Number of 

 inocula- 

 tions 



Percentage 

 ea 



of 



ch c 



infect 

 lass 



ion in 



Quick 



Unci 



as si 



fied 



Slow 



40 



33 





67 









38 



3 





92 





5 



36 









69 





31 



40 



8 





89 





3 



37 



8 





88 





4 



40 



23 





77 









36 









78 





22 



80 EVA FUCHS 



Table 11. Differentiation of races of stripe rust on the basis 

 of latent period 



Culture 



Race 7 from France 

 Race 8 from Germany 

 Race 32A from France 

 Race 32A from Switzerland 

 Race 3/55 from France 

 Race 3/55 from Netherlands 

 Race 3/55 from Switzerland 



For stripe rust, Straib (1940) and later on Schroder and Hassebrauk 

 (1964) stated that the differences between races were too small to serve 

 effectively in determination. 



The same seems to be true for differences in optimal germination 

 temperatures and germination speed where a lot of statistical work is 

 required, although these traits have some value (Manners, 1950; Fuchs , 

 unpublished) . 



Attempts to show differences in physiologic races with electrophoretic 

 patterns of spore proteins (Macko, Novacky, and Stahmann, 1967; Shipton 

 and Fleischmann, 1969) are too recent to be discussed. 



CLASSIFICATION, TAXONOMY 



As long as there is no general applicability of new methods, we 

 should apply classical race determination in cereal rusts, relying on 

 differential varieties or "resistance genes" (monogene differentials) . 



Taxonomy and nomenclature are problems every determinator of rust 

 races will meet. Kernkamp (1965, p. 822) stated, that either a world 

 center for race determination and nomination for each cereal rust has to 

 be established "with trained personnel, genetically pure differential 

 varieties, and growth chambers so that all identifications could be made 

 with the highest degree of precision," or that we carry on as is with 

 each investigator naming a race "as best he can with the differentials 

 and environment he has available." As shown in the survey of Johnson, 

 Green, and Samborski (1967) , some race determinators are already working 

 with "genes"; however, many of them are not. For my own purpose and as 

 a quintessence of the struggle with stripe rust "races" I am now going 

 to use the abbreviations of the differentials themselves for race 

 description in the sequence of their utility, which means: good differ- 

 entials first (major genes?), less good differentials later (minor genes?), 



