520 



NA TURE 



[October 28, 1909 



be admitted that Higgins, who stated these facts and 

 reasoned very justly upon them in his "Comparative 

 View of the Phlogistic and Anti-phlogistic Hypo- 

 theses " (1789), did not give any sign, by collating 

 them, that he felt himself on the threshold of a great 

 discovery. Again, Gay-Lussac and Humboldt, taking 

 up the study, for purposes of eudiometry, of the com- 

 bination of hydrogen and oxygen, found the ratio 

 between these gases to be 2 : i as nearly as they could 

 measure. This was in 1804. The observation arrested 

 Gav-Lussac's attention. Curious to find if other such- 

 like cases exist, he began work which resulted in the 

 discovery of his law, one of the most important in 

 science. 



Gay-Lussac, like Newton, did not form hypotheses. 

 The memoir in which he set forth his work is remark- 

 ably free from speculative matter. His conviction was 

 that " in natural science, and above all in chemistry, 

 generalisation should come after and not before a 

 minute knowledge of each fact." And assuredly the 

 history of Gay-Lussac 's law in science does show that 

 a " law of nature " may prove a dangerous weapon 

 to the man who puts it to theoretical and practical 

 uses, before its range and bearings in nature have 

 been accurately fixed. 



The law when published aroused the widest interest. 

 The world of science was just then pondering the 

 atomic theory in the form impressed on it by Dalton, 

 and it was obvious that theory and law must stand in 

 the most intimate relation to one another. Strangely 

 enough, the law was objected to by Dalton of all 

 people, and by him alone. In the second part of his 

 " New System of Chemical Philosophy," published in 

 18 10, he made strictures on it, and concluded : — " The 

 truth is, I believe, that gases do not unite in equal 

 or exact measures in any one instance ; when they 

 appear to do so, it is owing to the inaccuracy of our 

 experiments. In no case, perhaps, is there a nearer 

 approach to mathematical exactness, than in that of 

 one measure of oxygen to two of hydrogen ; but here 

 the most exact experiments I have ever made gave 

 I'gy hydrogen to i oxygen." Berzelius wrote to 

 Dalton protesting in the most courteous way against 

 the part of the atomic theory " which obliges you to 

 declare as inaccurate the experiments of Gay-Lussac, 

 on the volumes in which gases combine. I should 

 have thought rather that these experiments were the 

 finest proof of the probability of the theory ; and I 

 confess to you, that I will not so readily think Gay- 

 Lussac at fault, especially where the point is one of 

 good or of bad measurement." Nothing, however, 

 cuuld ever remove the distrust Dalton felt in the law. 



The chemists who accepted both Dalton 's theory and 

 Gay-Lussac's law had themselves to solve the problem 

 of defining the relation between the two. No more 

 than Dalton would Gay-Lussac do anything to help 

 them. Even so late as the year 1814, in his memoir 

 on iodine, and in the one on prussic acid of the 

 following year, he ignores the atomic theory. He 

 uses the word " molecule " for the sake of convenience, 

 and that is all. Yet there must be a connection 

 between the specific gravities, that is, the weights of 

 equal volumes, of different gases and their atomic 

 weights. This connection is the primary subject of 

 a paper by Prout, published in 1815. Here he 

 advanced his famous hypothesis that the atomic 

 weights of the elements are multiples of the atomic 

 weight of hydrogen, but there is good reason to think 

 that the hypothesis was conceived after the data had 

 been rounded off. 



Berzelius had already, in \%\t.. if not earlier, given 

 his solution of the problem. This was his " volume- 

 theory," that equal volumes of different gases contain 

 the same number of atoms. This hypothesis affords 

 NO. 20S7, VOL. 81] 



a basis of a purely physical kind for the determina- 

 tion of atomic weights, for it means that the atomic 

 weights of different gases stand in the same ratio to 

 one another as the weights of equal volumes of the 

 gases. 



The " volume-theor\'," plausible as it seems, in- 

 volved its author in difficulties one after another, 

 which finally became overwhelming. Or 1 arises as 

 soon as the theory is formulate''. Each atom of 

 hydrogen, on combining with chlorine, could, as 

 Berzelius and Dalton understood the atom yield only 

 one compound atom of hydrochloric acid. But the 

 volume of the hydrogen is half that of the hydro- 

 chloric acid which it produces, so that the atom of the 

 element occupies only half the volume of the compound 

 atom. Hence the theory must either be limited to 

 elements, or given up altogether. Years before 

 Dalton had to face the same difficulty i 1 the case of 

 nitric oxide. What he did at first was to abandon 

 outright the hypothesis that atoms of different gases 

 have the same volume, and then to object even to 

 Gay-Lussac's law. Dalton was "for thorough." 

 \\'hat Berzelius did was to make the " volume- 

 theory " apply only to the elements. 



In course of time another difficulty appeared. The 

 atoms of many important elements seemed to enter 

 into combination only by pairs. This strange result 

 arose in the following way. Berzelius began in the 

 year 1826 to ascribe the general formula RO to alJ 

 strong bases. Now, by the chemical equation for the 

 formation of a chloride from a base — RO+H,CL = 

 RCL-I- HjO — it is plain that the amount of acid needed 

 to form a chloride with one molecule of a base con- 

 tains two atoms of hydrogen and two of chlorine. 

 That is, as Berzelius saw, the hydrogen enters into 

 chemical combination in pairs, and so does the 

 chlorine atom. 



This, be it noted, involves a conception of the 

 element which is precisely the reverse of the modern 

 one. Hydrogen is now supposed to consist of physical 

 atoms, each of which can be halved when it enters 

 into chemical combination. The physical atom of 

 hydrogen is composed of two chemical ones. Berzelius 

 had formed the conception of a chemical atom com- 

 posed of two physical ones. It applied to quite a 

 large number of elements in addition to hydrogen, 

 namely, to chlorine, fluorine, bromine, iodine, nitro- 

 gen, phosphorus, antimony, and arsenic. 



The very natural comment on this was made by 

 Gmelin, that the " existence of the physical atom was 

 improbable and its adoption superfluous and trouble- 

 some." One could arrive at Gmelin 's system of 

 chemical formulse by suppressing every pair of 

 physical atoms in Berzelius's formulas, and putting in 

 a chemical atom instead. Thus H,0 became HO. 

 Nobody could help seeing that Berzelius's system 

 simply led the way to Gmelin 's. This was a great 

 blow to the " volume-theory," for Gmelin 's system 

 differs from Berzelius's only by leaving out the - 

 " volume-theory " and all its consequences. 



The above as an objection to the theory was per- 

 ceived and felt to be overwhelming only in course of 

 time. .As already explained, from the first the theory 

 could include in its scope only the elements. But 

 before long Berzelius had to limit the theory still 

 fuither. So long as it is applied to elements the 

 molecules of which are of the same degree of com- 

 plexity, hydrogen and oxygen, for instance, the 

 physical method of finding atomic weights is in agree- 

 ment with the chemical. The ratio H^/O^, which the 

 former method gives, is the same as the ratio H/O 

 given by the latter. But this is a matter of accident. 

 About the year 1826 Dumas succeeded in finding the 

 vapour-density of elements such as mercury and pbns- 



