144 ANNUAL OF SCIENTIFIC DISCOVERY. 



amount of heat generated by cosmical falls is therefore at least 

 9,000,000 times greater than in our terrestrial example. 



Rays of heat on passing through glass and other transparent bodies 

 undergo partial absorption, which differs in degree, however, accord- 

 ing to the temperature of the source from which the heat is derived. 

 Heat radiated from sources less warm than boiling water is almost 

 completely stopped by thin plates of glass. As the temperature of a 

 source of heat increases, its rays pass more copiously through diather- 

 mic bodies. A plate of glass, for example, weakens the rays of a 

 red-hot substance, even when the latter is placed very close to it, 

 much more than it does those emanating at a much greater distance 

 from a white-hot body. If the quality of the sun's rays be examined 

 in this respect, their diathermic energy is found to be far superior to 

 that of all artificial sources of heat. The temperature of the focus 

 of a concave metallic reflector in which the sun's light has been col- 

 lected is only diminished from \ to -|- by the interposition of a screen 

 of glass. If the same experiment be made with an artificial and 

 luminous source of heat, it is found that, though the focus be very 

 hot when the screen is away; the interposition of the latter cuts off 

 nearly all the heat ; moreover, the focus will not recover its former 

 temperature when reflector and screen are placed sufficiently near to 

 the source of heat to make the focus appear brighter than it did in 

 the former position without the glass screen. 



The empirical law, that the diathermic energy of heat increases 

 with the temperature of the source from which the heat is radiated, 

 teaches us that the sun's surface must be much hotter than the most 

 powerful process of combustion could render it. 



Other methods furnish the same conclusion. If we imagine the sun 

 to be surrounded by a hollow sphere, it is clear that the inner surface 

 of this sphere must receive all the heat radiated from the sun. At 

 the distance of our globe from the sun, such a sphere would have a 

 radius of 215 times as great, and an area 46, 000 times as large, as the 

 sun himself; those luminous and calorific rays, therefore, which meet 

 this spherical surface at right angles retain only ^g^r^ part of their 

 original intensity. 



If it be further considered that our atmosphere absorbs a part of the 

 solar rays, it is clear that the rays which reach the tropics of our 

 earth at noonday can only possess from g^^o' to WO~OTT f tue 

 power with which they started. These rays when gathered from a 

 surface of from five to six square meters, and concentrated in an area 

 of one square centimetre, would produce about the temperature which 

 exists on the sun, a temperature more than sufficient to vaporize 

 platinum, rhodium, and similar metals. 



A correct theory of the origin of the sun's heat must explain the 

 cause of such enormous temperatures. This explanation can be 

 deduced from the foregoing statement. According to Pouillet, the 

 temperature at which bodies appear intensely white-hot is about 

 1,500 C. The heat generated by the combustion of one kilogram 

 of hydrogen is, as determined by Dulong, 34,500, and according to 

 the more recent experiments of Grassi, o4,6G(3, units of heat. One 

 part of hydrogen combines with eight parts of oxygen to form water ; 

 hence one kilogram of these two gases mixed in this ratio would 

 produce 3,850. 



