Molecules of a Gas in a Field of Force. 647 



capable o£ expressing the results o£ experiments within 

 the limits of error. To the degree of approximation to 

 which (35) and (37) are simultaneously valid 



A 2 = A 3 e-tT-f (38) 



For tungsten in the neighbourhood o£ 2000° K, Langmuir * 

 finds A 3 = 34xl0 6 in the units necessary to give amperes 

 per sq. cm. In the units of equation (36) this gives 



A* = l-76xlO u (39) 



Values o£ A 2 of this order of magnitude, although in most 

 cases a little higher, have been given by different observers 

 for platinum, tantalum, molybdenum, and carbon. Except 

 for lime and baryta, the values of A 2 given by the non- 

 metallic conductors are often lower than 10 l0 . This is to be 

 expected from the theory, since even at the high tempera- 

 tures used many of these substances conduct so badly that Y t 

 will be a large quantity. It will moreover vary rapidly with 

 T. Thus the conditions with these substances approximate to 

 those required for the ordinary dynamics to apply. 



On the whole, the evidence for and against the reasoning 

 leading up to equations (13) which is furnished by thermionic 

 phenomena at present seems to be indecisive in character. 

 The agreement between the calculated and experimental 

 values (36) and (39) of A 2 is neither good enough to 

 establish the theory on a firm basis nor bad enough to 

 discredit it. Perhaps the strongest argument for the theory 

 which can be gathered from the thermionic field lies in the 

 fact that it accounts for the possession by the external elec- 

 trons of Maxwell's distribution of velocity, whereas the 

 internal electrons, as is indicated by the specific heats and 

 other phenomena, do not possess this distribution of velocity. 

 These remarks are made quite apart from a discussion of the 

 validity of equations (2) and (3), which underlie the whole 

 position, at any rate in the particular form in which it has 

 been considered. Equations (2) and (3) and their immediate 

 consequences are to be regarded as supported by (1) Wien's 

 theory o£ metallic conduction and its agreement with the 

 experimental results, (2) the general course of thermo- 

 electric phenomena at very low temperatures, and (3) the 

 equation o£ state of helium at low temperatures. 



University of London, 

 Kings College. 



* Luc. cit. 



