37 MAXWELL'S LAW 85 



The two series of experiments with the perforated brass 

 plate were carried out with slightly different arrangements 

 of the apparatus, and the second of these columns I have 

 calculated from the results given by Graham for the speeds 

 of efflux. 



The agreement between the observed times and the 

 square roots of the specific gravities may, on the whole, be 

 considered excellent, and may serve as proof of the correct- 

 ness of the theoretical law that has been established ; it 

 will not, therefore, be necessary to quote further observa- 

 tions made by Graham with altered values of the pressure. 

 The greatest deviation from the law is exhibited by hydro- 

 gen, and, doubtless, for the reason assigned by Graham, 

 that for this light gas, which undergoes very great friction 

 in narrow tubes, the plate was not thin enough and the tube 

 not short enough. 



Not only does the law hold for the efflux of gases into 

 vacuum, but the times of efflux of different gases are also 

 proportional to the square roots of their specific gravities if 

 under otherwise similar circumstances especially, therefore, 

 under equal pressures they stream into a space filled with 

 air or gas the atmosphere, for instance. 



The ground on which we are entitled to extend in this 

 way the applicability of the law is very simple in the case 

 of the back pressure being small in comparison with that 

 which causes the efflux ; the speed of efflux, in fact, will 

 not be materially diminished by a slight resistance. The 

 back pressure may, indeed, increase up to half the value of 

 the pressure that forces out the gas without the speed 

 sensibly falling off. This remarkable fact was first noticed 

 by St. Venant and Wantzel, 1 and has been confirmed by 

 later observers. On the kinetic theory we should explain 

 the matter thus : that on issuing from the orifice and coming 

 into the space which is less densely filled with gas, the 

 molecules very seldom collide with others, since the latter 

 also have most of them only just emerged from the orifice, 

 and, therefore, for the most part are moving in the same 

 direction and with nearly the same speed ; they will therefore 

 1 Journ. de VEcole Polyt. xvi. 1839, p. 101. 



