Proust, Joseph Louis
PROUST, JOSEPH LOUIS
(b.. Angers, France, 26 September 1754; d. Angers, 5 July 1826)
The second son of Joseph Proust, an apothecary, and Rosalie Sartre, Proust received his early education under the supervision of his godparents and continued it at the local Oratorian collège. He was then apprenticed to his father, to study pharmacy and to succeed him. Around 1774, despite parental opposition, he went to Paris to continue his training.
In Paris, Proust attached himself to Clérambourg, the apothecary, and studied chemistry with Hilaire-Martin Rouelle (not his elder brother, Guillaurne-Francois Rouelle, as is often reported), with whom he became friends, ever after referring to him as his “master.” In 1776 Proust secured the position of pharmacien en chef at the Salpêtrière. He published his first papers while at this hospital.
At the end of 1778, Proust took his first position in Spain, the country in which he spent the major part of his professional life. The post was a professorship in chemistry at the recently established Real Seminario Patriótico Vascongado at Vergara. This school was the creation of the first and most important of the “enlightened” societies which came into existence in the second half of the eighteenth century to bring modern learning and culture to Spain: the Real Sociedad Económica Vascongada de Amigos del Pais, established in 1765. Royal authorization had been obtained for professorships at the school in experimental physics, chemistry, mineralogy, and metallurgy. Proust stayed in Vergara only a short time, returning to France in June 1780.
During the next five years Proust’s most important enterprises were those connected with Pilâtre de Rozier. Until 1784 Proust taught chemistry at the Musée, founded by Pilatre in 1781. Also during this period Proust was involved with Pilâtre, as well as the physicist Jacques Charles, in aerostatic experiments that culminated in the ascent of Proust and Pilâtre in a balloon at Versailles on 23 June 1784, in the presence of the king and queen of France, the king of Sweden, and the French court.
In 1785 Proust again contracted to take a teaching position in Spain, this time at the invitation of the Spanish government and through the intermediary of Lavoisier. The offer was for a financially remunerative post, and after some hesitation Proust accepted and went to Spain in 1786. He first taught in Madrid; but in 1788 he moved to Segovia as professor of chemistry at the Royal Artillery College, where chemistry had been made a required subject. Proust taught and experimented there and also conducted geological and mineralogical surveys and analyses for the government. Assessments of his success as a teacher are contradictory; and there is some evidence that toward the end of this period he wished to return to France.
On 30 June 1798 Proust married a French resident of Segovia, Anne Rose Chêtelain Daubigné. They had no children. In April 1799 Proust was brought from Segovia to Madrid to head a newly organized chemical laboratory that amalgamated the previously existing facilities in Madrid and Segovia, and to oiler public courses in chemistry. The opulence of his laboratory became legendary. While in Segovia, Proust began to publish his papers on definite proportions and it was while he was in Madrid that the controversy with Berthollet over this issue took place.
Proust returned to France toward the end of 1806, for reasons not altogether clear (one account suggests that it was to settle his patrimony with his brother Joachim), and remained there. In 1808 Napoleon’s forces invaded Spain; Carlos IV abdicated; and during the ensuing political dislocation and resistance, Proust’s laboratory establishment was dispersed. Left in reduced circumstances by this sequence of events, Proust settled in Craon but spent considerable time at a family farm that he had inherited in the Loire valley. One of the few bright features of his return to France was his cordial reception by Berthollet. Both Proust and his wife were ill after 1810; she died in 1817, and Proust then moved to Angers, where in 1820 he took over the pharmacy of his brother Joachim, who was in poor health.
Awards came to Proust, although he remained aloof from the French scientific establishment. On 12 February 1816 he was elected to succeed Guyton de Morveau at the Institut de France. In 1819 he became a chevalier of the Legion of Honor, and in 1820 he was granted a pension by Louis XVIII.
Proust’s published papers are notable for their clear style and exposition. Testimony differs over his abilities as a lecturer and teacher, and there is reason to believe that he was at times negligent of his duties. Proust left Pilâtre de Rozier’s Musée in 1784 after criticism of his teaching, and there is evidence that his effectiveness as a teacher in Spain was less than his early French biographers suggested it had been.
Proust’s historical importance derives primarily from his abilities as an analyst: his development of the use of hydrogen sulfide as an analytical reagent; his use of quantitative methods—consistently giving the results of his analyses in terms of percentage weight composition and sometimes (for oxides and sulfides of the same metal) the weight of oxygen or sulfur in comparison with a constant weight of the metal (without, however, drawing any conclusion from the results)—and most significant, his enunciation of the law of definite proportions.
