244 



ANTS. 



to be sought in tlir gnmp of wasp-like forms represented by the Sco- 

 liiche. Tlunnidae and .Mutillidae. He leaves the Scoliidae and Thynnidae 

 out of consideration >n account of the un-antlike structure of their 

 male genitalia, and regards the Mutillidae, or velvet ants, as having 

 man}- points <>f resemblance to the Ponerinae and especially to the Cera- 

 pachysii. Existing Mutillids, however, present two highly specialized 

 characters which would seem to eliminate them also from the hypo- 

 thetical ancestral series : they are, so far as known, parasitic and their 

 females are wingless. Emery does not attach much weight to these 

 considerations, for it is not from the modern Mutillidae that he would 

 derive the ants, but from their more ancient, entomophagous and non- 

 parasitic precursors. He believes, 

 however, that the ancestral female 

 ants were wingless and have re-ac- 

 quired wings by inheritance from the 

 males. For reasons to be given in the 

 next chapter, in connection with the 

 wingless females of the Dorylinae, I 

 am unable to accept this view and am. 

 therefore, inclined to regard the 

 apterous condition of the female 

 Mutillids as a very serious obstacle 

 to Emery's hypothesis. Hancllirsch 

 (1908) has shown that in seeking the 

 ancestors of ants it is necessary to 

 turn to forms in which both sexes are 

 winged. In this respect, and also in 

 being non-parasitic and in hunting 

 their prey under ground, the Scoliidae 

 seem to present a more faithful pic- 

 ture of the ancestral ants than the existing Mutillidae. 



That the phylogenetic relations of the Ponerinae to the four other 

 formicine subfamilies are capable of rather satisfactory formulation is 

 evident from Forel's statement (10,03^) that if, as can hardly be doubted, 

 the Ponerinae represent the living remnants of the primitive formicine 

 stock, a stock in turn derived from the Scoliids, " the four other sub- 

 families must all be considered as specialized and more or less parallel 

 derivatives of the Ponerinae, i. e., as all arising from this common stock, 

 but without connection among themselves. 



'The Dorylinae, arising' directly from the Cerapachyii (a tribe of 

 Ponerinae), have no direct points of relationship with the three other 



FIG. 140. Worker of Aneure- 

 1ns sinwtii of Ceylon. (Emery.) a, 

 From above : b, from the side. 



