402 BROOKLYN BOTANIC GARDEN MEMOIRS 
influence in a striking way the capacity of a parasite for infecting 
hosts. 
The extent of the stability of specialized races must vary greatly 
in different cases. Some of the forms above considered are doubtless 
as fixed and constant as many parasites which show structural differ- 
ences. Mention may be made of the races of Melampsora Tremulae 
as illustrating a good case of physiological species. In studying the 
races of Erysiphe graminis one also gets a strong impression of their 
constancy and definiteness and they seem as real as though separable 
by structural features. 
In other cases, however, this is not true. One is not impressed 
with the definiteness of races in Puccinia Helianthi, P. Hieraciu and 
others. These are not characterized by well-defined host limitations. 
Perhaps one is justified in distinguishing different types of specializa- 
tion as physiological species, races, strains, etc. 
But little data are available for comparing the specialization of 
the same fungus in widely separated localities. The specialization of 
Puccinia graminits is apparently the same in Sweden and Russia but 
it has taken a quite different course in North America. Treboux 
(159, 160) finds a different condition in Puccinia coronata in southern 
Russia from what Eriksson (37, 42) finds in Sweden and Miihlethaler 
(102) finds in Switzerland. Carleton’s (26) results in the United 
States also diverge widely from those of Eriksson. Arthur (11, 13) 
does not find evidence for specialization in Puccinia Ribesit-Caricis 
in this country, while Klebahn (87, 91) reports several fairly well- 
defined races of this rust in Europe. On the other hand, there seems 
to be no essential difference in the specialization of Erysiphe graminis 
in Europe and North America. The same sharp host limitation seems 
to occur in both countries. Further data are necessary before we are 
able to determine the relation between the specialization of parasites 
in different regions. 
In a few cases the races are also characterized by minor structural 
differences. Freeman and Johnson (57) and later Stakman and 
Piemeisel (149) have noted variations in the size and shape of the 
uredospores which distinguish the races of Puccinia graminis. Kle- 
bahn (87, 91) has noted similar differences in the spores of the various 
races of Puccinia Absinthi, Melampsora Larici-epitea and others. 
Fischer (46, 47), as well as others, has suggested that physiological 
specialization is a starting point for the origination of forms distinct 
on structural grounds. The suggestion is plausible, for it is possible 
to arrange a series of forms ranging from races differing in host rela- 
tions through all stages to others showing constant structural differ- 
ences. 
