308 PHENOMENA, ATOMS, AND MOLECULES 



cules are independent of those of the incident molecules ; in other words, 

 that a = (i — yv) = i.oo. 



If we consider the case of argon in contact with platinum, for which 

 Soddy and Berry found a = 0.85, we obtain : 



mi = 40, 



W2 = 195, 



a = 0.85, 



^ = 0.72, 



Y = o.i7, 



I - ^v = 0.85, 



I — yv = 0.965. 



The accommodation coefificient for viscosity thus decreases with in- 

 creasing density of the gas, but remains nearly equal to unity. 



The experimental work on slip has shown that the accommodation 

 coefficient for viscosity is nearly unity, but the accuracy of the results has 

 not yet been sufificient to determine whether tlie values calculated by Baule's 

 theory are correct. 



Baule attempts to calculate the coefficient v by making various as- 

 sumptions as to the manner in which the molecules on the surface are 

 "shadowed" by others, but the validity of the assumptions seems so doubt- 

 ful that the resulting equations have little more value than empirical 

 equations. 



On the whole, Baule's theory marks a distinct advance in our under- 

 standing of the mechanism of the exchange of energy between a gas and a 

 soUd, but closer consideration of the assumptions underlying Baule's theory 

 shows that they cannot correspond to the facts. 



Baule fails to take into account the attractive forces between the atoms, 

 although it is evident that such forces must be of prime importance in 

 determining the amount of condensation and reflection. The phenomena 

 of cohesion, condensation and adsorption are all manifestations of these 

 attractive forces, and Sutherland ^^ has shown that even in the gaseous 

 state the attraction between molecules considerably decreases their mean 

 free path. 



We know from the work of Bragg and others that the atoms of crystals 

 are arranged according to a space lattice structure in which the identity 

 of molecules usually disappears. 



In a crystal of sodium chloride the forces holding the crystal together 

 evidently act between adjacent sodium and chlorine atoms ; in other words, 

 they are of the nature of what we have usually called chemical forces. 



^^ Phil. Mag., 36, 507, 1893. 



