282 
AMERICAN JOURNAL OF BOTANY 
[Vol. io, 
lateral veins (as seen from the upper or ventral leaf surface) were recorded. 
Petiole diameter was measured at the mid-region of the petiole by a microm¬ 
eter caliper. Blade thickness was also measured by a micrometer caliper 
at two symmetrically situated points at right and left of the midrib in a 
region where there were no projecting veins and where mesophyll thickness 
alone could be determined. These two measurements were almost always 
identical, but when different were averaged. Blade area was determined by 
outlining the blade on standard-weight paper, cutting this out, and weighing 
the cut-outs. Blade volume was calculated by multiplying area by 
thickness. 
Constants were determined for the whole population and for each of the 
four groups into which it was divided: vertical upper and vertical lower— 
the upper and lower members of the vertically oriented leaf pairs; and 
horizontal right and horizontal left—the right and left members of the 
horizontally oriented pairs as seen by an observer facing the apex of the 
twig (fig. i). These constants are set forth in table I. 
An inspection of this table brings out certain facts as to the differences 
between these leaf groups in dimensions , shape , and variability. 
Differences in Dimensions 
A comparative study of the means shows that in all measurements 
(except blade thickness) the vertical upper leaves are markedly smaller 
than the lower ones, and that the members of the horizontal pairs are in all 
cases intermediate between these two. A significant fact which the figures 
bring out is that the combined size of the two members of the vertical 
pairs approximately equals the combined size of the two members of the 
horizontal pairs. In other words, the plant has a given amount of leaf 
material available at a given node (of course this amount decreases as we 
approach the apex of the twig), but this material is divided between the 
two leaves at this node in varying proportions, depending on the orientation 
of the particular leaf pair which arises there. 
Perhaps the most striking fact brought out is that the four leaf groups 
are of almost exactly the same blade thickness. Whatever are the factors 
which produce the differences between these groups in other dimensions, 
they evidently do not affect this one. This uniformity cannot be due to 
lack of delicacy in measurement, for decided differences in thickness were 
recorded between different leaves, the readings running from 0.12 to 
0.25 mm. Leaves on the same twig all tended to have the same thickness, 
0 but twigs differed among themselves, those from relatively shaded positions 
in the tree tending to be thinner than those which were more exposed to 
the sunlight. If light intensity determines blade thickness, as commonly 
supposed, we must infer that on a given twig the leaves in these four posi¬ 
tions receive approximately the same degree of illumination. 
