c 233 n 
 
 IX. An Account of some Experiments and Observations on the 
 constituent Parts of certain astringent Vegetables ; and on their 
 Operation in Tanning. By Humphry Davy, Esq. Professor of 
 Chemistry in the Royal Institution. Communicated by the Right 
 Hon. Sir Joseph Banks, Bart. K. B. P. R. S. 
 
 Read February 24, 1803. 
 
 The discovery made by M, Seguin, of a peculiar vegetable 
 matter which is essential to the tanning of skin, and which is 
 possessed of the property of precipitating gelatine from its solu- 
 tions, has added considerably to our knowledge of the constituent 
 parts of astringent vegetables. 
 
 Mr. Proust has investigated many of the properties of this 
 substance; but, though his labours, and those of other chemists, 
 have led to various interesting observations, yet they are far 
 from having exhausted the subject. The affinities of tannin have 
 been hitherto very little examined; and the manner in which 
 its action upon animal matters is modified by combination with 
 other substances, has been scarcely at all studied. 
 
 At the desire of the Managers of the Royal Institution, I 
 began, in September, 1801, a series of experiments on the sub- 
 stances employed in the process of tanning, and on the chemical 
 agencies concerned in it. These experiments have occupied, 
 ever since, a considerable portion of my leisure hours ; and I now 
 presume to lay before the Royal Society an account of their ge- 
 neral results. My chief design was, to attempt to elucidate the 
 
 MDCCCIII. H h 
 
234 ’ Davy’s Experimerits and Observations 
 
 practical part of the art ; but, in pursuing it, I was necessarily 
 led to general chemical inquiries concerning the analysis of the 
 different vegetable substances containing tannin, and their pe- 
 culiar properties. 
 
 I. OBSEx^.VATIONS ON THE ANALYSIS OF ASTRINGENT VEGETABLE 
 
 INFUSIONS. 
 
 Tlie substances that have been supposed to exist most gene- 
 rally in astringent infusions are, tannin, gallic acid, and extractive 
 matter. 
 
 Tlie presence of tannin in an infusion, is denoted by the pre- 
 cipitate it forms with the solution of glue, or of isinglass. And, 
 when this principle is wholly separated, if the remaining liquor 
 gives a dark colour w'ith the oxygenated salts of iron, and an 
 immediate precipitate with the solutions of alum and of muriate 
 of tin, it is believed to contain gallic acid, and extractive matter. 
 
 The experiments of MM. Fourcroy, Vauouelin, and 
 Seguin, have shown that many astringent solutions undergo 
 a change by exposure to the atmosphere ; an insoluble matter 
 being precipitated from them, A precipitation is likewise occa- 
 sioned in them by the action of heat ; and these circumstances 
 render it extremely difficult to ascertain, with any degree of 
 precision, the quantities of their constituent parts, as they exist 
 in the primitive combination. 
 
 Affer trying several experiments on different methods of 
 ascertaining the quantity of tannin in astringent infusions, I was 
 induced to employ the common process of precipitation by gela- 
 tine, as being the most accurate. 
 
 This process, however, requires many precautions. The tan- 
 ning principle in different vegetables, as will be seen hereafter, 
 demands for its saturation different proportions of gelatine; and 
 
on the constituent Pai'ts of astrmgent Vegetables. 235 
 
 the quantity of the precipitate obtained by filtration, is not always 
 exactly proportional to the quantities of tannin and gelatine in 
 solutions, but is influenced by the degree of their concentration. 
 Thus, 1 found that 10 grains of dry isinglass, dissolved in two 
 ounces of distilled water, gave, with solution of galls in ex- 
 cess, a precipitate weighing, when dry, 17 grains; whilst the 
 same quantity, dissolved in six ounces of water, produced, all 
 other circumstances being similar, not quite 1 5 grains. With 
 more diluted solutions, the loss was still greater; and analogous 
 effects took place, when equal portions of the same solution of 
 isinglass were acted on by equal portions of the same infusion 
 of galls diluted in different degrees with water; tlie least 
 quantity of precipitate being always produced by the least con- 
 centrated liquor. In all cases, when the weak solutions ^^■ere 
 used, it was observed, that the residual fluid, though passed two 
 or three times through the filtre, still remained more or less 
 turbid and opaque; so that it is most likely that the deficiency 
 arose from the continued suspension of some of the minutely 
 divided solid matter in the liquid mass. 
 
 The solutions of gelatine, for the purposes of analysis, should 
 be employed only when quite fresh, and in as high a state of 
 saturation as is compatible with their perfect fluidity. I have 
 observed, that in cases when they approach towards the state of 
 jelly, their power of acting upon tannin is materially altered, 
 and they produce only a very slight precipitation. As the de- 
 gree of fluidity of solutions of gelatine is influenced by their 
 temperature, I have found it expedient, in all comparative ex- 
 periments, to bring them and the astringent infusions on which 
 they are designed to act, as nearly as possible to a common 
 degree of heat. My standard temperature has been between 6’o 
 
 H h 3 
 
*3^ Davy’s Experiments a?id Observations 
 
 and 70° Fahrenheit; and the solutions of gelatine that I have 
 used, were made by dissolving 120 grains of isinglass in 20 
 ounces of water. 
 
 In ascertaining the proportions of tannin in astringent infu- 
 sions, great care must be taken to prevent the presence of any 
 excess of gelatine; for, when this excess exists, I have found 
 that a small portion of the solid compound formed is redissolved, 
 and the results of the experiment otherwise affected. It is not 
 difficult to discover the precise point of saturation, if the solution 
 of isinglass be added only in small quantities at a time, and if 
 portions of the clear liquor be passed through a filtre at different 
 periods of the process. The properties of these portions will 
 indicate the quantities of the solution of gelatine required for 
 the completion of the experiment. _ 
 
 That the composition of any precipitate containing tannin and 
 gelatine may be known with a tolerable degree of precision, it 
 is necessary that the isinglass employed in the solution, and the 
 new compound formed, be brought as nearly as possible to the 
 same degree of dryness. For this purpose, I have generally ex- 
 posed them, for an equal time, upon the low^er plate of a sand- 
 bath, which was seldom heated to more than 150°. This method 
 I have found much better than that of drying at the temperatures 
 of the atmosphere, as the different states of the air, with regard 
 to moisture, materially influence the results. 
 
 Mr. Hatchett has noticed, in his excellent Paper on Zoo- 
 phytes, &c.* that isinglass is almost wholly composed of gelatine. 
 
 I have found, that 100 grains of good and dry isinglass contain 
 rather more than 98 grains of matter soluble in water. So that, 
 when the quantity of isinglass, in any solution employed for 
 
 • Phil. Trans, for 1800, page 327. 
 
on the constituent Parts of astringent Vegetables. 237 
 
 acting upon an astringent infusion, is compared with the quan- 
 tity of the precipitate obtained, the difference between them will 
 indicate the proportion of tannin, as it exists in the combination. 
 
 After the tannin has been separated from an astringent in- 
 fusion, for the purpose of ascertaining its other component parts, 
 I have been accustomed to evaporate the residual liquor very 
 slowly, at a temperature below 200°.* In this process, if it 
 contains extractive matter, that substance is in part rendered 
 insoluble, so as to fall to the bottom of the vessel. When the 
 fluid is reduced to a thick consistence, I pour alcohol upon it. 
 If any gallic acid or soluble extractive matter be present, they 
 will be dissolved, after a little agitation, in the alcohol; whilst 
 the mucilage, if any exist, will remain unaltered, and may be 
 separated from the insoluble extract, by lixiviation with water. 
 
 I have made many experiments, with the hope of discovering 
 a method by which the respective quantities of gallic acid and 
 extractive matter, when they exist in solution in the alcohol, 
 may be ascertained ; but without obtaining success in the results. 
 It is impossible to render the whole of any quantity of extrac- 
 tive matter insoluble by exposure to heat and air, without at the 
 same time decomposing a portion of the gallic acid. That acid 
 cannot be sublimed, without being in part destroyed ; and, at 
 the temperature of its sublimation, extractive matter is wholly 
 converted into new products. 
 
 Ether dissolves gallic acid; but it has comparatively little 
 
 * M. Deyeux has shewn, ( Annales de Cbimie, Tome XVII. page 36,) that in 
 tlie pirocess of evaporating solutions of galls, no gallic acid is carried over by the 
 water, at a temperature below that of ebullition. Many astringent infusions, however, 
 lose a portion of their aromatic principle, even in cases when they are not made to 
 boil ; but this substance, though evident to the smell, in the water that comes over, 
 cannot be detected by chemical reagents. 
 
238 M}\ Davy’s Experiments and Observations 
 
 action upon extractive matter, I have been able, in examining 
 solutions of galls, to separate a portion of gallic acid by 
 means of ether. But, when the extractive matter is in large 
 quantities, tliis method does not succeed, as, in consequence of 
 that affinity which is connected with mass,* the greatest part of 
 the acid continues to adliere to the extract, 
 
 Alumine has a strong attraction for extractive matter; but 
 comparatively a weak one for gallic acid.-f When carbonate of 
 alumine is boiled for some time with a solution containing ex- 
 tractive matter, the extractive matter is wholly taken up by the 
 earth, with which it forms an insoluble compound; but, into 
 this compound, some of the gallic acid appears likewise to enter; 
 and the portion remaining dissolved in the solution is always 
 combined with alumine. 
 
