430 Journal of Agricultural Research voi. x, no. 9 



investigate thoroughly the possibility of changing the parasitic tenden- 

 cies and capabilities of any biologic form by the use of so-called bridging 

 hosts, by confining a form for a long period of time to uncongenial hosts 

 in an attempt to increase the virulence, and by determining the effect 

 of ecological factors on the rust. The results of the studies of the plas- 

 ticity of biologic forms will be published in a subsequent paper. The 

 present paper deals mainly, therefore, with the identity of the biologic 

 forms on various hosts. 



The problem has received considerable attention both in the United 

 States and in foreign countries. Hitchcock and Carleton (15, 16),^ 

 Carleton (5, 6, 7), Arthur (i, 2), Bolley (3), Bolley and Pritchard (4), 

 Freeman and Johnson (14), Johnson (19), Pritchard (22, 23), Mercer (21), 

 Stakman (24, 25), Stakman and Jensen (26), and Stakman and Piemeisel 

 (27, 28, 29), have done work on various phases of the question in the 

 United States. The taxonomic work of Arthur and the extensive inocu- 

 lation experiments of Carleton, followed by the work of Freeman and 

 Johnson with the biologic forms on cereals, have especially laid the 

 foundation for further work. In Europe Eriksson (8-1 1), Eriksson and 

 Henning (12, 13), and Jaczewski (18) have made extensive investiga- 

 tions. The results are well summarized by Klebahn (20). Although 

 the results obtained by these investigators are usually in general agree- 

 ment, yet there often are differences which indicate clearly the necessity 

 for thoroughgoing work in different regions. A detailed discussion of 

 these results, in so far as they bear directly on the problem under con- 

 sideration, will be postponed until after the writer's results are given. 

 Two examples, however, may be cited to show the necessity for further 

 work. 



Jaczewski (18, p. 353) states that in Russia Dactylis glomerata is im- 

 mune to P. graminis avenae, but is infected by P. graminis secalis. 

 Eriksson (11, p. 601), on the other hand, found that in Sweden the same 

 grass is susceptible to P. graminis avenae and Carleton (6, p. 63) found 

 the same to be true in this country. Carleton (6, p. 54) also reports 

 successful infection with P. graminis tritici. Again, Carleton (6, p. 64) 

 gives the following as hosts for P. graminis avenae: Oats {Avena sativa 

 patula, A. sativa orienialis, and A. sativa nuda — cultivated varieties), 

 Dactylis glomerata, and Arrhenatherum elatius. Pritchard (22, p. 181) 

 obtained results indicating that one form of rust infected rye, oats, Hor- 

 deum juhatum, Agropyron tenerum, A. repens, and Avena jatua. Of these, 

 Carleton (6, p. 56-57) found that Hordeum jubatum was certainly a host 

 for P. graminis tritici and Agropyron tenerum probably so, whereas 

 neither was a host for P. graminis avenae, facts which have been con- 

 firmed by the writers. Moreover, the writers have been unable to infect 

 Agropyron tenerum and Hordeum juhatum with P. graminis from oats, 



' Reference is made by number to " Literature cited," p. 493-495. 



