THE DEVELOPMENT OF THE ATOMIC THEORY.
II. THE VARIOUS ACCOUNTS of the Origin of
DALTON'S THEORY.

By ANDREW NORMAN MELDRUM, D.Sc.
(Carnegie Research Fellow).

THE origin of Dalton's theory remains one of the outstanding problems in the history of chemistry. Yet the amount of material at hand for the study of the subject is considerable. Dalton's note-books, discovered within the last twenty years in the rooms of the Manchester Literary and Philosophical Society, contain material of the highest value for the purpose. Also, there are on record important accounts of the genesis of the theory by three different persons. One is given by William Charles Henry, another by Thomas Thomson, and another by Dalton himself. Although there are yet other accounts these three are the only ones that need be considered in detail in this paper.

The main object of the paper is to offer a suggestion, by way of explanation of how these various accounts arose, and to that extent of the discrepancies between them. It will become more and more evident in the course of the paper that Dalton is the person responsible for these discrepancies.

a

1. The Influence of 7. B. Richter. William Charles Henry was the pupil, the friend, and finally the biographer of Dalton. He held a conversation with Dalton on the subject of the origin of the theory, in which special importance was given to the influence of J. B. Richter. "The speculations which gave birth to the atomic theory were first suggested to Mr. Dalton by the experiments of Richter on the neutral salts table was formed exhibiting the proportions of the acids and the alkaline bases constituting neutral salts. It immediately struck Mr. Dalton that if these saline compounds were constituted of an atom of acid and one of alkali, the tabular numbers would express the relative weights of the ultimate atoms, These views were confirmed and extended by a new discovery of Proust," &c. (W. C. Henry, "Memoirs of Dalton," p. 84).

This narrative received strong support from William Henry (the father of W. C. Henry), who held more than one conversation with Dalton on the subject. The following is part of a minute, dated February 13, 1830, of one of these conversations: "Confirmed the account he before gave me of the origin of his speculations leading to the doctrine of simple multiples, and of the influence of Richter's table in exciting these views" (Ibid., p. 63).

was more intimate than that of any other man of science, Peter Clare excepted. In the preface to his biography of Dalton, W. C. Henry refers with just pride to Dalton's "almost lifelong friendship with my father, never shadowed by even a passing cloud"; and he refers also to "his early favourable notice of and unceasing benevolent regard towards myself, thoughtfully manifested in his last bequest to me of what he had most prized in life." This was the bequest of all his chemical and philosophical instruments and apparatus. Further illustration of this friendship is to be found in the dedication of Dalton's "New System of Chemical Philosophy" (vol. i., Part 2) to William Henry (along with Humphry Davy), and of Henry's "Elements of Experimental Chemistry" (6th Ed., 1810) to Dalton. There is no room for doubt that the reports of these conversations with Dalton are perfectly authentic. W. C. Henry states that he noted down Dalton's expressions" immediately after each lesson," and the passage which has been quoted, regarding the influence of Richter, is copied, he says, "verbatim from my own journal when his pupil" (Ibid., p. 84). Nevertheless, Henry knew there was something wrong. The date of his conversation with Dalton was February 5, 1824, and he says, " on reviewing in conversation, after the lapse of twenty years, the labours of the past, Dalton himself may have failed in recalling the antecedents of his great discovery in the exact order of sequence" (Ibid., p. 86).

Again, the Richter story is strongly challenged by Thomas Thomson. "When I visited him in 1804 at Manchester both Mr. Dalton and myself were ignorant of what had been done by Richter on the same subject." Again, "Nobody knows better than myself that Dalton was ignorant of what Richter had done about ten years. before him" (Proc. Phil. Soc. Glasgow, 1845-6, ii., pp. 86, 88). This shows conclusively that Dalton said nothing about Richter to Thomson.

