( 4 02 ) 



The studies have been confined to North America and necessarily 

 of course to those species which are known in both phases, eighteen 

 having been selected for this purpose. While not yielding any 

 results from which impartant generalizations may be made, it may 

 be worth while to present some of the facts as follows: 



A. Geographic distribution of the two sets of hosts very similar 



or almost coinciding — 

 (i) G. Juniperi-virginianae, (2) G. globosum. 



B. Geographic distribution of the two sets not coinciding, one 



much restricted and included within the other — 



a. Range of aecial hosts included within the range of the 



telial hosts — 

 (1) G. trachysorum, (2) G. exterum, (3) G. floriforme, (4) 

 G. juniperinum, (5) G. germinate, (6) G. corniculans, (7) 



b. Range of telial hosts included within range of aecial hosts — 

 (1) G. Botryapites. 



C. Geographic distribution which is common to both sets only a 



portion of the distribution of either set — 

 (1) G. Nelsoni, (2) G. Davisii, (3) G. inconspicuum, (4) 

 G. Blasdaleanum, (5) G. Betheli, (6) G. clavariaeforme, 

 (7) G. Nidus-avis, (8) G. cornutum. 

 While it is impossible to escape the conviction that some extra- 

 ordinary relationship must exist between the sets of hosts occupied 

 by a heteroecious form in its alternate phases the above facts 

 scarcely indicate that the two sets would have been selected 

 because of geographic distributions as now existing. If similar 

 distributions or constant associations of the antithetic hosts are 

 in any measure accountable for their utilization, and it is possible 

 to conceive that they may have had considerable determining 

 power, it is most likely that it was the distribution in very early 

 times, when these fungi were in an unstable and perhaps weakly 

 parasitic condition, that may have had an effect. It seems that 

 very little is to be gained from a study of present day conditions 

 which have become altered by so many factors. The fact that 

 so many species have the range of the aecial hosts included 

 within the range of the telial hosts is noteworthy and calls atten- 

 tion to the fact that in this regard there is a resemblance to the 

 conditions prevailing in most heteroecious species which possess 



