of an Electric Current by Magnetic Action. 119 
One of the most troublesome operations in these experiments 
is that of determining the thickness of the metal strips exam- 
ined. Owing to the inevitable slight roughness of the surface, © 
direct measurement with calipers is likely to give too great a 
thickness. On the other hand the density of many specimens 
is subject to considerable uncertainty and, therefore, the indi- 
rect method by means of weight and density, which is the 
method used, cannot be applied with full confidence. More- 
over the latter method gives, at best, only the average thick- 
ness of the strip, whereas I have heretofore assumed, and in 
this article still assume, that the effective thickness of the strip 
18 the thickness of that part which lies between the two arms 
4,4. My practice of late has been, therefore, to determine the 
average thickness by the weight and density method, making 
use of the best data available, and to estimate the thickness be- 
tween the arms by adding to, or subtracting from the average 
thickness, according as the calipers indicated the thickness at 
that place to be greater or less than the average. In case of sev- 
eral of the strips to be mentioned hereafter this correction was 
Correr, Zinc AND THEIR ALLOYS. 
_. More than three years ago (B. A. Report,1881), I found that 
if the rotational power of copper is called —, that of zinc 18 
re t the same time a specimen of brass, exact composi- 
Hon unknown, had been found to lie between copper and zinc 
in this respect, but nearer the copper, having in fact a small 
~ Totating power. 
S Through the kindness of Prof. Trowbridge and Mr. E. K. 
hla I had at command in the summer of 1883 sp te 
. OY of copper and zinc in widely varying proportions. Spec- 
— of these alloys had been analyzed chemically by Mr. 
tevens, but as they-had lain-for some time in an exposed posi- 
ton after he had-finished his work upon them, I feared a 
