VALUE OF THE BRITISH ASSOCIATION UNIT OF RESISTANCE. 
227 
counted and registered, and as a check a counter was attached to the axle and read 
at the end of the windings ; the two records agreed in both cases. The resistance of 
the coils after winding was 84'9 and 82'7 B.A. units respectively. Four diameters 
were measured in every layer. 
“ The measurements of length were made by means of the cathetometer, and I find 
in my book a record of a comparison between it and the beam-compass scale, which 
had been tested'"" I believe ; there are also measurements of the thicknesses of the 
walls of the channel and its width.” 
The numbers actually used in calculating the value of M were furnished by 
Professor Chrystal from these measurements. The error between the cathetometer 
and beam-compass appeared in one or two measurements to amount to 1 in 2000, but 
in the majority of cases it was very much less, its mean value being, perhaps, 1 in 
10 ,000, irrespective of sign. On the whole then we may, without sensible error, treat 
the cathetometer scale as accurate. 
In another letter, Professor Chrystal gives the following extract from his note¬ 
book of the direct results of the observations. He says : “ You remember that each 
diameter is given as the mean of four. 
“The first diameter is through the slit, the next 45° from it in the direction of the 
sun’s motion, the letter on the coil being up, and so on. 
“ Here is the entry in my book for the fourth layer in coil A. 
No. of layer. 
No. of turns. 
Cathetometer. 
Cathetometer. 
Difference. 
Mean. 
4 
26 
781-40 
280-14 
501-26 
781-30 
279-95 
50135 
781-27 
279-58 
501-69 
780-60 
279-52 
501-08 
501-35 
“ You will observe in the above extract that the two intermediate diameters are 
greater. This happens in most layers. At any rate the diameter perpendicular to 
that through the slit is in the great majority of cases the greatest, as might be 
expected.” 
Professor Chrystal’s measurements then gave us as the mean of four observations 
m different positions the value of the external diameter of each winding, and also the 
total number of windings in each layer. In each coil there were 30 layers and in 
each layer about 26 windings. In coil A the total number of windings was 797 and 
in coil B it was 791. 
Let the external diameter of the layers be d 1} cl, 2 , c/ 3 , &c., and let the number of 
turns in a layer be 26 +%, 26+%, &c. 
* This beam-compass has again been tested by Mr. Dodds during the present year and found correct. 
All our measurements of length are referred to it. (Nov., 1882.) 
2 G 2 
