252 



CHAPTER III 



fasciatus Say {tigrinus Sext.; vide Hynes, 1954) may be due to bird-transport, 

 I am unable to judge. Segerstrale (1954, p. 68 a.f.) accepts this as the normal mode 

 of dispersal for G. lacustrts G. O. Sars. 



5. Animals without special properties of active or passive dispersal. The ant 

 Ponera coarctata Latr. (fig. 35). The sexuals, males as well as females, are winged in 

 this species but do not make ordinary nuptial flights, the mating as a rule taking 

 place on the ground near the nest, or even within the nest. At any rate the fertilized 

 queens never fly (Wheeler, 1900; Michener & Michener, 195 1, p. 141-142). The 

 dispersal of this ant is therefore functionally the same as for a constantly wingless 

 species. 



Ponera coarctata inhabits a curiously broken-up area. Besides the forma typica 

 of Europe and the Mediterranean region and the sbsp. pennsylvanica Buckl. of 

 eastern North America, there is one subspecies endemic to South Africa (Natal) 

 and one to Australia (Queensland) (Creighton, 1950, p. 47). The subfamily Po- 

 nerinae includes the most primitive of living ants (Wheeler, 1900) and must be 

 very old, geologically speaking, not as a group alone, but also as actual species. 

 This is confirmed by the fact that the extinct Ponera atavia Mayr, known by all 

 casts from the Baltic amber (about the age of more than 40 million years), comes 

 very close to coarctata (Wheeler, 191 5, p. 39-40). 



The present distribution of Ponera coarctata must be interpreted as the remnant 

 of an almost world-wide area in Tertiary time and the occurrences on both sides of 

 the Atlantic do not constitute evidence in favour of the theory of a direct land-con- 

 nection (cf. the distribution of the beetle family Cupedidae, fig. 60). 



The conclusions arising from the study of animals showing an Amphiatlantic, 

 or similar, distribution are: — 



Their present area seems understandable without accepting a late (Pleistocene) 

 land-connection between the two continents. The existence of such (in one of the 

 interglacial periods) would undoubtedly have resulted in a much richer faunal 

 exchange in both directions, and a less "asymmetrical" range of distribution in 

 the individual, actually existing cases. 



On the other hand, it is not correct to regard almost all Amphiatlantic species 

 as peripheral relicts of an earlier circumpolar distribution, as do Hulten (1937) and 

 Deevey (1949), nor to emphasize that "der Terminus 'amphiatlantische Arten' 

 in faunagenetischer Hinsicht gegenstandslos wird" (Reinig, 1937, p. 22). Dispersal 

 across the sea is no negligeable part of faunal and floral history. 



