174 memoirs of the new york botanical garden 



Comparisons with other rust geneil\ 

 The relationship of AUodiis to other genera of the tribe Dicae- 

 omae is a most interesting subject. In many cases there are 

 species properly referred to Dicaeoma and Dasyspora which are 

 more closely related to species of Allodiis than they are to any 

 species in their respective genera. A good example of this sort of 

 relationship is afforded by A. effusa on Viola lobata. There is an 

 undescribed rust collected on Viola ocellata from the Rocky 

 Mountains, the teliospores of which are like the former species, 

 yet the rust has a full life cycle and is referred to the genus Dicae- 

 oina. Another undescribed rust collected at Dunsmuir, Cali- 

 fornia, on Viola lobata possesses only pycnia and telia and the 

 teliospores are almost identical with A . effusa. Another example 

 of the same kind of relationship, is found on Brodiaea, and 

 other examples undoubtedly exist. Here we have what to all 

 appearances seems to be three very closely related species or 

 forms of one species, yet each is placed in a separate genus. To the 

 phanerogamic taxonomist such a separation is most confusing and 

 would be considered the height of inconsistency, for it fails to 

 show what a well co-ordinated classification should, namely — a 

 well-defined arrangement of species, families, etc., to show their 

 true relationship to each other and to other groups. It must be 

 conceded, however, that with the fungi, taxonomists are working 

 on a different basis from those who work with flowering plants 

 and animals. While in the fungi the term "species" may repre- 

 sent in a psychological way just as definite a thing as it does to 

 taxonomists of higher organisms, yet in reality the existing rela- 

 tionship is entirely different. Only among individuals where 

 interbreeding of distinct lines of descent exists, do true species 

 occur and therefore in the parasitic fungi the term species has an 

 entirely different interpretation. From this standpoint it would 

 seem to be unimportant whether a classification of such a group as 

 the rusts shows all real relationships so long as the result is a 

 clearer exposition of the nature and life histories of the organisms 

 dealt with. 



So far as comparisons have been made there is no apparent 

 close relationship between species in the genera Allodiis and 

 Bitllaria. A careful comparative study of these two genera is 

 much needed. 



