2—
Water Analysis and the Hegemony of Chemistry, 1800–40

It does not seem of much practical consequence of what our mineral waters are composed, for we have gone on with them hitherto, content with analyses of the most wretched kind, if the latest analyses of some of them be correct.^[1]
Lancet

During the first part of the nineteenth century chemistry became in Britain the sort of profession ordinary people made their livings at and analytical disputes of the sort considered earlier increasingly took place among self-defined chemists, struggling to forge a new profession. The typical eighteenth century chemist had been a well-to-do eccentric (Cavendish), a dabbling clergyman (Hales), or an unschooled manufacturer. By contrast, by 1850 a large group of trained 'practical' chemists made their livings as analysts and were doing useful work in industry, commerce, government, law, and education. This transformation occurred first in Scotland. There, beginning about 1750, university chemists began to involve themselves in local industries: agriculture, the extraction of alkali from kelp, bleaching, and later sulphuric acid and dyeing. In the north of England in the early nineteenth century there developed a similar set of opportunities for chemists in the mining, metal working, and textile industries, and soon in the new artificial alkali industry.^[2]

Around London chemists became involved in brewing, paper-making, and the new gas industry. In the metropolis there was also an active trade in scientific lecturing: to medical students, burdened with chemistry requirements by the Apothecaries Act of 1815; to artisans in the mechanics institutes; or to the genteel patrons of the Royal Institution or its imitators.^[3] For water analysis London was

the dominant locus. As potable water analyses became increasingly important in public health policy-making during the second half of the century, London chemists with close ties to the Local Government Board or with experience as expert witnesses secured greater credibility for their analyses and methods than did their northern rivals.

It is noteworthy that practical chemistry expanded during the years of the chemical revolution, at the same time that scientific chemistry was acquiring its familiar form of elements and compounds. Yet it would be wrong to think that the expansion of practical chemistry was only made possible by the progress of pure chemistry. Instead, as Bud and Roberts have shown, relations between practical and scientific chemistry were complicated, with the modern distinction between pure and applied science only emerging during the period.^[4] The practical and the academic were not different breeds of chemist, nor was contributing to science unrelated to manufacturing, analysis, or any other form of commercial chemistry.

And the transformation of chemistry was not even wholly a matter of new knowledge, be it pure or applied. What chemists offered for sale was credibility, authority, and rationality, as much as it was new knowledge. Their success was made possible by the spreading belief that chemists and chemistry were the providers of profitable techniques and true answers. Morris Berman has argued that this credibility was not directly a function of competence, for in many cases early nineteenth century chemists claimed authority over technological territory which they were not competent to hold.^[5] Instead, the hegemony of chemistry was a result both of the needs of industrial society for an authority which could settle social problems by technical means and of the ability of chemists to establish themselves as providers of that authority.^[6] Water analysis illustrates Berman's point. By mid century chemists had wrested from medical men the final say both on the medicinal and the pathogenic potential of waters. Yet their achievement of this authority was not mainly a result of any specific pieces of knowledge that chemical analysis could demonstrably deliver. To be sure, there had been significant achievements in water chemistry; some medically important substances, notably iodine, had been discovered in a number of springs. But enormous and recognized areas of ambiguity and ignorance remained; that chemists held 'water' was a reflection of the fact that a combination of social needs and aggressive marketing had made chemistry one of the most important sources of social authority.

The Marketing of Chemistry:
Brande and Taylor

To get a better idea of the context of water analysis in early nineteenth century Britain it will be helpful to review the careers of two of the main promoters of London chemistry, William Thomas Brande and Alfred Swaine Taylor. Neither figures very largely in most histories of chemistry, nor was either primarily a water analyst. Yet between 1820 and 1850 these two, Brande overseeing a Royal Institution–Royal Society nexus, Taylor ensconced as professor of chemistry and medical jurisprudence at Guy's Hospital, were among the most important chemists in London. They were in demand as consultants and expert witnesses; were active as authors, editors, lecturers, and publishers, and they influenced the careers of numerous younger chemists.

Brande (1788–1873) has received more attention than Taylor, due to his involvement with the Royal Institution during the Davy–Faraday era. Brande came from a family of German apothecaries who had accompanied George I to England and had continued to serve the Hanoverians for the rest of the eighteenth century. The Brandes were thus well-off and oriented toward chemistry. During the first decade of the nineteenth century, Brande, in his teens, studied medicine and chemistry at Hunter's Windmill Street School and at St George's Hospital. By 1808, at age 20, he was teaching a variety of subjects at both schools and at the Cork Street Medical School. He was elected F.R.S. in 1809 at 21, made a professor at Apothecaries Hall in 1812, and at the Royal Institution in 1813. Subsequently he became master of Apothecaries Hall, coinage officer to the Royal Mint, and an examiner at the University of London. His Manual of Chemistry was one of the most respected texts of the day and he advocated his views as editor of the Royal Institution's Quarterly Journal of Science, Literature, and the Arts .

