24t) PETCH : THE FUNGI 



spheres fragments of wood, and violet-black hyphae and 

 spores, but these latter may have been obtained by eating 

 some of the comb, and cannot be taken as proving that the 

 sexed insects share the functions of the workers. 



The non-occurrence of injured spheres on the comb suggests 

 that each is devoured as a whole. If so, although the spheres 

 are abundant, there does not appear to be at any time a 

 sufficiently large number to supply food for the crowded hosts 

 of larva?, unless the growth of the fungus is abnormally rapid. 

 There are however in most large nests a number of combs 

 which contain only a few workers and soldiers, and the fungi 

 on these might conceivably constitute temporary food reserves. 



It may be mentioned that in the case of workers, soldiers, 

 and larvie enclosed in a damp chamber with the object of 

 carrying out feeding experiments, the workers and soldiers 

 died before the larvae. 



Other Fungi on the Comb. 



The purity of the termite fungus cultivation has been re- 

 marked by all observers, though as has already been shown 

 this purity is only apparent. On a normal comb just removed 

 from the nest one never finds any fungus except the spheres 

 and their mycelium, but the growth of Xylaria (?iigripes^) 

 takes place regularly in bell glass cultivations after two days, 

 before the development of any other fungi, and both its lux- 

 uriance and regularity of occurrence support the view that its 

 mycelium vegetates in the comb substance, though it has not 

 been found possible to distinguish it from the mycelium of the 

 11 spheres." 



The comb material is probably at first sterilized by its pas- 

 sage through the alimentary canal of the termites. DOflein's 



suggestion that it is only partially sterilized, the spores of the 

 cultivated fungus being uninjured, is scarcely credible : it would 



be more reasonable to suppose thai the thick-walled ascospores 



Kylarxa escape injury. If the oomb substance is sterilized, 



however, tins would necessitate an inoculation of every nev 



