AN AMERICAN CERAPACHYS. 189 



dition of the female Dorylinae is not of so much value in view of 

 the fact that some true Ponerinae (Leptogenys, c. g.} have wing- 

 less ergatoid queens. The characters drawn from the male 

 genitalia are not of themselves sufficiently important to deter- 

 mine the allocation of the Cerapachyinae in a particular subfamily. 



My own observations on the living Ccrapachys aiigustce, though 

 very fragmentary, incline me to accept Forel's views and to regard 

 the Cerapachyinae as true Ponerinae. There was certainly 

 nothing in the habits of the insects to remind me of the Dory- 

 line ants notwithstanding their striking morphological resem- 

 blance, especially in the shape of the head, and the blindness of 

 the workers, to certain species of Eciton (E. schmitti Emery, E. 

 sttmic/irasti Norton, E. wheclcri Emery). The small colony ; 

 with its queen so like the workers in size and structure, and the 

 slow movements of the insects, all very closely resemble the con- 

 ditions found in many lowly organized Ponerinae, e. g., Stigina- 

 tomiua, and I may add also, Proccratiinn and Sy spin net a to judge 

 from the account (/// littcris) of my friend, Rev. P. J. Schmitt, 

 O.S.B., who has been so fortunate as to observe living colonies 

 of these rare ants. Moreover, the fact that specimens of Cera- 

 pachyinae are rare in collections would seem to show that the 

 nests of these insects are not at all populous. It is safe to say 

 that if C. angnstce were like the timid Ecitons in its habits, it would 

 have been impossible to find a remnant of the colony on the day 

 following the ransacking of the nest. Morphologically the 

 Ponerinae are certainly a very heterogeneous assemblage of forms, 

 but this is merely what we should expect to find in so ancient 

 and extensive a group. The addition of the tribes of the Cera- 

 pachyinae to this subfamily would increase but little the already 

 existing heterogeneity, since these ants are closely related to 

 forms like Proccratiinn and Sysphincta which Emery formerly re- 

 garded as Dorylinse but has since placed with the Ponerinae. 

 When the Cerapachyinae are included with the Ponerinae, the 

 Dorylinae become a more homogeneous subfamily, while the 

 Ponerinae represent the diversified and often peculiarly specialized 

 survivors of the ancient stock from which all the other subfamilies 

 of ants have been descended. 



It is an interesting and probably significant fact that all the 



