(398 ) 
development attained by the family it seems only logical to 
accept the presence or absence of one or both of them as a 
character of primary importance. Two practical inconveni- 
ences, however, must be admitted. First, the two mem- 
branes are often confounded in descriptive works and, as 
either of them may or may not form an annulus on the stem, 
it is often impossible to determine from the literature which 
organ is really present. Second, the rudiment of a veil 
consisting of a more or less evident powdery or waxy coating 
is said to exist in the very young stages of even the most 
pronounced gymnophyllous species, while in many that are 
clearly cryptophyllous the veil entirely disappears at an early 
stage in the growth of the pileus. However, there are prob- 
ably very few cases in which a careful examination of the 
younger stages will not show at once to which of the two 
series any given species belongs. Only three tribes are 
here recognized, the Cantharelleae, the Lactarieae and the 
Agariceae, the last divided in two series as above indicated. 
The characters selected for the separation of genera are 
for the most part the old familiar ones, although others have 
been utilized when necessary to carry out the fundamental 
idea of making each generic name stand as far as possible 
for a compact, clearly related assemblage of species. The 
system adopted may therefore be considered as an amplifica- 
tion of that of Karsten or at least as being based on the same 
fundamental idea. Further study, and the discovery of the 
yet unknown multitudes of species which unquestionably exist 
in our territory, will result in a considerable increase in the 
number of genera to be recognized, and itis hoped that it may 
also lead to a better understanding of relationships and to a 
more natural grouping. The arrangement adopted in the fol- 
lowing key is largely a matter of convenience, and yet the 
attempt has been made, so far as is possible in a lineal ar- 
rangement, to bring related genera together. 
The family as here limited is taken in a narrower sense 
than has been done by Patouillard and others. This, too, is 
also done more as a matter of convenience than as expressing 
any fixed views as to natural limits. 
