasitoids of H. cubana. This is because another species of H eteropsylla from Brazil appears to be specific and highly damaging to a serious introduced weed, the giant sensitive plant (Mimosa invisa). This weed is responsible for losses of several millions of dollars a year in the moister, warmer areas of Queensland and is a major problem in many Pacific countries. The Brazilian H eteropsylla has been established in recent years in Queensland and Western Samoa on M. invisa, and appears to be causing serious damage to the weed. Several Pacific countries are waiting to follow suit if successful control is achieved.

Micronesica Suppl. 3, 1991

Unfortunately tests by CIBC in Trinidad have shown that the two parasitoids of Heteropsylla cubana will also attack the Brazilian Heteropsylla and thus they are likely to interfere with the biological control of Mimosa invisa, should they be introduced. It is hoped that additional, more specific, parasitoids may be found in Central or South America where H eteropsylla cubana occurs naturally and is not regarded as a pest. Australia and the Pacific islands in this instance share a common interest, namely to ensure that only the most highly specific natural enemies of H eteropsylla cubana are introduced to our region or indeed to countries bordering it. Cooperative Pest Control

In these few examples of many that could have been chosen, it is abundantly clear that there are great advantages of mutual interaction in biological control activities between Australia and the oceanic Pacific. Never in the past has biological control of arthropod and weed pests in the Pacific been more needed, and never has it been more attainable for a significant proportion of the important pests of the region. These assertions can be justified on many grounds. In relation to the need: 1. The steady increase in Pacific populations in modern times has progressively eroded traditional forms of agriculture, particularly those based on the use, in rotation, of part only of available land at any one time. A far higher proportion or agricultural land than ever before is now being used to produce the larger quantity of food needed and there is a correspondingly far greater dependence now on a reliable, sustained yield. It is thus even more important than before for crop losses to be kept to a minimum. 2. Traditional crops, such as coconut, taro and yams, have been supplemented by many new vegetables and fruits and these are becoming progressively more important in the economy (including cabbage, capsicum, cocoa, corn, cucumber, egg plant, tomato, mango, pawpaw, watermelon, etc.). These crops are attacked by a range of pests introduced with them. 3. At least three quarters of the major insect pests and all of the major weeds have been introduced to the region-the majority of them during this century. New pests are arriving steadily and the rate of arrival will increase with increasing tourism. 4. All pesticides have to be imported to the Pacific region, requiring foreign exchange. 5. There is a rapidly growing world recognition of the need to limit, as far as practicable, the use of pesticides because of a variety of undesirable side effects. 6. When carried out properly, biological control offers for many pests (particularly introduced pests in relatively simple island ecosystems) an opportunity for sustained and selective suppression of pest populations to densities below economic injury levels. Where effective biological control is possible, it is clearly the method of choice.

Waterhouse: Advantages of Biological Control Interactions

Table 1. Criteria relevant to an assessment of the importance of a pest to a country or region. 1. Importance of crop(s) affected: • area of crop • export value ·• subsistence value • size of population affected • social importance

2. Importance of the pest: • estimate of losses-actual potential • acceptable threshold of damage • cost of existing control • environmental/social cost • quarantine considerations 3. Biological control: • previous successes • size and cost of a program • perceived chances of success • conflict of interest (if any) 4. Are appropriate alternative methods of control effective and economic? • chemical control • cultural control • use of resistant varieties

In relation to being attainable: 1. Where important Pacific pests have already been brought under effective biological control elsewhere, the chances are high of repeating this success. In such instances the time scale is likely to be short (1 to 3 years) and the costs low. 2. With rapid air transport, care in eliminating unwanted fellow travellers, and better knowledge of culturing techniques, it is far simpler than ever before to import and establish natural enemies in a new country. It is not necessary to describe again here the assembling of information from the south western Pacific in which 47 major pests of the region were selected as possible biological control targets-30 insects, a mite, a snail and 17 weeds. There are, of course, many additional pests of generally lesser importance, or of importance to a limited region and, regrettably, serious new pests continue to appear from time to time. The information presented in 'Biological Control: Pacific Prospects' indicates that promising natural enemies are already known for between 15 and 20 of the 47 pests. Categorized as promising, are agents that have already been used in another country with some success and those for which there is already information suggesting that they may be valuable. A start has already been made on several of these target pests and it is perhaps worth noting that, of the 6 additional pests in Supplement 1 (1987) to Biological Control: Pacific Prospects, 3 are already targets for biological control investigations.

Micronesica Suppl. 3, 1991

What then are the main constraints to a substantial increase of biological control activities in the region? The principal constraint identified by all countries in the region is the shortage of funds. Other constraints included the shortage of trained staff in many countries and the need to upgrade quarantine and rearing facilities. Crucial to a sustained increase in biological control activity in the Pacific is the selection of appropriate targets. In addition to an early success in as many countries as possible, if biological control is to contribute most effectively to Pacific problems, possible targets must be placed in some priority order relating to the importance of a particular pest to a country or region. In this regard, the set of criteria drawn up by a biological control workshop in Tonga in 1985 provides a useful guide (Table 1). There will inevitably be gaps in supplying the documentation sought, but even a first attempt will help to provide each country with a logical basis for selecting projects for early action. Highly important also, the documentation will provide the very sort of information that will help to persuade donors to lend their support. References

Waterhouse, D. F. & Norris, K. R. 1987. Biological Control: Pacific Prospects. Inkata Press, Melbourne. 454 pp. Waterhouse, D. F. & Norris, K. R. 1989. Biological Control: Pacific Prospects. Supplement 1. Australian Centre for International Agricultural Research, Canberra. 123 pp.

Micronesica Suppl. 3: 93-98, 1991

Banana Skipper, Erionota thrax (L.) (Lepidoptera:Hesperiidae) in Papua New Guinea: a New Pest in the South Pacific Region D. P. A.

SANDS,

SANDS

CSIRO, Division of Entomology, Private Bag No. 3, Indooroopil!y, Queensland, 4068

and

ARURA

Department of Agriculture & Livestock, Bubia, Lae, Papua New Guinea

Abstract-Banana skipper, Erionota thrax (L.) (Lepidoptera:Hesperiidae) a butterfly of Southeast Asian origin, was first recorded in Papua New Guinea in 1983. Since 1987, it has spread throughout the mainland from sea level to 2500 m. The larvae feed on all species of bananas reducing fruit yields and preventing the traditional use of leaves for other purposes. No hosts other than Musaceae were recorded for E. thrax in Papua New Guinea. A joint survey by CSIRO and the Papua New Guinea Department of Agriculture & Livestock investigated the natural enemies and abundance of E. thrax as a preliminary to biological control of the pest in Papua New Guinea. Of three egg parasitoids found attacking E. thrax, only Ooencyrtus erionotae Ferriere significantly influenced its abundance. Natural enemies of larvae and pupae were uncommon and an important larval parasitoid of E. thrax in Southeast Asia, Apanteles erionotae Wilkinson, was absent. E. thrax could spread from Papua New Guinea, posing a threat to bananas throughout the south Pacific region. Introduction Banana skipper, Erionota thrax thrax (L.), sometimes referred to as the banana leaf roller, is a minor pest in its native range of Southeast Asia but became a major pest in Guam, Mauritius, Hawaii, and Papua New Guinea, where introduced without natural enemies (Waterhouse & Norris 1989). Two other subspecies, E. thrax mindana Evans from southern Philippines and E. thrax hasdrubal Fruhstorfer from northern Moluccas, are not known to occur outside of their native range. Larvae cause damage by leaf rolling and feeding on the leaves of banana plants, reducing fruit yields when defoliation exceeds 20% (Ostmark 1974). In Papua New Guinea, bananas are the second most important crop after sweet potato and in some provinces are the staple crop (King & Bull 1984). In August 1983, a specimen of E. thrax was collected at Vanimo, in northwestern

Micronesica Suppl. 3, 1991

Papua New Guinea (H. Roberts pers. comm.). E. thrax spread to the East Sepik Province in July 1986 and by mid 1987, was found defoliating banana plants in the Madang and Morobe Provinces. Several natural enemies of E. thrax have been identified in Southeast Asia (Hoffmann 1935). These include the egg parasitoid Ooencyrtus erionotae Ferriere (Encyrtidae) and the larval parasitoid Apanteles erionotae Wilkinson (Braconidae). Both have proven to be valuable agents when introduced to control E. thrax in countries where it has become an exotic pest (Waterhouse & Norris 1989). In Papua New Guinea, 0. erionotae was reported attacking eggs of E. thrax (Arura 1987). It is not known if this parasitoid was already established on another host or if it gained entry to the country at the same time as the pest. We discuss the distribution and hosts, damage to banana plants and studies associated with a program for biological control of E. thrax in Papua New Guinea. A threat to banana production elsewhere in the South Pacific region is recognised, should E. thrax spread further without its natural enemies. Materials and Methods ASSESSMENT FOR LEAF DAMAGE AND ABUNDANCE OF E. thrax STAGES Sites with bananas present were selected and total number of banana plants and number with stages of E. thrax was recorded. Ten banana plants were selected at random from a diagonal transect to measure the number of stages of E. thrax per leaf, the number of leaves with damage, and to estimate percent of each leaf damaged and the position on the plant. NATURAL ENEMIES OF E. thrax Any predators were collected and retained with the host. For parasitoid emergence, E. thrax stages were held in plastic tubes measuring 9.5 X 1.5 em diameter, each with a firm cotton wool plug. For eggs the following were recorded: (i) Number of eggs/mass, (ii) Number masses with parasites, (iii) Number eggs/mass parasitized, and (iv) Identity of parasites. For larvae and pupae, the number of parasites emerging, whether hyperparasitized and the instars of larval hosts were recorded.

Results DISTRIBUTION AND HOST RANGE OF E. thrax Since its initial build up in abundance in northern Papua New Guinea (PNG) during 1987, E. thrax has established up to altitudes of 2,500 m in the Eastern Highlands Province, to 2,060 m in the Finisterre Ranges and all coastal areas including the Western Province, less than 3 km from the Australian Torres Strait Islands (Fig. 1). Immature stages of E. thrax were collected from Musa sapientum Linn. and M. textilis Nee, Musa spp. (including sections Australimusa and Eumusa) and several varieties. No immature stages of E. thrax were found on H eliconia bihai

Sands et al.: Banana Skipper in PNG

PAPUA NEW GUINEA

1983

(mainland)

vii. 1986

1987

c::··. . ....

1988 1V

1989

•

Pt,~b~ 200 km

Figure 1.

Establishment and spread of Erionota thrax in Papua New Guinea.

