156 MEMORIAL OF JOSEPH III'.MiY. 



Through a hole in the screen the end of a sensitive thermo-pile was 

 projected, the wires of which were connected with a galvanometer. 

 By slightly moving the smaller end of the telescope, different parts 

 of the image of the sun could be thrown on the end of the thermo- 

 pile, and by the deviation of the needle of the galvanometer, the 

 variation of the heat was indicated. In this way it was proved 

 that tlic spots radiated less heat than the adjacent parts, and that 

 all parts of the sun's surface did not give off an equal amount of 

 heat. 



XI. Another series of experiments was made with what was 

 called a thermal telescope. This instrument consisted of a long 

 hollow cone of pasteboard, lined with silver leaf and painted out- 

 side with lampblack. The angle at the apex of this cone was such 

 as to cause all the parallel rays from a distant object entering tin' 

 larger end of the cone to be reflected on to the end of a thermo- 

 pile, tho poles of which were connected with a delicate galvan- 

 ometer. When the axis of this conical reflector was directed toward 

 a distant object of greater or less temperature than the surrounding 

 bodies, the difference was immediately indicated by the deviation of 

 the needle of the galvanometer. For example, when the object 

 was a horse in a distant field, the radiant heat from the animal was 

 distinctly perceptible at a distance of at least several hundred yards. 

 When this instrument was turned toward the celestial vault, the 

 radiant heat was observed to increase from the zenith downward; 

 when directed, however, to different clouds, it was found to indi- 

 cate in some cases a greater, and in others a less, degree of radiation 

 than tin- surrounding space. When the same instrument was 

 directed to the moon, a slight increase of temperature was observed 

 over that of the adjacent sky, hut this increase of heat was attrib- 

 uted to the reflection of the heat of the sun from the surface of the 

 moon, and not to the heat of the moon itself. To show that this 

 hypothesis is not inconsistent with the theory that the moon lias 

 cooled down to the temperature of celestial space, a concave mirror 

 was made of ice and a thermo-pile placed in the more distant focus; 

 when a flame of hydrogen, rendered luminous byaspiral platinum 

 wire, was placed in the other focus, the needle of the galvanometer 

 attached to the pile indicated a reflection of heat, care being taken 



