¹ The term “calorique” for the fluid of heat was first introduced in de Morveau, L. B. Guyton, Lavoisier, A. L., Berthollet, C. L. and de Fourcroy, A. F., Méthode de Nomenclature Chimique (Paris, 1787), 31.Google Scholar

² Petit married Arago's sister in November 1814.

³ In fact Arago became permanent secretary of the Académie des Sciences only in 1830.

⁴ For evidence of Dulong and Petit's fear of plagiarism see Berzelius's letter to Alexandre Marcet, 27 April 1819, in Jac. Berzelius Bref, ed. Söderbaum, H. G. (Uppsala, 1912–1932), i, Part 3, 193Google Scholar. This is cited hereafter as Berzelius, Bref.

⁵ Dumas, J. B. A., “Éloge historique de Henri-Victor Regnault”, Mém. Acad. Sci., xlii (1883), p. xlviii.Google Scholar

⁶ Petit, A. T. and Dulong, P. L., “Sur quelques points importants de la théorie de la chaleur”, Ann. Chim. Phys., x (1819), 395–413Google Scholar. Dulong, who seems to have overcome any hesitation he may have felt by 12 April, is identified as the reader in Académie des Sciences. Procès-verbaux des Séances de l'Académie tenues depuis la fondation de l'Institut jusqu'au mois d'août 1835 (Hendaye, 1910–1922), vi, 437Google Scholar. This is cited hereafter as Procès-verbaux.

⁷ Petit, and Dulong, , op. cit. (6), 405.Google Scholar

⁸ All biographers of Dumas state that he came to Paris for the first time in 1823, some three years after Petit's death. This strongly suggests that Dumas's account of the discovery of Dulong and Petit's law originated with Arago.

⁹ Such an opinion would not have been entirely original. In Thomas Thomson, 's Annals of Philosophy, ix (1817), 428Google Scholar, J. B. Emmett, a minor English writer, had already proposed that specific heats should be measured not for equal weights or volumes but for weights in proportion to the atomic (or molecular) weights of the substances concerned, but there is no evidence that Dulong and Petit knew of this. They certainly did know, however, of a result somewhat similar to their own law which had been derived by John Dalton in 1808, but any possibility that Dalton was a precursor is easily ruled out (see below, section 7).

¹⁰ The latter is the view taken by Charles Laurens in the most detailed of the existing biographies of Dulong. See Girardin, J. and Laurens, C., Dulong de Rouen. Sa Vie et ses Ouvrages (Rouen, 1854), 67.Google Scholar

¹¹ Despite the first restoration of the Bourbon monarchy in 1814, it was not until March 1816, after the second restoration, that the title Académie Royale des Sciences was restored.

¹² The subject appears on page 2 of a pamphlet announcing the subjects for prizes to be awarded by the First Class of the Institute in 1816 and 1817. The pamphlet was printed for the public meeting held on 9 January 1815. An announcement also appeared in Hachette, J. N. P., Correspondance sur l'École Polytechnique, iii (issue for 05 1815), 250.Google Scholar

¹³ Fourier's paper appeared (in two parts) only in 1824 and 1826, as “Théorie du mouvement de la chaleur dans les corps solides”, Mém. Acad. Sci., iv (1819–1820), 185–555, and v (1821–1822), 153–245Google Scholar. For comments on the remarkable delay in publication, which seems to have been the result of criticisms made by the influential judges for the competition (Laplace, Lagrange, Legendre, Malus and Haüy), see Darboux, G. in Oeuvres de Fourier (Paris, 1888), i, pp. vi–viiiGoogle Scholar, and Arago, D. F. J., “Éloge historique de Joseph Fourier”, Mém. Acad. Sci., xiv (1838), cxii–cxiii.Google Scholar

¹⁴ Fourier, , Mém. Acad. Sci., iv (1819–1820), 202–203Google Scholar. The assumption that Newton's law could be extended from the case of a hot body losing heat in a cooler surrounding medium to the case of heat flow between two points in a solid is, of course, unfounded. Fourier's cursory and quite inadequate attempt to justify the assumption appears on pp. 200–201 of the first part of his paper as published in the Academy's Mémoires.

