Journal of Agricultural Research 
Vol. V, No. 8 
3H 
integrate different temperatures if there are differences, and as the air 
is very thoroughly stirred by the electric fan previously mentioned, it 
is probable that the resistance thermometer shows the average tempera¬ 
ture of the air of the chamber. In the interior view in Plate XXXV, 
figure 1, two of the frames are plainly shown with a wide-mesh wire 
screen before the resistance wire to protect it against contact with any 
object that would cause a short circuit between two parts of the wire, 
as well as against injury. 
The six coils are connected in series by well-insulated No. 16 copper 
wire, and similar wire leads from the terminals of the series, through a 
rubber stopper in a small opening in one wall of the chamber, to a special 
switch on the observer’s table, by which they may be connected in one 
arm of the Wheatstone bridge. The purpose of the switch is to provide 
means for using with these coils the same bridge that is used with other 
coils for measuring the temperature of the walls of the chamber and that 
of the body of the subject, as explained later in this paper. This switch 
must be designed to avoid the error that would result from introducing 
appreciable resistance of the switch contacts into the bridge circuits. 
The connections between the bridge and the resistance coils include a 
compensating lead to eliminate from the measurement of the resistance 
of the coils the effect of both the resistance of the leads and any change 
in their resistance due to change of temperature. The contact that is 
moved along the slide wire of the bridge, to restore balance when the 
resistance of the thermometer coils has changed, is in series with the 
battery, so that contact resistance introduces no error in the measure¬ 
ment. 
The six coils have a total resistance of about 20 ohms at 20° C. Since 
the resistance of nickel wire varies approximately 0.4 per cent per degree 
at the usual temperatures of the experiments, their total change in resist¬ 
ance would be close to 0.08 ohm for a change of 1 0 in the temperature of 
the air of the chamber. The resistance of the slide wire of the Wheatstone 
bridge will balance the bridge circuit for the change of resistance in the 
coils that would result from a change of 5 0 in the temperature. By means 
of several coils of manganin wire, which may be connected in series with 
the slide wire, the total range of the bridge may be extended, but under 
usual experimental conditions the temperature of the air is allowed to 
change as little as possible. Whether the change is large or small, it 
must be measured accurately. A change of resistance in the thermom¬ 
eter coils resulting from a change of o.oi° in the temperature of the air 
will upset the balance of the bridge sufficiently to cause a deflection of 
the sensitive reflecting D’Arsonval galvanometer that indicates when the 
bridge is balanced. The balancing point of contact may be moved along 
the wire a distance sufficiently small to restore the balance, and the scale 
of the slide wire will indicate the distance. 
