656 
POPULAR SCIENCE REVIEW. 
lined with clay to prevent the fluid from escaping. The nitro- glycerine is 
first poured in ; then water, which floats on its surface ; and there is no 
necessity for tamping. Ignition is effected by means of a slow match, having 
a percussion cap at one end. The explosive efiect of this compound has 
been proved by experiments to be very great. In one instance a perpen- 
dicular hole, in. in diameter, was bored 14 feet from the outer side of a 
dolomitic rock, and was tamped at the depth of seven feet, as its lower 
extremity was found to communicate with a cavity filled with clay. A pint 
and three-quarters of nitro-glycerine was poured in, and filled up five feet 
of the hole. The result of the explosion was the production of two fissures, 
one 50 and the other 20 feet long— an immense mass of rock being separated 
from the main body. In another experiment, a hole was made of the same 
diameter and depth, at 10 feet distance from the face of the rock. The 
charge was a pint and a half, and filled about two feet of the hole, which 
happened to be enlarged, from natural causes, in one place. Sand was placed 
above the nitro-glycerine. The rock was rent, and a large portion of it 
shattered ; the whole amount thrown down being about 3,500 cubic feet. 
The gases produced by gunpowder would have been lost in the cavities. 
The Preparation of Oxygen from Binoxide of Manganese. — M. Carlevaris, 
to obtain its oxygen from the peroxide of manganese, mixes it with siliceous 
sand, and exposes the mixture to a suitable temperature ; 
M,0,+Si0, = Si0„ MnO-fO. 
The evolution of oxygen commences as soon as the temperature of redness is 
reached ; and with a material containing about 90 per cent, peroxide, about 
4,000 cubic feet of oxygen per ton will be obtained. Hence, taking into 
account the cost of the material, the fuel, and the manipulation, the price of 
the gas will still be very moderate. 
Soiv to detect Nitro-henzol in Oil of Bitter Almonds. — A method for the 
detection of the above-named impurity, based upon a knowledge of the rela- 
tive actions of sodium, alcohol, and nitro-benzol, has been suggested by Herr 
Dragendorff. A clean piece of sodium placed in pure oil of bitter almonds 
causes a slight evolution of gas, and the metal becomes covered with a white 
flocculent coating. If alcohol is present, the evolution of gas is greater, and 
the white coating falls away from the metal. Pure nitro-benzol similarly 
treated with sodium gives off a little gas, darkens but very slightly, and no 
white flocculi separate. But if a very small amount of alcohol is present 
with the nitro-benzol, an active evolution of gas takes place, and the liquid 
very quickly becomes of a dark brown or even black colour. Ten per cent, 
of alcohol can easily be recognized in this way. To detect nitro-benzol in 
bitter almond oil, Dragendorff adds four or five drops of alcohol to ten or 
fifteen drops of the suspected oil, and places a piece of clean sodium in the 
mixture. If nitro-benzol is present, the metal becomes covered with a yelloAV 
or brown coating, the colour being darker as the proportion of nitro-benzol is 
greater. — Zeitscht. fur Analyt. Chem., No. 4. 
Itelative (Composition of Green and Dried Tobacco Leaves. — Much as has 
been written upon the subject of tobacco and its composition, there is still 
considerable doubt as to the relative chemical composition of the fresh and 
prepared leaves. Hitherto it has been believed that the essential principle 
