January 12, 1893] 



NATURE 



249 



in the early years of the present century, with apparatus 

 in many respects inferior to that of the present day, and 

 with scarcely any aid from chemical manufacturers in 

 preparing pure materials and reagents, but with unsur- 

 passed manipulative skill and the most honest criticism 

 of his own work, produced the first fairly trustworthy 

 list of numbers representing the proportions by weight 

 in which the elements combine. Berzelius began work 

 in this direction in 1807, his attention having been 

 attracted by Richter's investigations ; but soon after- 

 wards he became acquainted with Daltons new atomic 

 theory of the nature of combination, and appears to have 

 been impressed with its great importance, and at the 

 same time with the need of more exact experimental 

 data for its support and development. The wonderful 

 accuracy of Berzelius's work generally is illustrated, as 

 Prof. Mallet points out, by the fact that his number for 

 oxygen, i6'o2i, becomes 15'894, almost exactly agreeing 

 with the latest determinations of the present day, if the 

 weighings of Dulong and Berzelius's three experiments on 

 the synthesis of water be corrected for the buoyancy 

 of the air. Since Berzelius many other chemists have 

 worked in the same field, but his most worthy successor 

 in such labours has undoubtedly been Stas. With greatly 

 better resources in the way both of apparatus and material, 

 with equal earnestness in seeking for the truth, with 

 equal intelligence and skill he took up the task which 

 became that of the largest part of his scientific life, and 

 for a more limited list of elements than Berzelius had 

 investigated, produced results of a degree of accuracy 

 which it is high praise to say would have delighted no 

 one more than Berzelius himself He aimed at the de- 

 termination with greater precision than any one before 

 him had attained of the atomic weights of some ten or 

 twelve of the elements. But by so determining these 

 constants he endeavoured also to settle several general 

 questions of fundamental importance in regard to matter 

 as studied by the chemist. 



Thus it has generally been assumed as true beyond 

 dispute since the early part of the present century, that 

 the mass of an atom of a given element is a constant 

 quantity. This has, however, occasionally been doubted, 

 and Stas himself considered the question as one requiring 

 examination. His researches, however, lend no support 

 to it. On this point Prof Mallet expresses himself 

 strongly in favour of the orthodox view. 



Assuming that the atomic weights are immutable values, 

 the question arises. Are they commensurable t This is 

 the much-discussed hypothesis of Prout, the origin and 

 development of which is very fully discussed by Prof 

 Mallet. A widespread feeling at one time undoubtedly 

 existed among chemists that Prout's hypothesis, that the 

 atomic weights of the other elements are integer mul- 

 tiples of that of hydrogen, if not true in its original form 

 would ultimately prove to be so at least in a modified 

 form. That Stas began his work under the influence of 

 this feeling is clear from his own words : — 



"Je le dis hautement lorsque j'ai entrepris mes re- 

 cherches, j'avais une confiance presque absolue dans 

 I'exactitude du principe de Prout." 



But his experimental results clearly contradicted the 

 hypothesis, and he satisfied himself that the atomic 

 weights of the elements which he determined with such 

 precision could not with truth be represented by integer 

 multiples of the atomic weight of hydrogen, or the half 

 or the fourth of this unit. In his own words : — 



" Aussi longtemps que, pour I'dtablissement des lois 

 qui r<5gissent la matiere on veut s'en tenir Texperience, 

 on doit considerer la loi de Prout comme une pure 

 illusion. La simplicitd de rapport de poids que pre- 

 suppose I'hypoth^se de Prout entre les masses qui 

 interviennent dans Taction chimique, ne s'observe done 

 point dans I'expdrience ; elle n'existe point dans la 

 r^alitd des choses." 



NO. 12 1 I, VOL. 47.] 



The great majority of chemists — Prof. Mallet remarks— 

 at the present day, are probably agreed in believing that 

 the hypothesis of Prout has been shown by Stas to be 

 untenable. But the fact that so many well determined 

 atomic weights, referred to hydrogen as unity present 

 numbers nearly apptoaching integers, is very striking and 

 calls for further investigation. Stas himself is quoted as 

 admitting this much. Prof Dewar, in the course of the 

 discussion after the paper was read, drew special atten- 

 tion to this question and gave several most striking 

 instances of the nearer approach to whole numbers which 

 resulted from a recalculation of the accepted values, using 

 the lower value for oxygen (15 "87) which so many recent 

 researches tend to support, although on the other hand, 

 of course, some of the values now near to whole numbers 

 are considerably thrown out. Evidently there is ample 

 opportunity for further experimental investigation of this 

 all-important problem, and it is impossible — notwith- 

 standing the extraordinary degree of accuracy attained 

 by Stas — to formulate any final conclusion. The supreme 

 interest attaching to the problem was clearly recognised 

 by Stas himself, as the following words show : — 



" Au point de vue de la philosophie naturelle, la 

 portce de I'idee de Prout est immense. Les elements 

 des corps composes que nous considdrons comme des 

 corps simples en egard a leur immutabilite absolue 

 pour nous, ne seraient eux-memes que des corps com- 

 poses. Ces elements, dont la dccouverte fait la gloire 

 de Lavoisier et a immortalise son nom peuvent etre 

 considerds ainsi comme ddrivant de la condensation d'une 

 matiere unique : nous sommes naturellement conduits a 

 I'unitc de la matiere, quoi qu'en realitd nous constations 

 sa plurality, sa multiplicity." 



This quotation is almost alone sufficient to show that 

 Stas was a philosophical chemist of the highest order, 

 and not a mere mechanical worker, as has sometimes 

 been supposed ; his unwearied attention to minutest 

 details has undoubtedly served to completely overshadow 

 the philosophical motives and aspirations by which he was 

 guided. 



Stas also endeavoured to obtain evidence with regard 

 to the possible dissociation of the elements at high 

 temperatures and to this end purified his materials with 

 every imaginable precaution. The skill with which he 

 carried out his operations is attested by the statement 

 made by Mr. Crookes,the chairman at the reading of Prof. 

 Mallet's paper, that he had seen in Stas's laboratory a large 

 mass of potassium chloride, v/hich Stas had been years 

 in preparing, and in which he had failed to find a trace 

 of sodium even spectroscopically — such an achievement 

 appears almost inconceivable to the chemist. Stas, in 

 fact, in the course of his work investigated the methods 

 of analysis to be used with a degree of rigour, and dis- 

 covered and applied refinements upon older methods of 

 experiment with a degree of patience and skill, such as 

 had never before been used in chemical investigation. 

 Only those who are thoroughly conversant with such 

 work can fully appreciate his labours ; they probably 

 will agree that owing to the multitude and diversity of the 

 precautions to be taken, his work is the most difficult 

 hitherto attempted, and that he stands unsurpassed 

 among all who have undertaken the execution of exact 

 physical measurements. 



A lengthy section of Prof. Mallet's paper is devoted to 

 the consideration of the objects to be aimed at and the 

 methods to be pursued in future work. He advocates 

 the repetition by competent hands of some one at least 

 of Stas's fundamental results, calling attention to Stas's 

 own emphatic expression of the wish that this should be 

 done. It is also most important that no distinction 

 should be made between rare and common elements, and 

 that the atomic weights of all should be determined with 

 the least possible delay and the highest attainable degree 

 of accuracy. Certain of the elements particularly calL 



