Dyce 1846
William Dyce, Observations on Fresco-Painting, in: Sixth Report of the Commissioners on the Fine Arts. With Appendix, London [William Clowes and Sons] 1846, s. 11–19.
Appendix
No. 4
OBSERVATIONS ON FRESCO-PAINTING, BY MR. DYCE.
Mr. Dyce having visited the Continent for the purpose of inspecting the best examples of fresco-painting, previously to the execution of his fresco in the House of Lords, the Commissioners requested him to state the result of his inquiries; particularly with respect to the fittest preparation of walls for the reception of frescos, the use of particular colours, and the general methods of finishing frescos. In compliance with this request, Mr. Dyce has obligingly communicated the following observations: –
August 31st, 1846.
SIR,
I HAVE the honour to reply to your communication of the 10th current, in which you request me to state, for the information of Her Majesty’s Commissioners on the Fine Arts, the results of any inquiries made by me on the subject of Fresco-painting during a recent visit to Italy, particularly on the following points: –
First, on the fittest preparation of walls for the reception of frescos; and, secondly, on the use of particular colours, and on the general methods of finishing frescos.
On the first of these heads, the information communicated in the earlier Reports of the Commissioners is so ample and satisfactory, that little more, I conceive, need be said. It seems to be clearly ascertained, that although a surface of brick is best adapted to the process of fresco-painting, the chances of durability are greater when lathing is employed. This general conclusion, however, which is more or less applicable under all circumstances, still leaves undetermined the question of preference in any particular case; and I am inclined to think that in this country, unless special precautions have been taken, it will always be advisable to resort to the use of battens and lathing. The danger to be apprehended from a surface of brick, when the wall is of recent or comparatively recent construction, at least in such buildings in the metropolis as are likely to be decorated with frescos, is not so much the transmission of damp from the soil or from the roof, as the exudation of the salts with which London bricks are highly charged, especially such as are employed in the construction of interior walls. However dry a wall of this kind may appear to be, or be in reality, every fresh wetting* of the surface will, for a period of indefinite length, cause the bricks to throw out certain salts in solution, which effloresce on the surface of the plaster laid on them. This saline moisture penetrates even through cements; but if it were otherwise, or if means were taken to render the surface of the bricks impervious to damp, it is clear we should have lost thereby the very quality of absorbency, on account of which bricks are preferred to lathing for the uses of fresco-painting. Of course the precaution might be taken to line, or construct, such portions of a wall as were intended to receive frescos with a better kind of bricks than I have referred to; but if the usual sort has been employed (as has, for instance, been the case in the New Palace of Westminster), it seems to be absolutely necessary either to adopt the ordinary method of lathing, or some expedient like that suggested in the Second Report of the Commissioners at pp. 51, 52, by which it is proposed to apply a hydrofuge to the surface of the bricks, and then to restore the necessary degree of absorbency by lining the surface, so prepared, with tiles roughened to afford a hold for mortar.
* That is, to a degree sufficient to secure the adherence of a coating of plaster.
I confess, however, that judging by the little experience I have had, I am inclined to doubt whether the advantage of bricks over lathing in point of absorbency, if it has not been over-estimated, is not, at most, so inconsiderable as to render it of little moment which of them be employed. The inconvenience arising from the more rapid drying of plaster on lathing is chiefly felt in the execution of objects, such as draperies, which cover a large space, are painted with colours that must be laid on while the intonaco is in a very fresh state, and happen to be so arranged that the whole must be completed in one day. I say, happen to be so arranged,—because supposing such an inconvenience were always to be anticipated, it would be easy, beforehand, to make such an arrangement as would obviate it. But after all, the inconvenience is not great; and perhaps it might be entirely done away by the simple expedient of increasing the thickness of the preparatory coatings of mortar. If one may judge by the durability of ancient Roman plastering, no risk is incurred by an increase in the thickness of mortar, even to the extent of an inch and a half or two inches, provided the laths are strong enough to bear the weight; and I imagine that if the ordinary mixture of lime and rough sand, or pounded marble, were not sufficiently absorbent (though there is no reason to doubt this), other compositions might be found better adapted to the purpose.*
* Perhaps the mortar commended by Anton Francesco Doni, in his little work entitled “Disegno” (Ven. 1549), might answer the required conditions. It was compounded of lime and roughly-pounded pumice-stone.