Now that we have access, thanks to Roscoe and Harden's "New View of the Origin of Dalton's Atomic Theory," to the valuable material contained in Dalton's notebooks, we can carry the critical process further than Henry and Thomson did. The notebooks show, as Roscoe and Harden point out, that Dalton had been busily engaged during the year 1803 on the atomic theory, and that he was investigating the non-metallic elements then, and not Richter's acids and bases at all. The date of the earliest reference to Richter is April 19th, 1807 (Roscoe an Harden, "New View of the Origin of Dalton's Ator Theory," p. 79; see also pp. 7-10, 46, 91-94). Th really no room for doubt that Dalton's declarat 1824 and 1830 to one and the same effect regar influence of Richter must be set aside (R

2. The Composition of Marsh-gas and Olefiant Gas. Thomas Thomson says that the theory first occurred to Dalton during his investigation of marsh-gas and olefiant gas. The discovery of the composition of these gases led to the discovery of the law of multiple proportion, and the theory was then devised in order to explain the law. His

exact words are:

"Mr. Dalton informed me that the atomic theory first occurred to him during his investigations of olefiant gas and carburetted hydrogen gas, at that time imperfectly understood, and the constitution of which was first fully developed by Mr. Dalton himself. It was obvious from the experiments which he made upon them that the constituents of both were carbon and hydrogen, and nothing else. He found, further, that if we reckon the carbon in each the same, then carburetted hydrogen contains exactly twice as much hydrogen as olefiant gas does. This deter mined him to state the ratios of these constituents in numbers, and to consider the olefiant gas a compound of one atom of carbon and one atom of hydrogen; and carburetted hydrogen of one atom of carbon and two atoms of hydrogen. The idea thus conceived was applied

to carbonic oxide, water, ammonia, &c., and numbers were given representing the atomic weights of oxygen, azote, &c., deduced from the best analytical experiments which chemistry then possessed (Thomas Thomson, History of Chemistry," ii., 291).

64

It

This narrative has passed muster for many years. was accepted with reservations by W. C. Henry ("Memoir of Dalton," p. 80) and Angus Smith (" Memoir of Dalton," p. 231), and by Roscoe and Schorlemmer without objection "Treatise on Chemistry, Non-metallic Elements," 1877, p. 36). It is of special importance because, more than any other version of the origin, it has influenced the opinion of the generality of chemists.

There is, nevertheless, the best reason for thinking that marsh-gas and olefiant gas did not have the effect which it assigns to them of leading to the theory.

Indeed, in 1811, Dalton connected the theory in its early days with the oxides of nitrogen :-"I remember the strong impression which at a very early period of these inquiries was made by observing the proportion of oxygen to azote, as 1, 2, and 3, in nitrous oxide, nitrous gas, and nitric acid, according to the experiments of Davy" (Nicholson's Journ., 1811, xxix., 143). Thomson must have seen the necessity of abandoning the marsh-gas and olefiant gas story, for he said in 1850:-"Dalton founded his theory on the analysis of two gases, namely, protoxide and deutoxide of azote" (Proc. Phil. Soc. Glasgow, 1850, iii., 140). Roscoe and Harden (loc. cit., p. 28) point out that Dalton had been busily engaged on the theory the year before he investigated these gases. He had even arrived at the fundamental ideas of his system, and had constructed a table of atomic weights by September 6th, 1803.

Obviously, Thomson's account of the origin of the theory is untrustworthy, inasmuch as marsh-gas and olefiant gas had no part in the matter. The question arises, who is responsible for the error, Thomson or Dalton? Before answering this question it is necessary to consider carefully the relations between the two men and the circumstances under which Thomson's narrative arose.

Thomson, unlike the Henrys, was not a personal friend of Dalton. He had made an adverse criticism of a certain theory of which Dalton was the author, and the author had made a stiff rejoinder (see Nicholson's Journ., 1804, viii., 145; and Annals of Philosophy, 1814, iv., 65). He thereupon paid a visit to Manchester with the object of rriving at a full understanding of the matter in question. e date of the interview was August 27th, 1804, and it then, by a fortunate accident, that Thomson learnt of mical atomic theory of Dalton.

it is certain that Thomson and Dalton were not tly in frequent communication with one another The sketch of the theory, which Thomson 1807, was accompanied by the note:-"In Dalton, I must warn the reader not to decide

ct.

upon the notions of that philosopher from the sketch which I have given, derived from a few minutes conversation, and The mistakes, if any from a short written memorandum. occur, are to be laid to my account, and not to his; as it is extremely probable that I may have misconceived his meaning in some points" (Thomas Thomson, "System of Chemistry," 3rd Ed., 1807, iii., 425).

Nevertheless, this footnote errs on the side of caution. Thomson showed both zeal and care in the matter, for His sketch of the theory, it strongly interested him. giving the first account of it ever published, was based on notes of what Dalton told him, made during the interview, and only one phrase in it is open to objection.

