Page:Collected Physical Papers.djvu/64

44 began occasionally to be affected, though rather feebly. But when this was reduced to 13 mm. there was no uncertainty; a measurable, though small, portion of the radiation was now found to be always transmitted.

I now increased the angle of incidence to 45°, and observed that the minimum thickness, which at 30° just allowed a small portion of radiation to be transmitted, was not sufficiently small to allow transmission at the increased angle of incidence. The thickness had to be reduced to something between 10·3 mm. and 9·9 mm. for the beginning of transmission.

With an angle of incidence of 60°, the minimum thickness for total reflection was found to lie between 7·6 mm. and 7·2 mm.

The minimum effective thickness is thus seen to undergo a diminution with the increase of the angle of incidence.

In the following experiments I kept the angle of incidence constant, and varied the wave-length. I used three different radiators, R1, R2, and R3; of these R1 emitted the longest, and R3 the shortest waves.

The following method of experimenting was adopted as offering some special advantages. If a cube of glass be interposed between the radiator and the receiver