132 Mr Bevan, On the Joule-Thomson Effect. 



This equation for a great many substances appears to be very 

 accurately followed, even at temperatures in the region of the 

 critical point. This equation represents then the best empirical 

 formula which has as yet been obtained for the behaviour of gases 

 and liquids. The quantities a and b in Ramsay and Young's 

 formula are however found not to be any simple function of the 

 volume, so that this equation alone does not carry us very far. 



If however we consider the equation 



Rd /, 6 , ,„\ 



and the equation 

 we obtain 



G €u 



</>! (6) in the form -7,+^, 

 so that equation (5) becomes 



p.- JW(l + $(£ + £)) (7). 



In thus deriving a form for the function tf> 1 (0) we obtain forms 

 for the functions f x (v) and f 2 (v). But these functions determined 

 in this way do not give very good values for the numbers obtained 

 by experiment by Ramsay and Young. The functions are far too 

 simple to give anything but a fairly good approximation to these 

 numbers. 



The equation (7) is equivalent to van der Waals' equation if 

 we neglect products of the small quantities, for we have from 

 van der Waals' equation 



P + ^)(v-b) = R0, 



pv = Rd + bp — 



"(^fiS-w)) 



The form we obtain for the cooling effect from equation (7) is 



G P \T~ B 



If we regard G p as constant this is the same form as used by 

 Rose Innes in an empirical formula which agrees well with the results 

 of Thomson and Joule's experiments. In the case of carbon- 



