China: Its History, Arts, and Literature/Volume 1/Chapter 14
Chapter XIV
MANUFACTURING PROCESSES
THE questions of the preparation of pâte and glazing material by Chinese potters, of the application of the glaze and of the mode of selecting and employing colouring matter, though investigated from time to time by Europeans residing in China, are still more or less obscure. It would be useless, therefore, to attempt to enter into the minutiæ of these matters. Brief reference may be made, however, to processes of which a general knowledge has been acquired; namely, those followed at the Ching-tê-chên factories in the times of their great prosperity; that is to say, during the reigns of Kang-hsi, Yung-ching, and Chieng-lung, the period comprised between 1661 and 1795. Whether these processes differed from those in vogue during the Ming dynasty, and if so, in what the difference consisted, there are unfortunately no means of determining. But it may reasonably be assumed that the differences were not considerable, for certainly in many directions the achievements of the Ming potters were at least equal to those of their successors in the Tsing dynasty. Information with respect to these questions of manufacture is chiefly derivable from the Annals of Fu-liang, as translated by M. Julien, from the letters of Père d'Entrecolles, from the annotations of M. Salvétat, and from minor sources.
The material employed in the manufacture of Chinese porcelain is composed of two ingredients, petuntse and kaolin. Of these the former is fusible and gives to the ware that transparency which characterises true porcelain; the other is infusible, and its presence enables the mass to support the temperature necessary to transform the fusible element into glass. Both are found in the mountains of a district distant about sixty miles from Ching-tê-chên, whither they are brought by boat, having first undergone preliminary manipulation and been reduced to the form of bricks. M. Salvétat's analysis shows that these materials practically correspond in all their constituents with those drawn from the mines at Saint-Yrieix and used at the Sèvres factories. At Ching-tê-chên the bricks undergo further treatment, which need not be described here since it presents no novel or noteworthy features. The kaolin and the petuntse are then mixed—in equal portions for the finest porcelain; in the ratio of four parts of kaolin to six of petuntse for middle-class ware; and in the ratio of one part of kaolin to three of petuntse for inferior ware. M. Salvétat has analysed the masses of four qualities of Chinese porcelain. His results, side by side with corresponding figures for the Sèvres pâte, are as follows:—
| Chinese Porcelains. | |||||
| Sèvres Porce- lain. | |||||
| First Quality. |
Second Quality. |
Third Quality. |
Fourth Quality. | ||
| Silica | 69.0 | 70.0 | 73.3 | 69.0 | 58.0 |
| Alumina | 23.6 | 22.2 | 19.3 | 21.3 | 34.5 |
| Iron Oxide | 1.2 | 1.3 | 2.0 | 3.4 | 34.5 |
| Lime | 0.3 | 0.8 | 0.6 | 1.1 | 4.5 |
| Magnesia | 0.2 | trace | 0.6 | 1.1 | 4.5 |
| Manganese Oxide, | 0.1 | trace | 0.6 | 1.1 | 4.5 |
| Potash | 3.3 | 3.6 | 2.5 | 3.4 | 3.0 |
| Soda | 2.9 | 2.7 | 2.3 | 1.8 | 3.0 |
It is well known that the nature of porcelain biscuit depends largely on the amount of iron oxide it contains, the impure colour of bad ware being chiefly due to the presence of that mineral. It was to kaolin in its mass that the porcelain of China owed its beauty—its whiteness, its transparency, and its timbre. Europe knew nothing of kaolin until the eighteenth century. Western keramists could produce only the well known pâte tendre, fusible at a temperature of 700° less than that supported by the pâte dure of China.