Linn, Ensete glaucum (Roxb.) Cheesman, Nypa fruticans Wirmb., Metroxylon sp., Calmus spp., Ravena/a madagascariensis J. Gmelin, Cocus nucifera Linn., Areca catechu Linn. or Elaeis guineensis Jacq., even when infested Musa spp. plants were growing close by. In the laboratory, groups of 10 newly-hatched of E. thrax died without feeding when placed on fresh foliage of H. bahai, C. nucifera and E. guineensis, whereas all larvae placed on M. sapientum as controls, began feeding. ABUNDANCE OF IMMATURE STAGES OF

E. thrax

AND DAMAGE TO BANANA PLANTS

Summaries from surveys carried out in three Provinces of PNG, in MayJune 1988 and the Western Province, in March-April 1989 are shown in Table 1. E. thrax was present at all localities surveyed in the Morobe, Eastern Highlands and Madang Provinces in 1988 and Western Provinces in 1989. The maximum number of leaves infested with E. thrax stages was 94% at Bulolo, Morobe Province. The highest level of defoliation per site was 70% near

Micronesica Suppl. 3, 1991 Table 1.

Leaf damage to banana plants and abundance of Erionota thrax stages.

Number of Localities

Province Morobe Madang E. Highlands Western

32 7 2 14

Table 2.

Dates

x% Leaves with Stages (±SE)

x% Total Plants Defoliation

v.-

217 41 20 83

VI.

1988 iii-vi. 1989

37.0 29.3 29.2 23.9

Morobe Madang E. Highlands 1 2

No. Eggs Larvae

40.7 ± 4.0 27.1 ± 3.2 56.3 31.7 ± 6.1

6.8 3.8 10.8 2.7

Pupae

3.9 1.3 3.3 4.7

3.1 1.4 1.5 0.7

Parasitization of Immature Stages of E. thrax (May-June 1988). Eggs

Province

x No. Stages per Leaf

Larvae

No. Masses

Total Eggs

34 0 32

511

32.7

338

25.7

% Parasitized 1 No.

191 17 24

Pupae

% Parasitized

No.

% Parasitized2

0 0 0

172 7 11

5.8 0 0

by Ooencyrtus erionotae by Brachymoria sp. and unidentified tachinid

Busu Village in the Markham Valley and the maximum number of leaf rolls was 6 per leaf, at Goroka, Eastern Highlands Province. At several localities individual plants were completely defoliated, leaving only the leaf stalks, by as few as 2-3 larvae per leaf. 0. erionotae was present at all of the defoliated sites. When sites in the Morobe and Madang Provinces were revisited in MarchApril 1990, levels of infestation were much lower than in 1987 and 1988, except at Boana, Morobe Province. NATURAL ENEMIES OF E. thrax Three egg parasitoids were reared: Ooencyrtus erionotae, an unidentified Ooencyrtus sp. and an unidentified Anastatus sp. Of these, only 0. erionotae contributed to significant mortality of E. thrax (Table 2). Egg parasitization by 0. erionotae averaged 30% (total 849 eggs) from all localities for the period MayJune 1988 but reached 82% on one occasion in the Morobe Province. Eggs on Cephrenes mosleyi (Butler) were never attacked in the field by 0. erionotae even when present (on different host plants) with E. thrax in the same localities. Eggs of C. mosleyi were attacked by a different, unidentified Ooencyrtus sp. However, in recent studies in the laboratory, 0. erionotae completed development in eggs of Cephrenes augiades (Felder), showing that 0. erionotae will attack other hosts when confined in an artificial environment.

Sands et al.: Banana Skipper in PNG

Larvae of E. thrax were occasionally attacked by predatory Hemiptera but not by any Hymenoptera. Only 10 of 190 pupae collected from the field were parasitized (ca. 5%), two by a Brachymeria sp. and the remaining eight, by tachinid Diptera. The same Brachymeria sp. was reared from C. mosleyi. Diseased pupae (possibly by a microsporidian) frequently supported the development of sarcophagid Diptera. Discussion

After an initial build up in abundance, with accompanying serious defoliation of banana plants in Papua New Guinea during 1987-88, E. thrax numbers appeared to decrease by March 1990. However, some outbreaks particularly in the Markham Valley continued to cause problems. The decrease in abundance of E. thrax can in part, be attributed to egg parasitization by 0. erionotae but the level of biological control is not considered effective. Since the arrival of E. thrax in PNG, defoliation ofbanana plants has lowered fruit yields as well as disrupted the traditional use of banana leaves for: (i) wrapping ripening fruit, to prevent attack by birds and bats, (ii) as table cloths, (iii) sheltering firewood from rain, (iv) covering food to prevent settling by flies and (v) making baskets (Waterhouse & Norris 1989). Adults of E. thrax are active from sunset until about two hours after dusk (unpubl. obs.) and specimens are frequently attracted to lights. Lights in boats and loading aircraft may attract adults of E. thrax, enabling movement between Pacific countries. However, their powerful flight may be sufficient for international dispersal, especially movement from mainland Papua New Guinea to the Bismarck Archipelago, the Torres Strait Island and eventually, mainland Australia and the Solomon Islands. Should E. thrax establish in countries elsewhere in the southwestern Pacific without natural enemies, banana plants will be adversely affected as previously seen in Mauritius, Guam and Hawaii. A collaborative project between Australia CSIRO and the PNG Department of Agriculture & Livestock is assessing the potential for biological control of E. thrax in PNG. Since the egg parasitoid, Ooencyrtus erionotae is already present, the larval parasitoid Apanteles erionotae would appear to be the most promising agent to achieve control. Following an assessment of the abundance of E. thrax and its natural enemies in PNG, a decision was made to conduct further tests on the larval parasitoid, A. erionotae. These tests, carried out on a population from Guam, are being carried out in a quarantine facility in Papua New Guinea. Acknowledgements We thank Dr. Gary Denton and Professor R. Muniappan who kindly provided cultures of A. erionotae for further study. We are grateful to Mr. Fred Dori and R. Muthappa, DAL for their collaboration and the Australian Centre for International Agricultural Research, for supporting this project.

Micronesica Suppl. 3, 1991

References Arora, M. 1987. Overseas news. Australian Entomological Society, News Bulletin. 23: 130. King, G. A. & P. B. Bull, 1984. The Papua New Guinea Biological Foundation banana collection. IBPGR Regional Committee for Southeast Asia, Newsletter 8: 3-4. Hoffman, W. E. 1935. Observations on a hesperid leaf-roller and a lace-bug, two pests of banana in Kwangtung. Lingnan Sci. 14: 639-649. Ostmark, H. E. 1974. Economic insect pests ofbananas. Ann. Rev. Entomol. 19: 161-176. Waterhouse, D. F. & K. R. Norris, 1989. Biological control: Pacific prospects, Supplement 1. Australian Centre for International Agricultural Research, Canberra, 123 pp.

Micronesica Suppl. 3: 99-101, 1991

Biological Control of Some Introduced Pests in the Federated States of Micronesia NELSON

EsGUERRA

College of Micronesia, Land Grant Program P. 0. Box 1179, Kolonia, Pohnpei Federated States of Micronesia 96941

Abstract-An active biological control program is being pursued to suppress the populations of introduced pests into the Federated States of Micronesia. Natural enemies of orange spiny whitefly, southern green stink bug, leaf footed bug, Egyptian fluted scale and Siam weed have been introduced. Introduction of natural enemies of giant sensitive weed, spider mites, and coconut termite are being planned. Introduction A number of insect pests have been accidentally introduced to the Federated States of Micronesia (FSM) over the past few years. Leaving behind their natural enemies, with ample food and environment favorable for their development and survival, these pests have threatened the production of existing crops and new crops on the island nation. In most cases, indigenous natural enemies have had no impact on them. As a result, a number of them have become serious insect pests and are now the target of bio control programs. Current Control Programs Orange spiny whitefly-Aleurocanthus spiniferus Quiantance The orange spiny whitefly has been a major pest of citrus on Yap, Pohnpei, Truk and Kosrae. In 1988, control of the orange spiny whitefly on citrus on Pohnpei was achieved by using the parasite, Encarsia smithi (Silvestri) introduced from Guam. On Kosrae, satisfactory control of the whitefly was achieved in the early 80's using the same parasite. Through Agricultural Development in the American Pacific (ADAP) projects on Biocontrol Agent Exchange, the same parasite was released on whitefly infested citrus trees in Chuuk on November 1, 1989 and in Yap State in 1990. Southern Green Stink Bug-Nezara viridula (Linn.) The stink bug is a perennial pest of many crops on Pohnpei. On November 22, 1989, the College of Micronesia Land Grant Programs received parasitized eggs of N. virudula from the Hawaii Department of Agriculture. These eggs were

100

Micronesica Suppl. 3, 1991

incubated and adults of the parasite Trissolcus basalis (Wollaston), that emerged were released on cucumber plantings infested with the stinkbugs. The collection of stinkbug eggs is being continued to confirm the establishment of the parasite. Leaf Footed Bug-Leptoglossus australis (Fabr.) Although present on Guam and Belau, the leaf footed bug occurs only on Pohnpei but not in other states of the FSM. Both nymph and adult bug suck the sap from developing fruits, flowers, leaves, and stems of the host plants such as bittermelon, cucumber, watermelon, pumpkin, waxgourd, guava, citrus, beans, tomato, eggplant, and zuccini. As a result of their feeding the flowers fall off, the fruits fail to form, or fruits become deformed at maturity. It is suspected that the parasite, T. basalis also attacks eggs of the leaf footed bug. Egyptian fluted scale-/cerya aegyptiaca (Douglas) This mealybug is a serious pest of breadfruit on Mokilloa atoll. It has been observed to be present on the main island of Pohnpei, Kosrae, Yap and Chuuk but the population has never reached a level where control measures need to be initiated. The predatory coccinellid beetle, Rodolia pumila Weise is an effective biocontrol agent of this mealybug. Adults were collected from the main island of Pohnpei and released on breadfruit trees infested with the mealybug in Mokilloa atoll. Siam Weed-Chromolaena odorata (L.) King and Robinson C. odorata is an important weed of pastures, vacant areas, roadsides and also farm lands. The weed is present in Kosrae, Yap, Pohnpei and Belau. On Pohnpei, the weed is particularly abundant in Kitti and Madolenihmw. An arctiid moth, Pareuchaetes pseudoinsulata Rego Barros was received from Guam, mass multiplied and field released. It has been established in the field at Palikir in October, 1990. Planned Biological Control Projects

Giant Sensitive Plant-Mimosa invisa Martius ex Colla. The giant sensitive plant is a major weed on the island of Yap. It occurs on roadsides, vacant areas, pastures and even cultivated lands. It is an aggressive weed and can take over main crops planted in the area. Since it is thorny, the weed interferes with the farming activities. A request was made to the South Pacific Commission in February 1989 for the psyllid, a natural enemy of M. invisa. The Biological Control Officer of South Pacific Commission has agreed to supply the psyllid to Yap State. Spider mites-Tetranychus spp. Spider mites caused serious damage to cassava, papaya, sweet potato and other crops throughout the FSM in 1989. Affected leaves showed brownish spots and fell off in severe cases of infestation. Some susceptible varieties of cassava

Esguerra: Biological Control in FSM

101

were rendered leafless even during the early stages of plant growth. We plan to introduce the predaceous mites, Phytocelus persimilus from California for control of spider mites in the FSM. Coconut termite This termite is becoming an important pest in the FSM. It attacks coconuts on Pohnpei, Chuuk and Yap. Specimens of this termite have been sent to Australia for identification. We plan to initiate a biological control program utilizing an entomophagous nematode.