There can be little question that the general tendency in late eighteenth-century chemistry toward quantitative analysis, associated particularly with the expression of weight composition which became common especially in the 1780’s, served as the background to the formulation of the law of definite proportions. Yet the exact relationship between this development and the enunciation of the law remains obscure. The assumptions behind gravimetric analysis are not necessarily the same as the fundamental tenet of the law of definite proportions: that combining substances can combine in only a small number of fixed proportions.
Thus, to speak, as some have done, of Proust’s law as a commonplace of eighteenth-century chemical thought is unwarranted, at least on the basis of the limited study that this question has received. There were, however, precursors of Proust’s law in the literature of the late eighteenth century, notably Robert Dossie (1759), G. F. Venel (1765), L B. Guyton de Morveau (1786), and Thomas Thomson (1801).1
These authors shared a common viewpoint and stated what I would call the “principle of constant saturation proportions.” Their argument ran roughly as follows: Reacting substances had a unique proportion of combination, the saturation proportion, at which—and only at which—chemical combination took place. Observationally, the saturation concept derived from neutralization reactions and from saturation of solvent by solute. The satiation at constant proportions was due to the constant intensity of the affinity forces between the two reagents. Dossie’s statement is a good summary and illustration of this principle.
In bodies that will commensurate strongly with each other, the specific attractions are in many cases limited only to certain respective proportions. For in some kinds, after they are combined together in a certain proportion, the compound becomes neutral, or indifferent with regard to further quantities of any of its constituents; in the same manner, as if those specific attractions, by which it was formed, had been wholly wanting.2
In addition to the chemical formulation of the principle of constant saturation proportions, the doctrine of fixed mineral species, enunciated by Romé de risk (1784), Haüy (1793, 1801), and Dolomieu (180l),3 presented defining characteristics of constant chemical composition and fixed crystal form. Although the mineralogical doctrine of fixed species seems to have developed more or less independently of the chemical principle of constant saturation proportions or of Proust’s law, it was to have indirect relevance to Proust. since he was led to make his most comprehensive statement of the law of definite proportions in answer to a challenge actually directed at the mineralogists.
Proust’s own formulation of the law of definite proportions was published rather suddenly in a paper on iron oxides, “Recherches sur le bleu de Prusse” (1794). Until further research is done on Proust’s lecture notes and other materials from the Segovia period just prior to and encompassing this paper, it will be difficult to form an impression of the origins of his law.
Proust introduced his problem at the start of the 1794 paper:
If iron were, as is thought, susceptible of uniting with oxygen in all proportions between 27/100 and 48/100, which seem to be the two extreme terms of its union with this principle, ought it not to give as many diverse combinations with the same acid as it can produce different oxides?
He drew the following conclusion:
A great number of facts prove, on the contrary, that iron does not at all stabilize indifferently at all the degrees of oxidation intermediary between the two terms which we have just cited; and despite the different degrees of oxygenation through which one believes iron can pass when its sulfate is exposed to the air, only two sulfates of this metal are known.
At the end of the paper, Proust generalized his conclusion for other oxidizable substances:
I terminate … by concluding from these experiments the principle which I established at the beginning of this memoir; namely, that iron is, like several other metals, subject by that law of nature which presides over all true combinations, to two constant proportions of oxygen. It does not at all differ in this regard from tin, mercury, lead etc. and finally from virtually all of the known combustibles.4
Over the next thirteen years, and especially after 1797, Proust elaborated this conclusion in a series of papers in which he tried to show that most metals formed two distinct oxides at constant proportions—which he called the minimum and maximum—and that these two metal oxides were capable of forming two separate series of compounds, lead being recognized as an exception in forming three. Proust also asserted that there was only one sulfide per metal, with the exception of iron, which he came to recognize had two.
Although Proust’s papers on metallic oxides appeared quite abruptly in the 1790’s, interest in the different saturation proportions of metallic oxides had been growing in the previous decade with the recognition of the pneumatic chemistry and its gravimetrie methods. Lavoisier had published two papers on the analyses of iron oxides (Proust cited data from the second). Proust’s interest in metallic oxides and sulfides undoubtedly was also stimulated by his professional concern in Spain with metallurgical and mineral analysis. Spain with metallurgical and mineral analysis.
