42 
JOHN H. EHLERS 
B. Final Experiments 
1. Material and Location. — ^When the work was resumed in the 
autumn of 1913 the station in the arboretum was abandoned. An 
Austrian pine growing on the university campus was selected for the 
final experiments. This tree is about 40 years old and about 15 
meters high. It is protected from direct winds on the north and west 
by the library, and on the south by the physics laboratory. To 
give the operator access to the leaves and to hold the outdoor part of 
the apparatus, a wooden platform 4 meters high was constructed. 
Upon this platform a wooden screen was set up to protect the apparatus 
from the sun's rays. Three pairs of heavy insulated copper wires 
connected the platform with the ground floor of the physics laboratory 
where the measuring apparatus was kept. 
2. Method. — The first two months were spent in trying out various 
methods. A potentiometer method was finally adopted as best suited 
for the purpose. The principle of the method consists in opposing 
over the slide wire of a Wheatstone bridge the fall of potential due to 
a suitable current from a cell against the electromotive force produced 
by the difference in temperature between the two junctions, one of 
which is in the leaf. By moving the sliding contact, the portion of 
the bridge wire included in the galvanometer circuit {A to B, figure i) 
may be changed until the fall of potential over this portion is equal 
and opposite to the resultant E.M.F. due to difference in temperature 
of the junctions. This condition is indicated by a zero deflection of 
the galvanometer. The displacement along the wire per degree 
temperature difference is determined empirically by placing the 
junctions in baths of known temperature. 
The great advantage of this method over the first one employed 
is that, since it is a zero method, no current flows in the galvanometer 
circuit and variations in resistance due to temperature changes and 
other causes may be neglected. Furthermore, its accuracy is inde- 
pendent of any assumption of proportionality between galvanometer 
deflection and current. 
3. Apparatus. — ^The apparatus in its essential parts consisted of a 
slide wire Wheatstone bridge, galvanometer, storage cell, resistance 
boxes, rheostat, and voltmeter. Figure i shows the arrangement. 
The galvanometer used was of the d'Arsonval type with a resistance 
of 30.6 ohms, and was made by the Eberbach and Son Co., of Ann 
Arbor, Michigan. The Wheatstone bridge was made by the same 
