PROFESSOR KOPP ON THE SPECIFIC HEAT OF SOLID BODIES. 
191 
with those of the corresponding compounds of such elements as have an atomic heat 
= 6'4. The average atomic heat of compounds RS and RS is 11*9, according to the 
determinations in § 83, while those of chlorine compounds RC1 and R Cl (§ 84) =12*8, 
that of the corresponding bromine compounds =13*9, and of the corresponding iodine 
compounds =13*4. In comparing more complicated sulphur compounds, sulphates, for 
instance, with other compounds of analogous composition, the same is met with ; 
although such complicated compounds are of little value in giving data for deciding on 
such small differences. The specific heat of the simpler phosphorus compounds has not 
been investigated ; for more complicated compounds, although they point to a smaller 
atomic heat for P than 6 - 4, the above remark also applies. 
The determinations of the specific heat of silicium give for this element also a smaller 
atomic heat than 6*4 (compare § 82), and the same conclusion results from a comparison 
of the atomic heats of Si 0 2 , and the oxides, R 0 2 , of the silicates R Si 0 3 , and the oxides 
R 2 0 3 . The atomic heat to be assigned to silicium cannot as yet be settled with any 
degree of certainty. Direct determinations, varying considerably from each other, give 
a specific heat mostly greater than 4; while the numbers obtained indirectly, and them- 
selves also not closely agreeing, are partly considerably smaller. If in the sequel I put 
the atomic heat of silicium at 3*8, corresponding to the lowest number found for the 
specific heat of this element, I do so for want of other and more certain data. I con- 
sider this number as quite uncertain. 
The atomic heat of boron , from the direct determinations of the specific heat, is con- 
siderably smaller than 6 *4 ; and the atomic heats of boron compounds confirm this, as 
was discussed in §§ 93 and 95. By comparing the atomic heats of such boron and sul- 
phur compounds as contain along with boron and sulphur the same elements in the 
same proportions, the atomic heat of boron is found to be half that of sulphur. The 
atomic heat of KB0 2 =16*8 is exactly half that found for K 2 B0 4 =33*6; the atomic 
heat of BbB 2 0 4 = 26’5 is almost exactly equal to that for RbS0 4 =25*7. Taking the 
atomic heat of S, in accordance with the above discussion, at 5-4, that of B would be 
2 ’7; the numbers obtained directly for the atomic heat of boron (§ 82) from the expe- 
riments on the specific heat of this element agree with sufficient accuracy. In the sequel 
I take the atomic heat of B at 2-7. A smaller number is obtained in other compari- 
sons ; for instance, of the atomic heats of B 2 0 3 and of the oxides R 2 0 3 , or of the salts 
R B 0 2 and the oxides R 2 0 2 ; but in such indirect determinations of the atomic heat, 
where such small numbers are to be determined, as is here the case with the atomic heat 
of boron, the results are very uncertain, owing to the fact that the entire uncertainty in 
the atomic heats of the compounds, and in the assumption that the elements correspond- 
ing to boron in compounds of analogous composition have really the atomic heat, =6*4, 
is thrown on the final result. 
Lastly, carbon also, from the direct determinations of its specific heat (§ 82), has a 
much smaller atomic heat than 6 ’4. The same result follows from a comparison of the 
atomic heats of carbon compounds : the atomic heat of the carbonates, R 2 € 0 3 =28 , 4 as 
mdccclxv. 2 D 