Micronesica Suppl. 3: 103-107, 1991

Distribution and Control of Chromolaena odorata (Asteraceae) R.

MUNIAPPAN AND

MARUTANI

Agricultural Experiment Station College of Agriculture and Life Sciences University of Guam, Mangilao, Guam U.S.A. 96923

Abstract-Chromolaena odorata (L.) King and Robinson is a neotropical weed, spread all over the humid tropical regions of Asia, Africa and Micronesia. This paper reviews mechanical, cultural, chemical and biological control methods against this weed. Of these methods, biological control either alone or in combination with others seems to be the most promising one to suppress this weed. Introduction

Chromolaena odorata (L.) R. M. King and H. Robinson (Siam weed) is a neotropical weed. It is a perennial shrub with allelopathic properties (Ambika & Jayachandra 1980). The weed is widely distributed in the tropical and subtropical Americas from southern Florida to southern Bolivia (Cruttwell McFadyen 1989). The spread of this weed to Asia, Africa, and Micronesia has been reported by Biswas (1934), Bennett & Rao (1968), Ivens (1974), Muniappan & Marutani ( 1988), and Cruttwell McFadyen ( 1989). The importance of C. odorata as a weed in various agricultural fields has been reviewed by Holm et al. ( 1977) and Ambika & Jayachandra (1990). C. odorata was introduced to Guam in early 1960's (Seibert 1989), and it has also established on the neighboring islands of Rota, Tinian, Aguijan, and Saipan in the Northern Mariana Islands. This noxious weed has become a problem on rangelands, roadsides and waste lands since early 1980's on these islands. It has also spread to Caroline islands: Yap and Palau (Muniappan et al. 1988) and Pohnpei and Kosrae in the late 1980's. The migration and spread of C. odorata in Micronesia has been steady and fast. Cruttwell McFadyen ( 1989) has warned of the possibility of introduction of this weed into Australia from Asia in the near future. There is also the possibility of C. odorata spreading to the South Pacific Islands from Micronesia and Asia. Mechanical Control

Manual slashing and use of motorized bush cutters and tractor drawn equipment are commonly used for clearing C. odorata. Olaoye (1977) found slashing to cause rapid regeneration even though repeated slashings eventually caused the death of this weed. Ojuederie et al. (1983) reported slashing of C. odorata 1.5 m

104

Micronesica Suppl. 3, 1991

diameter around the coconut trees to increase the yield. Manual weeding is mostly done in places where cheap and plentiful labor is available. Use of motorized bushcutters and tractor drawn equipment is also limited because of the restricted accessibility of areas where this weed is growing (Erasmus 1988). Slashing and burning are carried out in some places. Top portions of C. odorata burn readily during the dry seasons, while bases of the plants are resistant to fire, and they coppice immediately after rains. Seeds also germinate well after a fire (Liggitt 1983). Cultural Control

Cover crops such as Pueraria javanica, Pueraria phaseoloides, Calopogonium caeruleum, Desmodium ovalifolium and M oghania macrophylla have been tried for suppression of C. odorata and found to be not effective (Ambika & Jayachandra 1990). However, Salgado (1972) reported that Tephrosia purpurea grown as a cover crop in coconut plantations was effective in suppressing C. odorata in Sri Lanka. In tropical southern China, kikuyu grass (Pennisetum clandestinum) and Surinam grass (Brachiaria decumbens) are grown as pasture crops to suppress the growth of this weed (Wu Renrun, pers. comm.). Chemical Control

In India, Nair (1973) reported that Gramoxone at 0.3% ofthe concentration was not effective in controlling this weed, however, Rai (1976) found that Gramoxone in combination with 2,4-D sodium salt was effective. Mathew et al. (1977) recommended a combination of 1.5liters ofGramoxone and 0.75 kg Fernonoxone for control of this weed. Tumaliuan & Halos (1979) also reported that Gramoxone was effective in control of this weed in the Philippines. Soerjani et al. (1975) found that Picloram was effective for control of C. odorata in Indonesia. Madrid (1974) reported that Picloram at 1 kg/ha and Dicamba at 2 kg/ha as the recommended rate to control the weed. Erasmus (1988) and Liggitt (1983) reviewed chemical control of C. odorata in Africa. In general, timing of herbicide application was important in control of C. odorata. Plants were most susceptible when herbicides were applied at the young seedling stage or to the regrowth after slashing. Biological Control

Since C. odorata is an exotic plant and has become a serious weed in Asia, Africa and Micronesia, it is a good target for a classical biological control program. In 1966, the Nigerian Institute for Oil Palm Research requested the Commonwealth Institute of Biological Control (CIBC) to investigate the natural enemies of C. odorata. As an outcome of this investigation, several insects and mites were identified attacking C. odorata (Crutwell1974). Pareuchaetes pseudoinsulata Rego Barros (Lepidoptera:Arctiidae) was introduced to Ghana, Nigeria, India, Sri

Muniappan & Marutani: Chromolaena odorata

105

Lanka and Sabah (Malaysia) by the CIBC in the early 1970's (Julien 1987). The occurrence of P. pseudoinsulata in the Palawan Island of the Philippines was noted in 1985 (Torres 1986, Alterrado 1986) even though it was not introduced to the Philippines. Possibly, it was accidentally introduced from Sabah to the Palawan island. Since then, it has spread to Mindanao and Vasayas islands. P. pseudoinsulata was introduced and established on Guam in 1985 (Seibert 1989) and subsequently in Rota, Tinian, Saipan and Aguijan islands in the Marianas. Laboratory rearing and field release of the insect are in progress in Yap, Palau, and Pohnpei and shortly to be taken up in Kosrae. P. pseudoinsulata cultures have been shipped from Guam to South Africa, Thailand and Ghana for multiplication and field release. The seed feeding weevil, Apion brunneonigrum Beguin-Billecoq (Coleoptera:Curculoinidae), has been related in India, Sabah, Sri Lanka, Ghana, Nigeria (Julian 1987) and Guam (Seibert 1989). No recoveries of this weevil have been made in these countries. Acalitus adoratus Keifer (Acarinae:Eriophyidae) has been identified as one of the natural enemies of C. odorata in the neotropics and recommended as a biological control agent (Cruttwell 1977, Cock 1984). A. adoratus has been observed in Thailand even though it was not introduced for biological control (Napompeth et al. 1988). Similarly, Muniappan et al. (1988) observed this mite in Palau and Yap in Micronesia. The larva of Mescinia sp. near parvula Zeller (Lepidoptera:Pyralidae) bores into the stem of C. odorata (Cruttwell 1977, Cock 1984). A few moths of this species were released in Guam in 1985 (Seibert 1989). Melanagromyza eupatroiella (Diptera:Agromyzidae) was also identified as a natural enemy of C. odorata in the neotropics (Cruttwell 1974). Efforts to introduce M. eupatoriella into Guam were discontinued in 1987 as there was the high incidence of parasitism in Trinidad where the collections were made by CIBC. The Institute de Rechereche Pour les Huiles et Oleagineux (IRHO) in cooperation with CIBC is undertaking investigations for further exploration of natural enemies of C. odorata in the neotropics for eventual field release of suitable candidates in West Africa. The government of South Africa is also investigating in South America for promising natural enemies to be released in the Natal region. References

Ambika, S. R. & Jayachandra. 1980. Suppression of plantation crops by Eupatorium weed. Current Science. 49: 874-875. Atterrado, E. D. 1986. Pareuchaetes pseudoinsulata, caterpillar on Chromolaena odorata: a new Philippine record. Philippine J. Coconut Studies 11: 59-60. Bennett, F. D. & V. P. Rao. 1968. Distribution of an introduced weed Eupatorium odoratum Linn. (Compositae) in Asia and Africa and possibilities of its biological control. PANS Section C. 14: 277-281. Biswas, K. 1934. Some foreign weeds and their distribution in India and Burma. Indian Forester 60: 861-865.

106

Micronesica Suppl. 3, 1991

Castillo, A. C., F. A. Moog & C. Pineda. 1977. Introduction of ipil-ipil in "gonoy" infested pastures. Philippine J. Animal Industry 32: 1-10. Cock, M. J. W. 1984. Possibilities for biological control of Chromolaena odorata. Tropical Pest Management 30: 7-13. Cruttwell, R. E. 1974. Insects and mites attacking Eupatorium odoratum in the neotropics. 4. An annotated list of the insects and mites recorded from Eupatorium odoratum L. with a key to the types of damage found in Trinidad. Commonwealth Institute of Biological Control. Tech. Bull. 17: 87-125. Cruttwell, R. E. 1977. Insects and mites attacking Eupatorium odoratum L. in the neotropics. 6. Two eriophyid mites, Acalitus adoratus Keifer and Phyllocoptes cruttwellae Keifer. Commonwealth Institute of Biological Control. Tech. Bull 18: 59-63. Cruttwell McFayden, R. E. 1989. Siam Weed: A new threat to Australia's north. Plant Protection Quarterly 4: 3-7. Dharmadhikari, P.R., P. A. C. R. Perera & T. M. F. Haasen 1977. The introduction of Am malo insulata for the control of Eupartorium odoratum in Sri Lanka. Commonwealth Institute ofBiological Control. Tech. Bull. 18: 129135. Erasmus, D. J. 1988. A review of mechanical and chemical control of Chromolaena odorata, in South Africa. Proc. First International Workshop on Biological Control of Chromolaena odorata, Bangkok. pp. 34-40. Ag. Exp. Sta., U. Guam, Mangilao. Holm, L. G., D. L. Plucknett, J. V. Pancho, & J.P. Herberger. 1977. The World's Worst Weeds, Distribution and Biology. Univ. of Hawaii Press, Honolulu, 609 pp. Ivens, G. W. 1974. The problem of Eupatorium odoratum L. in Nigeria. PANS 20: 76-82. Julian. M. H. 1987. Biological Control of Weeds: A World Catalogue of Agents and their Target Weeds. 2nd Edition. Commonw. Agri. Bur. Int. Wallingford, 144 pp. Liggitt, B. 1983. The invasive alien plant Chromolaena odorata with regard to its status and control in Natal. Monograph 2, Institute ofNatural Resources, Republic of South Africa, 41 p. Madrid, M. T. Jr. 1974. Evaluation of herbicides for the control of Chromolaena odorata (L.) R. M. King and H. Robinson. Philippines Weed Sci. Bull. 1: 25-29. Mathew, M., K. I. Punnoose & S. N. Potty. 1977. Report on the results of chemical weed control experiments in the rubber plantations in South India. J. Rubber Res. Inst. Sri Lanka. 54: 478-488. Muniappan, R. & M. Marutani. 1988. Ecology and distribution of Chromolaena odorata in Asia and the Pacific. Proc. First International Workshop on Biological Control of Chromolaena odorata, Bangkok, Thailand pp. 21-24. Ag. Exp. Sta., U. Guam, Mangilao. Muniappan, R., M. Marutani & G. R. W. Denton. 1988. Introduction and establishment of Pareuchaetes pseudoinsulata Rego Barros (Arctiidae) against

