5.82 
ON RECENT CHANGES IN 
(C 2 H 3 0)' in CJI 3 OHO (acetic acid). It may also attack itself to elements 
or groups of a bivalent or higher order, of course, in appropriate proportion ; 
two atoms of (HO)' are required for a bivalent compound, three atoms for one 
that is trivalent, and so on. 
In this manner are constructed the metallic and other hydrates, c.g .:— 
Slaked Lime.Ca" | jjq 
Hydrate of Copper.Cu" J||q 
Sulphuric Acid.(SO/| ||q 
Glycol. (C, J I 4 )''{ho 
(HO 
Phosphorous Acid.P'" ] HO 
(HO 
(HO 
Phosphoric Acid.(PO)"' < HO 
(HO 
r ho 
Silicic Acid.Si"" | jjq 
Uo 
Now it might be inquired what would be the effect of supplying a multi¬ 
valent radicle with an insufficient number of atoms of this group HO? The 
result would be simply a repetition of the same kind of non-saturation or 
tendency to enter into combination observed in the case of HO itself. Equi¬ 
valent proportions of some other substance would come to supply the defi- 
f H O 
ciency. (C 2 H 4 )" j ^ would be ready to complete itself by taking in for in¬ 
stance Br, or Cl, giving rise to (C 2 H 4 )" j ; or Bi'" ^ ^ in a similar 
manner would become saturated by adding two more atoms of Cl or one 
(Cl 
atom of O", etc., producing Bi'" j q„ (oxychloride of bismuth). 
There is, I think, no need to question the facts which have been detailed 
in the case of hydroxyl, H O ; although this body cannot be separated from the 
compounds described and exhibited in the free state, precisely in the form 
shown when in combination, this circumstance properly viewed lends assist¬ 
ance to the ideas involved. Eor it is in consequence of this same incomplete¬ 
ness that the group HO cannot appear free in that shape. It follows the 
rules which govern the construction of elementary molecules ; when liberated 
from other forms of combination it exercises its affinity upon another group 
like itself and makes its appearance as H O H O or (H 0) 2 . And so with 
other similar ‘radicles.’ 
It must not be supposed, however, that instances are wanting of unsaturated 
free molecules. A familiar example will be recognised in ordinary ammo- 
niacal gas (N H 3 )" ; this body and its analogues have a constant tendency to 
pass to a saturated state; not indeed by assumption of hydrogen, but by 
taking on HC1 (NH 4 C1), I 2 (NH 3 I 2 ), etc., according to circumstances. 
Stannous chloride (SnCJ 2 )" greedily absorbs Cl 2 and yields the stannic 
chloride, Sn01 4 . Carbonic oxide (CO)" combines readily with O or Cl 2 , pro¬ 
ducing respectively carbonic acid gas (C0)"0, or phosgene gas (CO)"Cl 2 , both 
of which are saturated. 
The rule seems to be, then, that the coefficient of replacing-power of a 
