120 THE AGARICACEAE OF MICHIGAN 



known that the spore prints fade or change after a time, and 

 hence old lierbarium spore-prints are not reliable unless accom- 

 panied by careful notes of the print when fresh. 



The genus may be divided into four natural groups which are 

 here considered as subgenera : Compactae, Eigidae, Subrigidae and 

 Fragiles. Of these, the first and last correspond to the tribes of 

 that name in Fries. (Hymen. Europ.) As shown in a former paper, 

 (Kauflfman, Mich. Acad. Sci., Kep. 11, p. 60, 1909), the forking of 

 the g'ills and the striations on the pileus are not very reliable for 

 the characterization of the main groups. It has seemed practicable 

 to establish a new division, viz., the Subrigidae, to include forms 

 with a pruinose or velvety dry pellicle and rather firm consistencj^, 

 which are out of place elsewhere, and seem to be closely related. 

 Some have divided the genus into two large groups on the basis of 

 the mild and acrid taste (Massee, British Fungus Flora, Vol. III.). 

 Others have used the spore-color (Schroeter, Pilze Schlesiens and 

 Henniugs, Eugler. u. Prantl Pflanzenfamilien). Earle has raised 

 the five ^'tribes" to generic rank (Bull. X. Y. Bot. Gard. 5, p. 373, 

 1909), and finally, Maire has proposed a division of the genus into 

 eight sections based in part on microscopical characters (Soc. Myc. 

 de. France, Bull. 26, p. 120, 1910). The last author appreciates that 

 the groups of Fries are fairly natural and has kept the main fea- 

 tures, while emphasizing the presence or absence of cystidia-like 

 spicules on the surface of pileus and stem. These "cystidia" cause 

 the velvety or pruinose character which I have used in the group 

 Subrigidae. Further studies of all young buttons and their develop- 

 ment will aid materially in a proper arrangement, especially with 

 reference to the character of the margin of the very young pileus. 



The claim of Maire (1. c.) that microchemical tests can be used 

 to advantage, has been given a trial in ten of the following species. 

 This w^ork was done at my request by Dr. W. B. McDougall in our 

 laboratory during the summer of 1912. The results are appended 

 under the corresponding descriptions of the species studied. 



The abbreviations of Maire are used as follows : G = Tincture 

 of Guaiac. S V=Sulfovanilline. F S^Sulfoformalin. The last 

 two are prepared as follows: 



Sulfovanilline. 



Distilled water 2.cc. 



Sulfuric acid, pure 2.cc. 



Vanilline (c. p.) 25 g. 



