May, 1923] 
STAKMAN — SPECIES CONCEPT 
243 
investigated thoroughly. For ten years the writer and various colleagues 
have tried in every conceivable way to change the hereditary parasitic 
capabilities of P. graminis tritici and P. graminis secalis. Considerable 
work also was done with P. graminis phleipratensis , P. graminis avenae , 
P. graminis agrostis , and P. graminis tritici-compacti. It was impossible to 
induce hereditary changes, or, indeed, any fundamental changes, although 
the growth of these fungi, like that of other plants, is influenced by environ¬ 
mental conditions. These biologic forms were as constant genetically as 
were the species of wild and cereal grasses upon which they were cultured. 
There was no evidence whatever that the inheritance of physiologic char¬ 
acters by these biologic forms depends any less upon real germinal specificity 
than does the inheritance of structural characters in morphologic species. 
Reed states that “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.” Dox concluded that 
species of Penicillium and Aspergillus could not acquire new ability to 
produce enzyms by any special methods of nutrition; and Brierley was 
unable to “educate” Botrytis cinerea unless the initial culture consisted of 
a mixed population, although a form with colorless sclerotia did suddenly 
appear from a single-spore strain. This phenomenon, however, can be 
explained on the basis of known principles of genetics. Brierley points out 
clearly that it is quite essential to use pure lines of the organism in “fungus- 
educability” studies. This point can not be emphasized too strongly. 
Any one is likely to obtain very striking evidence of rapid changes of biologic 
forms unless his supposed biologic form itself is pure. For example, until 
a few years ago it was supposed that the tritici form of P. graminis could 
change readily. But the so-called P. graminis tritici itself consists of at 
least thirty-seven biologic forms which can be distinguished from each 
other readily by their action on certain pure-line varieties of various species 
of Triticum. All of these forms develop normally on various pure lines of 
Triticum compactum and apparently also on several wild grasses. It would 
be strange, in using such mixed cultures, if changes were not observed. 
Those hosts which were attacked by several of these forms naturally would 
appear to act as bridges to the normally immune forms. The longer one 
works with these forms, the deeper becomes the conviction that they 
represent as real, as constant, and as genetically pure entities as do morpho¬ 
logical species. 
But many biologic forms differ from each other not only physiologically 
but morphologically as well. The forms of P. graminis which are separable 
on the basis of their action on different genera of host plants (in the United 
States) can be recognized by the size, shape, and color of the urediniospores 
and also by the size of teliospores and aeciospores, provided these spores 
are developed on hosts of the same approximate degree of susceptibility 
and in approximately identical environmental conditions. The differences 
