1884. 41 Trans. N. Y. Ac. Sci. 
of the atmosphere is a nearly constant quantity. Its absorption through 
the stomata is certainly not rapid enough to seriously diminish this 
quantity ; for the molecules taken up are immediately replaced by 
others from the surrounding atmosphere, maintaining a practically 
constant tension of the gas, without any action on the part of the 
plant. -Chemical analyses of the atmosphere from widely separated 
localities, and under differing conditions, show that the amount varies 
only very slightly from one part of COs: in twenty-five hundred parts 
of air, measured by volume. And the total amount taken by plants 
from the air during any period is small when compared with the total 
quantity present—this loss being constantly replaced in various ways. 
For these reasons it appears that Mr. Allen’s theory is hardly in ac- 
cordance with known facts, and must be abandoned if something more 
rational can be suggested. 
It has seemed to me that the chief agency of the production of mar- 
ginal indentations is to be looked for from a mechanical, rather than a 
chemical or vital cause, and this combined with more or less well- 
defined types of venation, no matter how originated, and also with 
relative amounts of sap nutrition and consequent relative vigor of dif- 
ferent leaves on the same plant, or of those borne by young plants or 
shoots as compared with those of older individuals or branches. 
That there is a general plan in the vein-arrangement, is apparent 
from the fact that in genera where some species bear nearly entire- 
margined, and others serrate, cut, or even compound leaves, or where 
this circumstance obtains in related genera,* the venation is generally 
similar ; and this is true even of some orders.+ We know also that the 
vein system has existed as long as the plants have ; for in the Sassafras, 
Liriodendron, etc., of the Lower Cretaceous, it is the same as in living 
species of these genera, or similar, and is recognized by students of 
palzo-botany as a valuable means of determining genera. The origin 
of these various plans of venation, the ‘‘ architectural ground-plans ”’ 
of leaf-structures, I do not propose to discuss. 
The amount of nutriment furnished the leaves by the roots is an 
important element in the discussion. This is relatively greatest at 
the time of vernation, and as other leaves appear on a growing branch 
or stem, above the earlier ones, part of the sap is diverted into these 
newer lamine at the expense of those below; and this consideration 
will explain the fact that on land-plants bearing heterophyllous leaf- 
blades, the lower leaves are generally the most divided, although, from 
their earlier supply of sap, the largest. In germination the earliest 
leaves are mostly entire, from a similar cause. The nearer approach 
* In Negundo, the leaflets placed together form a perfect maple leaf. 
+ The peculiarly characteristic venation of Me/astomacee. 
