PHYSIOLOGIC SPECIALIZATION AND VARIATION 211 



differences even after repeated transfer, whereas others reverted 

 to the normal. 



Importance of physiologic specialization and variation. 

 Thus far emphasis has been placed upon the fact that specialized 

 races exist, and their possible origin has been considered. The 

 significance and practical application of these facts and theories 

 cannot have been kept from mind during the perusal of this dis- 

 cussion. Their importance in the field of plant pathologv is not 

 believed to be properly appreciated; indeed, it can scarcely be 

 overestimated. For a period of years these problems have engaged 

 the attention of many students of the rusts and smuts, especially 

 Stakman and his associates. In a report Stakman (1936) has sum- 

 marized them in their application to the need for plant quaran- 

 tines and to the breeding of varieties resistant to disease. 



Investigations at the Minnesota Agricultural Experiment Sta- 

 tion, at, the Dominion Rust Research Laboratory, and in Australia 

 [Waterhouse (1929)] are in accord in showing the relationship 

 between the presence of barberry and the existence of numerous 

 races of Puccinia graminis tritici. Race 34 seems to be the only 

 one present in quantity in Australia, whereas in the wheat-grow- 

 ing belt of North America approximately 150 races are known 

 to exist. In addition, new races are continuously being developed 

 as the result of hybridization on the barberry. The unrestricted 

 introduction of the North American races into Australia or other 

 continents might easily result in epidemics of rust on varieties of 

 wheat that are highly resistant to races of rusts already present in 

 these countries. This supposition is supported by Stakman's 

 (1936) observations on the rust epidemic on the varieties Ceres 

 and Thatcher in 1935. Both had previously rather uniformly re- 

 sisted rust for a term of years. It should be added that in 1935 

 no races of rust capable of infecting Vernal emmer were isolated 

 from uredinial collections made in reoions where barberries are 

 absent, whereas three capable of infecting this variety of emmer 

 were isolated from barberries or from grain growing near them. 

 Several of the races isolated that year from barberries were new 

 biotypes, showing that hybridization and segregation in rusts are 

 taking place in nature. Experiences of this sort should convince 

 the most hard-headed unbeliever that barberries should be 

 eradicated. 