Brande was not without genuine scientific talent, but he made little attempt to apply himself to original research. Instead his business was selling chemistry. To Brande almost any technical problem was being trifled with if it was not informed by chemistry. As Berman has noted, the Royal Institution became for him the proper forum for any 'controversy involving cost–benefit analysis, technical consultancy, litigation and patents, government testimony, and problems of pollution.'^[7] That contemporary chemistry was often incapable of providing useful guidance did not matter. Brande, like others we shall see, discoursed with an air of authority, especially to

Parliamentary select committees on issues he knew little about. On occasion his ignorance was exposed by a cross-examining barrister but in the long haul Brande's programme of making chemistry an indispensable authority succeeded amazingly well.^[8]

While Brande ruled over general and technical chemistry, Alfred Swaine Taylor (1806–80) was the foremost toxicologist and forensic chemist of the day. Son of an East India captain, Taylor studied medicine at St Thomas's and Guy's Hospitals in the mid 1820s, working with the famous surgeon Astley Cooper. In 1828 he went to Paris where he studied chemistry with Gay-Lussac and toxicology with Orfila. He became the professor of medical jurisprudence at Guy's in 1831; in 1832 he began sharing the Guy's chemistry lectureship with Arthur Aikin. He was elected F.R.S. in 1845 and F.R.C.P. in 1853, wrote the authoritative text on medical jurisprudence, edited the London Medical Gazette , and co-authored with Brande an elementary text, On Chemistry (1863).^[9]

Like Brande's, Taylor's scientific contributions were modest, empirical investigations, dealing with the detection of poisons and other phenomena of forensic interest. For neither man are the usual indicators of scientific achievement satisfactory; both came to hold numerous positions of trust, appeared regularly in courts of law and before select committees of Parliament, and influenced the careers of others without being prominent knowledge-producers themselves. Most importantly, their careers typified those of a great many practical chemists—with decent laboratory skills, passing familiarity with the contents of the journals, tolerable lecturing talents, good connections, and untouchable confidence one could make a decent living in London as a practical chemist. A few, like Davy and Faraday, tried to keep a distance between their serious science and practical work that subsidized it, and looked with scorn upon the Brandes of the world; yet there were many more Brandes and Taylors.^[10]

The Contexts of Mineral Water Analysis

Mineral water analysis embodied these more general tensions in British chemistry. They were the source of the double standard discussed in the last chapter. Clients who commissioned water analyses wanted results listed as salts. Chemists complied, despite their inability to confirm that the salts they listed were actually in the water. But to understand fully how water analysis fit into the growing

profession we need to consider more exactly what were the purposes of analyses and the big issues of water chemistry.

Why analyse water at all? By the mid eighteenth century a complete water analysis had become a lengthy and complicated matter. In one sense mineral water analysis seems a science without purpose. Supposedly mineral springs were to be known by their effects, not their constituents. Many had proven their worth long before there were chemical means either of distinguishing waters from one another or of determining their composition. Analysis was rarely the means by which new mineral waters were discovered nor was it instrumental in protecting patients from ingesting poisonous salts—two of the main uses one might expect.^[11]

Yet analyses were done, published in books, papers, and tracts, and made much of. It would be wrong to impose rigid categories on these works, but one can recognize three distinct contexts in which mineral water analyses were done. First, many were advertisements. A chemical analysis was scientific legitimation for the medical claims being made. Quantitative analysis could show, more convincingly than any testimonial, how closely the contents of an unknown spring resembled those of well-known resorts. (Of course, defenders of the well-known spring might counter with analyses done with improved techniques and revealing new uniquenesses.)

A second reason to analyse mineral waters was to establish a basis for imitating them. There had been attempts at artificial mineral water manufacture in the mid seventeenth century; by the nineteenth imitating famous waters had become an important industry. An understanding of carbon dioxide chemistry brought with it the ability to manufacture artificial carbonated waters, still known in some places as 'minerals.' Here too there were bitter controversies. Upholders of the old spas denied natural waters could be imitated. Entrepreneurial chemists argued that they could not only imitate natural waters but improve upon them by adding active ingredients and leaving out deleterious salts.

The third class is a catch-all category. It includes works concerned mainly with increasing natural knowledge and analyses undertaken as lab projects by chemists-in-training such as those done by Hofmann's students. The main scientific rationale for mineral water analyses was geological; the soluble substances that impregnated ground water were clues to the composition of the earth's interior and the sources of subterranean heat. Among British analysts, Richard Kirwan exemplified this perspective. He saw mineral

water analysis as the key to 'sound notions of universal geology, a science intimately connected with the principles of morality and religion. . . . Arising from unknown depths, . . . [mineral waters] alone announce to us . . . the awful operations therein transacted.'^[12]

It is tempting to see the first two contexts as applications of the third, which would represent a core of pure science. Yet a split between pure and applied science was only emerging during the period, and to impose such a pattern on early nineteenth century water analysts would be to impose an unfamiliar and unacceptable set of distinctions. To appreciate how these contexts fit together and how the most brazen advertising might be accepted as a scientific contribution, we need to look more closely at these contexts themselves and consider how mineral water analysis reflected aspects of the Baconian ethos of early nineteenth century British science.