Proust’s strategy for proving his assertions about metallic oxides was well developed in his paper on Prussian blue and changed but little in general outline thereafter. He tried to show that there were two oxides of each metal, each oxide having a set of well-defined physical and chemical characteristics. Any metallic salts formed in chemical reactions had to have one of the oxides as its base. In oxidation or reduction reactions, Proust attempted to show that the oxide was raised (or lowered) directly from minimum to maximum (or vice versa) with no intermediate oxide produced.
As for oxides at intermediate proportions, Proust’s strategy was to identify them as mixtures of the maximum and minimum oxides, mixtures of one of the oxides with uncombined metal, or a compound that really was not an oxide. Similarly, in the case of sulfides, Proust tried to show that reports of sulfides in variable proportions (such as those later offered by Berthollet) really concerned various kinds of mixtures or solutions, not true chemical compounds. This distinction between “true” compounds and mixtures or solutions became the crucial—and weakest— part of his strategy.
To separate what he took to be mixtures of the two oxides, Proust employed the diversity in chemical properties of the maximum and minimum compounds—for example, solubility in alcohol or some other solvent. Separating oxide from excess metal could also be accomplished by this means, as well as by more strictly physical ones, such as washing the oxide or heating a sulfide to sublimate the excess sulfur without changing the essential “physionomie” of the compound.
Proust consistently adopted a posture of empiricism, particularly in his controversy with Berthollet; indeed, there is little evidence of concern on his part with establishing any detailed theoretical underpinnings for his assertions of constant proportions. He chided Berthollet for elaborating an all-encompassing theoretical system, claiming that chemistry was not yet “mature” enough for such an approach but required more experimental data. Yet one can detect in Proust’s justification of definite proportions more than the mere generalization from laboratory data that he claimed for his method. There were occasionally echoes of chemical affinity theory: “Election and proportion are the two poles around which the whole system of true combination rotates invariably, as much in Nature as in the hands of the chemist.” And in one of his rare formulations of chemical combination in something like molecular terms, he wrote:
For example, when a glass of potash is exposed to free air, every molecule of carbonic acid which approaches it is seized instantly by the number of alkali molecules which are needed to transform it into the carbonate. The attraction is there, as one knows; it keeps watch, it presides over this number. This reaction thus introduces into the potash new portions of carbonate, but of a complete carbonate.5
But even more general and fundamental to Proust’s assertion was his belief in a natural principle of order, which he saw as regulating all chemical unions in nature as well as in the laboratory. This pondus naturae, as he called it, assured the invariability of chemical reaction and the constancy of the proportions of the product, regardless of the circumstances under which the reaction took place.6 In 1788, six years before his first paper on definite proportions, Proust had already expressed this idea in a comment on the presence of phosphoric acid in minerals as well as in laboratory-produced compounds: “Presided over by the same laws, the ones and the others [minerals and artificially produced substances] are always alliances of choice and proportion.”7
This notion of an overarching principle of proportion in nature and in the laboratory permeated Proust’s writings on definite proportions. It remained disconcertingly empty of content; at times there is, indeed, more than a hint of rhetorical flourish in his invocation of it. Yet in one aspect it was endowed with some specificity: Proust held tenaciously to the idea that each metal could form only two oxides (with rare exceptions) and one sulfide (also with rare exceptions). Two oxides and one sulfide seem to have had an almost numerological significance for Proust; and he later defended this view despite Berzelius’ analyses that contradicted it.
By 1801, through his refinements of the theory of chemical affinity, Berthollet had arrived at the proposition that chemical combination was, in principle, not necessarily restricted to definite proportions; combining substances could unite in a continuum of weight proportions between the minimum and the maximum. Berthollet attributed those combinations that seemed to occur in fixed proportions to physical properties—such as volatility, or tendency toward precipitation or crystallization—which tended to halt chemical reactivity and, hence, to stabilize the combinations. But there was nothing intrinsically “natural” about such compounds; and where reagents did not tend to produce compounds strongly endowed with such physical characteristics, a continuum of proportions might be produced.
Berthollet published a comprehensive exposition of his chemical theory, including his ideas on combining proportions, in Essai de statique chimique (1803). Despite his obvious disagreement with Berthollet’s ideas, Proust was not the main object of Berthollet’s attack on this issue in the Essai. Rather, in the theoretical sections of the first volume Berthollet devoted his most elaborate criticisms to Haüy’s and Dolomieu’s doctrine of fixed mineral species. But in the second volume, Proust did come under attack for his position on metallic oxides and sulfides—an attack strong enough to provoke his response in 1804.