¹⁵ Although much of the evidence against the law appears inconclusive to modern eyes. See Martine, G., Essays Medical and Philosophical (London, 1740), 233–247Google Scholar; Erxleben, J. C. P., Novi Commentarii Societatis Regiae Scientarum Gottingensis, commentationes physicae et mathematicae classis, viii (1777), 74–95Google Scholar; Dalton, J., A New System of Chemical Philosophy (Manchester, 1808), i (Part 1), 12 and 108–123Google Scholar. The work with the greatest influence in France, however, was probably that of Delaroche, François, published in the Journal de Physique, lxxv (1812), 201–228Google Scholar. Although Delaroche's paper was read to the Institute on 3 June 1811, it was not published until September 1812 and so was in all likelihood unknown to Fourier, who was working in Grenoble, at the time he submitted his entry on 28 September 1811.

¹⁶ For biographical details see Biot, J. B., “Notice historique sur M. Petit”, Ann. Chim. Phys., xvi (1821), 327–335Google Scholar. This biographical sketch also appeared in Journal de Physique, xcii (1821), 241–248.Google Scholar

¹⁷ See Biographie Nouvelle des Contemporains, ed. Arnault, A. V., Jay, A., Jouy, E. and Norvins, J. (Paris, 1823), ix, 57Google Scholar (article “J. H. Hassenfratz”). The reasons for Hassenfratz's resignation which have been cited here are given by the Director of Studies at the École in a MS. note dated 20 October 1815. Another note, in the report of the Conseil de Perfectionnement of the École and dated 30 October 1815, states that Petit took over from Hassenfratz on 1 September 1814. Both of these items are in the archives of the École Polytechnique and were consulted on my behalf by the librarian, Monsieur A. Moreau.

¹⁸ From information on Dulong's record card at the École Polytechnique. For other biographical details see Girardin and Laurens, op. cit. (10).

¹⁹ For a detailed account of the society which met at Arcueil and of the work which was performed there see Crosland, M. P., The Society of Arcueil (London, 1967).CrossRefGoogle Scholar

²⁰ Dulong, P. L., “Mémoire sur une nouvelle substance détonnante”, Mémoires de Physique et de Chimie de la Société d'Arcueil, iii (1817), 48–63.Google Scholar

²¹ See pp. cxcviii–cc of Georges Cuvier's report on the work of the First Class for the period 1813–15 (physical sciences) in Mémoires de la Classe des Sciences Mathématiques et Physiques de l'Institut de France (1813, 1814, 1815).Google Scholar

²² Dulong, P. L. and Petit, A. T., “Recherches sur les lois de dilatation des solides, des liquides et des fluides élastiques, et sur la mesure exacte des températures”, Ann. Chim. Phys., ii (1816), 240–263Google Scholar. Dulong was the reader (see Procès-verbaux, v, 514).Google Scholar

²³ Dulong, and Petit, , op. cit. (22), 241–242.Google Scholar

²⁴ Ibid., 243 and 263. For Gay-Lussac's work see Ann. Chim., xliii (1802), 137–175.Google Scholar

²⁵ Dulong, and Petit, , op. cit. (22), 263n.Google Scholar

²⁶ On the effect of the unrest see Pinet, G., Histoire de l'École Polytechnique (Paris, 1887), 74–102Google Scholar and Berthollet's letter to Berzelius, 27 August 1815, in Berzelius, , Bref, i, Part i, 54–55Google Scholar

²⁷ Fourcy, A., Histoire de l'École Polytechnique (Paris, 1828), 331.Google Scholar

²⁸ The decision to publish at this stage was almost certainly connected with Dulong's (unsuccessful) candidature for a place in the physics section of the First Class. See Crosland, , op. cit. (19), 167.Google Scholar

²⁹ A decision which was fully justified, if we are to judge by the two insubstantial entries which have survived in the archives of the Académie des Sciences.