Muniappan & Marutani: Chromolaena odorata

107

Chromolaena odorata in the Western Caroline Islands. J. Biol. Control. 2: 141-142. Nair, P. N. 1973. The effect of gramoxone application on Eupatorium odoratum. Indian Forester 99: 43-48. Napompteh, B., N. Thi Hai & A. Winotai. 1988. Attempts on biological control of Siam weed, Chromolaena odorata in Thailand. Proc. First International Workshop on Biological Control of Chromolaena odorata, Bangkok, pp. 5762. Ag. Exp. Sta., U. Guam, Mangilao. Ojuederie, B. M., G. 0. Iremiren & S. N. Utulu. 1986. Effects of various interrow slashing regimes and size of weeded rings on the early growth, flowering and bunch yield of the oil palm. J. Nigerian lnst. Oil Palm Research 6: 322334. Olaoye, S. 0. A. 1977. The effect of slashing on the performance of Eupatorium odoratum Linn. (Siam weed) in Nigeria. Proc. 7th Nigerian Weed Science Soc. Conf. pp. 70-79. Rai, S. N. 1976. Eupatorium and weedicides. Indian Forester 102: 449-454. Salgado, M.L. M. 1972. Tephrosia purpurea (Pila) for the control of Eupatorium and as a green manure on coconut estates. Ceylon Coconut Plrs. Rev. 6: 160-174. Seibert, T. F. 1989. Biological control of the weed, Chromolaena odorata (Asteraceae), by Pareuchaetes pseudoinsulata (Lep.: Arctiidae) on Guam and. the Northern Mariana Islands. Entomophaga 34: 531-539. Soerjani, M., A. Soedarsan, S. Mangoensoekarjo, T. Kuntohartono & !\1. Sundaru. 1975. Weed problems and prospects for chemical control in Indonesia. 5th Conference of the Asian Weed Science So. pp. 18-22. Torres~ D. 0. 1986. Potential insect biological control agents against the weed Chromolaena odorata R. M. King and H. Robinson in the Philippines. Philippines Entomol. 6: 535-536. Tumaliuan, B. T. & S. C. Halos. 1979. Screening herbicidal preparations and mixture for clearing reforestation areas. Sylvatrop Philippines Forest Research J. 4: 151-159.

Micronesica Suppl. 3: 109-116, 1991

Occurrence of the Giant African Snail in the Ogasawara (Bonin) Islands, Japan K. TAKEUCHI 1, S. KoYANO and K. NUMAZAWA Ogasawara Subtropical Agricultural Research Centre, Ogasawara-mura, Chichijima, Tokyo, 100-21, Japan

Abstract-The giant African snail, Achatina fulica Bowdich, was introduced from Taiwan into Ogasawara in the 1930's. The population gradually increased during the 1940's and reach high densities in the 1950's. During this period, snails spread over the entire islands and caused serious damage to agricultural crops in spite of hand picking and baiting with metaldehyde. The predatory land snail, Euglandina rosea Ferussac, was introduced from Hawaii to Ogasawara in 1965. However, its population density was very low and it was restricted to a few areas in Chichijima Island. In 1985, we started ecological studies on the life history of A. fulica and the factors responsible for fluctuation of its population density. In 1986, the population density of A. fulica started to decline sharply at all the infested areas. The reasons for this sudden decline of population after a 30-year build up have yet to be identified. Introduction

The giant African snail, Achatina fulica Bowdich, has spread to the IndoPacific region from its original home of East Africa. In Japan, it is distributed in Ogasawara, Okinawa and Amami Islands (Koyano et al. 1989). In Ogasawara, A. fulica was first introduced to Chichijima Island in the 1930's (Mead 1961) and, in the 1950's, the population of A. fulica reached a high level in Chichijima Island (Mead 1961 ). Currently, A. juliea is distributed in Chichijima, Hahajima, lwojima and Marcus Islands. In the former two, A. fulica has caused serious damage to agricultural crops. A. fulica is designated as a key plant quarantine pest in Japan. Ecological studies of A. fulica have continued since 1985 with the aim of controlling it in Ogasawara. Life History

The life cycle of A. fulica in Ogasawara is shown in Fig. 1. In Ogasawara, the average temperature ranges from about l8째C in January and February to 1 Current address: Tokyo Metropolitan Agricultural Experiment Station, 3-8-1 Fujimi-Chyou, Tachikawa-shi, Tokyo 1983, Japan

110

Micronesica Suppl. 3, 1991

MARCH

Inactive

~ DECEMBER

~eath

(INTER

SPRING

--~----~--~--------~~------------~----~---JUNE

AUTUMN

SUMMER

• ••• ••

Egg

~ 1

y 1

SEPTEMBER

Figure 1. Life cycle of A. fulica on Chichijima and Hahajima Islands (modified from Yasuda & Suzuki).

27°C from July to September. The annual precipitation over the last 20 years has less than 1300 mm. Almost all the snails become inactive in late November and bury themselves in the soil. Activity resumes in April. Mating behavior has been observed in every season. The snails lay eggs from April to December, although most are laid between May and July with only a few between October and December. Aestivation is observed in the dry climate of mid summer. Autumn-born juveniles do not possess the tolerance for low temperature and are assumed to die. Thus, A. fulica seems to reproduce once a year in Ogasawara. Moreover, results of the outdoor research indicate that A. fulica needs one and a half to two years to grow from egg to adult. The life cycle of A. fulica in the Okinawa island is similar to Ogasawara (Suzuki and Yasuda 1983). Compared with other countries the life cycle in Japan is somewhat lengthy (Chang 1984). Distribution and Fluctuation of Population

Ogasawara Islands are located 27°N and 142°E in the Pacific, about 1,000 km south of Tokyo under subtropical conditions. Chichjima is the main island of Ogasawara and is about 25 km 2 with 1,500 residents. The central part of this

Takeuchi et al.: Giant African Snail in Bonin Is.

111

island is about 300 m above sea level and the residential area is limited to the northern part of the island. A. fulica habitats are confined almost exclusively to areas around human residence with hardly any noted in the natural forests (Aoki 1978). We investigated the distribution of A. fulica in this island in 1985 and 1989 (Fig. 2). In 1985, live snails were found in 39 sites out of 52. The population size was about 75% of that estimated by Numazawa et al. (1988). However, 4 years later, in 1989, the population had dwindled to 36%. The distribution of A. fulica is restricted to few areas. Similar reductions in population and distribution were observed in Hahajima Island. Snails were smaller in size in the residential areas compared with those in abandoned agricultural areas and around the natural forests. In the former the rate of reduction of the population density from 1985 to 1989 was higher than in the latter. The fluctuation of A. fulica population at one of the research sites in the residential area is shown in Fig. 3. More than 100 adult snails with shell lengths greater than 40 mm in 10 m 2 areas were found in 1985, compared with 30 in 1986, 10 in 1987, and only a few in 1988. Similarly, more than 60 juvenile snails per 10m2 with shell lengths of20 to 40 mm were found in 1985 and have become rare since 1986. Fig. 4 shows the fluctuation of the percentage of snails with eggs and the density of egg clutches, juvenile and adult snails in the site with white popinac, Leucaena glauca Benth. The percentage of snails with eggs and the clutch density were very low from 1985 to 1986 resulting in a lack of juvenile snails in 1986. Natural Enemies

There are some natural enemies of A. fulica in Ogasawara (Table 1). The most important one is the carnivorous snail, Euglandina rosea Ferussac which

Figure 2. Distribution of A. fulica on Chichijima Island in 1986 and 1989. Solid circle (e) present: Open circle (0), no snail.

112

Micronesica Suppl. 3, 1991

ADULT (>40nm)

0 .. ..... ..... .. .. ..... ...... ... .... ... ... .. ... ..... .. ....... ... ~--~--~- -········ · '- ·'··.. ··· ····· ····· ······· · ·-~--~--

80[

J U V EN I L E (20--40nin)

0 ---'j-A..._,......s_oL....'N.. a. .'_.0-~A~M~J ~ ~ tt<?U.......... 7 tt=:i'? ..........J_A...._S......._..O---:N~O-A~M-=-J"':-LJ~A~S_.O_N~D_.F~A~M~J.__J A-S........a..O-...a...D_,_F

1985

1986

1987

YEAR

1988

Figure 3. Population fluctuation of A. fulica at one of the research sites in the residential area on Chichijima Island.

1~1 ~

••

01 c<:S

• • :~t ~~~ J u v e n i I e (20- 40tml)

-----....

...

<l..>

J uven

:z:.

0.41 0.2

No. o f

c I u t c h e s / rd

%snails AS 0

1985

i I e ( < 20tml)

iloiloilA~-'loA~~~

0-'l£8£

with

eggs

]\_

B88~ss ~~o~ J AS 0 N AM

eeAenAss ssP j AS 0 N D M M J

J M MJ

1986

Year

1987

\{'\

1988

Figure 4. Fluctuation of the percentage of snails with eggs and the density of egg clutches, juvenile and adult snails in the site with White popinac, Leucaena glauca Benth on Chichijima Island.

was introduced from Hawaii in 1965. E. rosea was introduced to Hawaii from Florida in 1955 (Davis and Butler 1964). The E. rosea population increased suddenly 8 years after its introduction and reduced A. fulica population by about 20%. Since then E. rosea has been introduced to other Pacific regions. A. fulica

Takeuchi et al.: Giant African Snail in Bonin Is.

