KOSTAS GAVROGLU

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

When discussions about problems and issues in the philosophy of science need examples for corroborating claims, criticisms and alternative schemata, they almost always refer to physics. Though there has been a growing interest in the philosophy of biology and despite the substantial contributions to this field, for many years and for many people the paradigmatic case for philosophy of science has been that of physics. Chemistry has been ignored and part of the explanation accounting for such an absence of chemistry from the philosophical discussions has been the dominance of the question of reductionism. Reductionism has, in fact, been the primordial curse of chemistry: Many scientists (though not necessarily the chemists) felt that every new theoretical success brought chemistry nearer to physics and that almost always a new theoretical schema in chemistry owed its legitimacy because of its closeness to the culture of physics. Hence, in attempting to straightjacket chemistry within a standard philosophical analysis, there is always the danger that one contributes to the further strengthening of the feeling that the content of a philosophical discussion relating to chemistry is more or less dependent on having clarified similar issues in physics. Such clarifications, though they are very useful, should by no means be considered as necessary prerequisites for dealing with the philosophy of chemistry. My purpose in this paper is not to argue that chemistry needs “its own” philosophy in a relatively autonomous manner from that of physics. Nor will I attempt to propose ways standard philosophical problems (such as the issue of meaning or explanation) are particularised in the case of chemistry. What I would like to attempt is to articulate some philosophical and theoretical issues which result from the theoretical particularity of chemistry and chart the space of these philosophical problems which are indigenous to chemistry. Furthermore, these issues are dealt with from a historian’s perspective and such a task appears to be a necessary undertaking for doing history of chemistry. The role of theory in chemistry, the strategies of theory building and the legitimisation of the dominant discourse of the various subdisciplines of chemistry are the three main problems I shall be concerned with. These are not wholly independent from the implications of reductionism and realism for chemistry. Nevertheless, in examining the ways chemists attempted

Synthese 111: 283–304, 1997. c 1997 Kluwer Academic Publishers. Printed in the Netherlands.

284

KOSTAS GAVROGLU

to deal theoretically with the classic problems of chemistry, the historian is invariably confronted with the chemists’ particular attitude on how much one could and should “borrow” from physics in order to construct a chemical theory and what is the methodological status of empirical observations for theory building in chemistry. The choices of the chemists and the schemata they proposed brought into being new research traditions, articulated new strategies of experimental manipulation, implied a different role for mathematics in each tradition, and gave rise to different styles of research within these traditions. It is the confluence of all these processes that eventually became decisive in forming the chemists’ culture. Such considerations may appear to be historical (and to some sociological) questions, and to a certain extent they are. They involve, however, a number of intriguing philosophical and theoretical dimensions. The use of rules, for example, which is a constitutive aspect of the chemists’ culture forms a framework where it becomes possible to accommodate more than one theory. What is interesting in the case of chemistry and which is radically different when compared to physics, is that these theoretical schemata are used in a complementary manner. Finally, and notwithstanding the less or more violent invasions of mathematics into chemistry during the last two centuries and the remarkable successes of such invasions, the chemists’ culture has been conditioned by laboratory practices, their theoretical constructs are always very sensitive to the exigencies of the laboratory, and theory building is strongly dependent on using as inputs the experimentally measured values of various parameters. In other words, the use of semi-empirical methods in constructing theoretical schemata, has always had a far stronger legitimacy in chemistry than in physics. These characteristics instead of being viewed as aspects of the theoretical particularity of chemistry, they are very often being viewed as the shortcomings of chemistry with respect to physics. But such an attitude gives credence to the arguments of those who adopt the reductionist view. Chemistry, it has been claimed, has all these characteristics because it cannot by itself and without becoming physics develop a one “true” theory at any period nor can it fully assimilate mathematics into its practice. If, however, we accept such an argument and consider these characteristics as shortcomings rather than expressions of the theoretical particularity of chemistry, then we should claim that the history of chemistry is to a large extent a long history of devising empirical rules, innovating approximate and phenomenological theoretical schemata and in order to reach a “serious” theoretical level, chemistry is always dependent on physics and follows developments in physics. Such an unquestioned reign of reductionism has had many victims and the historiography of chemistry was

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

285

among the first to suffer. As a result, quite often the history of chemistry has been projected as a partially fulfilled expectation of a long wait for physics to provide the “true” basis for chemistry. But such considerations have no rapport with the history of chemistry. If they did, then the chemists’ self-consciousness would be nothing more than a shared collective false consciousness. A community and its culture cannot be formed on the beliefs that what they think they do is a job others – the physicists – do “more” properly. Many of those who have physicists dreaming of a final and finished theory, by the very same token, have chemists traverse their history by living the nightmare of temporary and unfinished theories. At a Conference in 1977 commemorating half a century of valence theory, Charles Alfred Coulson gave the closing talk. He was a mathematician by training and a writer of what became one of the standard text-books on valence. He claimed that the fifty years of valence theory meant fifty years of changing ideas about the chemical bond. He divided this period into three parts. He put great emphasis on the first third of the period when chemists were concerned with identifying the electronic nature of the bond, and in “escaping from the thought forms of the physicist”. Coulson could not have expressed any better the often misunderstood relationship between chemistry and physics. Borrowing and using concepts first proposed by physicists, and, subsequently, “escaping from the thought forms of the physicists” had really been the dominant trend, not only in the first years of quantum chemistry, but also in much of the history of physical and structural chemistry in the latter part of the 19th century as well. While chemistry was striving to free itself from the “thought forms of the physicist”, the theoretical issues raised by this escape have been, on the whole, ignored by philosophers of science. I am not, of course, saying that there have been no philosophical discussions of the problems of chemistry, since questions about the phlogiston and the Daltonian atoms have been very systematically discussed. But the neglect of the philosophical problems arising out of the particularities of chemistry after the development of microphysics since the last quarter of the 19th century is also a fact. 1. STRONGLY BONDED PARTS: REDUCTIONISM AND REALISM Though it is acknowledged that a reductionist view of biology is rather problematic, there is a rather peculiar uneasiness about chemistry. This is more so if we realise that in the case of chemistry – more than in any other discipline – the issues related to reductionism, scientific realism, and theory building are strongly interrelated and none can be dealt with separately from the rest.