Full acquaintance with the composition of Chinese porcelain pâte and ability to reproduce its qualities long ago put an end to the interest excited by its first appearance among European experts. But the same cannot be said of Chinese glazes. Among these there are some that still defy imitation and remain more or less a mystery. In fact, much of the beauty of old Chinese wares is due to the texture and tone of their glaze; its peculiar lustre, solidity, and velvety softness constituting features as attractive as they are incomparable. The notes of M. d'Entrecolles, with reference to the subject of glaze and its preparation, are worth translating in full. "In addition," he writes, "to boats laden with petuntse and kaolin which line the banks at Ching-tê-chên, one sees others filled with a white liquid substance. I had long known that this was the glazing matter which gave the porcelain its whiteness and lustre, but I was ignorant of its composition which I have now at length learned. This glazing material is obtained from the hardest rock. Although the rock from which petuntse is obtained is also employed in making the glazing matter, only the parts that are whitest and show the most distinctly green marks are chosen. These green marks, or spots, are caused by the presence of oxide of manganese. The stone has to be well washed, in the first place, after which it is subjected to processes similar to those employed with the ordinary petuntse. When, by these processes, the purest portions have been separated, there is added to every hundred parts by weight of the liquid one part of fibrous gypsum, which has been previously brought to a red heat.’’ M. Salvétat considers that the rdle played by this gyp- sum is purely mechanical. It facilitates the precipita tion of minute impurities held in suspension in the liquid. ‘‘The glazing matter, thus prepared, is never used alone. There is added to it another substance which may be called its soul, and which is prepared as follows: — A mass of lime is taken and sprinkled with water to reduce it to powder. A bed of fern is then made and on this is placed a bed of the slacked lime. Over this is placed another bed of fern and then another of lime and so on. The whole mass is subsequently roasted, and when it is entirely calcined, the ashes are spread upon another bed of fern, and the process of piling and burning is gone though de novo. This is repeated six or seven times, and even oftener for very choice glazes. According to the history of Fu-liang, the wood of the Diospyrus Kaki was formerly used instead of ferns, but this practice has been abandoned owing to the scarcity of the tree. Perhaps the change of process is responsible for the deterioration in the quality of modern porcelain.” M. d’Entrecolles, it must be remembered, wrote these notes in 1720-25; that is to say, during the closing years of the reign of Kang-hsi, an epoch generally regarded as particularly prolific in fine wares. His reference to the deterioration of porcelain must be taken in the sense that many of the grand monochromes of the Ming dynasty were, in his day, esteemed above anything subsequently produced, though with such an estimate connoisseurs of the present age are not likely to agree. "The quality of the lime and the nature of the ferns," he continues, "contribute appreciably to the excellence of the glazing matter, and I have remarked that the lime and ferns of certain localities are much more highly esteemed than those of others. The workmen, having obtained a sufficiency of these ashes of lime and ferns, throw them into an urn filled with water. Then to every hundred parts by weight there is added one part of fibrous gypsum. The mixture is well shaken and left to settle, until there floats on the surface a scum which is removed and placed in a second urn. This purifying process is repeated several times, until a species of paste is formed at the bottom of the second urn. The water is then carefully drained off and the residue is the matter which is mixed with the glazing liquid obtained from petuntse as described above. It is essential that the mixture should be of uniform consistency, and to determine this, pieces of petuntse are plunged into it. By examining the surface of these when they are withdrawn, it is possible to estimate the intimacy in which the components of the glaze are mingled. With regard to the quantities in which the lime and fern compound is added to the petuntse liquid, ten parts by weight of the latter go to one part of the former."
This account shows plainly that the quality of the glazing material depended largely on the expertness and care of those that prepared it. The industry appears to have been a specialty at one time, and it is recorded that frauds were often practised by the manufacturers, who increased the bulk of the liquid by adding water and fibrous gypsum, the result of course being an evolution of sulphuric acid in the furnace and consequent imperfections in the surface of the porcelain. An idea of the pains lavished on the manufacture may be gathered from the "History of Ching-tê-chên Keramics," where it is stated that the potters of the Hung-woo era (1368-1398) used to dry the unbaked vases during a year, at the end of which time they were replaced on the wheel, thinned down and covered with the glazing material. Then, after having been again thoroughly dried, they were put in the oven. When they emerged, if any faults were found in the glaze, it was removed on the wheel by means of a tool, and the piece, having been re-glazed, was again fired. "By these means," writes the author of the history, "a glaze lustrous and rich as congealed fat was obtained." It is, indeed, the perfection of their glazes which places the porcelains of China at the head of all the keramic wares in the world.
To what particular methods of manufacture is this excellence of glaze due? The question has naturally received attention in Europe, especially at the hands of the well-known chemist, M. Salvétat, of the Imperial Factory at Sèvres. "The glaze," he writes, "of European porcelain is generally composed—the fact is certain in the case of Sèvres ware—of pure pegmatite, finely ground and attached by immersion to the ware while in the condition of 'biscuit.' In Germany certain substances are mixed with the felspar to modify its fusibility: often kaolin is added to produce greater resistance to the action of heat. But at Sèvres, for a long time back, the pegmatite of Saint Yrieix is employed alone. If, in the factory's early days, an artificial glaze was used, it was because the potters were ignorant of the advantages to be derived by employing the volcanic rocks of Haute-Vienne. About the year 1780, the glaze was composed of a mixture of twenty parts of sand from Fontainebleau, twenty-four parts of porcelain stone, and six parts of chalk. This composition, tolerably difficult to melt, had disadvantages that led to its abandonment. At any rate, it is with the processes now actually in use that I compare those of China, and I believe myself perfectly secure in the conclusions I record here. If the addition of lime or of other matters to modify the fusibility of the vitreous substances employed in glazing porcelain, is exceptional in Europe, in China, on the contrary, the use of pure petrosilex without any admixture is confined to special cases. The porcelains of China and Japan are generally covered with composite glazes, obtained by mixing various materials in proportions determined by the nature of the ware. The substance which the Chinese add to petrosilex to render the latter more fusible, is lime. This is clearly shown both by the translation of Chinese books, and by analyses which have been made either of glazes taken from finished porcelains, or of specimens sent by Père Ly, marked 'lime' by him, and specified as part or the glazing material. I must repeat that it is the lime alone which seems, in my opinion, to play an active rôle in the calcined mixture of fern-leaves and lime, having regard to the small quantity of fern ashes that enter the mixture. For the rest, these ashes contain only silica and insignificant quantities of phosphoric acid. I do not think that any influence should be ascribed to the soluble salts obtained from them; moreover certain passages that treat of glazing material seem to show that it is sometimes composed by mixing chalky earth to the felspathic quartz which forms the basis of the glazing matter. It is also explained that the mixture of caustic lime and ashes is finely ground and washed before being mixed with the petrosilex, the carbonised film which the carbonic acid of the atmosphere forms on the surface of the liquid being carefully removed to be mixed with the petrosilex. This practice has no other object than the production of a perfectly pure lime. We are apparently given to understand that it is the film of regenerated lime which constitutes the useful element, and that the fern ashes which sink to the bottom of the vessel in which the washing is effected, are thrown away as of no value. Whatever be the true action of the fern ashes, whatever be the real consequence of calcining the lime and the ferns, judging only by the rough figures which analyses give, it is evident that the presence of the lime, which enters in very minute quantities only into the glaze of European porcelain, but which, on the contrary, is found in considerable proportions in the glaze of Chinese porcelain—sometimes as much as twenty-five per cent by weight—establishes a very salient difference between the two productions."