Analysis As Travelogue: Frederick Accum, 1808, 1819

Two striking examples of the way mineral water analysis might be used to advertise a particular spa are the reports on the mineral waters at Cheltenham (1808) and Thetford (1819) by the Anglo–German chemist, entrepreneur, and food-adulteration crusader Frederick Accum. Accum (1769–1839), an émigré German apothecary, exemplified the sort of chemistry Brande was practising and advocating. He spent most of his career in a wide-ranging private practice that included analysing whatever was brought in to be analysed, as well as a good deal of lecturing and laboratory instruction. He also wrote a great deal: texts on chemistry and mineralogy, treatises on gas lighting and manufacture, food adulteration, dietetics, brewing, bread making, wine making, and building materials. He also consulted and served as an expert witness, built and sold experimental apparatus, and promoted speculative technical ventures such as supplying London with gas.^[13] To Humphrey Davy, defender of the honour of science, Accum was 'a cheat and a quack,'^[14] yet his accomplishments in food analysis and gas manufacture were real enough.

Though the reports on Cheltenham and Thetford were published in the Philosophical Magazine , one of the main British forums for research in physics and chemistry, portions of them read like travel brochures.^[15] Accum used stirring accounts of scenery, salubrity, and sociability to introduce the details of analysis. Cheltenham was one of those 'choice and suitable spots . . . particularly favourable to the

curative effects of mineral waters.' It had 'uncommon fertility . . . romantic scenery . . . [which would] present a picture dear to the man of taste as well as to the invalid.' The rain there 'seldom prevent[ed] . . . walking or riding for any length of time,' the houses 'in point of taste and elegance may vie with any modern buildings whatever.' The food was good and the natives known for their longevity.^[16] Eleven years later when Accum reported on Thetford (not a leading spa, but trying to become one), he reused the Cheltenham hyperbole. Here too were the 'houses, which in point of taste and elegance may vie with any modern buildings,' the 'uncommon fertility,' 'romantic scenery,' the 'picture dear,' the rain that never precluded outdoor exercise and so on.^[17]

Accum was candid about the promotional potential of his analyses. The Thetford spring had failed as a resort in the mid eighteenth century and had been closed, 'till [now] a happier spirit of research seems once more likely to liberate it . . . and to diffuse those benefits it is so well calculated to ensure.'^[18] As his biographer Browne makes clear, Accum was expected in such cases to act as a publicist.^[19] He insisted that in praising Cheltenham, he intended 'no invidious comparisons . . . with other springs.' He was just telling truths that needed to be told: 'unbiassed as I stand, a humble labourer in the field of chemical science, it is merely my wish to furnish a clear idea of the nature and composition of those fountains of health, so as to present truth in a simple form, and to establish it upon legitimate foundations; in order to enable the medical practitioner to select in a judicious manner the springs so bountifully given to the spot by the hand of Nature, and to apply them with advantage in the routine of his profession.'^[20]

Implicit in Accum's apology were two claims for the utility of analysis. First, analysis was to demonstrate that Cheltenham's water really did have medicinal properties; it could not be legitimately dismissed. Second, Accum implied that guided by chemistry, Cheltenham's physicians might now rationalize their physic. This latter idea had been central for Bergman: once chemists had correlated composition with medical effects, physicians would be able to prescribe with great specificity certain doses of certain waters to certain patients. As Thomas Garnett claimed, chemistry 'emboldens the practitioner to make trials of the efficacy of mineral waters, in cases in which a person ignorant of chemistry would never think of, and which it would be rash to attempt without previous knowledge of the properties and composition.' Mere experience would never

make a physician, he added.^[21] In fact, while chemists often gave lip service to the idea, this second rationale was mostly pretence, a way of augmenting the reputation both of the analyst and the spa analysed. There are two reasons for thinking that this was the case.

First, physicians had made good livings prescribing the waters long before Accum. They made their reputations matching regimen to constitution. So, while qualitative analysis might have a small utility in matching a particular class of patients—say those with skin diseases—with the right sort of waters (sulphurous), a quantitative analysis represented a degree of precision that would be unlikely to have much of a bearing on the doses the doctor prescribed. As John Barker put it with regard to the Cheltenham water, 'it may indeed be . . . proved, that iron enters into its composition, which may be known as well by the taste. But what purpose will it answer to calculate that there are about four grains in every quart?'^[22] And given the notorious discrepancies in methods and results of analysis, it would have been an exceptionally naive doctor who would have committed his practice to the pronouncements of a chemist. Linden complained that misbegotten chemistry had led to claims of opposite diseases (constipation and diarrhoea) being cured by the same water.^[23]

Second, analysts themselves were often perversely vague when it came to explaining what medicinal uses their analyses indicated. Many (Accum included) went directly from listing their analytical operations or giving tables of constituents to statements of medical effects without explaining how the tests led to the statements (or for that matter to the composition). To an extent, of course, such knowledge would be expected of physicians and wealthy invalids, yet the contrast between details of reagent tests and reticence about routes of inference suggests that it was the appearance of thoroughness that was to impress the reader.^[24] Analyses indicated the relative goods the rival springs promised to deliver and symbolized too that someone knew what was going on, that the medicinal environment one was to encounter was comprehended and would be applied in a precise and rational way.

