Pacific Science (1984), vol. 38, no. 2
© 1984 by the University of Hawaii Press. All rights reserved
The Extinction of Endemic Species by a Program of Biological Control1
Bryan Clarke,2 James Murray,3 and Michael S. Johnson4
ABSTRACT: Land snails of the genus Partula, inhabiting the high islands of the Pacific Ocean, have provided exceptional opportunities for studies on the origin and differentiation of species: The endemic taxa of Moorea, in French Polynesia, have been particularly well studied.
In an attempt to control the numbers of the giant African snail, Achatina fulica, which is an agricultural pest, a carnivorous snail, Euglandina rosea, has been introduced into Moorea. It is spreading across the island at the rate of about 1.2 km per year, eliminating the endemic Partula. One species is already extinct in the wild; and extrapolating the rate of spread of Euglandina, it is expected that all the remaining taxa (possibly excepting P. exigua) will be eliminated by 1986-1987.
Euglandina has been introduced into many other oceanic islands, and it appears that more than a hundred endemic species are at risk. These observations point to a serious danger in programs of "biological control."
Since the time of Darwíns the endemic species of oceanic islands have held a special interest for students of evolution. Archipelagos of volcanic origin, such as the Galapagos and Hawaiian islands, are natural laboratories in which the phenomena of speci-atiort can be studied (Lack 1947, Carson and Yoon 1982). The fauna and flora of such islands are, however, unusually susceptible to the introduction of competitors, parasites, and predators.
In this paper we report the progressive extinction of land snails (genus Partula), endemic to the island of Moorea in French
1 Manuscript accepted 10 January 1984.
2 Museum National d'Histoire Naturelle et École Pratique des Hautes Etudes en Polynésie Frangaise, Centre de l'Environnement de Mooréa, B.P. 12, Mooréa, French Polynesia. Permanent address: University of Nottingham, Department of Genetics, University Park, Nottingham NG7 2RD, England.
3 Museum National d'Histoire Naturelle et École Pratique des Hautes Etudes en Polynésie Frangaise, Centre de l'Environnement de Mooréa, B.P. 12, Mooréa, French Polynesia. Permanent address: University of Virginia, Department of Biology, Charlottesville, Virginia 22901.
4 University of Western Australia, Department of Zoology, Nedlands, Western Australia 6009.
Polynesia, by an introduced molmscan predator (Euglandina rosea (Ferussac)).
The classic studies of Crampton (1932) and later investigations (Murray and Clarke 1980, Murray, Johnson, and Clarke 1982) have shown that these snails provide exceptional opportunities for study of the origin and differentiation of species. Their unique combination of ovoviviparity, low mobility, short generation time, ease of culture in the laboratory, and extensive genetic polymorphism (at both the morphological and molecular levels) make them almost ideal material for the study of ecological and evolutionary genetics.
Their loss is not merely a tragedy for students of genetics, it is also a Warning about the potentially devastating effects of some programs in "biological control"; and it exemplifies a threat to more than a hundred other species. Because the Partula of Moorea have been mapped in great detail, we are able to document rather precisely the progress of extinction.
the Partula of moorea
Moorea is one of the Society Islands and lies about 20 km to the northwest of Tahiti, in almost the exact geographical center of the
97
98
PACIFIC SCIENCE, Volume 38, April 1984
MAHAREPA
PAPETOAI
N
AFAREAITU
MAATEA
Figure 1. The island of Moorea, showing the principal topographical features.
Pacific Ocean. The island is volcanic in origin, and its age is approximately 1.2 million years (Jackson 1976). It is about 12 km in diameter, and its highest peak rises to 1207 m. Nine other peaks exceed 700 m (Figure 1).
Crampton (1932) and Crampton and Cooke (1953) described eleven species of Partula, all endemic to the island. Two of these {Partula solitaria Crampton and Partula diaphana Crampton and Cooke) have since been relegated to the genus Samoana (Kondo 1973). Of the remaining nine, one (P. den-droica Crampton) is clearly a geographical race of P. suturalis Pfeiffer and another (P. olympia Crampton) a geographical race of P. tohiveana Crampton (Clarke and Murray 1969, Murray and Clarke 1980). The number of Partula "species" is thus reduced to seven (P. taeniata Morch, P. exigua Crampton, P. suturalis Pfeiffer, P. tohiveana Crampton, P. aurantia Crampton, P. mooreana Hartman,
and P. mirabilis Crampton). Their status as good biological species is, however, in question because each, at some point on the island, gives evidence of hybridization or intergra-dation with another. For example, P. tohiveana tohiveana and P. suturalis are the sympatric terminal elements of a ring-species including P. tohiveana olympia. The pattern is such that, potentially at least, genes could flow from any one "species" to any other (with the possible exception of P. mooreana). Nonetheless, at any particular place as many as four of the taxa may coexist without interbreeding. When they do so, they are clearly distinct in genetics, morphology, ecology, and behavior (Johnson, Clarke, and Murray 1977, Murray and Clarke 1980, Murray, Johnson, and Clarke 1982).