The conclusion of this eminent expert as to the role played by lime does not seem to cover the whole ground. He truly notes that the Japanese also—who indeed acquired the art of porcelain manufacture from China—added lime to their glazing material. But he fails to note that the glazes of Japan could never bear comparison with those of China in lustre, depth, and solidity. The one point in which the manufacturing processes of China differed emphatically from those of both Europe and Japan was that the Chinese potters applied the glazing material to the unstoved piece, whereas the potters of Japan and Europe alike applied it to ware which had received a preliminary firing; that is, to ware in the state of biscuit. Does it not seem reasonable to ascribe the marked difference of the results in the two cases, in part at least, to this salient difference of process? An example which supports this idea may be found in European keramics. It is French majolica, the manufacture of which was commenced at Nevers by Antoine Conrade, in 1644, and continued with much success at Rouen and Moustiers. This was faience decorated au grand feu, the decorator using metallic oxides which, under the influence of heat, became incorporated with the glaze. The makers of the ware never quite overcame the difficulties of the process. Their productions are not remarkable for technical perfection. But it has never been denied that the glaze of this French majolica possessed qualities of solidity and softness, lustre and depth, which no other manufacturing process gave. M. Salvétat and other writers have not hesitated to conclude that because the Chinese potters did not avail themselves of the technical facilities offered by porcelaine dégourdie as a recipient of glazing material, they were ignorant of those advantages. Such an inference sounds harsh and unwarranted. The Chinese of the Kang-hsi era, at all events, fully understood the secret of producing couleurs de demi-grand feu: in other words, they knew how to apply enamels to porcelain lain already fired, and how to vitrify and fix them by a second firing at reduced temperatures. Yet while possessing this knowledge, the Chinese expert, in the great majority of cases, did not depart from his old-fashioned method of applying the glazing material to the unstoved piece and subjecting both pâte and glaze to the same degree of temperature. Is it conceivable that the great technical inconvenience of the latter process would have been wantonly endured by a keramist skilled also in the former, unless some compensatory advantage were obtained? The more reasonable supposition is, that, though the absorbent properties of porcelaine dégourdie and the ease of glazing it were well known at Ching-tê-chên, the Chinese keramist deliberately chose the incomparably more troublesome and unsafe plan of applying his glazes to the unstoved pâte, because by no other process could he obtain the lustre, depth, and softness so highly prized by his country's connoisseurs. Japan, China's pupil and confessedly her inferior in the technicalities of porcelain manufacture, always used the su-yaki-gama, or kiln for stoving porcelain before glazing. Her potters, however, adopted this simpler process, not wholly from choice, but because the materials immediately available to them for manufacturing the porcelain mass were too refractory to permit varieties of composition such as those habitually resorted to in China. In this branch of his art the Ching-tê-chên expert was deeply versed. Some of his most delicate and beautiful monochromatic glazes are found, not on true porcelain, but on fine stone-ware. The Japanese, on the other hand, had to content themselves with a lower range of technical excellence, and they naturally eschewed a process which, in their case, offered no adequate compensation for its great difficulties. That they thoroughly appreciated and would gladly have emulated the brilliant velvet-like glazes of China, there can be no doubt. But they never succeeded in producing anything of comparable beauty, although, so far as concerns composition their glazes did not differ from those of the Chinese potters sufficiently to account for the signally superior results obtained by the latter. On the whole, it appears a reasonable conclusion that the exceptional qualities of Chinese glazes were due in part only to the nature of the materials employed, and that they owed something to the troublesome and seemingly unscientific method of their application.