³⁰ Procès-verbaux, vi, 164 (17 03 1817)Google Scholar and Ann. Chim. Phys., iv (1817), 302–303Google Scholar. In fact the wording of the subject was slightly modified for the new competition. It was now specified that the observations of the mercury-in-glass thermometer should be made “comparativement à la marche du thermomètre à air” and between 20°C and 200°C rather than between 0°C and 200°C. Otherwise the wording was identical.

³¹ Dulong, P. L. and Petit, A. T., “Recherches sur la mesure des températures et sur les lois de la communication de la chaleur”, Ann. Chim. Phys., vii (1818), 113–154, 224–264 and 337–367Google Scholar. That the importance and high quality of the work were immediately recognized in England as well as in France is apparent from the fact that Thomson, 's Annals of Philosophy, xiii (1819)Google Scholar, contained not only a full translation but also a lengthy, and generally favourable, criticism by Thomson himself (on pp. x–xiv, xvi–xviii and xix–xxiii).

³² Comte, A., Cours de Philosophie Positive (Paris, 1835), ii, 534Google Scholar; Poisson, S. D., Théorie Mathématique de la Chaleur (Paris, 1835), 6Google Scholar; Lamé, G., Cours de Physique (Paris, 1836), i, pp. i–iiGoogle Scholar; Whewell, W., A History of the Inductive Sciences (London, 1837), ii, 485.Google Scholar

³³ If Dulong and Petit had used Berzelius's most recent values for atomic weights, as given for example in Thomson, 's Annals of Philosophy, iii (1814), 362–363Google Scholar, they would have found that the atomic heats of no fewer than five of the seven metals whose specific heats they had determined were very nearly constant. An equally striking result would have been obtained if they had used John Dalton's figures given in the latter's New System (Manchester, 1810), i (Part 2), 546.Google Scholar

³⁴ Procès-verbaux, vi, 292 (2 03 1818).Google Scholar

³⁵ These are most easily referred to in Berzelius, J. J., Essai sur les Proportions Chimiques (Paris, 1819)Google Scholar, a translation of part of his Lärbok i Kemien (Stockholm, 1818), iii.Google Scholar

³⁶ In their choice of the elements whose atomic weights should be left unchanged, Dulong and Petit were probably influenced by Berzelius, who had long been convinced that the molecular formulae of compounds containing oxygen were analogous to those in which the oxygen was replaced by sulphur. Thus the quantities of oxygen and sulphur in pairs of analogous compounds yielded atomic weights for those elements which were thought not to be subject to the usual uncertainty concerning the numbers of different atoms in compounds. On this see especially Berzelius's letter to Berthollet, in Ann. Chim., lxxvii (1811), 68Google Scholar. Although I have assumed here that Berzelius's value for the atomic weight of platinum was not modified for the purposes of the 1819 paper, it should be pointed out that the precise figure used by Dulong and Petit is in some doubt. In their table of results (reproduced opposite) the weight for platinum appears as 11.16, whereas in his Essai Berzelius gives it as 12.15 (the atomic weight of oxygen on this scale being unity). Since the product 11.16 × 0.0314 is in fact 0.3504 and not 0.3740, as stated in the last column of the table, it is clear that the figures given for platinum are not to be relied upon and that some error, possibly involving a misprint of the atomic weight of platinum, was made. For a discussion of some possible explanations of this curious mistake see Van Spronsen, J. W., “The history and prehistory of the law of Dulong and Petit as applied to the determination of atomic weights”, Chymia, xii (1967), 165Google Scholar. The point remains, however, that a simple fraction of Berzelius's atomic weight for platinum was certainly not used.

³⁷ On the importance of Dulong and Petit's law in removing this doubt see note 86.

