F. W. Clark—Molecular Heat of Similar Compounds. 841 
NaCl, 21401; KCl, ‘17295—all by Regnault; RbCl, 112, 
Kopp. From these, multiplying by the atomic weights, we get 
the following molecular heats: 
LiCl, 11°99. 
NaCl, 12°52 
Cl, 12°88. 
RbCl, 13°54 
Here we have a gradual increase, accompanying an increase 
m the atomic weights. A similar increase is found in Kopp’s 
Avogadro’s and Neumann’s determinations for NaCl and Cl, 
KCl, 12°88, 
KBr, 13°47. 
KI 13°60. 
Again there is a slight increase accompanying the rise in 
atomic weight. The chloride, bromide, and iodide of silver, 
and the same series for lead, illustrate this still farther, although 
it is not worth while to cite the figures here. For sodium, 
starting with the fluoride, we have about the same thing, only 
the bromide proving, perhaps, an exception. But since only 
one determination of the specific heat of the bromide has ever 
been published, it is likely that even this compound may be 
brought into line 
Wo series of oxides are very perfect, as follows: As,O,, 
12786; Sb,O,, 09009; Bi,O,, ‘06053; all by Regnault. 
SiO, (quartz), 186; TiO, (rutile), -157; SnO, (tinstone), 
0894; all by K These determinations give us the follow- 
ing molecular heats: 3 : 
As,O,, 5°31. SiO, i 
Sb,0,, _ 26°31. U5: 12°37, 
Bi,O,,.. 28°3 SnO, 13°40. 
3. 
If, instead of single determinations, we take the mean of all 
series of compounds. Exceptions (even seeming exceptions) 
Am. Jour, Sco1.—Tuirp Serres, VOL. VIII, No. 47.—Nov., 1874. 
22 