A word of explanation may be added here with reference to the expressions couleurs de grand feu and couleurs de demi-grand feu. In the case of the former, the colouring matter is mixed with the glaze, applied to the raw pâte, and exposed to the full heat of the porcelain furnace. In the case of the latter, the colouring matter is added to a fusible base, is applied to the ware already baked, and is subjected subsequently to a reduced temperature under which the base vitrifies and adheres to the glaze at the same time as the colours develop. The method of mixing the colouring matter with the glaze, and exposing the finished piece to a high temperature was adopted by the Chinese in the case of their richest monochromatic and polychromatic glazes, with exceptions to be presently noted. The couleurs de demi-grand feu were employed in the decoration of enamelled porcelain. The colours were obtained from metallic oxides. Binoxide of copper gave green, bluish green, and turquoise blue; oxide of cobalt, blue; oxide of antimony, yellow; oxide of iron, brown and vermilion; oxide of manganese, violet and black; arsenical acid or stannic acid, white; chloride of gold, certain shades of carmine; and oxide of chromium, certain tints of green. A third range of colours was developed at a still lower temperature. These are known as couleurs de petit feu. They are of the nature of pigments rather than enamels, and the Chinese potters do not seem to have employed them until the middle of the eighteenth century.
The glazes noted above as exceptions to the rule of application to the surface of the crude pâte, were yellow, violet, peacock green, "king-fisher's wing" and "cucumber-rind" green. These were applied to porcelain in the form of biscuit, but, on the other hand, the temperature to which they were subsequently subjected almost justifies their inclusion in the category of couleurs de grand feu.
It is evident that great difficulty attended the manufacture of couleurs de grand feu according to the Chinese process. The potter, in order to apply his glaze, had to handle the piece before stoving, when the soft clay was sensible to the slightest pressure. To pour the glazing material over it in this condition, or to immerse it in a bath of glaze, with such delicacy of manipulation that no distortion of form occurred, was a feat requiring much practical skill. To facilitate the operation a foot of clay was left adhering to the piece until the process of glazing was completed, and on the removal of this superfluous clay, the rim, or it might be the whole under surface, of the base remained unglazed. Sometimes the latter defect was partially remedied, but in scarcely any cases was glaze applied to the rims at the bottom of Chinese wares.
When egg-shell porcelain or other very delicate ware had to be glazed, the methods of aspersion or immersion while the clay was still soft, presented additional embarrassment. It is said that the difficulty of handling such pieces inspired the invention of the process of insufflation, to which are due some very beautiful effects. But examination of the curiously mottled glazes of the Chün-yao and other chefs-d'œuvre of the Sung Dynasty, shows that the practice of some such method must have preceded by a considerable interval the manufacture of egg-shell porcelain. In applying glaze by insufflation, the potter used a tube having its orifice covered with gauze, through which he blew the glazing material. In this way a dappled appearance, known by collectors as soufflé, was obtained. The results varied according to the patience with which this process was performed and the degree of consistency of the glaze. Evidently the tone of the colour could be changed by repeating the insufflation; in this respect the process may be described as stippling. Again, by blowing on different colours, intermingled clouds of variegated speckles made their appearance. Finally, by increasing the consistency of the glazing material and the quantity of kaolin it contained, there was produced a shagreened surface varying in roughness from the rind of a lime to a dust of millet seed.
As to the means by which the Chinese manufactured the much-admired egg-shell porcelain, there is no certain information. In Europe, ware of great thinness is obtained by a simple and ingenious device. The porcelain pâte, having been brought to a liquid state, is poured into a mould of dry plaster. That portion of the pâte which comes into direct contact with the plaster is deprived of its moisture by the absorbent properties of the plaster, and adheres to the surface of the mould. The mould being then reversed and the liquid pâte run off, there remains a thickness of porcelain varying with the absorbent properties of the plaster and the time during which these are suffered to act. The Chinese keramist may have been acquainted with this process, but evidence points in the other direction. At all events, he succeeded in manufacturing porcelain so thin that, according to his own description, it seemed to consist of glazing material only.
