280 (JOPELANI). 
r 
them could possibly be redrawn so that the Philippine species, at least, 
would form natural gi'oups; but they would not be ^readily enough 
distingixishable to constitute convenient genera. I therefore prefer, with 
Milde, to regard all the species of both as Athyrla. 
For similar reasons it is impracticable to group the species with anas- 
tomosing veins in a genus by themselves (as An'isogonium) , for Aniso- 
gonium cordifolium and the other species with practically entire pinnge 
may, but more probably do not, have a common ancestry; while A. 
dccussaium and A. esculcntum certainly had two other distinct lines of 
free-veined ancestors. Therefore, a natural classification of Atliyrium, 
sensu latiore, will not maintain the old genera, with their usually 
accepted limits, as subgenera. 
In general, the stoiit, harsh species constitute a comj^aratively rec- 
ognizable group, which may still be called Diplazmm, as a minor group, 
since most of its members have borne that name, and since the type of 
the genus Diplazlum is in this group, if in either. However, it is my 
personal opinion that this group is a biologic rather than an altogether 
natural one, and that several of the groups of "I>ip?azm" have originated 
separately in Euathynum. A. umhrosum, A. silvaticum, A. sor so gone use, 
A. nigripcs, A. cyatheaefolium, and A. Whitfordi may well all have 
originated independently of each other in Euathyrium; and each has its 
own relatives or descendants in Diplazlum. Likewise, the more delicate 
species, including most of the plants commonly called AthyriUm, and the 
type of the genus, may be designated as Euathyria. Diplazlum is a 
tropical group, while it is to the north of the Philippines that Euathyrlutti 
reaches its greatest development in species. 
Although as natural as a large and primitive genus well can be, Atliy- 
rium is practically indefinable. The Asplenieae as a whole are naturally 
and conveniently distinguished from the Aspidieae by the elongate sori. 
These are found practically without exception in the other genera, and 
in most of the species of this one. But of late there have been included 
in Atliyrium, a number of species the sori of whieh seem to have been 
handed down unchanged from ancestors anterior to the separation of the 
two tribes, and other species are included which have apparently inherited 
the subelongate and unstable sori which must have preceded the fixing of 
the Athytiuiu type. The typical sorus of Euathynum is intermediate 
between that of Dryoptcrls and those of DlpJazium and Asphnium, 
Xature offers us no character and no group of characters by which we 
can certainly distinguish the primitive Athyrla from their near relatives 
in Dryoptcrls.' If we try to make the line between the genera a natural 
one, we must decide the critical cases on their individual merits, accord- 
ing to their probable affinity to recognized members of cither genus; and 
such characters as pubescence, texture, or margin, in general valueless as 
generic characters, may decide our judgment as to the affinity of single 
jy 
