OK CERTAIN TERMITE NESTS. 203 



light apparently make no difference : the result is always the 

 same, and it has not been found possible to develop the spheres 

 further. 



On the other hand, when a sphere is suspended in a hanging 

 drop of water, hyphse grow out from it in all directions and 

 convert it into a tangled mass in which the original regular 

 arrangement of the conidial branches is quite obscured. A 

 better growth is obtained by using an extract of the comb : if 

 the latter is soaked in water at the ordinary temperature, a 

 deep brown, feebly acid liquid is obtained, which after sterili- 

 zation by boiling is an excellent medium for the cultivation of 

 the comb fungi. The addition of a small quantity of tartaric 

 acid hinders the development of bacteria which otherwise stop 

 all growth in three or four days. 



If the sphere is separated into its component cells and these 

 are suspended in a hanging drop, it is found that growth 

 arises exclusively from the oval conidia ; and a careful exami- 

 nation of an uninjured sphere shows that if the conidia are not 

 separated only the terminal cell of a chain produces a hypha. 



Growth from the separated conidia is far more rapid in 

 comb extract than in water : the number of cells formed is 

 about the same, but they are shorter in the latter case and the 

 resulting hyplne appear more monihform. Figures 7 and 8 

 show four days' growth in water and comb extract respectively. 

 At the end of twenty-four hours the original cell (marked by a 

 cross in the figures) has produced a hypha at both ends, and 

 effected a total growth of sometimes 500 /t : these hypha 1 

 frequently emerge obliquely and curve into line .with the spore. 

 The growing apex is somewhat inflated. In many cases the 

 original spore gives rise later to one or two additional hyplue 

 on either side. Branching arises by side buds immediately 

 below the septa, as in the sphere, but through the further 

 growth of the hypha the branches are ultimately at right 

 angles to it. 



The conidia germinate readily in comb extract agar, with or 

 without dextrose. The best results were obtained by using 