With respect to the stoving of Chinese porcelain, there are few specially interesting points to be noticed. The temperature to which the kiln was raised fell short of that employed in Europe by from two to three hundred degrees. The furnace, in ordinary cases, was kept alight for about thirty-six hours, and after its extinction, the contents of the kiln were left to cool gradually. The latter process occupied four or five days, and care was taken to exclude the air during the cooling. It should be mentioned that the heating also was gradual, the maximum temperature not being developed until several hours after lighting the furnace. Various effects were produced by the introduction of currents of air while the porcelain was incandescent, but these have already been spoken of in the context of the wares they concern. M. d'Entrecolles, describing the practice in vogue at the beginning of the eighteenth century, when he was in China, says:—"The furnace is heated during a day and night, after which two men take it in turn to feed wood incessantly to the fire. The quantity of wood consumed by each furnace is as much as a hundred and eighty charges. Judging by what Chinese books tell us, this quantity of wood could not have sufficed in former times. We are assured that although the furnaces were only half as large then as they are now, two hundred and forty charges of wood were fed to each, with an additional twenty in wet weather. The preliminary burning was kept up for seven days and seven nights, and on the eighth day, the fire was raised to its strongest heat. The conclusion cannot be avoided that the porcelain of former times was much thicker than that of the present day (1720). Another point of difference is also observable. Formerly the furnace was not opened until ten days had elapsed from the time of extinguishing the fire, in the case of large pieces, and five days in the case of small. An interval of some days still separates the opening of the oven and the extinction of the fire where large porcelains are concerned, since otherwise they would crack. But as for small pieces, if the fire has been put out in the evening, for example, they are taken from the oven the following morning." M. d'Entrecolles' deduction is doubtless accurate with regard to the greater body of the heavy stone-wares of early epochs. But something must be placed to the account of the thick, viscous glazes, which required to be stoved more gradually as well as more intensely, giving results that well repaid the labour and cost expended in producing them.
Speaking generally, it may be said that the genius of the Chinese keramist was mechanical rather than artistic. His choicest pieces owed their value to excellence of glaze, delicacy of colour, or infinitely patient use of the graver's and moulder's tool. When he chose figure subjects to adorn his pieces, he too often fell into the mistake, without possessing the graphical ability, of the early Delft potters, who thought that elaboration of detail could supply the place of artistic conception. And the grotesque perspective of the Chinese decorator helped to heighten the confused effect of his scenes. Even in depicting a landscape he was generally the slave of conventionality. His rocks and trees bore little resemblance to anything common in nature. When he treated floral subjects he was certainly happier. But even then he seldom rose above the level of a copyist, for his designs were faithfully borrowed from the pictorial scrolls with which his country abounded. So, too, of his shapes, and of the wealth of diapers, arabesques, and scroll patterns which he employed. For these he went direct to the innumerable bronzes which the genius of Chinese workers in metals—genius exercised through long centuries—had bequeathed to the nation. It is true that upon some specimens of Kai-pien blue-and-white and of Famille Rose decoration is found which in grace, delicacy, and artistic conception leaves nothing to be desired. But these are notable exceptions. In a majority of cases the mechanical element is conspicuous. When the connoisseur comes, however, to consider the quality of the Chinese potter's pâtes; the remarkable skill with which he varied their composition according to the nature of their covering; the marvellous softness and lustre of his glazes; his extraordinary range of fancy in respect of colours, and the brilliancy and purity of his enamels, it is necessary to admit that in everything relating to the technique of his art, he has no equal. The remarks upon this subject by the eminent expert, M. Alphonse Salvétat, in his preface to M. Stanislas Julien's Histoire et Fabrication de la Porcelaine Chinoise, are of much interest. "Variety," he writes, "of fonds colorés de grand feu has made the reputation of Chinese porcelain quite as much, perhaps, as originality of decoration and rich harmony of painting. The analyses which I have been able to make of these colours, as well as the syntheses which I have tried, justify me in regarding as tolerably exact the greater part of the recipes given in Chinese books, at any rate in the case of those where the synonyms are easily found. Some of the Chinese colours de grand feu have not been as yet reproduced on European porcelains. I may cite specially the clear, bluish green colour known as céladon, so much sought after by amateurs, and the reds, sometimes orange, sometimes bordering upon violet, which owe their colour to protoxide of copper. These tints are of great delicacy and brilliancy. It would be a matter of real interest to reproduce them for use on our porcelains. But this is not the place to give recipes which render their production probable, perhaps possible. It is to be regretted that the detailed instructions given by M. Bronginart, at different epochs, to travellers setting out for China, or the letters which I have myself addressed to persons living in that distant country, have not brought us more complete ideas upon this subject than those acquired by examining or analysing pieces which we have been permitted to study. The want of isolated materials and their absence in either a crude or prepared form, as they were furnished by the consignment of Père Ly, so far as concerns decorations in couleurs de moufle induce us to think, as has been often said, that those colours are no longer manufactured in China and that the methods of producing them are lost. There are also found, in certain Chinese colours, shades which appear accidental, as céladons and reds, and which prove that those people owed much to chance in their manufactures. The same glaze gives different results under varying conditions. It has been well demonstrated by me that empirical essays only could have led to the discovery of most of the colours which we seek to imitate. This remark applies above all to colours obtained by mixing, in variable proportions, ferruginous manganese and cobaltiferous earth with white glazing material already prepared. It is evident that colours obtained by such mixing could not always present identical tints; that the state of the atmosphere in the furnace could change them; that they would be more or less green, more or less black, according to the original composition of the materials employed in their preparation and according to the proportions in which those materials were mixed. "The fond laque (golden brown or dead-leaf colour) of China presents, like the colours we have just enumerated, variations of tone. Sometimes it is clear; sometimes it is deep; sometimes it is like bronze. These differences are due to the proportion of oxide of iron which enters into the components of the mass, as well as to the influences of the gas which envelops it in the kiln. . . . The colour, as is known, can be perfectly imitated in France by adding to ordinary white glazing material a quantity of oxide of iron.
