200 
rorULAR SCIENCE REVIEW. 
a reciprocal activity. As, however, tlie energy consumed by the first 
decomposition cannot be again produced, the pristine sugar cannot be repro- 
duced. In place of this the products suffer a new and deeper decomposition 
into alcohol and carbonic acid. Berthelot examined the condition of the 
simultaneous hydrogenation and oxidation of the sugar. A battery of six or 
eight Bunsen cells was employed, the poles of which were connected with 
an oscillating commutator which could be alternately positive and negative 
twelve to fifteen times in a second ; and the electrodes were cylinders of pla- 
tinum black. This apparatus, when immersed in acid water, evolved at 
each of the poles hydrogen and oxygen alternately. The apparatus was so 
regulated that it evolved no gas, the water being reformed at the moment 
after its decomposition. When placed in a watery solution of glucose, which 
was in some instances neutral, in others acid, and in others alkaline, alcohol 
was decomposed but only in very small quantities (some thousandths), the 
greater part of the glucose resisting the decomposition. So limited a change 
does not justify one in drawing conclusions. The production of alcohol in 
the cold, however, and by means of sugar, is a fact of great interest. — 
Compt. rend., lxxxvii., 949. 
The Chemical Composition of Sea-water . — Jacobsen has set himself the 
task of deciding the question whether the composition of sea-water taken 
from different seas and oceans, and different depths, possessed the same com- 
position, and whether the discrepancies observed in analyses were due to 
errors of manipulation. For this purpose he examined the composition of 
forty-six specimens of sea- water, collected on board the “ Gazelle ” during 
the expedition of 1874-1876, for every possible locality and depth. The 
constituents which were determined were chlorine, sulphuric acid, and 
calcium carbonate. The chlorine showed only a very slight variation ; the 
salt corresponding to the chlorine amounted in the highest case to 1-8140, in 
the lowest case to 1-8047, the mean being 1*80936. The chlorine was 
determined in fifteen specimens. When it is remembered that these results 
are influenced by the unavoidable errors of chlorine determinations and the 
determination of salt, one will not be disposed to ascribe to the found 
irregular variations any significance of weight, but will not hesitate to say 
that the relative amounts of chlorine contained in oceanic waters show no 
considerable variation. The sulphuric acid was determined in 166 specimens 
of water. It constituted in the mean 6-493 per cent, of the entire salt 
present ; the greatest difference (0 35 per cent.) lay between the maximum 
6‘69 per cent, and the minimum 6*34 per cent. The author remarks that here 
again the variation would be less if the unavoidable error of the areometric 
determination of salt could be eliminated. There are grounds, however, for 
believing that the amount of sulphuric acid present in water is somewhat 
less constant than the amount of chlorine. On the other hand, attention 
must be directed to the fact that any regular variation in the properties of 
sulphuric acid, depending on the place or the depth from which the water 
has been taken, was not observed. The determinations of calcium carbonate 
were made in thirty-nine samples of water. The mean result was in 10,000 
parts of water 0*269 parts of lime carbonate, the maximum being 0*312 parts, 
and the minimum 0-220 parts. So far from referring these variations in the 
results to differences in the sources whence the waters were taken, or regarding 