While Accum is unusual in so openly making his sales pitch in an ostensibly scientific article, he was by no means unique. A Dr Evans, writing in the Philosophical Magazine in 1805 on 'Sutton Spa, near Shrewsbury,' alluded to the splendid scenery and low prices of this yet-undeveloped spa.^[25] Others mentioned that their analyses had been undertaken at the request of the proprietor of the waters, and

that their purpose in writing was to ensure that the particular spring received its due share of attention.^[26] In some cases, portions of the analyst's report were excerpted for incorporation into promotional materials, as was the case with A W Hofmann's 1854 report on Harrogate.^[27]

Glorifications of bucolic scenery were also standard fare, even among more moderate writers like Saunders, who spoke of the Bristol Hotwell as 'one of those choice and favoured spots that are peculiarly calculated for the pleasure and comfort of the invalid . . . the whole adjacent country abounds with beautiful scenery and romantic prospects.' Because environment was held to contribute to cure, and because not all beneficially situated springs had been sufficiently utilized, such brazen advertising could almost be viewed as disinterestedly in the service of the public good. Saunders wrote: 'it is merely advantage of situation or accidental causes that have given some of these [springs] a superior reputation over the rest; and where this is owing to beauty of site or local conveniences, it is well merited, as these circumstances have no small share in the general plan of cure, by enabling the invalid to employ daily exercise, and giving that irresistible charm to the spirits, which the sight of a beautiful or romantic country almost always excites.'^[28]

To the modern mind it may seem strange to find such undisguised commercialism in scientific journals. While journal standards varied, Accum's excesses do raise questions about the consulting scientist's relationship with his client. Early nineteenth century sensibilities on such issues differ from those prevalent today, and even then there was no consensus on these matters. Some, like Davy, were scornful of such performances, but such men were condescending toward commercial chemistry in general. In part, the ethos which legitimated articles like Accum's was a Baconian ethos, but before considering the ways in which mineral water analysis was Baconian, we need to consider the second context of mineral water chemistry, the use of analysis to guide the synthesis of artificial waters.

Torbern Bergman and the Synthesis of Artificial Mineral Waters

One component of Torbern Bergman's systematic approach to mineral water analysis was the suggestion that each analysis be checked by dissolving the constituents found in the analysis in distilled water. If the synthetic water possessed the medicinal qualities of the

original, the analysis could be assumed accurate; synthesis was the check. As mentioned in the last chapter, such syntheses were difficult and few, if any, chemists customarily used synthesis to verify their analyses. But Bergman had another interest in synthesis. If a mineral water could be successfully synthesized, it could be produced on a large scale and its benefits made much more widely available.

This question of whether one could expect analysis to guide synthesis was thus not only one of the most philosophically problematic but also one of the most socially significant questions water chemists faced. An industry that could manufacture artificial mineral waters would threaten the exotic and expensive springs. Already by the later eighteenth century some of the best known continental springs—Seydschutz, Seltzer, Spa, Pyrmont—were exporting bottled water in sealed bottles to ensure authenticity.^[29] Never of strong constitution, Bergman was himself a consumer of these bottled mineral waters and it was in part his dissatisfaction with their expense, frequent unavailability, and variable quality that led to his interest in synthesis.^[30] By the mid 1770s he was using his own imitations of several continental waters to treat his own illnesses and those of a few patients.

Bergman was realistic about the opposition a large-scale program of synthesis would arouse.

From the very nature of the thing it must be obvious, that an invention of this kind, however useful, cannot possibly be universally pleasing.—Many who are incapable of ascertaining or judging of the truth, will distrust it, not without reason, on account of its novelty;—many contend, that to imitate nature is impossible, without considering, that when the component parts are thoroughly known, the success of the process cannot in any degree depend upon the hand which combines them. Some who prescribe, and others who sell the foreign waters, condemn the artificial, for obvious reason; and not a few are urged by motives too trivial to be detailed.^[31]

To Bergman this resistance was resistance to progress. The advantages of a free market in mineral water manufacture seemed straightforward. In Sweden such an industry would make the waters available year around (they were unavailable in winter and spring); it would ensure better quality, a lower price, and stop the flow of money 'out of the kingdom.'^[32] But such resistance was also resistance to science. Those who insisted on the inimitability of natural springs on such grounds as that their properties resulted from 'a certain degree of fermentation, as they are pleased to call it,' were

[Full Size]

