October 2, 1890] 



NATURE 



539 



and perhaps also that of Rogers ; the prodflcts of the 

 action are aluminium and chlorine or fluorine. To the 

 second class may be ascribed the process of Kleiner, 

 perhaps that of Rogers, and that of Bernard Brothers ; the 

 products of the action are the same as the first class, but 

 the aluminium fluoride destroyed by electrolysis is in part 

 restored to the bath "by causing the fluorine vapours 

 evolved to act on alumina or beauxite placed somewhere 

 about the anode." To the third class may be ascribed 

 the processes of Henderson, Hall, and part of Heroult's 

 patented process ; here the products are said to be 

 aluminium and oxygen, which by contact with the carbon 

 anode is converted into carbon monoxide or carbon 

 dioxide, and the cryolite, &c., used as solvent for the 

 alumina are said to remain unchanged. The fourth class 

 contains only Heroult's process, and of that only the 

 latter half of his claims. The fifth class contains the 

 process of Monckton, Cowles, Menges, and Farmer. 



As regards the details of these various processes re- 

 ference must be made to the book. The processes 

 of Cowles, Hall, and Heroult are reported as being in 

 active and very extensive use by the several companies, 

 and if one is to believe the glowing reports that are pub- 

 lished they are very successful indeed ; thus Hall claims 

 to extract 50 per cent, of aluminium from alumina, instead 

 of the theoretical 52-94 per cent, while the fluorides used 

 waste only very slightly, and require replenishing to the 

 extent of a small fraction of the weight of the metal 

 made ; and with his latest improvements aluminium is 

 not to cost more than half a dollar a pound ! 



The scientific investigation of these processes is either 

 kept secret, or, alas, has scarcely been attempted ; and 

 yet the surest and quickest way to establish a process on 

 a sound commercial footing is to thoroughly investigate 

 the conditions regulating every reaction, and not merely 

 those conditions relating to the principal reaction, for 

 those relating to the ubiquitous " impurity " are at least 

 of equal, if not of greater, importance. In describing 

 the above processes, the author introduces scientific and 

 numerical discussions on several points ; but the work 

 would have been more valuable to the increasing number 

 of metallurgists interested in the subject if the book had 

 bristled more with hard facts expressed in figures, and 

 with references to volume and page where the original 

 might be found. 



Chapter xii., of 31 pages, is a summary of the very 

 many processes that have been proposed for the "' Reduc- 

 tion of Aluminium Compounds by other means than 

 Sodium or Electricity." Many of the accounts record the 

 partial success of actual trials, and deserve considera- 

 tion ; but many are but little more than written hopes 

 and imaginations. 



As far as the end of this last chapter, the subject-matter, 

 with the exception of chapters iii. and iv., is purely 

 chemical, and relates, indirectly or directly, to the primary 

 production of aluminium or of certain of its alloys. 

 From this point the book deals with the manner of work- 

 ing aluminium, the preparation of its alloys, and the pro- 

 perties which characterize them. This metallurgical part 

 of the book may be considered as being introduced, as 

 far as aluminium itself is concerned, by chapters iii. and 

 iv. and some four pages of chapter xiii., which describe 

 the " Purification of Aluminium," and refer very briefly J 

 NO. 1092^ VOL. 42] 



to Mallet's preparation of the pure metal. The special 

 methods found suitable for the analysis of aluminium 

 and its alloys are described in the last chapter of the 

 book. 



Chapter xiii., of 29 pages, describes fully the methods 

 of working in aluminium — casting, rolling, annealing^ 

 soldering, &c., &c. In speaking of the uses of aluminium, 

 the author says, when referring to its lightness, 



" but I would say a word or two about the popular fallacy 

 of aluminium replacing steel as a constructive material, 

 ... or in any position where its strength is of import- 

 ance, ... it is forgotten that it is only one-third as 

 strong." 



The aluminium alloys are considered in the next three 

 chapters. Chapter xiv., of 30 pages, describes many alloys, 

 of which the following two classes are especially important, 

 as they promise to enter largely into commerce. The 

 alloys with copper and nickel mostly contain but a very 

 small proportion of aluminium, but nevertheless are 

 superior to ordinary German silvers for strength and fine- 

 ness of grain. Those containing copper and zinc, and 

 known as aluminium brass, possess exceedingly valuable 

 working qualities, are three and four times as strong as 

 ordinary brass, and containing mostly only 2 or 3 per cent, 

 of aluminium are further recommended by their low cost. 

 Chapter xv., of 32 pages, describes the alloys with copper ; 

 of these the most important contain 5 or at most 10 per 

 cent, of aluminium, and are known as aluminium bronzes ; 

 and full accounts of the methods of working and tests of 

 the strengths of the metals are given. Chapter xvi., of 

 31 pages, describes the "Aluminium-Iron Alloys," and 

 is a very interesting account of a difficult but ex- 

 ceedingly important subject. The chapter is divided 

 into three parts, dealing with the effects produced by 

 adding trifling quantities of aluminium to steel, to 

 wrought-iron, and to cast-iron ; in almost all cases the 

 castings are quite free from blow-holes ; and in certain 

 cases the metal becomes more fluid, allowing of castings 

 being more readily made. Cast wrought-iron sounds like 

 a paradox, but it is not one, for, by adding a small amount 

 of aluminium to wrought-iron that has been heated 

 until it has become pasty, the latter immediately 

 liquefies, and can then be poured into moulds, making 

 castings as sound as if they were of grey cast-iron. The 

 author discusses at some length the probable explana- 

 tions of the effect of adding aluminium to the various 

 kinds of iron, and his conclusions may be very briefly 

 stated as being : (i) addition of very small quantities of 

 aluminium, i.e. o'oi to o*i per cent., causes the destruction 

 of carbonic oxide or dioxide, or of the oxygen compounds, 

 as oxide of iron, disseminated mechanically, and which 

 at the moment of setting give rise to the formation of 

 these gases ; hence the cast metal is free from blow- 

 holes, and, owing to the removal of suspended oxides, the 

 metals cease to be pasty and become quite fluid ; (2) 

 addition of aluminium in larger quantity, i.e. 0*2 to 0*5, or 

 even several per cents., converts the combined carbon — 

 that is, if there be any appreciable amount — into graphitic 

 carbon, and, according to the quality of the iron operated 

 on and the amount of aluminium added, has the effect of 

 rendering the castings free from a chilled surface, of 

 making the metal very uniform in texiure and hardness, or 



