458 



NA TURE 



[March 17, 1898 



volume is provided with a good index, and old-fashioned 

 people who might be disposed to look for copper, silver 

 and gold elsewhere will have no difficulty in finding a good 

 and clear account of them, notwithstanding that they 

 are sandwiched in between the metals of the alkalis and 

 the metals of the alkaline earths. 



The volume concludes with the history of those two 

 remarkable substances argon and helium, and winds up 

 by pointing out that " the existence of two substances 

 having the properties of argon and helium is not in any 

 way indicated in the periodic classification of the elements, 

 and it is probable that their discovery will lead to modi- 

 fications of that hypothesis," a statement with which 

 even the most fanatical of periodic chemists will find it 

 difficult to disagree. 



One is tempted to think that, if we were not so 

 familiar with it, hydrogen is the element upon which, 

 in connection with any system of classification, wonder 

 must be concentrated. Undoubtedly a positive element, 

 similar in general chemical character to a metal, yet its 

 volatility removes it far from any acknowledged metal, 

 and its low atomic weight separates it in the periodic 

 scheme from all the other gaseous elements. Its nearest 

 allies are the metals of the alkalis, and towards them it 

 stands, so far as physical character is concerned, much 

 in the same relation as nitrogen to arsenic and antimony 

 or oxygen to selenium and tellurium. And even as to 

 chemical relations its position is analogous to that of 

 nitrogen or of oxygen, notwithstanding that among the 

 negative elements chemical activity generally diminishes 

 as atomic weight increases ; while in the series Li, Na, 

 K, Rb, Cs, chemical activity increases with atomic weight. 

 The question of the molecular weights of the metals 

 is one of great interest. It has been known for a long 

 time, from the vapour densities of mercury, cadmium, 

 and zinc, that the molecules of these metals are mon- 

 atomic, while the semi-metal arsenic forms in the state 

 of vapour a molecule composed of four atoms. Potassium 

 and sodium are unmanageable, and no other metal is 

 volatile enough to afford by this process any trustworthy 

 results. But attempts have been made to estimate the 

 molecular weights of a great many metals by observ- 

 ation of the depression of the freezing points of certain 

 fusible metals— for example, of tin, lead, or bismuth 

 (Heycock and Neville), or of the lowering of the vapour 

 pressure of mercury (Ramsay) by solution in these fluids 

 of determinate quantities of the metals under con- 

 sideration. The general result of these experiments tends 

 to the conclusion that in nearly all such cases the mole- 

 cules are composed of one atom only. There is evidently 

 room for further experimental inquiry in this direction, 

 for there seems at present to be some conflict of testi- 

 mony. The molecule of arsenic in the vaporous state 

 is undoubtedly AS4, and its chief oxide is also admittedly 

 As40(i. In like manner the molecule of antimonious 

 oxide is Sb^Og, and by analogy one must conclude that 

 the vapour of antimony, although not actually observable, 

 must be tetratomic or Sb4. When these vapours con- 

 dense, either to the liquid or solid state, it would be 

 contrary to all experience in other cases to suppose that 

 they dissociate, and that the four atoms part company 

 from each other. Yet this is what the results derived 

 NO. 1 48 1, VOL. 57] 



from freezing point or vapour pressure experiments would 

 lead to. There is also the curious fact that the theory of 

 Dulong and Petit, which states that specific heat in the 

 solid elements is inversely as the atomic weight, points to 

 the atom, and not any congeries of atoms, as the physical 

 unit in the solid. So that, for example, arsenic and 

 cadmium, so different as regards vapour density, agree 

 in atomic heat ; thus — 



Specific Heat x Atomic Weight = Atomic Heat 

 Arsenic -0814 (Regnault) 744 = 605 



Cadmium "0542 (Kopp) iii'3 = 603 



Probably in the solid the molecules settle down into 

 some tactical relation to one another, in consequence of 

 which the atoms of neighbouring molecules are brought 

 as near together by cohesion as the atoms in any one 

 molecule are by chemical attraction. 



Turning again to the volume before us, there is much 

 that must be interesting to readers of all classes, and 

 especially in connection with the metallurgy of such 

 metals as sodium and aluminium, concerning which .Sir 

 Henry Roscoe is in a position to give trustworthy in- 

 formation. As is well known, sodium was first obtained 

 in 1807 by Sir Humphry Davy, who got it by the electro- 

 lysis of caustic soda. Subsequently Gay-Lussac and 

 Thenard succeeded in decomposing that compound by 

 means of metallic iron at a high temperature, but up to 

 quite recent times it has been usual to manufacture it by 

 distilling a mixture of dry sodium carbonate and charcoal, 

 the metal , passing over accompanied by the escaping 

 carbonic oxide. But even in well-conducted operations 

 the yield of metal corresponded to not more than about 

 one-third of the theoretical amount. 



A great improvement was introduced by Mr. Castner 

 some fifteen years ago when he replaced sodium car- 

 bonate by caustic soda, and employed an intimate mixture 

 of iron and carbon as the reducing agent. It is now 

 interesting to learn that he has accomplished another 

 great advance which consists in adapting the original 

 method to manufacturing purposes, and we are told that 

 whereas — 



"in the year 1807 Davy, with his battery of 100 cells, 

 found it impossible to produce the effects of decomposition 

 on pieces of soda of more than fifteen or twenty grains 

 in weight, the process has now been so amended that the 

 plant at the works of the Aluminium Company, Limited, 

 at Oldbury, manufactures about five tons of sodium per 

 week." 



What becomes of so large a quantity of this strange 



metal, considering that the .Aluminium Company no 



longer makes aluminium, and sodium is no longer used 



for this purpose anywhere, is a question which naturally 



arises. And the answer is supplied pretty well by the 



booL Sodium is used for the manufacture of cyanides, 



the consumption of which in the process of extracting 



gold from poor ores and tailings must be now enormous.| 



it is also used for the manufacture of sodium peroxide, 



for the production of metallic magnesium, and in the 



preparation of certain complex carbon compounds, of 



which the most important commercially is antipyrine. 



The history of the metal aluminium has also many 



I points of interest. The 'discovery of the metals sodium 



1 and potassium by Davy placed in the hands of chemists 



