May, 1909.] 97 



lived till 1898 before completing their metamorphosis. I have 

 recorded giving the nest some nufer pupae on A-Uguyt 6th, but some 

 must have been given the nest before August 2nd. Nevertheless, no 

 umhratus workers were observed to hatch during 1898. 



On August 13th I introduced some ants from nest No. 1 to a 

 fertile niger ? . After a time they got over their alarm, and were 

 fairly friendly to her. Next day, after leaving them together all night, 

 I put workers and ? into the nest at 12.30. The ? was at once 

 attacked, though some ants seemed to salute her. There was very 

 little excitement in the nest. At 4 p.m. she was dead. 



By June 18th, 1899, there were several full-grown larvae, which 

 became pupae on the 24th. 



Up to August 10th I had not observed any to hatch, but on that 

 date, after changinof the ants to a new nest, 1 discovered some frae- 

 ments of a yellow ant under some earth. Soon after I discovered 

 a yellow worker, undoubtedly umhratus, nearly dead. The glass of the 

 new nest being much cleaner, I could observe the ants better, and up 

 to September 20th I saw 329 umhratus workers hatch in the nest. 

 Some of them were killed by the nigers immediately on emerging, but 

 a few lived for a day or two and were then killed. In some cases the 

 cocoons were taken off pupae, and the pupae then thrown out. The 

 dead bodies were often used as food for the larvae, a thing ants never 

 do with their own dead. I should perhaps mention that I rescued a 

 number of these pupae with a few workers, and put them with a fertile 

 umhratus $ , and formed a small colony. Three niger workers got 

 enclosed with the others, and were shut in the new nest. They made 

 themselves at home, however, tended the pupae, fed the queen and the 

 workers, and were fed by the latter. One of these nigers supplied the 

 whole nest with food during the winter. 



During the winter 1899 — 1900, nest No. 1 was kept in a warm 

 room, and probably in consequence of this, the first larva changed to 

 a pupa as early as May 1st, 1900. The pupae began to hatch in the 

 beginning of July, and on the 6th there were six alive and well, moving 

 pupae about, &c., though some others were killed. 



On the 18th 1here were twenty alive and assisting in the work of 

 the nest. It was strange to see the black and yellow ants together in 

 the nest. 



I was then, unfortunately, obliged to be away from home for over 

 two months, and had to leave the nest in the care of people who did not 

 understand them. When I returned at the end of September, the only 

 ant alive was the queen, and she died soon. The nest and the box 



I 



