place, as is generally recommended. The final effect of aluming, in whatever manner performed, and whatever chemical changes may have taken place in it, consists in the combination of alumine with the stuff: this union has probably been imperfect, and the acids only partially separated, but becomes complete when the cloth has been boiled with madder, as in the case of printed stuffs. But an acid or an alkali may form a supercompound with the stuff, the coloring matter, and the alumine; for there are some colors which are changed by an acid, and restored by alkalis, or by calcareous earths, which take the acid from them, or vice versa; but this supercomposition does not take place with respect to those colors which are esteemed durable, being unchangeable by alkalis or acids, which are not strong enough to destroy their composition.

25. The attraction of alumine for animal substances is not, however, merely indicated by uncertain appearances, nor supposed for the purpose of being employed in explanations, but is proved by direct experiment. M. Berthollet united them together, by mixing an anima! substance with a solution of alum; a double exchange took place, the alkali entered into combination with the acid of the alum, and the alumine, combining with the animal substance, was precipitated. He also proved the affinity of alumine for animal substances by another experiment: having mixed a solution of glue with a solution of alum, he precipitated the alumine by an alkali, and the glue with which it had combined fell down along with it. This compound has the appearance of a semitransparent jelly, and dries with difficulty. Thus, in the preceding experiments, the alkali precipitated the alumine combined with the animal substance, from the uncrystallisable residue of the alum which had been boiled with the wool.

26. The affinity of alumine for most coloring substances, may also be shown by direct experiment. If a solution of a coloring substance be mixed with a solution of alum, a precipitation sometimes takes place; but if to the liquor we add an alkali, which decomposes the alum, and separates the alumine, the coloring particles are then precipitated, combined with the alumine, and the liquor remains clear: this compound has obtained the name of lake. In this experiment, too much alkali must not be added, because alkalis are capable of dissolving lakes in general. No direct experiment has however yet shown, that alumine attracts any vegetable substance except the coloring particles: its affinity for them seems much weaker than that which it has for animal substances; hence the acetite of alumine is a better basis for cotton and linen than alum is, and upon this depend the different means employed to increase the fixity of the coloring particles of madder in the dyeing of these sub

stances.

27. Metallic oxides have so great an affinity for many coloring substances, that they quit the acids in which they were dissolved, and are precipitated in combination with them. On the other hand, all metallic oxides have the property of uniting with animal substances; and these different compounds may be formed by mixing an alkali, saturated with an animal substance,

with metallic solutions. It is not surprising, therefore, that metallic oxides should serve as a bond of union between the coloring particles and animal substances; but, besides the attraction of the oxides for the coloring particles, and for animal substances, their solutions in acids possess qualities which render them 10ore or less fit to act as mordants: thus, those oxides which easily part with their acids, such as that of tin, are capable of combining with animal substances, without the aid of coloring particles; it is sufficient to impregnate the wool or silk with a solution of tin, although they be afterwards carefully washed, which is not the case with other metallic solutions. Some metallic substances afford, in combination, only a white and colorless basis; and some by the admixture of their own color, modify that which is proper to the coloring particles; but in many metallic oxides, the color varies according to the proportion of oxygen they contain, and the proportion of this is easily liable to change, Upon these circumstances their properties in dyeing chiefly depend.

28. The affinity of metallic oxides for substances of vegetable origin, seems much weaker than that which they have for animal substances: metallic solutions are, therefore, not well adapted to serve as mordants for colors in cotton or linen, except iron, the oxide of which unites firmly with vegetable substances, as is shown by ironmoulds, which are owing to a real combination of this oxide. Whenever the coloring particles have precipitated a metallic oxide from its menstruum, the supernatant liquor contains the disengaged acid, which is commonly capable of dissolving a portion of the compound of the coloring substance and oxide, so that the liquor remains colored; but sometimes the whole of the coloring particles aré precipitated, when the proportions have been accurately adjusted: this precipitation is facilitated, and rendered more complete, by the presence of the stuff, which assists, by the tendency it has to unite with the compound of oxide and coloring particles. Uncombined metallic oxides have also a very evident action on many coloring substances when boiled with them, and modify their color; the oxide of tin in particular increases the brightness and fixity of many.

