610 PROFESSOR CHRYSTAL ON THE DIFFERENTIAL TELEPHONE, 
There are two leading null methods in use for measuring resistances, with a 
ealvanometer as indicator,—the differential galvanometer method and 
Wheatstone’s bridge. I discuss the two corresponding methods for the 
telephone ; but it is the former more especially to which I wish to direct 
attention. 
The instrument which takes the place of the differential galvanometer, I 
call a differential telephone. It is simply an ordinary telephone wound double 
like a differential galvanometer. It is essential that this instrument shall stand 
the test of producing no sound when the same current is passed in opposite 
directions through its two coils. My first attempt to make such an instrument 
by winding the wires side by side in the ordinary way failed entirely. It was 
hopeless to attempt to compensate one coil, as is done in the case of a 
differential galvanometer ; for it is obvious from the nature of the telephone 
that there must be null magnetic force all over the core, and throughout the 
ferrotype plate. I found, however, that the desired arrangement could be got 
by twisting up the wires of the telephone together, and then winding the twist 
upon the bobbin. The instruments with which I have worked are constructed 
in this way, and appear by every test I have applied to them to answer their 
purpose perfectly. At all events, any defect, if discovered, will certainly be so 
small as not to affect their use in practice. As an illustration of the complete- 
ness of the compensation and of the delicacy of the instrument, I may mention 
that when the current of three small Le Clanché cells, interrupted by a tuning 
tork, passes through the coils in opposite directions, nothing whatever can be 
heard, although one turn of the line wire outside the instrument restores the 
sound very distinctly, and the current through one coil produces a hum that 
can be heard a long way off. 
One of the simplest applications of the differential telephone is to the 
comparison of coefficients of induction. In order to test this in practice, I 
fitted up the following rough arrangement which I mean to replace by a better, 
after I have settled by experience 
the most convenient disposition of 
the parts. af, yd are two pairs 
of coils mounted on two stems, so 
that they can be placed at any 
distances apart up to 30 cm. or so. 
AB, CD are the terminals of the 
two coils of the differential tele- 
phone, EK is a multiple are in 
one branch of which are Ba, and 
AB, in the other dy and CD; 
the arrangement is such that the current passes in opposite directions round 
Fig. 1. 
