PROTOBASIDIOMYCETES 387 



Heteroecism. In this order of fungi there has been developed 

 not only great diversity in form and character of spores, and in the 

 relationships of these types one to another, but also a very definite 

 relationship between the different spore types and the host plants. 

 Where more than a single spore type is present a species may 

 either complete its entire life cycle upon a single host, that is, 

 produce all spore forms in its life cycle on one host, or it may 

 require two plants for complete development (in a few cases 

 three) in regular order. The former group of rust fungi are 

 termed auta'cioiis and the latter heteroecious. Autcecism is the 

 rule among fungi generally. Heteroecism is better developed in 

 rusts than in any other group of living organisms, and it is with 

 one or two exceptions confined to the rusts, so far as the fungi 

 are concerned. There are more than 150 cases of heteroecism 

 which have been experimentally demonstrated in this order, and 

 this number will be greatly increased as the experimental work 

 proceeds. Upon such hosts as the grasses, sedges, rushes (Gram- 

 ineae, Cyperaceae, Juncaceae), and allied plants, the spore forms 

 produced are quite generally the uredo and teleuto stages, or one 

 or the other of these ; and in general, so far as the experimental 

 work has been carried, such fungi have other stages, at least an 

 aecidial stage, upon some dicotyledonous host. Indeed, in only 

 one group of cases (the species of Puccinia on Phalaris) is the 

 aecidial stage produced on another monocotyledonous host. Again, 

 in no case of heteroecism has the aecidiospore been found to be 

 capable of infecting also the host upon which it is borne. Since 

 the teleutospore germinates by the development of a promycelium 

 and sporidia, and in no other manner, it is precluded that the 

 teleutospore may infect directly the host upon which it is pro- 

 duced. The uredospores alone possess this capacity. 



In general, it would seem that the period of incubation may 

 vary from eight to twenty days during the growing season, for 

 most of the species of rusts. The time, however, will vary in the 

 same species under different climatic conditions. 



The terminology of spore combinations. Based upon the asso- 

 ciation of spore forms, that is, upon the number and kind of spores 

 present in a particular species, Schroeter has proposed certain very 

 convenient type names as below. First, however, it should be stated 