29. The compounds of oxides and coloring substances are similar to many other chemical compounds, which are insoluble, when the principles of which they are formed are properly proportioned; but which are capable of being supersaturated by an excess of one of the principles, and thence of becoming soluble. Thus a metallic oxide, united with a coloring substance to excess, produces a liquor, the color of which will be modified by the oxides; whereas, when the coloring matter is not in excess, the compound will be insoluble, or nearly so; these effects are very evident in the combination of iron with the astringent principle. Neutral salts such as nitre, and particularly muriate of soda, or common salt, act as mordants, and modify colors; but it is difficult to ascertain the manner in which they act. M. Berthollet found that the muriate of soda was contained, in substance, in the precipitates produced by some species

coloring particles, and that these precipitates retained a considerable degree of solubility; it would seem that a small part of the salt becomes fixed with the coloring particles and the stuff. Salts with calcareous bases also modify colors; but, as these modifications are nearly similar to those which would be produced by the addition of a small quantity of lime, it is probable that they are decomposed, and that a little of the lime enters into combinaation with the coloring particles and the stuff. By attention to this, we shall easily discern what combinations are formed by the agency of the different reactives, employed in the analysis of coloring substances; but we must not forget, that the mordants and the coloring particles have a mutual action on each other, which may change their properties. It is evident that, by varying the mordants, we may greatly multiply the shades obtained from a coloring substance; even to vary their mode of application may be sufficient: thus we shall obtain different effects by impregnating the stuff with the mordant, or by mixing the mordant with the bath; by applying heat, or using exsiccations, for we operate upon three elective attractions; that of the coloring particles, that of the stuffs, and that of the principle of the mordant; and many circumstances may cause variations in the result of these attractions; circumstances which merit further explanation. Exsiccation, or drying, favors the union of the substances which have an affinity for the stuff, and the decompositions which may result from that union; because the water which held these substances in solution, by its attraction, opposed the action of the stuff; but the exsiccation should be slow, in order that the substances may not be separated before their mutual attractions have produced their effect.

30. Considerable differences must be observed in the manner of employing the mordant, as the force of affinity between the stuff and the coloring matter is greater or less. When this affinity is strong, the mordant and the coloring substance may be mixed together; the compound thus formed, immediately enters into combination with the stuff. But, when the affinity between the stuff and the coloring particles is weak, the compound formed of the latter and the mordant may separate, and a precipitation take place, before it can be attached to the stuff; and hence it is, that the mordant which is to serve as the medium of union between the stuff and the coloring matter, must be combined with the former, before the application of the latter. It is from these differences that different processes must be followed in fixing coloring matters on animal and vegetable productions.

31. In judging of the effects of mordants, and the most advantageous manner of applying them, it is necessary to attend to the combinations which may be formed, either by the action of the ingredients of which they are composed, or by that of the coloring matter and the stuff. It is necessary, also, to take into consideration the circumstances which may tend to bring about these combinations with more or less rapidity, or that may render them more or less perfect. The action which the liquor in which the stuff is immersed

may have, either on its color or texture, must also be considered; and to be able accurately to judge of the extent of this action, we must know the proportions of the principles of which the mordant is composed; which of these principles remains in an uncombined state in the liquor, and the proportion or quantity which is separated.

32. The coloring particles have been hitherto considered only as substances capable of forming different combinations, by which their properties are modified; but they may be altered in their composition, either by other external agents, or by the substances with which they unite. The stability of a color consists in its power of resisting the action of vegetable acids, alkalis, soap, and more especially that of the air and light; but this power varies exceedingly, according to the nature of the color and the species of the stuff; for the same durability is not required in the colors of silk as in those of wool. There is not much obscurity in the action of water, acids, alkalis, or soap: it is a solution brought about by these agents: and it appears that a small quantity of acid, or of alkali, sometimes unites with the compound which gives the color; because the color is not destroyed, but only changed, and may be restored by taking away this acid; for instance, by chalk and ammoniac, or volatile alkali. But this is not the case with respect to the action of air and light.

