458 Dr. W. W. J. Nicol on 
tion thus obtained was cooled, crystals slowly separated out 
at the air temperature. Some crystals thus obtained were 
drained as completely as possible from the oily mother-liquor, 
and then removed and pressed between filter-paper ; in spite 
of the draining they were found to be imbedded in a mass of 
the solidified hydrated salt. On heating, the original crystals 
did not change form, and remained nearly transparent. The 
total loss of water on the mixture was 17 per cent., corre- 
sponding to Na,8,03 2aq. The crystals cannot therefore con- 
tain more than one molecule of water; and in view of their 
transparency and not changing in form on heating, it appears 
probable that they were anhydrous. The crystalline form was 
an obtuse rhombohedron. I have not been able to find any 
account of a salt containing less than five molecules of water. 
A solution of sodium sulphate, which was distinctly super- 
saturated at 20°C., readily dissolved a considerable quantity 
of the anhydrous salt when introduced as above, even at 
15°C. Another solution deposited anhydrous salt at 30° C., 
which redissolved on cooling to 20°C.; while Na,SO, 7 aq. 
crystallized out on cooling a few degrees. These crystals, 
however, quickly dissolved when the bottle was warmed in 
the hand. The most remarkable feature of these experiments 
was that in no instance was any caking of the anhydrous salt 
observed, as would have been the case had it combined with 
the water to forma hydrate before dissolving ; it is, of course, 
evident that no trace of the normal hydrate was in any case 
produced, for this would have determined the crystallization 
of the supersaturated solution. 
The only experiments in any way resembling those above 
described that [ have been able to find an account of, are 
those of Coppet and Thompson. 
Coppet* describes the manufacture of supersaturated solu- 
tions of sodium sulphate and other salts by the addition of 
anhydrous or partially dehydrated salt to water. He found 
that, in order to succeed, it was necessary to exclude unfiltered 
air and to add the salt in exceedingly small quantities at a 
time. With sodium sulphate he found that he could thus 
obtain a solution containing nearly five times as much anhy- 
drous salt as is contained in a solution prepared from the 
decahydrated salt at the same temperature. In this case the 
addition of the anhydrous salt must have been so slow, that 
when the ordinary saturated solution was produced no deca- 
hydrated salt remained undissolved; or it is quite possible 
that no decahydrated salt is formed by the addition of anhy- 
drous salt to any but very dilute solutions. 
* Comptes Rendus, 1871, vol. Ixxiii. p. 1824. 
