NATURE 



{^Nov. 4, 1880 



other without which no great scientific work can be 

 produced, viz., imagination. 



The work is divided into two books : the first, and 

 most valuable, treating of "Atoms," the second of 

 "Atomicity." The historical' introduction is very full, 

 and remarkable for the clear exposition of the work of 

 Ricbter, which was of so much importance in the subse- 

 quent development of the doctrine of atoms. The error, 

 which is still fallen into in some books, of attributing the 

 " law of proportionality '' to Wenzel is pointed out and 

 corrected. 



Full justice is not done to the work of Avogadro, on 

 which, confirmed as it has been by physical evidence, 

 rests the structure of modern chemistry. The distinction 

 between "integral molecules" and "elementary mole- 

 cules" was clearly stated by Avogadro in iSii, three 

 years before the date of the publication of Ampere's 

 letter to the Comte Berthollet. Ampere's attempt to 

 extend the hypothesis to facts concerning crystalline 

 bodies cannot be regarded as an improvement on the 

 simpler conception of Avogadro. 



But throughout this work there is a manifest resolve to 

 abate no jot nor tittle of that assertion, which, made with 

 the plenary knowledge of a chemical Philistine, sounded 

 the keynote of M. Wurtz's well-known " History of the 

 Atomic Theory." 



The statement on p. 42 of the reasons for adopting H., 

 as the standard of molecular weights is neither clear nor 

 satisfactory. The student might readily suppose that 

 this standard is adopted simply for the sake of conveni- 

 ence ; he might also be led to regard the statement 

 of Avogadro's law, on this page, as a deduction from some 

 vaguely-expressed relations between the number of atoms 

 in elementary molecules and the volumes occupied by 

 these atoms. 



Few text-books make clear the fundamental deficiency 

 of the Daltonian theory, viz., the absence of any trust- 

 worthy means for determining the weights of the "atoms'' 

 (or as we now say, molecules) of compound bodies. 

 Dalton, and Berzelius after him, laid down rules for deter- 

 mining these weighrs, but the rules of both chemists were 

 wholly empirical. "The atomic weight of an element is 

 the smallest amount of that element which combines with 

 unit-weight of hydrogen to form an atom of a compound," 

 but so long as the " atom " of the compound was undefined, 

 the atomic weight of the element could not be determined. 



Avogadro furnished chemists with a means of deter- 

 mining the molecular weights of all gasifiable bodies; and 

 in modern chemistry determinations of molecular weights 

 of many compounds of a given element, and analyses of 

 these compounds, must precede the determination of the 

 atomic weight of the element itself. 



The atomic weight of an element is the smallest amount 

 of that element— referred to hydrogen as unity— contained 

 in the molecule, that is in two gaseous volumes of any 

 compound thereof. For lack of a clear differentiation 

 between atom and molecule, and for lack of a definite 

 statement of how atomic weights are determined, the full 

 and valuable table, extending from p. 104 to 109, loses 

 much of its meaning. This table, by the way, very closely 

 resembles a table which occurs in Lothar Meyer's " Die 

 modernen Theorieen"; the alterations made by M. Wurtz 

 certainly do not add to the value of the table. 



Dalton's objections to the generalisation of Gay Lussac, 

 that "equal volumes of gases contain equal numbers of 

 atoms," was, as we now know, perfectly justifiable, but 

 on p. 35 Dalton is said to have repudiated "the solid sup- 

 port which the great French chemist gave to his ideas." 



Gay Lussac's generalisations could not be true, said 

 Dalton, because of such a reaction as that between 

 nitrogen and oxygen, wherein equal volumes of each 

 combine, and the product, nitric oxide, measures twice 

 the volume of either ; that is, there are, according to Gay 

 Lussac, twice as many atoms of nitric oxide as of oxygen 

 or nitrogen ; but as elementary atoms are chemically 

 indivisible, this is impossible. Berzelius obviated, or 

 rather shirked, the difficulty by applying Gay Lussac's 

 generalisations to elementary gases only, but a full 

 reconciliation between the views of Dalton and those of 

 Gay Lussac was only possible when Avogadro's fruitful 

 idea of the existence of molecules as distinct from atoms 

 was fully recognised in chemical science. 



In describing the physical methods for checking atomic 

 weight determinations, the law of Dulong and Petit is 

 stated in too absolute a manner ; if the data concerning 

 the specific heats of the elements are carefully considered, 

 it is evident that in many cases the value varies very 

 much with temperature, that in others no direct determi- 

 nation of specific heat has yet been made, and that the 

 law cannot be regarded as a final statement of the con- 

 nection between the specific beats and atomic weights of 

 the elements. 



The state of our knowledge with regard to the struc- 

 ture of molecules, indeed, renders a full understanding of 

 specific heat at present impossible. The dynamical theory 

 of gases has not yet been fully worked out in this 

 direction. 



Although the " law of Avogadro " is a deduction from 

 the dynamical theory of gases, and as such is invested 

 with an authority which no mere collection of empirical 

 facts can bestow upon it, yet nowhere in M. Wurtz's 

 book is this insisted upon. 



The compromise between an atomic and an equivalent 

 system of notation, which was so long adopted by 

 chemists, is well described and its evils fully laid bare. 



The objectors to Avogadro's law are more numerous 

 and more important in France than in this country or in 

 Germany, hence M. VVurtz devotes considerable space to 

 the subject of dissociation, which he discusses with much 

 clearness and wealth of illustration. 



The demonstration on pp. 121-123 of the monatomic 

 character of the mercury molecule is admirable. 



In the list of names of those who have pointed out 

 relations between the atomic weights of elements and 

 properties of their compounds, there is a serious omission, 

 viz., the name of A. R. Ncwlands. This subject of 

 relation between atomic weights and properties of com- 

 pounds is discussed on pp. 154-176. A better idea of 

 Mendelejeff's " periodic law " may be obtained from these 

 pages than probably from any other English text-book, 

 but surely it would have been well had the author more 

 explicitly acknowledged his indebtedness to Lothar 

 Meyer's work. The graphic representation of the rela- 

 tions between the atomic weights and physical properties 

 of the elements — taken from Meyer's book — has not 

 hitherto been in the hands of the English student. 



