THE FUNGI WHICH CAUSE PLAXT DISEASE 331 



In all of the rusts so far investigated that have an aecium or 

 primary uredinium there is in the aecio-mycehum or the primary 

 uredinio-mycelimn a fusion of uninucleated cells, gametes. This 

 cellular fusion is not, however, followed by a nuclear fusion until 

 after long delay; but the two nuclei remain in the fusion cell and 

 when this cell divides both nuclei divide mitotically and simul- 

 taneously but still independently of each other (conjugate division). 

 This process continues through the aecial sporophores, or uredinial 

 sporophores, and in the production of the spores, with the result 



J ^3. Cm., 



rf-% 



t 



Fig. 244. — Showing 

 Ronjugate nuclei and 

 degenerating cells in 

 conidiospore chain 

 of .£cidium. After 

 Sappin-TrouSfy. 



Fio. 245. — Gymnospo- 

 rangium, clavariiB- 

 forme, mitosis of a 

 nucleus in the promy- 

 eelium. After Black- 



FiG. 246. — Conjugate 

 nuclear division in 

 Gymnosporangium 

 clavarioeforme show- 

 ing four chromatin 

 masses. After 

 Blackmau. 



that the cells of all of these are binucleate. The conjugate divi- 

 sion continues further through the uredinia and until teUospore 

 formation occurs, the whole intervening series of cells being binu- 

 cleate. Prior to the formation of the promycelium and in the 

 teliospore the nuclei unite, reducing the cells again to an uninu- 

 cleate condition. 



In rusts which have only teliospores the binucleate condition 

 begins somewhere in the mycelium from which the teliospores 

 arise. 



It is generally held that the cellular fusion is a sexual act 

 with long delayed fusion of the sexual nuclei: and consequently 

 that the uninucleate phase is the gametophj^te; that the be- 



