Physicians, chemists, and the analysis of mineral waters: "The most difficult part of chemistry."

In the first volume of his Physical and chemical essays (1779), Torbern Bergman, professor of chemistry and pharmacy at Uppsala in Sweden, discussed the chemical analysis of natural waters, including mineral waters. He remarked that when a weighed quantity of a mineral water was evaporated to dryness the weight of the residues usually amounted to no more than a minute fraction of the original weight of water, yet it might contain six or eight ingredients including salts and earths, which had first to be identified and then separated to determine their proportions. Bergman remarked that, "An accurate analysis of waters is justly considered as one of the most difficult problems in chymistry."'l Most eighteenth-century mineral water analysts were inclined to agree. Quite apart from the problems of identifying small quantities of salts and earths with very similar chemical properties, there was no guarantee that the residue left after evaporation contained the same ingredients as the mineral water itself. J. A. Chaptal, professor of chemistry and a practising physician at Montpellier, observed that new compounds were sometimes formed during evaporation whilst others present in the mineral water were decomposed.2 In either case the results of analysis would be misleading. Thomas Garnett, writing about Harrogate waters in 1794, put the blame for the unreliability of mineral water analysis on the "very low state of chemical knowledge, together with the many difficulties which attended the examination of mineral waters . . .".3 To analyse a mineral water it was first necessary to know the chemical identity of the dissolved substances. This in itself was difficult since the quantities held in solution were usually very small and many of the compounds dissolved in mineral waters were so similar in their chemical properties that it was hard to distinguish between them. Yet for a reliable chemical analysis these closely related salts and earths had to be identified either in the mineral water itself or in the residues which remained after evaporation. In the latter case, the analyst also had to be aware of any chemical changes which might have occurred during the evaporation. It is clear