62 Journal of Agricultural Research voi. xvi.no. a 



A rather distinct and consistent specificity in shape and color was 

 exhibited by the rye rust urediniospores regardless of the host on which 

 they developed, length of time confined to it, or of origin and subse- 

 quent history. As seen from Table VI, the spore modes average more 

 than 5 n shorter and about 2.5 /x narrower than those of the spores of 

 P. graminis tritici; the shape is cylindiic-ellip'tic, and the color dull, 

 ashy-yellowish to grayish. The size of the spores is perceptibly affected 

 by cultural conditions, but not by the host plants. Under the same 

 circumstances the spores were in each case distinctly and consistently 

 smaller than those of P. graminis tritici. During the cloudy and cold 

 weather of midwinter and the excessive heat of the summer months the 

 spores grew only to their minimum size and again attained their maxi- 

 mum size in spring and fall. 



P. graminis avenae (Table VII) was thought to be an especially inter- 

 esting form for study in the attempt to determine the amount of varia- 

 tion which could be induced by growing it on different congenial hosts, 

 because it appeared from inoculation results that this biologic form 

 might possibly be considered as plastic. In contrast to P. graminis 

 tritici, P. graminis tritici-compacti, and P. graminis secalis, the uredinio- 

 spores of which are, as a rule, of a definite specific size and shape when 

 grown on a congenial host, those of P. graminis avenae are very 

 variable even when parasitizing oats. They may be ellipsoid, 

 ovoid, pyriform, or subglobose in shape. In size they are somewhat 

 longer than those of P. graminis secalis, and in width approach those of 

 P. graminis tritici. The spore color is bright cadmium-yellow. 



On Dactylis glomerata the spores show a tendency to shorten, the width 

 remaining practically the same as on oats. On Bromus tectorum, which 

 is only a tolerant host, the spores decreased both in length and width, 

 becoming nearly globose. The presence of equatorial germ pores, how- 

 ever, showed clearly that the rust was P. graminis. Both barley and 

 timothy are very uncongenial hosts for P. graminis avenae and both 

 uiedinia and spores are very small (13). This is similar to the behav- 

 ior of P. graminis tritici on resistant varieties of wheat on which 

 Stakman (11, p. 31) found the spores to be smaller than on the suscep- 

 tible wheat varieties. Similar results were obtained by the writers with 

 P. graminis tritici-compacti (see Table V). 



It will thus be seen that the size and shape of P. graminis avenae are 

 quite easily influenced by the host on which they grow; the color, how- 

 ever, remains constant. The size of uredinia is directly proportional to 

 the size of the spores, and vice versa. This morphological variation is 

 interesting because the rust is also versatile parasitically. 



