472 ON IMPRESSIONS OF COLD 
centrating the heat or cold projected from any remote body. 
Hence also the construction of the Pyroscope, a delicate instru- 
ment, adapted to distinguish and measure those peculiar impres- 
sions of heat or cold. It-consists of a differential thermometer, 
having one ball naked, and the other coated with gold or silver 
leaf; this metallic surface reflecting the greatest part-of the 
projections, while the vitreous surface absorbs almost the whole 
effect. Ifa broad plate of glass be interposed before a cubical 
canister holding hot or cold water, the action on the pyroscope, 
though concentrated by a reflector, will yet'be very much redu- 
ced. But, on removing the plate farther from the canister, and 
therefore nearer to the reflector, this action will be still more 
diminished ; so as at last to become.almost extinguished. Hence 
the projected influence of heat or cold consists not in any 
streaming matter, for the same proportion of the warm or fri- 
gid rays would evidently be intercepted by the glass screen in 
whatever part .af the route between the canister and the re- 
flector it was planted. 
But the question is completely decided, by varying the ex- 
periment. Let two glass plates be covered, each on one 
side with tin-foil ; applying the metallic surfaces now together, 
place the double sereen immediately before the hot canister, 
and the effect on the pyroscope will be greatly reduced ; re- 
verse the position of the plates, by exposing the coated sides, 
and the action on the sentient ball will be entirely extinguish- 
ed. The interposition of a screen would in every case detain 
the whole impressions, if it did not itself become affected by | 
them, and therefore come to act as a secondary but feeble pro- 
jecting surface. Hence the different influence of a metallic 
screen, in comparison with one of glass or paper. 
The impressions of heat or cold emitted from a body are 
not sent equally in all directions. Those projected perpendi- 
cular 
