1914] _ GATES—XEROPHILY 483 
palustris) and in the deciduous woody plants (Aronia melanocarpa, 
Nemopanthes mucronata, Salix pedicellaris, Cephalanthus occidentalis, 
Spiraea salicifolia, Gaylussacia baccata, Vaccinium corymbosum, 
and Larix laricina) is much higher than in the evergreen shrubs and 
trees. The more xeromorphic the structure of the leaves the lower 
is the transpiration, and the more exposed to winter conditions the 
more xerophytic is the structure. This argues well for the winter 
conditions as being the fundamental causes necessitating xerophily 
in the evergreen ericads. The difficulty is in not being able to 
obtain water sufficiently fast or in sufficient quantities to supply 
the demands of transpiration. This is very much accentuated in 
the winter, mainly on account of low temperatures. 
Although this xerophytic structure also reduces the water 
demands of these plants during the summer, there does not appear 
to be in summer any need of so thorough a xerophily, for neither 
the very extreme droughts and hot spells of the summers of 1911 
and 1912, nor the extreme evaporating conditions of the laboratory 
appreciably injured the many plants of Chamaedaphne observed or 
experimented upon. On the other hand, the continued severe dry 
cold during the winter of 1911-1912 killed thousands of plants of 
Chamaedaphne down to the snow line, below which they were 
efficiently protected, whereas during the average winter of 1910- 
Ig11, with but very little snow on the ground during the coldest 
weather (barely below —17°C.), all of the bushes of Chamae- 
daphne remained alive clear to the top. 
As these ericads are of distinctly northern phytogeographic 
affinities (HARSHBERGER 21, 22, and TRANSEAU 48), and conse- 
quently are less subject. to the extremes of summer conditions 
which obtain near Ann Arbor, almost at the southern limit of their 
range, and are more subject to severe winter conditions in nature 
farther north, and as the evergreen habit is a hereditary character, 
the xerophytism is real and has been brought about and is necessi- 
tated primarily by the winter conditions which these plants endure. 
Summary 
1. The determination of the rate of transpiration per unit area 
of leaf surface by weighing is a satisfactory approach to a knowledge 
