778 BuUetin American Museum of Natural History. [Vol. XXIII, 



tions from completed heads to their earUest development in the form of 

 a rich branching at the tip of the bundle of hyphiie forming the stem. The 

 otherwise parallel filaments ramify more and more, till the head is formed. 

 It should be noted that the ends of the filaments do not become thinner 

 while branching but always retain the thickness of the general mycelium. 

 The outer cells grow less rapidly than the others and after a time become 

 passive, thus forming an envelope Avhich later appears as a kind of peridium. 

 The limits of the envelope subsequently become more distinct through the 

 gradual drying up of the outer cells. The inner cells, on the contrary, 

 actively proliferate. The head continues to enlarge owing to the numerous 

 ramifications of the hyphfe, till it has become a sac-like apical thickening. 

 With this increase in size its spherical form changes to an oval. Some time 

 before it attains its complete development, a rapid formation of oidia takes 

 place in its interior, as the hyphse break up into very short oval cells. Only 

 here and there a few of the main filaments remain intact, bvit the lateral 

 branches and greater portion of the hj^hse everywhere break up into short 

 rows of oidia." These oidia are 8-25 ix long and 6-10 /j. broad and have one 

 or two vacuoles in their protoplasm. So complete is this resolution of the 

 hyphse of the head into oidia that a slight pressure on the cover glass causes 

 the dry peridium to burst and thousands of oidia to escape. Holtermann 

 found that the oidia are eaten by the termites, but he expressly states that 

 these insects also feed on dead leaves, stems, etc. When the insects are 

 removed from the garden, the cavities of the latter become stuffed with 

 masses of aerial hypha?, the ripe oidial heads wither up and alien fungi may 

 make their appearance. Holtermann does not believe that the termites 

 are instrumental in preventing these changes under normal conditions 

 since they occur even when termites are present, if the garden is exposed 

 to the light. The normal condition of the gardens may be due to their 

 confinement in dark subterranean chambers, where the spores of alien fungi 

 are unable to germinate. Holtermann is also of the opinion that the above 

 described fungus represents a form of the mushroom which he found growing 

 out of the nests and calls Agaricus rajap. This mushroom has an umber- 

 brown pileus and long gray stem. Its spores are rose-red. He succeeded 

 in growing these spores in a culture liquid, but no oidial heads were pro- 

 duced although the hypliPe sometimes bore club-shaped swellings. Oidia 

 from the termite gardens were also somii and slowly produced hyphse with 

 swollen ends and indistinguishable from those grown from the Agaricus 

 spores. This is not, however, conclusive proof of the identity of the two 

 fungi, although it seems to be regarded as such by Holtermann. 



Karawaiew (1901) has published in Russian an account of this same 

 fungus which he observed at Buitenzorg, Java. His article is accompanied 



