SAPROPHYTES AND PARASITES 79 



apex when nuiture, and then globose and warted. This stage 

 was formerly known as Accidium violae. Before either sper- 

 niogonia or aecidia appear there is always present a plentiful 

 mycelium in the tissues. The swelling of the petioles is caused 

 l)y the development of this mycelium, tlie cluster-cups l^eing 

 developed from the same mycelium as the spermogonia, and 

 consequently deeply innate and thoroughly endopliytal. The 

 mature aecidiospores, after voluntary separation from the chain, 

 will germinate within a few hours, but seldom after a period 

 of forty -eight hours. Each spore has several germ-pores, 

 perhaps four or six, but germination seldom proceeds from 

 more than one. This cylindrical tube continues growing until 

 it has acquired a considerable lengtli, the coloured contents of 

 the spore passing meantime along the tube to its extremity, 

 which finally enters one of the stomata of the proper host- 

 plant, and there, by branching and progressive growth, con- 

 stitutes a mycelium, presumably the mycelium which becomes 

 the spore-bed of the uredospores. If we return to the violet 

 leaves later in the year, we shall find the under surface 

 of many leaves exhibiting small raised pustules, which are 

 scattered all over the surface. These sori, or pustules of the 

 Uredo, are soon exposed by the irregular splitting of the 

 cuticle, and the light brown spores, resembling snuff, are freely 

 distributed. Examined more closely, each pustule will be 

 found to possess a spore -bed of compacted mycelium, from 

 which the uredospores grow, at the apices of rather short 

 hyaline threads or peduncles, which are soon absorbed, leaving 

 an elliptical pale-brown spore, with a shortly spinulose surface, 

 as the second stage of an alternation of generations, the problem- 

 atic spermatia being left out of the question. It must be remem- 

 bered that the origin of an uredospore-bed is not absolutely 

 resultant from a germinating aecidiospore, but it may also be 

 produced by a germinating uredospore, or by the germination 

 of a promycelial spore. This fact may be associated with the 

 other fact, that some species of Puccinia are known with whicli 

 no aecidium has yet been associated. The mature uredospores 

 have two, three, or four points of germination or germ-pores. 

 The germination takes place, as in the aecidiospores, within a 

 few hours, and in liki> manner tlie growing point enters one 