113

Table 1. Known species of natural enemies for A. fulica in Ogasawara. Vertebrata Bird: Mouse: Frog: Arthropoda Crab:

Hermit crab:

Mollusca

Land snail:

Platyhelminthes Flat worm: 1

Turdus dauma Latham 1 Rattus rattus Linnaeus R. norvegicus Berkenhout 1 Mus musculus boninensis 1 Bufo marianus Linne Geograpsus grayi2 (H. Milne-Edwards) Sesarma dahaani2 (H. Milne-Edwards) Metopograpsus messor2 (Forskal) Ocypoda cordimana 2 (Desmarest) Coenobita brevimanus Dana C. purpreus Stimpson C. perlatus H. Milne-Edwards Euglandina rosea Ferussac Bipalium sp. Geoplanidae sp. or spp.

Not confirmed by the present authors. From lga 1982.

density in Chichijima was very high for more than 30 years and it was hardly affected by E. rosea. The distribution of E. rosea was very limited, and also the population was low. Some of the terrestrial turbularian flatworms are said to be remarkable enemies of A. fulica. Muniappan (1986, 1987) reported that Platydemus manokwari de Beauchamp effectively reduced A. fulica in Guam and other islands. A different species of that worm occurs in Ogasawara. It does feed on A. fulica. It occurs in low density and its distribution is limited. Therefore, we consider it was not responsible for the reduction population of A. fulica. Some land crabs were reported to on feed A. fulica (lga 1982). Other natural enemies were also reported to feed on A. fulica, but no information is available on their effectiveness. Control

A. fulica has been controlled using metaldehyde pellets in farming areas. The recent reduction in population of A. fulica has reduced it to below pest density in Chichijima and Hahajima. A. fulica is distributed throughout Chichijima, while E. rosea occurs only in limited areas (Fig. 5). E. rosea also occurs near natural forests. More than 60 species of land snails endemic to Ogasawara occur in these natural forests (Kurozumi 1988). As these species are natural monuments, it is not advisable to

Figure 5. Distribution of two snails, A. fulica (e) and E. rosea (0), on Chichijima Island.

release E. rosea or P. manokwari in these areas. An understanding of the factors responsible for change in population density and life history are more important than biological control in Ogasawara.

Takeuchi et al.: Giant African Snail in Bonin Is.

Mead and Kondo(l949)

c 0

â&#x20AC;˘

Introduced ell ::J

c c...

1935

, ""

Introduced E.rosea

t ----------------

115

Starting decline

Return to Japan

Year

Figure 6. Supposed trend of A. fulica populations in Chichijima for last 55 years.

Conclusion

Since Ogasawara was returned to Japan, A. fulica remained at a high density for more than 30 years in spite of various control methods. A. fulica population has suddenly declined since 1986 in Chichijima and Hahajima islands (Fig. 6). The extremely low percent snails with eggs during 1985 and 1986 might have caused reduction in production of adults. Also our investigation to determine the possible climatic effects on the reduction A. juliea did not produce any positive results. Complex factors responsible for this phenomenon are currently being investigated. Our future research of A. fulica will mainly be aimed at understanding the factors that change the population of A. fulica. In addition, we plan to compare the population dynamics of other A. fulica populations in the Pacific region. Acknowledgements

We wish to express our thanks to Dr. 0. Mochida, Japanese government entomologist, National Agriculture Research Center, and Dr. R. Muniappan, University of Guam, for their helpful suggestions and critical reading of the manuscript. We also wish to thank Dr. K. Akutsu, N. Habu and M. Iga, Tokyo Metropolitan Agricultural Experiment Station for comments on the manuscript. Thanks are extended to Mrs. Y. Morimoto for her kind cooperation in preparing the English manuscript. References

Aoki, J. 1978. Investigations on soil fauna of the Bonin Islands. II. Ecological Distribution of the Agate Snail, Achatina fulica, and some possibilities of its ecological control. Edaphologia 18: 21-29.** *in Japanese **in Japanese with English summary

116

Micronesica Suppl. 3, 1991

Chang, W.-C. 1984. The cultivation of the Giant African Snail in commercial scale in Taiwan. Bull. Malacol. R.O.C. 10: 49-57. Davis, C. J. & Butler, G. D., Jr. 1964. Introduced enemies of the giant African snail, Achatina fulica Bowdich, in Hawaii. Proc. Hawaiian Entomol. Soc. 18: 377-389. Iga, M. 1982. Ecology and Control of Achatinafulica Bowdich. Japanese Journal of Applied Entomology and Zoology 36: 24-28.* Koyana, S., K. Numazawa & K. Takeuchi. 1989. Ecology of Giant African Snail in Japan. Plant Protection 43: 53-56.* Kurozumi, T. 1988. Species composition and abundance of land mollusks and factors affecting their extinction in the Ogasawara (Bonin) Islands. Ogasawara Research Nos. 14815: 59-109.** Mead, A. R. 1961. The Giant African Snail: a Problem in Economic Malacology. The University of Chicago Press, Chicago, U.S.A. Muniappan, R., G. Duhamel, R. M. Santiago & D. R. Acay. 1986. Giant African snail control in Bugsuk island, Philippines, by Platydemus manokwari. Oleagineux 41: 183-186. Muniappan, R. 1987. Biological control ofthe giant African snail, Achatinafulica Bowdich, in the Maldives. F.A.O. Plant Prot. Bull. 35: 127-133. Nunazawa, K., S. Koyano, N. Takeda & H. Takayanagi. 1988. Distribution and Abundance of the Giant African Snail, Achatina fulica Ferussac, in Two Islands, Chichijima and Hahajima, of the Ogasawara (Bonin) Islands. Jap. J. Appl. Entomology and Zoology. 32, No. 3: 176-181.** Suzuki, H., K. T. Yasuda. 1983. Studies on ecology and control of the giant African snail, Achatina fulica Bowdich, in Okinawa Island. Bulletin of the Okinawa Agricultural Experiment Station 8: 43-50.*

Micronesica Suppl. 3: 117-122, 1991

Possibilities for the Biological Control of the Breadfruit Mealybug, I eerya aegptiaca, on Pacific Atolls DOUGLAS

WATERHOUSE

ACIAR, GPO Box 1571 Canberra, ACT Australia 2601

Abstract-The introduced breadfruit mealybug, Icerya aegyptiaca is a major pest in Kiribati and some other Micronesian atolls. It has been brought under effective biological control in the high islands of Micronesia by the introduction of a predatory ladybird beetle, Rodolia pumila. Like most other ladybirds, this species searches effectively at high prey densities, but inefficiently at low densities. When introduced to the relatively restricted atoll environment it soon reduces I. aegyptiaca populations to low levels. It then dies out, apparently being unable to find prey, whereupon there is an upsurge in mealybug populations. One control strategy would be to re-introduce R. pumila whenever mealybug populations reach a pre-determined action level. However, and preferably, other natural enemies could be introduced that might co-exist at low scale densities. In particular, the parasitic fly, Cryptochetum grandicorne, may have appropriate characteristics. Origin and Distribution The Egyptian fluted scale or breadfruit mealybug, Icerya aegyptiaca (Douglas) is, despite its name, probably of Indian or Oriental origin. It was, however, first described in 1890 from specimens collected from a heavy infestation on fruit trees in Egypt. It now occurs widely in Asia, in tropical and subtropical Africa, in Australia and in Micronesia, but does not appear to be present in neighbouring Tuvalu. Doubt has recently been thrown on the authenticity of earlier records from Vanuatu, Fiji, and Tahiti (D. J. Williams pers. comm. 1990). In Micronesia it is known from the Federated States, the Marianas, Marshall Islands, Wake Island, Nauru and Kiribati (Waterhouse 1991). Beardsley ( 19 55) suggests that I. aegyptiaca may have gained entry into Micronesia from Taiwan. This was early this century or possibly late last century. At all events, the first record for the general region is 1893 from New South Wales (Maskell 1894). The first record from Kiribati, the country currently most seriously affected, is 1950 (Hall 1953). Life Cycle Most stages of I. aegyptiaca are present all year round in Micronesia. Only females are known: several casual references to males must be treated as dubious,

118

Micronesica Suppl. 3, 1991

since there are no specimens in collections. Unlike two related major pest species, I. purchasi Maskell and I. seychellarum (Westwood), which are both hermaphroditic, I. aegyptiaca is parthenogenetic. It has about 3 generations a year, with a peak of adult abundance in summer. Up to 200 eggs are laid into a waxy egg sac attached ventrally to the tip of the abdomen. The adult female is deep orange in color and has blackish legs. Marginally it has long, white, waxy processes and the body surface is covered with a white mealy secretion consisting of wax. Pest Status The greatest impact of I. aegyptiaca in the Pacific is on the breadfruit tree, Artocarpus altilis, which, together with coconut, provides essential food in low coral atolls. Although it may infest the fruit, the mealybug is usually situated along the midribs and larger veins on the underside of the breadfruit leaves. Large quantities of sap are sucked out by the mealybugs and this causes immature leaves and stems to dry up and die. Prolonged dry weather appears to favor the build up of heavy infestations. Heavy infestations have been reported to kill even mature breadfruit trees but, more often, trees are partially defoliated and the crop reduced, sometimes by more than 50%. In addition to direct effects from sap removal, the mealybugs produce large quantities of honeydew which acts as substrate for an abundant growth of sooty molds. This black growth covers the surfaces of all but the youngest leaves of heavily infested trees, seriously interfering with photosynthesis. Important food plants other than breadfruit that may suffer from heavy mealybug attack include banana, young coconut plants and citrus. Infestations may also occur on wild fig (te boro), babai, taro and many other plants. In its presumed Indo-Oriental native range I. aegyptiaca seldom causes damage and is most often found at low elevations in coastal regions. Nor is it a pest in Australia, although it is found from time to time on economic plants in the Northern Territory. Chemical Control Sprays containing white oil or some of the modern synthetic pesticides will control breadfruit mealybug but, under atoll conditions, these are expensive and difficult to apply to large trees. Natural Enemies The most important predators of Icerya are coccinellids of the genus Rodolia (Table 1), although several chrysopids also contribute to mortality. Of the ladybirds, R. cardina/is is best known for its spectacular control of the cottony cushion scale, Icerya purchasi in California, but R. pumila has been employed more often for biological control of I. aegyptiaca in Micronesia. Both species make massive

Waterhouse: Breadfruit Mealybug on Atolls Table 1.

119

Natural enemies of Icerya aegyptiaca (after Waterhouse 1990).

Predators

Parasitoids

COLEOPTERA Coccinellidae Coelophora inaequalis (Fabricius) Cryptolaemus montrouzieri Mulsant Harmonia arcuata (Fabricius) Pullus coccidovora (Anyar) Rodolia breviuscula Weise R . cardinalis (Mulsant) R . pumila (Weise) R . ruficollis Mulsant Rodolia sp. Scymnus sp.

