The Origins of Attitudes Genetic Bases and Acquisition ... Test phase: â¢ Participants ... Evaluative Conditioning and the Spreading Attitude
(EMAEE): Geography, Networks and Innovation, Marinus Ruppert building, Utrecht University, De Uithof, Utrecht, The Netherlands, 19-21 May 2005 . 1. Introduction The origins of technological innovation in the economic system remain a significant conce
the origin of species fallen angel dawn of reckoning blood bound ... through the stone age and the origins of ... psychoanalytic theory the ogres banquet east
except for the genus Viverricula, which is represented by a single species, Viverricula rasse (HORSFIELD), which is also found throughout southern Asia and was probably introduced to the island with man. Palaeontology shows that this fauna is an anci
systems has been carried out in a polymer — poly(dimethylsiloxane), or PDMS — the properties of which are entirely distinct from those of silicon2,3
After the decline of the Mycenaean palace society around 1200 B.C., ancient Greece entered the so-called Dark Age, traditionally dated c. 1150-750 (henceforth, all dates in the study are B.C. unless otherwise stated). By the early tenth century, the
The Origins of the Bulgars . Dilian Gurov . 1. Introduction . ... Artificial skull deformation has been characteristic for various Iranian speaking nomadic peoples,
The Origins of the Bulgars . ... and Volga Bolgar possibly indicates the power of the Bulgars in the ... Language and writing The language spoken by the Bulgars is
Emballonura atrata, Mormopterus jugularis and the endemic Myzopoda have no known systematic relationship with extant African genera or species. A clue to the origins of the endemic Myzopodidae may be found in a fossil
THREATENED FROGS OF Madagascar ... secrete toxins in some way, but as the species has only ... while others eat both plant
Biogéographie de Madagascar, 1996:457-465
ORIGINS AND AFFINITIES OF THE ANT FAUNA OF MADAGASCAR Brian L. FISHER Department of Entomology [email protected]
University of California Davis, CA 95616, U.S.A. e-mail:
ABSTRACT.- Fifty-two ant genera have been recorded from the Malagasy region, of which 48 are estimated to be indigenous. Four of these genera are endemic to Madagascar and 1 to Mauritius. In Madagascar alone,41 out of 45 recorded genera are estimated to be indigenous. Currently, there are 318 names of described species-group taxa from Madagascar and 381 names for the Malagasy region. The ant fauna of Madagascar, however,is one of the least understoodof al1biogeographic regions: 2/3of the ant species may be undescribed. Associated with Madagascar's long isolation from other land masses, the levelof endemism is high at the species level, greaterthan 90%. The levelof diversity of ant genera on the island is comparable to that of other biogeographic regions.On the basis of generic and species level comparisons,the Malagasy fauna shows greater affinities to Africathan to India andthe Oriental region. Thestriking gaps in the taxonomic composition ofthe fauna of Madagascar are evaluatedin the context of island radiations.The lack of driver antsin Madagascar may have spurred the diversification of Cerapachyinae and may have permitted the persistence of other relic taxa such as the Amblyoponini.
KEY W0RDS.- Formicidae, Biogeography, Madagascar, Systematics, Africa, India RESUME.- Cinquante-deux genres de fourmis, dont 48 considérés comme indigènes, sont COMUS dans la région Malgache. Quatre d'entr'eux sont endémiques de Madagascar et un seul de l'île Maurice. Parmi les 45 genres malgaches41 sont'considérés comme indigènes. Actuellement 3 18 espèces et sous-espèces sont décrites de Madagascaret 381 de la région Malgache. La faune des fourmis est pourtant l'une des moins connues de toutes les régions biogéographiques: on estime que 2/3 des espèces n'ont pas été décrites. Lié au long isolement de Madagascar,le taux d'endémisme est très élevéà l'échelle spécifique et pourrait dépasser 90%. Le niveau de diversité générique est comparable à celui des autres régions biogéographiques. Par comparaison, la compositiongénériqueouspécifiquede la faune de fourmis malgaches paraît plus étroitement liéeà celle de l'Afrique qu'à celle de l'Inde ou dela région orientale. Les lacunes frappantes dans la composition taxinomique de la faune Malgache sont évaluées dans le contextedesradiationsinsulaires.L'absencedesdorylines(fourmisvoyageuses)estsupposéeavoir stimulé la diversification des Cerapachyinae et probablement permis la survie de lignées ancestrales de fourmis comme celle des Amblyoponini. MOTS-CLES.- Formicidae, Biogéographie, Madagascar, Systématique, Afrique, Inde