The geographical ranges of the taxa prior to the introduction of Euglandina are shown in Figure 2 (pages 99 and 100).
Extinction of Endemic Species—Clarke, Murray, and Johnson
99
Figure 2. The ranges of the different species of Partula on Moorea prior to the introduction of Euglandina rosea: (A) Partula taeniata (N = P. taeniata nucleola, Sp = P. t. spadicea, St = P. t. striolata, Si = P. t. simulans, E = P.t. elongata); (B) Partula exigua (E) and P. mirabilis (M); (C) Partula suturalis (V = P. suturalis vexillum, S = P. s. strigosa, D = P.s. dendroicd); (D) Partula mooreana (M), P. tohiveana (O), and P. aurantia (A).
Extinction of Endemic Species—Clarke, Murray, and Johnson
101
THE INTRODUCTION OF Achatilta AND
Euglandina
The giant African snail, Achatina fulica Bowdich, is well known as an agricultural pest (Mead 1961, 1979). It was introduced into Tahiti (by someone who wished to breed it for food) in 1967 and spread rapidly to the other islands in the archipelago. Its numbers increased in Moorea to such an extent that the snails invaded human dwellings, and on one occasion two wheelbarrow-loads of snails were taken from the walls of a single house (R. Brosious, pers. comm.). Because Achatina, unlike Partula, eats living plants, it caused great damage to crops and gardens. The numbers of A. fulica apparently reached a peak on Moorea in 1978, and thereafter declined. The cause of this decline is unknown (see below), but the very large numbers of empty unbroken shells found all over the island suggest an epidemic disease (or perhaps predation by land planarians; see Mead 1979: 70). Living Achatina are still common on Moorea, and still cause damage to crops, but the scale of the problem has greatly diminished.
Euglandina rosea was introduced into Tahiti and Moorea in an attempt at the "biological control" of Achatina fulica. Euglandina rosea, which is native to Florida and central America, is a carnivorous mollusc that preys upon other land snails. Although it certainly eats Achatina, there is, as far as we know, little evidence that it is an effective agent of control (Mead 1979). Nevertheless, it has been widely recommended for this purpose by departments of agriculture in various countries. It is still so recommended by the South Pacific Commission.
Euglandina rosea was introduced into Moorea, with the approval of the Service de L'Economie Rurale and the Division de Recherche Agronomique, at the orange plantation of M. Nardi in Paopao on 16 March 1977 (J. L. Reboul, pers. comm.). It spread into the adjoining forests, and by 1980 its range covered approximately 4 km2, very nearly a third of the island. Its rate of spread between 1980 and 1982 was approximately 1.2 km per year (Figure 3). If the spread continues at the present rate, Euglandina will occupy the whole island by 1986 or 1987.
the effect of Euglandina on Achatina
The decline in the numbers of Achatina fulica on Moorea since 1978-1979 cannot rigorously be ascribed to the effects of Euglandina rosea. Living Achatina are still to be found in the territories occupied by Euglandina, and the decline in the number ofAcha-tina has been observed in parts of the island not yet reached by the carnivore, for example Aareo Valley and Vaihiiaiia Valley. Furthermore, a similar decline has occurred on the island of Huahine, to which E. rosea has not yet been introduced (Pointier and Blanc 1982).
the effect of Euglandina on Partula
Although the role of Euglandina in controlling Achatina is doubtful, there is unfortunately no doubt at all about its effects on Partula. It consumes them with great efficiency. Four individuals of P. taeniata were put in a plastic lunch-box with a single E. rosea. Within 24 hours they had all been eaten, leaving four empty shells.
During our field studies in 1982 we were unable, with one exception discussed below, to find any Partula in the territories now occupied by Euglandina rosea. Our 1982 surveys specifically included localities (in Paparoa, Mouaputa, Faamaariri, and Puutu Valleys) that had yielded large samples of Partula in 1967 and 1980. Four species (P. aurantia, P. suturalis, P. tohiveana, and P. taeniata) seem to have been entirely eliminated from the northeast of the island. Partula aurantia, which was restricted to this area (see Figure 2), must now be considered extinct in the wild state (the only living examples being those kept alive by James Murray in the laboratory). The subspecies P. taeniata striolata is also, we believe, extinct.
