100 F. W. Very— A Cosmic Cycle. 



temperature and viscosity increase, explosions again become 

 frequent or periodic. 



There is a limit to temperatures in outer layers immediately 

 below the photosphere, possibly due to viscosity alone, and at 

 any rate partly attributable to this cause. If the photosphere 

 shares in the thermal changes of external layers, we may get 

 some idea of outside temperatures from the distribution of 

 energy in the spectrum. Even if the thermal condition of the 

 photosphere is determined by some physical property which 

 prevents it, for example, from exceeding a certain maximum 

 temperature, and granting that there is a continual action tend- 

 ing to produce this maximum and therefore constant tempera- 

 ture, superficial radiation will always . cause the outer layer to 

 fall a little below the maximum, and this the more as the outer 

 atmosphere is more transmissive of photospheric radiation. 

 Allowing for the general absorption of the terrestrial and solar 

 atmospheres, the spectral energy-curve of the sun has its 

 maximum at 0*45 jut, corresponding, by Paschen's law of the 

 wave-length of maximum energy for a black body, to an abso- 

 lute temperature of 6424 centigrade degrees. It is not likely 

 that the naked photosphere of any star has its spectral maxi- 

 mum much beyond 0*45 ja, but the emissive power of the photo- 

 sphere may not be that of an absolutely black body, in which 

 case the temperature will be higher. 



On the supposition that the photosphere shares the varying 

 temperature of the outer layers, we might have surmised, in 

 the absence of the interlinear comparisons, that the heat in 

 layers near the photosphere would diminish while central 

 temperatures were increasing, viscosity correlating the two. 

 But it is necessary to recognize that we do not know the distri- 

 bution of energy in the spectrum of the naked photosphere of 

 a star except as revealed by interlinear comparisons, and not 

 even then except by a further correction for general absorption. 

 When these allowances have been made, it does not seem nec- 

 essary to assume any large variation in photospheric radiation, 

 and it appears probable that there are some special substances 

 which are precipitable in solid or liquid form between definite 

 temperatures, out of which the photospheric clouds are formed, 

 and whose presence serves as an indicator of the level within 

 the stellar sphere at which these temperatures are attained. 

 Thus in the sun, the photospheric level appears to be about 

 160,000 miles* below the limit of the coronium atmosphere. 



Sir William and Lady Huggins in their " Discussion of the 

 Evolutional Order of the Stars,"f suggest that in " the earlier 

 subdivisions of the white-star type, it is by no means certain 



* E. W. Maunder, The Indian Eclipse, 1898, p. 84. 



f An Atlas of Representative Stellar Spectra, p. 69, 1899. 