³⁸ For a detailed account of the “Irvinist” doctrines, with full references to the work of the leading “Irvinists” and an illustration (from John Dalton's New System of Chemical Philosophy) of the cylinders of caloric referred to below, see my paper, entitled “Dalton's caloric theory”, in Cardwell, D. S. L. (ed.), John Dalton and the progress of science (Manchester 1968), 187–201.Google Scholar

³⁹ Essays, chiefly on Chemical Subjects by the late William Irvine, M.D., F.R.S.Ed., and by his son William Irvine, M.D. (London, 1805), 71.Google Scholar

⁴⁰ Delaroche, F. and Bérard, J. É., Ann. Chim., lxxxv (1813), 169–176Google Scholar. This paper by Delaroche and Bérard won the Institute's prize competition in physics for 1812.

⁴¹ The main sources for the opinions of Lavoisier and Laplace are their “Mémoire sur la chaleur”, Mém. Acad. Sci. (1780), 355–480Google Scholar, and their “3^e mémoire … contenant les expériences faites sur la chaleur, pendant l'hiver de 1783 à 1784”, in Lavoisier, 's Mémoires de Chimie (Paris, 1805), 121–147Google Scholar. The latter paper was written in 1793. In the former paper (see especially p. 358) Lavoisier and Laplace were careful not to commit themselves on the question of the nature of heat, although Laplace later became a convinced calorist. In Britain, Joseph Black was usually cited as the leading exponent of the view that bodies contained both latent and sensible heat and in his Lectures on the Elements of Chemistry, ed. Robison, J. (Edinburgh, 1803), i, 193–197Google Scholar, he appears as a critic of the Irvinist doctrine that changes in specific heat could cause the emission or absorption of heat. But, despite his undoubted priority with regard to much in the opinions of Lavoisier and Laplace on heat, Black is of less importance here, concerned as we are with an episode in French science.

⁴² “Sensible” and “latent” were just two of a number of terms in use at this time to denote the states of caloric. In the second of the papers cited in note 41, for example, Lavoisier and Laplace used the adjectives “interposé” and “combiné”.

⁴³ Lavoisier, , Mémoires de Chimie, 138–147.Google Scholar

⁴⁴ Dulong and Petit made only one brief reference to Berzelius in their paper, on p. 412. Berzelius seems to have taken no part in the experimental work (see Berzelius's letter to Alexandre Marcet, 27 April 1819, in Berzelius, , Bref, i, Part 3, 193).Google Scholar

⁴⁵ The result was a joint paper by the two men in the Ann. Chim. Phys., xv (1820), 386–395.Google Scholar

⁴⁶ In Partington, J. R., A History of Chemistry (London, 1964), iv, 168CrossRefGoogle Scholar, it is stated that Berzelius adopted Lavoisier's theory of chemical heat in volume i of the first edition of his Lärbok i Kemien (Stockholm, 1808)Google Scholar, but a copy of the latter work has not yet been traced.

⁴⁷ Gilbert, L. W.'s Annalen der Physik, xxxvii (1811), 278–280Google Scholar, and Annales de Chimie, lxxix (1811), 249–251Google Scholar. Cf. the similar criticism made independently by Augustin Fresnel in a letter to his brother Léonor, 5 July 1814, in Oeuvres Complètes d'Augustin Fresnel (3 vols., Paris, 1866–1870), ii, 820–821.Google Scholar

⁴⁸ Phil. Trans., xcvii (1807), 42–44.Google Scholar

⁴⁹ Ibid., xcix (1809), 71–72.

⁵⁰ Berzelius, , Essai, 58–73Google Scholar. This is a translation, with only minor modifications, of pp. 49–63 of vol. iii of the Lärbok. See note 35 for bibliographical details.

⁵¹ Schweigger, J. S. C.'s Neues Journal für Chemie und Physik, vi (1812), 139–141Google Scholar. These comments appeared in French in Ann. Chim., lxxxvi (1813), 168–171.Google Scholar

⁵² Described in Davy, H., Elements of Chemical Philosophy (London, 1812), 95–96.CrossRefGoogle Scholar

⁵³ Berzelius to Davy, n.d., in Berzelius, , Bref, i, Part 2, 41.Google Scholar

⁵⁴ Berzelius, , Essai, 68–73Google Scholar. On the date of publication see Berzelius's letter to Alexandre Marcet, 1 June 1819, in Berzelius, , Bref, i, Part 3, 195Google Scholar. I am indebted to Dr. C. A. Russell for the latter reference.