In the "History of Chemistry," published in 1831, Thomson says:-"I wrote down at the time the opinions which he offered, and the following account_is_taken literarally from my journal of that date" (Thomas Thomson, "History of Chemistry," ii., 287). Then comes an account of the atomic theory, and on that there follows the passage already quoted, connecting marsh-gas and olefiant gas with the genesis of the theory. Here the question arises, is all this taken from the journal, both the sketch of the theory and of how it arose? Only an examination of the journal can settle this point, but I have not succeeded in ascertaining where it is kept, if, indeed, it is still in existence.

It must be admitted also that Thomson seems to become more and more positive regarding the genesis of the theory as time goes on. The account which I have been considering was published in 1831. Six years earlier he had advanced the same account in a more hesitating way :'Unless my recollection fails me, Mr. Dalton's theory was originally deduced from his experiments on olefiant gas and carburetted hydrogen" (Thomas Thomson, "An Attempt to Establish the First Principles of Chemistry by Experiment," 1825, i., 11). Yet there is no intrinsic improbOne cannot ability that Thomson's recollection is correct. doubt that during the interview Dalton was much less interested in the question of origin than in the theory itself. If Thomson inquired about the origin Dalton may have made the inquiry an opportunity of expounding the theory in terms of its latest triumph, namely, the composition of marsh-gas and olefiant gas.

He

3. The Amended Theory of "Mixed Gases.” There remains for consideration the account which Dalton gave in a lecture at the Royal Institution of London in 1810. It still exists in his own handwriting, and was found, along with his notebooks, in the rooms of the Manchester Literary and Philosophical Society. begins by discussing his physical atomic theory, which aimed at explaining the diffusion of gases. He entertained two diffusion hypotheses, the first of which originated in 1801, while an amended hypothesis he says, was formed in the year 1805. He had not at first "contemplated the effect of difference of size in the particles of elastic fluids." On consideration, he "found that the sizes must be different," and subsequently arrived at a different explanation of the mechanism of diffusion from the one he at first suggested.

He then introduces the subject of the chemical atomic theory:"The different sizes of the particles of elastic fluids under like circumstances of temperature and pressure being once established, it became an object to determine the relative sizes and weights, together with the relative number of atoms in a given volume. This led the way to the combination of gases other bodies besides elastic fluids, namely, liquids and solids, were subject to investigation, in consequence of their combining with elastic fluids. Thus a train of investigation was laid for determining the number and weight of all chemical elementary principles which enter into any sort of combination one with another " (Roscoe and Harden, loc. cit., pp. 16-17).

This narrative is certainly right on an essential point. It recognises that Dalton had been using a physical atomic theory, from which he passed to a chemical one. Here

there is a common ground of objection to both the communications made by Dalton, to Thomson and Henry respectively. They both ignore the connection, which certainly existed, between the physical and chemical theories. Thomson did not feel this defect, but Henry did. While not denying the influence of Richter, he sums up the evidence on the subject as "unequivocally demonstrating the genesis of the atomic theory as a general physical conception from the study of matter in the aeriform condition, and its first practical application in chemistry to gaseous bodies, and emphatically to such as combine in multiple proportions" (W. C. Henry, loc. cit., p. 84). There is no question here of extraordinary insight and discernment on Henry's part. He has simply considered the use Dalton had made of the physical atomic theory previous to forming a chemical one. There is therefore not much justification for Roscoe and Harden's suggestion that they had found in Dalton's narrative a new view of the genesis of his atomic theory. The view is to be found in Henry, and would be formed by any person who should read with understanding the "Essay on the Constitution of Mixed

Gases," which Dalton wrote in 1801.

There is, however, a fundamental objection to Dalton's narrative. It has a deceptive appearance of being historical. Dalton was a pioneer of science, and a pioneer is a man who must make many mistakes and experience many failures. He has taken a number of different scientific movements and marshalled them, so that they are invested with the appearance of a great, irresistible, strategical advance. On examination his narrative, in spite of its grand air, is found to throw much less light than it promises on the line of thought and train of investigation which he pursued. It is excessively abstract in tone, and avoids going into details and particulars and instances. It does not tell us what we want to know most, how and when Dalton arrived at the law of multiple proportions, and the part played by the law in the construction of the theory. Information on these matters is what is wanted, and any thing else is beside the point.