286

KOSTAS GAVROGLU

Paul Dirac’s pronouncement in 1929 that since the underlying laws governing the behaviour of the electrons became known, to do chemistry meant to deal with equations which were in principle soluble even though in practice they may only produce approximate solutions, has often been the starting point of discussions about reductionism in chemistry. Dirac’s comment appeared to have given the misleading impression that the question of reductionism has arisen only after the advent of quantum mechanics. Furthermore, the absoluteness with which Dirac expressed his view and the way it has been raised by some to the status of a dogma, set the tone for the ensuing discussions: the reactions, on the whole, to such a claim that all chemistry is physics has been to strive for an argument about the absolute and full autonomy of chemistry. And as a result what had been neglected was the realisation that the methodological significance of raising the discussion of reductionism in chemistry lies in the implicit understanding that what is at stake is neither the absolute reduction of chemistry to physics nor the absolute autonomy of chemistry. The two extreme cases (full reduction of chemistry to physics and no reduction at all) are both historically wrong, philosophically naive and lead to a methodological deadlock. The challenge lies in the ability to articulate the necessarily intermediate position of the relative autonomy of chemistry with respect to physics (Scerri 1991, 1993). In the discussions about scientific realism, there is an implicitly shared set of values which undermine the possible contributions chemistry can provide to these philosophical issues. It is believed, not unjustifiably, that the special role of mathematics in physics renders the problems of scientific realism to be clearly delineated. It is claimed that physics deals with the fundamental entities of the world and there are no intrinsic limitations as to how deep it could probe. Whether it studies the planets, billiard balls, atoms, nuclei, electrons, quarks or superstrings, it is still physics and the change of scale does not oblige the change, as it were, of the discipline itself – as it would be the case in biology and chemistry. Although, on the whole, I agree with this view, I have two reservations. The first is that it is not too clear to me that when we are discussing quarks or superstrings or the big-bang we are doing theoretical, rather than mathematical physics. Mathematical physics is often confused with theoretical physics, and I do not think that mathematical physics is the mathematically more exact treatments of problems in physics. During various periods, and especially among the British physicists of the dynamic tradition during the second half of the 19th century, mathematical physics has been practised without necessarily a reference to the underlying ontology. And even in those cases that it was possible to construct a model, this was taken to be a feature

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

287

which further justified the mechanical explanation, rather than the model itself being the depiction of the underlying physical structure. My second reservation is that the view which confines the study of realism predominantly to the problems of physics is more a matter of convenience rather than something which has a serious justification. Furthermore, such an attitude neglects the theoretical particularity of chemistry. And what is much more important, it supposes an absolute reductionism of chemistry to physics. If nothing else, these undeclared assumptions deprive philosophy of science of a vast area where issues about the ontological status of theoretical entities and the criteria for empirical adequacy for the acceptance of a theory have been discussed quite systematically. Ever since the end of the 19th century, chemists have been debating whether their science may not be the “science of bodies which do not exist”. Or whether the unsettling discovery of radium implied that “in relation to the ponderable, we seem to be creating a chemistry of phantoms”. The history of chemistry is also a history of the attempts of the chemists to establish its relative autonomy with respect to physics. Hence, and unlike the physicists, the chemists are obliged to proceed to ontological commitments which are unambiguous and clearly articulated, and they have little or no tolerance to an attitude which stipulates that these may be temporary commitments. Otherwise the chemist would be at a loss about the underlying ontology, and would never be sure whether chemistry should be doing the describing and physics the explaining. The chemists have passionately debated these issues, and the myth of the reflective physicist and the more pragmatic chemist is, if anything, historically untenable.

2. THE HIDDEN FACE OF THE CHEMISTS’ DISCOURSE: APPROPRIATING ENTITIES AND DISPOSSESSING CONCEPTS

At the centre of all these theoretical issues are the many facets of the emergent discourse of chemistry and the changes incurring as a result of the becoming of its many and new subdisciplines. Some of the issues related to the emergent discourse of chemistry are the ontological status of theoretical entities, the collective decision of where to stop when searching for building-blocks, the role of empirical data in theory building, the co-existence of more than one theories or theoretical schemata, the question of their corroboration and falsification, the explanatory strength of the theoretical schemata in chemistry, the legitimacy of the ensuing discourse, its reliance less on establishing “objective” criteria to which theories should conform and more on the consensus of the community and the metamorphosis of the praxis of its practitioners. The development of

288

KOSTAS GAVROGLU

chemical thermodynamics offers an interesting instance to discuss some of these issues. It is by no means the only such case and there are many instances where these questions come to the fore. Before discussing in a detailed manner an episode in the development of chemical thermodynamics, let me attempt to give some indications of the ways the previous considerations could be traced in two other cases One such case is the attempt of the British chemists to re-appropriate a concept which had been “snatched” from them by their fellow physicists. From about the mid-1880s the British chemists were expressing an increasing interest in the possibilities provided by the dynamic approach of the British physicists to understand more and more the behaviour of atoms. But the views put forth by the energeticists, the various developments in chemistry where the use of mathematics became a necessary ingredient for doing chemistry and the developments in physics where the understanding of the behaviour of the atoms was at the expense of the “billiard ball” ontology which was so appealing to the chemists obliged the chemists to re-acquisition what they felt was theirs in the first place. The British chemists found themselves between a world of ethereal vortices articulated by their fellow physicists and a world of anti-atomic energetics propagandised by the German physical chemists. In the early 1890s, after the triumphant period of organic chemistry, and in parallel with the developments in inorganic and most significantly, physical chemistry, the chemists were starting to reassert their presence in the discussions from which they had been left out. By the late 1870s the atom, a chemical entity par excellence, had been quite successfully appropriated by the physicists. It was not until the mid-1890s that the chemists started to re-assert their lost jurisdiction over the atomic theory. In his Presidential Address to the Mathematical and Physical Section of the British Association for the Advancement of Science meeting in 1873 James Clark Maxwell had talked about molecules. Though he had acknowledged that the ideas embodied in molecules were those of chemistry, he felt that molecules should not remain the chemists’ prerogative, since, as he said, there was a universal interest in molecules. His suggestion was to reformulate what were considered to be chemical laws in terms of “the laws of motion”. And in his article on the Atom for the 9th edition of the Encyclopaedia Britannica Maxwell, who was also the editor for the physical sciences for the Britannica, gave a long survey of Kelvin’s vortex atoms in the ether and he chose to ignore the Daltonian “billiard-ball” ontology. At the end of the 19th century, both physicists and chemists appealed to history in search for criteria to justify the conflicting ontologies adopted

