1907.] Wheeler, Fungus-groiving Ants of North America. 777 



taprohanes and fatalis in India and the Malay Archipelago (Singapore, 

 Java and Borneo.) He says : "These animals build their nests in the ground ; 

 once only did I find them nesting in a log. Notwithstanding their clandes- 

 tine mode of life, I have been able to investigate hundreds and hundreds 

 of their singular habitations, for I was able to find them easily by means of 

 a species of Agaricus which was always rooted in a termite nest. It was only 

 necessary to follow the stem of the pileus into the earth, although in some 

 cases I had to dig to a depth of a meter." Like Nietner, Holtermann refers 

 to the fungus gardens as "nests." These varied from the size of a walnut 

 to that of a man's head and were of a sponge-like structure, full of holes and 

 galleries containing termite eggs, larvtie and nymphs. The gardens were 

 found resting on the floor of the earthen chambers and were separated from 

 the walls by a space as broad as one's finger. They consisted of finely 

 comminuted vegetable substances (portions of dead leaves and stems) that 

 had passed through the bodies of the termites. Under the microscope 

 "the surfaces of the galleries were seen to be covered with a white felt-work 

 of mycelium. Usually the hAi^hpe were loosely united but sometimes they 

 were combined in strands. The individual hyphiie were richly septate but 

 showed no 'Schnallenbildung' at the septa." Even with the unaided eye 

 Holtermann could detect aerial hyphre projecting from the general felt- 

 work of the mycelium. "The terminal and often the penultimate cells 

 of these hyphfe were filled with strongly refractive, hyaline protoplasm, 

 whereas the remaining cells contained remarkably little plasma. The 

 terminal cells were often swollen and club-shaped. Sometimes the tip 

 even became spherical but only in its upper portion. In exceptional cases 

 the hyphfe anastomosed, most frequently through confluence of the terminal 

 cells." This mycelium ramified through the whole substratum which it 

 perhaps served to bind together. The swollen tips of the hyphre were often 

 aggregated to form bi'omatia like those of Atta, but Holtermann failed to 

 find them in all termite colonies, and believes that they may occur only in 

 the gardens of certain species. In addition to these structures he describes 

 others of a more interesting character, namely, small spherical bodies dis- 

 tributed everpvhere on the mycelial net-work. They were white, varied 

 from .25-2 mm. in diameter and were usually attached by a peduncle .5-1 

 mm. in length. The minute structure of these spherules which were not 

 abundant in the interior of the garden, is described as follows: "The pe- 

 duncle consisting of nearly parallel hyphte becomes wider below and loses 

 itself in the substratum; otherwise it is of uniform thickness and the head 

 is sharply marked oft' from its end. The rudiment of the head appears as 

 a distinct thickening at the tip of the stem and as soon as the head is estab- 

 lished the stem ceases to grow. In every chamber are found all the transi- 



