36 Trans. Acad. Sci. of St. Louis. 



fold radial limits here pointed out, it will still admit of wide 

 application throughout the heavens. In the present state of our 

 knowledge of the laws of gases, we refrain from any attempt 

 at fixing more definite limits to the Solar Nebula, which would 

 also depend on the temperature of the mass. For if the mass 

 could be kept sufficiently heated it might extend its bounds 

 far beyond the present limits of the Solar system. 



In concluding these remarks, the following curious illus- 

 tion of a reversible process is thought to be worthy of atten- 

 tion. Imagine a huge pipe made of some material impervious 

 to heat laid from the center of a great hot star like Canopus 

 to the center of the Sun, and suppose the two ends to be 

 closed by non-fusible pistons which freely transmit the heat 

 communicated along the pipe. Heat will flow steadily from 

 the hotter to the cooler source, and as the center of Canopus 

 is assumed to be much hotter than the center of the Sun, the 

 material at the latter point will receive a supply of heat 

 which will tend to elevate the temperature of the Sun's mass ; 

 but as heat cannot be supplied to the gaseous globe without 

 expanding its dimensions, the result will be an increase in its 

 diameter and a corresponding fall in its temperature. If the 

 flow of heat along the pipe is sufficiently great (it must of 

 course surpass the amount lost by surface radiation), and 

 kept up long enough, the compact mass of the Sun will be 

 expanded into a vast diffuse nebula filling the planetary 

 orbits ; and if the pipe then be intercepted the cold rare mass 

 will again slowly condense and rise in temperature, and the 

 planetary system will be formed anew ! • 



6. Conclusions based upon the fundamental law of temper- 

 ature. The following conclusions seem to be legitimate 

 inferences from the remarkably simple law of nature treated 

 above. 



(a.) The diffused nebulae are near the temperature of space. 



In the formula T = p , K is different from each body, but 

 always finite, and hence when B is infinite T is zero.* 



* Cf . Astronomical Journal, No. 458. 