⁵⁵ Mém. Acad. Sci., iv (1819–1820), pp. lxxxi–lxciii.Google Scholar

⁵⁶ On the work of Despretz, see note 67.

⁵⁷ Dulong to Berzelius, 15 January 1820, in Berzelius, , Bref, ii, Part 1, 13–14.Google Scholar

⁵⁸ Cf. the term “excited” used by Rumford to express a similar idea in Phil. Trans. (1798). 88 and 99.Google Scholar

⁵⁹ The originality of Dulong's interpretation is apparent only in the light of contemporary opinion, which was almost entirely committed to a “calorist” explanation of the heat produced in the rapid compression of a gas. On the latter see Kuhn, T. S., “The caloric theory of adiabatic compression”, Isis, xlix (1958), 132–140.CrossRefGoogle Scholar

⁶⁰ The Dictionnaire de Biographie Française (Paris, 1933—in progress), contains convenient biographical sketches of Clément and Desormes. Clément appears as Clément-Desormes, the name which he adopted after his marriage to Desormes's daughter.

⁶¹ Both men, moreover, had been candidates for admission to the Académie des Sciences on more than one occasion and as recently as 5 July 1819 Desormes had finally secured election as a corresponding member in the chemistry section (see Procès-verbaux, vi, 466).Google Scholar

⁶² The paper was never published, but an extract appeared in the Bulletin des Sciences par la Société Philomathique de Paris for 08 1819, pp. 115–118.Google Scholar

⁶³ On the discrediting of their results see below, and also notes 67 and 68.

⁶⁴ On earlier versions of this “law” proposed by James Watt and J. N. P. Hachette, see Black, , op. cit. (41), i, 190Google Scholar, and “Tableau de M. Clément-Desormes, relatif à la théorie générale de la puissance mécanique de la vapeur (extrait)”, Nouveau Bulletin des Sciences par la Société Philomathique de Paris (1826), 53Google Scholar. The extract in the Bulletin was prepared by Hachette, who signed the article.

⁶⁵ Given in Gay-Lussac, op. cit. (24).

⁶⁶ For an argument similar to that outlined here, in which the data for saturated vapour pressures are taken from Biot, J. B.'s Traité de Physique Expérimentale et Mathématique (Paris, 1816)Google Scholar, see Carnot, N. L. S., Réflexions sur la Puissance Motrice du Feu (Paris, 1824), 67–68n.Google Scholar

⁶⁷ In his Traité Élémentaire de Physique (1st edn., Paris, 1825), 100Google Scholar, G. M. Despretz stated that he had shown experimentally that Clément and Desormes, 's “law” could be applied to a number of vapours other than steam, but in the second edition of the work (Paris, 1827), pp. 113–114Google Scholar, he wrote that he had now disproved the law. In the pamphlet, Résumé des Travaux de Physique de M. Despretz (Paris, n.d. but, according to internal evidence, probably 1828), 3–4Google Scholar, Despretz described how he had announced his revised conclusions before the Société Philomathique.

⁶⁸ Despretz, C. M., Traité Élémentaire de Physique (4th edn., Paris, 1836), 212Google Scholar (and note) and Lamé, , op. cit. (32), i, 487Google Scholar. The former reference suggests that Dulong's experiments were performed at about the same time as Despretz's, since Despretz states that Dulong had supported him in his criticisms of the “law” during a discussion with Clément which had taken place at the Société Philomathique “some years” earlier. Dulong's experiments may well date from the period 1824–1829 when he was engaged in a great deal of work on vapour pressures and related matters in connection with a commission set up by the Académie.

