(401) 



fact that there are only two known species which inhabit hosts 

 outside of the families mentioned above, those being G. exterum, 

 whose aecial host is Porteranthus stipulates, an herbaceous plant 

 of the family Rosaceae, and G. gracilens, whose aecial hosts are 

 Philadelphus and Fendlera of the family Hydrangiaceae. 



One suggestion which naturally presented itself in this study 

 concerns the relation of the distribution of the hosts of the dif- 

 ferent phases. In many of the heteroecious rusts there occurs in 

 one phase a stage which is capable of reproducing itself upon the 

 host which bore it, and in such species it is obvious that the dis- 

 tribution of the phase possessing this repeating stage is not very 

 dependent on the other phase. Consequently one phase might, if 

 its hosts permitted, spread far beyond the distribution of the hosts 

 of the alternate phase. Examples of this situation are so common 

 as to scarcely need mentioning. It is a well-known fact that the 

 black rust of wheat and other cereals and grasses is widespread in 

 areas where the barberry bush is unknown. Another conspicuous 

 example is the leaf rust of rye which is common in North America 

 but which has its alternate phase on a member of the Borage family 

 which is practically unknown in North America, and the alternate 

 stage has never been detected on this continent. 



As has been previously stated, the genus Gymnosporangium 

 possesses no such repeating stage in the present known species 

 and the result is that the association of the hosts of the alternate 

 phases is a necessary condition for the perpetuation and distribu- 

 tion of the fungus. It is well known that the mycelium of the 

 telial stage of many species is perennial, and it has been claimed 

 in certain instances that the mycelium of the aecial stage may be 

 perennial (von Tubeuf 1907), but since the mycelium cannot spread 

 from one individual host to another the distribution of the species 

 is still dependent on the association of the alternate hosts. As 

 to just how close this association must be, no very definite limit 

 can be set. The spores, doubtless, under usual conditions can be 

 blown for distances varying from a few feet up to several miles. 

 Thaxter (1891) mentions a case where he is certain that infection 

 resulted from spores blown about eight miles. .It is not probable 

 that either under the most favorable or unusual conditions that 

 spores could be carried any very great distance and we are there- 

 fore in a favorable position to study the comparative distribution 

 of the alternate hosts in this genus. 