DIPTERA Cryptochetidae Cryptochetum grandicorne Rondani Tachinidae Masicera sp. HYMENOPTERA Eulophidae Tetrastichus sp. Pteromalidae Oricoruna arcotensis (Mani and Kurian)

inroads into heavy Icerya infestations, but search inefficiently when prey densities are reduced to low levels. Of the other predators, R. ruficollis is reported in Pakistan to feed voraciously on I. aegyptiaca, but only in heavy infestations, whereas Pullus coccidovora attacks eggs and first instar larvae in both high and low populations and might be worthy of further investigation. Among the parasitoids (Table 1) the fly Cryptochetum grandicorne appears to be the most important, causing 20 to as high as 90% parasitisation of I. aegyptiaca in India and 3 to 10% in Pakistan. This suggests that, under appropriate conditions, it could be very important in population regulation, although it has not so far been employed as a biological control agent. The related species, C. iceryae (Williston), is a very important parasitoid of Icerya purchasi and, in coastal areas of California, is a far more effective biological control agent than the much publicized R. cardinalis (Quezada & De Bach 1973). C. iceryae was introduced into Chile, where it alone is reported to keep I. purchasi under effective control (Gonzalez & Rojas 1966). All of the 200 or so species of the family Cryptochetidae whose biology is known are parasitoids of the scale family Margarodidae (Ferrar 1987). C. grandicorne ranges in distribution from the Mediterranean to Asia, but does not occur in the oceanic Pacific. It is easy to rear in small cages in the laboratory, adults mating readily in sunshine, but rarely otherwise. Sometimes more than one egg is inserted, always into first instar mealybugs, but only one larva develops per host (Thorpe 1934). Insufficient is known about the other parasitoids to comment on their possible value as biological control agents. Attempts at Biological Control

The first attempt at biological control of I. aegyptiaca was of the outbreak in Egypt in 1892. Rodolia cardinalis, which had recently had such a spectacular

Micronesica Suppl. 3, 1991

120

impact on I. purchasi in California, was introduced and soon led to successful control. Micronesia is the only other region of the world where I. aegyptiaca has proved to be a serious problem. In this region there have been many attempts at biological control, a few of the early ones involving Rodolia cardinalis, but, more recently, Rodolia pumila has been the species used. The history of introductions is somewhat confused for several reasons. One is uncertainty relating to which of the three pest species of I eerya was actually present, namely I. aegyptiaca, I. purchasi, or I. seychellarum. Another complicating factor is that R. pumila, the only widespread coccinellid now attacking I. aegyptiaca in Micronesia, was brought in 1941 to Saipan, probably from Taiwan, but was at that time referred to as R. cardinalis, although lacking the latter's characteristic spots (Beardsley 1955). Actually R. cardinalis from Hawaii had been released against I. purchasi in 1926 on Guam and rapidly brought this mealybug under control. However it was last recorded in the region in 1945 (Nafus & Schreiner 1989). At all events, the outcome of the releases was that R. pumila had become established on most of the high islands of Micronesia by the 1950's (Beardsley 1955, Chapin 1965) and that I. aegyptiaca is no longer regarded as a pest in these high islands (Schreiner 1989). In contrast with these results, are those for atolls, where R. pumila has been repeatedly introduced, but appears not to be able to persist. Information is scanty on whether or when it has really become securely established, whether it has died out at some time after temporary establishment, or whether it might even have been present from an earlier introduction, but overlooked because of its low numbers. Sample records (Table 2) suggest that it Table 2.

Some introductions of Rodolia pumila to Micronesia (after Waterhouse 1990). Year

Established (recovery date)

Saipan Palau

1948 1954

+ (1950) + (1957)

Palau

1964

Truk State Nama

Losap

1954

Marshall Is Jaluit

Palau

1953 1958 1961

1964

Source Palau State Ulithi Atoll

Comment

but serious outbreak in 1964 but serious outbreak in 1984 later reported absent

(1954) but eliminated by typhoon

Waterhouse: Breadfruit Mealybug on Atolls

121

is unable to maintain itself on atolls for more than, at most, a few years. The situation is, perhaps, best documented in Kiribati (Table 3) where R. cardinalis was definitely established for a period on both Butaritari (1953) and Marakei ( 1962), but a few years later could not be found. R. pumila was introduced on at least 5 occasions and was established at least twice, leading both times to control, although it could not be found a year or two afterwards when I. aegyptiaca once again built up to pest numbers. The most likely explanation is that of Schreiner (1989), that on small atolls R. pumila died from starvation once mealybug populations were reduced to very low levels. Since Rodolia spp. are reported to be specific predators of I cerya and related scales, they would find it difficult, if not impossible, to find suitable alternative prey on most atolls which generally have a very limited insect fauna compared with that of high islands with their far more diverse habitats. Schreiner (1989) also suggests that typhoons may play a part in eliminating R. pumila, as apparently one did in Jaluit after the 1961 introduction. Although they may play an important role in some regions, typhoons occur only rarely near the equator, as in Kiribati, where elimination of R. pumila can occur without their intervention. Whatever the explanation, if Rodolia pumila is to be relied upon for the biological control of the breadfruit mealybug, arrangements will be necessary for re-introductions every few years, as suggested much earlier by Vandenburg ( 1928). If this tactic is to be adopted, mealybug populations should be monitored at regular intervals and introductions arranged when a pre-determined level of abundance is attained. A procedure should be established for regularly obtaining healthy stocks of R. pumila from a convenient source, possibly from Guam or Palau. Adequate precautions would be essential to prevent the introduction of damaging parasitoids or diseases at the same time. An alternative strategy would be to investigate the possibility of using as yet untested natural enemies, the first choice being the parasitic fly Cryptochetum grandicorne. Although it parasitizes only members of the scale family Margarodidae to which Icerya belongs, and hence on atolls would have few, if any, hosts other than Icerya available to it, it appears to have characteristics (such as Table 3. Introductions to Kirbati of Rodolia spp. (after Waterhouse 1990). Source

Rodolia cardinalis Fiji Hawaii

Rodolia pumila

Marianas Guam Palau Guam Palau

Year

Established

1953

1962

(?)1971 1975 1977 1978 1979

Established briefly

Established briefly on Butaritari, but later died out

Comment Established on Butaritari, but later died out Established on Marakai, but later died out

122

Micronesica Suppl. 3, 1991

requiring a single host only for development as contrasted with many for Rodolia) which may enable it to co-exist with /. aegyptiaca at low population densities. Thus, it might well survive at the low host densities, which R. pumila brings about. R. pumila should not be introduced simultaneously, but might be considered if C. grandicorne alone fails to produce an adequate reduction in mealybug populations. It is hoped that, in the near future, the Australian Centre for International Agricultural Research will fund the CSIRO Division of Entomology to evaluate C. grandicorne, and perhaps other natural enemies, for the control of breadfruit scale in Kiribati.

References Beardsley, J. W. 1955. Fluted scales and their biological control in United States administered Micronesia. Proc. Hawaii Entomol. Soc. 15: 391-399. Chapin, E. A. 1965. Coleoptera: Coccinellidae. Insects of Micronesia 16: 189254. Ferrar, P. 1987. A guide to the breeding habits and immature stages of Diptera Cyclorrhapha. Part 1. Entomonograph 8. E.J. Brill. Leiden. 478pp. Gonzalez, R. H. & S. Rojas. 1966. Estudio anclitico del control biol6gico de plagas agricolas en Chile. Agricultura Tecnica 26: 133-147. Hall, W. J. 1953. Outbreaks and new records. Gilbert Islands. FAO Plant Protection Bulletin 2: 44. Maskell, W. M. 1894. Further coccid notes: with descriptions of several new species, and discussion of various points of interest. Trans. Proc. N. Z. Inst. 26: 65-103. Nafus, D. & I. Schreiner. 1989. Biological control activities in the Mariana Islands from 1911 to 1988. Micronesica 22: 65-106. Schreiner, I. 1989. Biological control introductions in the Caroline and Marshall Islands. Proc. Hawaiian Entomol. Soc. 29: 57-69. Thorpe, W. H. 1934. The biology and development of Cryptochaetum grandicorne (Diptera), an internal parasite of Guerinia serratulae (Coccidae). Q. J. Microscopical Sci. (N.S.) 77: 273-304. Vandenberg, S. R. 1928. Report of the entomologist. Report of the Guam Agricultural Experiment Station for 1926: 15-19. Waterhouse, D. F. 1991. Prospects for the biological control of the breadfruit mealybug, Icerya aegyptiaca. Australian Centre for Agricultural Research, Canberra (in press).

Micronesica Suppl. 3: 123-127, 1991

Cultural Methods of Pest Control on Taro (Colocasia esculenta Schott) in American Samoa S. FATUESI, P. TAUILI'ILI, F. TAOTUA and A. VARGO Land Grant Program American Samoa Community College Pago Pago, American Samoa 96799

Abstract-In December 1989, a Rapid Rural Appraisal (RRA) of taro agriculture was held in American Samoa. A survey of 32 farmers revealed several traditional methods of insect, disease and weed control. Planting Coleus blumei Beuth in or around a plantation was alleged to control the cluster caterpillar (Spodoptera litura (Fabricius)) and the taro planthopper (Tarophagus proserpina (Kirkaldy)). "Smoking" a field with a torch of coconut fronds 3 times a week reportedly controlled the planthopper. Allowing chickens to roam taro fields in search of cluster caterpillars for food, and leaving a cluster caterpillar-infested field fallow for 3 to 5 months supposedly suppressed cluster caterpillar numbers. Control of weeds was accomplished by mulching with coconut fronds and banana leaves. Using the shade of papaya and banana trees and growing taro in high density or in an intercropped scheme were also identified as weed control measures. Pulled and slashed weeds are frequently used as mulch, and the mat of mulch is periodically turned over to break up any rooted weeds. Paragrass was cited as a "fertilizer" weed and "mile-a-minute" (Mikania micrantha Kunth) was regarded as a helpful plant. Herbicides promoted the growth of a different, more persistent weed population which was more difficult to control. Taro corm rot was controlled by selecting disease-free planting material, replanting in fallowed areas, and avoiding areas near Hibiscus tiliaceus L. trees. Introduction Pacific Islanders have developed traditional practices that help sustain their resources to support viable agricultural systems in their fragile ecosystems. These practices not only assist in maintaining soil fertility but also address the challenges of pest management. Taro (Colocasia esculenta Schott) is a staple crop in several Pacific Island agricultural systems. The purpose of this study is to document traditional pest management practices in the production of taro in American Samoa. Methods and Materials From October 31 to November 10 1989, a Rapid Rural Appraisal (RRA) of taro agriculture was held in American Samoa, bringing together multidiscipli-