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

289

by the respective communities: The past was invoked by the physicists to argue that the ether and its dynamical view was the correct and promising trend to follow. The chemists reverted to their reading of the past where the Daltonian atomic hypothesis appeared to be reigning supreme. And the reference of both groups to the history of their own subjects was also to strengthen the view that conflicting ontologies aside, both the dynamical approach as well as atomism were distinctly British trends. The prospects promised by the early successes of the dynamical approach did not leave the chemists uninterested. In fact, in the early 1880s it even appeared that dynamical theory may even provide an explanation of why atoms are held together to form a molecule. In 1882 J. J. Thomson had shown that a complex mechanism of vortex rings could account for the mechanism of valence and proposed that the pairing of two vortex rings is “what takes place when two elements of which these vortex rings are atoms combine chemically”. Many of the dominant figures of the chemical establishment during their public lectures overemphasised the need for co-operation with the physicists. Roscoe in 1884 stressed that one “of the noteworthy features of chemical progress is the interest taken by physicists in fundamental questions of our science” (Roscoe 1884, 659–669 quote on p. 666). In 1885 Henry Armstrong discussed one of the cardinal issues of chemistry which was the understanding of the nature of chemical change, a subject which, he emphasised “requires the immediate earnest attention of chemists and physicists”. In 1888 J. J. Thomson published his lecture notes at the Cavendish under the title Applications of Dynamics to Physics and Chemistry. Though chemistry was cruelly reduced to physics, there did emerge a dynamical framework which could accommodate the physical and chemical phenomena together. In 1893 Emerson Reynolds placed great emphasis on the kinds of problems he called the “physico-chemical” problems. In 1895 Ralph Meldola argued that the “one great desideratum of modern chemistry is unquestionably a physical or mechanical interpretation of the combining capacities of atoms” (Meldola 1895, 639–655 quote on p. 643). But all this talk came to an abrupt end in 1896, the year Wilhelm Ostwald launched his campaign for energetics at the Lubeck meeting of the Society of German Scientists and Physicians. It was this new brand of physical chemistry which made it so “difficult : : : to determine where chemistry left off and where physics began”. Soon afterwards, British chemists started to complain quite vigorously. In 1902 Edward Divers who was the vice-president of the Chemical Society delivered the presidential address to the chemical section of the BAAS. His grudge was against the physicists who had never been “satisfied with the hard, indivisible ball of specific substance and definite mass which

290

KOSTAS GAVROGLU

has served chemistry so well”. He denounced the physicists who had manipulated the atom, making, in fact, ‘even a vortex ring of it’. Divers’ main thrust was that all this meddling with the atoms led to a situation that only a ‘few chemists can understand’ (Divers 1902, 557–576 quote on p. 568). In contrast to the physicists’ practice, the chemists had not meddled with the atom’s interior. Divers argued that chemistry dealt with tangible truths and did not need the help of mechanical models in order to legitimate its various laws. It is important to note that these sentiments were expressed only a few years after all those verbal commitments by the chemists on the common agenda between them and the physicists and they reflected the changing sentiments of the chemists in the way they would pursue their disciplinary politics after the turn of the century. There were two aspects of Divers’ talk which are particularly relevant to my argument. The first was his attempt to have the community agree on an unambiguous ontology if they were to counteract successfully the attacks of the energeticists. The second was the abandonment of the atom and the embrace of the molecule as the basic unit of chemistry, as a way of escaping the disconcerting mess that he felt the atom was facing with the electrons, X-rays and radioactivity. In 1906 two years after his Faraday Lecture in London, Ostwald indicated that he was having second thoughts about his denial of the existence of atoms. The next year, Arthur Smithells, the forceful spokesman of British chemistry, was telling his audience at the BAAS meeting that when compared to 20 years ago, everything appeared to be more promising in chemistry. Though the discovery of radioactivity did mark a new epoch in the history of chemistry, there did not appear to be any reasons to start questioning any of the constitutive aspects of chemistry. His pronouncement was that the chemists were perplexed, not so much by the new ideas, but by the invasion “of chemistry by mathematics”. It appeared that there was a feeling among chemists who were feeling “submerged and perishing in the great tide of physical chemistry which was rolling up into our laboratories” (Smithells 1907, 469–502 quote on p. 477). Smithells concluded by commanding that chemistry should not be invaded by mathematical theorists! In 1909 Henry Armstrong pursued the same aggressive strand. Concerning his views about the energeticists, Armstrong boasted that, even though his attitude was one “complete antagonism towards the speculations of the Ostwald school”, he claimed, nevertheless, to have been the first English chemist to publicly remark that Ostwald’s investigations were of the highest importance. But now Ostwald had changed his mind, and Armstrong warned his fellow chemists in a most dogmatic manner about

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

291

the dangers of dogmatism. He reminded his audience how Ostwald had “charged his test tubes with ink instead of chemical agents” and how he tried to convince chemists to become adherents of the cult of his school. He warned his audience of the dangers “of uncontrolled literary propagandism in science”. But that was not the end. Armstrong had one more account to settle. He appealed to the physicists to make themselves more acquainted with the methods of the chemists and to stop speculating unnecessarily. He felt that such an attitude by the physicists was all the more necessary, especially since they appeared to “dictate a policy” to the chemists, without making much effort to understand their methods of “arriving at the root conceptions of structure and the properties as conditioned by structure. It is a serious matter that chemistry should be so neglected by physicists” (Armstrong 1909, 420–455, quote on p. 423). I believe, however, that it was Joseph Larmor’s Aether and Matter which more than any other work, can be considered as being a concrete response to the necessity for a consensual theoretical framework for both the physicists and the chemists. It was published in 1900 and it was substantially the same text for which two years earlier he had been awarded the Adams Prize. Larmor’s work was a study of the relations of “micro-discontinuity to macro-continuity” (Whyte 1960) and his ether attempted to play a unifying role, explaining all natural phenomena. Electromagnetism would no more be considered as being solely based on the idea of the continuum without the necessary changes due to microscopic considerations after the discovery of the electron. It should be stressed that Larmor’s Aether and Matter affords a chemical reading and chemical considerations were not a minor contributory factor in its completion. Larmor was quite knowledgeable about various chemical problems, and especially about chemical affinity. Ten years earlier in the introduction of his A Treatise on the Motion of Vortex Rings J. J. Thomson had thanked Larmor for his “valuable suggestions” and soon afterwards Larmor had produced his own theory to account for chemical phenomena. “I am now deep into atomism” he wrote to Oliver Lodge in 1894 (Larmor to Lodge April 30, 1894. UCL MS ADD.89. Buchwald 1985, p. 153). This work attempted to create a consensus not only on the issue of ontological commitments, but on the issue of methodological preferences as well. Thus, I think, that Larmor was pursuing an agenda where it was clear to him that the ontological issues that had to be settled between the physicists and the chemists were closely related with the methodological issues involved in setting up a consensual theoretical framework. Decisions about the former necessitated decisions about the latter. The new consensus

