OF CERTAIN TERMITE NESTS. 2.~J-~> 



to differences in the agarics cultivated, and not to differences 

 in the intelligence of the ants. The intervention of the latter 

 need only be demanded to limit or prevent the growth of those 

 hyphee which emerge from a sphere to form others connected 

 with it, and if the spheres are eaten as soon as they are fully 

 formed, this condition is attained. It may be re-stated that 

 such hyphae do grow from the termite spheres just before they 

 decay, and it is conceivable that if the necessary conditions 

 could be realized a mycelium similar to that of Entoloma 

 microcarpum would be produced. 



The spheres of the termite comb show a far higher develop- 

 ment than any of the similar structures which occur in ants' 

 nests. They possess a distinct stalk, are of definite shape, 

 and consist of definitely arranged hyphae bearing conidia of 

 two kinds. But this may be considered to be due to an advance 

 in complexity of the fungus rather than to any specialization 

 induced by the termites. 



Seeing that these "conidia" when in situ only produce a 

 hypha from the terminal cells, it is probably incorrect to style 

 them " conidia." They are never thrown off from the termite 

 sphere. It might be argued that the insects eat the spheres 

 before they are ripe, but in that case we should expect them 

 to develop further when the termites are removed, instead of 

 producing only hypha?. The cells of the sphere of Entoloma 

 microcarpum behave in exactly the same way. It seems most 

 probable therefore that the termite spheres are part of a 

 normal mycelium and not a conidial formation. 



Entoloma microcarpum arises directly from the mycelial 

 cluster of spheres and is gymnocarpous. Its pileus is at first 

 conico-campanulate with an acute umbo, then almost plane; 

 its colour is livid grey when moist, becoming darker towards 

 the umbo ; when dry it is dirty white. Young examples are 

 slightly silky and striate; older ones may be radially streaked 

 owing to the splitting of the surface layers. The marg'n is 

 irregular, at first incurved, and in old specimens sometimes 

 reflexed. The pileus is often split to the centre, and the Mesh 



9(8)06 (2h 



