Aug. 27, 1917 Puccinia graminis on Cereals and Grasses 431 



although it is quite probable that Hordeum jubatum may be infected very 

 rarely and weakly. Carleton's work was done largely in the southern 

 wheat-growing area of the Mississippi Valley, while Pritchard's was 

 done in North Dakota. This suggests a possible geographical specializa- 

 tion of the stemrust, and the writers therefore attempted to obtain 

 inoculating material from as many different sources as possible. 



REGION COVERED BY THE SURVEY 



The work reported in this paper was confined mainly to the upper 

 Mississippi Valley, especially IMinnesota, North Dakota, South Dakota, 

 northern Iowa, northern Nebraska, northeastern Wyoming, Montana, 

 and part of the Red River Valley of Manitoba, Canada. Preliminary 

 work was also done in a small part of the intermountain area of the 

 Northwest, especially in Washington and Idaho. It would be highly 

 desirable to extend the work over the cereal-producing regions of the 

 entire United States and Canada. 



EXPERIMENTAL METHODS 



The rusted grasses were collected in the field, kept in separate envelopes, 

 and immediately sent to the Minnesota Experiment Station, where all 

 inoculations were made. Great care was always taken to avoid any 

 mixing of spores on different grasses. When collecting, the hands were 

 thoroughly washed after each species of grass was handled; or, when this 

 was impossible, the grass was grasped with the collecting envelope as a 

 protection against contaminating the hands with spores. 



Inoculations were usually made with comparatively fresh uredinio- 

 spores, although it was sometimes necessary to use rather old material. 

 Inoculations were made in a few cases a month or longer after the grasses 

 had been collected. The viability of the spores had sometimes been 

 impaired somewhat, but at other times they germinated readily. In 

 making the inoculations every precaution was taken to maintain aseptic 

 conditions, and the writers are confident that erroneous conclusions were 

 not drawn as a result of accidental infection when working with different 

 strains. Usually one person worked with only one strain during the day, 

 although it was sometimes necessary to work with more. The plants to 

 be inoculated were grown in a greenhouse compartment where no rust 

 was kept. They were grown under cages made of two layers of a fine- 

 mesh muslin with a dead-air space between the two. 



The inoculations on cereals were all made on the leaves or sheaths of 

 seedlings. It was found that the results were comparable with those 

 obtained on older plants. The inoculations on grasses were sometimes 

 made on old plants. In general, the best results were obtained on young, 

 vigorously growing plants, although with some grasses the opposite 

 seemed to be true. Although the writers observed no cases of immunity 



