782 Bulletin American Museum of Natural History. [Vol. XXIII, 



that has passed through the bodies of the insects, are dish-shaped, and there 

 may be several piled one on top of the other in a single chamber. They 

 are perforated with galleries filled with the termites and their larvae. "On 

 taking one of these brown cakes in the hand, one can see with the unaided 

 eyes that its whole surface is covered with a fine bloom of fungus mycelium. 

 When broken open the interior of the galleries is found to be covered with 

 peculiar white spherules about as large as a pin-head (1-2 mm. in diam.)." 

 Doflein's description of the minute structure of these spherules is less 

 explicit than that of Holtermann and Tragardh, but he actually saw the 

 termites swallow these bodies when they were presented on the point of 

 a sterilized needle. They were eaten by the larval workers and soldiers 

 and by the adult kings and queens, but the adult workers and soldiers would 

 not take them. The intestines of the latter contained only comminuted 

 wood in which no fungus elements could be found. Doflein, is, therefore, 

 of the opinion "that in this species the larvfe are fed with a concentrated 

 and easily assimilated food in the form of mycelial spherules, and that these 

 constitute the permanent food of the sexual forms, w^hereas the larvae of 

 the workers and soldiers are not fed with these after reaching a certain age 

 but with other substances [dead Avood] instead. This suggests the further 

 inference that this food may play an important role in the differentiation 

 of the castes of Termes obscuriceps Wasmann." 



Doflein found that when the fungus garden of this insect is placed in 

 the light under a bell-jar to protect it from evaporation "the termite fungus 

 can easily be induced to fructify, a peculiarity in which it differs from the 

 fungus cultivated by the South iVmerican leaf-cutting ants. In the course 

 of a few days numerous long, club-shaped fruiting organs grow up out of 

 the dense mass of hyphae, which has developed in the meantime. As time 

 goes on these club-shaped bodies develop pilei, which, as Mr. Green of Pera- 

 denyia informs me, are now known to be those of an Agaricus, a fact which 

 is also indicated by my own observations. While the fungus is growing up 

 freely in this manner, one is surprised to find alien fungi gradually making 

 their appearance in the garden, and other objects in the neighborhood taking 

 on the usual mouldiness. The tendency of the termite fungus to grow as a 

 pure culture must therefore be very great. This is the case even when very 

 few termites are present. Hence the purity of the culture cannot be ascribed 

 to a ceaseless weeding process carried on by the termite workers, like that 

 assumed by INIoeller in the case of the South American Attce." 



When the garden is left under the bell-jar the under surface of the latter 

 soon becomes wet, showing that the fungus gives off a great deal of water. 

 In a day or two the termites become suffocated, although masses of these 

 insects hermetically sealed between pairs of watch glasses manage to live in 