 I have not, in any instance, been able to separate gallic acid 
 and extractive matter perfectly from each other; but I have 
 generally endeavoured to form some judgment concerning their 
 relative proportions, by means of the action of the salts of alu- 
 mine, and the oxygenated salts of iron. Muriate of alumine 
 precipitates much of the extractive matter from solutions, without 
 acting materially upon gallic acid; and, after this precipitation, 
 some idea may be formed concerning the quantity of the gallic 
 acid, by the colour it gives with the oxygenated sulphate of iron. 
 In this process, however, great care must be taken not to add 
 the solution of the sulphate of iron in excess ; for, in this case, 
 the black precipitate formed with the gallic acid will be redis- 
 solved, and a clear olive-coloured fluid only will be obtained. 
 
 * See Berthollet, Recherches sur les Lois de V Affiniie Mem. de I'Listitul ISfa^ 
 tional. Tome 111 . p. 5. 
 
 f See Fiedler, Journal de Cbimie, par ]. B. Van-mons, Tome I. pag. 85. 
 
on the constituent Parts of astringent Vegetables. 239 
 
 The saline matters in astringent infusions, adhere so strongly 
 to tlie vegetable principles, that it is impossible to ascertain their 
 nature with any degree of accuracy, by means of common re- 
 agents. By incineration of the products obtained from the eva- 
 poration of astringent infusions, I have usually procured carbo- 
 nate of lime and carbonate of potash. 
 
 In the different analyses, as will be seen from the results 
 given in the following sections, I have attended chiefly to the 
 proportions of the tanning principle, and of the principles pre- 
 cipitable by the salts of iron, as being most connected with 
 practical applications. 
 
 With regard to the knowledge of the nature of the different 
 substances, as they exist in the primitive astringent infusion, we 
 can gain, by our artificial methods of examination, only very 
 imperfect approximations. In acting upon them by reagents, 
 we probably, in many cases, alter their nature; and very few 
 of them only can be obtained in an uncombined state. The 
 comparisoh, however, of the products of different experiments 
 with each other, is always connected with some useful conclu- 
 sions ; and the accumulation of facts with regard to the subject, 
 must finally tend to elucidate this obscure but most interesting 
 part of chemistry. 
 
 II. EXPERIMENTS ON THE INFUSIONS OF GALLS. 
 
 I have been very much assisted in my inquiries concerning 
 the properties of the infusions of galls, by the able Memoir of 
 M. Deyeux, on galls.* 
 
 The strongest infusion of galls that I could obtain, at 5^° 
 * Annales de Chimie, Tome XVII, pag. 1. 
 
240 Davy’s Experiments and Observations 
 
 Fahrenheit, by repeatedly pouring distilled water upon the 
 best Aleppo galls broken into small pieces, and suffering it to 
 remain in contact witli them till the saturation was complete, 
 was of the specific gravity i.o 58 . Four hundred grains of it pro- 
 duced, by evaporation at a temperature below 200°, fifty-three 
 grains of solid matter ; which, as well as I could estimate, by the 
 methods of analysis that have been just described, consisted of 
 about ^ of tannin, or matter precipitable by gelatine, and 
 of gallic acid, united to a minute portion of extractive matter. 
 
 100 grains of the solid matter obtained from the infusion, 
 left, after incineration, nearly 4^ grains of ashes ; which were 
 chiefly calcareous matter, mixed with a small portion of fixed 
 alkali. The infusion strongly reddened paper tinged with litmus. 
 It was semitransparent, and of a yellowish-brown colour. 
 Its taste was highly astringent. 
 
 When sulphuric acid was poured into the infusion, a dense 
 whitish precipitate was produced; and this effect was constant, 
 whatever quantity of the acid was used. The residual liquor, 
 when passed through the filtre, was found of a shade of colour 
 deeper than before. It precipitated gelatine, and gave a dark 
 colour with the oxygenated sulphate of iron. 
 
 The solid matter remaining on the filtre, slightly reddened 
 vegetable blues ; and, when dissolved in warm water, copiously 
 precipitated the solutions of isinglass. M. Proust,* who first 
 paid attention to its properties, supposes that it is a compound 
 of the acid with tannin ; but I suspect that it also contains 
 gallic acid, and probably a small portion of extractive matter. 
 
 * The fact of the precipitation of solution of galls by acids, was noticed by M. 
 Dize'. See Annales de Chimie, Tome XXXV. p. 37, 
 
on the constituent Parts of astringent Vegetables. 241 
 
 This last substance, as is well known, is thrown down from its 
 solutions by sulphuric acid ; and I found, in distilling the pre- 
 cipitate from galls by sulphuric acid, at a heat above 212°, that 
 a fluid came over, of a light yellow colour, which was ren- 
 dered black by oxygenated sulphate of iron ; but which was not 
 altered by gelatine. 
 
 Muriatic acid produced, in the infusion, effects analogous to 
 those produced by sulphuric acid ; and two compounds of the 
 acid and the vegetable substances were formed ; the one united 
 to excess of acid, which remained in solution ; the other con- 
 taining a considerable quantity of tannin, which was precipi- 
 tated in the solid form. 
 
 When concentrated nitric acid was made to act upon the 
 infusion, it was rendered turbid; but the solid matter formed was 
 immediately dissolved with effervescence, and the liquor then be- 
 came clear, and of an orange colour. On examining it, it was 
 found that both the tannin and the gallic acid were destroyed ; 
 for it gave no precipitate, either with gelatine or the salts of iron, 
 even after the residual nitric acid was saturated by an alkali. 
 By evaporation of a portion of the fluid, a soft substance was 
 obtained, of a yellowish-brown colour, and of a slightly sourish 
 taste. It was soluble in water, and precipitated the nitro-muriate 
 of tin, and the nitrate of aluraine; so that its properties approached 
 to those of extractive matter ; and it probably contained oxalic 
 acid, as it rendered turbid a solution of muriate of lime. 
 
 When a very weak solution of nitric acid was mixed with the 
 infusion, a permanent precipitate was formed ; and the residual 
 liquor, examined by the solution of gelatine, was found to con- 
 tain tannin. 
 
 A solution of pure potash was poured into a portion of the 
 
 MDCCCIII. I i 
 
24,2 Mr. Davy’s Experiments and Observations 
 
 infusion. At first, a faint turbid appearance was perceived; but, 
 by agitation, the fluid became clear, and its colour changed from 
 yellow brown to brown red ; and this last tint was most vivid 
 on the surface, where the solution was exposed to the atmo- 
 sphere. The solution of isinglass did not act upon the infusion 
 modified by the alkali, till an acid was added in excess, when a 
 copious precipitation was occasioned. 
 
 The compound of potash and solution of galls, when eva- 
 porated, appeared in the form of an olive-coloured mass, which 
 h.ad a faint alkaline taste, and which slowly deliquesced when 
 exposed to the air. 
 
 Soda acted upon the infusion in the same manner as potash; 
 and a fluid was formed, of a red-brown colour, which gave no 
 precipitate to gelatine. 
 
 Solution of ammonia produced the same colour as potash and 
 soda, and formed so perfect an union with the tannin of the 
 infusion, that it was not acted upon by gelatine. When the 
 compound liquor was exposed to the heat of boiling water, a 
 part of the ammonia flew off, and another part reacted upon 
 the infusion, so as to effect a material change in its properties. 
 A considerable quantity of insoluble matter was formed ; and 
 the remaining liquor contained little tannin and gallic acid, but 
 a considerable portion of a substance that precipitated muriate 
 of tin, and the salts of alumine. 
 
 When the experiment on the ebullition of the com.pound of 
 the infusion and ammonia was made in close vessels, the liquor 
 that came over was strongly impregnated with ammonia ; 
 its colour was light yellow, and, when saturated with an acid, 
 it was very little altered by the salts of iron. The residual fluid, 
 after the process had been continued for some time, as in the 
 
on the constituejit Parts of astringent Vegetables. 243 
 
 other case, precipitated gelatine slightly, but the salts ofalumine 
 copiously ; and it gave a tinge of red to litmus paper. 
 
 When solution of lime, of strontia, or of barytes, was poured 
 in excess into a portion of the infusion, a copious olive-coloured 
 precipitate was formed, and the solution became almost clear, 
 and of a reddish tint. In this case, the tannin, the gallic acid, and 
 the extractive matter, seemed to be almost wholly carried down 
 in the precipitates; as the residual fluids, when saturated by an 
 acid, gave no precipitate to gelatine, and only a very slight tint 
 of purple to oxygenated sulphate of iron. 
 
