N. I. Vavilov 53 



With respect to wheat the ajjjjlication of fungal reactions is a 

 specially grateful method, because in this instance we have at our 

 disposal many narrowly specialized parasites, in other words many 

 sensitive physiological reagents, as brown and yellow rusts, mildew, 

 etc. When one test is not sensitive enough to distinguish our races, 

 we may have recourse to a second, a third, and so on. 



In the above-mentioned paper (16) many examples of such a division 

 of varieties of oats and wheat into races are given. As a, perhaps, 

 interesting result, I may mention that many races were found to exist 

 in some of the less common varieties, as Avena diffusa As. and Gr. 

 var. hrunnea Kcke., var. cinerea Kcke., var. montana Al., and in several 

 rare varieties of Triticum vulgare Vill. 



In the species Triticum dicoccum, which is represented by many 

 varieties and races, it was ascertained that there exist two groups of 

 races : one immune to brown rust, the other susceptible to it. The 

 genetic significance of this will be dealt with later on. Here we must 

 observe only, that the resistant forms of Tr. dicoccum, which were in 

 our collection, are morphologically, in the structure of ears and leaves, 

 very like several varieties of Tr. durum. The degree of immunity of 

 these latter to brown rust is also nearly the same as in the resistant 

 races of Tr. dicoccum. 



The most interesting example of the application of this method to 

 the systematic study of wheat I met with, is the following : 



In the investigation of 580 sorts belonging to the species Tr. vulgare, 

 which in general is very susceptible to mildew {Erysiphe graminis D.C.) 

 and to different rusts, we found to our surprise a spring race which was 

 perfectly immune to mildew. Notwithstanding many attempts at arti- 

 ficial infection in the field, in the greenhouse or under bell-jars, this 

 race remained quite immune. Not one pustule of mildew was found 

 on this wheat, whereas other races of Tr. vulgare in these conditions 

 were severely attacked by the above fungus. This wheat proved also 

 to be relatively very immune to brown rust (P. triticimi). The seeds 

 of this wheat were obtained from a German seed merchant under the 

 name of " Persian Wheat." 



Such an extraordinaiy immunity as distinguished this race from 

 other races of Tr. vulgare and from the majority of I'aces belonging 

 to other species made me pay exceptional attention to this form, and 

 the preliminary investigations so far made have revealed many further 

 peculiarities in it. Although a number of morphological characters 

 show that it belongs to e(jnnnon wheat, namely to Tr. vulgare var. 



