BASIDIOMYCETES 439 



spermatia of a different strain. In other words it is now recognised that where- 

 as some Rusts are homothallic, others like Puccinia graminis are hetcrothailic 

 (see p. 416) ; the latter require the association of hyphae from two comple- 

 mentary strains before cluster-cups and aecidiospores are developed. This 

 association may be brought about by the overlapping of pustules. Insects 

 visiting the spermogonia for " nectar " bring about a transference of spermatia 

 and these may become intimately associated with flexuous hyphae of the 

 opposite sex. Spermatial nuclei then enter the hyphae and the subsequent 

 nuclear divisions and migrations bring about the cytological condition necessary 

 for the development of aecidiospores. In all cases, nuclear pairing is believed 

 to occur at the base of young aecidia. It has been traced in Phragmidium 

 violaceum, as consisting in the passage of the nucleus of one cell into a neigh- 

 bouring cell (Fig. 338, A), very much as has been seen in certain apogamous 

 Ferns (Fig. 403, p. 508). But here the nuclei do not fuse at once ; the receptive 

 cell remains bi-nucleate, and divides as such into a chain of bi-nucleate spores 

 (a) and sterile intercalary cells (2) (Fig. 338). Each chain of spores appears to 

 be initiated in this way. A similar process has been seen in P. speciosum, and 

 other Uredinales. This discovery made it possible to relate the life-history 

 of the Uredinales to that of other plants. The state with paired nuclei is 

 held as the correlative of the diploid sporophyte, and that with a single 

 nucleus in each cell as the haploid gametophyte. The events may then 

 be summarised as follows : — 



A. Sporophyte on Grass B. Gametophyte on Barberry 



(paired nuclei). (uni-nucleate). 



i. Mvcelhim. viii. Mycelium. 



ii. Uredospore (repeating the ix. Spermatium. 



mycelium). • x. Nuclear association. 



iii. Teleutospore (winter's rest). xi. Aecidiospore (bi-nucleate) 

 iv. Nuclear fusion in germinating (Fig. 338, D). 



teleutospore. xii. Infection of Grass-leaf. 



v. Basidium ( =promycelium). 



equivalent of a spore-tetrad, 



with reduction. 

 vi. Basidiospore ( =sporidium). 

 vii. Infection of Barberry plant. 



There are many of the Uredinales that do not show so elaborate a series 

 of stages in the life-cycle as the Rust of Wheat. The uredospores may be 

 omitted, or the aecidia ; or some may have neither aecidia nor uredospores. 

 The spermogonia may also be omitted : so that in these last the life consists 

 of a repetition of teleutospores with subsequent germination. 



The Rust disease is difficult to cheek, and its distribution is world- 

 wide. One obvious measure would appear to be to remove the 

 alternative host, the wild Barberry. This has been done in wheat- 

 growing districts in England, but definite consequences from it are 



