& 
Chemistry and Physics. 313 
to Sas the chemical ‘fo orce: but in no case y except the first 
has any attempt been made to investigate the interaction of 
the two. The atomic theory explains such cases of indefinite com- 
bination by distinguishing between atomic and molecular union, — 
and it regards the complex products as resultin gre- 
gation of dissimilar molecules, in each of whi ch, however, the law 
of — ite proportion rigorously holds. It is necessary, how- 
ever, many cases to assame ees existence of very complex 
molecules, and the doctrine of in F phenowenn, certainly gives a 
the mmiking oe. If a crystal may enclose aici wholly 
foreign to mposition, why may it not enclose an excess 
either of sts sg han air ts? There can be no doubt ink ¢ in such : 
Cases the amount of material thus enclosed depends to agreater = 
oF Tess extent on the chemical force, and there is a continuity in : 
t which is not, and gives at least the appearance that the S 
t so controlled may be variable. | 
the assumption, however, of the existence of sen cules and “ 
= mpounds, the cater “of molecules of such compounds: 
ae that would form at any one time raust depend—certainly among — 
= conditions—on the strength of the —* = cede we 
while the molecules of the definite compound would 
_ 4& greater or less extent. f the ec stit- 
. us might arise such a regulated variation 0: nasthe 
_ Writer’s investigation indicated. The extent of the for mation of 
- Molecules of definite yenreese in the menstruum would obyi-— 
ously whose action would determine _ 
on of these 
2 ot ce aiky ‘but the ae; 
| setae ng rh ity as gai onee a sudden eh ge. 
des lization is always a break of continuity, and al ears. 
t talline struct structure may, as we yet a8 the seen, enclose | orel 
| to a ve large extent, et, as ee 
Salig ts ll, Tare Sunres, V Sa ETE No. = 1883. 
t is a very probable hypothesis, that when we bring to- 
in solution or in fusion substances capable of yielding defi- — 
is form in such a menstruum. we dened expect ct that 2 
_ together, the crystalline structure thus formed would Poi - a 
: Uents of this compound if present in excess, and the amount of such 
~uclosure must depend on ng relative proportion of the molecules 
. in the menstruum; and this roportion, as we have seen,must have = 
ae been already determined be the strength of the chemi cal force, = 
ition ) 
