BOTANY: W. TRELEASE 
33 
more than 6 ranks being very exceptional except in some tropical species 
with leaves venulose above and dull beneath, and in some of our northern 
forms. 
One of the characters most available and significant in the classifi- 
cation of the species of Phoradendron is a fundamental difference in 
their leaves. By far the larger number of species have unmistakable 
leaves, but our western group to which P. calif ornicum and P. juniperi- 
num belong have their foliage reduced to short thin scales which resemble 
the leaves of the related genus Arceuthohium or Razoumofskya so closely 
that species of either genus are commonly to be found in herbaria as 
representative of the other. Unlike typical foliage leaves, these scales 
do not disarticulate, though a constriction at the base of the scales in 
two forms affords partial ground for their specific recognition: one 
species of the Mexican mountains, P. minutifolium, has almost equally 
small though fleshy disarticulating leaves, and two of the South Ameri- 
can species, P. tunaeforme and P. fragile, are characterized by bearing 
small scale-like leaves only. Such species are very likely to be mis- 
taken for some representatives of the related genus Dendrophthora, 
which differs technically in its 1 -celled anthers. 
If any species of the United States, for example P. Eatoni of the ever- 
glades of Florida, is compared with any West Indian or South American 
species, for example P. rubrum of the Bahamas, the latter will be found 
to possess constantly in addition to its foliage one or more pairs of scale- 
leaves at least on the lowermost foliage internode of every branch. 
Comparable with the scales of the flowering spikes and with the stem- 
scales of P. juniperinum, etc., these cataphyls afford by their presence 
or absence what proves to be one of the most important of characters 
for the primary division of the genus Phoradendron. Usually cataphyls 
do not subtend flowers or spikes, apparently serving no function further 
than the protection they may afford the shoot in its earliest development; 
but in P. crassifolium and P. craspedophyllum spikes are regularly and 
characteristically found in the axils of some of the cataphyls, and less 
characteristically in a few other cases. 
Never found on any species of the United States, absent from three- 
fourths of those of Mexico and Central America, but invariably present 
on all of the South American and West Indian species, these scales are 
usually confined in the latter to the basal joint of each branch, though 
in cases of true or cymose forking they are found on all joints — since 
only basal joints are then present. In a very small percentage, only, 
of the tropical species with percurrent or monopodial branching, e.g., 
P. flavens and P. crassifolium and their allies, cataphyls are found on all 