"The mineral cobalt, as it is found in China, that is to say, peroxide of cobaltiferous manganese, mixed simply with white glazing material, gives a blue colour au grand feu, sometimes pale, sometimes deep. According to the quantities added and the cobaltiferous richness of the mineral employed, blues of a character more or less violet are obtained.
"When one examines attentively the manner in which black glazes were produced on Chinese porcelain, one sees that all are not simularly manufactured. In some cases the black results from the thickness of the coloured glaze; in others the superposition of various colours of different shades produces a tint of such intensity that it appears black. Sometimes the black is obtained by superposing brun de aque on a blue ground: sometimes, again, it is produced by the inverse process of superposing blue on brun de laque.
"I conclude here my examination of the principal fonds de grand feu which characterise Chinese porcelains. Evidently the manufacture comprises processes which give products very interesting, original, and beautiful. But these processes are often only modifications of those in more general use. At the same time, it is to be observed, inasmuch as the fact establishes another point of divergence between European and Oriental manufacturing methods, that among the colours peculiar to China, some have evidently been applied to biscuit, that is to say, to porcelain already stoved. Looking closely at these colours, they are seen to be cracked, and the crackle, which is very fine, forms a net-work with very small meshes. The analysis of these colours, or the simple test of touching them with fluor hydric acid, shows that there is a considerable proportion of oxide of lead in their composition. This naturally places them in the catalogue of colours used for decorative purposes, which may be called couleurs de demi-grand feu. We find nothing analogous to them in our own colours. To name them, will sufficiently indicate those which we qualify thus. They are: violet, turquoise blue, yellow, and green. . . . While admitting, however, that nothing recalling these productions is made in Europe, we hasten to add that it would be easy to imitate them; for the green and the turquoise blue owe their colours to copper; the yellow is obtained from antimony, and the violet from oxide of manganese slightly cobaltiferous. We are convinced that some synthetical experiments would lead at once to the imitation of these kinds of products. I may add that I prepared, some time ago, fusible colours which can be easily applied to the biscuit of Sèvres porcelain. But as they are stoved at a low temperature, they differ considerably from the colours prepared by the Chinese. Besides, borax enters into their composition.
"The couleurs de moufle claim our attention now, and in order to continue the comparison I have commenced between the porcelain of China and the corresponding European product, I shall briefly recall the colours which compose the palette of potters in Europe, especially at Sèvres. I will say a few words of the principal conditions they must fulfil. It will be easy to appreciate the differences to which I have still to direct attention.
"The colours should be able to attach themselves firmly to the surface of the porcelain, and at the same time to acquire by fusion the vitrification which is an indispensable feature in the éclat of the decoration. They are obtained by mixing either an oxide, or a compound of different colouring metallic oxides, with a vitreous flux, the composition of which varies with the nature of the colour it is desired to obtain. That which is most generally employed is known as fondant aux gris. It is used for greys, blacks, reds, blues, and yellows, and its ingredients are six parts of red lead, two of silicious sand, and of liquid borax. The colours are obtained ordinarily by mixing three parts of the flux, by weight, with one part of metallic oxide, so that the composition may be generally expressed as follows:—
| Silica | 16.7 |
| Oxides of lead | 50.0 |
| Borax | 8.3 |
| Colouring oxides | 25.0 |
| 100.0 |
Sometimes the mixture of oxide and flux is melted or ground before using; sometimes, on the contrary, the oxides are simply mixed with the flux. The colour obtained is immediately employed without any prefatory fusion or calcination. If the colour is to be produced by a combination of the oxide with the flux, as is the case when oxide of cobalt is employed, it is necessary to melt the oxide previously with the flux, in order that the colours may have the required tone. But if the colouring matter belongs to the oxide itself, and if the latter has only to be disseminated and not brought into a state of combination with the flux, it must not be melted before use. The various colours produced by peroxide of iron belong to this category. Were the peroxide of iron melted with the flux, the colour would be perceptibly altered, and the second fusion undergone during the stoving of the decoration would produce a further change. The assortment of colours when made of preparations succinctly indicated here suffices to produce the chefs d'œuvres of oil-painting. All these colours should be able to melt at the same time, and to present, after stoving, an uniformly vitrified mass. This condition is essential. The paintings obtained with Chinese colours are far from satisfying these conditions as to equality of thickness and vitrification. Some are brilliant, perfectly melted, and applied enough thickly to appear in relief on the surface of the porcelain. Rose tints obtained from gold, green, and yellow belong to this category. Others, such as iron-reds, and blacks, are generally almost quite without vitrification, or show only a little vitrification at places where they are thin.[1] Their thickness is always much less than that of vitrified colours. Chinese pictures, too, have a character quite different from that of our pictures. Neither the figures nor the flesh are modelled. Black outlines define all the colours. The tones are not shaded. The colours are applied in flat tints which the painter afterwards damasks with different colours or with metals; but the operation of grinding and mixing different colours upon the palette, a method which gives so much resource to our painters, does not appear to be practised by the Chinese.
