692 Bulletin Americayi Miiseum of Natural History. [Vol. XXIII, 



seen continually coming out of the latter craters laden with brown pellets 

 which they cast aside, while into the gray craters a stream of the same kind 

 of workers is entering in an almost continuous procession laden with green 

 leaves. Some small workers also stand around the openings. On disturb- 

 ing the nest one is severely attacked by the largest workers. With their 

 sharp jaws, worked by enormous muscles, they can bite so severely as to 

 bring the blood; in fact, a small artery in my little finger was severed by 

 one of these workers. The wounds were as much as 4 mm. in length. 

 Nevertheless Mr. Bradbury, a native and myself attacked the nest with a 

 shovel and dug into it deeply. Thousands of the large workers rushed out 

 at us. The half-naked Indian ran away and I had to retreat from time 

 to time with bleeding hands. But the interior of the nest was laid bare. 

 This consisted of a number of great cavities, 15 to 20 cm. long and 8-12 cm. 

 high and each was nearly always filled with a fungus garden, which looks 

 very much like the single garden of the Acromyrmex species. In the laby- 

 rinth of this gray to brown garden live thousands of the smallest and medium 

 sized workers, together with the whole ant brood. Colossal female larvfe 

 are there found covered with a regular envelope of larvae of all sizes, so that 

 they have the appearance of hedge hogs. The workers held fast to the 

 larvae so tenaciously that I could take them in my hands and even kill them 

 in alcohol without their losing their hold.... The large species of Atta 

 therefore have not only one but hundreds of fungus gardens. The fungus 

 chambers communicate with one another by means of broad galleries 2-3 

 cm. in diameter. The lower portion of the garden is uniformly light rust- 

 red with white fungus patches, whereas the upper portions are more gray. 

 The dark brown portions seem to represent the residuum. The fungus 

 garden is so friable that it is impossible to remove it without destroying its 

 form. How the old mvth, or nonsense, that these Afta species line their 

 nests with leaves could have originated and could even be revamped by 

 McCook is incomprehensible to me .... All the pupse are naked, that is, 

 not enclosed in cocoons. The workers have the habit of carrying their 

 straying sisters exactly like our species of Formica (the carried ant is rolled 

 under the head of the carrier)." In another place (1899, 1900&) Forel says 

 that Acromyrmex octospinosus carries its sister workers in the reverse posi- 

 tion, i. e., like Myrmica. He also describes (18966) very briefly the nests 

 and distribution of A. cephalotes and JcEvigata. The latter also has very 

 large but deeply subterranean nests. It lives more in the mountains at and 

 above an altitude of 1,000 m. and so far down in the ground that Forel could 

 not reach the fungus gardens. Cephalotes is intermediate; its nests are 

 nearly as large as those of sexdens and the fungus-gardens have a very 

 similar structure and arrangement. The colonies of cephalotes and espe- 