292

KOSTAS GAVROGLU

was not only a matter of devising a new theory as such, but of articulating the novelty of the theoretical approach as well. One could also discuss another case, that of quantum chemistry. I shall not discuss it analytically, since it has been examined elsewhere in a detailed manner (Sopka, 1988; Schweber, 1990; Simoes 1993; Gavroglu, Simoes, 1994; Gavroglu 1995). From its very beginning chemists were not certain as to whether quantum chemistry belonged to the domain of physics or chemistry. They hotly debated where and how its boundaries are to be drawn, what would be the methodological priorities, the ontological commitments and, above all, what would be the character and extent of the practitioners’ allegiances to physics and chemistry. And most importantly, the question as to the character of theory in quantum chemistry dominated the minds of many chemists. The problem of the chemical bond contributed more than any other problem, so that quantum chemistry could articulate its own autonomous language with respect to both physics and chemistry, chart its own theoretical agenda and formulate its own theoretical framework. The point, however, to emphasise is that the beginnings and the establishment of quantum chemistry involved a series of issues which transcended the question of the application of quantum mechanics to chemical problems. Quantum chemistry developed an autonomous language with respect to physics and what appeared to be disputes over methods were, in fact, discussions concerning the collective decision of the chemical community about methodological priorities and ontological commitments. And the outstanding issue to be settled in the community turned out to be the character of theory for chemistry and, therefore, a reappraisal of the praxis of the chemists. 3. AN IDIOSYNCRATIC PROGRAM IN ‘DE-ENTROPISING’ CHEMISTRY At about the same period as chemists were trying to reappropriate the atom from the physicists, they were also trying to dispossess entropy which was a rather paralysing concept of the physicists. It should be noted that the possibilities offered by thermodynamics to chemistry did not automatically lead to the formulation of chemical thermodynamics and its adoption by the chemists. There ensued a stage of adapting chemical thermodynamics to the exigencies of the chemical laboratory. Chemical thermodynamics had to appeal to the chemists not only because it provided a theory for chemistry, but also because it formed a framework sufficiently flexible to include parameters which could be unambiguously determined in the chemical laboratory. Entropy was a key concept in thermodynamics, but it was, in effect, a physicist’s concept. Chemists, not always successfully, attempted

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

293

to bypass the incapacitating role of entropy. Among these attempts there was J. D. van der Waals’ theory of mixtures and his insistence on using those aspects of Gibbs’ work which would be best suited for defining a program at the Physical Laboratory of the University of Leiden. Another attempt was that of G. N. Lewis in whose idiosyncratic theoretical agenda, foundational issues of chemical thermodynamics played a prominent role. Both van der Waals and Lewis, though deeply committed to thermodynamics, proposed two notions which were much better suited to the chemists’ laboratory culture. They were both different expression for free energy which could be unambiguously determined at the laboratory. Van der Waals proposed the notion of -surface and Lewis introduced fugacity and activity. An aim shared by both was the definition of entities which could be of practical use to experimentalists by avoiding in the definitions a direct reference to entropy. They both made efforts to propose visualizable entities, something which was not independent of the special relations of each with particular laboratory practices. The choices made by van der Waals and Lewis displayed a peculiar complementarity. Van der Waals was a physicist, Lewis was a chemist and both have been among the protagonists in articulating the constitutive elements of physical chemistry. The former’s path to physical chemistry was from molecular physics, the latter’s was from chemistry. Van der Waals’ alter ego was the doyen of the Dutch experimentalists, Heike Kamerlingh Onnes. Lewis was an experimentalist himself. Van der Waals’ theory of mixtures was put to test by Kamerlingh Onnes in an experimental program spanning nearly twenty years of studying the -surface. Lewis’ successive re-formulations of thermodynamics and the central role he assigned to fugacity and activity led him to start, and for many years continue, an experimental program around the systematic measurements of free energies of many different substances. In 1873 Gibbs published two articles (1873a,b), where he proposed some graphical methods for the representation of thermodynamic properties. His purpose was to use the concept of entropy, which was still not fully clarified, together with the Second Law of Thermodynamics in order to systematically deal with all the issues related with the equilibrium of a physical system. As he explicitly put it: “Although this principle [i.e. maximisation of entropy] has by no means escaped the attention of physicists, its importance does not appear to have been duly appreciated. Little has been done to develop the principle as a foundation for the general theory of thermodynamic equilibrium” (Gibbs 1878, 441). The use of thermodynamic diagrams which rested upon the concept of entropy was important, not only because they facilitated the study of substances through thermodynamics, as Gibbs showed in these articles, but also because they