Figure 2.1
However rigorous the attempts to certify their quality, as with official seals
pictured here, bottled mineral waters were of variable quality and expensive to
boot. Progressive chemists like Torbern Bergman envisioned a synthetic
mineral water industry to remedy these shortcomings
(D W Linden,  A Treatise on Chalybeat Waters , pp xix–xx).

clinging to an obsolete alchemical perspective. Such views were held. John Barker of Cheltenham, for example, held that 'there are specific properties in almost every mineral water, wherein it differs from every other of the same class. Nay, there are qualities in the water, and even in the spirit of every common spring, whereby it is peculiarly different, in many respects, from all others.' The belief that chemistry could be used to 'elucidate things of so high a nature, and enable us to imitate them' was in Barker's view 'a gross mistake, the crude conception and immature production of a deluded mind.' In rejecting this view Bergman was advocating a chemistry fully representative of the spirit of the englightenment: the materials of the earth were compounds, lawfully combined, of simple substances and

whether they were put together in 'the bowels of the earth . . . [or] artificially added . . . can make no difference in the result.'^[33]

In making the link between the progress of society and the advance of science Bergman was placing great trust in analytical chemistry. Not only was he assuming that he could isolate and identify all the constituents of the waters he analysed, but that he knew which were pharmacologically active. Indeed, so confident was Bergman that he saw no need to duplicate natural waters exactly. It was quite proper to leave out inactive constituents and salts that might be harmful.^[34]

British proponents of artificial mineral waters enlarged on Bergman's arguments. Synthetic waters would be more accessible, affordable, and effective.^[35] Rather than sticking to mere imitations, chemists could improve on natural waters by leaving out harmful ingredients and increasing concentrations of active ingredients. One might even go so far as 'to form new and valuable compounds [in artificial mineral waters] which are no where to be met with in a natural state.'^[36] Ironically, the very chemistry that cast doubt on the accuracy of analysis—Murray's—would resolve one of the major problems of the artificial waters industry. Murray's argument that a rearrangement of salts occurred as a water evaporated, and therefore that the salts found in a residue were not necessarily the salts in the water, implied that one could supply constituents in highly soluble forms, rather than struggling to dissolve the insoluble compounds found in residues.^[37]

Yet the conception of social progress so central to Bergman's justification of artificials turned out to be double-edged, for there was as much opportunity for quackery in the artificial water industry as in the natural, and rather than ending the oligopolistic control of mineral waters, the rise of the synthetic industry simply provided an opportunity for a new group of chemist–entrepreneurs. There was thus ample room for quarrels, for attacks of synthesizers on anti-synthesizers and vice versa.^[38] The most important British centre of artificial mineral water manufacture was F A Struve's Royal German Spa at Brighton, where continental mineral waters were imitated.^[39] Successful and well respected in Britain, Struve was accused by a continental chemist of ignoring the latest analyses and hence trading under false colours in representing his concoctions as imitations. Again there were financial interests involved; Struve was competing with continental spas. The Oxford chemistry professor Charles Daubeny defended Struve, noting that emulation was less important

than effectiveness, and that the criticism was 'scarcely candid.'^[40]

The Logic of Mineral Water Analysis

Ultimately, as the more sober commentators on the imitation issue pointed out, the question was one of the adequacy and completeness of analysis. To be confident that one had imitated a mineral water one had to be confident that analytical chemistry in its current state was good enough to detect the active constituents in the water. Yet the abilities of analysts were continually changing and improving: Daubeny (1795–1867), who served Oxford as professor of chemistry, rural economy, and botany (not all at the same time), wrote in 1836 of 'chemists, [who] in the pride of half knowledge . . . [had] smiled at the faith reposed' in a spring which their analysis had showed to be devoid of medicinal properties, yet which had later turned out to have active components not yet discovered at the time of the analysis.^[41] Daubeny was writing in the wake of the discovery of iodine and bromine in many springs, and while iodine was not yet clearly recognized as the cure for goitre, it did seem to him that these elements—and other only recently detected components of mineral waters such as manganese, zinc, strontium, potassium, lithium, and phosphoric and fluoric acids—might account for 'the unexplained virtues attributed to certain mineral waters.'^[42] Given this history of discoveries it was unwise for chemists to think that they finally knew all the components of a mineral water.

In principle Daubeny had a resolution for this kind of problem: one ought to insist on a correlation between medical effects and chemical composition. 'To refuse credence to the reports given by medical men with respect to the salutary or injurious effects of a particular water, merely because the chemist can discover in it no active principle, would seem a proceeding not less unphilosophical, than . . . treating as fabulous the accounts given of stones that had fallen from the sky, because . . . [we] did not understand how such ponderous masses could have continued suspended in it,' he wrote. Likewise, 'granting that a spring possesses peculiar virtues, we must suppose that it differs, either in its mechanical, or chemical properties, from the rest.'^[43]

In fact this sort of correlation was both correlation and explanation and that made it problematic. As Daubeny posed it, the problem was one of inductive reasoning. One started with an empirical finding, medical effects. The validity of a theory explaining

these effects, i.e. what chemical analysis supplied, was determined by the consistency of that theory with empirical findings. If an inconsistency between theory and observation arose one was to discard theory: medical truths were to drive out chemical truths. Once generalizations had been obtained, they would permit recognition of anomalies—waters whose effects could not be explained by their composition—and lead to progress. If a water had different effects than others of like composition, one simply looked for new ways in which it differed chemically and physically. In this way new medicinal substances might be discovered.

