TRANSACTIONS OF SECTION B. 671 
accurate balance being struck with the amount of carbon found in the plant at the 
end of the experiment, 
Up to within a few years ago we had no means of even approximately deter- 
mining the actual rate at which the assimilatory process goes on in a plant other 
than that afforded by its increase in weight in a given time. Such experiments, 
necessarily extending over weeks or months, can, at the best, only give us certain 
average results, and consequently afford no measure of the activity of assimilation 
under fixed conditions of insolation. In the year 1884 Sachs, who had for some 
time been at work on the formation of starch in leaves under the action of sunlight, 
found that the accumulation of freshly assimilated material in a leaf may, under 
favourable conditions, go on.so rapidly as to give rise to a very appreciable increase 
of weight in the leaf lamina within the short space of a few hours. By obsery- 
ing at different times of the day the varying dry weight of equal areas of large 
leaves, Sachs obtained an approximate measure of the rate of the assimilatory 
process which he could express in terms of actual number of grams of substance 
assimilated by a unit area of leaf in unit of time. In this manner he was able to 
show, for instance, that a sunflower leaf, whilst still attached to the plant, increases in 
weight when exposed to bright sunshine at the hourly rate of about one gram per 
square metre of leaf area. In the case of similar leaves detached from the plant, 
and of course under conditions in which the products of assimilation were entirely 
accumulated in the leaf, he found an increase in weight of rather more than 
1} gram per square metre per hour. 
I was able to confirm this work of Sachs in the course of an investigation on 
the Chemistry of Leaves which I made with Dr. G. H. Morris in 1892-93, and 
there can be no doubt that the variations in the weight of leaves can be used as a 
fair index of the activity of a leaf in assimilating, but it is not a method which 
admits of much refinement of accuracy, owing, amongst other things, to the want 
of perfect symmetry in the leaves as regards thickness and density of the lamin 
and to the possible migration of the assimilated material into the larger ribs, 
which of course cannot be included in the weighings, 
It is evident that a far better plan of measuring the rate of assimilation 
under varying conditions would ke the estimation of the actual amount of carbon 
dioxide entermg a given area of the leaf ina certain time, and it was to the 
perfection of a method of this kind that Mr. Escombe and I first turned our 
attention. 
In all previous attempts to measure the rate of ingress of carbon dioxide, such 
as those of Corenwinder, and more recently still of Mr. F. F. Blackman, it has been 
necessary to use air containing comparatively large quantities of carbon dioxide, 
amounting to 4 per cent. and upwards. Interesting and useful as such experiments 
undoubtedly are from the point of view from which they were undertaken, we 
must not lose sight of the fact that such conditions are highly artificial, and very 
far removed from those under which a plant finds itself in the natural state, where 
its leaves are bathed with air containing not 4 or 5 per cent. but only -03 per cent. 
of carbon dioxide. I shall have oceasion later on to show how remarkably the rate 
of intake of carbon dioxide into a plant is influenced by extremely small variations 
in the tension of that gas, and that on this account no deduction can be drawn as 
to the rate of assimilation under natural conditions from any experiments in which 
the air contains even so small an amount of carbon dioxide as 1 per cent. 
Before proceeding further in this direction, however, it will be well to consider 
the amount of carbon dioxide which must enter a leaf in a given time in order to 
produce an influence on its weight comparable with that indicated by the Sachs 
method of weighing definite areas. For this purpose I will consider a leaf with 
which we have made many experiments—that of Catalpa bignonioides. It is a 
very symmetrical leaf and a good assimilator, and since the intake of carbon 
dioxide takes place only on the under side, the question to which I wish to draw 
your attention can be stated in a simple manner. When such a leef is subjected 
to a modified form of the half-leaf weighing method of Sachs, into the details of 
whici I cannot here enter, it may, under favourable conditions, show an increase in 
dry weight equal to about one gram per square metre per hour. Since this 
