116 Progress in Science. [January, 
amalgams thus obtained is quoted ; as instances, we mention the following :— 
Hg, 98370; K, 17560; Na, 0-036. Hg, 98420; I, 1303) Na, cosmo: 
Hg, 98,044; K, 0934; Na, 0641. Potassium amalgam—formula, K,Hg2,; 
in percentages—Hg, 98°41; K, 1°59. Sodium amalgam, NazH¢gy2, containing 
1°88 per cent sodium. 
Dr. Berthelot, who has worked long and successfully on the chemistry of 
the different varieties of carbon, has now treated on the properties of the 
carbon met with in the Cranbourne (near Melbourne, Australia) meteorite. 
This carbon must be considered as having been first in state of solution in 
molten iron, and to have separated therefrom on cooling. Next, the carbon 
obtained from oxide of carbon, by the decomposition of that gas by iron at a 
relatively low temperature, is considered. From the reactions of these 
substances, the author comes to the conclusion that native graphite, about 
the origin of which very little is known, is certainly not, at least as far as 
experiments can throw light on this point, derived from either anthracite or 
from decomposed masses of meteoric iron which contained carbon. 
The presence of milk sugar in a juice of vegetable origin has been demon- 
strated by Dr. G. Bouchardat, who has examined a specimen of sugar obtained 
(1837) from the juice of the Achras sapota, at Martinique, the specimen being 
preserved among the objects known as belonging to the Matiére Médicale de 
Mérat. By treating a portion of the sample of the sugar alluded to with 
boiling alcohol at go per cent, there was left undissolved a substance which, 
on further investigation, was found to possess all the physical and chemical 
properties of milk sugar. A careful quantitative analysis of the sample 
established the fad that it contains 55 per cent of cane sugar and 45 per cent 
of milk sugar. This latter variety was also found to exist in the ripe fruit of 
the same tree recently brought from Egypt and analysed by the author. 
As a reagent for alcohol, Dr. Berthelot recommends benzoic chloride. 
When this compound is put into conta with water it is only very slowly 
decomposed, but if the water contains any alcohol (even as little as 1 in 1000 
parts of water) benzoic ether is at once formed; this ether is set free by a 
single drop of aqueous solution of caustic potassa, the odour of the ether 
being very peculiar and prominent. 
From a lengthy memoir on the nature of the sea-water along the coast of 
Bohuslan (Sweden) by Professor J. L. Ekman, we quote the most interesting 
point—viz., that, as regards the quantity of salt contained in the sea-water on 
the west coast of Sweden, there is greater difference than for any other now 
known sea. The coast alluded to is, on the one hand, washed by the North 
Sea, and on the other by the Baltic. In the more northerly part the water at 
the surface contains rather less than 2 per ¢ent salt, at 60 feet depth 2°5 per 
cent, at go feet depth 3 per cent; in the more southerly portion the discre- 
pancies are greater, but at 600 feet depth 3°5 per cent salt is met with. The 
average quantity of salt of the oceans is 3°44 per cent; while in the Atlantic, 
from the equator to from 55° to 60°N. latitude, the quantity ofsalt in the surface- 
water is 3°606 per cent, and at depths of from 500 to 10,000 feet 3°578. 
LIGHT. 
Conversion of cane sugar in the state of solution into glucose under the 
influence of light has now been proved by E. M. Raoult. The author placed, 
on May ro last, a concentrated solution of sugar in water in glass tubes, and 
sealed the tubes while boiling; these were placed close to each other in the 
same locality and under identically the same conditions, with the exception 
that one of the tubes was kept completely in the dark, the other tube being 
exposed to bright daylight. On October 20 the tubes were opened, and the 
contents examined. The solutions were perfectly clear, and did not, on being 
microscopically examined, exhibit the least trace of vegetable matter. The 
fluid in the tube exposed to light yielded an abundant red-coloured precipitate 
with the cupro-potassic reagent, thereby indicating the presence of glucose, 
while the contents of the tube kept in complete darkness did not manifest that 
reaction at all. 
