HEAT FROM SMALL CYLINDERS IN A STREAM OF FLUID. 
385 
chronograph record. At various lengths along this arm could be clamped a light fork 
designed to hold the specimens of wire under test. The latter formed part of a Kelvin 
double bridge, electrical connection being obtained through a central mercury- 
connecting switch and overhead wires to the remainder of the bridge. By means of 
a rheostat it was possible at each speed to adjust a measured current through the wire 
so as to bring its resistance to a value corresponding to a predetermined temperature. 
In this way it was possible in the case of each wire to vary in any chosen way the 
various factors of temperature, air-velocity and heat-loss. The general arrangement 
of apparatus and details are drawn in Diagram II. and shown photographically in 
Plate 8. 
(ii.) Details of Resistance Bridge. 
Connecting A and 
B 
In order to measure the temperature of a length of platinum wire heated by an 
electric current under given conditions of wind-velocity, it is necessary to design a 
form of resistance-bridge suitable for 
an accurate determination under these 
conditions. The well-known connections 
of the Kelvin double bridge at once 
recommend themselves^ for the purpose 
and are shown diagrammatically in fig. 2. 
A represents the platinum wire whose 
resistance it is required to measure. B is 
a ten-metre bridge wire of No. 23 S.W.G. 
manganm wire. 
is a resistance S including that of the 
ammeter and leads to the fork on the 
rotating arm. The ratio-resistances a, 
b, a, (3 were in the neighbourhood of 
10 5 ohms, so that the current in the wire 
A is to a very close approximation that indicated by the ammeter. A and B were 
connected through an adjustable rheostat to the 110-volt mains. The following 
formulae are given for future reference. When the galvanometer is balanced it is 
not difficult to prove that 
S 
a 
a 
A = a § 
B b B a +(3 + S\b (3. 
(34) 
Also if I be the current in the platinum wire, k that in the ammeter, i that through 
the coils (a, b), and j that through (a, (3), we have 
I = k[ l + <V(a + /3)], j — k . S/(a + (3), i = I [A/a + a/a. S/(a + /3 + S)]. (35) 
We notice from (34) that when the ratio coils are so adjusted that a lb — aff3, then, 
3 D 
VOL. CCXIV.-A. 