Another point connected with this subject is not unimportant. Practical men in this country to whom the question has been referred say, without exception, that so far as the preparatory coatings of mortar are concerned, no purpose whatever is served by allowing so long a period as is generally prescribed to elapse between the application of these and the execution of the picture. When the mortar, they say, is once thoroughly dry, which, in the absence of counteracting influences, it is in the course of a week or two, it undergoes of itself no further change; and they conclude that the prescribed delay must have reference to a possible bad condition of the wall to be painted, and is intended to allow time for its complete desiccation and for the exudation of saline matter. But if, as is the case when lathing is employed, the transmission of moisture from the wall is intercepted, they believe that the mortar, when it is quite dry, and provided it is suitable in other respects, is in as fit a condition for the operations of the artist as ever it will be.†
† Experience only can test the justness of this opinion; yet, even at present, this is not altogether wanting. The whole of the frescos submitted for exhibition in Westminster Hall, and those executed at the commands of His Royal Highness Prince Albert in the gardens of Buckingham Palace, were painted on recently laid mortar; and, so far as I am aware, none of them have shown defects attributable to that cause.
In the preparation of mortar for fresco, pozzolana is not likely to be employed in this country; but it may be noticed, as an objection to its use, that its purplish-pink colour almost invariably in course of time rises to the surface, if a wall is affected by damp from the interior. Of this I have observed many instances in Italy, particularly in the works in the Campo Santo at Pisa. Even those of Benozzo Gozzoli are stained with this pink hue, although pozzolana certainly did not enter into the composition of the whole arricciato prepared for his works, but was laid (along with other ingredients), in the form of a cement, on certain spots only where flaws of some kind had probably appeared after the arricciato was dry.‡
‡ It deserves notice, however, that the masters of Gozzoli’s time were fond of a pink colour in architecture, middle ground of landscapes, rocks, and the shadows of light clouds; and no doubt a good deal of the pink now pervading Gozzoli’s works existed originally much as we see it in the missal illuminations of the quattro-centisti.
2. With respect to the second particular, referred to in your letter, I must premise by explaining that all pigments adapted for fresco-painting may be used with about equal facility, provided they are mixed with an adequate proportion of white, i. e. of the washed lime employed as white in fresco. In that case, whether they are applied solidly, or in the form of thin washes, their adherence to the intonaco, and the effect they will produce when dry, may be calculated upon with tolerable certainty.
If, however, tints be required of greater depth than can be thus obtained, and for which colours must be applied in a pure or nearly pure state – that is to say, with little or no white – the use of certain pigments becomes difficult and embarrassing. In the first place, some colours, such as Indian red, burnt vitriol, umber, and others, deepen in tint as they dry. When they are used in a pure state, this increase of depth may be depended upon, and allowance made for it; but if a small quantity of white be added with the view of reducing the strength of the tint, the result is extremely uncertain. Sometimes the white seems to disappear, and produce no effect whatever; at other times it has much more effect than was intended. Portions of even the same mixture used on different days will dry of different degrees of depth. But as these differences of tint depend on the state of the intonaco when the colour is applied, and on the manner of its application, I have no doubt that a little experience would suggest remedies for this kind of inconvenience.
But, secondly, there is another kind, for which the remedy seems as yet to be a problem. Some pigments (one of them at least most necessary and important) will scarcely adhere to the intonaco at all, or require such precaution as renders their use extremely difficult. If, for instance, a tint of pure ultramarine be laid on the intonaco in such a body as will entirely cover it – and this can only be done by going over the space twice or three times, allowing a certain interval to elapse between the successive layers of colour – it will be found in nine cases out of ten, that the first layer has adhered perfectly, the second partially and in streaks, and the third not at all. The same results take place frequently in the case of vermilion,* sulphuret of cadmium, black, both of Cologne earth and burnt peach-stones, and, to a certain extent, of burnt vitriol.
* I am disposed to think that the directions given by old authors for the preparation of vermilion for fresco-painting are to be looked upon much in the same light as their rule of allowing a long interval between the first plastering of a wall and the execution of a picture on it. In both cases the remedy may have answered the purpose; but as the evil to be remedied was misapprehended, the remedy might in many cases have been dispensed with. In the instance of plastering, it is quite clear that old writers attributed the mouldiness and other appearances on the surface of mortar to the nature of the mortar itself, whereas the cause lay in the wall on which it was laid. See, for example, Vasari’s mistake respecting the efflorescence on Michael Angelo’s works on the ceiling of the Capella Sistina, which he attributed to the nature of the lime prepared from travertine. San Gallo, the architect, evidently knew better; and as Condivi relates (Vita di M. A. Buonaroti, 1553, p. 24), explained to Michael Angelo that he had used the lime in too fluid a state.* So in the case of vermilion. It was thought that there were some qualities in this pigment which must be destroyed by the action of lime before it could be used with safety: whereas in truth these qualities were mere adulterations with red lead, red orpiment, &c. Pure vermilion is not acted upon by lime.
* The true explanation, however, was that he had wetted the wall too much.
