2 
work, for instance, was Brenner’s “Klima nnd Blatt bei der G-attnng 
Quercus.” 2 
When geographical can be added to genetic unity, then the influence 
of local differences in environment, in the selection of adaptive pecul- 
iarities and the evolution of species should be especially clear and 
instructive. Such a combination of unities is usually impossible in a 
work of any scope because of lack of material. However, the homoge- 
neous family of Polypodiacece has presented to me, within the radius of 
an easy day’s walk in the wilderness back of San Ramon, Mindanao, 
material sufficiently ample to allow me to draw a variety of conclusions. 
While directly engaged in collecting on one visit, and on excursions at 
irregular times as official work on the physiology of the coconut per- 
mitted on a second one, I collected in this neighborhood 186 species in 
this family. I do not know that any other family anywhere in the 
world will yield such an amount of material; certainly no other will 
yield material so diverse in form and adapted to such varying local 
conditions. 
In this family are xerophytic, trophophytic, mesophytic and almost 
hydrophytic plants. It includes terrestrial and epiphytic herbs, vines, 
and subarboreous species; with a remarkable number of peculiar adap- 
tations to special conditions, such as the humus-cups of Thayeria and the 
ant-chambers of Lecanopteris. 
Such remarkable structures as the two just mentioned have ever been 
favorite subjects of study; but of greater real importance than the study 
of these curiosities, interesting because of their rareness, is the causal 
interpretation of more com m on phenomena, such as length of stipe; 
its articulation; the presence or absence of indusia; the permanence of 
root hairs; and ciliate or serrate margins. For the explanation of each 
of these characteristics this family, as developed at San Ramon, furnishes 
concurrent evidence in two or more genera. I believe that the applica- 
tion of rich material to such everyday, but too often ignored problems as 
these will be regarded as a real service. The range of forms and habitats , 
could have been widened by including some of the neighboring orders, 
taking in the aquatic Ceratopteris, the climbing leaves of Lyg odium, the 
arboreous Oyatheacece , and the peculiarly xerophytic, sometimes rootless 
Hymenophyllacece , but the sacrifice of genetic homogeneity is too great 
a price to repay me for extending to this wider -field. 
The subjects in this paper are taken up in the following order: 
I. The origin and geographical affinity of the San Ramon fern flora. 
II. Local physiography and classification by environment. 
III. Adaptations to common environment and to special conditions. 
IY. Systematic application of the results. 
-Flora ( 1902 ), 90 , 114 - 160 . 