124

Micronesica Suppl. 3, 1991

nary team from the areas of agricultural economics, agricultural extension, agronomy, animal science, environmental psychology, entomology, geography, soil science, plant pathology and weed science to document traditional agricultural practices associated with the growing of taro. Focus was given to pest management, soil fertility and soil conservation practices. Three teams, each with at least one member from each representative discipline, were organized to conduct the RRA interviews with the local farmers. To accommodate time constraints, 32 farmers were selected to interview. These farmers represented a broad range of taro producers in American Samoa-from commercial farmers to backyard hobby farmers and near-subsistence level farmers. Other factors considered in farmer selection were: age of farmer, years of farming experience, location of farm, site elevation, and slope of land. The RRA differs from other survey techniques in that the questions are open-ended, with subsequent questions evolving from answers given by the farmer. Consequently, the teams needed to discuss the answers received throughout the day in order to reformulate questions or redefine questioning strategy. Two basic questions were of primary concern: what the farmer perceived as major pests and what control measures were used to manage these pests. If additional pests were observed by the team, their presence was brought to the attention of the farmer for his opinion as to their pest status. Results and Discussion Four major pests of taro were identified, in descending order of importance: weeds, taro planthopper, taro armyworm and corm rot. Generalized farmer observations, opinions and control methods are listed for each pest. WEEDS

Weeds are regarded as the major obstacle to taro production in American Samoa because the farmer reportedly spends most of his time at weed control. Major weeds include Honolulu rose (Clerodendrum philopinum Schau) and Paspalurn sp. Some farmers regard a few types of weeds as beneficial. Many farmers used pulled weeds as mulch between taro plants. Sometimes this mat of mulch is turned over several times during the growing season in order to disrupt any rooted weeds. Besides functioning as a weed suppressant, the mulch slowly decomposes, thus providing a recyclable nutrient source for the growing taro or subsequent crops. Mikania micrantha Kunth, commonly known as "mile-a-minute", is a ubiquitous, fast-growing vine that is regarded as a "helpful" weed by many farmers. It is wrapped around banana plants during drought periods, in order to catch dew and hold moisture near the banana. Some farmers report Mikania to be the "weed of choice" to overrun a taro plantation during fallow periods, as it is easy to pull and leaves the soil moist and friable.

Fatuesi et al.: Cultural Methods of Taro Pest Control

125

In addition to mulching with cut weeds, weed control practices include mulching with coconut fronds and banana leaves, planting taro in a high density scheme, or intercropping with a variety of other groups, such as banana, papaya, breadfruit, etc., to provide shade to suppress weed growth. While about 60% of the surveyed farmers used the nonspecific herbicide Paraquat, use throughout the 6 to 10 month growing season was minimal. Paraquat application can be categorized generally into three main schemes: 1. Spraying during preplanting only, using Paraquat 3 to 14 days before planting. This may or may not be followed by slashing these dead weeds for mulch. 2. Spraying 1 to 3 times during the growth of the taro. 3. Spraying initially to clear the field, then handweeding the rest of the time. The advantages and disadvantages of Paraquat were clearly recognized by the farmers. Though an additional expense, herbicides were seen as labor-saving devices which allowed the farmer to service a large area in a short time when compared to traditional methods. However, herbicides were also criticized for causing leaf damage, "spoiling" both the soil and taro corms, and excessive expense. Some farmers noticed that the use of herbicides promoted the growth of different weeds, usually grasses, which were more difficult to control. Farmers also observed that weeds grew back more quickly in fields sprayed with herbicides than those where weeds were pulled or slashed. INSECTS

Taro Planthopper (Tarophagus proserpina Kirkaldy) Farmers report that the taro planthopper can be totally devastating to the crop if there is a severe attack on plants under 3 months old. More commonly, the taro planthopper is an occasional pest on taro plantations. Farmers have developed several strategies for controlling planthopper populations. Planting Coleus blumei Benth, pate in Samoa, around or within a field is one of them. Various explanations are offered to explain its mode of action. Some believe that the smell of pate repels the planthopper. Others contend that pate attracts the planthopper, which then feeds on its supposedly toxic juices and dies. Opinions vary widely regarding pate's effectiveness: belief that it works sometimes; belief that it works only when it is first planted; belief that it is ineffective in controlling any insects; and belief that it harbours nematodes which affect the banana. The dark red or purple-colored variety of pate is thought to be the most effective in controlling the planthopper. Another method of planthopper control is through "torching" or "smoking" the plantation. Coconut leaves are bundled and then set afire. Walking through the fields with these torches causes the planthopper to jump from the plant. The belief is that the planthoppers die as their wings get singed or that the smoke drives them from the field. This practice is performed 3 times a week until the planthoppers are gone. "Smoking" is sometimes thought to be effective against the cluster caterpillar as well.

126

Micronesica Suppl. 3, 1991

A predator of planthopper eggs, Cytorhinus fulvus Knight, lives in close association with the planthopper. It is rarely noticed or recognized by farmers as a biological control probably because of its small size and cryptic hiding behavior. Cluster Caterpillar (Spodoptera litura (Fabricius)) Cluster caterpillars are perceived as a long established pest in American Samoa. Outbreaks reportedly occur every 3 to 5 years and are location specific. Severe infestations often occur after hurricanes. Some believe that cluster caterpillars appear after alternating short periods of rain and sun. Others report "more worms" during the dry season. Honolulu rose is an observed alternate host of the cluster caterpillar. Farmers have devised various strategies for dealing with the cluster caterpillar. One method is planting pate in taro fields. Some believe that its scent repels the cluster caterpillar, while others conjecture the caterpillar prefers pate to taro. Another method of control is smashing the cluster caterpillars and their eggmasses, or handpicking and burning them. Chickens are often brought to, or raised in the field because they reportedly peck the cluster caterpillars from the plant. The scratching and feeding of the chickens around the base of the plant was also mentioned as a good weed control as well as advantageous in aerating the soil around the plant. Allowing the weeds to grow unchecked in a cluster caterpillar-infested field is another control strategy. Farmers believe that when weeds take over the field, the taro will be "hidden from the cluster caterpillar" and the cluster caterpillar will leave the field. After 3 months taro can be planted again with success. PLANT DISEASE

Corm rot is the most serious plant disease of taro and reportedly can be controlled by selecting disease free planting material. Additionally, replanting taro in a "different hole" will prevent the problem. Taro planted near the "fau" tree (Hibiscus tiliaceus L.) was reported to be more susceptible to corm rot. Farmers advised that taro should not be planted near coconut trees unless the soil is deep, since the shallow roots of the coconut would interfere with the growth of the taro corm. There is also a problem with corm rot developing on taro planted in rocky areas when the saprolite layer is near the surface. Discussion

In American Samoa, traditional methods of pest management have been used to combat a number of pests. Weeds have generally been controlled by hand pulling, slashing and mulching with weeds, coconut fronds, and banana or other leaves. These methods not only suppress weeds but serve to replenish the soil through nutrient recycling. Insect damage is generally minimal, possibly due to several interacting mechanisms. For instance, taro is planted either in monoculture or is intercropped with one more other crops such as banana, coconut or yams. When monocultured,

Fatuesi et al.: Cultural Methods of Taro Pest Control

127

it may be in an isolated plot in the forest or surrounded by weeds, or other crops. In either case, it seems probable that the plant diversity of these systems offers increased environmental opportunities for natural enemies, thereby increasing biological control (Altieri & Letourneau 1980). The multi-story, dense foliar cover associated with intercropping is less visually stimulating to insects which cue in on plants silhouetted against a contrasting background, such as the bare earth or the bright horizon. Similarly, plant diversity provides camouflage and thereby renders the at-risk crop less visible to pests (Perrin & Phillips 1978). To complement the planting strategy method of pest control, Samoans have also adapted mechanical methods, such as smashing insect eggs and larvae, and up-rooting or slashing weeds. Though time consuming, these methods are environmentally sound and conform to the traditional Samoan practice of the family working together. It is common to see an entire family, from children to the elderly, involved in traditional taro production, even in the area of pest management. The planting of pate, the use of chickens and the use of smoking torches are methods developed over generations of careful observation. While statistically sound evidence of their effectiveness has yet to be collected, continuous, widespread practice of these methods offers some credence in their capabilities. Also, the potential use of these methods at other locations offers hope in the continuous search for effective, alternative methods of pest control. Acknowledgements We would like to thank all those who comprised the LISA survey team in American Samoa. Special appreciation is due to Joe O'Reilly for the RRA methods, Larry Hirata and the Cooperative Extension Service of the American Samoa Community College for their assistance in selecting farmers and community representatives to interview, the Land Grant Staff for their logistic support throughout the RRA arid to Sharon Hanover for her assistant in the final preparation of this paper. This research was supported by the Low Imput Sustainable Agriculture Grant of U.S. Department of Agriculture. References Altieri, M. A. & D. K. Letourneau. 1982. Vegetation management and biological control in ecosystems. Crop Protection 1: 405-430. Perrin, R. M. & M. L. Phillips. 1978. Some effects of mixed cropping on the population dynamics of insect pests. Ent. Exp. Appl. 24: 385-393.

Micronesica Suppl. 3: 129-133, 1991

Automated Identification of Insects in Flight A. MOORE 1 Agricultural Development in the American Pacific Project Agricultural Experiment Station, University of Guam, Mangilao, Guam 96923

Abstract-Insect wingbeat frequency has been used as a taxonomic character. To test the feasibility of developing instrumentation which monitors the identity and population density of flying insects, wingbeat frequencies of the mosquitoes, Aedes aegypti (L.) and A. triseriatus Say, were recorded using a microcomputer-based instrument which measured light reflected off the wings of individuals in flight. The wingbeat frequency and other spectral components from 403 individual recordings were used to train an artificial neural network. The trained network correctly identified the species and sex of mosquitoes in an independently recorded group of 57 mosquitoes. This technology has potential use for detecting and monitoring a wide range of flying insects. Interspecific and intraspecific differences in wingbeat frequencies have been used to identify insects in flight (Reed et al. 1942, Sotavalta 1947, Sawedal & Hall 1979, Greenbank et al. 1980, Farmery 1982, Riley et al. 1983, Schaefer & Bent 1984, Unwin & Corbet 1984, Rose et al. 1985). To assess the feasibility of developing instrumentation which monitors the identity and population density of flying insects, Moore et al. ( 1986) measured wingbeat frequencies for individuals from two species of mosquitoes, Aedes aegypti (L.) and A. triseriatus Say. by recording changes in the intensity of light reflected off wings during free flight (Fig. 1). Spectral analysis showed that each recording contained the wingbeat frequency plus several harmonics (Fig, 2). A univariate discriminant function calculated using wingbeat frequencies from one group of mosquitoes (which will be referred to as the 'training set') was used to identify the species and sex of individuals from a second, independently reared group ('test set') with an accuracy of84%. Multivariate discriminant functions calculated using wingbeat frequencies and amplitudes of the first four harmonics did not improve accuracy, implying that characteristics of the frequency spectrum, other than the wingbeat frequency, were not useful for identification. However, recent analysis of data from this experiment using a new pattern recognition technique from the field of artificial intelligence shows that the frequency spectra are rich in information useful for identification. An artificial neural network (Stanley 1989) was trained using the Present address: Department of Entomology, CTAHR-Maui Research, University ofHawaii, P. 0. Box 269, Hawaii 96790. 1