294

KOSTAS GAVROGLU

helped clarify the concept of entropy itself. The challenging problem of the equilibrium of mixtures, the “general theory of equilibrium”, was the ideal playground to test and apply these diagrams and the corresponding theoretical methods. In a major article, published in two parts in 1875–77, Gibbs introduced certain fundamental thermodynamic functions, namely the free energy , the enthalpy  and the chemical potential  (or  for a single substance), which were defined not only for a single substance but also for mixtures of different constituents. These thermodynamic functions yield a set of methods which could in principle be used interchangeably to study all the problems of the equilibrium of heterogeneous substances. Gibbs, however, did not actually make use of all his thermodynamic functions in order to study mixtures with the graphical methods. Of all his thermodynamic functions, he chose to describe (and not actually draw) the graphical representation of the  -surface for a single substance, and to give some hints for the application of the  -method for the case of mixtures of two or three substances; the - and -methods were completely ignored in the rest of his article, as regards to their graphical treatment. 4. THE problematique ON VAN DER WAALS’ The first person to have successfully applied the method of the -surface to the study of binary mixtures was van der Waals. Van der Waals turned to the study of mixtures, because he felt that some of the deviations from the predictions of his equation of state were due to the influence of admixtures. As long as a complete theoretical account for mixing was lacking, van der Waals’ insistence on calling upon the influence of admixtures was an ad hoc move to save the theory. It is worth noting that van der Waals’ first attempt to deal with mixtures, in his 1880 article, was not to construct an equation of state for them as was the case with the pure substances, but he chose, instead, to discuss the coexistence of phases in mixtures. If nothing else, this was an indication of his belief, that the problem of coexistence was the fundamental problem for mixtures. He, then, drew some interesting conclusions on the behaviour of binary mixtures, and these were mainly in the form of semiempirical laws and qualitative arguments. But, why did van der Waals chose to use the -surface for the study of binary mixtures while everyone else – including Gibbs – had declined to follow this path? Why did van der Waals use the -method, and not any other thermodynamic function of Gibbs? Why had he not applied the same method, namely the -line, to the study of single substances, and only applied it to the more complicated case of mixtures? How did it occur to van

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

295

der Waals to use the -method, or in other words, was there some necessity which led him, in a stronger or weaker sense, to this method? Answers to these may help us understand significant methodological dimensions of van der Waals’ program. In his construction of a theory of mixtures, Van der Waals imposed certain methodological constraints. Constructibility and visualizability became the dominant criteria. Van der Waals’ commitment to these criteria led him to his choice of the -method among the possible alternatives. It should also be noted that in the case of mixtures, one has to do with the depiction of properties and not of entities; in other words, what is rendered visualizable with are the thermodynamic properties of mixtures, and not any “picture” of the arrangement of the molecules. Van der Waals’ ontological commitments and more specifically his belief in the reality of molecules was expressed in his work involving the equation of state rather than the -surface (Gavroglu 1990). In contrast to his ontological premises concerning the existence of molecules which may have guided him to the specific equation of state, it was his methodological premises which were decisive in his choice of for mixtures.

5. G. N. LEWIS’ PHYSICAL CHEMISTRY Gilbert Newton Lewis is best remembered because of his ingenious proposal that chemical bonding – both the ionic type as well as the mysterious homopolar type – could be explained in terms of the shared electron pairs. It was a semi-empirical schema first proposed in 1913 and whose justification became possible only after the advent of quantum mechanics, and specifically in the papers of Heitler and London in 1927 and 1928. It would be misleading to assess Lewis’s theory of valence on its own merits alone. Lewis had a “theoretical agenda” and his theory of valence was an integral part of such an agenda. His work in chemical thermodynamics and the special theory of relativity displayed the same trends as his work on valence: an attempt to propose a theoretical framework within which phenomena can be accounted for in a more unifying manner by making use of fewer assumptions and more rigorous (mathematical) derivations. Lewis announced his program in physical chemistry in the opening paragraph of the first paper he published by himself The advance of modern physical chemistry has been largely due to the application to physico-chemical problems of the first and second laws of thermodynamics and the gas law – the latter both directly and by analogy. Upon this basis the whole theoretical treatment of chemical equilibrium rests at present. For this reason it may not be without interest to attempt to express the relations deduced from these laws in a single equation, the most

296

KOSTAS GAVROGLU

convenient and the most general possible, which may serve to systematise a part of our present knowledge, and perhaps point out new laws. (Lewis 1899, 3)

For Lewis thermodynamics could be assimilated in chemistry only if it became possible to work with parameters which could be unambiguously related to situations one meets in the laboratory, rather than seeking the extension of parameters originally defined for ideal systems to problems occurring in the laboratory. Thermodynamics could loose all its appeal for the chemists if it remained a theory which could be formulated in terms of parameters that the chemists use, but which they cannot be unambiguously measured in the laboratory. For example, it was notoriously difficult to determine exactly partial pressures and concentrations which were the parameters in terms of which most of the equations of chemical thermodynamics were formulated. Lewis proposed to base chemical thermodynamics on the notion of escaping tendency or fugacity, which he considered as being closer to the chemists’ culture, more fundamental than partial pressure and concentration and exactly measurable. Lewis hoped that this new concept would become the expression for the tendency of a substance to go from one chemical phase to another. If any phase containing a given molecular species is brought in contact with any other phase not containing that species, a certain quantity will pass from the first phase to the second. Every molecular species may be considered, therefore, to have a tendency to escape from the phase in which it is : : : .The quantity which we shall choose [to express this tendency quantitatively for a particular state] is one which seems at first sight more abstruse than [thermodynamic potential, vapour pressure, solubility in water], but is in fact simpler, more general and easier to manipulate. It will be called fugacity. Obviously, the fugacity of a system which is less stable is greater than that of another system is more stable. (Lewis 1907, 54)

After discussing fugacity whose experimental determination involved the difficult measurements of osmotic pressures, Lewis proposed to reformulate chemical thermodynamics in terms of a new notion, that of the activity of a substance, which was, in effect, its fugacity divided by the product of the gas constant and the absolute temperature. The activity of a species was, according to Lewis, the “perfect measure of the tendency of a species to take part in any chemical reaction” (Lewis 1907, 284). Even though Lewis admitted that absolute activities of ions in a solution could not be determined, he offered a method whereby it could become possible to determine the ratio of the activities of a substance at two different concentrations and this quantity was quite sufficient for most of the tasks confronted by the chemists. Lewis, then, showed how activities were related to changes in the free-energy in reactions as well as to the electromotive force of the galvanic cells. “He showed that both usable data regarding the

