Oct. 28, 1918 Plasticity of Biologic Forms of Puccinia graminis 223 



On account of the weight of the above-cited evidence the role of bridg- 

 ing hosts in breaking down biologic-form specialization has been given 

 fairly general credence. 



But Eriksson (5), Jaczewski (/o), Freeman and Johnson (9) and Stak- 

 man {18) could detect no clearly appreciable influence of barberry on the 

 parasitism of biologic forms of stemrust. Stakman {18), Stakman and 

 Jensen (19), and Stakman and Piemeisel {20), in rather limited experi- 

 ments, could not duplicate the results of Freeman and Johnson with P. 

 graminis nor those of Johnson with P. graminis phleipratensis ( = P. 

 phleipratensis) . Stakman (18), however, obtained results indicating that 

 possibly changing the host metabolism by the use of anesthetics and fer- 

 tilizers might increase the parasitic capabilities of the rust slightly, thus 

 giving some support to the work of previous investigators (12, 14, 15, ij) 

 along similar lines. 



On account of the undoubted theoretical and practical importance of 

 the problem it seemed desirable to make extensive experiments with a 

 number of forms of rust from different hosts from different regions. 

 Work was therefore begun in the summer of 1914 and continued unin- 

 terruptedly since that time. 



Although there was already considerable evidence on the question of 

 the effect of barberry on the rusts, it seemed desirable to do still further 

 work. An attempt was therefore made to determine the possible role of 

 barberry as a bridging host and also to determine its possible effect as a 

 reinvigorator of the rust. 



Material was obtained from as many different sources as possible for 

 the work with cereals and grasses as intermediaries or bridging hosts. 

 While much of the work with so many different strains might appear to 

 be superfluous, nevertheless there has been some idea that strains of the 

 same biologic form might differ somewhat parasitically. Naturally, 

 therefore, it would be desirable to get data on as many different strains as 

 possible. 



Most of the work was done with the tritici and secalis forms because 

 they are the most important economically in the spring-wheat States and 

 because theoretically it seems probable that bridging should take place 

 with these two forms on account of the close similarity of the rusts to each 

 other in many respects. It has already been pointed out by the writers 

 {21) that the tritici and secalis forms have many hosts in common. 

 Thus, barley, various species of Elymus, Hystrix, Hordeum, and Agro- 

 pyron are about equally congenial hosts for both rust forms. It would 

 seem that if the idea of bridging is well founded, and if the host plants 

 actually do exert a distinct and permanent effect on the rusts, these com- 

 mon hosts ought to unify rust strains which are grown on them. 



Barley {Hordeum spp.) was used more than any of the other forms in 

 attempted bridging on account of the fact that Freeman and Johnson 