Yet there is one novel element in Dalton's account. This is the suggestion that the formation of the chemical atomic theory took place subsequently to the amendment of the diffusion theory. But, as the notebooks show, the chemical theory was formed in 1803. Hence, Roscoe and Harden conclude that 1805, the date which Dalton assigns to his amended diffusion theory, should be 1803 (Ibid., p. 25). Reasons will be given later in a paper on Dalton's theory of "mixed gases" for thinking that the date 1805 is right after all. Hence, Dalton's narrative is doubtful on the only point in which it presents any novelty.

himself.

Conclusion.

There are in existence yet other accounts of this matter. One is given by Dalton's pupil, Joseph A. Ransome (W. C. Henry, loc. cit., pp. 220-222), and another by Dalton This was in the lecture which he delivered in Manchester on October 19th, 1835 (Manchester Times, October 25, 1835). The main feature which all the accounts have in common is that each originated with Dalton. Thomson's narrative and Henry's and Ransome's were based on conversations with him, and there is no ground for impugning their accuracy any more than his good faith. The natural explanation of the existence of so many and various accounts is that Dalton was simply deficient in historical instinct. He did not perceive the difference between describing the genesis of his theory and expounding the theory itself.

A man who makes history as Dalton did need not be a good historian. The account of the origin of the chemical theory in his own handwriting is no more satisfactory than the others which came from him at second-hand. Apparently, Dalton never had in his mind a precise view of how the theory developed, and when invited to give one he produced, on the spur of the moment, an account to which he did, or did not, adhere on the next occasion. The University, Manchester.

ON THE CALCULATION OF OPTICO-CHEMICAL CONSTANTS.

PART I.-ALIPHATIC HYDROCARBONS.

By H. STANLEY REDGROVE, B.Sc. (Lond.)., F.C.S.

§ 1. Introductory.

IT has been pointed out by the author (see "On the Calculation of Thermochemical Constants," Arnold, 1909, pp. 1-7) that the usually-adopted method of cal. culating the specific influences of atoms and unsaturated links on additive physico-chemical properties is fundamentally incorrect, inasmuch as the influences of the ethane and certain other links are ignored. If the influences of these links are not assumed to be zero, the data relative to the subject obtainable in the present state of knowledge reduce to a set of indeterminate equations. The difficulty is overcome and an unjustifiable assumption avoided by the author's general method of calculating physico-chemical constants, which depends upon the use of certain derived

constants chosen so as to be obtainable from the indeterminate equations referred to above.

This method has been used successfully for the calculation of thermo-chemical constants. It is now proposed to adopt a somewhat similar procedure in the case of opticochemical constants. The fundamental constants employed will have, in general, the same general significations as before; but since their specific significations will be entirely different, capital Greek letters will be employed to designate them (instead of small Greek letters as in the case of thermo-chemical constants) in order to avoid any possible confusion.

referred to above, that, in general, if C represents the It has been shown by the author in the monograph influence of a carbon atom irrespective of its valences, H the influence of a hydrogen atom together with the link binding it to carbon, and L1, L2, L3 the influences of the of C, H, L1, L2, and L3 are not obtainable in the present ethane, ethene, ethine links respectively, then the values state of knowledge; but the values of the following derived constants-the so-called "fundamental constants" -can readily be calculated :-A=C+4H, B = 2H - L1, r-4H-L2, A=6H-L3. From these constants the theoretical value for any aliphatic hydrocarbon can at once be obtained.

§2. The Data Employed.

We shall avail ourselves of the experimental work of various chemists. In cases where more than one determination has been carried out on the same substance we shall, as a general rule, choose those results which by their consistency with the main body of data show themselves to be the more accurate. The actual experimental data used in the present paper are exhibited in Table I. In this table, Column I gives the number of each substance, by which reference to it can be made; Column 2 gives the name and Column 3 the formula of each substance; Column 4 gives t, the temperature at which each determination was carried out; Column 5 gives d1t, the density of each substance at temperature t compared with water at 4° C.; Column 6 gives ua, the refractive index of each substance for the a line of the hydrogen spectrum; (= Fraunhofer's C line); Column 7 gives Ma, molecular refractivity for the Ha line calculated according to the formula of Lorentz-Lorenz, according to which (μ2 - 1)m M = where m is the molecular weight of the (μ*+2)d substance in question; Column 8 gives u, the refractive index for the Hy line; Column 9 gives M1, the molecular refractivity for the Hy line calculated in a similar manner from μ7; Column 10 gives My-Ma, the molecular dispersion; the footnotes give the name of the experimenter to which each result is due, and the references to the original memoirs from which the data have been taken.

the