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

297

activities of substances could be obtained in the laboratory and that such data could be used in equations describing not simply changes in physical state, but also chemical processes” (Servos 1989, 144). In 1907 Lewis published a paper titled ‘Outlines of a new system of thermodynamics in chemistry’. From a methodological point of view, it is his most significant paper. It is the paper where, among other things, he explicitly articulated the overall approach to chemical thermodynamics. He started by stating that there are, basically, two approaches in thermodynamics. The first made use of entropy and the thermodynamic potential and it had been used by Gibbs, Duhem, and Planck. The second approach, where the cyclic process was applied to a series of problems, had been used by van t’Hoff, Ostwald, Nernst, and Arrhenius. The first method was rigorous and exact and had been, mainly, used by physicists, whereas chemists preferred the second. According to Lewis, the main reason for the chemists’ preference was the difference between the physicists’ notion of equilibrium and that of the physical chemists. Lewis was not particularly satisfied with the latter method since he felt that its application “has been unsystematic and often inexact, and has produced a large number of disconnected equations, largely of an approximate character” (Lewis 1907, 259). The reason for this was that nearly all of the equations had been based on the assumption that it was possible to treat the vapour of a substance as a perfect gas, or a solution as a perfect solution – one that obeyed the laws of an infinitely dilute solution. These assumptions presupposed a kind of continuity from the case of an ideal gas to a real one and, hence, explored the possibilities of a corresponding behaviour. The laws derived from such assumptions were, then, intrinsically approximate laws. Hence, Lewis, wanted to investigate the implications of the deviations found from the predictions of these laws. One expects that Lewis, whose knowledge of mathematics was quite impressive, would attempt to apply the more exact methods of the physicists. Instead he developed the methods already in use among the physical chemists “in such a way as to render them exact. : : : The aim is to develop by familiar methods a systematic set of thermodynamic equations entirely similar in form to those which are now in use, but rigorously exact” (Lewis 1907, 260). Lewis’ work in physical chemistry went hand in hand with his attempts to delineate the autonomous status of physical chemistry by trying to develop its own characteristic discourse, rather than to try to reduce it to physics. Even though many aspects of the proposed theory may be very similar to the respective physical theory, Lewis’ aim was to articulate not so much the theory of physical chemistry, but rather the theory of physical chemistry by

298

KOSTAS GAVROGLU

emphasising the significance of the measured quantities for theory building in chemistry.1 6. G. N. LEWIS’ METHODOLOGICAL COMMITMENTS The strongest themes in Lewis’ theoretical agenda run concurrently in his papers on physical chemistry, relativity and valence and, to a large extent, they define his methodological commitments. These themes include the unification and generalisation of existing formulations, rigorous derivations of semi-empirical rules, a preference of what is convenient rather than a preoccupation with what is actually true, and visualizability of the proposed mechanisms. Among his themes, the one dearest to Lewis and which dominated most of his work in physical chemistry and thermodynamics was his attempt to formulate a general expression which can account for as many phenomena as possible and from which most of the existing empirical rules, semiempirical relations, or, even, rigorously derived formulas can be produced. In his paper on “The law of physicochemical change” published in 1901, he claimed that his researches had been crowned with the success of finding a single law which is simple, exact and general enough “to comprise in itself many laws and yet concrete enough to be immediately applicable to specific cases” (Lewis 1901–02, 49). In the mathematical paper he published with E. B. Wilson on “The space-time manifold of relativity” in 1912, he declared that mechanics had been given a more general treatment, and the conservation laws of momentum, mass and energy were shown to be special deductions from a single general law which stated the constancy of the newly introduced four dimensional vector of extended momentum (Lewis 1912, 392). In his classic paper on valence published in 1916, he argued that the proposal about electron-pairing could now account for both kinds of bonds. The striking differences in properties between the extreme polar and the extreme nonpolar types with respect to which “fundamental distinctions have been made between the two types, and which seem so unconnected, are in fact closely related, and the differences are all due to a single cause” (Lewis 1916, 763–764). In 1930, in one of his most speculative papers, he declared that he “decided to present certain ideas” which would solve one of the outstanding problems of quantum mechanics by showing that almost all the rules which stipulated the exclusion of certain quantum states can be shown to be direct mathematical consequences of quantum mechanics (Lewis 1930, 1144). And when Joseph Mayer wrote to Lewis that he looked forward to reading his new paper concerning the foundations of thermodynamics about which he had heard enthusiastic comments from

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

299

Oppenheimer, Lewis, in answering, did not seem to be feeling a strong urge of humility towards his former collaborator: “My paper on thermodynamics in which I derived classical thermodynamics, as well as the whole theory of fluctuations, from a single extremely simple assumption, is much more fundamental than Gibbs’ “Statistical Mechanics” and its successors” (Mayer 1931a,b). 7. SOME GENERAL COMMENTS The experimental program of Heike Kamerlingh Onnes, who was the first to liquefy helium in 1908, had an appreciable effect in influencing van der Waals to make the particular theoretical choices. The -surface was visualizable and it could be applied in a practical manner. The -surface could actually be constructed according to experimental measurements and be used to facilitate the experimental study of the properties of mixtures, and this was actually the case at the Cryogenic Laboratory at Leiden, under the direction of Kamerlingh Onnes. The mutual influence between Kamerlingh Onnes and van der Waals is evident in their regular correspondence, which concerned mainly technical matters and their regular bi-weekly meetings prior to or following the meetings at the Academy in Amsterdam. We should also not overlook the dedication of van der Waals’ book on mixtures in 1900 to Kamerlingh Onnes, where we read: “the fortunate co-operation between theory and experiment, of which the following pages are but a token, is to be ascribed to you, in the first and decisive place” (van der Waals 1900, dedication). An experimental project was launched around in 1890 at Leiden under the title “Contributions to the knowledge of van der Waals’ -surface” and it lasted for about two decades. In this project, the constructibility and visualizability of the -surface became one of its most celebrated features and it can be argued that the study of mixtures with the aid of the -surface constituted a heuristic factor in the process of the liquefaction of helium. was the theoretical instrument which allowed van der Waals to study the properties of substances as – in essence – properties of their molecules. To clarify this point, we should add that from the form of the -surface, and more specifically from the generation of “plaits” upon it, one could draw conclusions about the physicochemical affinity of the different mixed substances. Through a “phenomenological” study of the -surface, one could draw conclusions with ontological implications since the interpretation of the plaits on the -surface expressed molecular properties. This was actually done in Leiden, where it was used as a means to accomplish a more general goal: the goal of classifying substances, according to physico-

