202 PROTOBASIDIOMYCETES [CH. 



division takes place (fig. 175). The origin of the trinucleate cell by the 

 fusion of three fertile cells has been observed, and no doubt it may arise by 

 the migration of a second vegetative nucleus. 



.•^C7"r-' 



r.;.'^.:: 





::-^;;?^?u 



'■\ ■,-- 



I'^'g' I75- ^2'f"«M /"fla^ww Niels. ; conjugate division, X2280; 

 after Blackman and Fraser. 



At the periphery of the aecidium the cells cut off from the basal cells 

 divide in the usual way, so that two cells corresponding respectively to the 

 aecidiospore and the intercalary cell are formed from each. The upper 

 (aecidiospore) cells acquire very thick striated walls, lose their contents 

 and form a sheath or pseudoperidium about the sporogenous part. The 

 behaviour of the lower cells varies considerably ; Kurassanow has shown 

 that in some cases they are quite small, like typical intercalary cells, while 

 in others they are relatively well developed and form part of the p.seudo- 

 peridium. This is especially the case where the tissue to be broken through 

 by the developing aecidium is dense or extensive. Centrally the pseudo- 

 peridium arches over the contents of the aecidium. In this region it is 

 derived from the cells first cut off by the central basal cells. These, like the 

 others, divide transversely, and one of the daughter cells, usually the outer 

 one, corresponding to the aecidiospore, becomes one of the elements of the 

 pseudoperidium (Dittschlag, Kurassinow). When the aecidium reaches 

 maturity the pseudoperidium pushes through the epidermis of the host and 

 is then itself ruptured and exposes the ripe spores. It becomes torn and 

 recurved so that the characteristic cluster-cup is produced (fig. 176). The 

 pseudoperidium is sometimes much elongated and cylindrical or inflated, 

 producing the forms known as roestelia {Gyinnosporangium), and peri- 

 dermium (Coleosporiunt, Cronartium. and allied genera), so-called from their 

 old generic names, or it may be represented only by a few paraphyses or 

 altogether absent (Phragmidium, Melampsord). The latter forms, to which 

 the term caeoma is applied, are probably primitive. 



In the majority of cases, after the fertile and sterile cells have been 

 formed and nuclear association has taken place, the basal cells each give 

 rise to a single chain of spores, but occasionally {Piiccinia Falcariae (fig. 

 177), Endophylliim Sempervivi) they may branch and thus produce two or 

 more spore-rows. In certain other species the basal cells regularly form 

 a number of lateral buds or branches and each of these is cut off as a spore 