INTRODUCTION Madagascar with its large size and varied topography, offers a diverse array of habitats occupied by a highly endemic and species rich ant fauna. In this manuscript, 1 discuss the species richness and endemism on the island, and provide a summary of the generic affinities ofthe Malagasy ant fauna to the Afrotropical and Oriental In: W.R. LOURENçO (ad.) Editions de I'ORSTOM, Paris
B. L. FISHER
biogeographic regions. 1 propose possible origins for the Malagasy fauna and evaluate the striking gaps in the taxonomic composition of the fauna in the context of island radiations. 1 contrast this pattern with that found for India. These themes are fùrther elaborated in FISHER (in press a). '
SPECIESDIVERSITY AND ENDEMISM
In the following discussions and analyses, Madagascar includes the coastal islands (e.g., Nossi Bé, St. Marie),and the MalagasyRegion refers to Madagascarand neighboringislands of the IndianOcean:Mauritius,Reunion,Seychelles,Rodrigues, Aldabra, Farquhar, Chagos, andthe Comoros Islands.In 1893, DALLAT o m listed 1 19 Malagasy species (excluding subspecies), and WHEELER(1922) enumerated 237 species (excludingsubspecies) for the MalagasyRegion.There are currently a total of 288 species (38 1 includingsubspecies) forthe Malagasyregionand 243 species (3 18 including subspecific names) in Madagascar alone. The small increase innewspecies described since 1922 does not reflect a saturation of collecting of a well-known fauna, but the lack of new material.Recent collections by the author, G.D. ALPERT et al., D.M. OLSON,and P.S. WARD, suggest that there may be over 1000 species in the Malagasy region. Thelevel of endemismisextraordinary, with 90% ofthetaxa endemic to Madagascar alone, and 96% endemic to the Malagasy region. The level of endemism may even be greater in the eastern forests of Madagascar. A recent ant survey of a wet tropical closed forest site in the RNI d'hdringitra found an estimated 100% of the 134 ant species inthis mountainous regionto be endemicto Madagascar (FISHER,in press b). ikFFINITIES TO AFRTCA AND THE ORIENT
To investigate the origin and affinities ofthe ants of Madagascar, 1 compared the number of genera in common with neighboring land masses of India and Africa which were al1 once united in Gondwana (SMITHet al., 1994). To fùrther understand afKnities with the Africanantfauna, 1 havecompareh the genera of Madagascarwith geographically distinct subregions within Africa: West Africa(W.A.), East Africa @.A.), and southern Afiica (S.A.) (For a discussion, definition, and the list of genera of these regions and subregions,refer to FISHER, inpress a). Zn Table 1, 1list the 52 genera recorded fiom the Malagasy region. Forty-eight are estimated to be indigenous to the region, of which4 are endemic to Madagascar and 1 to Mauritius. For the island of Madagascar, 45 genera are recorded, of which 41 are considered to be indigenous. The classification of a genus as indigenous or introduced is based on a subjective evaluation of historical distributionrecords and the biology of the species. The richnessof 48 genera in Madagascar is comparable with other tropical regions. In a comparison of the log of the total numberofgenera present ineachof the biogeographic regions and subregions, and of the log of the area of each region, the Malagasy region is not depauperate. The slope of the log-log relationship has a z-value of 0.224 (R2 = 0.61) and the log number of genera inthe Malagasy region falls very close to this line.
ANT FAUNA OF MADAGASCAR
3 Genera Shared,
O Unique Fig. 1. Three genera are shared between the Malagasy region and Peninsularhdia (included Sri Lanka, but excludes Bhutan, Nepal, Pakistan, Bangladesh, and Assam) and are not found in the Afrotropical region. These genera are not uniquely shared since they are found in other biogeographic regions. Three genera, Aphaenogaster, Kyidris, and Vollenhovia, are shared between the Malagasy region and India and are not found in Afi-ica. These three genera are found in other biogeographic regions and thus are not uniquely shared between the Malagasy region and India (Fig.1). Nine genera are sharedbetween the Malagasyregionand Afi-ica and are not recorded fiom India: Anopolepis Discothyrea Glamyromyrmex Melissotarsus Mystrium Prionopelta Pristomyrmex Simopone Terataner Melissotarsus, and Terataner are the only genera uniquelysharedbetween the Malagasy region and Africa (Fig. 2). Simopone, Mystrium, Prionopelta, Discothyrea, and Pristomyrmex are also found in the Oriental biogeographic region and it is possible that these genera may have gone locally extinct in India.
