44 Journal of Agricultural Research voi. xvi, no.z 



a given host might change the urediniospore dimensions of a biologic 

 form. 



Recently Long {8) working with Puccinia ellisiana Thuem. and 

 P. andropogonis Schw., whose secial hosts are certain species of Viola 

 and Pentstemon, respectively, found he could change the morphol- 

 ogical characters of the urediniospores of these rusts by reversing 

 their aecial hosts. Thus P. ellisiana after passing through Pentstemon 

 as the aecial host acquired the morphological characteristics of the 

 urediniospores of P. andropogonis ; likewise, P. andropogonis assumed 

 the characters of P. ellisiana after passing through Viola sp. as the 

 aecial host. 



Attention has been called several times to the fact that when a bio- 

 logic form of stemrust develops weakly on a partly resistant host the 

 urediniospores usually are appreciably smaller than they normally are. 

 It seems, therefore, that physical factors might also influence the spore 

 size by affecting the vigor of the rust. But, although a great deal of 

 work has been done on the effect of environmental factors on the severity 

 of rust attacks, the possible correlation between the degree of vigor 

 of the rust fungus and the size of the spores has not been investigated 

 thoroughly. 



Ward {15, p. 2J4) noted tnat differences of temperature, illumination, 

 drouth, etc., affecting the transpiration, assimilation, and other proc- 

 esses of the seedlings, also affect the period of germination, incubation 

 and maturation of the rusts. Fromme {4, p. 50^-509) has tabulated 

 a number of recorded observations of this nature. 



Johnson (7, p. 47) found the cardinal temperature of Pticcinia graminis 

 on wheat, barley, and oats to vary from about 35 to 90° F. Butler 

 and Hayman (i, p. 11) have not succeeded in producing rust arti- 

 ficially on plants grown in the open in the hot weather, in India, and 

 they doubted — 



whetlier the spores have power to infect when exposed to temperatxires exceeding 

 100° F. 



Christman (2, p. 106) — 



found by experiment that [in Wisconsin] in the cooler weather of spring the incu- 

 bation period following inoculation with uredospores is usually lengthened to 

 between three and four weeks. 



Although there was considerable evidence on the effect of these 



environmental factors on rust, their effect was investigated again, 



especially for the purpose of getting evidence of the effect on spore 



morphology. 



EXPERIMENTAL METHODS 



The methods employed in these experiments were (essentially the 

 same as those described by Stakman and Piemeisel (14). But it was 

 thought advisable to obtain additional data on the following points: 