The controversy through 1807 consisted of a paper or two every year from both parties to the dispute. Proust’s most elaborate rebuttal of Berthollet’s views, and defense of his own, was presented in “Sur les oxidations metalliques” (1804). Its arguments are typical. His strategy against Berthollet was twofold: to demonstrate inconsistencies and even contradictions between Berthollet’s physicalistic and theoretical explanations for the appearance of some oxides at fixed proportions; and to show experimentally that instances of what Berthollet took to be oxides at intermediate proportions between minimum and maximum were really mixtures either of the two oxides or of the metal and one of its oxides.
To demonstrate such inconsistencies and contradictions, Proust noted that Berthollet had suggested that when metallic oxides were produced having stable proportions, those proportions were endowed with a greater degree of condensation than the metal itself. But Proust pointed out that two of Berthollet’s examples—arsenic and antimony—had oxides at minimum that were more volatile than the metal. Berthollet also had argued that volatilization of the metal favored its rapid oxidation to maximum, because the metal could mix more thoroughly with the oxygen in the air; on the other hand, the solid state was a hindrance to complete oxidation because of the reciprocal attraction of the metallic particles. Proust pointed out that arsenic, which volatilized easily in the metallic state, formed the oxide at minimum readily, whereas tin, lead, antimony, copper, and bismuth, all of which volatilized with much greater difficulty than did arsenic, formed oxides at maximum by simple calcination.
Proust demonstrated that there were no oxides at intermediate proportions, either by showing that some of the unoxidized metal was mixed with the oxide by separating it mechanically (as by washing) or by separating mixtures of the two oxides through use of their different chemical properties. He also corroborated his results with mineral analyses: in the case of iron ores, for example, he claimed always to obtain oxides at either the minimum or maximum proportion, never at intermediate ones. Proust did express some uneasiness over his methodology and reasoning in the separation of the oxides in the latter case:
Does one say that it is the means of analysis which changes the state of the oxide for another which did not exist in the mineral? I reply … that insofar as this has not been demonstrated, it would [best] conform to the true principles of science to admit nothing provisionally, or beyond that which the facts set forth at present.8
More difficult than the oxides for Proust to defend—and more openly attacked by Berthollet—were his views on metallic sulfides, particularly pyrite ores and the various antimonial compounds and complexes (glasses, livers, and others). Proust contended that all metals and semimetals, except iron, formed only one sulfide each; Berthollet had called the sulfides that occurred in variable proportions and the antimonial sulfide-oxide complexes “solutions”—and he considered them to be undifferentiable from other types of compounds. Proust seized on this use of “solution” to distinguish sharply these as well as other types of solutions (including salts in water) from true chemical compounds at fixed proportions. It was over this issue of sulfides that Proust was led to make this important distinction. But when challenged by Berthollet to provide criteria for distinguishing between “true” compounds and what Berthollet called “solutions,” Proust was repeatedly unable to oblige.
Proust’s most detailed consideration of the nature of chemical combination was given in answer not to Berthollet, but to a critic of Haüy’s fixed mineral species concept. Proust’s own view was that minerals were complexes of metallic oxides, sulfides, and other simple binary compounds. The binary compounds were the true combinations; the mineral complexes were merely “secondary assemblages of mineralizations, united and dissolved in all sorts of proportions”
This argument raised the question of what was a true combination. The heart of Proust’s answer was that real compounds “have been given to us only under the rigorous condition of one proportion or two at most,” whereas other types of intermixtures, such as solutions, can be obtained “in a latitude of proportions the extremes of which are infinitely separated.”9
The circularity of Proust’s definition of true “combination” and his failure to provide criteria for his distinction between compounds at definite proportions and other homogeneous systems have been noted in the literature on the Berthollet-Proust controversy. Proust was doubtless at a disadvantage here vis-à-vis Berthollet; on the other hand, his intuitive sensitivity to these practical distinctions was surer than Berthollet’s.
The immediate impact of Proust’s work can be considered under two related questions: Did Proust triumph over Berthollet and convince contemporary chemists that he was right? What was the relation of Proust’s work to the development and implementation of the chemical atomic theory?
With regard to the first question, former historians of chemistry, perhaps seduced by the neat historical and logical relationship between Proust’s law and Daltonian chemical atomism, were inclined to credit Proust with a more decisive victory over Berthollet and more general success in imposing his conviction on the chemical community of his time than the evidence warrants. The series of exchanges with Berthollet ended in 1807, unmarked by any obvious triumph on either side; each protagonist continued to maintain his position thereafter. Proust had scored notable successes against Berthollet’s physicalistic explanation for the appearance of compounds at fixed proportions, but Proust remained unable to answer Berthollet’s challenge to produce criteria, other than definite proportions, for “true” chemical combination.