The single exception is Partula exigua. This taxon is restricted to the northeast (see Figure 2), although it is closely related to P. taeniata and hybridizes with it (Murray and Clarke 1980). Prior to the introduction of Euglandina it tended generally to be less common than its sympatric relatives (P. taeniata, P. suturalis, and P. aurantia). In 1982 we found a few scattered   individuals    of   P.    exigua   or
102
PACIFIC SCIENCE, Volume 38, April 1984
Figure 3. The spread of Euglandina rosea on Moorea. The black circle represents the site of its introduction in 1977. The vertically hatched area represents its range in 1980, and the horizontally hatched area its range in 1982. The black dotted line represents its expected range in 1984, and the serrated line its expected range in 1986. The star represents the site of a secondary introduction (in Faataai Valley) the success of which is unknown. If it proves successful, the spread of Euglandina is likely to be faster than shown in this figure.
taeniata/exigua hybrids surviving where the other Partula were extinct (in Faamaariri and Mouaputa Valleys). We are not yet able to give a reason for the "escape" of some P. exigua, or to tell whether the phenomenon is transient or lasting.
An incidental consequence of the spread of Euglandina is the raising of Partula tohiveana to the status of a good biological species. Until 1980 P. tohiveana and P. suturalis formed a ring-species, the zone of intergrad-ation being to the east of Mt. Mouaputa (Murray and Clarke 1980). Populations from this zone have now been eliminated, and the remaining populations of tohiveana are sym-patric with, and distinct from, suturalis.
The rate at which Euglandina is spreading (Figure 3), and the known ranges of the Partula taxa (Figure 2), allow us to make clear predictions about the years in which the taxa are expected to become extinct in the wild. These are given in Table 1. The majority are expected to die out by 1986.
THE THREAT TO Partula ON OTHER ISLANDS
H. E. Crampton, in addition to his work on Moorea, wrote two further evolutionary monographs, one on the Partula of Tahiti (1916), the other on the Partula of Guam and Saipan (1925). These snails, too, are under
Extinction of Endemic Species—Clarke, Murray, and Johnson
103
threat. In Tahiti Euglandina rosea was established during 1974 and now occupies the whole of the Temarua Valley, above the village of Papara, and the entire plateau of Taravao, above the village of Afaahiti. In these areas the native Partula appear to have been eliminated (Pointier and Blanc 1982). Because Tahiti, with a diameter of about 30 km, is a larger island than Moorea, some of the Tahitian Partula may survive longer than the Moorean ones, perhaps until the 1990s. The situation is probably worse in Guam and Saipan. Not only is Euglandina rosea well established on both islands, but two other carnivorous snails {Gonaxis quadrilateralis (Preston) and Gonaxis kibweziensis (E. A. Smith)) also have been introduced (Mead 1979).
In view of the rapid spread of Achatina among the Society Islands, both by accidental and by intentional transport, it seems very probable that Euglandina will reach many other islands within the next few years.
THE THREAT TO OTHER GENERA OF LAND SNAILS
We have not studied the effect of Euglandina on the two Moorean species of Samoana {Samoana diaphana (Kondo) = Partula diaphana Crampton and Cooke, and Samoana attenuata (Pease) = Partula solitaria Crampton), but if they prove to be as susceptible to Euglandina as the other Partulidae we can expect them to be eliminated from the island before 1987. The same is true of the arboreal snail Trochomorpha pallens Pease, which seems now to be absent from the territories occupied by Euglandina.
The introduction of Euglandina and other carnivores has been implicated in the extinction of many endemic snails in the Hawaiian islands (van der Schalie 1969, Hadfield and Mountain 1980, Kondo, pers. comm.). Euglandina has been successfully introduced into Mauritius, the Seychelles, La Reunion, the New Hebrides, and New Caledonia (Mead 1979). The number of endemic species that are endangered or already extinct as a result of the introductions must now be well over a hundred.
TABLE 1
Predicted Years of Extinction of Partula from Moorea
	PREDICTED YEAR
TAXON	OF EXTINCTION
Partula taeniata striolata	already extinct
Partula taeniata spadicea	1984
Partula taeniata elongata	1986
Partula taeniata simulans	1986
Partula taeniata nucleola	1986-1987
Partula exigua	uncertain (see text)
Partula aurantia	already extinct
Partula tohiveana olympia	1983
(P. olympia Crampton)	
Partula tohiveana tohiveana	1984
Partula mirabilis	1985
Partula mooreana	1985
Partula suturalis dendroica	1984
(P. dendroica Crampton)	
Partula suturalis strigosa	1986
Partula suturalis vexillum	1986-1987
REMEDIAL MEASURES
We see no hope of stopping the spread of Euglandina on the islands where it has already become established. There is, however, a very good case for discouraging new introductions, whether by governmental programs, by private enterprise on the part of local farmers, or by accident.