 When the solutions of the alkaline earths were used only in 
 small quantities, the infusion being in excess, a smaller quantity 
 of precipitate was formed, and the residual liquor was of an 
 olive-green colour ; the tint being darkest in the experiment 
 with the barytes, and lightest in that with the lime. This fluid, 
 when examined, was found to hold in solution a compound of 
 gallic acid and alkaline earth. It became turbid when acted on 
 by a little sulphuric acid ; and, after being filtrated, gave a black 
 colour with the solutions of iron, but was not acted upon by 
 gelatine. 
 
 When a large proportion of lime was heated for some time 
 with the infusion, it combined with all its constituent principles, 
 and gave, by washing, a fluid which had the taste of lime-water, 
 and which held in solution only a very small quantity of ve- 
 getable matter. Its colour was pale yellow ; and, when saturated 
 with muriatic acid, it did not precipitate gelatine, and gave only 
 a slight purple tinge to the solutions of the salts of iron. The 
 lime in combination with the solid matter of the infusion, was of 
 a fawn colour. It became green at its surface, where it was ex- 
 posed to the air; and, wdien washed with large quantities of water, 
 
 I i 2 
 
244 Davy’s Experhnents and Observations 
 
 it continued to give, even to the last portions, a pale yellow 
 tinge. 
 
 Magnesia was boiled in one portion of the infusion for a few 
 hours ; and mixed in excess with another portion, which was 
 suiFered to remain cold. In both cases, a deep green fluid was 
 obtained, which precipitated the salts of iron, but not the solu- 
 tions of gelatine; and the magnesia had acquired a grayish- 
 green tint. Water poured upon it became green, and acquired 
 the properties of the fluid at first obtained. After long washing, 
 the colour of the magnesia changed to dirty yellow ; and the 
 last portions of water made to act upon it were pale yellow, and 
 altered very little the solutions of iron. > 
 
 When the magnesia was dissolved in muriatic acid, a brownish 
 and turbid fluid was obtained, which precipitated gelatine and 
 the oxygenated salts of iron. So that there is every reason to 
 believe, that the earth, in acting on the astringent infusion, had 
 formed two combinations ; one containing chiefly gallic acid, 
 which was easily soluble in water; the other containing chiefly 
 tannin, which was very diflicultly soluble. 
 
 Alumine boiled with the infusion became yellowish-gray, and 
 gave a clear white fluid, which produced only a tinge of light 
 purple in the solutions of iron. When the earth* was employed 
 in very small quantity, however, it formed an insoluble com- 
 pound only with the tannin and the extract; and the residual liquor 
 was found to contain a gallate of alumine with excess of acid. 
 
 The oxides of tin and of zinc, obtained by nitric acid, were 
 boiled with separate portions of the infusion for two hours. In 
 both cases, a clear fluid, which appeared to be pure water, was 
 
 * Mr. Fiedler, I believe, first observed the action of alumine upon tannin. Van- 
 Mgns’s Journal, Vol. I, page 86. 
 
on the constituent Parts of astringent Vegetables. 245 
 
 obtained ; and the oxides gained a tint of dull yellow. A part of 
 each of them was dissolved in muriatic acid. The solution ob- 
 tained was yellow: it copiously precipitated gelatine; and gave 
 a dense black with the salts of iron. Mr. Proust,* who first 
 observed the action of oxide of tin upon astringent infusions, 
 supposes that portions of tannin and gallic acid are decomposed 
 in the process, or converted, by the oxygen of the oxide, into 
 new substances. These experiments do not, however, appear to 
 confirm the supposition. 
 
 M. Deyeux observed, that a copious precipitation was 
 occasioned in infusion of galls, by solutions of the alkalis 
 combined with carbonic acid. Mr. Proust has supposed that 
 the solid matter formed is pure tannin, separated from its so- 
 lution by the stronger affinity of the alkali for water ; and he 
 recommends the process, as a method of obtaining tannin. 
 
 In examining the precipitate obtained by carbonate of potash 
 fully combined with carbonic acid, and used to saturation, I 
 have not been able to recognise in it the properties which are 
 usually ascribed to tannin : it is not possessed of the astringent 
 taste; and it is but slightly soluble in cold water, or in alcohol. 
 Its solution acts very little upon gelatine, till it is saturated with 
 an acid; and it is not possessed of the property of tanning skin. 
 
 In various cases, in which the greatest care was taken to use 
 no excess, either of the astringent infusion or of the alkaline 
 solution, I have found the solid matter obtained possessed of 
 analogous properties ; and it has always given, by incineration, 
 a considerable portion of carbonate of potash, and a small quan- 
 tity of carbonate pf lime. 
 
 The fluid remaining after the separation Tf the precipitate, 
 
 * Annales de Cbimic, Tome XLII. p. 69. 
 
2^6 Mr. Davy’s Experime?iis and Observations 
 
 was of a dark-brown colour, and became green at the surface, 
 when it was exposed to the air. It gave no precipitate to solu- 
 tion of gelatine ; and afforded only an olive-coloured precipi- 
 tate with the salts of iron. 
 
 When muriatic acid was poured into the clear fluid, a violent 
 effervescence was produced ; the fluid became turbid; a precipi- 
 tate was deposited ; and the residual liquor acted upon gelatine and 
 tlie salts of iron, in a manner similar to the primitive infusion. 
 
 M. Deyeux, in distilling the precipitate from infusion of galls 
 by carbonate of potash, obtained crystals of gallic acid. In fol- 
 lowing his process, I had similar results ; and a fluid came over, 
 which reddened litmus-paper, and precipitated the salts of iron 
 black, but did not act upon gelatine. 
 
 When the precipitate by carbonate of potash was acted upon 
 by warm water, applied in large quantities, a considerable por- 
 tion of it was dissolved ; but a part remained, which could not 
 in any way be made to enter into solution ; and its properties 
 were very different from those of the entire precipitate. It was 
 not at all affected by alcohol ; it was acted on by muriatic acid, 
 and partially dissolved ; and the solution precipitated gelatine 
 and the salts of iron. It afforded, by incineration, a considerable 
 portion of lime, but no alkali. 
 
 In comparing these facts, it would seem, that the precipitate 
 from infusion of galls, consists partly of tannin and gallic acid 
 united to a small quantity of alkali, and partly of these vegetable 
 matters combined with calcareous earth ; and it will appear pro- 
 bable, when the facts hereafter detailed are examined, that both 
 'the potash and the lime arc contained in these compounds in a 
 state of union with carbonic acid. 
 
 The solutions of carbonate of soda and of carbonate of 
 
071 the co7istituent Pai'ts of astringe7it Vegetables. 24^'/ 
 
 ammonia, both precipitated the infusion of galls in a manner 
 similar to the carbonate of potash ; and each of the precipitates, 
 when acted on by boiling water, left a small quantity of insoluble 
 matter, which seemed to consist chiefly of tannin and carbonate 
 of lime. 
 
 The entire precipitate by carbonate of soda produced, wlieii 
 incinerated, carbonate of soda and carbonate of lime. The pre- 
 cipitate by carbonate of ammonia, when exposed to a heat 
 sufficient to boil water, in a retort having a receiver attached to 
 it, gave out carbonate of ammonia, (which was condensed in 
 small crystals in the neck of the retort,) and a yellowish fluid, 
 which had the strong smell and taste of this volatile salt. After 
 the process of distillation, the solid matter remaining was found 
 of a dark brown colour; a part of it readily dissolved in cold 
 water, and the solution acted on gelatine. 
 
 The residual fluid of the portions of the infusion which had 
 been acted on by the carbonates of soda and of ammonia, as in 
 the instance of the carbonate of potash, gave no precipitate with 
 gelatine, till they were saturated with an acid; so that, in all 
 these cases, the changes are strictly analogous. 
 
 The infusion of galls, as appears from the analysis, contains 
 in its primitive state calcareous matter. By the action of the 
 mild alkalis, this substance is precipitated in union with a por- 
 tion of the vegetable matter, in the form of an insoluble com- 
 pound. The alkalis themselves, at the same time, enter into 
 actual combination with the remaining tannin and gallic acid ; 
 and a part of the compound formed is precipitated, whilst anotiier 
 part remains in solution. 
 
 When the artifleial carbonates of lime, magnesia, and barytes, 
 were separately boiled with portions of the infusion of galls for 
 
24,8 Mr. Davy’s Experiments and Observations 
 
 some hours, they combined with the tannin contained in it, so 
 as to form with it insoluble compounds ; and, in each case, a 
 deep green fluid w'as obtained, w'hich gave no precipitate to 
 gelatine, even when an acid was added, but which produced ti 
 deep black colour in the solutions of the salts of iron. 
 
 Sulphate of lime, when finely divided, whether natural or 
 artificial, after having been long heated with a small quantity 
 of the infusion, was found to have combined with the tannin of 
 it, and to have gained a faint tinge of light brown. The liquid 
 became of a blue-green colour, and acted upon the salts of iron, 
 but not upon gelatine ; and there is every reason to suppose, 
 that it held in solution a triple compound, of gallic acid, sulphuric 
 acid, and lime. 
 