The evidence in Proust’s favor in the chemical community during the controversy is not much clearer, at least concerning the particulars of his position. Proust’s analyses and the ensuing controversy with Berthollet had aroused interest in the numbers of existent metallic oxides and sulfides and their weight proportions, but it is difficult to find any kind of agreement between the various published analyses— much less agreement with Proust’s particular position. An index to the reception of Proust’s work and views is found in the early editions of Thomas Thomson’s A System of Chemistry (1802; and ed., 1804). In these editions Thomson, who a few years earlier had proposed a version of the constant saturation principle and who, in the third edition of the textbook, was to give the first published account of Dalton’s atomic theory, in fact adopted a compromise position between Proust and Berthollet (before their controversy had even erupted!). Recognizing that only a limited number of metallic oxides and their derivative compounds could be formed, Thomson cited Proust but disregarded his limit of two oxides per metal; and in later theoretical parts of the text, he explained these stable proportions in Bertholletian physicalist terms.10
The evidence of Proust’s influence on Dalton is disappointing. While Dalton was certainly in a position to know of Proust’s work and of the controversy with Berthollet. there is no evidence that Proust’s work played any role in the genesis of Dalton’s chemical atomic theory—even though Dalton was engaged in controversy with Berthollet over the issue of mixture versus true chemical compound (the air) in the early years of the Berthollet-Proust controversy, and Dalton’s theory was to provide the theoretical foundation for Proust’s position.
Indeed, the first published attempt to incorporate Proust’s findings systematically into the atomic theory was not by Dalton but by Thomson. With suitable caution, not to say reluctance, Thomson introduced into the third (and Daltonian) edition of his textbook (1807) a short table exhibiting what he took to be multiple proportions for ten metals and semimetals.11
In the table, even though he derived much of his data from Proust’s analyses, Thomson gave Proust no particular recognition. Moreover, one can understand Thomson’s reluctance to have much confidence in the results of his table, given some of the lengths to which he had to go in order to square his data with multiple proportions. For example, in the case of the three oxides of lead, Thomson had to take the mean of his and Proust’s weight proportions for two of them in order to arrive at reasonably simple proportions.
Dalton gave Proust very scant notice in the first two volumes of A New System of Chemical Philosophy (1808, 1810). It was apparently left to Berzelius (1811) to give Proust his due by establishing the logical relationship between Proust’s work (and the controversy with Berthollet) and Daltonian atomism, as well as to give Proust proper credit for the law of definite proportions.12
When Proust was confronted with Berzelius’ post-Daltonian stoichiometry in Thenard’s Traité de chimie élémentaire, théorique et pratique, I and II (1813-1814), he reacted adversely—partly out of pique at Thenard’s attribution of the law of definite proportions to Berzelius—and passed over the brief exposition of the law of multiple proportions in this textbook, describing it merely as “another relationship” between sulfides.13
Second to his mineral and inorganic analyses associated with the law of definite proportions, Proust’s most important work lay in organic chemistry and chemical technology. In 1799 he succeeded in isolating grape sugar, and he suggested that it could be manufactured and used to supplement the relatively expensive and uncertain supplies of cane sugar from the West Indies. After Proust’s return to France and publication of his researches there, this idea attracted the attention of the Napoleonic government, which in 1810 offered him 100,000 francs to help establish a factory for grape sugar production. In 1818 he announced the discovery of what he called oxide caséeux (leucine) in cheese, which shortly afterward was independently discovered by Braconnot. He also devoted considerable time to attempts to prepare nutriments that could be cheaply made and conveniently stored.
1. R.Dossie, Institutes of Experimental Chemistry, I (London, 1759), 11–13; [G. Vend], “Mixte et mixtion,” in Encyclopédie ou dietiomaire raisonné des sciences des arts et des métiers, X (Paris, 1765), 587; L. B. Guy ton de Morveau, “Affinité,” in Encyclopédie méthodique. Chymie, pharmacie et métallurgies I (Paris, 1786–1789), 560–563; T. Thomson, “Chemistry”; in Encyclopaedia Britannika 3rd ed.f supp., I (Edinburgh, 1801), 342–344.
2. Dossie, op. cit., 11.
3. See S. Mauskopf, “Minerals, Molecules and Species” in Archives internationales d’histoire des sciences.23 (1970), 185–206.