The appearance of their paintings, when closely examined, recalls that of the stained glass mosaics manufactured with so much art in the thirteenth century, and in which the whole design and all the modelling of figures and accessories were the result only of red or brown strokes applied to fragments of white coloured glass.
"Having regard to the thickness of the colours employed and yet to the lack of intensity, is many cases, of the tone obtained, one is led to conclude that these colours, compared with our own, contained but a small proportion of the principal colouring matters. The conclusion has been fully verified by experiments. They prove that the colours by means of which the Chinese obtained such remarkable results, in respect of brilliancy and harmony of decorative effect, had much more analogy with the vitrified substances known as enamels than with anything else.
"Whatever be the origin of the colours used for the decoration of porcelain in China, they all present, simultaneously with great simplicity, a character of similarity which cannot escape attention. The flux, which is not distinct in the colour, is always composed of silica, of oxide of lead in proportions somewhat variable, and of a greater or less quantity of alkalies (soda and potash). This flux retains, in a liquefied state, as silicates, some hundreths only of colouring oxides: heir number is exceedingly limited. The colouring matters are, oxide of copper for greens and bluish greens; gold for reds; oxide of cobalt for blues; oxide of antimony for yellows; arsenical acid and stannic acid for whites. Oxide of iron and oxides of impure manganese (which give, one a red, the other a black) constitute the only exception, doubtless because it is impossible to obtain these colours by the process of liquefaction from the oxides mentioned above.
"This special composition of the colours used in China leads to special fashions in the pictures they are employed to produce, and it is to this fact that porcelains manufactured by the Chinese and the Japanese owe their distinctive aspect. Some colours are applied directly, just as they are procured in the market. Others, before being used, require a varying addition, doubtless determined beforehand by experience. They are all thus brought to develop uniformly at a fixed temperature. An assortment obtained in Canton, where it was taken from the table of a Chinese painter, gives an example of a palette fully prepared for use. The necessary additions must have been already made, and we see that the white lead added was in very small quantities, if even that disclosed by the analysis is not due to the beginning of a change in the colour during the operation of grinding. It would carry me too far were I to compare in succession each of the Chinese colours examined above with its corresponding colour as used in Europe, whether at Sèvres or elsewhere. I shall limit myself, therefore, to noting in a few words and in a general manner the essential differences that distinguish the two palettes. It will thus be possible to appreciate naturally the opposing aspects of European porcelains and of similar wares manufactured in China or Japan; aspects so different that it is impossible, even at first sight, to confound the productions of the two countries.
"I have said that in Europe colours for painting hard porcelain are formed by mixing certain oxides with certain fluxes. And I have noted that the Chinese colours differ completely as much in respect of the nature of the constituents of the flux as in the proportions of the colouring oxide. Differences no less distinct are observed when we consider the state in which the colouring matter presents itself. Finally, the two assortments admit of no further comparison when we come to establish a parallel between the substances employed as principal colouring matters in the two cases.
