THE TEMPERATURE OF LEAVES OF PINUS IN WINTER 5 1 
same conditions. Now since respiration is but feeble at 0° C. and 
decreases rapidly as the temperature falls [Maximow (16), Matthaei 
(14)], and since nearly all the readings were taken at 0° C. and below, 
and since the wound made in the leaf was very small, any increase in 
temperature due to wounding would not be appreciable. 
(4) Loss of Heat by Conduction Along the Lead Wires. — This was, 
in all probability, the source of greatest error and could not be entirely 
overcome. As already stated, two leaves were fastened together 
and the lead wires drawn through them at an angle in order to give 
as large a contact surface as possible. To determine what the loss 
from this source might be, the following experiment was made. A 
small glass tube 15 cm. long with a bore of 5 mm. was furnished with 
two side arms and two minute openings. Three thermo-j unctions, 
such as were used in taking the leaf temperatures, were inserted in 
this tube as shown in figure 3. Through this tube kerosene oil was 
siphoned from a bath kept at a temperature 
of 10° C. above the room temperature. The 
leads of junction i were in contact with the 
heated oil the entire length of the tube, and 
this junction should therefore record accu- 
rately the temperature of the flowing oil. 
The leads of junction 2, on the other hand,, 
were in contact with the oil for a distance of 
.5 mm. to I mm. for the advance lead and 
4 mm. for the copper — a condition similar to 
that of the junction in the leaf. By connect- 
ing junctions x and 2 with the potentiometer, 
any difference in temperature between these 
junctions due to loss of heat by conduction 
from junction 2 could be measured. By con- 
necting junctions I and 3 in the same way 
the temperature gradient between the flowing 
oil and the surrounding air could be deter- 
mined. A three-way switch enabled the oper- 
ator to measure both in rapid succession. Re- 
peated trials gave an average displacement on 
the bridge for junctions i and 3 of 84.5 cm. which at the room tem- 
perature was equivalent to a temperature difference of 9.7° C. Junc- 
tions I and 2 gave an average displacement of .36 cm. This repre- 
FiG. 3. Showing the 
details of the three thermo- 
j unctions. 