4. “Recherches sur le bleu de Prusse,” in Journal de physique, 45 (1794), 334–335, 341.
5. “Sur les oxidations métalliques,” ibid., 59 (1804), 329.
6. “Recherches sur le cuivre,” in Annales de chimie, 32 (1799), 31.
7. “Lettre de M.Proust à M.d’Arcet sur un sel phosphorique calcaire naturel,” in Journal de physique, 32 (1788), 241.
8.Ibid., 59 (1804), 332.
9. “Sur les mines de cobalt, nickel et autres” ibid., 63 (1806), 369–370.
10. T. Thomson, A System of Chemistry (Edinburgh, 1802), I, 230–231, III, 196–203.
11.A System of Chemistry, 3rd ed. (Edinburgh, 1807), III, 520–522.
12. J. J. Berzelius, “Essai sur les proportions déterminées dans lesquelles se trouvent rétinis les éléments de la nature inorganique,” in Annates de chimie, 78 (1811),5.
13. “Sixième lettre sur l’incertitude de quelques oxidations,” in Journal de physique, 81 (1815), 258.
I. Original Works. MSS of chemical lectures by Proust, which L. Silván dates from Proust’s tenure at Segovia, are in the archives of the Diputación Provincial de Guipúzcoa in San Sebástian, Spain. A set of nine letters to Proust (including some by Berthollet) is in the archives of the Academie des Sciences; all are from 1810 or later. See also H. David, “Une correspondancc inédite du grand chimiste, Joseph Louis Proust,” in Revue d’histoire de la pharmacie (1938), 266–279. Unfortunately, David only paraphrases the letters, some of which date from the 1790’s. Proust also edited Anales del Real laboratorio de quimica de Segovia (1 and 2 , 1791 and 1795) and was an editor of Anales de historia natural1 and 2 (1799–1800); title changed to Anales de ciencias naturales, 3–7 (1801–1804), contributing to both. Proust published many papers, most of them in either Annales de chimie or Journal de physique. Some were first published in Spain. Adequate lists of them (although not complete or without errors) are in Poggendorff, II , cols. 536–538; and Royal Society Catalogue of Scientific Papers, V, 31–33.
II. Secondary Literature. There is still no adequate general study of Proust and his place in the history of chemistry. With regard to biography, there are studies by both French and Spanish authors. The French tend to be ignorant about his Spanish career; the Spanish tend to be sketchy about his life in France. The following are useful but often erroneous: H. David, “Une correspondance inédite”; and M. Godard-Faultrier, Notice biographique sur le chimiste J. L. Proust (Angers, 1852). Two opposing appraisals of Proust’s career in Spain are J. Rodriquez Carracido, Estudios históricocriticos de la ciencia española (Madrid, 1897), 153–166 (negative); and J. Rodriquez Mourelo, “L’oeuvre de Proust en Espagne,” in Revue scientifique, 54 (1916), 257–266 (positive). More recent studies on Proust by Leandro Silvan are “Proust en Vergara,” in Boletin de la Real sociedad vascongada de amigos del pais, 1 (1945), 237–247; Los estudios cientificos en Vergara a fines del sigh XVIII. Monografia Vascongada no. 12 (San Sebastián, 1953); and El quimico Luis José Proust (Vitoria, 1964), particularly rich in information on Proust’s Spanish career. Background studies on the law of definite proportions and the Berthollet-Proust controversy are H. Gueriac, “Quantification in Chemistry” in Quantification: A History of the Meaning of Measurement in the Natural and Social Sciences (Indianapolis, 1961), 64–84; and S. Mauskopf, “Thomson Before Dalton,” in Annals of Science, 25 (1969), 229–242, on the principle of constant saturation proportions.
The best and most recent study on the law of definite proportions and the controversy is S. C. Kapoor, “Berthollet, Proust and Propositions,” in Chymia, 10 (1965), 53–110, although Kapoor’s focus is on Berthollet, not Proust. R. Hooykaas, “The Concept of “Individual’ and “Species’ in Chemistry,” in Centaur us, 5 (1958), 307–322, is oriented more toward the doctrine of fixed mineralogical species. J. R. Partington, A History of Chemistry, III, 640–650, has much useful information.
The following older studies are still very valuable: I. Freund, The Study of Chemical Composition (Cambridge, 1904), 127–143; and A. N. Meldrum, “The Development of the Atomic Theory: (1) Bertholle’s Doctrine of Variable Proportions,” in Memoirsand Proceedings of the Manchester Literary and Philosophical Society, 54, no. 7 (1910), 1–16.