130

Micronesica Suppl. 3, 1991

SENSOR

•

PhotoDiode Amplifier

EQUIPMENT FOR ON-LINE ANALYSIS

Analog-to-~::

Digital ~~: Converter =~~ •a:•:•:•:•:•:•:•:•o"o"o"o"o"o"•:•:•:•:•:•:•:•:•o::

Computer ··:·:·:·:·:·:·:·:·: ·:·:·:·:·:·.·.··:·:·:·:·!·!•!·:·:·:·:·:·:·:·:·:·

EQUIPMENT FOR OFF-LINE ANALYSIS

Figure 1. Schematic diagram of equipment for recording and analyzing insect wingbeat frequencies.

training set and correctly identified the species and sex of every mosquito in the test set with a high degree of confidence. This new technology may be useful in developing a remote sensing monitoring system which automatically counts and identifies insects in flight. Following is a brief description of the experimental methods. Details can be found in Moore et al. ( 1986). An instrumentation system for recording and analyzing insect wingbeat frequencies was constructed (Fig. 1). The photosensor, consisting of a photodiode and amplifier (Unwin & Ellington 1979), detected fluctuations in light intensity caused by reflections off individual mosquitoes flying through a light beam. Digital recordings of the signals were made with a microcomputer (IBM PC) equipped with an analog-to-digital converter (LabTender, Tecmar Inc., Solon, Ohio) under the control of a program which simulates a digital oscilloscope (SCOPE2, Moore Scientific, Kula, HI). A change in light intensity caused by a mosquito flying in front of the sensor triggered storage of 512 samples at a rate of 10 kHz (Fig. 2A). Each signal was converted to a 256 cell frequency spectrum using the fast Fourier transform (Cooper 1981). The frequency spectrum for each signal contained a harmonic series with the fundamental at the wingbeat frequency (Fig. 2B). In the original analysis, the wingbeat frequency and amplitudes of the first four harmonics were extracted from each signal in the training set (n = 403; approximately 100 signals for each species-sex combination). Discriminant functions based on several combinations of these variables were calculated and were tested by using them to identify signals from the test set (n = 57; approximately 15 signals for each species-sex combination). The function based on wingbeat frequency alone identified the correct species and sex 84% of the time. Accuracy did not improve when the amplitudes of the harmonics were used in the calculations.

Moore: Automated Identification

131

1.-----------------------------------------------~

::> ....

:J (L

::E <(

-1

-2+---~~--~----~----~--~----~----~--~----~--~

TIME (ms)

0.3.----------------------------------------------

~ 0.2

....::>

0.1

1000

2000

3000

FREQUENCY (Hz)

4000

Figure 2. (A) Signal produced by the flight movements of a female Aedes aegypti mosquito. (B) Frequency spectrum.

5000

132

Micronesica Suppl. 3, 1991

In the recent analysis, an artificial neural network in which the input layer contained 256 neurons (one for each cell in the frequency spectra), the hidden layer contained 127 neurons, and the output layer contained four neurons (one for each of the four species-sex combinations). Each fact in the training set consisted of 256 numbers representing the amplitude of each cell in the frequency spectrum plus one of four identifying codes corresponding to the species-sex combination of the mosquito which produced the signal. After training to completion, the network was tested by using frequency spectra of signals in the test set as input. The species and sex of the mosquito causing each signal were identified correctly. Furthermore, each insect was identified with a high degree of confidence as indicated by the fact that the relative firing rate for the output neuron representing the correct species-sex combination ranged from 88% to 100% with a mean of 98%. This study demonstrates the feasibility of developing instrumentation capable of counting and identifying insects in flight. Even though morphologically similar species were used in the experiment, each signal, lasting only one-twentieth of a second, contained information enabling identification of species and sex with a high degree of confidence. With further development, this type of instrumentation could become an important tool for research and pest control. A flight monitor designed for the field could use either the sun or an appropriate artificial light source, such as a red or infra-red laser. Possible applications include continuous monitoring of several sympatric populations (useful for ecological and biological control studies), pollination studies, measurement of diurnal activity cycles, and evaluation of attractants and repellents. The system I envision will be able to count and identify several species flying through a defined volume of airspace. Such a monitor will be a useful entomological tool. A medical entomologist could use it to determine what species of mosquitos are present, how numerous they are, and when they are flying. A biocontrol specialist might monitor the flight activity of a host insect and several parasitoids and predators. References

Cooper, J. W. 1981. Introduction to Pascal for Scientists, Wiley, New York. Farmery, M. J. (1982). The effect of air temperature on wingbeat frequency of naturally flying armyworm moth (Spodoptera exempta). Ent. Exp. Appl. 32: 193-194. Greenbank, D. 0., G. W. Schaefer & R. C. Rainey. 1980. Spruce budworm (Lepidoptera: Tortricidae) moth flight and dispersal: new understanding from canopy observations, radar, and aircraft. Mem. Entomol. Soc. Can. 110. Moore, A., J. R. Miller, B. E. Tabashnik & S. H. Gage. 1986. Automated identification of flying insects by analysis of wingbeat frequencies. J. Econ. Entomol. 79: 1703-1706. Reed, S C., C. M Williams & L. E. Chadwick. 1942. Frequency of wing-beat as a character for separating species races and geographic varieties of Drosophila. Genetics 27: 349-361.

Moore: Automated Identification

133

Riley, J. R., Reynolds, D. R., and Farmery, M. J. (1983). Observations of the flight behaviour of the armyworm moth, Spodoptera exempta, at an emergence site using radar and infra-red optical techniques. Ecol. Entomol. 8: 395-418. Sawedal, L. & R. Hall. 1979. Flight tone as a taxonomic character in Chironomidae (Diptera). Ento mol. Scand. Suppl. 10: 139-143. Schaefer, G. W., & G. A. Bent. 1984. An infra-red remote sensing system for the active detection and automatic determination of insect flight trajectories (IRADIT). Bull. Entomol. Res. 74: 261-278. Sotavalta, 0. 1947. The flight-tone (wing-stroke frequency) of insects (Contributions to the problem of insect flight 1.). Acta Entomol. Fenn. 4: 1-114. Stanley, J. 1989. Introduction to Neural Networks. California Scientific Software, Sierra Madre. Unwin, D. M. & S. A. Corbet. 1984. Wingbeat frequency, temperature, and body size in bees and flies. Physiol. Entomol. 9: 115-121. Unwin, D. M. & C. P. Ellington. 1979. An optical tachometer for measurement of the wing-beat frequency of free-flying insects. J. Exp. Bioi. 82: 377-378.

MICRONESICA Devoted to the natural sciences of Micronesia and related areas Some recent subjectsMarine algae of New Caledonia Five new wrasses from the Marshall Is. Population history of Nauru Vascular plants of Puluwat Atoll Biological control in the Marianas Seven new species of hermit crab Isotope variation in Holocene reef corals Cook Islands plants Long-distance seed dispersal New genus of parasitic crustacean Freshwater algae from Yap Marianas bird records Mammals of Tinian Chamorro fish names Chamorro fertility Reptiles of Rota Micronesian high island horticulture New orchid from Guam Chemistry of Guam waters

MICRONESICA ORDER FORM

0 0 0

Please enter my subscription to start with vol. 24 ( 1991) or , at _ _ $15 (individual) or _ _ $25 (institution), post paid per year. Most back issues are available: please enquire. Please send _ _ copy(ies) of Supplement 2 (Micronesian Archeology, 1990) at $35, post paid. Please send _ _ copy(ies) of Supplement 3 (Exotic Pests in the Pacific, 1991) at $7.50 post paid. Payment (U.S. check or money order) or purchase order enclosed. 0 Please bill me.

Name------------- Institution------------Mailing a d d r e s s - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Z i p _ _ _ _ Country _ _ __ Billing address (if d i f f e r e n t ) - - - - - - - - - - - - - - - - - - - - - -

MICRONESICA A Journal of the University of Guam Devoted to the Natural Sciences in Micronesia The

journal of natural sctenct:~ Ill Llw ·,v. ~·~¥... P"rific

EDITOR:

CHRISTOPHER

s. LOBBAN

The Marine Laboratory, University of Guam UOG Station, Mangilao, Guam 96923, U.S.A.

If your research area includes

• • •

anthropology, archeology, biogeography, coral biology, crustacean systematics, demography, ecology ethnology, geology, herpetology, ichthyology, island biology, mammalogy, marine biology, Pacific studies, phycology, reef studies, tropical vegetation, or zoology ...

You should see Micronesica regularly. Micronesia is a region with tremendous cultural and biological diversity that attracts researchers from all over the world. Micronesica is a fully-refereed journal with international authorship and readership that acts as a forum for studies on all <!spects of the natural sciences, broadly interpreted. Because the region comprises numerous widely scattered islands, it provides models of island biology and island cultures, and its place near the center of Indo-West Pacific biological diversity makes it important in tropical biology. We bring you the scientific richness of the region twice a year. See sample of recent papers overleaf Micronesica is published in June and December each year

SUBSCRIBE! and ask your library to subscribe Subscription form overleaf

INSTRUCTIONS FOR CONTRIBUTORS Scientific research reports, notes, review papers, bibliographies, and book reviews in anthropology, biology, and related fields are accepted on the basis of their originality and their pertinence to Micronesia and the adjacent Pacific areas. Descriptions of new species will be considered formal papers, no matter how short; information in range extensions will be considered notes, no matter how long. The manuscripts must be written in English, but a summary in another language is acceptable. Each manuscript will be reviewed by at least two members of the Editorial Board or by specialists other than board members in whose field the paper lies. Manuscripts should be sent via airmail to The Editor, The Marine Laboratory, University of Guam, UOG Station, Mangilao, Guam 96923, U.S.A. The original and two clear copies of text and artwork are required; the original will be retained in the editorial office while the copies are sent out for review. Authors must follow the guidelines printed in the June issue (from 1990), or available from the Editor. Papers which deviate from the required format may be returned for revision before review.

9 â&#x201E;˘ The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences-Permanence of Paper for Printed Library Materials, ANSI Z39 .48-1984.

Typeset, printed and bound by Edwards Brothers, Ann Arbor, MI