 We owe to Mr. Proust, the discovery that different solutions 
 of the neutral salts precipitate the infusion of galls ; and he 
 supposes, that the precipitation is owing to their combining with 
 a portion of the water which held the vegetable matter in solu- 
 tion. In examining the solid matters thrown down from the 
 infusion, by sulphate of alumine, nitrate of potash, acetite of 
 potash, muriate of soda, and muriate of barytes, I found them 
 soluble, to a certain extent, in water, and possessed of the power 
 of acting upon gelatine. From the products given by their inci- 
 neration, and by their distillation, 1 am however inclined to 
 believe that they contahi, besides tannin, a portion of gallic acid 
 and extractive matter, and a quantity of the salt employed in the 
 primitive solution. 
 
 It is well known, that many of the metallic solutions occasion 
 dense precipitates in the infusion of galls; and it has been gene- 
 rally supposed, that these precipitates are composed of tannin and 
 extractive matter, or of those tw'o substances and gallic acid, united 
 
on the constituent Parts oj astringent Vegetables. 249 
 
 to the metallic oxide ; but, from the observation of different pro- 
 cesses of this kind, in which the salts of iron and of tin were 
 employed, I am inclined to believe, that they contain also a 
 portion of the acid of the saline compound. 
 
 When the muriate of tin was made to act upon a portion of 
 the infusion, till no more precipitation could be produced in it, the 
 fluid that passed through the filtre still acted upon gelatine, and 
 seemed to contain no excess of acid; for it gave a precipitate 
 to carbonate of potash, without producing effervescence. The 
 solid compound, when decomposed by sulphuretted hydrogen, 
 after the manner recommended by Mr. Proust, was found 
 strongly to redden litmus-paper, and it copiously precipitated 
 nitrate of silver ; whereas, the primitive infusion only rendered 
 it slightly turbid ; so that there is every reason to believe, that 
 the precipitate contained muriatic acid. 
 
 By passing the black and turbid fluid, procured by the action 
 of solution of oxygenated sulphate of iron in excess upon a 
 portion of the infusion, through finely-divided pure flint, con- 
 tained in four folds of filtrating paper, I obtained a light olive- 
 green fluid, in which there was no excess of sulphuric acid, and 
 which I am inclined to suppose was a solution of the compound 
 of gallic acid and sulphate of iron, with superabundance of me- 
 tallic salt. I have already mentioned that gallic acid, when in 
 very small proportion, does not precipitate the oxygenated salts 
 of iron ; and Mr. Proust, in his ingenious Paper upon the Dif- 
 ference of the Salts of Iron, has supposed that, in the formation 
 of ink, a portion of the oxide of iron in union with gallic acid 
 is dissolved by the sulphuric acid of the sulphate. This comes 
 near to the opinion that they form a triple compound : and, in 
 reasoning upon the general phenomena, it seems fair to 
 
 MDCCCIII. K k 
 
2^0 Mr. Davy's Experiments and Observations 
 
 conclude, that, in the case of the precipitation of tannin by the 
 salts of tin and of iron, compounds are formed, of tannin and 
 the salts ; and that, of these compounds, such as contain tin are 
 Slightly soluble in water, whilst those that contain iron are 
 almost wholly insoluble. 
 
 In examining the action of animal substances upon the infu- 
 sion of galls, with the view of ascertaining the composition cf 
 th.e compounds of gelatine, and of skin, with tannin, I found that 
 a saturated solution of gelatine, which contained the soluble 
 matter of 50 grains of dry isinglass, produced from the infusion 
 a precipitate that weighed nearly 91 grains ; and, in another 
 instance, a solution containing 30 grains of isinglass, gave about 
 56 grains ; so that, taking the mean of the tv/o experiments, 
 and allowing for the small quantity of insoluble matter in 
 isinglass, we may conclude, that 100 grains of the compound of 
 gelatine and tannin, formed by precipitation from saturated 
 solutions, contain about 54, grains of gelatine, and 46 of tannin. 
 
 A piece of dry calf-skin, perfectly free from extraneous matter, 
 that weighed 180 grains, after being prepared for tanning by 
 long immersion in water, was tanned in a portion of the infu- 
 sion, being exposed to it for three weeks. When dry, the leather 
 weighed 295 grains ; so that, considering this experiment as 
 accurate, leather quickly tanned by means of an infusion of 
 galls, consists of about bi grains of skin, and 39 of vegetable 
 matter, in 100 grains. 
 
 After depriving a portion of the infusion of all its tanning 
 matter, by repeatedly exposing it to the action of pieces of skin, 
 I found that it gave a much slighter colour to oxygenated sul- 
 piiateof iron, than an equal portion of a similar infusion which 
 had been immediately precipitated by solution of isinglass; 
 
on the constituent Parts of astringent Fegetables. 25 < 
 
 but I am inclined to attribute this effect, not to any absorption . 
 of gallic acid by the skin, but rather to the decomposition of it 
 by the long continued action of the atmosphere; for much in- 
 soluble matter had been precipitated, during the process of 
 tanning, and the residuum contained a small portion of acetous 
 acid. 
 
 In ascertaining the quantity of tannin in galls, I found that 
 500 grains of good Aleppo galls gave, by lixiviation with pure 
 water till their soluble parts were taken up, and subsequent slov/ 
 evaporation, 185 grains of solid matter. And this matter, exa- 
 mined by analysis, appeared to consist, 
 
 Of tannin - - - - 130 grains. 
 
 Of mucilage, and matter rendered insoluble by 
 evaporation - - - - 12 
 
 Of gallic acid, with a little extractive matter - 31 
 
 Remainder, calcareous earth and saline matter 12 
 The fluid obtained by the last lixiviation of galls, as M. 
 Deyeux observed, is pale green ; and I am inclined to believe, 
 that it is chiefly a weak solution of gallate of lime. The ashes 
 of galls, deprived of soluble matter, furnish a very considerable 
 quantity of calcareous earth. And the property which M. 
 Deyeux discovered in the liquor of the last lixiviations, of be- 
 coming red by the action of acids, and of regaining tlie green 
 colour by means of alkalis, I have observed, more or less, in all 
 the soluble compounds containing gallic acid and the alkaline 
 earths. 
 
 Kk 2 
 
252 
 
 Mr. Davy's Experiments and Observations 
 
 III. EXPERIMENTS AND OBSERVATIONS ON CATECHU OR TERRA 
 
 JAPONICA. 
 
 The extract called catechu is said to be obtained from the 
 wood of a species of the Mimosa,* which is found abundantly 
 in India, by decoction and subsequent evaporation. 
 
 There are two kinds of this extract ; one is sent from Bom- 
 bay, the other from Bengal ; and they differ from each other 
 more in their external appearance than in their chemical com- 
 position. The extract from Bombay is of an uniform texture, 
 and of a red-brown tint, its specific gravity being generally 
 about 1.39. The extract from Bengal is more friable, and less 
 consistent ; its colour is like that of chocolate externally, but, 
 when broken, its fracture presents streaks of chocolate and of 
 red-brown. Its specific gravity is about 1.28. Their tastes are 
 precisely similar, being astringent, but leaving in the mouth a 
 sensation of sweetness. They do not deliquesce, or apparently 
 change, by exposure to the air. 
 
 The discovery of the tanning powers of catechu, is owing to 
 the President of the Royal Society, who, concluding from its 
 sensible properties that it contained tannin, furnished me, in 
 December, 1801, with a quantity for chemical examination. 
 
 In my first experiments, I found that the solutions of catechu 
 copiously precipitated gelatine, and speedily tanned skin ; and, 
 in consequence, I began a particular investigation of their 
 properties. 
 
 The strongest infusions and decoctions of the two different 
 kinds of catechu, do not sensibly differ in their nature, or in 
 their composition. Their colour is deep red-brown, and they 
 
 * See Kerr. Medical Observations, Vol. V. page 155. 
 
on the constituent Parts of astringent Vegetables. 253 
 
 communicate this tinge to paper ; they slightly redden litmus- 
 paper; their taste is highly astringent, and they have no per- 
 ceptible smell. 
 
 The strongest infusions tliat I could obtain from the two 
 kinds of catechu, at 48° Fahrenheit, were of the same specific 
 gravity, 1.037. decoction, I procured solutions of 
 
 1.102, which gave, by evaporation, more than of their weight 
 of solid matter. 
 
 Five hundred grains of the strongest infusion of catechu from 
 Bombay, furnished only 41 grains of solid matter; which, 
 from analysis, apjieared to consist of 34 grains of tannin, or 
 matter precipitable by gelatine, and 7 grains that were chiefly 
 a peculiar extractive matter, the properties of which will be 
 hereafter described. The quantity of solid matter given by the 
 strongest infusion of the Bengal catechu, was the same, and 
 there was no sensible difference in its composition. Portions 
 of these solid matters, when incinerated, left a residuum which 
 seemed to be calcareous ; but it was too small in quantity to be 
 accurately examined, and it could not have amounted to more 
 than of their original weights. 
 