To many chemists this was an unacceptable way of posing the problem, for it represented not the correlation of medicine and chemistry but the subjugation of chemistry to medicine. Indeed, this was precisely the argument being made. 'Chemistry,' wrote Barker of Cheltenham, was a 'good servant to physic, though a very bad master.' It provided 'imperfect knowledge . . . apt to mislead weak minds.'^[44] Where Daubeny had treated medical effects as empirical findings capable of being unambiguously demonstrated, most chemists saw chemical composition as the only thing that could be empirically determined. Claims of medical effects were unproved assertions—unsound theories explaining why invalids got well from diseases they probably either didn't have or would have recovered from without the waters. To them mineral water chemistry was predicated on the idea that medicinal properties had to be deduced from chemical composition. There were good grounds for this view. Evidence of the medicinal virtues of springs was almost entirely in the form of testimonials; there were no controlled clinical experiments demonstrating the efficacy of mineral waters in certain conditions. Already it was beginning to be admitted that a great deal of the healthfulness of spas was due to relaxation, regimen, and climate, hence it was quite defensible to refuse to accept any claims for medicinal virtues which could not be confirmed by analysis.^[45] Chemists could cite older writers (and moderns like Barker) who had regarded waters of a certain spring as irreducibly unique, 'exquisitely formed by the hand of nature, produc[ing] effects very different from those of any other mineral waters about this place.' In place of such obscurantism they could point out that 'analysis shews however, that this water must possess less active and stimulant powers than any of the others.'^[46]

There was no good resolution for this problem. Meredith Gairdner, according to Daubeny the best (certainly one of the more sober)

of modern writers on the subject, recognized both horns of the dilemma. He regarded Murray's chemistry, for example, as having shown 'the great, and in many cases dangerous, errors into which the physician might fall, who, a priori , judged entirely of the medicinal effects of a mineral spring from the results of chemical analysis' and admitted that medical effects were often 'the reverse of what we should expect from . . . composition.' He called for 'impartial experience,' but admitted 'although I assume experience to be our principal guide in judging of the real effects of any spring, . . . let it not be supposed that I undervalue chemical analysis, or am of the number of those who regard them [mineral waters] as specifics prepared by the Hand of Nature for the cure of the more obstinate maladies with which human nature is afflicted. This would be to render the whole a system of mystical empiricism, and to place an insurmountable barrier to the acquirement of any true theory of their action, totally incompatible with the present state of medical science.' A new era of mineral water chemistry, founded in Murray's ideas of the pharmacological activity of acids and bases, was the only way out of this dilemma.^[47]

The Foundations of Authority:
Baconians and the Ideology of Progress

Neither the utility of chemistry for advertising mineral waters nor the possibility of making new and stronger mineral water medicines explains the authority mineral water chemists achieved as the determiners of the medicinal potency of waters. These contexts of water chemistry only reflect that authority or, at most, contributed to it in a small way. Nor was that authority a function of the degree of certainty mineral water chemistry had attained, for while it is undeniable that the capabilities of analytical chemistry were improving during the period, by the late 1840s chemists were still likely to come up with quite different compositions for water from the same spring. In part that authority derived from the authority chemistry in general was acquiring, but it also reflected chemists' success in placing mineral water analysis within particular traditions of scientific progress. By making it clear how far chemists had come, how worthy were the programs of investigation they were pursuing, and how useful had been their results so far, these traditions made it possible to accept an authority that was not truly authoritative.

To the modern mind the contrast between the seriousness analytical chemists attached to their own pronouncements and the catalogue of contradictory results they produced is amazing and appalling. One wonders how a community of scientists could go on, year after year, contradicting one another without seeing something as dreadfully wrong and determining either to insist on a standard method of analysis, or to exclude incompetent analysts from the profession, or to decide, like Murray, that unwarranted inferences were completely undermining their practice. So rapidly were the contents claimed for various springs changing that Brande's Quarterly Journal of Science observed, 'It does not seem of much practical consequence of what our mineral waters are composed, for we have gone on with them hitherto, content with analyses of the most wretched kind, if the latest analyses of some of them be correct.'^[48] In one sense the writer was correct: whether chemists were correct mattered little; what mattered was that patients trusted chemistry and went to the spas and got themselves cured (or got themselves to believe they were cured). But most chemists were not so cynical and the observation sheds no light on how they saw their endeavour.

To understand how analyses could be taken seriously we need to go beyond the social circumstances in which analysis was done and understand the motifs or ideologies of analysis, the throwaway rhetoric of context and significance with which analysts began the reports they published in scientific journals. In early nineteenth century Britain analysts' reports reflect two prominent motifs which may be labelled 'Baconianism' and 'Enlightenment.'