300

KOSTAS GAVROGLU

chemical criteria. In this sense the method of the -surface for studying mixtures was a constitutive element of Leiden’s style of doing physical chemistry. The constructed moulds became heuristic artefacts. Since was a representation of properties, it was not necessary to be explicit about the underlying ontology. The use of the -method legitimised a particular approach to the study of mixtures without requiring a commitment as to the reality of molecules. Its success provided the prestige to pursue an independent approach to physical chemistry without coming into any conflict with the energeticists. It was a method which stemmed from the work of Gibbs, which was also highly esteemed by Ostwald and the energeticists. Perhaps this is one of the reasons why van der Waals in his Nobel speech in 1908 – after Ostwald’s public admission as to the reality of atoms – said for no apparent reason, that he always believed in the reality of molecules. The Leiden group under Kamerlingh Onnes did not commit themselves to a particular ontology in an explicit manner for a long time. They studied the properties of substances and concentrated in measuring deviations from the predicted values and proceeded to a classification of substances through these deviations. This was accomplished through physicochemical methods and criteria. It was quite different, for instance, from Boltzmann’s style, who did use van der Waals’ theories, but his method was to postulate and check a number of different molecular models until he achieved a satisfactory mathematical model of the molecular forces. It was also different from the energeticists’ style, who were also conducting work on the foundations of chemistry. The Dutch followed their own programme in physical chemistry, where there was no need to be explicit about the ontology, but, instead, the emphasis was to propose visualizable heuristic devices. The direction of Lewis’ approach and of his proposed solutions was not towards the discovery of hitherto unknown physical principles in order to provide a more solid physical basis for chemical thermodynamics. Instead he insisted on defining quantities which were capable of being experimentally determined in the chemists’ laboratories. The only relatively reliable answers provided by chemical thermodynamics were for dilute solutions of strong and weak electrolytes. To deal with all the remaining situations which comprised the overwhelming cases chemists were confronted with in their laboratory work, chemical thermodynamics had to be further adjusted to the needs of real solutions and chemical reactions. Determining activities by measuring free energies was particularly convenient for the chemists. It was possible to construct a table with the affinities of a particular substance and a large number of its chemical reactions

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

301

by measuring the free energy of the formation of a substance from its elements. It was in this way that activity became such a crucial quantity in Lewis’ thermodynamics. The use of activities made chemical thermodynamics relevant to non-ideal systems. This new quantity allowed the chemists to be able to transform the free-energy values obtained when one of the constituents of a reaction was in a non-ideal state into corresponding values when all the constituents were in standard states. The standard state was the state of a substance with the activity coefficient being equal to one. To perform any conversions it was only necessary to measure the activities of substances in their non-standard states. “The problem of converting free energies of various states into free energies in standard states is the problem of determining the activities of the various substances concerned” (Lewis and Randall 1923, 291). Arthur Noyes, the Director of the Research Laboratory of Physical Chemistry at MIT, was convinced of the immense usefulness of such a compilation of free energy changes. It was a complex operation and in 1905–1906 he tried to involve several laboratories in the measurement of free energies of formation. Failing to secure such a collaboration, Noyes decided that the measurements would be done at MIT and Lewis took charge of the program. By the time Lewis left for Berkeley in 1912 exact values of free energies were determined for many substances and there were impressive improvements in the experimental techniques. Lewis’ concept of activity, invented as part of a personal quest to reform chemical thermodynamics, ended up being a useful tool both in the study of the anomaly of strong electrolytes and in the calculation of free energy values. Lewis’ original idea, that the concepts of fugacity and activity might play a central role in the formal structure of thermodynamics, never caught on, but by 1920 physical chemists everywhere were beginning to use activities and activity coefficients in their calculations. That activities and activity coefficients became part of the vocabulary of physical chemistry was due in large part to the support given to Lewis and his ideas by Noyes and his associates at MIT (Servos 1989, 149). Van Fraassen has asserted, that descriptive excellence at the observational level is the only genuine measure of any theory’s success and that one’s acceptance of a theory should create no ontological commitment whatsoever beyond the observational level (van Fraassen 1980). This argument has been counteracted by the claim that observational excellence or empirical adequacy is only one of the epistemic virtues among others of equal or comparable importance. And it was also argued that the ontological commitments of any theory are totally blind to the distinction of what is

302

KOSTAS GAVROGLU

and what is not humanly observable and so should be our own ontological commitments. How can this problematique be recast to accommodate chemistry whose theoretical, and to a large extent historical, particularities are such as to admit the simultaneous existence of more than one theoretical schemata with “equivalent” empirical adequacy? Claiming that these considerations refer to mature and closed theories, is a way of avoiding to deal with the problem by talking, in effect, about physics rather than chemistry. But chemists have been debating similar questions since the last quarter of the 19th century. Many times these debates have been portrayed as discussions among the more sophisticated protagonists which did not really touch the rank and file. I do not think so. The settling of these issues affected deeply laboratory practices and research agendas and, as I tried to show, were never snubbed by the chemists as a whole.

ACKNOWLEDGMENTS

I thank Th. Arabatzis, E. Scerri and T. Tsiantoulas for their useful comments. NOTES 1

An important aspect of Lewis’s work was thermodynamics and his repeated attempts to formulate thermodynamics on what he considered to be an axiomatic basis where the emphasis was on defining parameters and procedures which would appear convenient to the chemists. Lewis discussed the foundational issues of thermodynamics and proposed ways to generalise the standard treatments in the book on thermodynamics with Merle Randall published in 1923, and returned to the subject in 1928 with two papers written with J. E. Mayer and in 1931 in a paper he wrote by himself.

REFERENCES

Armstrong, H.: 1909, Proceedings of the British Association for the Advancement of Science (BAAS), 420–455. Buchwald, J.: 1985, From Maxwell to Microphysics, Chicago University Press, Chicago. Coulson, C. A.: 1970, ‘Recent Developments in Valence Theory – Symposium Fifty Years of Valence’, Pure and Applied Chemistry 24, 257–87. Dirac, P.: 1929, ‘Quantum Mechanics of Many-Electron Systems’, Proceedings of the Royal Society A123, 714-33. Divers, E.: 1902, Proceedings of the British Association for the Advancement of Science (BAAS), 557–575. Gavroglu, K.: 1990, ‘The Reaction of the British Physicists and Chemists to van der Waals’ Early Work and to the Law of Corresponding States’, Historical Studies in the Physical Sciences 20, 199–237.