 The strongest infusions of catechu acted upon the acids and 
 pure alkalis in a manner analogous to the infusion of galls. 
 With the concentrated sulphuric and muriatic acids, they gave 
 dense light fawn coloured precipitates. With strong nitrous 
 acid they effervesced ; and lost their power of precipitating tiie 
 solutions of isinglass, and the salts of iron. The pure alkalis 
 entered into union with their tannin, so as to prevent it from 
 being acted upr n by gelatine. 
 
 When the solutions of lime, of strontia, or of barytes, were 
 
254 ) I^avy’s Experiments and Observations 
 
 poured into the infusions, copious precipitates, of a shade of light 
 brown, were formed; and the residual fluid assumed a paler 
 tint of red, and was found to have lost its power of precipitating 
 gelatine. 
 
 After lime had been boiled for some time with a portion of 
 the infusion, it assumed a dull red colour. The liquor that 
 passed from it through the filtre had only a faint tint of red, 
 did not act upon gelatine, and seemed to contain only a very 
 small portion of vegetable matter. Pure magnesia, when heated 
 with the infusion, acted upon it in an analogous manner; the 
 magnesia became light red, and the residual fluid had only a 
 very slight tinge of that colour. With carbonate of magnesia, 
 the infusion became deeper in colour, and lost its power of pre- 
 cipitating gelatine; though it still gave, with oxygenated sulphate 
 of iron, a light oliv-e precipitate. 
 
 The carbonates of potash, of soda, and of ammonia, in their 
 concentrated solutions, produced only a slight degree of turbid- 
 ness in the infusion of catechu; they communicated to them a 
 darker colour, and deprived them of the power of acting upon 
 gelatine ; tiiough this power was restored by the addition of an 
 acid. 
 
 After the mixture of the solution of carbonate of potash and 
 the infusions had been exposed to the atmosphere for some 
 hours, a brown crust was found to have formed upon its surface, 
 and a slight precipitation had taken place. 
 
 The salts of alumine precipitated the infusions, but less co- 
 piously than they precipitate the infusion of galls. A similar 
 effect was produced by nitrate of potash, sulphate of magnesia, 
 prussiate of potash, and many other neutral salts. 
 
on the constituent Parts of astringent Vegetables. 255 
 
 The nitrate, or acetite, of lead, in concentrated solution, when 
 poured into the infusion, produced in it a dense light brown 
 precipitate, which gave to the fluid a gelatinous appearance. 
 After this effect, there was no free acid found in it; and both 
 the tannin and the extractive matter seemed to have been car- 
 ried down, in union w ith a portion of the metallic salt. 
 
 The solution of muriate of tin, acted upon the infusion of 
 catechu in a manner similar to that in which it acts upon the 
 infusion of galls. 
 
 The least oxygenated sulphate of iron produced no change in 
 the infusion. With the most oxygenated sulphate it gave a dense 
 black precipitate, which, when diffused upon paper, appeared 
 rather more inclined to olive than the precipitate from galls. 
 
 Tl’.e infusions w’ere precipitated by the solution of albumen. 
 
 The precipitates by gelatine had all a pale tint of red-browm, 
 wdiich became deeper w'hen they were exposed to tlie air. The 
 compound of gelatine and the tannin of the strongest infusions 
 of catechu appeared, by estimation of the quantit}^ of isinglass 
 ill the solutions used for their precipitation, to consist of about 
 ■41 parts of tannin, and 59 of gelatine. 
 
 Of two pieces of calf-skin which w^eighed, when dry, 132 
 grains each, and w’hicli Iiad been prepared for tanning, one was 
 immersed in a large quantity of the infusion of catechu from 
 Bengal, and the other in an equal portion of the infusion of 
 that from Bombay. In less than a montirtlicy were found con- 
 verted into leather. When freed from moisture, by long e.xposure 
 in the sunshine, tliey were weighed. The first piece had gained 
 about 34 grains; and the second piece 35I- grains. The leather 
 was of a much deeper colour than that tanned with galls, 
 
2^6 Mr. Davy’s Experiments and Observations 
 
 and on the upper surface was red-brown. It was not acted on 
 by hot or cold water ; and its apparent strength was the same 
 as that of similar leather tanned in the usual manner. 
 
 In examining the remainder of the infusions of catechu, in 
 which skin had been converted into leather, I found in them 
 much less extractive matter than I had reason to expect, from 
 the comparative analysis of equal portions of the unaltered in- 
 fusions made by solutions of gelatine. At first, I was inclined 
 to suppose that the deficiency arose from the action of the 
 atmosphere upon- the extractive matter, by which a part of it 
 was rendered insoluble. But, on considering that there had been 
 very little precipitation in the process, I was led to adopt the 
 supposition, that it had entered into union with the skin, at the 
 same time with the tannin ; and this supposition was confirmed 
 by new experiments. 
 
 Both kinds of catechu are almost wholly soluble in large 
 quantities of water; and, to form a complete solution, about 
 i8 ounces of water, at 52°, are required to 100 grains of extract. 
 The residuum seldom amounts to of the original weight of 
 the catechu ; and, in most cases, it is found to consist (ihiefly of 
 calcareous and aluminous earths, and of fine sand, which, by 
 accident or design, had probably been mixed with the primitive 
 infusion at the time of its evaporation. 
 
 A considerable portion of both kinds of catechu is soluble in 
 alcohol ; but, after the action of alcohol upon it, a substance 
 remains, of a gelatinous appearance and a light brown colour, 
 which is soluble in water, and is analogous in its properties 
 to gum or mucilage. 
 
 The peculiar extractive matter in the catechu, is much less 
 
on the constituent Parts of astringent Vegetables. 257 
 
 soluble in water than the tanning principle; and, when a small 
 quantity of water is used to a large quantity of catechu, the 
 quantity of tanmn taken up, as appears from the nature of the 
 'strongest infusion, is very much greater than that of the extrac- 
 tive matter 
 
 The extractive matter is much more soluble in warm water 
 than in cold water ; and, when saturated solutions of catechu 
 are made in boiling water, a considerable quantity of extrac- 
 tive matter, in its pure state, falls down, as the liquor becomes 
 cool. 
 
 The peculiar extractive matter of the catechu may be like- 
 wise obtained, by repeatedly lixiviating the catechu, when in 
 fine powder, till the fluids obtained cease to precipitate gelatine ; 
 the residual solid will then be found to be the substance in 
 question. 
 
 The pure extractive matter, whether procured from the 
 Bombay or Bengal catechu, is pale, with a faint tinge of red- 
 brown. It has no perceptible smell ; its taste is slightly astrin- 
 gent; but it leaves in the mouth, for some time, a sensation of 
 sweetness, stronger than that given by the catechu itself. 
 
 Its solution in water is at first yellow-brown ; but it gains a 
 tint of red by exposure to the air. Its solution in alcohol does 
 not materially change colour in the atmosphere ; and it is of an 
 uniform dull brown. 
 
 The extractive matter, whether solid or in solution, was not 
 found to produce any change of colour upon vegetable blues. 
 
 It became of a brighter colour by the action of the alkalis ; but 
 it was not precipitated from its solution in water Dy these bodies, 
 nor by the alkaline earths. 
 
 L 1 
 
 MDCCCIII. 
 
2^8 Mr. Davy’s Experiments and Observations 
 
 The aqueous solution of it, when mixed \vith solutions of 
 nitrate of alumine and of muriate of tin, became slightly turbid. 
 
 To nitrate of lead, it gave a dense light brown precipitate. 
 
 It was not perceptibly acted upon by solution of gelatine; 
 but, when solution of sulphate of alumine was added to the 
 mixture of the two fluids, a considerable quantity of solid matter, 
 of a light brown colour, was immediately deposited. 
 
 To the solution of oxygenated sulphate of iron, it communi- 
 cated a fine grass-green tint; and a green precipitate was de- 
 posited, which became black by exposure to the air. 
 
 It was not precipitated by the mineral acids. 
 
 Linen, by being boiled in the strongest solution of the ex- 
 tractive matter, acquired a light red-brown tint. The liquor 
 became almost colourless ; and, after this, produced very little 
 change in the solution of oxygenated sulphate of iron. 
 
 Raw skin, prepared for tanning by being immersed in the 
 strong solution, soon acquired the same kind of tint as the linen. 
 It united itself to a part of the extractive matter; but it was not 
 rendered by it insoluble in boiling water. 
 
 The solid extractive matter, when exposed to heat, softened, 
 and became darker in its colour, but did not enter into fusion. 
 At a temperature below that of ignition, it was decompounded. 
 The volatile products of its decomposition were, carbonic acid, 
 hydrocarbonate, and water holding in solution acetous acid and 
 a little unaltered extractive matter. There remained a lig-ht and 
 very porous charcoal. 
 
 In considering the manner in which the catechu is prepared, 
 it would be reasonable to conclude, that different specimens of 
 that substance must differ in some measure in their composition, 
 
071 the constituent Parts of astringent V egetables. 259 
 
 even in their pure states ; and, for the purposes of Commerce, 
 they are often adulterated to a considerable extent, with sand 
 and earthy matter.* 
 
 In attempting to estimate the composition of the purest cate- 
 chu, I selected pieces from different specimens, with which I 
 was supplied by the President, and reduced them together into 
 powder ; mixing, however, only those pieces which were from 
 catechu of the same kind. 
 
