June 17, 1904.] 



SCIENCE. 



911 



of electric heating to take off tlie 'peak of 

 the load,' so to speak, in our open-hearth 

 steel furnaces— that is, to furnish the last 

 few hundred degrees of necessary tempera- 

 ture while the combustion of gas furnishes 

 the lower range— is a distinct commercial 

 possibility. Already our steel works are a 

 network of electrical appliances for run- 

 ning cranes, charging machines, hoists and 

 cars, and the step is not a long one to em- 

 ploy this already-present agent to help 

 the heating gases over the heaviest part of 

 their work, the bringing up of the charge 

 to tapping heat. 



Next to iron in natural abundance is 

 calcium, and our chemists and metallurgists 

 are only beginning to appreciate its possi- 

 bilities. Occurring as almost chemically- 

 pure calcium carbonate, in inexhaustible 

 quantities and at the cost of only a few 

 cents a ton for quarrying, the question of 

 producing the metal cheaply is the partic- 

 ular task of the electrochemist. The cost 

 of the metal is practically the cost of its 

 reduction, and there is no doubt that the 

 eleetrometallurgist can and will solve this 

 question as he has that of aluminium and 

 silicon. Calcium is, at temperatures above 

 a red heat, the strongest metallic base ex- 

 isting, and is therefore the most powerful 

 pyrochemical reagent. By its use many 

 problems may find their solution, such as 

 the complete deoxidation of melted metals, 

 the reduction of rare elements and many 

 other interesting reactions. 



Magnesium does not occur quite so plenti- 

 fully as calcium, but still it is so common 

 that 99 per cent, of the present cost of ma- 

 king magnesium must be charged against 

 the process used and only 1 per cent, against 

 the raw material. There is no reasonable 

 doubt but that careful study put upon 

 the electrolytic production of magnesium 

 woiild result in its being produced at a 

 fraction of its present cost. It is certainly 

 a metal which, at its present price, has very 



limited uses, but, with a specific gravity 

 of only 1.72, a capacity for being hardened 

 and strengthened like aluminium and the 

 property of forming valuable alloys with 

 copper and with aluminium, it is certain 

 that its cheap production would mean an- 

 other metal added to those in everyday use. 



There is yet another metal of kindred 

 character worth considering. Beryllium 

 occurs in the gem beryl as silicate of alu- 

 minium and beryllium. The mineral is 

 found massive in large enough quantities 

 to form a commercial source of the metal. 

 The separation of the beryllium oxide from 

 the silica and alumina is not a very difficult 

 chemical operation, but could probably be 

 simplified by the application of Hall's 

 process of differential reduction in the elec- 

 tric furnace. The reduction of beryllium 

 oxide to metal dissolved in a fused bath of 

 alkaline and beryllium salt is a step which 

 would probably yield to investigation, while 

 the collecting of the metal floating upon the 

 bath should offer no greater difficulties than 

 does the collecting of sodium. 



With a brilliant white color, specific 

 gravity 1.6, malleable, ductile, forming fine 

 alloys, there are a large number of possible 

 applications for beryllium if it can be ob- 

 tained cheaply, and it is to the electro- 

 chemist that we must look for the solution 

 of this problem. 



In the electrolytic refining of metals, 

 copper was the first to yield commercial 

 results, silver next, then gold, lead and 

 bismuth. Yet there are others awaiting 

 conquest. The electrolytic refining of 

 nickel, zinc, antimony and tin has been 

 attempted, but not yet commercially mas- 

 tered; that of aluminium is an attractive 

 question because of the economy it would 

 produce. Instead of purifying by costly 

 chemical methods four tons of aluminium 

 ore, we have the alternative problem of 

 refining one ton of impure aluminium, and 

 with a large margin of difference in com- 



