258 PBTCH : THK FUNG] 



sphere in their natural position produce a hypha at the apes 

 of a branch, and this also obtains in the chains of cells in the 

 Entoloma microcarpum sphere, as well as in the less deter- 

 minate " Kohlrabikopfe" found in the nests of Apterostigma 

 Wasmanni, Cyphomyrmex auritus, and C. strigatus. In 

 general there is a close resemblance between the behaviour 

 of these various " conidial " formations. 



The occurrence of agarics appears to be a much rarer pheno- 

 menon in the nest of the leaf-cutting ants than in those of 

 the termites. When the fungus garden develops them, the 

 upper surface is covered with a continuous layer of brown 

 hypha 1 . The growth of mycelium is more abundant, though 

 the remains of the leaf balls can still be found amongst it. 

 The hypha? of this upper crust are continuous with the 

 white hypha' in t he more loosely built "garden." The agaric 

 is a Pholiota (Pholiota gongylophora) and grows in clusters 

 directly on the upper layer, from which it is impossible to 

 remove them without at the same time breaking away a 

 considerable mass of the " garden. " This condition seems to 

 resemble Entoloma microcarpum more nearly than the termite 

 agaric : in the former the agaric arises directly from the cluster 

 of spheres and these remain attached when it is uprooted. 

 There is no continuous sheet of hypha? over (he termite comb 

 when it produces agarics, though the mycelium is I hen more 

 abundant. The nests of the leal-cutting ants are often 

 covered merely by leaves, sticks. &c. and consequently 

 .in enormous development of the stalk of bhe agaric does not 

 bake place. Midler was fortunately able to germinate the 

 spores of Pholiota gongylophora and to grow from (hem a 

 mycelium which developed the '* Kohlrabihaufchen." 



T. fi<l< mm, iii and T. obscitriceps belong )<» the same sub- 

 genu and all bhe fungi of (heir nests are identical. Moller 

 found Pholiota gongylophora on one occasion on the nest of 



Mia discigt ra . bhree ol her occurrences are noted but I be speciee 

 of ant is not recorded He states, however, thai (he " Kohlrabi- 

 haufchen" in all peciee of the genus Atta which he 



