F. W. Very — A Cosmic Cycle. 101 



that a true photosphere after the pattern of a solar one exists. 

 In these early, and therefore still very diffuse stars, we may see 

 deep down into the star, and the continuous spectrum may 

 come from a thick region of dense gas." 



Experiments have been made which seem to show that con- 

 densed gases may glow with a continuous spectrum, but the 

 phenomenon is an obscure one and subject to the interpretation 

 that we are not dealing with a pure gas, but that, in some way, 

 complex molecules similar to those of liquids* have been 

 formed by the pressure. We have no examples of such spec- 

 tra at very high temperatures, and even at low ones gaseous 

 radiation is commonly limited to particular spectral regions, 

 often very narrow, and comes from layers of small depth. f 



In another place Sir William and Lady Huggins^: say : " The 

 progress of contraction of the stellar mass with increasing age 

 will not only exalt the violence of the convection currents, but 

 also increase the density of the gases, though at the same time 

 probably, the nearer approach of the photosphere towards the 

 star's boundary will have the effect of making the increase of 

 density at and just above the photosphere less than it would 

 otherwise be." I have already noted that quiet gaseous con- 

 vection does not increase with age, but diminishes on account 

 of there being greater viscosity as the temperature rises. In 

 like manner Lane's law that a cooling sun is getting hotter 

 would be a complete paradox, were it not understood that, at a 

 layer of given density, the temperature must always be falling. 

 ¥or a short time the position of a layer of a particular density 

 in a contracting star may travel outwards and at a more rapid 

 rate than the temperature-change in the opposite radial direc- 

 tion at that point ; but in* the end and throughout the greater 

 part of its history, the position of a layer of given density 

 must travel inwards, and the photospheric layer (which, as we 

 have seen, is possibly at constant temperature, and anyhow 

 does not change its temperature through a very wide range), 

 because it is able to radiate more powerfully than the gaseous 

 layers above it, will increase its distance from the outer boun- 

 dary, until in the event of final liquefaction of the region 

 below the photosphere, we might expect to find a relatively 

 small nucleus encompassed by a vast atmosphere. After this 

 stage, the nucleus having ceased to contract to any great extent, 

 the gradual condensation of the atmosphere will undoubtedly 

 cause the limiting surfaces in question to approach. 



In the scheme which I have outlined there are not two 



* See the facts concerning complex molecules of water- vapor and oxygen in my 

 research on '"Atmospheric Radiation," Bulletin G. U. S. Dept. of Agriculture, 

 Weather Bureau, pp. 99-100 and 103, 1900. 



fSee F. Paschen, Wied. Ann., vol. li, p. 34, 1894; alsoF. W. Very, Atmospheric 

 JRadiation, p. 61. 



X An Atlas of Representative Stellar Spectra, p. 70. 



