CHRISTENSEN: ON THE GENUS DRYOPTERIS 35 
variable. By examining a series of tropical Dryopterids, 
we soon find that such a character as the presence or 
absence of indusia is of no generic value, and even difficult 
to use as a specific character. A very large number of 
species, which have been referred to Phegopteris, are 
really indusiate, but the indusia are to be seen only in the 
very young sori. Moreover, we find very often that two 
forms, which can scarcely be distinguished by any other 
character, differ from each other only by the one being 
_indusiate, the other exindusiate, but that, notwithstand- os 
: ing this difference, the two forms are closely 
. on ihe venation, if ee (i. e., forming a few 
Gener: 
e) or not. founded on such characters 
arg Phegopteris, ee are, therefore, not groups of 
related species, as natural genera should be, but may 
include a number of species that are often of a very re- | 
mote affinity, e. g., D. tetragona and D. phegopteris, both ee 
sometimes referred to Phegopteris; on the other hand, _ 
species referred partly to one, partly to another genus, ou 
are intimately related, e. g., D. patens, often referred to - 
Lastrea, and D. mollis se nenree _— is aN ph- 