33. Scheele observed, that the oxygenated muriatic acid rendered vegetable colors yellow, and he attributed that effect to the property it had of taking up the phlogiston which entered into their composition. Berthollet has shown, that the properties of the oxygenated muriatic acid were owing to the combination of its oxygen with the substances exposed to its action; that it commonly rendered the coloring particles yellow; but that, by a continuance of its action, it destroyed their color; without determining in what this action consisted. Fourcroy afterwards made several observations on the action of oxygen on the coloring particles, which throw a great deal of light on the nature of the changes they undergo, chiefly when watery solutions of them are left exposed to the air, or have been subjected to a boiling heat. He observed that, in consequence of the action of the air, vegetable decoctions formed pellicles, which lost their solubility, and underwent successive changes of color; he marked the gradations of color thus produced, and concluded, from his observations, that oxygen entered into the composition of the coloring particles; that when it combined with them, their shade was changed; that the more they received, the more fixed did their color become; and that the best method of obtaining permanent unchangeable colors, for painting, was to choose such as had been exposed to the action of the oxygenated muriatic acid.

34. In considering the effects of air on colors, it is necessary to make a distinction between those produced by metallic oxides, and those produced by the coloring particles. Berthollet is of opinion that the modifications of the former are entirely owing to different proportions of oxygen, but from observation he has been led to

form a different opinion respecting the modifications of the latter. He observed, that the oxygenated muriatic acid exhibited different phenomena with the coloring particles; that sometimes it discharged their colors, and rendered them white; that most frequently it changed them to a yellow, fawn, or root-colored, brown, or black, according to the intensity of its action; and that, when their color appeared only discharged or rendered white, heat, or a length of time, was capable of rendering them yellow. He compared the effect produced by the oxygenated muriatic acid, when the particles are rendered yellow, fawn-colored, or brown, with the effect of a slight degree of combustion, and showed that they were the same; that they were owing to the destruction of the hydrogen, which, combining with the oxygen, more easily, and at a lower temperature than charcoal does, leaves it predominant, so that the natural color of charcoal is more or less blended with that which before existed. This effect becomes very evident, when sugar, indigo, or the infusion of the gall-nut, or of sumach, are exposed to the action of oxygenated muriatic gas; the sugar and the indigo assume a deep color, and afford indisputable marks of a slight combustion; the infusion of the gall-nut, and that of sumach, let fall a precipitate, which is not far from being pure charcoal or carbon. These appearances are analogous to those which are observed in the distillation of organised substances; in proportion as the hydrogen is extracted in the form of oil, or of gas, the substance grows yellow and at length there remains only a black coal. If the hydrogen be expelled from an oil, by heat, it grows brown, evidently in the same way. 35. Berthollet also found, by other experiments made on alcohol and ether, that the oxygen united to the marine acid, had the property of combining with the hydrogen, which abounds in these substances, and of thereby forming water. He therefore supposes, that when the oxygenated marine acid renders a color yellow, fawn-colored, or brown, the effect proceeds from the coloring matter having undergone a slight combustion, by which more or less of its hydrogen has been converted into water; and that the charcoal, thus rendered predominant, has communicated its own color. The art of bleaching linen by means of the oxygen of the atmosphere, of the dew, and of the oxygenated marine acid, he also supposes to depend on this change of the coloring matter. The coloring particles of the flax are rendered soluble in the alkaline lixivia, the action of which ought to be alternate with that of the oxygen. These coloring particles may be afterwards precipitated from the alkali, and by evaporation and drying become black, and prove the truth of this theory, both by the color they have acquired, and by the quantity of charcoal which they yield on being analysed. But the alkaline solution of the coloring matter of linen which is of a dark brown color, loses its color almost entirely, by the addition of a certain quantity of oxygenated muriatic acid; and the same effect is observable in many other substances, which have assumed a color originating from a commencement of combustion. A piece of linen, which appears white, may grow yellow in process of time, particularly