Articles on mineral water analysis reflected the resurgence of Baconianism in the early nineteenth century.^[49] Mineral water analysis was Baconian in a number of respects. First it relied on an army of fact-gatherers, whose contributions, when sorted and organized, would lead to accurate knowledge. These contributors were not co-ordinated as well as they would have been at Salomon's House (or under the auspices of the British Association) and there were great problems with comparability and completeness. Kirwan complained of 'a labyrinth of particular facts, betwixt which we can trace no connection, nor consequently apply to no useful purpose' and believed that 'to select, . . . compare, repeat, and correct where need should be found, and occasionally add to these, . . . [could] be undertaken and properly executed only by a society of skilful and well-informed persons, instituted for that particular purpose.'^[50] Gairdner admitted in 1832 that 'it cannot be denied that the subject is involved in much

obscurity . . . but this should be a stimulus to renewed exertion; observing without intermission, and applying the Baconian philosophy at every step, we may arrive at results that . . . now perhaps would be rejected as visionary.'^[51]

Second, mineral water chemists saw their enterprise as Baconian inasmuch as they equated scientific progress with more facts, to be gained from continued study of known springs or discovery of new springs. No conceptual rearrangement was foreseen. When Daubeny wrote about the progress of analysis he wrote of the new substances discovered in springs, not of the theoretical changes that had given chemists the new entities to look for or the new means to find them.^[52]

Third, water analysis was Baconian in that water analysts saw themselves as building a foundation for technical and medical progress. When bromine and iodine were discovered in a few mineral waters, Daubeny undertook a nationwide survey of their presence. Even though medical effects had not yet been demonstrated for these substances, he regarded the project as worthwhile on the grounds that their existence might explain hitherto unexplained medical properties of some springs. Hence Daubeny's research was Baconian not in the sense that it promised direct benefits, but in the sense that he deemed it worthwhile to collect facts that might have social utility in the future. Even those whose analyses were essentially advertisements presented their findings as new knowledge which added to the common good.^[53]

Fourth, like the histories of the trades undertaken during the early years of the Royal Society, the mineral water analyses were intended to contribute knowledge that both described—the analyst was contributing to a catalogue of what nature offered the invalid—and provided a basis for generalization—once the relations between composition and medical effects had been discovered, medical treatment could be carried on in a rational way and discoverers of new springs could know their medical properties without having to conduct lengthy and inconclusive clinical experiments.^[54]

Finally, and again in common with the early years of the Royal Society, investigators of mineral waters found themselves forced to tolerate a great range of competence (and motive) among their colleagues. A part of the Baconian ethos was the idea that all information was grist for the mill, though of course much of it might need to be sifted out. Writers like Granville and Daubeny recognized that a great deal of the mineral water literature came from persons not competent in chemistry and was 'manifestly dictated by selfish

motives,' yet they did not dismiss it on those grounds.^[55] Instead they assumed that as chemistry progressed what truths there were in these works would become apparent while their falsehoods would disappear.

'Enlightenment,' the second motif, was more specific to mineral water chemistry, though it is certainly connected with broader intellectual currents of the age. In introducing their analyses mineral water chemists sometimes gave a short history of the discovery and progress of the medicinal use of mineral waters. In its most developed form, this history had four stages. First came discovery: at some point in the distant past ordinary folk, acting perhaps through instinct, had discovered the beneficial effects of the water and become accustomed to using it to cure their ills. A stage of primitive explanation followed. To explain why some springs had such special powers they had relied on pantheistic (or papist) superstitions: the powers of springs came from saints or spirits. The third stage was skepticism: with the coming of the Reformation (or some other religious authoritarianism) such pantheism had been condemned, and the springs, deprived of an animistic rationale, yet lacking scientific warrant owing to the primitive state of analytical chemistry, had fallen into unwarranted neglect. These were 'the days of intellectual bondage.' Yet by instilling a spirit of independent inquiry the Reformation had also given birth to the fourth stage of the enlightened scientific present when analytical chemistry, finally matured, had redeemed these forgotten springs.^[56]

It will be apparent how nicely this history fit into the larger history of protestant and industrial Britain in the early nineteenth century. It was the triumph of enlightenment over superstition, a scientific confirmation of the common sense of good, plain folk. This version of history was particularly attractive to those trying to develop unknown springs into important resorts. It suggested that there were far more medicinal springs than currently recognized. The ideology also elevated chemistry above medicine: through the centuries of neglect the doctors had either failed to recognize these springs or been unable to persuade others of their powers. While medicine could not provide proof, chemistry, by contrast, was progressive: one could, or would soon be able to, offer a final answer to the question of what, if any, medicinal properties a spring possessed.^[57] To opponents of chemistry the same prospect was a threat. Barker confessed himself 'so old fashioned [as] to think, that their uses [of mineral waters] . . . were much better known . . . in the last, and even some preceding

generations, than the present.'^[58]