“We have seen that oxides employed in the palette of the Chinese were confined to oxide of copper, gold, antimony, arsenic, tin, and oxide of impure cobalt, which gives sometimes blue, sometimes black ; finally, to oxide of iron, which produces a species of red. We have seen that in the colours of Europe, where use is made of various oxides cited above, much advantage is derived from substances unknown to the Chinese. Thus the tone of pure oxide of cobalt is modified by combination with oxide of zinc or aluminum; sometimes with aluminum and oxide of chromium; pure oxide of iron furnishes a dozen different shades from orange red to deep violet; pale or deep ochres, yellows, or browns are obtained by combining various proportions of oxide of iron, oxide of zinc, and oxide of cobalt or nickel; the browns are obtained by increasing the proportions of oxide of cobalt contained in the composition that gives the ochres ; the blacks, by suppressing the oxide of zinc in the same preparations. The shades of yellow are varied by adding either oxide of zinc or tin to lighten them, or oxide of iron to render them deeper. Oxide of chromium, pure or combined with oxide of cobalt or oxides of cobalt and zinc, gives yellowish greens and bluish greens which vary from pure green to almost pure blue. Metallic gold gives the purple of Cassius, which can be changed at will into violet, purple, or carmine. To this list may be added oxide of uranium, the chromates of iron, of baryta, of cadmium, which give useful colours, and in conclusion we may notice the recent employment of metals inoxidisable by heat, the discovery and preparation of which require a knowledge of chemistry not possessed by the Chinese. All these different principal colouring matters are found on European porcelains in a state of simple mixture; on Chinese porcelains, on the contrary, the oxides are dissolved, and this circumstance justifies us in comparing them to another species of production, well known in China and frequently met with in European industry also. In the vitreous compounds called 'enamels' in Europe, we find not only the same colours produced by the same oxides, but also a flux analogous and sometimes identical. Transparent enamels, as is well known, are vitreous compounds of ingredients varying according to the desired degree of fusibility, and coloured by fractions of oxides. The blues are furnished by oxide of cobalt; the greens by deutoxide of copper; the reds by gold. Opaque enamels, as yellow or white, owe their colour and their opacity either to antimony, or to arsenical or stannic acids, isolated or mixed. The resemblance that examination of the colours employed in China establishes between them and enamels, is fully confirmed by the manner in which the colours behave under the influence of heat. Experiments have been made upon Chinese and European wares with the assortments of Chinese colours obtained by M. Itier and Père Ly. On Chinese porcelain, the colours developed at a temperature lower than that employed to fix floral decoration at the Sèvres factory. But upon the Sèvres porcelain, no sooner were the colours developed than they came off in flakes. Direct experiments long ago showed that enamels cannot be used in the decoration of European porcelains owing to this grave defect. Whatever be the cause to which, in the case of European porcelain, this want of adherence on the part of the enamels is due, I think that it is to be found in the difference between the natures of the glazes of the two wares. The more fusible pâte of Chinese porcelain had to be covered with a glaze more fusible than that used in Europe, and it is the introduction of lime into the glazing material which, by diminishing the infusibility of the latter, and, perhaps, modifying its expansion, approaches its physical properties to those of enamel. If the aspect of Chinese porcelains differs from that of ours, if the harmony of their painting seems more varied, these things are, I believe, the necessary result of the methods employed in China. All the colours employed there have but little colouring matter; they have no value unless they are given a thickness which imparts to them a degree of relief impossible to obtain otherwise. The harmony of their decoration results from the nature and composition of their enamels.
"It remains only to say a few words of the processes of stoving the tender colours, or couleurs de demi-grand feu. Chinese books, and plates that we have seen separate the furnaces into two divisions—open and closed. The former are similar to the furnaces employed by enamellers. I am not aware that they are used anywhere in Europe for stoving decorated porcelain except in Germany. Even in China, the danger of breakage limits the employment of such furnaces to the stoving of small pieces. Large specimens are baked in closed furnaces, the arrangement of which is like that of our moufles. It seems, however, that these furnaces have a circular shape, which would make them resemble porcelain ovens of small dimensions. . . . If the reader has well understood my object in comparing the porcelain industries of China and of Europe, he will see the advantages that can be drawn by studying attentively the processes indicated in Chinese books for obtaining certain fonds de couleurs au grand feu. The reproduction of céladon as it is made in China, of the rich reds and lustrous blues so prized by amateurs, the imitation of crackle large and small, would confer advantages the more certain in that these products have a character of quite special originality, that they are much in vogue, and that, up to the present, attempts to imitate them in Europe have not been successful."
M. Salvétat might have added that attempts to reproduce these remarkable works are unsuccessful in China itself. Doubtless many causes have combined to bring about this result, but the most important, as well as the most comprehensive, is, perhaps, the loss of patronage. Patience that knew no weariness, and painstaking that kept no count of time, were distinguishing characteristics of the old artists, but they were characteristics that owed their development less to inspiration than to circumstance. In China, under successive Emperors from the Tang dynasty down to the middle of the seventeenth century, the keramist's masterpieces were destined for Imperial use. He was able to be sure that whatever excellence he might obtain, at whatever cost, would be more than adequately appreciated. Yet, even after so many long years of prosperous achievement—sufficient to have crystallised into a natural endowment the transmitted skill of fifty generations—a brief withdrawal of Court patronage had power to paralyse art. During the years that intervened between the fall of the Ming dynasty (1644) and the accession of the Emperor Kang-hsi (1661), the outcome of the best factories scarcely deserve to be called mediocre; and again, although the reigns of that sovereign and his two successors are memorable as a period of renaissance culminating in hitherto unparalled perfection, the illiberal policy of subsequent Emperors was the signal for an almost immediate loss of everything but tradition. Since the beginning of the present century, China has produced little that deserves to be classed with the works of her old masters. No longer are found the depth and softness of paste, rich velvety lustre of glaze and brilliancy of enamels that distinguished, as they are infallible evidences of, her keramic efforts prior to 1800; while in paintings, bronzes, and lacquer, the same marked inferiority is manifest.
- ↑ M. Salvétat is speaking of colours used in enamelled decoration. His remarks do not apply to black and iron-red monochromatic glazes.