sepia, | 
¢ 
F. W. Clarke— Volume of Water of Crystallization. 481 
tion, the salts themselves were to undergo a change olume, 
it Is evident that, in such a variety of compounds, we could 
the remainder. We might, indeed, get similar remainders for 
a series of salts of equal hydration, but we could certainly hope 
for nothing of the kind in ‘comparing compounds with one, two, 
our, five, six, seven, ten, twelve, and eighteen molecules of 
water of crystallization. 
1,0,, 4-487, Ditte (74-4). 1,0,,H,O (HIO,), 4-269, Ditte (82-4), 
K,0, 2°656, Karsten (35:4). 
(548). CaO, 3-180, Filhol (17-6). CaO, HO, 2-078, Filhol 
; , Filhol (22-4) SrO, H,O, 3625, Filhol 
(33°5). BaO, 5-456, Filhol (28-0). BaO, H,O, 4-495, Filhol 
(38). Mn,O, (braunite), 4°752, Rammelsberg (33-2). Mn, Og. 
H,0 (manganite), 4-335, Rammelsberg (40°6). Fe,0,, 5°037, 
H. Rose (31-7). O,, H,O (githite), 4°37, Yorke (40-7). 
1,0, (sapphire), 4-0001, Schaffgotsch (25-7). eg Hy 
(diaspore), 3-45, J. L. Smith (35-0). B,O,, 1°803, Davy (38°8). 
B,0,, 3H, 0, 1-4347, Stolba (86-4). 
Arranging these serially and subtracting, we get the follow- 
h r 
_ Mg remainders to represent the wa 
In [,0,, H,0, 8°0 In Mn,0,, H,0, 74 
~ K,O0, HO, 19°4 H Gees Hass 9°0 
“ CaO, H,0, 17% “"A),0,, 1,0. 9°3 
“ SrO, H,0, Yes at, ets, 15°9 
“ BaO, H,0, 10-0 
It would be easy to carry this out still farther, but alread 
of uniformity is striking enough. One compound, 
however, is worth noting. The molecular volume of K,O, H,0, 
halved as it ought to be for KHO, is almost the exact mean 
between the values for ice, 19°6, and K,O, 354. This, of 
Course, is what we should expec 
we would naturally expect, viz., that when an oxide unites 
With water to form a hydrate, both undergo condensation. 
ondly, what supports the same idea, that the volumes thus 
bitrarily calculated for water average much lower than the 
values obtained from crystallized salt. 
