622 
Harper . — Defoliation : its Effects upon the 
the consumption of a considerable quantity of material has been shown 
clearly by the fact that growing trees actually lose in dry weight while 
their leaves are expanding from the winter buds, especially in the case of 
deciduous trees like the larch. 1 The unfolding of the leaves, therefore, 
represents the expenditure of capital, and if they are lost early in the 
season, before an adequate interest may have accumulated, the resources of 
the plant are diminished. The primary physiological effect of an early 
defoliation is a more or less severe degree of starvation. 
Accordingly, most of the facts recorded in this paper are purely 
starvation phenomena. Slow starvation of a growing tree is met by 
economy in expending the formative materials, and in the woody cylinder 
this may be expressed in two distinct ways : firstly, by a reduction in the 
amount of growth ; secondly, by a decrease in the proportion of ‘ mechanical 
tissues ’ ; for thick-walled fibrous cells require in their formation more 
material than an equivalent development of thin-walled, water-conducting 
tracheides, and are of less immediate importance to the tree. The amount 
of growth is indicated by the breadth of the annual ring, and in timber so 
simple in structure as that of the larch, the thick-walled tissue is, generally 
speaking, identical with the zone of so-called autumn wood. 2 Therefore 
both the ring-breadth and the development of autumn wood must be con- 
sidered in dealing with the effects of defoliation. 
The trees to be investigated were felled in October and November, 
1 91 1, after growth for that year had ceased. Complete cross-sections of 
the trunk were sawn out at equal distances apart, every four or five feet as 
the case might be. (In the tables these sections are numbered successively 
from the base of the tree upwards.) From each section a small rectangular 
block from 8 to 12 mm. broad was cut out at the circumference opposite to 
each of the four points of the compass, N., S., E., and W., the north side of 
the tree having been marked before felling and shown by an arrow on the 
surface of each section as it was cut out of the trunk. From the surface of 
each block was cut a thin, transparent shaving for microscopic examination 
and measurement, and the breadth of each annual ring as a whole and of its 
zone of autumn wood taken separately was measured along three different 
radii separated tangentially by a mean distance of 2-5 mm. Thus three 
sets of measurements were made for each little block, making twelve in all 
for each complete cross-section of the trunk, and the average of the twelve 
is taken as the mean radial increment for the tree at any particular height 
from the ground at which a cross-section may have been made. 
1 Ramann and Bauer, p. 67. 
2 It need scarcely be remarked that this so-called 1 autumn wood ’ is not formed in the autumn, 
and on the other hand the ‘ spring wood 5 is being made during the first half of the summer. There 
is therefore little to be gained by changing the term ‘ Herbstholz ’ into ‘ Sommerholz’, as has been 
done by Hartig, and since a translation of Strasburger’s still more suggestive expression 1 Spatholz ’ is 
not yet generally used in this country, the terms ‘ spring wood ’ and ‘ autumn wood ’ are employed 
throughout this paper (cp. Hartig (1), p. 13, and (3) p. 276; Strasburger, p. 501). 
