21 
rainy and the dry season. The lower boundary is approximately the 
back line of the coastal plain. It is marked in most places by large- 
leaved epiphytic aroids or scandent ferns and sometimes by luxuriant 
Calamus. The high forest gets its name from the many high trees, but 
in the size of these there is the utmost diversity. Of Whitford’s for- 
mations 10 I construe it as including the Anisoptera-Strombosia, the 
Dipterocarpus-Shorea, and the Shorea-Plectronia. I have not adopted 
these names because so fine a stratification does not appear justified by 
my material, and because I do not know how far they, being descriptive, 
are applicable — that is, whether or not the trees the names of which they 
bear, are dominant, or representative, or represented, here. On the 
other hand, the classification and nomenclature which I am using 
are generally applicable though not exclusive . 11 
There are, indeed, a number of societies, some very distinctly localized 
and locally characteristic of different altitudes, but the narrower strata 
as entities are without other than the weakest structural or floristic 
characters. 
From the bionomic standpoint, nearly all the ferns of the high forest 
fall readily into one or another of three not very unequal divisions, rea- 
sonably definable by the environment, by the plants’ structures and 
systematically. These are terrestrial plants of humid places, terrestrial 
plants of dry places, and epiphytes. In each of these divisions in the 
high forest many ferns have wide distribution, but no fern of one division 
is common in another and the dubious ground between the moisture- 
loving and the drought-enduring terrestrial species has but a very limited 
representation. 
The hygrophilous geophytes grow along the river and its branches, 
in narrow gulches usually without running water, and near the upper 
boundary of the high forest, in the sheltered cirque-like basins which 
are the characteristic abode of tree-ferns and of Diplazium Williamsi. 
Diplazium tenerum occurs along the river, never high enough to escape 
floods, and Nephrodium Foxii is exceedingly common in the same places. 
N. pliilip pinense, A-spidiiim irriguum, Leptochilus inconstans and Polypo- 
dium rivulare have the same restricted range along creeks in the 
mountains near Manila, as Odontosoria does throughout the Islands. 
Except the Aspidium (and, as compared with most Aspidia, it is so) 
these are all small, or finely dissected, or both. The Aspidium is 
usually found with its fronds torn and broken by freshets. N. ferox 
and Odontosoria retusa are strictly confined to stream banks, but high 
enough up and with long enough stipes to escape floods. Leptochilus 
10 This Journal (1906), 1, 384. 
11 Beside the formations represented at San Ramon, we have in the Philip- 
pines, prairies in northern Luzon, and a montane brush above the mossy forest 
on Apo. The savanna-wood can he construed broadly to include cogonal, parang, 
and pine lands. 
