The Philippine journal of science. [Vol. 64, no. 1]

64, 1-2

Copeland: Hymenophyllum

5

There is, however, a considerable group of species, ascribed
to Hymenophyllum, characterized both by elongate receptacles
and by involucres cleft part-way down but not to the baseoften not nearly so. This group includes Meringium and Amphipterum of Presl, regarded by him as intermediate genera.
That expression-intermediate-in Presl's time might express
merely a combination of diagnostic characters. Today, if carefully used, it appraises this combination of characters as evidence of affinity. If the group seemed to be primitive in other
respects, the features just mentioned would suffice to fix it quite
positively as really related to Trichomanes, and therefore really
intermediate. In fact, though, instead of being apparently
primitive in other respects, it is the one group of species ascribed
to Hymenophyllum that displays the greatest structural specialization. Because this is so, I am constrained to believe that the
features of resemblance to Trichomanes are not due to mutual
phylogenesis. While I postulate H. polyanthos as primitive in
a monophyletic genus Hymenophyllum, I cannot also believe that
a group with serrate margins and crested involucres-conspicuous features nowhere suggested by any sure species of Trichomanes-is more related to Trichomanes than are the many
species with entire margins and crestless involucres.
The foregoing propositions, which are not true, or are hard to
accept, all arise from the basic assumption that Hymenophyllum
is a natural genus, a monophyletic group, homogeneous in comparison with Trichomanes, containing some primitive element
ancestral to the rest of the genus. They have largely disappeared as problems, because I have come to the positive conclusion that Hymenophyllum is not a group of this kind. If
there is a distinction in this respect, Hymenophyllum is even
less homogeneous than Trichomanes. But the task of recognizing and identifying the natural groups collected in Hymenophyllum has proven incomparably more difficult than in
Trichomanes.
As far as my work is concerned, this might well be because
the groups in Trichomanes were largely recognized and defined
by my predecessors; but this course of history may fairly be
ascribed to the fact that the groups in Trichomanes are more
easily recognized, and have therefore invited study. Mettenius,
Prantl, and Giesenhagen, all studying intensively the anatomy
of these plants, as a foundation for taxonomic work at which
they never arrived, devoted much attention to species of Tricho