ENDOTHIA PAEASITICA AND RELATED SPECIES. 53 



factor in determining their distribution. As already stated, while 

 Endothia fhiens occurs on Quercus, it has been found toward the 

 southwestern limit of its range (northern Mississippi and Alabama) 

 only on Castanea, and in Tennessee the writers have sixteen collec- 

 tions of this species on Castanea and only three on Quercus. In 

 this same region, E. gyrosa is everywhere abundant on Quercus. In 

 numerous inoculations with E. gyrosa and E. fuens on oak it has 

 been found that E. gyrosa is more generally successful than E. fluens. 

 Moreover, E. gijrosa occurs abundantly on Liquidambar and Fagus 

 in this region, thus providing more numerous sources of infection 

 for this species than for E. fuens. It seems highly probable, there- 

 fore, that E. gyrosa^ with its greater affinity for oak and greater 

 opportunity for infection, may occupy the available oak roots to 

 the exclusion of E. fluens^ even though climatic conditions are favor- 

 able for the growth of the latter species. Castanea rarely serves as 

 a host for E. gyrosa^ consequently, on this host E. -jiuens meets with 

 little competition and is very abundant. 



In the northeastern limit of its range, Endothia fluens has been 

 found only on oak roots. "V^Hiether it grows naturally on chestnut 

 in this region can not well be determined, since practically all the 

 chestnut trees here are dead or badly diseased with E. parasitica. 

 E. gyrosa is rare in this region, but E. jiuens here evidently comes 

 into competition with VaUa fmstum-coni (Schw.) Curtis, which is 

 common on exposed roots of various species of Quercus. 



Since none of the species of Endothia in America extends to the 

 limits of its host species, climate probably has an important part in 

 determining their present ranges. 



In this connection it is of interest to compare several life zone and 

 climatic maps which have been published with the range maps of 

 the various species of Endothia. The map entitled " Life zones of 

 the United States," by C. Hart Merriam (50, pi. 14), is based largely 

 on a study of animal life. Merriam deduces from his studies the 

 conclusion that the northward distribution of animals and plants is 

 determined by the total quantity of heat and their southern dis- 

 tribution by the mean temperature of the hottest part of the year. 

 The life zones which he outlines show, however, a striking relation 

 with the known ranges of Endothia in America. With the exception 

 of a single locality for Endothia gyrosa in Michigan, all the known 

 localities for E. gyrosa and E. -flu-ens fall within the Upper Austral 

 and Lower Austral zones. All the known localities for E. fluens and 

 all the region where E. gyrosa has been found abundantly fall within 

 the humid divisions of these zones. The northeastern limits of the 

 Upper Austral coincide very closely with those of E. gyrosa; its 



