220 WHEELER. 



new subgeneric name for the Australian and Papuan forms with 

 11 -jointed antennse. 



In this connection attention may be called to the fact that the 

 females of Nothosphinctus bear a surprising resemblance to those of 

 the related tropical genus Acauthosticlius, as will be seen by consulting 

 Emery's description and figures of the female of A. quadratus (Zool. 

 Jahrb. Abt. Syst. 8, 1895, p. 693, PI. 14, Fig. 4 and Gen. Insect. Fasc. 

 118, 1911, PI. 1, Figs. 4 and 4b). As Emery and Forel have remarked, 

 these females resemble the only known females of the Dorylinse 

 {Dorylua, Ecifou, ^Icnictus, Leptanilla) and may therefore have con- 

 siderable phylogenetic significance. In previous publications I have 

 described similar females in several Ponerine genera {Onychomyrmex, 

 Paranomopone, Leptogcnys.^ 



Turning to the other genera of the Cerapachyini we find that our 

 knowledge of the females is even more incomplete than in Sphincto- 

 myrmex and Eiisphinctus. The female Cerapachys imcrincnsis Forel 

 of Madagascar has well-developed wings and Phyracaces pubescens 

 Emery of Borneo and Ph. turncri Forel of Queensland were described 

 from dealated females, and I have winged females of an undescribed 

 Phyracaces from the Congo, but the females of Parasyscia augustce 

 Wheeler and Phyracaces elegans sp. nov. show no traces of having 

 borne wings, though the thorax is of the same structure as in the 

 winged females. The same is true of the female of an undescribed 

 species of Syscia recently taken in Fiji by Dr. W. M. Mann. In the 

 Indian Lioponera Jongitarsis the female is ergatoid. These genera 

 therefore exhibit various stages in the reduction of the normal winged 

 female to the ergatoid type of Eusphinctus s. str. while Nothosphinctus 



3 Consistency with the views here advanced would require that Ctenopyga 

 Ashmead should be regarded as a distinct genus and not as a subgenus of 

 Acanthostichus, since the females of the two known species, C. texanus Forel 

 and C. townsendi Ashmead, are winged and quite different from that of A. 

 quadratus. A too consistent following of the example of Andre and myself 

 would, however, lead to difficulties in such cases as the European Harpago- 

 .venus sublcevis, an ant which in Sweden has only apterous, ergatoid females, 

 but in Saxony has winged females of normal structure. I have perhaps over- 

 (>mphasized the differences between the paleotropical Eusphinctus and the 

 neotropical Sphinctomyrmex, but my procedure may at least deter zoogeog- 

 raphers from citing Sphinctomyrmex as evidence of a former antarctic land 

 connection between South America and Australia. It is often just such 

 im])erfectly known genera, which are confidently cited in support of ancient 

 land-bridges; e. g. the genera M yrmecocijst us and Melophor us. When care- 

 fully studied the Okl World forms referred to Myrmecocy.stus are seen to be 

 gcnerically distinct and are now referred to Cataglyphis, while the South 

 American forms referred to Melophor us really belong to a distinct genus, 

 Lrisiophanes. 



