AUSTRALIAN CERAPACHYINI. 219 



me more acceptable than Forel's supposition that we are here con- 

 cerned with a peculiar dimorphism of the worker caste comparable 

 with that observed in Poncra cdunrdi? The colonies referred to 

 Eu,s-phiucfus s. str. are polygynic whereas those referred to Sphwcto- 

 vii/nnc.v s. str. by Emery and Forel are monogynic, a difference natur- 

 ally correlated with the size and development of the gaster in the 

 females of the two groups. Since the reproductive function is dis- 

 tributed over several queens in Eusphindus, each is more nearly of 

 the normal stature of the largest workers than in the Hphindomyrmcx 

 colony. 



I therefore agree with Andre in leaving only S. stall of Brazil in the 

 genus Sphindomyrmex and in regarding all the paleotropical species 

 as belonging to the genus Eusphindus, but believe that the latter 

 should be divided into at least three subgenera, as follows: 



1. Eusphindus s. str. ^Yorkers and females with 11-jointed 

 antennae, with entire or emarginate pygidium, the workers blind, the 

 females with eyes and ocelli. Habits hypogteic. 



2. Nothosphindus subgen. nov. Workers and "females with 12- 

 jointed antennte; the former blind and with entire pygidium, the 

 latter with emarginate pygidium and either blind or with very minute 

 eyes and the anterior ocellus. Habits hypogteic. 



3. Zasphindus subgen. nov. Workers large, dark colored, with 

 12-jointed antennfe and well developed eyes, but without ocelli. 

 Females unknown. Habits probably epiggeic. 



The worker of E. cribratus Emery of New Guinea has an entire 

 pygidium and belongs with the Australian species in Eusphindus 

 s. str. but the workers of the Indian species E. furcatus Emery and 

 taylori Forel have a notched pygidium. Should future investigation 

 show that the pygidial characters are correlated with other peculiari- 

 ties or with different types of female, it may be advisable to restrict 

 the subgenus Eusphindus to the two Indian species and to suggest a 



2 Cases of dimorphism of the worker among the Ponerinae are extremely 

 rare. The most remarkable case is that of the African Megaponera fcetens 

 Fabr., which has two types of workers differing not only in size, but also in 

 sculpture and the structure of the antennae. As no winged females were 

 known of this species Emery believed that their place in the colonies was 

 usurped by the large workers but Arnold has recently discovered and described 

 the true female (Ann. South Afr. Mus. 14, 1915, p. 48 nota, Fig. 6). It is 

 ergatoid and therefore resembles the females of E. nianni and imbecillis. 

 Among many specimens of M. foetens recently collected in the Congo by 

 Mr. H. O. Lang of the American Museum of Natural History I find both 

 forms of worker in each colony. 



