652 Journal of Agricultural Research voi. xni, no. 12 



can not infect wheat. Rye and v/heat are, therefore, differential hosts 

 for the two forms. But the tritici and the tritici-compacti forms can not 

 be distinguished from each other by their action on grasses, soft wheats, 

 oats, barley or rye. The differential hosts for these two forms are less 

 widely separated taxonomically. It is only by their action on a limited 

 number of different varieties of wheat that they can be distinguished; 

 but the differences on the varieties which do serve as differential hosts 

 are extremely sharp and very consistent. 



In order to definitely establish the identity of the rust in question, it 

 was further cultured on the following differential hosts for P . graminis 

 tritici and P. graminis tritici-compacti (in addition to those mentioned 

 above): Kanred P762 (Kans. 2401), P1066 (Kans. 2415), P1068 (Kans. 

 2414), Barletta (Minn. 1178), Marquis (Minn. 1239), and Royalton (Minn. 

 1037) — all vulgare wheats. 



Club wheat has been found by Stakman and Piemeisel ^ to be highly 

 susceptible to both P. graminis tritici and P. graminis tritici-compacti, 

 while Haynes bluestem was found very susceptible to the tritici form and 

 highly resistant to tritici-compacti. The Kansas varieties, P762, Pi 066, 

 and P1068, have been described by Melchers and Parker^ as decidedly 

 resistant to P. graminis tritici compacti under greenhouse conditions. 

 Barletta, Marquis, and Royalton, on the other hand, are extremely 

 resistant to P. graminis tritici-compacti. Inoculation experiments 

 proved all of these varieties to be very suspceptible to the rust found in 

 Oklahoma. The results of the inoculations, which extended over a 

 period of six months, during which 10 successive transfers were made, 

 are given in diagram i . 



This diagram shows that all the wheat varieties tested were susceptible 

 to the new rust. These varieties, as already mentioned, are differential 

 hosts for P. graminis tritici and P. graminis tritici-compacti (Table I). 

 By using these differential hosts it was possible to ascertain that the new 

 rust was not a mixture of P. graminis tritici and P. graminis tritici- 

 compacti. ^ 



EXPIyANATlON OF DIAGRAM I 



In diagram i Haynes bluestem wheat is represented by W, club wheat by C, Kanred 

 P762 by 762, P1066 by '66, P1068 by '68, Barletta by Ba, Marquis by Ma, Royalton 

 by Ro, and barley by B. Transfers are indicated by dashes; thus, C — B— means 

 that the rust was transferred from club wheat to barley and all other hosts indicated 

 in the same vertical column, as Ba (Barletta), Ma (Marquis), 762 (Kanred P762), etc. 

 The results of inoculations are represented in the form of a fraction, the denominator 

 indicating the total number of leaves inoculated and the i.umerator the number 

 which became infected. 



1 Stakman, E. C, and Piemeisel, F. J. op. cit. 



2 Melchers, L. E., and Parker, J. H. tiirk;; varietiixs op hard red winter wheat resistant to 

 STEM RUST. In Phytopathology, v. 8, no. 2, p. 79- 1918. 



