196 PROCEEDINGS OF SECTION A. 
ing currents were measured on a Siemens dynamometer (No. 
2686), which had previously been checked with a Kelvin electric 
balance (No. 528), the error being less than 4 of 1 per cent. 
The ballistic galvanometer was calibrated, both from an earth 
inductor and a current inductor before and after each test. 
The conclusions arrived at are :— 
(a) That the cast steel made by the Heskett process is fairly 
suitable for the work, although it is not so good as imported 
steel. The elbow of the curve being rather low. 
(6) That the crucible cast steel is inferior to the Heskett steel. 
For the sake of comparison a welded ring of common annealed 
wrought-ir-n was tested at the same time, and the results are 
included in the diagram. 
7. — ELECTROMAGNETIC REFLECTION AND 
REFRACTION. 
By Proressor A. MacAutnay, M.A. 
8.—NOTE ON SPECIFIC INDUCTIVE CAPACITY OF 
SELENIUM. 
By T. P. V. MaApseEn. 
[ Abstract. ] 
Tue method adopted was that used by Prof. Threlfall and Mr. 
Brearley in their work on the Specific Inductive Capacity of 
Sulphur, described in the “ Philosophical Transactions,” Vol. 187 
(1896). The selenium was first prepared by Prof. Threlfall, and 
cast into a plate by Mr. Brearley. After the first set of ex- 
periments, the cake was broken up and recast. On _ being 
heated, the selenium first became sticky, then suddenly solidi- 
fied, and finally melted. After another series of experiments, 
the cake was again broken up, and found very free from air 
bubbles, and remelted. The results for the first plate gave an 
average of 5.61; for the second, 5.78; and the third (one ex- 
periment only), 5.83. 
9.—-A GRAVITY BALANCE. 
By Proressors THRELFALL AND POLLOCK. 