 Two hundred grains of the pow'der procured in this way, 
 from the catechu of Bombay, afforded by analysis, 
 
 Grains. 
 
 Tannin - - - - -509 
 
 Peculiar extractive matter - - - 68 
 
 Mucilage - - - - _ 
 
 Residual matter; chiefly sand and calcareous earth 10 
 
 The powder of the Bengal catechu gave, by similar methods 
 of analysis, in 200 grains, 
 
 Grains. 
 
 Tannin _ _ « « « 
 
 Peculiar extractive matter - - “73 
 
 Mucilage - - - - - 16 
 
 Residual matter; sand, with a small quantity of calca- 
 reous and aluminous earths - - - 14, 
 
 In examining those parts of the catechu from Bengal which 
 were differently coloured, I found the largest proportion of 
 tannin in the darkest part of the substance ; and most extractive 
 matter in the lightest part. It is probable that the inequality of 
 composition in this catechu, is owing to its being evaporated 
 
 • One specimen that I examined, of the terra japonica of commerce, furnished, by 
 Incineration, | of sand and earthy matter ; and another specimen, nearly i. 
 
 LI 2 
 
26o Mr. Davy’s Experiments and Observations 
 
 and formed without much agitation ; in consequence of which, 
 the constituent parts of it that are least soluble, being first pre- 
 cipitated, appear in some measure distinct from the more soluble 
 parts, which assume the solid form at a later period of the 
 process. 
 
 From the observations of Mr. Kerr,* it would appear, that 
 the pale catechu is that most sought after in India ; and it is 
 evidently that which contains most extractive matter. The ex- 
 tractive matter seems to be the substance that gives to the 
 catechu the peculiar sweetness of taste which follows the impres- 
 sion of astringency ; and it is probably this sweetness of taste 
 which renders it so agreeable to the Hindoos, for the purpose 
 of chewing with the betle-nut. 
 
 IV. EXPERIMENTS AND OBSERVATIONS ON THE ASTRINGENT IN- 
 FUSIONS OF BARKS, AND OTHER VEGETABLE PRODUCTIONS. 
 
 The barks that I examined were furnished me by my friend 
 Samuel Purkis, Esq. of Brentford; they had been collected in 
 the proper season, and preserved with care. 
 
 In making the infusions, I employed the barks in coarse 
 powder; and, to expedite the solution, a heat of from loo to 
 120° Fahrenheit was applied. - 
 
 Tlie strongest infusions of the barks of the oak, of the Lei- 
 cester willow, and of the Spanish chesnut, were nearly of tiie 
 same specific gravity, 1.05. Tlieir tastes were alike, and strongly 
 astringent; they all reddened litmus-paper; the infusion of the 
 Spanish chesnut bark producing the highest tint; and that of 
 the Leicester willow bark the feeblest tint. 
 
 Two hundred grains of each of the infusions were submitted 
 
 ^ Medical Observations, Vol. V, page 155. 
 
on the constituent Parts of astringent Vegetables. 261 
 
 to evaporation ; and, in this process, tlie infusion of the oak 
 bark furnished 17 grains of solid matter; that of the Leicester 
 willow about i 6 f grains; and that of the Spanish chesnut nearly 
 an equal quantity. 
 
 The tannin given by these solid matters was, in that from the 
 oak bark infusion, 14 grains; in that from the willow bark 
 infusion 14^ grains; and in that from the Spanish chesnut bark 
 infusion 13 grains. 
 
 The residual substances of the infusions of the Spanish chesnut 
 bark, and of the oak bark, slightly reddened litmus- paper, and 
 precipitated the solutions of tin of a fawn colour, and those of 
 iron black. The residual matter of the infusion of the willow 
 bark, did not perceptibly change the colour of litmus ; but it 
 precipitated the salts of iron of an olive colour, and rendered 
 turbid the solution of nitrate of alumine. 
 
 The solid matters produced by the evaporation of the infu- 
 sions, gave, by incineration, only a very small quantity of ashes, 
 which could not have been more than of their original 
 weights. These ashes chiefly consisted of calcareous earth and 
 alkali ; and the quantity was greatest from the infusion of 
 chesnut bark. 
 
 The infusions were acted on by the acids, and the pure alkalis, 
 in a manner very similar to the infusion of galls. With the so - 
 lutions of carbonated alkalis, they gave dense fawn-coloured 
 precipitates. They were copiously precipitated by the solutions of 
 lime, of strontia, and of barytes; and, by lime-water in excess, 
 the infusions of oak and of chesnut bark seemed to be deprived 
 of the whole of the vegetable matter they held in solution. 
 
 By being boiled for some time with alumine, lime, and mag- 
 nesia, they became almost colourless, and lost their power of 
 
2^2 Mr, Davyds Experiments and Observations 
 
 acting upon gelatine and the salts of iron. After being heated 
 with carbonate of lime and carbonate of magnesia, they were 
 found deeper coloured than before; and, though they had lost 
 their power of acting on gelatine, they still gave dense olive- 
 coloured precipitates with the salts of iron. 
 
 In all these cases, the earths gained tints of brawn, more or 
 less intense. 
 
 When the compound of the astringent principles of the in- 
 fusion of oak bark with lime, procured by means of lime-water, 
 was acted on by sulphuric acid, a solution was obtained, which 
 precipitated gelatine, and contained a portion of the vegetable 
 principles, and a certain quantity of sulphate of lime; a solid 
 fawn-coloured matter was likewise formed, which appeared 
 to be sulphate of lime, united to a little tannin and extractive 
 matter.^ 
 
 The solutions were copiously precipitated by solution of 
 albumen. 
 
 The precipitates they gave with gelatine were similar in their 
 appearance ; their colour, at first, was a light tinge of brown, but 
 they became very dark by exposure to the air. Their composition 
 was very nearly similar ; and, judging from the experiments 
 on the quantity of gelatine employed in forming them, the 
 compound of tannin and gelatine from the strongest infusion 
 of oak bark, seems to consist, in the lOO parts, of 59 parts of 
 
 * M. Merat Guillot proposes a method of procuring pure tannin, ( Annales de 
 Chim'ie, Tome XLI. p. 325.) which consists in precipitating a solution of tan by 
 lime-water, and decomposing it by nitric or muriatic acid. The solution of the 
 solid matter obtained in this way in alcohol, he considers as a solution of pure tannin ; 
 but, from the experiments abovementioned, it appears, that it must contain, besides 
 tannin, some of the extractive matter of the bark ; and it may likewise contain saline 
 matter. 
 
on the constituent Parts of astringent Vegetables. 2^53 
 
 gelatine and 41 of tannin; that from the infusion of Leicester 
 willow bark, of 57 parts of gelatine and 43 of tannin ; and that 
 from the infusion of Spanish chesnut bark, of 61 parts of gelatine 
 and 39 of tanniiv. 
 
 Two pieces of calf-skin, which weighed when dry 120 grains 
 each, were tanned; one in the strongest infusion of Leicester 
 willow bark, and the other in the strongest infusion of oak bark. 
 The process was completed, in both instances, in less than a 
 fortnight ; when the weight of the leather formed by the tannin 
 of the Leicester willow bark was found equal to 161 grains; 
 and that of the leather formed by the infusion of oak bark was 
 equal to 164 grains. 
 
 When pieces of skin were suffered to remain in small quan- 
 tities'of the infusions of the oak bark, and of the Leicester willow 
 bark, till they were exhausted of their tanning principle, it was 
 found, that though the residual liquors gave olive-coloured pre- 
 cipitates with the solutions of sulphate of iron, yet they were 
 scarcely rendered turbid by solutions of muriate of tin ; and 
 there is every reason to suppose, that a portion of their extractive 
 matter had been taken up with the tannin by the skin. 
 
 I attempted, in different modes, to obtain uncombined gallic 
 acid from the solid matter produced by the evaporation of the 
 barks, but without success. When portions of this solid matter 
 were exposed to the degree of heat that is required for the pro- 
 duction of gallic acid from Aleppo galls, no crystals were formed; 
 and the fluid that came over gave only a brown colour to the 
 solution of salts of iron, and W'as found to contain much acetous 
 acid and empyreumatic oil. 
 
 When pure water was made to act, in successive portions, 
 upon oak bark in coarse p owder, till all its soluble parts were 
 
26*4 Mr. Davy’s Experiments and Observatmis 
 
 taken up, the quantities of liquor last obtained, though they did 
 not act much upon solution of gelatine, or perceptibly redden 
 litmus-paper, produced a dense black with the solution of sul- 
 phate of iron : by evaporation, they furnished a brown matter, 
 of which a part was rendered insoluble in water by the action of 
 the atmosphere; and the part soluble in water was not in any 
 degree taken up by sulphuric ether ; so that, if it contained gallic 
 acid, it was in a state of intimate union with extractive matter. 
 