The peculiar difficulty of using ultramarine in fresco, arises from the facility with which all moisture is absorbed, and entirely sucked out of it by the intonaco.† In the case of other pigments, such as terre verte, terra di Siena, and the ochres, when applied in a pure state, though the excess of water used with them is absorbed by and diffused through the intonaco, the colours still remain more or less in the state of paste, at least in as moist a condition as the intonaco itself, and form with it a homogeneous mass; but ultramarine, if the layer of it exceeds a certain thickness, becomes dry and mealy on the surface, and for want of moisture the outer portions of colour, which are beyond the medium of action between the lime and the carbonic acid of the atmosphere, are excluded from the crystallization of the lime, and remain in the form of powder on the surface when the plaster is dry.
† The use of vermilion and cadmium might be dispensed with; but, notwithstanding the high authority of Luini, in whose frescos in Milan there appears to be not a particle of ultramarine, I fear we cannot give up this colour. It remains to be proved whether the factitious ultramarine or cobalt can be used as substitutes for ultramarine of lapis lazuli; but if we except these and smalt, which is more unmanageable than ultramarine, there are no other blues at the command of the fresco painter. The blue carbonate of copper (mountain blue), formerly much employed, is of an exquisite tint; but it can only be used in distemper; and, even then, it is sooner or later converted into the green carbonate. In the Campo Santo of Pisa there are draperies and skies painted with the blue carbonate, which show all the states intermediate between blue and the brilliant green to which the pigment changes. So also in the arabesques in the Loggia of the Vatican by Raffaelle and his scholars. In some instances the deep blue grounds of panels in these arabesques on which figures are painted, have changed so uniformly and advantageously to a brilliant green, resembling the English pigment called emerald green, that it is difficult to suppose they were not originally of that colour. It is only by observing other panels where the change is partial that we discover the original tint.
It must be remembered that I am speaking now of pure colour, and the production of dark and rich hues, which cannot be obtained alla prima, but require two or even three separate layers of colour, the first of which must be applied when the intonaco is in a fresh state.‡ No difficulty would occur in the case of ultramarine, if the intended effect could be given alla prima; but a rich and solid effect is not thus attainable. On the contrary, although a single layer of ultramarine, provided the intonaco be in a fit condition, will adhere sufficiently well, yet it dries of a lighter and more chalky hue than the tint actually applied ought to have produced. There are two reasons for this: – in the first place, the adherence of the colour depends, if not on its actually touching the lime, at least on the penetration of the lime water of the intonaco; and this, though perfectly pellucid while the surface is wet, deposits, as it evaporates, the lime held by it in solution, and produces on the tint about the same effect as if so much lime had been actually mixed with it. But, secondly, it is barely possible to lay a tint on the fresh intonaco without stirring it up more or less. Even with the utmost care, a certain portion of lime mixes with the colour, and, of course, alters its tint wherever the mixture has taken place. This then is another cause of the bleached and streaky look of such colours as ultramarine laid on alla prima.
‡ No reference is here made to the case of transparent washes of pure colour. These must be applied when the intonaco has lost its freshness; a circumstance which will indicate what pigments are proper to be used in a transparent form, whether they are laid on the virgin intonaco or on some coloured ground prepared to receive them.
This action of the intonaco takes place more or less on all colours applied to it while it is fresh; hence, in fact, the necessity so much insisted upon by old authors, of going over every tint at least twice, even if there were no other reason for it. “Fresco-painting,” says the Jesuit Pozzo, “has this peculiarity, that the first colours which touch the lime soon become faint and lose a great part of their brilliancy. On this account, it is necessary to go over the work again with a greater body of the same colours.” This is necessary, first, to obviate the effect of any accidental mixture of the lime of the intonaco with the colour laid on it; and, secondly, to prevent the crystalline action of the intonaco from reaching the surface of the colour in greater strength than is required for fixing it. To secure these objects, a certain interval must elapse between the first and second applications of colour. The first coat must have become so firm as to receive the second without being stirred up or mixed with it; but as this implies that the intonaco has become more solid and absorbent than before, it will be at once perceived wherein lies the difficulty attending the application of a second coat of ultramarine. The water by which it is applied, is so entirely and rapidly absorbed by the intonaco that there remains too little moisture to act as the medium of crystallization; and the colour in consequence dries in the form of powder. It is not meant to be affirmed, that a second layer of ultramarine never adheres; but the chances are, that with such precautions as are possible in ordinary practice, the adherence will most frequently be imperfect or partial. If the object to be painted, a piece of drapery for instance, covers a considerable space, the probability is, that the colour will adhere unequally, owing partly to the unequally moist state of the intonaco, which cannot be prevented, and partly to the manner in which, through inadvertence, or with a view to some pictorial effect, the colour may have been applied to particular portions of the surface.
