

PHYSICAL CONSTANTS OF HYDROGENIUM. 169 
1:6, and the alteration of volume is equivalent to saying in the formation of 
this condition 7 litres of gas are condensed into 1 cubic centimetre. The 
maximum charge of hydrogen the palladium would take was given to No. 7 
experiment, and this very nearly accords with the composition Pd, H,. In 
treating masses of palladium the above proportion was never exceeded. 
As all the above experiments were made with bar palladium increasing to 
saturation, it was necessary to find similar values for a mass decreasing from 
saturation, the hydrogen being expelled by heating. For this purpose a piece 
of plate palladium was employed. When the plate was fully charged for the 
first time, and the specific gravity of the hydrogen found as in the previous 
experiment, the result was a density of 0°621, the plate in this case contained 
021 grammes of hydrogen. When a portion of the hydrogen was expelled by 
heat, leaving 0°127 grammes in the plate, the density of the hydrogen was 0-633, 
and where only 0:0495 gramme was left, the value was 0°615. The mean specific 
gravity of the three different alloys of plate was 0°623, nearly the same as pre- 
viously found for the bar. After palladium has been treated with hydrogen, 
and heated, the metal becomes porous, and often blisters, thus rendering specific 
gravity determinations very uncertain. The curious discovery of GRAHAM, that 
the strain on the particles of palladium produced by wire-drawing induces such 
a curious shrinking in the length after the hydrogen has been expelled, suggests 
the importance of investigating how a saturated piece of bar would resist the 
action of sustained pressure or tension. If the formation of the alloy is attended 
with an increase of volume in excess of the sum of the volumes of the consti- 
tuents regarded as in the solid state, then partial decomposition ought to occur 
under great pressure. Experiments on this subject are left over for the present. 
After palladium has been used in the above experiments, and the last trace of 
hydrogen is removed by heating, the specific gravity is found to have diminished 
from 12:0359 to 11:9546. If this final value is taken in calculating the mean 
density, then the average result of the three hydrides of plate is 0-707 for the 
specific gravity of the condensed hydrogen. The mean of the first and second 
series of experiments is thus 0°664. A piece of palladium, weighing 31 grammes, 
was fused in a lime crucible with the oxyhydrogen flame, and the specific gravity 
was now found reduced to 105549. When this sphere of palladium was hydro- 
genised for a very long time (twenty-four hours) only 0:0684 grammes of hydro- 
gen were absorbed, and by further treatment nothing was added. The specific 
gravity of the hydrogen in this case was 0°655, nearly identical with the mean 
of the values found from the plate experiments. 
Specific Heat Observations. 
The apparatus devised for this purpose is represented in the plate appended 
to this paper. The calorimeter used in the experiments had the form (A) shown 