 Two pieces of calf-skin, which weighed when dry 94 grains 
 each, were slowly tanned ; one by being exposed to a weak in- 
 fusion of the Leicester willow bark, and the other by being acted 
 upon by a weak infusion of oak bark. The process was completed 
 in about three months ; and it was found, that one piece of skin 
 had gained in weight 14 grains, and the other piece about 
 grains. This increase is proportionally much less than that which 
 took place in the experiment on the process of quick tanning. 
 The colour of the pieces of leather was deeper than that of the 
 pieces which had been quickly tanned ; and, to judge from the 
 properties of the residual liquors, more of the extractive matters 
 of the barks had been combined with them. 
 
 The experiments of Mr. Biggin* have shown, that similar 
 barks, when taken from trees at different seasons, differ as to 
 the quantities of tannin they contain : and 1 have observed, that 
 the proportions of the astringent principles in barks, vary con- 
 siderably according as their age and size are different; besides, 
 these proportions are often influenced by accidental circum- 
 stances, so that it is extremely difficult to ascertain their distinct 
 relations to each other. 
 
 In every astringent bark, the interior white bark (that is, the 
 
 • Phil. Trans, for 1799, p. 299. 
 
on the constituent Parts of astringent Vegetables. 265 
 
 part next to the alburnum) contains the largest quantity of 
 tannin. The proportion of extractive matter is generally greatest 
 in the middle or coloured part : but the epidermis seldom fur- 
 nishes either tannin or extractive matter. 
 
 The 'vdiite cortical layers are comparatively most abundant in 
 young trees ; and hence their barks contain, in the same weight, 
 a larger proportion of tannin than the barks of old trees. In 
 barks of the same kind, but of different ages, which have been 
 cut at the same season, the similar parts contain always very 
 nearly the same quantities of astringent principles; and the 
 interior layers afford about equal portions of tannin. 
 
 An ounce of the white cortical layers of old oak bark, fur- 
 nished, by lixiviation and subsequent evaporation, 108 grains of 
 solid matter; and, of this, 72 grains were tannin. An equal 
 quantity of the white cortical layers of young oak produced 111 
 grains of solid matter, of which 77 were precipitated by gelatine. 
 
 An ounce of the interior part of the bark of the Spanish 
 chesnut, gave 8 g grains of solid matter, containing 63 grains of 
 tannin. 
 
 The same quantity of the same part of the bark of the Lei- 
 cester willow, produced 117 grains, of which 79 were tannin. 
 
 An ounce of the coloured or external cortical layers from the 
 oak, produced 43 grains of solid matter, of which 19 were tannin. 
 
 From the Spanish chesnut, 41 grains, of which 14 were 
 tannin. 
 
 And, from the Leicester willow, 34 grains, of which 16 were 
 tannin. 
 
 In attempting to ascertain the relative quantities of tannin in 
 the different entire barks, I selected those specimens which 
 appeared similar with regard to the proportions of the external 
 
 MDcccni. M m 
 
266' Mr. Davy’s Experiments and Observations 
 
 and internal layers, and which were about the average thickness 
 of the barks commonly used in tanning, namely, y an inch. 
 
 Of these barks, the oak produced, in the quantity of an ounce, 
 6i grains of matter dissolved by water, of which 29 grains were 
 tannin. 
 
 The Spanish chesnut 53 grains, of which 21 were tannin. 
 
 And the Leicester willow 71 grains, of which 33 were tannin. 
 
 The proportions of these quantities, in respect to the tanning 
 principle, are not very different from those estimated in Mr. 
 Biggin’s table.* 
 
 The residual substances obtained in the different experiments, 
 differed considerably in their properties ; but certain portions of 
 them were, in all instances, rendered insoluble during the process 
 of evaporation. The residuum of the chesnut bark, as in the 
 instance of the strongest infusion, possessed slightly acid pro- 
 perties ; but more than \ of its weight consisted of extractive 
 matter. All the residuums in solution, as in the other cases, 
 were precipitated by muriate of tin ; and, after this precipitation, 
 the clear fluids acted much more feebly than before on the salts 
 of iron; so that there is great reason for believing, that the 
 power of astringent infusions to precipitate the salts of iron 
 black, or dark coloured, depends partly upon the agency of 
 the extractive matters they contain, as well as upon that of the 
 tanning principle and gallic acid. 
 
 In pursuing the experiments upon the different astringent 
 infusions, I examined the infusions of the bark of the elm and 
 of the common willow. These infusions were acted on by re- 
 agents, in a manner exactly similar to the infusions of the other 
 barks : they were precipitated by the acids, by solutions of the 
 
 * Phil. Trans, for 1799, p. 263. 
 
on the constituent Parts of astringent Vegetables. 267 
 
 alkaline earths, and of the carbonated alkalis ; and they formed, 
 with the caustic alkalis, fluids not precipitable by gelatine. 
 
 An ounce of the bark of the elm, furnished 13 grains of 
 tannin. 
 
 The same quantity of the bark of the common willow, gave 
 11 grains. 
 
 The residual matter of the bark of the elm, contained a con- 
 siderable portion of mucilage; and that of the bark of the 
 willow, a small quantity of bitter principle. 
 
 The strongest infusions of the sumachs from Sicily and 
 Malaga, agree with the infusions of barks, in most of their pro- 
 perties ; but they differ from all the other astringent infusions 
 that have been mentioned, in one respect ; they give dense pre- 
 cipitates with the caustic alkalis. Mr. Proust has shown, that 
 sumach contains abundance of sulphate of lime ; and it is pro- 
 bably to this substance that the peculiar effect is owing. 
 
 From an ounce of Sicilian sumach, I obtained 165 grains of 
 matter soluble in water, and, of this matter, 78 grains were 
 tannin. 
 
 An ounce of Malaga sumach, produced 156 grains of soluble 
 matter, of which 79 appeared to be tannin. 
 
 The infusion of Myrobalans* from the East Indies, differed 
 from the other astringent infusions chiefly by this circumstance, 
 that it effervesced with the carbonated alkalis ; and it gave with 
 them a dense precipitate, that was almost immediately redis- 
 solved. After the tannin had been precipitated from it by 
 gelatine, it strongly reddened litmus-paper, and gave a bright 
 black with the solutions of iron. I expected to be able to procure 
 
 • The Myrobalans used in these experiments are the fruit of the Terminalia Che- 
 bula. Retz. Obs. Bolan. Fasc. V. p. 31. 
 
 Mm2 
 
268 Mr. Davy’s Experiments and Observations 
 
 gallic acid, by distillation, from the Myrobalans; but in this 
 I was mistaken ; they furnished only a pale yellow fluid, which 
 gave merely a slight olive tinge to solution of sulphate of 
 iron. 
 
 Skin was speedily tanned in the infusion of the Myrobalans ; 
 and the appearance of the leather was similar to the appearance 
 of that from galls. 
 
 The strongest infusions of the teas are very similar, in their 
 agencies upon chemical tests, to the infusions of catechu. 
 
 An ounce of Souchong tea, produced 48 grains of tannin. 
 
 The same quantity of green tea, gave 41 grains. 
 
 Dr. Maton has observed, that very little tannin is found in 
 cinchona, or in the other barks supposed to be possessed of 
 febrifuge properties. My experiments tend to confirm the ob- 
 servation. None of the infusions of the strongly bitter vegetable 
 substances that I have examined, give any precipitate to gelatine. 
 And the infusions of quassia, of gentian, of hops, and of cha- 
 momile, are scarcely affected by muriate of tin ; so that they 
 likewise contain very little extractive matter. 
 
 In all substances possessed of the astringent taste, there is 
 great reason to suspect the presence of tannin ; it even exists in 
 substances which contain sugar and vegetable acids. I have 
 found it in abundance in the juice of sloes; and my friend Mr. 
 Poole, of Stowey, has detected it in port wine. 
 
 V. GENERAL OBSERVATIONS, 
 
 Mr. Proust has supposed, in his Paper upon Tannin and 
 its Species,* that there exist different species of the tanning 
 principle, possessed of different properties, and different powers 
 
 * Annales dc Cb'mie, Tome XLI. p. 332. 
 
on the constituent Parts of astringent Vegetables. 26g 
 
 of acting upon reagents, but all precipitable by gelatine. This 
 opinion is sufficiently conformable to the facts generally known 
 concerning the nature of the substances which are produced in 
 organised matter; but it cannot be considered as proved, till the 
 tannin in different vegetables has been examined in its pure or 
 insulated state. In all the vegetable infusions which have been 
 subjected to experiment, it exists in a state of union with other 
 principles; and its properties must necessarily be modified by 
 the peculiar circumstances of its combination. 
 
