250 Dr. S. Chapman and Dr. F. W. Dootson : 



A mixture containing nearly equal proportions of each, by 

 volume, was introduced into the apparatus at atmospheric 

 pressure ; the outer connexions were closed, and with the 

 intermediate stopcock open one bulb was heated up to 230° C, 

 the other being kept at the temperature of running water 

 from the mains. The apparatus was left in this state for 

 four hours, to allow some time for the thermal diffusion to 

 take effect, and the stopcock was then closed. The proportion 

 of C0 2 in the mixture contained in each bulb was then 

 determined by analysis of samples drawn off (after cooling). 

 The result was as follows : — 



(1) 



/Hot bulb (230° C.) 44-9 per cent. H 2 : 55"1 per cent .C0 2 .. 

 /Cold bulb (10° 0.) 41-3 per cent. H 2 : 58-7 per cent. CO*. 



Two further experiments on similar lines, but with rather 

 different proportions of the two gases, were made, the time 

 allowed being four hours in each case, as before ; the results 

 were as follows : — 



/ 9 n /Hot bulb (240° C.) 85'3 per cent. H 2 : 14*8 per cent. C0 2 . 

 ^ "/Cold bulb ( 10° C.) 83-4 per cent. H 2 : 16-6 per cent. C0 2 .- 



m / Hot bulb (200° C.) 53-2 percent. H 2 : 46-8 per cent. C0 2 . 

 {6) LColdbulb (10° C.) 51*0 per cent. H 2 : 49-0 per cent. C0 2 . 



An attempt was also made to perform a similar experiment 

 using a mixture of sulphur dioxide and hydrogen, but it 

 was unsuccessful at first : the hot bulb was raised to a low 

 red heat, and the gases reacted to form sulphur and water. 

 Later on, however, with a slightly different form of the 

 apparatus, the phenomenon was clearly demonstrated also 

 with these gases. In the new apparatus cylindrical tubes of 

 wide bore took the place of the spherical bulbs, and the hot 

 side was heated by an electric furnace formed of resistance- 

 wires wound round a cylindrical tube of asbestos. The hot 

 tube was of 200 c.c. capacity, and the cold one of half this 

 size ; they were joined by a tube of 10 mm. diameter, pro- 

 vided with a stopcock of 8 mm. diameter. In the first 

 experiment with this apparatus the gases were analysed after 

 only two hours, the result being as follows : — 



f Hot bulb (220° C.) 58-3 per cent. H 2 : 41-7 per cent. kS0 2 .. 

 (4 J /Cold bulb (10° C.) 57-9 per cent. H 2 : 42-1 per cent. S0 2 .. 



The differences hardly exceed the limits of experimental 

 error, the time being insufficient for the attainment of the 

 steady state : with this form of apparatus the diffusion 



