April 1, 1892.] 



KNOWLEDGE 



75 



combination are mathematically exact. For half a century ' 

 the scientific ivorld had accepted the dictum that the laws 

 of chemical combination were Ms mathematiques, but the 

 original experiments on which these laws were based were 

 far from being models of accuracy. This fact was admitted 

 by Stas. who undertook the laborious task of a re-examina- 

 tion of those laws, with a view to settle by the most exact 

 methods whether these laws were in fact of mathematical 

 exactness, or, like so many physical laws, only Jois limites 

 or approximate relations. Li 1865, five years after the 

 date of his first series of researches, appeared the 

 XoureUcs Rtcherrhes sur hs Lois ths Proportions Chimiijues. 

 In this work Stas repeated the more important of 

 his former determinations of atomic weights, ■with 

 additional precautions. He also subjected to the most 

 rigorous tests the laws of definite, constant, and 

 equivalent proportions which had hitherto rested on the 

 comparatively rough experiments of Dalton, Wollaston, 

 and other workers of the early part of the present century. 

 In this great work Stas confirmed, on the one hand, 

 his previous conclusion that Proufs hypothesis was 

 unsupported by experiment, but showed on the other that 

 the laws of chemical combination, hitherto accepted on 

 insufficient data, were, as far as experiment could prove, 

 actual and veritable mathematical laws. It is impossible 

 to over-estimate the benefit conferred upon science by a 

 man who has the courage to devote years of patient labour 

 to the re-examination of points such as this, and the re- 

 investigation of supposed laws which have been accepted 

 on the evidence of insuflicient experimental data. Such 

 work is much needed in the chemical world at the present 

 time, when a vast superstructure of theory is being built 

 upon a comparatively small number of approximate ex- 

 periments with regard to the behaviour of substances in a 

 state of solution. 



From the point of view of the working practical chemist 

 the most important aspect of Stas's researches is that 

 relating to ths preparation of chemical substances in a 

 state of purity. Since Stas's time chemists have not been 

 satisfied with the approximate purification of substances 

 which in general sufficed the earlier experimenters. The 

 approximate isolation or purification of substances is the 

 first step in a chemical research ; the complete purification 

 is the most difficult and the most important part of exact 

 research in the science. Stas's methods of purification 

 have served as a model for all subsequent experimenters. 

 In order to give a general idea of the character of his work 

 we will describe a method he adopted for the purification 

 of sUver, a substance which is, as he says, the " pivot " of 

 his determinations. Silver is a substance which, as Stas 

 showed, can be obtained in a state of almost perfect 

 purity. The way in which it resists oxidation, and the 

 distinctive character and insolubility of certain of its salts, 

 would lead one to suppose tliat its complete purification 

 would be very readily eflected. That this is not exactly 

 the case will be evident from the following description of 

 Stas's method. In order not to make the description 

 tmduly long, we omit the special methods of ptirifying the 

 reagents used in the work. These reagents are water, 

 nitric acid, hydrochloric acid, caustic potash, and mili:- 

 sugar. Each of these had to be submitted to special pro- 

 cesses of purification, lest their use should introduce foreign 

 substances into the silver. 



Coinage silver was taken, and dissolved in very dilute 

 nitric acid. Any gold present is left undissolved. The 

 solution of the nitrate is evaporated to dryness, and 

 heated till no more nitrous fu-nes are evolved. The 

 salt is then dissolved in a small quantity of water. On 

 filtering, any platinum present is left behind. The filtrate 



is then diluted with about thirty times its volume of water 

 and an excess of hydrochloric acid added. All the silver 

 is then precipitated or thrown down in the form of the 

 insoluble chloride of silver. Any copper and u-on present 

 remain in solution. The liquid is poured oflf and the pre- 

 cipitate washed, first with dilute hydrochloric acid and 

 then with water, tUl the washing appears to be pure water 

 containing no trace of copper or of hydrochloric acid. Tliis 

 washing of a large quantity of a precipitate is a very lengthy 

 and tedious operation, requiring days or weeks, according 

 to the quantity of the precipitate. The washing is efiected 

 in this case by shaking up the precipitate with water in a 

 stoppered flask, allowing the precipitate to settle, and 

 pouring off the liquid. All the operations with chloride of 

 silver were carried out in a room lighted by artificial Ught, 

 since daylight, as is well known, effects a chemical change 

 in the composition of chloride of silver. The chloride of 

 silver, purified as above, is brought on to a cloth (previously 

 washed with hydrochloric acid) and the water squeezed 

 out. After drying, the silver chloride is pounded fine in a 

 mortar, and reduced to the metallic state by warming for 

 forty-eight hours with a solution of caustic potash and milk 

 sugar (both carefully purified). The finely-divided metal 

 is then fused, with special precautions to prevent access of 

 impurities. By this process Stas hoped to obtain an ingot 

 of perfectly pure silver, but found that, besides very slight 

 traces of other substances, there remained an appreciable 

 quantity, 2 parts in 100,000, of silica. Experience con- 

 vinced Stas that no substance can be obtained absolutely 

 pure except by distillation. He therefore subjected the 

 silver obtained as above to the process of distillation from 

 one cavity to another in a hollowed block of quickhme, 

 made from white marble. The cavity having been pre- 

 viously heated by the oxy-hydrogen flame, in order to drive 

 off any volatile substances such as soda, the silver was 

 placed in the cavity and fused. No scum appeared on the 

 surface, showing the absence of certain impurities such as 

 iron, which imder these circumstances would form a slag. 

 The heat from the oxy-hydrogen flame was then increased 

 tiU the metal began to boil. The vapour had at first a 

 strong yeUow tinge, showing that sodium was still present. 

 This, however, soon disappeared, the\apom' of the silver 

 showing no colom- beyond a &int blue tinge. The absence 

 of any green tint showed that the substance was free from 

 copper. The metal having completely distilled into the 

 second cavity, or receiver, in the lime block, it was 

 found that absolutely no residue remained, the small 

 quantity of silica, and any similar fixed substance of 

 an acid character, having combined with the Ume, 

 and any oxidizable material having been burnt away by 

 the flame of the oxy-hydrogen blow-pipe. By the above 

 process Stas beUeved that he had obtained silver 

 absolutely pure. Subsequently, however, Dumas showed 

 that silver thus prepared absorbs, after distillation but 

 whde still molten, a certain quantity of oxygen which does 

 not combine chemically with the silver but remains 

 ' occluded "' in the metal. The elaborate precautions 

 adopted by Stas were therefore not successful in obtaining 

 even this well-known and characteristic substance in a 

 state of perfect purity, though he subsequently determined 

 the amount of oxygen present. But the practical chemist 

 owes to Stas a proper appreciation of the difficulties 

 attending the purification of substances, an appreciation 

 of the necessity for taking every means to overcome these 

 difficulties, and a knowledge of methods for the carrying 

 out of this class of work ; methods elaborated by Stas 

 thirty years since, and which yet form the basis of many 

 of the recent researches on the detenoination of atomic 

 weights. 



