412 
DR. F. P. BURT AND DR. E. C. EDGAR ON 
explosion vessel itself was cut down and drastically cleaned by consecutive treatment 
with stannous chloride and strong hydrochloric acid, and a hot solution of chromic 
anhydride in fuming nitric acid. After the acid had been washed out, the final 
content of distilled water was replaced by dust-free air drawn in over cotton-wool. 
The vessel was then dried at room temperature in a stream of air passed over calcium 
chloride and filtered through cotton-wool. 
On continuing the determinations, condensible gas was still found in the hydrogen 
residue. The only sources of impurity yet to be examined was the hydrogen. A 
glass spiral was inserted in the train between the palladium tube and the pipette, 
and cooled in liquid air during the filling. A small quantity of gas was condensed in 
the spiral, and the amount of impurity in the hydrogen residue was diminished, 
though not entirely eliminated, by this treatment; facts which at once suggested a 
flaw in the palladium tube. The tube was accordingly cut down and examined, and 
a crack was discovered. The phosphoric oxide in the drying trains had not been 
distilled in oxygen, and was probably the source of contamination. It is of interest 
to note that hydrogen passed over impure phosphoric oxide apparently contains some 
carbon compound which is not condensed at the temperature of liquid air ; perhaps 
carbon monoxide. A new palladium tube, the third described in the preface, was 
procured and set up, and in the following series of determinations, the amount of 
condensible gas in the hydrogen residue was absolutely negligible. In some 
Table III. 
1. 
2. 
3. 
4. 
5. 
Hydrogen 
excess. 
Hydrogen 
residue. 
Difference. 
Ratio of 
combining 
Atomic weight 
of 
volumes. 
hydrogen. 
c.c. 
c.c. 
c.c. 
1 
3-213 
2-323 
0-890 
2-00290 
1-00773 
2 
3-116 
2-221 
0-895 
2-00292 
1-00774 
o 
O 
3-990 
3-092 
0-898 
2-00292 
1-00775 
4 
3-128 
2-240 
0-888 
2-00289 
1-00773 
5 
2-630 
1 • 754 
0-876 
2-00285 
1-00771 
6 
3-424 
2-521 
0-903 
2-00294 
1-00776 
7 
2 ■ 495 
1-584 
0-911 
2-00297 
1-00777 
8 
3-247 
2-341 
0-906 
2-00295 
1-00776 
9 
3-573 
2-655 
0-918 
2-00299 
1-00778 
10 
2-552 
1-675 
0-877 
2-00286 
1-00771 
Mean . 
2-00292 
±0-00001 
1-00774 
1__ 
Maximum deviation, 1 in 
14,000 