Together, these ideologies made it possible to take seriously Accum's hyperbole about standing 'unbiassed . . . a humble labourer in the field of chemical science.' Accum made it clear that his purpose was to enlighten ('to furnish a clear idea of the nature and composition of these fountains of health'); to convince by unimpeachable evidence ('to present truth in a simple form'); and to legitimate medical use of the Cheltenham waters ('to establish it [the composition] on legitimate foundations').^[59] In his Thetford analysis, he represented himself as coming to put an end to long-standing doubts about the efficacy of the waters there. He noted that a mid eighteenth century analysis by Dr Matthew Manning, a local physician, had been well done for its time, 'but the science of chemistry at the time . . . was not sufficient to enable him [Manning] to trace their [the constituents of the water] true combinations,' and hence the spa had not succeeded. Since that time however, in no other part of analytical chemistry had there been 'greater acquisition, in point of real matter of fact,' than mineral water chemistry and now 'modern chemistry [was able to give] . . . us clear and accurate information as to the nature and qualities of all the foreign matters.'^[60]

Accum's representations are ironic in two senses. First, the 'happier spirit of research' in which he placed himself and understood his abilities was rapidly disappearing in the face of Murray's criticisms. Second, Manning, the object of Accum's condescension, had made a similar offer of enlightenment three quarters of a century earlier, in his explanation of what his analysis would do for Thetford. Manning's analysis was

to establish its [the Thetford Spring's] virtues on the principles of sound science, that no one should, henceforth, presume to refuse them his assent. This analysis has, happily, succeeded beyond my utmost expectations; having most clearly proved these waters to abound in all those mineral substances required for the cure of chronic complaints .^[61]

Ducking Disagreement;
Avoiding Anomalies

The juxtaposition of the claims of Manning and Accum, both supremely confident that enlightenment had finally arrived, helps make sense of one of the most remarkable characteristics of early nineteenth century British mineral water analysis: the toleration of

discordant results. One of the diagrams in figure 2.2 is taken from an 1847 paper by Merck and Galloway, two of Hofmann's protégés, on the waters at Bath. The paper included their new analysis of the Bath waters (done by Hofmann's 'usual method') and this comparison of their results with those obtained by five earlier analysts, going back to Richard Phillips' analysis in 1806. As the authors noted previous analysts had disagreed significantly as to what was in the Bath waters: 'Besides great differences in the quantitative analysis, we find discrepancies even in regard to the presence and absence of certain constituents.'^[62] Indeed it was these discrepancies that seemed to warrant a new analysis.

We can understand the tolerance of this range of results only in terms of the optimism that pervaded chemistry in the early nineteenth century. At each stage practitioners felt they finally had gotten the chemistry right, just as we do now; hence each subsequent analysis was to make clear which if any of the past efforts had been pretty nearly right and which wrong. If there was anything ironic it was the longevity of this optimism. The first of the authors Merck and Galloway considered, Richard Phillips, had raised the same issues in his report, wondering how the analysts of the eighteenth century could have come up with anywhere from 17 to 34 grains/quart of solid matters in the Bath waters.^[63] Likewise, there was widespread recognition among chemists that incompetent and fraudulent analyses were sometimes done, and that there was need for more uniformity in technique and more stringency in qualification. Yet it was always another who was incompetent and fraudulent. Not until the mid '70s when these concerns finally led to the formation of the Institute of Chemistry and the Society of Public Analysts would the community of chemists find sufficient organization to tackle these problems.^[64]

There was also a wholly benign explanation for the discrepancy, the expectation that nature varied enormously. Often chemists explained discrepant analyses of a particular spring by attributing these differences to variations in the waters themselves. While this excuse could plausibly be stretched only so far, it did have the advantages of deflecting public criticism from chemists' competence and techniques, of dissipating incipient intra-professional conflict, and of suggesting the need for more analyses so that the full range of variability could be determined.^[65]

These perspectives carry over into the potable water analysis of the second half of the century. There too we find a continuing strong

[Full Size]

Figure 2.2
A troubling question for the early nineteenth century mineral water analysts was
how closely should independent analyses of the same water agree. Note the figure
for fixed salts in the Herapath and Herapath table of analyses of the Dead Sea,
ranging from 15.15 to 38.5 grains. On the Merck and Galloway table note Noad's
listing of the carbonate as carbonate of soda while the other analysts list it as
carbonate of lime. John Murray's insights explained such discrepancies
(Phil Mag  3rd series 31 [1847]: 67,  J Chem Soc  2 [1849]: 344).

belief that a society in which chemistry informed decisions in matters of health and industry was vastly superior to one in which it did not. There too there was a willingness to accept as legitimate, analyses done under circumstances where those funding the analyses had a direct financial stake in the outcome of the analyses. There too one finds discrepant results (and more importantly, discrepant interpretations). Finally, there too one sees the conviction that the troublesome problems in assessing water quality lay in nature rather than in chemistry, and would be resolved by more chemistry.