PROFESSOR KOPP ON THE SPECIFIC HEAT OF SOLED BODIES. 
189 
by subtracting from the specific volume of the oxide that of the metal in it, and con- 
sidering the remainder as the volume of oxygen. It would follow from this that the 
specific volume of oxygen in suboxide of copper is much greater (about four times as 
great) than in oxide of tin. But if the atomic heat of oxygen be deduced by sub- 
tracting from the atomic heat of the oxide that of the metal in it, it is found that the 
atomic heat of oxygen in suboxide of copper and in oxide of tin gives almost exactly 
the same number. Hence it does not seem that the state of condensation in which a 
constituent may be contained in a compound has any material influence on the atomic 
heat of this constituent. 
99. From all that has been said in the foregoing paragraphs the following must be 
adhered to. (1) Each element in the solid state, and at a sufficient distance from its 
melting-point, has one specific or atomic heat, which may, indeed, somewhat vary with 
physical conditions, different temperature, or density for instance, but not so consider- 
ably as to be regarded in considering in what relations the specific heat stands to the 
atomic weight or composition; and (2) that each element has essentially the same 
specific or atomic heat in compounds as it has in the free state. On the basis of these 
two fundamental laws it may now be investigated what atomic heats individual elements 
have in the solid free state and in compounds. Indirect deductions of the atomic 
heats of such elements as could not be investigated in the solid free state are from 
these propositions admissible : that from the atomic heat of a compound containing such 
an element the atomic heat of everything else in the compound is subtracted, and the 
remainder considered as the expression for the atomic heat of that element. Such in- 
direct determinations of the atomic heat of elements may be uncertain, partly because 
the atomic heat of the compounds is frequently not known with certainty, as is seen 
from the circumstance that analogous compounds, for which there is every reason to 
expect the same atomic heat, are found by experiment to have atomic heats not at all 
agreeing ; but more especially because the entire relative uncertainty in the atomic 
heats for a compound, and for that which is to be subtracted from its composition, is 
concentrated upon a small number, the residue remaining in the deduction. But 
when such deductions are made, not merely for individual cases, but for different com- 
pounds, and for entire series of corresponding compounds, they may be considered suffici- 
ently trustworthy to make the speculations based upon them worthy of attention. Of 
course in indirectly deducing the atomic heat of an element, its simpler compounds, 
and those containing it in greatest quantity (measured by the number of atoms), promise 
the most trustworthy results. 
100. For Silver , Aluminium , Arsenic , Gold, Bismuth, Bromine , Cadmium, Cobalt, 
Copper, Iron, Mercury , Iodine, Iridium, Potassium, Lithium, Magnesium , Manganese, 
Molybdenum, Sodium, Nickel, Osmium, Lead, Palladium, Platinum, Rhodium, Antimony, 
Selenium, Tin, Tellurium, Thallium, Tungsten, and Zinc, it may be assumed, from the de- 
terminations of their specific heat in the solid state (§ 82), that their atomic heats, in 