B. L. FISHER
The number of uniquely shared genera is very low, but itdoes suggest that the ant fauna of Madagascar has closer ailinitiesto Afi-ica than to India andthe Orient (2 genera uniquelyshared with Afi-ica, O withIndiaand the Orientalregions). Two additional analyses support the close relationship between Mrica and Madagascar. Results fi-om a cluster analysis (single linkage) and an analysis of faunalsimilarityusing the Jaccard Index (MAG~JRRAN, 1988) of the genera present in each biogeographic region, indicate that the Malagasy and African Regions are most similar (see FISHER,in press a). The Jaccard Index also indicatesthat within Mica, the southern and East Afi-ica regions are the most similarto Madagascar. The close relationship of M i c a and Madagascar is also supported at the level of species-groups and species. For example, of the 36 described species of Tetramorium in the Malagasy region (BOLTON, 1979), 29 are endemic, 4 are shared with Afi-ica, and 3 are pantropical tramp species. Bolton organized the taxa into 8 species groups of which 3 are endemic to Madagascar, 2 are sharedwith Afi-icaand 3 are widespread. In addition, 3 indigenous species fiom the Malagasy region are shared with Afiica, while none are shared with the Oriental or Indo-Australian regions. Af€inities at the speciesgroup and species level with the Oriental and Indo-Australian regionsmay exist in other pantropical genera, suchas Strumigenys and Pheidole which have numerous undescribed species in collections awaiting revision. Nineteen genera are found in Africa and India or the Orient but not in Madagascar: Aenictus Calyptomyrmex Carebara Cataglyphis Centromyrmex Cryptopone Dorylus Epitritus Leptanilla Messor Myrmicaria Oecophylla Paedblgus Pheidologeton Polyrhachis Probolomyrmex Rhoptromyrmex Psezcdolasius Sphinctomyrmex Ifwe assume that the ant genera found on Madagascararrivedprimarily by dispersal (see discussion of the origin of genera below), the nesting behavior of the nineteen genera listed above may help explain which ant genera successllly colonized the island. These taxa nest ineither the ground, in termite mounds, inthe leaf litter or iri rotten Wood, or self constructed nests. Polyrhachis is the only genus that has some species that nest in plant cavities. BROWN (1973) suggestedthat those ants that nest in preformed plant cavities are best suited for surviving long distance dispersal by rafting across an ocean. Ants could weather such a journey formany months by sealing off the entrance of the nest and feedingon their own larvae. A similar method of dispersal could also have been used by a hibernating ancestor of the present day lemursto rafi across the MozambiqueChannel. The arriva1 ofthe driverants (Aenictm and Dorylus) and Leptanilla to Madagascar may also havebeenhinderedby the fact that queens are wingless, and coloniesreproduce by budding. The driver ants which dominatethe forest floor, and the weaver ants (Oecophylla) which dominantthe forest canopy in Afiotropicaland Oriental tropical regions are absent from the Malagasyregion.Weaver ants andespeciallydriver ants are important predators of other ants and have been shown to influence ant population structure and the diversity of antcommunities (HOLLDOBLER & WILSON, 1990; GOTWALD, 1995).
ANT FAUNA OF MADAGASCAR
Because of their absence from Madagascar? the population dynamics of Malagasy ant communities may differ greatly from other Old World ant communities;A comparison of Afi-ican and Malagasy ant communities provides a natural experiment for evaluating the effect of these dominant African ants. .’
9 Genera Shared 2 Unique
Fig. 2. Nine genera are shared between the Malagasyregion and Africa andare not recorded from Peninsular India. Of these,2 genera are uniquely found in the Malagasy and Afrotropical regions.