⁶⁹ See M. P. Crosland, “The first reception of Dalton's atomic theory in France”, in Cardwell, , op. cit. (38), 274–287.Google Scholar

⁷⁰ Dulong, P. L., “Recherches sur la décomposition mutuelle des sels insolubles et des sels solubles”, Ann. Chim., lxxxii (1812), 273–308.Google Scholar

⁷¹ Ann. Chim. Phys., i (1816), 411–416Google Scholar. Prout's papers had appeared in November 1815 and February 1816, in Thomson, 's Annals of Philosophy, vi (1815), 321–330 and vii (1816), 111–113.Google Scholar

⁷² See Dulong's letter to Berzelius, 8 January 1822, in Berzelius, , Bref, ii, Part 1, 36–37Google Scholar. Among the other members of the board were Berthollet, Chaptal, Gay-Lussac, Thenard, Chevreul, Biot and Arago.

⁷³ Dulong, P. L., “Mémoire sur les combinaisons du phosphore avec l'oxigène”, Mémoires de Physique et de Chimie de la Société d'Arcueil, iii (1817), 405–452Google Scholar. The paper was read on 1 and 15 July 1816.

⁷⁴ Procès-verbaux, vi, 101–103 (21 10 1816).Google Scholar

⁷⁵ Ibid., vi, 103. Berthollet's paper was read immediately after the report on Dulong's work. Crosland states (op. cit. (69)), that the paper is now missing from the archives of the Académie des Sciences.

⁷⁶ But even in 1816 there is evidence that opinions were beginning to change. For example, although the main text of the first edition of Thenard, L. J.'s Traité de Chimie Élémentaire, Théorique et Pratique (Paris, 1813–1816)Google Scholar contained no mention of the atomic theory, a translation by Collet-Descotils of W. H. Wollaston's celebrated paper on 1814 on chemical equivalents was included as an appendix, on pp. 247–254, to the last (fourth) volume, published early in 1816 (on this date see Procès-verbaux, vi, 33 (4 03 1816))Google Scholar. By 1817, when the first volume of the second edition appeared, Thenard revealed a very sympathetic attitude to Dalton's theory (see Traité (2nd edn.), i, 21–23).Google Scholar

⁷⁷ Ann. Chim. Phys., ii (1816), 320–339Google Scholar. The first part of this lengthy paper had appeared in June 1816 in the same issue of the Annales as a shortened version of Dulong's paper on the subject.

⁷⁸ See, for example, Ann. Chim. Phys., ii (1816), 174n.Google Scholar

⁷⁹ This is evident also in a letter of 5 August 1816 from Dulong to A. M. Ampère quoted in Lemay, P. and Oesper, R. E., “Pierre Dulong, his life and work”, Chymia, i (1948), 175.Google Scholar

⁸⁰ See Berzelius's letter to Dulong, 5 November 1819, in Berzelius, , Bref, ii Part 1, 10–11.Google Scholar

⁸¹ It is interesting that Dulong should have chosen to mention Laplace in this context, when Berthollet was generally recognized as the leading opponent of the atomic theory in France. It seems likely that a genuine personal attachment to Berthollet (on which see Dulong's letter to Berzelius, 20 November 1823, in Berzelius, , Bref, ii, Part 1, 45)Google Scholar mitigated Dulong's criticisms.

⁸² Dulong to Berzelius, 15 January 1820, in Berzelius, , Bref, ii, Part 1, 12.Google Scholar

⁸³ See, for example, Buchdahl, G., “Sources of scepticism in atomic theory”, The British Journal for the Philosophy of Science, x (1959), 120–134.CrossRefGoogle Scholar

⁸⁴ On scepticism in England, with special reference to Davy, see Brock, W. H. and Knight, D. M., “The atomic debates: memorable and interesting evenings in the life of the Chemical Society”, Isis, lvi (1965), 5–25Google Scholar. For convenient statements of the influential views of Berthelot, P. E. M., see his La Synthèse Chimique (2nd edn., Paris, 1876), 154–171Google Scholar, and also Comptes Rendus hebdomadaires des Séances de l'Académie des Sciences, lxxxiv (1877), 1189–1195.Google Scholar

