82 BOTANICAL GAZETTE [FEBRUARY 
observations made by the horn hygroscope, as given in the above- 
quoted paper. 
METHOD. 
The apparatus employed was one of Callendar’s recorders, a 
platinum thermometer,’ or resistance thermometer, in which the 
difference between the temperatures of two fine platinum wires is 
recorded on a revolving drum.3 In the experiments illustrated 
by figs. I, 2, 3, 5, 6, 12, 13, 14, 15, the resistances were so 
arranged that a vertical fall of 24™™ in the tracing made rep- 
resents a difference of 1° C. But as some of the figures have 
been reduced in scale it is simpler to state that the distance 
between the horizontal lines represents one-sixth of a degree C. 
In figs. 4, 7, 8, 9, 10, rz, the value of 1° C. is uncertain, owing 
to an omission in my notes. In these figures the distance 
between the horizontals represents either 0.1° or 0.2° C. 
The fine platinum wires, the difference of whose temperature 
is recorded, are disposed in fine zigzags on plates of talc 10x 3" 
which will be referred to as “‘ bulbs.” The leaf is inverted, so that 
its stomatal surface is free, and its astomatal surface is in con- 
tact with the wire. It is generally necessary to hold the petiole 
in a clamp, and to press the leaf against the bulb by a pair of 
springs, or bya thread having weights at each end. An unavoid- 
able drawback of the apparatus is that, owing to the current run- 
? The instrument was made by the Scientific Instrument Company at Cambridge. 
I am much indebted to my brother, Horace Darwin, for help in management of the 
apparatus. 
3A resistance thermometer, or, as it is often soar a platinum thermometer, con- 
sists of a fine platinum wire, which is exposed to the t temperature it is desired to 
measure. As its temperature alters, its eee? resistance also aiters, and by measur- 
ing resistance the temperature is fou thermometers are usually made of 
platinum, but this is not necessarily he eh ; a fine wire of copper or of other metal 
can be ae in adady cases with equally good results. The resistance is usually 
measured by a Wheatstone bridge; the Callendar recorder is an automatic Wheat- 
stone bridge, which is continually measuring the resistance, and a 
moves a pen so that its distance from a fixed point is proportional to He temperature. 
This pen marks a line on a drum driven by clock-work, and thus 
ement wh 
Electric Recorders, a saisighine cbibal | fas by the Cambridge University Press. 