In an attempt to preserve the Moorean taxa of Partula we have established breeding colonies in our laboratories. Other colonies have been set up by the Jersey Wildlife Preservation Trust and by the Zoological Society of London. However, even if these colonies can successfully be maintained, a valuable genetic resource will have been lost.
We believe that our observations on the effects of Euglandina may be worth the attention of the workers and governmental agencies concerned in attempts to control Achatina and other pest species. They show that programs of "biological control" may have very dangerous consequences to local faunas.
ACKNOWLEDGMENTS
We are very grateful to Bernard Salvat, Rene Galzin, and Gilbert Vergonzanne for
104
PACIFIC SCIENCE, Volume 38, April 1984
their hospitality, and much practical help, at the Centre de FEnvironnement, Moorea; to Jean-Pierre Pointier, Charles Blanc, and Yoshio Kondo for allowing us to mention their unpublished work; and to the Science and Engineering Research Council for financial support. We also thank the Fauna and Flora Preservation Society, and the Onaway Trust, for financial help in setting up breeding colonies of Partula.
LITERATURE CITED
Carson, H. L., and J. S. Yoon. 1982. Genetics and evolution of Hawaiian Drosophila. Pages 207-344 in M. Ashburner, H. L. Carson, and J. N. Thompson, eds. The genetics and biology of Drosophila, Vol. 3b. Academic Press, London.
Clarke, B.,and J. Murray. 1969. Ecological genetics and speciation in land snails of the genus Partula. Biol. J. Linn. Soc, 1:31-42.
Crampton, H. E. 1916. Studies on the variation, distribution, and evolution of the genus Partula. The species inhabiting Tahiti. Carnegie Inst. Wash. Publ. 228:1-311.
---------. 1925. Studies on the variation, distribution, and evolution of the genus Partula. The species of the Mariana Islands, Guam and Saipan. Carnegie Inst. Wash. Publ. 228A: 1-116.
---------. 1932. Studies on the variation, distribution, and evolution of the genus Partula. The species inhabiting Moorea. Carnegie Inst. Wash. Publ. 410:1-335.
Crampton, H. E., and C. M. Cooke. 1953. New species of Partula from southeastern Polynesia. Occ. Pap. Bernice P. Bishop Mus. 21:135-159.
Hadfield, M. G., and B. S. Mountain. 1980. A field study of a vanishing species, Achatinella mustelina (Gastropoda, Pulmonata), in the Waianae Mountains of Oahu. Pac. Sei. 34:345-358.
Jackson, E. D. 1976. Linear volcanic chains on the Pacific floor. Pages 319-336 in G. H. Sutton, M. H. Manghnani, and R. Moberly, eds. The geophysics of the Pacific Ocean basin and its margin. American Geophysical Union, Washington.
Johnson, M. S., B. Clarke, and J. Murray. 1977. Genetic variation and reproductive isolation in Partula. Evolution 31:116-126.
Kondo, Y. 1973. Samoana of the Society Islands (Pulmonata-Partulidae). Malacol. Rev. 6:19-33.
Lack, D. 1947. Darwin's finches. Harper Bros., New York. 204 pp.
Mead, A. R. 1961. The giant African snail: A problem in economic malacology. University of Chicago Press, Chicago.
---------.   1979.  Economic malacology with
particular reference to Achatina fulica. Pages 1-150 in V. Fretter and J. Peake, eds. The pulmonates vol. 2B. Academic Press, London.
Murray, J., and B. Clarke. 1980. The genus Partula on Moorea: Speciation in progress. Proc. R. Soc. Lond. B 211:83-117.
Murray, J., M. S. Johnson, and B. Clarke. 1982. Microhabitat differences among genetically similar species of Partula. Evolution 36:316-325.
Pointier, J. P., and C Blanc. 1982. Achatina fulica dans les iles de la Société. Rapport de Mission, Museum Naturelle d'Histoire Naturelle, Antenne de Tahiti, B. P. 562, Papeete, Tahiti.
van der Schalie, H. 1969. Man meddles with nature—Hawaiian style. Biologist 51:136— 146.