if exposed to a certain degree of heat, if the oxygenated parts have not been removed by a sufficiently strong lixivium. In the same manner, the green parts of vegetables are rendered white by the oxygenated muriatic acid, but beome yellow when boiled.

36. From these facts it appears, that oxygen is capable of whitening, or rendering paler. the coloring matters with which it unites, perhaps by having produced the effects of a slight combustion upon them; or possibly these effects take place only afterwards in a gradual manner, but more rapidly, when the whole is exposed to a certain degree of heat. It is extremely probable, that in all cases a part of the oxygen unites with the coloring matter, without being combined with the hydrogen in particular, and that it is in this way that oxygen acts, in rendering the coloring matter of flax more easily soluble in alkalis. In many other cases oxygen has evidently an influence on the changes which take place in the coloring particles of vegetables; these particles are formed chiefly in the leaves, flowers, and inner bark of trees; by degrees they undergo a slight combustion, either from the action of the atmospheric air which surrounds them, or from that of the air which is carried by a particular set of vessels into the internal parts of vegetables.

37. Berthollet, therefore, supposes we may explain how the air acts upon coloring matters, of an animal, or a vegetable nature; it first combines with them, renders them weaker and paler, and by degrees occasions a slight combustion, by means of which the hydrogen which entered into their composition is destroyed; they change to a yellow, red, or fawn-color; their attraction for the stuff seems to diminish; they separate from it, and are carried off by water: all these effects vary, and take place more or less readily, and more or less completely, according to the nature of the coloring particles; or rather, from the nature of the properties which they possess, in the state of combination into which they have gone. The changes which occur in the colors, produced by the union of the coloring particles with metallic oxides, are effects compounded of the change which takes place in the coloring particles, and of that which is undergone by the metallic oxide.

38. The light of the sun considerably accelerates the extinction of colors. It ought, therefore, if this theory be well founded, to favor the combination of oxygen, and the combustion thereby induced. Sennebier, who has given many interesting observations on the effects of light on different substances, and particularly on their colors, attributes these effects to a direct combination of light with the substances. And the effects of light on the color of wood, have long ago been noticed; it preserves its natural appearance while kept in the dark, but when exposed to the light, it becomes yellow, brown, or of other shades. The same writer also remarked the varieties which occur in this particu lar in different kinds of wood, and found, that the changes are proportioned to the brightness of the light, and that they take place even under water, but that wetted wood underwent these changes less quickly than that which was dry;

