[ 223 ] 
IY. Experiments on the Value of the British Association Unit of Resistance. 
Part K I. by 11. T. Glaze brook, M.A., Fellow and Assistant Lecturer of Trinity 
College, Cambridge, Demonstrator of Physics in the Cavendish Laboratory, and 
J. M. Dodds, B.A., Fello w of St. Peter s College. 
Part II. by II. T. Glazebrook and E. B. Sargant, M.A., Trinity College. 
Communicated by Lord Hayleigh, M.A., F.R.S, 
Received May 24,—Read June 15, 1882. 
Part I. 
Before leaving Cambridge, Professor Chrystal, in the summer of 1878, wound with 
great care two large coils of about 50 centims. in diameter to be used for a redeter¬ 
mination of the value of the British Association unit of resistance. A galvanometer 
of special construction with double coils, one of thick wire the other of thin, was also 
wound by him and mounted with Weber’s suspension. 
The coils were to be placed at a known distance apart, so that the coefficient of 
mutual induction could be calculated. 
The apparatus remained unused till the spring of last year, when the experiments 
described in the present paper were commenced. 
The method used is similar to those employed by Kirchhoff (Pogg. Ann., lxxvi.) 
and Howland (American Journal of Science and Arts, vol. xv., 1878). The coefficient 
of mutual induction of the two coils is determined by calculation from the geometrical 
data. A current is passed through one coil, the other being in circuit with a ballistic 
galvanometer, and the induced current, produced when the primary current is broken 
or reversed, is measured by the throw of the galvanometer needle. The primary 
current itself being then measured by some method, we have enough data to deter¬ 
mine in absolute measure the resistance of the secondary circuit and the galvanometer. 
But this resistance can be measured in terms of the B.A. unit, and hence a value 
obtained for the latter. 
Professor Howland measured his primary current by means of a second galvano¬ 
meter, the constant of which he compared with that of the ballistic galvanometer both 
by direct experiment and by calculation. 
In our experiments the value of the primary current was obtained by passing by 
means of a shunt about ubli part of it through the ballistic galvanometer ; we thus 
eliminate from our equations the values of the galvanometer constants, as well as the 
