699 
the Respiratory Exchanges of Leaves. 
a sample, it was necessary to ensure that the changes of composition to be 
measured should be relatively large. For this reason several leaves were 
enclosed together in a chamber of minimal volume specially constructed for 
the purpose. This chamber (shown in plan and sectional elevation in 
Fig. i) was designed in collaboration with Dr. F. F. Blackman and made 
by the Cambridge Scientific Instrument Company. It consisted of two 
parts, a heavy circular base of brass (a), and the flat chamber itself (b), 
carried by a brass disc (c) which was accurately ground into the base, and 
so when greased made an air-tight junc- 
tion with it. There was a small tubu- 
lure (d) at the top of the chamber to 
which a manometer could be attached, 
and another (e) at the side of the base, 
leading into the small space between the 
disc and the base. This latter tubulure 
could be connected to an apparatus for 
withdrawing samples of the enclosed air, 
which was similar to that used by 
Aubert. 1 This chamber had the special 
advantage that it could readily be opened 
to renew the air or examine the condition 
of the leaves enclosed in it, and as readily 
closed again perfectly air-tight, working, 
when greased, like a well-ground stopper. 
No attempt was made to investigate 
respiratory exchanges during exposure 
to chloroform vapour, as its presence in 
the air would have introduced errors 
into the analyses, and there appears to 
be no satisfactory method of removing it 
before analysis. Leaves were therefore 
chloroformed in a separate vessel, with 
a capacity of about a litre, by adding a 
measured volume of liquid chloroform on 
a piece of cotton-wool. The duration of the exposure varied from 2 to 
25 minutes in different experiments, and the dose from 0-05 c.c. to i-o c.c. 
(0-075 to 1 ’5 g rm -) of liquid chloroform to the litre of air. 
The procedure here outlined has not proved in all cases sufficiently 
sensitive for the satisfactory investigation of changes in the respiratory 
composition which has occurred during a given interval. Improvements have been introduced into 
the technique by which the range of difference between analyses of the same sample as well as the 
absolute error are now very much reduced. See loc. cit. 
1 Rev. gen. de Bot., iv, 1892. 
Fig. 1. Plan and sectional elevation 
of leaf-chamber. 
