260 WHEELER. 



panded and separating the dark dorsal and ventral sclerites from one 

 another. L. unicolor evidently has not developed this honey-storing 

 habit. In all the other species the individuals which become repletes 

 must begin their development in this direction as soon as they hatch 

 from the pup?e and before their integument has hardened. In several 

 nests I found a certain number of callows that had already Ijecome 

 perfect repletes. The repletes of Lcptomyrmcx are not helpless like 

 those of the American Myrmccocystus mexicanus and melliger and their 

 varieties, but run about easily though not so rapidly as the ordinary 

 workers. They are, however, much more interested in the brood tlian 

 the ordinary workers. As soon as the nest is disturl)ed every replete 

 grabs a larva or pupa and makes off with it. If possible they all retire 

 to the ceiling of the innermost nest chamber and there huddle side by 

 side, each quietly holding its larva or pupa in its jaws. Though the 

 ordinary workers occasionally carry away some of the brood, they are 

 conspicuously less interested in this occupation than their replete 

 sisters. Turner states that these individuals do not leave the nest, 

 but on two occasions at Kuranda, Queensland, I found repletes of 

 L. varians var. rvficcps returning to their nests along a road through 

 the scrub. 



Numerous larvse and in most cases pupse were found in the nests 

 during October and November, but all the pupse collected proved to 

 be workers and I saw no males in any of the nests. Moreover, the 

 most careful scrutiny failed to reveal the presence of anything like a 

 queen or female in any of the nests and on several occasions I must 

 have captured the entire colony. Nor was it possible to detect among 

 the workers any individual or indi^•iduals that showed signs of func- 

 tioning as queens of the colony. If winged queens existed they would 

 certainly have been taken by this time, for these ants are rather abun- 

 dant in many localities in New South Wales and Queensland. I be- 

 lieve, in fact, that no true queens occur in the genus Lcptomyrmcx, but 

 that there are in each colony one or more fertile workers which supply 

 the eggs that develop into workers and males. We have here one of 

 the interesting problems to be solved by the resident Australian ento- 

 mologist. In this connection I may state that some other Australian 

 ant-genera are peculiar in lacking queens, notably Rhitidoponera and 

 Diacamma. In still another series of Ponerine genera {Onychomyrmex, 

 Paranomopone and Leptogcnys) the queens are wingless and extremely 

 worker-like. 



The Lcptomyrmcx larva is very peculiar. That of Z. tiigrivcntris 

 is shown in Fig. 4a. The head (b) is violin-shaped, with very small. 



