HYDROGENIUM; THE RELATION OF HYDROGEN TO PALLADIUM. 79 
volumes of hydrogen were taken up in twenty minutes), and an opportunity be gained 
of observing if the density of the hydrogenium remains constant, or if it varies with the 
proportion in which hydrogen enters the alloy. In the following statement, which in¬ 
cludes the three experiments already reported, the essential points only are produced:— 
Table. 
Volumes of 
hydrogen occluded. 
Linear expansion in 
millimetres. 
Density of 
Hydrogenium. 
From 
To 
329 
496T89 
498-552 
2-055 
462 
493-040 
496-520 
1-930 
487 
370-358 
373 126 
1-927 
745 
305*538 
511-303 
1-917 
867 
488-976 
495 656 
1-898 
888 
556-185 
563-652 
1-977 
936 
609-144 
618-923 
1-708 
If the first and last experiments only are compared, it would appear that the hydro¬ 
genium becomes sensibly denser when the proportion of it is small, ranging from l - 708 
to 2-055. But the last experiment of the table is perhaps exceptional; and all the 
others indicate considerable uniformity of density. The mean density of hydrogenium, 
according to the whole experiments, excluding that last referred to, is 1-951, or nearly 
2. This uniformity is in favour of the method followed for estimating the density of 
hydrogenium. 
On charging and discharging portions of the same palladium wire repeatedly, the 
curious retraction was found to continue, and seemed to be interminable. The follow¬ 
ing expansions, caused by variable charges of hydrogen, were followed on expelling the 
hydrogen by the retractions mentioned :— 
Elongation. Retraction. 
1st Experiment 9-77 millims.9*70 millims. 
2nd „ 5*765 „ .6"20 „ 
3rd „ 2-36 „ .3-14 „ 
4th „ 3*482 „ .4-95 „ 
23-99 
The palladium wire, which originally measured 609-144 minims., has suffered, by four 
successive discharges of hydrogen from it, a permanent contraction of 23-99 millims.; 
that is, a reduction of 3-9 per cent, on its original length. The contractions will be ob¬ 
served to exceed in amount the preceding elongations produced by the hydrogen, par¬ 
ticularly when the charge of the latter is less considerable. With another portion of 
wire it was carried to 15 per cent, of its length by the effect of repeated discharges. The 
specific gravity of the contracted wire was 12-12, no general condensation of the metal 
having taken place. The wire shrinks in length only. 
In the preceding experiments the hydrogen was expelled by exposing the palladium 
placed within a glass tube to a moderate heat short of redness, and exhausting by means 
of a Sprengel tube ; but the gas was also withdrawn in another way, namely, by making 
the wire the positive electrode, and thereby evolving oxygen upon its surface. In such 
circumstances a slight film of oxide of palladium is formed on the wire, but it appears 
not to interfere with the extraction and oxidation of the hydrogen. The wire mea¬ 
sured— 
Difference. 
Before aharge .... 443 25 millims. 
With hydrogen . . . 449-90 „ -}-G’65 millims. 
After discharge . ... 437-31 „ -—5-94 „ 
The retraction of the wire therefore does not require the concurrence of a high tem¬ 
perature. This experiment further proved that a large charge of hydrogen may be re- 