fact, that colors become brighter by their union with a small portion of oxygen. It is on this account found necessary to air stuffs when they come out of the bath, and sometimes even to take them out of it from time to time, expressly for this purpose; but the quantity of oxygen which, thus becoming fixed, contributes to the brightness of the color, is very considerable in some cases and the deterioration of shade soon begins. But the action of the air affects not only the coloring matter and the stuff, but also metallic oxides, when they are employed as intermedia; because the oxides, which have at first been deprived of a part of their oxygen by the coloring particles, may absorb it again. Those then, the color of which varies according to their proportion of oxygen, have thereby an influence in effecting the changes which the stuff undergoes. It is undoubtedly to this cause that the change observable in the blue given to wool, by sulphate of copper, or blue vitriol, and logwood, is to be attributed. This blue soon becomes green by the action of the air: now copper, which has a blue color, when combined with a small proportion of oxygen, assumes a green one by its union with a larger quantity. The change which the coloring particles undergo, may indeed contribute to this effect; but the coloring particles of the logwood, which have themselves a dark color, should rather become brown by combustion, than grow yellow, which would be necessary in order to produce a green with the blue. It has been observed, that coloring particles in a state of combination were less disposed to be changed by the action of the air, than in an uncombined state. This is generally the case, but there are some exceptions; an alkali, for instance, produces a contrary effect. A matrass half filled with an infusion of cochineal, was exposed to the light, over mercury; a similar matrass contained an infusion of cochineal made with a little tartar; and in a third, a small quantity of alkali had been added to the infusion. The second matrass appeared least altered in the same space of time, and in it the absorption had been least considerable. In the third, the color of the liquor became first brown, and was then discharged; and the absorption of air, though inconsiderable, was greater than in the two others. On evaporation it assumed a brown color; and left a residuum of a yellowish brown. 44. Similar experiments having been made on different coloring substances, the alkali was found to darken their color, which grew more and more brown, and promoted the absorption of air. Madder appeared to be the only exception to this rule its color, which became darker at first, stood better than that of the infusion made without alkali. The general effect of alkalis on the coloring particles is consonant to that which it produces on many other substances, such as sulphur; it favors the absorption of air, because it has a strong affinity for the substance which is the result of that absorption. From this effect of alkalis, a fact which has been observed by Becker may be explained; viz. that a vegetable infusion, rendered green by an alkali, becomes gradually yellow, if left exposed to the air, and that when the yellow is completely formed, acids cannot restore the original color: but that this is not the

case, when a vegetable color, reddened by an acid, has been kept in like manner for some time. Those instances in which acids have been employed, which act by giving off their oxygen, must be excepted, for in these there is an extraction of the color.

45. From the above remarks on mordants it must appear very obvious that the practical dyer ought to be exceedingly careful in his selection of substances, giving the preference to those that most readily resist the action of the causes which we have specified.

46. It may not be improper to notice the action of these acids on animal substances, in consequence of its intimate counexion with the subject of mordants. It was observed by M. Brunwiser, that wood, on being exposed to the action of the air, assumed different colors: this led him to endeavour to ascertain whence those colors arose, and to produce them by artificial means. He remarked that on moistening the surface of wood, particularly young wood, with nitric acid, it assumed a yellow color; and that, by applying in the same way the muriatic and sulphuric acids, the wood assumed a violet color. Hence he inferred that, as all colors are produced by a mixture of yellow, blue, and red, all those colors which are seen in the leaves, fruits, and flowers of trees, are owing to the coloring particles which exist in the wood, and are there kept in a state of disguise, by the action of an alkali; that the mineral acids, by taking up this alkali, set the coloring particles at liberty; and that the fixed air, by penetrating the leaves, fruits, and flowers, produces naturally the same effect, by combining with the alkali which kept them disguised.

47. M. de la Folie informs us that having immersed a skein of white silk in nitrous acid of the strength generally used in commerce, the silk in three or four minutes assumed a fine jonquille yellow. He washed it several times in water, that it might not be affected by any adhering acid; the color sustained several trials to which he submitted it, and the silk preserved its lustre unimpaired. When dipped into an alkaline solution, a fine orange color was the result. Dr. Gmelin observes, that he has given a fine brimstone color to silk, by keeping it for a day in cold nitric acid, or some hours only, when the acid was warm. Boiling with soap and water diminished the brightness of this color; and it was changed to a fine lemon color, by being kept for twelve hours in an alkaline solution; but, when the solution was employed hot, a fine gold color was produced. The different solutions of metals in nitric acid communicated a more or less deep yellow to silk, as did also the solution of alumine in the same acid; but those of the calcareous earth and magnesia had no effect whatever.

48. M. Berthollet also found, that the oxyge nated muriatic acid has the property of tinging animal substances yellow; but that it does not give them so deep a color as the nitrous acid, and it weakens them much more than that acid when properly diluted; so that the nitrous acid is far preferable for the different purposes of art. It, therefore, appears that the nitrous acid, diluted with a certain quantity of water, gives silk