Parr .— The Response of Pilotolus to Light. 183 
all lateral radiation. In each end of the apparatus was a metal box (d) with 
adjustable slit (e) in which the light source was enclosed. 
At 154 cm. (60*5 in'.) from the light source a lens (1), having a focal 
distance of 71*12 cm. (28 in.), was placed. Seventy-nine cm. (31*1 in.) 
from each lens a carbon bisulphide prism (p) deflected^a spectrum into the 
experimental chamber through adjustable slits in a screen (;;/). It will be 
noted that the spectrum thrown 011 the screen was at twice the distance from 
the lens of that reported by Laurens, Day, and Gross. Much better 
spectral results were thus obtained. The thermopile (t) was connected 
in series with the galvanometer (g), and deflexions of the galvanometer were 
read with telescope and scale at a distance of 1 metre. A tube («), with 
the inside completely blackened, led from the thermopile to the Hefner 
lamp (h), and served to exclude all lateral radiation and draughts from the 
thermopile. This apparatus was located in a darkened basement room 
of the Natural History Building. The walls of this room were blackened, 
and all lights used in connexion with the apparatus were adjusted so that no 
stray rays could enter. The temperature and moisture conditions were 
comparatively constant. The room was thoroughly aired for two hours 
before each experiment. 
The junctions of the thermopile were composed of bismuth and silver 
welded with tin and covered with platinum-black and lampblack. When 
a ray of light falls upon this surface at the junction of the two metals it is 
absorbed as heat and transformed by the junctions into electromotive 
force. A deflexion of the galvanometer in series with the thermopile 
indicates the intensity of the radiation. The galvanometer used to indicate 
the energy absorbed had a moving astatic system suspended by a fine 
quartz fibre. It was shielded from magnetic disturbance by tubes of soft 
iron having small apertures through which the scale reflected by the 
