458 RESEARCHES ON FUNGI 



clamp-connections, i.e. become secondary, there can be but little 

 doubt that the fungus is homothallic and not heterothallic. More- 

 over, it is also clear that the stellate chlamydospores are produced 

 only in a secondary mycelium and never in a primary one. It is 

 therefore to be supposed that each chlamydospore contains at least 

 one pair of nuclei, ( + ) and (— ). If now under natural conditions, 

 as seems probable, N. asterophora is propagated from generation to 

 generation by means of chlamydospores, the mycelium throughout 

 such successive generations doubtless remains continuously secondary 

 in its character. However, whenever basidiospores are produced, 

 the pair of nuclei, ( + ) and (— ), cut off in each young basidium, 

 fuse together, thus completing a sexual act, the fusion being 

 followed by two successive nuclear divisions accompanied by nuclear 

 reduction. Thus, as usual in Hymenomycetes, the basidiospores of 

 N. asterophora come to be haploid in their nature and, on germina- 

 tion, give rise to haploid mycelia. We must therefore conclude 

 that, in N. asterophora, it is only when basidiospores are produced 

 that the sexual act is completed ; whilst, when the fungus is pro- 

 pagated from generation to generation by chlamydospores alone, 

 the completion of the sexual act (owing to the life-history not 

 including basidiospore-production) is rendered impossible. A similar 

 condition doubtless occurs in Ptychogaster alhus. 



In the Uredineae, as is now well known, the mycelium which 

 gives rise to the uredospores, the uredospores themselves, and the 

 mycelia which the uredospores produce all contain pairs of nuclei 

 which divide conjugately, i.e. they are all in the secondary or 

 diploid sexual phase, while it is only in the teleutospore that nuclear 

 fusion takes place, thus completing the sexual act. When, therefore, 

 a Rust fungus, e.g. Puccinia Iridis, P. epiphylla, or P. Chrysanthemi, 

 omits its teleutosporic stage and propagates itself solely by uredo- 

 spores, it is clear that its mycelium remains continuously secondary 

 in character throughout successive generations, thus affording us 

 an exact parallel with Nyctalis asterophora when propagated from 

 generation to generation by chlamydospores alone. ^ 



^ Concerning Rust fungi which produce teleutospore-sori only under special 

 conditions, my colleague, Dr. G. R. Bisby, has been good enough to supply me with 

 the following information : " There are several rusts which seldom produce telia, 



