3 18 Bulletin American Museum of Natural History. [Vol. XXI, 



observations : First, several large nests which I located during Septem- 

 ber, 1904, were found to have been deserted when I revisited them 

 during August, 1905. Second, in a particular locality the number 

 of abandoned is much greater than the number of inhabited nests. 

 Third, on September 16, 1905, 1 actually saw a large colony of maricB 

 in the act of excavating a new nest in a bunch of grass. Such changes 

 of domicile can be readily effected on account of the simple architect- 

 ure of the nest and the ease and rapidity with which the sand is ex- 

 cavated. Both maricB and gagates resemble the species of Eciton not 

 only in their habit of moving everywhere in files and their probably 

 not infrequent changes of domicile, but also in the singular habit 

 when in their nests of hanging to one another by means of their claws 

 till they form bunches sometimes nearly as large as one's fist. 



These ants resent any disturbance of their nests with all the power 

 of their mandibles and anal glands. The secretion from the latter 

 seems to be very volatile and does not have the rancid butter or 

 "Tapinoma odor" of many Dolichoderinse, like the species of Tapi- 

 noma, Forelius, Iridom-yrmex, Dorym,yrm.ex, Liom-etopum, and some of 

 the tropical species of Dolichoderus, but a peculiar smoky or pungent 

 odor, fainter in m,ari(B and stronger and of a somewhat different 

 character in gagates. 



It is difficult to keep these ants in artificial nests of the Fielde or 

 Janet patterns, as they seem to be very restless and so indifferent to 

 the sunlight that the chambers cannot be readily opened or cleaned. 

 The original Lubbock nest, with its contrivance for permitting the 

 ants to seek their food on an open platform, would probably be better 

 adapted to both of the species. 



The sexual phases of maricB and gagates make their appearance 

 at different times. I infer this from the fact that on August 19 and 

 20, 1905, I found the nests of mari(S containing male and female 

 pupae, many mature males, and quite a number of callow females. 

 Single dealated and winged females were also seen running over the 

 sand, so that August 20 is approximately the date of the nuptial 

 flight of this species. On September 16 and 17 I again opened several 

 nests but in only one did I find sexual forms. These were all mature, 

 apparently belated females, and there were only worker pup^. The 

 nests of gagates, however, were searched in vain on all of these dates 

 for males and females and their pupas. It is certain, therefore, that 

 the colonies of this variety throw off their winged phases earlier in the 

 summer, probably during July and early August. In my collection 

 there is a male Dolichoderus taken June 29, 1905, at Lakehurst, New 