⁸⁵ In his Leçons sur la Philosophie Chimique (Paris, 1837), 271Google Scholar, J. B. A. Dumas cited the discovery of Dulong and Petit's law as the “beau moment” of the atomic theory. For a similar view see, for example, Wurtz, A., Leçons de Philosophie Chimique (Paris, 1864), 24.Google Scholar

⁸⁶ The approximate nature of the law, which was soon widely recognized, greatly restricted its usefulness in this respect, but the law did allow a choice to be made in cases where there was uncertainty concerning the number of atoms of various elements making up a “compound atom”. Berzelius, for example, had halved many of his atomic weights by 1826 in the light of the law and also of Mitscherlich's work. See Poggendorf, J. C.'s Annalen der Physik und Chemie, vii (1826), 414.Google Scholar

⁸⁷ Despite their disagreement with many of his opinions, Dulong and Petit seem to have held Dalton in genuine esteem. A courteously inscribed copy of their prize-winning memoir of 1818, which they sent to Dalton, is now in the Burndy Library, Norwalk, Conn. I wish to thank Mrs. A. Matthysse, Librarian of the Burndy Library, for providing me with a description of this copy of the memoir.

⁸⁸ Dalton, , New System (Manchester, 1827), ii, 280–281 and 293–297.Google Scholar

⁸⁹ Petit, and Dulong, , op. cit. (6), 406–407Google Scholar. Dulong and Petit almost certainly believed, with Berzelius (see Berzelius, , Essai, 52–55)Google Scholar, that equal volumes of all elementary gases at a given temperature and pressure contained the same number of atoms and they would therefore have expected by their law that the volume specific heats of such gases would be identical. This prediction was very nearly borne out by the experimental results given in Delaroche, and Bérard, , op. cit. (40), 157.Google Scholar

⁹⁰ Dalton, , op. cit. (15), 66–75Google Scholar. For a fuller account of Dalton's view see Fox, op. cit. (38).

⁹¹ A claim made in Henry, W. C., Memoirs of the Life and Scientific Researches of John Dalton (London, 1854), 68Google Scholar, and in the Dictionary of National Biography, article “John Dalton”.

⁹² Petit, and Dulong, , op. cit. (6), 407–408.Google Scholar

⁹³ This view of the period 1820–1850 clearly requires more detailed documentation than is possible here, but an illustration of the sort of changes which had taken place since the early years of the century is provided in Thomas Thomson, 's An Outline of the Sciences of Heat and Electricity (London, 1830), 335Google Scholar, where Thomson put the then widely-held opinion that the question of the nature of heat was unsolvable, whereas in his A System of Chemistry (Edinburgh, 1802), i, 259Google Scholar, he had considered the question as settled (in favour of caloric).

⁹⁴ A convenient account of this later work, in which F. E. Neumann, Victor Regnault and Hermann Kopp figured particularly prominently, appears in Freund, I., The Study of Chemical Composition (Cambridge, 1904), 361–384Google Scholar. Dulong himself participated in the work after Petit's death, but with little success.

⁹⁵ The evidence for this is found in a set of notes taken by Auguste Comte at Petit's lectures on physics at the École Polytechnique in the winter of 1814–1815. A copy of the notes was kindly supplied to the author by Monsieur D. Cantemir, archivist of the Maison d'Auguste Comte, Paris.

⁹⁶ For an account of Fresnel's work see Émile Verdet's introduction to Fresnel, , op. cit. (47), i. pp. ix–xcix.Google Scholar

⁹⁷ Details of this and a number of other developments in the period 1815–1820 will be given in a joint paper which the author is preparing in collaboration with Dr. J. R. Ravetz. However, evidence of Dulong's support for Fresnel will be found in the letter of 15 January 1820, from Dulong to Berzelius. quoted in section 5.

⁹⁸ See his “Complément au mémoire sur la diffraction”, dated 10 November 1815, in Fresnel, , op. cit. (47), i, 59–60Google Scholar; also his letters of 5 July 1814 and 11 July 1814 to Léonor Fresnel, ibid., ii, 820–822 and 827–829.