 From the experiments that have been detailed it appears, that 
 the specific agencies of tannin in all the different astringent 
 infusions are die same. In every instance, it is capable of en- 
 tering into union with the acids, alkalis, and earths ; and of 
 forming insoluble compounds with gelatine, and with skin. The 
 infusions of the barks aftect the greater number of reagents in a 
 manner similar to the infusion of galls ; and, that this last fluid 
 is rendered green by the carbonated alkalis, evidently depends 
 upon the large proportion of gallic acid it contains. The infu- 
 sion of sumach owes its characteristic property, of being pre- 
 cipitated by the caustic alkalis, to the presence of sulphate of 
 lime ; and, that the solutions of catechu do not copiously pre- 
 cipitate the carbonated alkalis, appears to depend upon their 
 containing tannin in a peculiar state of union with extractive 
 matter, and uncombined with gallic acid or earthy salts. 
 
 In making some experiments upon the affinities of the tanning 
 principle, I found that all the earths were capable of attracting 
 it from the alkalis : and, so great is their tendency to combine 
 with it, that, by means of them, the compound of tannin and 
 gelatine may be decomposed without much difficulty ; for, 
 after pure magnesia had been boiled for a few hours with 
 
syo Mr. Davy’s Experiments and Observations 
 
 this substance diffused through water, it became of a red-brown 
 colour, and the fluid obtained by filtration produced a distinct 
 precipitate with solution of galls. The acids have less affinity for 
 tannin than for gelatine; and, in cases where compounds of the 
 acids and tannin are acted on by solution of gelatine, an equi- 
 librium of affinity is established, in consequence of which, by 
 far the greatest quantity of tannin is carried down in the inso- 
 luble combination. The different neutral salts have, compara- 
 tively, feeble powers of attraction for the tanning principle ; but, 
 that the precipitation they occasion in astringent solutions, is 
 not simply owing to the circumstance of their uniting to a por- 
 tion of the water which held the vegetable substances in so- 
 lution, is evident from many facts, besides those which have 
 been already stated. The solutions of alum, and of some 
 other salts which are less soluble in water than tannin, pro- 
 duce, in many astringent infusions, precipitates as copious as 
 the more soluble saline matters ; and sulphate of lime, and other 
 earthy neutral compounds, which are, comparatively speaking, 
 insoluble in water, speedily deprive them of their tanning prin- 
 ciple. 
 
 From the different facts that have been stated, it is evident 
 that tannin may exist in a state of combination in different sub- 
 stances, in which its presence cannot be made evident by means 
 of solution of gelatine; and, in this case, to detect its existence, 
 it is necessary to have recourse to the action of the diluted acids. 
 
 In considering the relations of the different facts that have 
 been detailed, to the processes of tanning and of leather-making, 
 it will appear sufficiently evident, that when skin is tanned in 
 astringent infusions that contain, as well as tannin, extractive 
 matters, portions of these matters enter, with the tannin, into 
 
on the constituent Parts of asiringetit Vegetables. ^71 
 
 chemical combination with the skin. In no case is there any 
 reason to believe that gallic acid is absorbed in tliis process ; 
 and M, Seguin’s ingenious theory of the agency of this sub- 
 stance, in producing the deoxygenatioii of skin, seems sup- 
 ported by no proof’s. Eveii in the formation of glue from 
 skin, there is no evidence which ought to induce us to suppose 
 that it loses a portion of oxygen ; and the effect appears to be 
 owing merely to the separation of the gelatine, from the small 
 quantity of albumen with which it was combined in the orga- 
 nised form, by the solvent powers of water. 
 
 The different qualities of leather made with the same kind of 
 skin, seem to depend very much upon tlie different quantities 
 of extractive matter it contains. The leather obtained by means 
 of infusion of galls, is generally found harder, and more liable 
 to crack, than the leather obtained from the infusions of barks ; 
 and, in all cases, it contains a much larger proportion of tannin, 
 and a smaller proportion of extractive matter. 
 
 When skin is very slowly tanned in weak solutions of the 
 barks, or of catechu, it combines with a considerable proportion 
 of extractive matter; and, in these cases, though the increase 
 of weight of the skin is comparatively small, yet it is rendered 
 perfectly insoluble in water ; and is found soft, and at the same 
 time strong. 
 
 The saturated astringent infusions of barks contain much less 
 extractive matter, in proportion to their tannin, than the weak 
 infusions; and, when skin is quickly tanned in them, common 
 experience shows that it produces leather less durable than the 
 leather slowly formed. 
 
 Besides, in the case of quick tanning by means of infusions 
 of barks, a quantity of vegetable extractive matter i$ lost to the 
 
27J2 Mr. Davy’s Experiments and Observations 
 
 manufacturer, which might have been made to enter into the 
 composition of his leather. These observations show, that there 
 is some foundation for the vulgar opinion of workmen, con- 
 cerning what is technically called the feeding of leather in the 
 slow method of tanning ; and, though the processes of the art 
 may in some cases be protracted for an unnecessary length of 
 time, yet, in general, they appear to have arrived, in consequence 
 of repeated practical experiments, at a degree of perfection 
 which cannot be very far extended by means of any elucidations 
 of theory that have as yet been made known. 
 
 On the first view it appears singular that, in those cases of 
 tanning where extractive matter forms a certain portion of the 
 leather, the increase of weight is less than when the skin is 
 combined with pure tannin ; but the fact is easily accounted for, 
 when we consider that the attraction of skin for tannin must be 
 probably weakened by its union with extractive matter; and, 
 whether we suppose that the tannin and extractive matter enter 
 together into combination with the matter of skin, or unite with 
 separate portions of it, still, in either case, the primary attraction 
 of tannin for skin must be, to a certain extent, diminished. 
 
 In examining astringent vegetables in relation to their powers 
 of tanning skin, it is necessary to take into account, not only 
 the quantity they contain of the substance precipitable by gela- 
 tine, but likewise the quantity, and the nature, of the extractive 
 matter; and, in cases of comparison, it is essential to employ 
 infusions of the same degree of concentration. 
 
 It is evident, from the experiments detailed in the Illd section, 
 that of all the astringent substances which have been as yet 
 examined, catechu is that which contains the largest propor- 
 tion of tannin ; and, in supposing, according to the common 
 
on the constituent Parts of astringent Vegetables. 273 
 
 estimation, that from four to five pounds of common oak bark 
 are required to produce one pound of leather, it appears, from 
 the various synthetical experiments, that about half a pound of 
 catechu would answer the same purpose.* 
 
 Also, allowing for the difference in the composition of the 
 different kinds of leather, it appears, from the general detail of 
 facts, that one pound of catechu, for the common uses of the 
 tanner, would be nearly equal in value to 2^ pounds of galls, 
 to 7^ pounds of the bark of the Leicester willow, to 11 pounds 
 of the bark of the Spanish chesnut, to 18 pounds of the bark of 
 the elm, to 21 pounds of the bark of the common willow, and 
 to 3 pounds of sumach. 
 
 Various menstruums have been proposed for the purpose of 
 expediting and improving the process of tanning, and, amongst 
 them, lime-water and the solutions of pearl-ash : but, as these 
 two substances form compounds with tannin which are not de- 
 composable by gelatine, it follows that their effects must be 
 highly pernicious ; and there is very little reason to suppose, that 
 any bodies will be found which, at the same time that they 
 increase the solubility of tannin in water, will not likewise di- 
 minish its attraction for skin. 
 
 • This estimation agrees very well with the experiments lately made by Mr. Purkis, 
 upon the tanning powers of Bombay catechu in the processes of manufacture, and which 
 he has permitted me to mention. Mr. Purkis found, by the results of different 
 accurate experiments, that one pound of catechu was equivalent to seven or eight of 
 oak bark. 
 
 MDCCCin. 
 
 Nn 
 
L 274 ] 
 
 X, Appendix to Mr. William Henry’s Paper, on the Qiianliiy of 
 
 Gases absorbed by Water, at differeiit Temperatures, and under 
 
 different Pressures. (See Page 29). 
 
 Since my Paper was printed, I have found that the numbers 
 assigned in it, as indicating the quantities taken up by water, 
 of some of the more absorbable, and of all the less absorb- 
 able gases, are rather below the truth. The accuracy of these 
 numbers I was led to doubt, by a suspicion that due atten- 
 tion had not always been paid, in my former experiments, to 
 the quality of the unabsorbed residuum. For, the theory which 
 Mr. Dalton has suggested to me on this subject, and which 
 appears to be confirmed by my experiments, is, that the absorp- 
 tion of gases by water is purely a mechanical effect, and that its 
 amount is exactly proportional to the density of the gas, consi- 
 dered abstractedly from any other gas with which it may acci- 
 deptally be mixed. Conformably to this theory, if the residuary 
 gas contain or any other proportion, of foreign gas, the 
 
 quantity absorbed by water will be &c. short of the 
 
 maximum. The proof of these propositions would lead me into 
 a minuteness of detail, not suited to the present occasion ; I 
 therefore hasten to communicate the results of my latest expe- 
 riments. 
 
 The report which I have already given, of the quantity of 
 Carbonic Acid Gas, absorbed under the ordinary pressure 
 of the atmosphere, I find no reason to correct; but, of Sul- /-' 
 PHURETTED Hydrogen Gas, I liave effected a larger absorption