JUNE 19, 1902] 
rich crimson as the sun sank behind the lofty range of moun- 
tains west of the Albert Nyanza, and, contrasted with the deep 
green foliage of the river banks, the scene was striking. But 
by far the most superb sight was the fall itself under the light 
of a full moon that night. For the identification of the rock 
specimens from the falls, consisting of biotite gneiss, mica 
schist, garnetiferous mica schist, and quartz, I have to thank 
Mr. G. T. Prior, of the Natural History Section of the British 
Museum. C. STEUART BETTON. 
THE MANUFACTURE AND USES OF SCDIUM. 
THE manufacture and uses of metallic sodium is the subject 
of an interesting article by Mr. James .D. Darling in the 
January number of the Journa/ of the Franklin Institute, from 
which we take the following facts. 
NATURE 
189 
melting point (800° C.) of the chloride and to its corrosive action 
when in the molten state. 
The process introduced by Mr. Darling involves the electro- 
lysis of sodium nitrate with the liberation of sodium and of 
nitrogen peroxide, which is then converted into nitric acid. 
The decomposition cell consists of a cast-iron pot set in a 
brick furnace. At the bottom of the pot is a 6-inch layer of 
refractory insulating material, and on this rests a cup 30 inches 
high, 16 inches outside diameter, with walls 4 inches thick. This 
cup is made of two sheets of perforated steel, between which is 
a mixture of ground deadburned magnesite and Portland cement 
which has been mixed with water and allowed to set hard. The 
space between the cup and the pot is filled with sodium nitrate 
(M.P. 313°C.) and the cup itself with melted caustic soda 
(M.P. 320°C.). The cast-iron pot acts as the anode, and 5 per 
cent. of the current is advantageously shunted through the metal 
walls of the cup. The kathode consists of a short length of 4-inch 
° Fic. 1,.—General View of Sodium Furnaces. 
From the time of its isolation in 1801 until 1858, the cost of 
sodium was exceedingly high. In 1858, Deville perfected the 
process of manufacture by heating sodium carbonate, chalk 
and coal in an iron retort and condensing the sodium vapour 
thus produced. The price then fell from 2000 francs to 10 
francs per kilo. No further important advance in the manu- 
facture was made until the late Mr. Castner took up the subject, 
and after using with great advantage a modification of Deville’s 
process in which carbon was replaced by a compound of carbon 
and iron and carbonate of soda by caustic soda, he succeeded 
in 1890 in making Davy’s original method available on the large 
scale, that is to say, the method of decomposing fused caustic 
soda by the electric current. Most of the sodium used to-day is 
made by this process. 
Attempts to make sodium from its cheapest compound, the 
chloride, have so far been unsuccessful, owing chiefly to the high 
NO 1703, VOL. 66] 
| wrought-iron pipe reaching nearly to the bottom of the cup. 
| Each furnace takes a current of about 400 amperes at an average 
| E.M.F. of 15 volts, and external heat is used only when starting 
up or when changing the cups, which have a life of 425 to 450 
hours. When the current is passed, nitrogen peroxide and 
| oxygen are liberated at the anode and escape through a hole in 
the cover of the pot. Sodium is liberated at the kathode, and 
rises to the top of the cup, where at intervals of an hour it is 
dipped off with a spoon and preserved under mineral oil. _ 
The aim of this new process is to decompose sodium nitrate 
in such a way that the sodium is likerated ina medium which 
will not oxidise it, and to get the nitrogen peroxide for the 
manufacture of nitric acid. How this is done will be evident 
from the description just given; the sodium ions of the fused 
nitrate travel through the walls of the porous cup to the fused 
caustic soda ; they act upon the caustic soda until it is converted 