The lack of driver ants in Madagascar may have spurred the diversification ofthe tribe Cerapachyinae (Cerapachys and Sirnopone) on the island(HOLLDOBLER & WILSON, 1990; OLSON& WARD,in press). The morphologically diverse and species rich Cerapachys on the island include species that are morphologically convergent to the driver ant genusAenictus found in Afi-ica and exhibit armyant habits of raidingthe nests of other ants (BROWN,1975; pers. obs.). In the absence of Dorylinae, Cerapachys may have been fiee to develop and capitalize on their army ant-like predatory behavior, and thus represent a monophyletic island radiation of species. On the other hand, this clade of Cerapachys may have been found once in other biogeographic regions, but now only persists in Madagascarin the absence of driver ants. Mystrzum, in the tribe Amblyoponiniwhichpossessesmanymorphologically primitive characters, represents another genus that may have been able to persist in the absence of competition fiom more recentlyevolvedgenera that failed to colonize Madagascar.Currently there are 6 describedspecies of Mystrium from Madagascar,
B. L. FISHER
whileonly one speciesfromWest M c a andonespeciesfrom Australian, and Australian regionsare known.
the Oriental, Indo-
ORIGINS OF THE ANT FAUNA
The origin of theant faunaof Madagascar mustbe inferred fromOur understanding of geological history andOur limited fossilrecords of ants. The high numberof extant ant genera found in Baltic amber fossils (Oligocene) suggests that many extant genera are at least 35 million years old (H~LLDOBLER& WILSON,1990). The almost complete lackof extant genera found in Cretaceous and Eocene deposits implies that most modern taxa may not have been abundant or even extant before the Oligocene. If the fossil data accurately reflects the history of ants, then al1 modern genera most likely evolved after the complete separation of Madagascar fiom M c a (around 120 Ma; R A E ~ O W I T Zet al., 1983) and aRerthe separation or India fiom Madagascar (approx. 90 Ma;SToREY et al., 1995). An alternative hypothesis, though not supported by the fossil record, is that some extant genera are much older and may date back to around the mid-Cretaceous at the time of the break up of Gondwana. BAROMU ~ A NetI al. (1992) offer some indirect evidence to support this. They hypothesized that the Old and New World army ants are each others closest relative, and thus, form a monophyletic group. Since al1 modern taxa have wingless queens and reproduce by budding, it is most parsimonious to assume that the ancestor of this cladealsohadwinglessqueens. The presentdistribution of Ecitoninae in the New World and Dorylinae in the Old World, may be explained by a common ancestor that was present before the complete breakup of South America and Afi-ica which is estimated to have occurred between 84 Ma and 106 Ma (PI"N et al., 1993). Since Dorylinae are absent from Madagascar, this common ancestor may have evolved after the complete separation of Madagascar fi-om &ca more than 120 Ma (RAEIINOWITZ et aZ., 1983).Other genera whichhaveapresentdayGondwanaland distribution, such as in the tribe Amblyoponini, may also be much older than predicted fiom the fossil record.
In summary, Madagascar can be describedas an isolated landmass that has a highly endemicfaunacharacterized by closeaffinities to the Afrotropical region with many groups that have apparently undergone unique island radiations, and relict taxa that have numerous plesiomorphic characters. Thesepatterns can be explained by the proximity of Madagascar to Africaand its longhistory of isolation.PeninsularIndia, the sister continent of Madagascar, does not follow these predictions. Along with Madagascar, it began separating fiom Africa 150 Ma and remained joined with Madagascar until around 90 Ma before splitting off, moving northward, reaching its present position approximately 50 Ma (RABINOWITZ et al., 1983;.STOREY et al., 1995). Therefore India wasisolated with Madagascar for 60 millionyearsandcompletelyisolated for an additional 40 million years. India is over five timesas large as Madagascar, but contains only 3 endemic ant genera, Aneuretus and Stereomyrmex in Sri Lanka, andIndomyrma in the Western Ghats in India, two of the most mesic localities inthe region.
ANT FAUNA OF MADAGASCAR 463
What happened to the endemic fauna that would be predicted from India’s long isolation? Did the endemic taxa go extinct after initial contact with Asia due to the invasion of superior competitors? Did India experience dramatic climatic changes that drove taxa to extinction, leaving the two endemic taxa in the mesic localities? Or was India in contact with other landmasses via land bridgesor filters duringits sojourn in the Indian Ocean, andtherefore not an isolated land massfor 100 million years? BRIGGS(1989) suggested that during the Cretaceous, India moved northward and became in direct or close contact with Northern Afiica. Faunal interchangethat occurred during this contact with Afiica, or during later contact with Asia could have allowed superior competitors to colonizeIndia,drivingendemic taxa to extinction.The mechanism for the absence of endemic ant taxa is not known, but the pattern has been noticed for vertebrate groups (I~RIGGs,1987; S m , 1984).
