644 MR. A. D. MICHAEL ON THE [DeC. 1 , 



to become empty once more. I then placed in the cells some dead 

 Ants which I had soaked in carmine stain ; the next morning the 

 ahmentary canals of the Gamasids were all bright red, while the rest 

 of their bodies was uncoloured. I repeated this several times with 

 the like result, and on one occasion when a very clear Gamasid, 

 which had lately changed from the nymph, had been supplied with 

 a stained Ant and the cell then removed to the stage of the microscope, 

 I saw the Gamasid mount on the body, plunge its trophi into it, 

 and then I could plainly see the small streams of carmine liquid 

 passing down the canal as the Gamasid sucked, and I afterwards 

 dissected out the alimentary canals of some of the Gamasids and 

 found them filled with red matter ; the ordinary contents of course 

 are not of that coloui". These Gamasids would undoubtedly feed on 

 the dead body of any small freshly-killed insect which might be 

 found in the nest. The Gamasid hereinafter referred to as Lcelaps 

 vacua I also found would feed and thrive on the dead Ants, &c. ; 

 but Lcelaps acuta I could not get to feed in a similar manner, and 

 it did not live long in the cells. The above facts made it seem 

 probable to me that the Gamasince were present either as scavengers, or 

 else for the purpose of sharing the feast in the case of small insects 

 killed by the Ants ; possibly the friendly conduct of the Ants points 

 rather to the former than the latter conclusion. 



In the nests of the same Ant I found three smaller species of 

 Lcelaps, none of which I could find elsewhere, and which, as far as I 

 know, are unrecorded ; I propose calling them L. Jlexuosa, L. vacua, 

 and L. acuta. The first-named is in one respect a singular creature, 

 viz., as regards the mandible of the male. The mandibles of the 

 Gamasince and Vropodince are usually chelas ; very retractile, and 

 capable of being wholly withdrawn within the body. The two arms of 

 the chela are often different, particularly in the male, one arm, oftenest 

 the movable, having frequently some appendage or other com- 

 phcation, often of very strange form, but both arms are almost 

 always directed forward. In the present species the fixed arm is 

 most minute, a mere spike, while the movable arm is very long, 

 horn-hke, and doubly curved and undulated, both perpendicularly 

 and laterally ; so that the two mandibles cross and cannot be with- 

 drawn into the body (fig. 6 a). 



In the nests of the same Ant, Camponotus herculeanus, I also 

 found a handsome bright-crimson TJropoda belonging to the section 

 with sculptured backs ; it was present in large numbers in one nest, 

 and in small numbers in one or two others, and was found on the 

 walls of the passages and chambers, and also, most abundantly, on 

 the outside of the cocoons of such pupse of the Ant as were en- 

 veloped in a cocoon ; there were often three or four Uropodce on a 

 single cocoon. I could not ascertain that the cocoons were in any 

 way injured, but the iiroyoda appeared to get a thread or two of 

 the cocoon loose, and this it held on to firmly, as well as I could 

 ascertain, by holding it with the flattened femora of the first pair of 

 legs. \ never found any of the Uropodcn either upon the adult Ants 

 or their larvae or upon such pupse as were not enclosed in a cocoon. 