PHILOSOPHICAL ISSUES IN THE HISTORY OF CHEMISTRY

303

Gavroglu, K. and Simoes, A.: 1994, ‘The American, the Germans and the Beginnings of Quantum Chemistry: The Confluence of Diverging Traditions”, Historical Studies in the Physical Sciences 25, 1–63. Gavroglu, K.: 1995, ‘Appropriating the Atom at the End of the 19th Century: Chemists and Physicists at Each Others’ Throat’, in Philosophers in the Laboratory, edited by V. Mosini (ed.), Accademia Nazionale di Scienze Lettere e Arti, Modena, pp. 93–106. Gavroglu, K.: 1995, Fritz London (1900–1954), A Scientific Biography, Cambridge University Press, Cambridge. Gibbs, J. W.: 1873a, ‘Graphical Methods in the Thermodynamics of Fluids’, Transactions of the Connecticut Academy of Arts and Sciences 2, 309–342. Gibbs, J. W.: 1873b, ‘A Method of Geometrical Representation of the Thermodynamic Properties of Substances by Means of Surfaces’, Transactions of the Connecticut Academy of Arts and Sciences 2, 382–404. Gibbs, J. W.: 1875/77, ‘On the Equilibrium of Heterogeneous Substances’, Transactions of the Connecticut Academy of Arts and Sciences 3, 108–248, 343–524. Gibbs, J. W.: 1878, ‘Equilibrium of Heterogeneous Substances’, American Journal of Sciences and Arts, 3rd ser. 16, 441–58. Kamerlingh Onnes, H., Keesom, W. H.: 1912, ‘Die Zustands-gleichung’, Encyclopaedie der Mathematischen Wissenschaften Art. V 10, 615–945 CPL Suppl. 23. Klein, M. J.: 1974, ‘The Historical Origins of the Van der Waals Equation’, Physica 73, 28–47. Larmor, J.: 1900, Aether and Matter, Cambridge University Press, Cambridge. Lewis, G. N.: 1899, ‘The Development and Application of a General Equation for Free Energy and Physico-Chemical Equilibrium’, Proc. Am. Acad. 30, 3–38. Lewis, G. N.: 1901–1902, ‘The Law of Physico-Chemical Change’, Proc. Am. Acad. 37, 49–69. Lewis, G. N.: 1907, ‘Outline of a New System of Thermodynamic Chemistry’, Proc. Am. Acad. 43, 259–93. Lewis, G. N.: 1908, ‘A Revision of the Fundamental Laws of Matter and Energy’, Phil. Mag. 16, 707–17. Lewis, G. N.: 1913, ‘Valence and Tautomerism’, J. Am. Chem. Soc. 35, 1448–55. Lewis, G. N.: 1916, ‘The Atom and the Molecule’, J. Am. Chem. Soc. 38, 762–85. Lewis, G. N.: 1923, Valence and the Structure of Atoms and Molecules. The Chemical Catalogue Company, New York. Lewis, G. N. and E. B. Wilson: 1912, ‘The Space-Time Manifold of Relativity. The nonEuclidean Geometry of Mechanics and Electromagnetism’, Proc. Am. Acad. 48, 389–457. Lewis, G. N. and R. C. Tolman: 1909, ‘The Principle of Relativity and nonNewtonian Mechanics’, Phil. Mag. 18, 510–23. Lewis and Randall: 1923, Thermodynamics, McGraw-Hill, New York. Maxwell, J. C.: 1871, Theory of Heat, Longmans, London. Maxwell, J. C.: 1874, ‘Van der Waals on the Continuity of the Gaseous and Liquid States’, Nature 10, 476–79. Reprinted in Maxwell 1890, Vol. 2, pp. 407–415. Maxwell, J. C.: 1875, ‘On the Dynamical Evidence of Molecular Constitution of Matter’, J. Chem. Soc. 13, 493–508. Reprinted in Maxwell 1890, pp. 418–438. Maxwell, J. C.: 1890, The Scientific Papers of James Clerk Maxwell, W. D. Niven, 2, 1965. Mayer, J.: 1931a, May 29th, Letter to G. N. Lewis, Lewis Papers, College of Chemistry, University of California, Berkeley, CA.

304

KOSTAS GAVROGLU

Mayer, J.: 1931b, June 4th, Letter to G. N. Lewis, Lewis Papers, College of Chemistry, University of California, Berkeley, CA. Meldola, R.: 1895, Proceedings of the British Association for the Advancement of Science (BAAS), 639–655. Roscoe, H.: 1884, Proceedings of the British Association for the Advancement of Science (BAAS), 659–669. Scerri, E.: 1191, ‘Electronic Configurations, Quantum Mechanics and Reduction’, British Journal for the Philosophy of Science 42, 309–325. Scerri, E.: 1993, ‘Correspondence and Reduction in Chemistry’, in Correspondence Invariance and Heuristics. Essays in Honour of Heinz Post, S. French and H. Kamminga (eds.), Boston Studies in Philosophy of Science, No. 148, Dordrecht, Holland, pp. 45–64. Simoes, A.: 1993, ‘Converging Trajectories, Diverging Traditions: Chemical Bond, Valence, Quantum Mechanics and Chemistry 1927-137’, Doctoral dissertation, University of Maryland. Smithells, A.: 1907, Proceedings of the British Association for the Advancement of Science (BAAS), 469–502. van Fraassen, B. C.: 1980, The Scientific Image, Clarendon Press, Oxford. van der Waals, J. D.: 1873, Over de Continuiteit van de Gas- en Vloeistoftoestand, Leiden: Sijthoff. van der Waals, J. D.: 1881 Die Continuitaet des gasfoermigen und fluessigen Zustandes, F. Roth, Leipzig: Barth. van der Waals, J. D.: 1890a, ‘Theorie Moleculaire d’une Substance Composee de Deux Matieres Differentes’, Arch. Neerl. 24, 1–56. van der Waals, J. D.: 1890b, ‘The Continuity of the Liquid and the Gaseous States of Matter’, R. Threlfall, J. F. Adair, Physical Memoirs, selected and translated from Foreign Sources, under the Direction of the Physical Society of London 1(3), London: Taylor and Francis. van der Waals, J. D.: 1899, Die Continuitaet des gasfoermigen und fluessigen Zustandes, zweite verbesserte Aufgabe, 1 Theil, Barth. van der Waals, J. D.: 1900, Die Continuitaet des gasfoermigen und fluessigen Zustandes, 2 Theil, Binaere Gemische, J. J van Laar, Leipzig: Barth. van der Waals, J. D.: 1910, ‘The Equation of State for Gases and Liquids’, Nobel Lectures (Physics) 1901–1921, 254–265. Whyte, L. L.: 1960, ‘A Forerunner of Twentieth Century Physics’ (A Review of Larmar 1900) Nature 186, 1010-14. Department of History and Philosophy of Science University of Athens Athens Greece