Our understanding of the origins and affinities of the ant faunaof Madagascar is far from complete. We understand even lessabout the biogeographic patterns of ants within the island. Almost everyone of the 48 indigenousgeneraisinneed of revision. Nonetheless, evidence does suggest that the Malagasy ant fauna is more closely related to the Afrotropical fauna than to the Oriental or Indo-Australian faunas. The absence of ecologically dominant genera found in Afi-ica, such as weaver and driver ants, may have created unique opportunities for the ants of Madagascar, allowing some to persist and others to radiate. The absence of a high numberof endemic genera in India, Madagascar sister landmassfor 60 million years, poses interesting biological and geological questions.
ACKNOWLEDGMENTS 1 thank P.S. Ward for comments onthe manuscriptand for hisconstant encouragementandtrainingin al1 aspects of antsystematics. Studies in Madagascar could not have been completed without the willing support of the Malagasy people and the fùnding by World Wide Fund for Nature-Madagascar, National Geographic Society, and the National Science Foundation.
BOLTON, B., 1979. The ant tribe Tetramoriini (Hymenoptera: Formicidae). The genus Teframorium Mayr in the Malagasy region and in the New World. Bull. Br. Mus. (Nat. Hist.) Entomol., 38:129-181. BRIGGS, J.C., 1987. Biogeography and plate tectonics. Elsevier, Amsterdam, x + 204p.
BRIGGS, J.C., 1989. The historic biogeography of India: isolation or contact? Syst. Zool., 38:322-332. BROWN, W. L., Jr., 1973. A comparisonof the Hylean and Congo-West Africanrain forest ant faunas. In: B.J., Meggers, E.S. Ayensu& W.D. Duckworth (eds.) Tropical forest ecosystems in Africa and South America: a comparative review. pp. 161-185. Smithsonian Inst. Press, Washington, D.C., viii + 350p. DALLA TORRE, K. W. VON, 1893. Catalogus Hymenopterorum hucusque descriptonun systematicus et synonymicus. Vol. 7. Formicidae (Heterogyna). W. Engelmann, Leipzig, 289p. FISHER, B.L.(in press a). Biogeography and ecology ofthe ant fauna of Madagascar. J. Nat.Est. FISHER, B.L. (in press b). Ant diversity patterns alongan elevational gradientin the Réserve Naturelle Intégrale d'Andringitra. Fieldliana. GOTWALD,W. H., Jr., 1995. Army ants: the biology of social predation. Corne11 University Press, Ithaca, xviii + 302p. HOLLDOBLER, B., & E.O. WILSON, 1990. The ants. Cambridge, Harvard Univ. Press, Mass., xii + 732p. MAGURRAN, A. E., 1988.Ecologicaldiversity Princeton, New Jersey, 179p.
and its measurement.PrincetonUniversityPress,
OLSON, D.M., & P.S. WARD (in press). The ant fauna (Hymenoptera: Formicidae)of Kirindy Forest (tropical dry forest) in western Madagascar. Primate Report. (Spec. Vol.). PITMAN, W.C. III., S. CANDE, L. LABRECQUE, & J. PINDELL, 1993. Fragmentation of Gondwana: the separation of Africa from South America, In: P. Goldblatt (ed.) Biological relationships between Africa and South America. pp 15-34.: Yale University Press, New Haven,p.630
RABINOWITZ, P.D., M.F. COFFIN, & D.FALVEY, 1983. The separation of Madagascar and &ca. Science, 220:67-69.
SAHNI, A., 1984. Cretceous-Paleocene terrestrial faunas of India: lack of endemism during drifting of Indian Plate. Science, 226:441-443. SMITH, A.G., D.G. SMITH, & B.M. FUNNELL, 1994. Atlas of Mesozoic Cambridge University, Cambridge,ix + 99 p.
and Cenozoic coastlands.
STOREY, M., J. J. U O N E Y , A.D. SAUNDERS, R.A. DUNCAN, S.P. KELLEY & M.F. COFFIN, 1995. Timing of hotspot-relatedvolcanismandthebreakup of Madagascar and India: Science, 267352455. WHEELER, W.M.,1922.Antsof the American Museum Congo expedition. A contribution to the myrmecology of Africa. Lx. A synonymic list of the ants of the Malagasy region. Bull. Am. Mus. Nat. Hist., 45:1005-1055.
MADAGASCAR ANT OF FAUNA
Table 1 . Ant Genera found in the Malagasy region. Endemic genera are indicated in bold, and genera known only fiom probable introduced species are noted with an *. Genera that are unique to Mauritius, Seychelles, and Reunionare also indicated.