192 
ME. EOBEET MALLET ON VOLCANIC ENERGY. 
He can scarcely have looked at Bischoff’s own statements, or their insecure founda- 
tions would not have escaped his acute mind. 
Bischoff’s experiments were conducted by two methods. 
In one he fused the rock in clay crucibles holding only a few pounds, determined the 
capacity of the empty crucible by the weight of its contents in mercury, and the volume 
of the unfused rock by its specific gravity and weighing. He then fused the rock, and 
to obtain its volume when liquid he dipped the depth of the surface below the brim of 
the crucible by means of a graduated iron wire, and then permitted the whole to cool. 
He then filled up over the surface of the solidified rock with mercury until the iron 
clipping-wire touched the liquid ; from the weight of this mercury he infers the volume 
of the contraction; and filling now the whole vacuity of the crucible with mercury, he 
obtains, by weighing and deduction of these latter volumes of mercury from that which 
the empty crucible originally contained, the volumes of the rock when fused and when 
cold. 
Substantially this is Bischoff’s method ; though for all his steps and apparent pre- 
cautions his original lengthy memoir must be consulted. 
Now it scarcely needs to be shown at any length that reliable results by this complex 
and indirect method are perfectly impossible, even were not one condition which really 
vitiates the whole found to be entirely disregarded. 
He was obliged to use mercury because his crucibles absorbed water ; but the strong 
capillarity of mercury and its high specific gravity could not fail to introduce great 
errors in the deduction from its weight in such very limited volumes. 
The supposed expansion of the crucible and enlargement of its capacity when heated 
was taken by endeavouring to measure the internal diameter of its brim when cold and 
when containing the liquid rock, and it is assumed the expansion affected all parts of 
the crucible alike, so that its capacity, in fact, was proportionate to the cube of this one 
imperfectly measured dimension. But this cannot be true, from what we know of the 
changes of form of pottery in baking, even were all parts of the crucible heated alike. 
But what of the contraction of the crucible, in common with all other earthenware, by 
heating, which depends not only on the temperature to which it is raised, but on the 
time of its exposure to the heat 1 
Of this which is certain to have largely affected Bischoff’s 
results, we do not find a word of remark. 
Bischoff admits the difficulty of this class of experiment, and 
he seems in the end to have but little faith in his results, for he 
adopts another method suggested to him by Altiiaus, an Ober- 
bauinspector at Saynerhiitte. This consisted in filling a cast-iron jst 
globe (fig. 10) or shell, M N, provided with a funnel-shaped neck, 
with the fused rock. During the setting by cooling of the fused 
mass a solidified crust formed across the neck d e , and as soon as 
this had reached quite across it was assumed (upon the most manifestly insufficient 
