September, 1907] Mycolo^ical Bulletin Xo. 81. 357 



baria referred to tliese two species, not one was accompanied by descriptive 

 notes that were ot any value. Yet we have at least a half dozen good 

 species here represented, cuiild the distinctive characters be clearly es- 

 tablished. 



"The species of this family are not only comparatively rare and local 

 in distribution, but they are often intermittent in appearance. The writer 

 once found three different species in a space not over ten feet square, and 

 a fourth in the same woods a short distance away. But not one of the 

 four was found anywhere in that i-egion in the next four successive years, 

 altliough the ground was searched over repeatedly each year. 



DISTRIBUTION. 



"The gei)gra])liica] (]istriI)Ution of these plants appears to be largely 

 influenced by latitude. P.nt collections of IJasidiomycetous fungi from 

 the region we-t of the Mississipi river have been so few and incomplete 

 thrt general conclusions respecting distribution in this region can not be 

 confidently drawn. The following areas may be recognized as possessing 

 each a characteristic and somewhat distinctive hydnaceous flora. (1) The 

 northeastern United States south to North Carolina and Tennessee and 

 west to the Great Plains. (2) The Southern States west to Lousiana. (5) 

 The Gulf region including the West Indies and the immediate borders of 

 the Gulf. (4) The north Pacific coast including Oregon and Wash- 

 ington. It seems probable that Canada and northern New England to 

 Greenland may represent another distinct floral distribution, but collections 

 in this region have been too meager to suggest more than a possibility. 

 These remarks on distribution are based on specimens actually seen by 

 tha writer, and do not include the various species reported in catalogues 

 and local floras without accompanying specimens. The material examined 

 .has come chiefly from the following states: Maine. Massachusetts. Con- 

 necticut. New York, New Jersey. West Virginia. Ohio. Indiana. Kentucky. 

 .'Mabama. Louisiana. Cuba. Honduras. Oregon. Washington. This study 

 of distribution cannot be satisfactorily supplemented by published local 

 floras, for in consequence of the confused conception of species in this 

 family, such lists are wholly unreliable except when verified by actual 

 Specimens, and these are often lacking. A comparison of two collections 

 on which such floras have been based, quickly reveals how utterly untrust- 

 worthy are these lists of species as a means of determining distribution. 

 The plants referred 1o Hydinnii imbricaitun by Alabama collectors are 

 totallv distinct from the plant referred to the same species by the New 

 England botanists. Professor Earle has noted that Ifydiium rcfciidum as 

 collected by him in Connecticut, was a very different thing from the plant 

 of that alliance with which he had been acquainted in Alabama. 



"It is lidiH'd tlipt the present contribution may lead to a clearer con- 

 ception respecting the species of this family and be a means of stimulating 

 a more exact study of the distribution of these plants. It can hardly be 

 expected that all confusion has been removed or that all errors have been 

 avoided. The source of many of our present difliculties is to be traced 

 back to the work of early European botanists, whose material is either 

 inaccessible or has long since passed into an irrecoverable oblivion. The 

 rtnthor believes that in the majority of cases, with respect to the species 

 included in this paper, he has formed a clear conception of them in his 

 own mind and has endeavored to present that conception as definitely and 

 distinctly as he was able in the accompanying descriptions and synopses. 

 Whether he has in all cases made an absolutely correct determination, 

 especially in the case of species referred to old European types, he cannot 

 State with complete confidence." 



