Diffusion of Hydrogen through Hot Platinum. 9 
as the mean of two concordant. determinations. A similar 
experiment with the pump, &c., connected at N gave the 
value of V as 576 c.c. 
The aim of the first series of observations at atmospheric 
pressure was to test whether there was any per ceptible change 
produced in the rate of diffusion by the heating in one experi- 
ment. This was done by raising the temperature of the tube 
in steps of about 140° from 576° to 1136°, making a 
measurement at each temperature, and then repeating “the 
observations in the inverse order on the downward grade. If 
there were an effect of the kind looked for, there ought to be 
a difference between any two observations at the same 
temperature, the amount of which should increase with the 
interval between the times at which they were taken. As 
the rate of escape of the hydrogen is considerable at the 
highest temperatures, there was a gradual diminution of the 
total pressure of the hydrogen in the apparatus from 75:9 to 
74:1 cms. during the experiment. The numbers were cor- 
rected to a pressure of 76:0 cms. by assuming that the flow 
was proportional to the square root of the pressure, an 
assumption which will be justified later. The numbers which 
were obtained are given in the subjoined table. 
TABLE. 1-—P= 76:0 ems. 
| aeeen at of Diffusion. (Q). 
By Sista «i 4655x1079 | 4:60 x1079 
mtg? VIN ks | 205 «10° | 2015x1072 
BG. ec cectnene:|, | 37910. h 620 x167° 
aa aR | Nose TAG © 1542x1074 
BIBS. J cect. cdc | 3564 x 10~ ee 
The observations were taken in the order indicated by the 
arrows. It will be seen that the numbers in the right-hand 
column are all slightly smaller than the corresponding 
numbers on the left. In no case, however, is the difference 
greater than the admissible experimental cae and it was 
not confirmed by later experiments. 
_ After having made a sufficient number of Pier 
observations to ‘thoroughly test the working of the apparatus, 
a series of measurements was made, in the manner already 