In the foregoing observations, I have supposed the application of pure ultramarine; but as the unbroken colour of this pigment is of too ethereal a quality for the true representation of local tints* of drapery, it is obvious that one means of facilitating the use of it will often arise out of the necessity of reducing its tint by an admixture of other pigments more manageable than itself. For example, the tint may be reduced and deepened by the addition of terra di Siena and colcothar of vitriol, in such proportions as to constitute, with the blue, a neutral or blackish tint;† or it may be inclined towards green or purple by varying the proportions. Still it must be admitted that a remedy of this kind is but of partial application.
* By local tint I understand that colour of an object, which independently of any effect of light and shade upon it, is identified with its surface. The blue of the sky, which ultramarine represents with so much truth, is not a local tint, but a colour which suggests the notion of space and vacuity. Connect the idea of surface with a painted sky, and it ceases to be a true resemblance. On the other hand, give an ethereal effect to blue drapery, that is to say, destroy the identity of its colour and surface, and it becomes unreal and unsubstantial, an effect which is extremely frequent in frescos painted since the beginning of the 16th century.
† If great depth be required, black may be added; but, by itself, black (whether of burnt Cologne earth, or of burnt peach-stones) will not promote much, if at all, the adherence of ultramarine.
Another remedy in use among the old masters (on which it is to be regretted the documentary information is not very explicit), was the employment of some kind of tempera in the application of ultramarine to the wet plaster. Cennino Cennini, speaking of the two kinds of tempera used for painting a secco, expressly says, that the second sort (prepared from the yolk of eggs only) was of universal use, in muro, in tavole, in fresco;‡ and Armenini, who certainly did not contemplate, as a general rule, the application of blues in tempera when the wall was dry, makes an exception in the case of il smalto con altri simili azzurri, when he directs that colours for fresco-painting should be ground in water alone.§ Armenini does not inform us with what which the blues were to be mixed; but Palomino, who probably only followed the tradition of earlier times, recommends goats’ milk for colours like mountain green and smalt, which adhere with difficulty. How far the caseous element of milk, which, as is well known, forms with quicklime a cement insoluble in water, may combine in a similar manner with the lime of the intonaco during its absorption of carbonic acid,* is a question which might with advantage be submitted to the test of experiment. If it does, a better vehicle than milk would probably be found in the solution of pure caseum in water; but if caseum (whether in this state, or in combination with the other constituents of milk) simply acts as a glue, it might be advantageously replaced by other gums not liable to the putrefaction which caseum undergoes when kept moist for a certain time; a condition of their employment being, of course, that their adhesive qualities are not destroyed by the action of the wet intonaco.†
‡ Trattato, part iii. cap. 72. The word “muro” is evidently employed here to signify mural painting a secco, as distinguished from mural painting a fresco, and from painting a secco on panel, signified by the words “in tavole.”
§ Veri Precetti della Pittura, 1678, lib. ii. 64.
* It appears to form with recently-slaked lime a mixture, which, though not so powerful as that known by the name of “Vancouver’s Cement,” is sufficiently so to be used as a tempera in house painting; and probably with old lime it would be equally efficacious. See on this subject “Mémoire sur la Peinture au lait,” par A. Cadet-de-Vaux. Par., 1804.
† A solution of starch might answer to the required conditions.
With respect to the use of milk, the few trials I have made, seem to evince that it answers equally well as a vehicle for ultramarine, whether the colour be applied to the bare intonaco or to a ground of red or some other colour laid on the intonaco as a preparation; but it happens that both economy and convenience point out the latter as preferable. The expense of ultramarine is so considerable, that it becomes important to produce the intended effect with the least possible waste of colour; and this purpose a dark ground greatly facilitates, by showing ultramarine as a comparatively light and opaque pigment instead of the transparent and dark one which it appears on a light ground.
If it be true, as Palomino affirms, that lime destroys the colour of ultramarine, this fact would afford an additional reason for the use of a ground of some pure colour under the blue as a protection to it. But the statement of Palomino is not confirmed by any ancient authority; and is, I think, for other reasons, doubtful. The bleached appearance of ultramarine in many ancient frescos may be accounted for without supposing any want of durability in the pigment itself, since the same bleached and faded look will (as I have already observed) be imparted to other pigments, of which the durability has never been questioned, if the precaution of weakening the force of the lime by repeated layers of colour is not taken.‡ But the difficulty, in the case of ultramarine used a buon fresco, is to take this precaution. If it be applied while the intonaco is fresh enough to secure its adherence, the lime whitens it; if to prevent this, the intonaco is allowed to harden, the ultramarine will not adhere; so that Palomino’s conclusion, respecting the impossibility of using ultramarine with any safety in pure fresco, was not far from the truth, though his premiss might be erroneous. Of course, if tempera such as milk were employed with the second layer of the pigment, the difficulty would be got over; but in that case, the first layer done a buon fresco might be dispensed with for the reasons already given, viz., that the ultimate effect of the blue is more easily calculated upon, and less colour is required to produce the intended effect when it is applied to a dark ground. That these were the reasons which led formerly to the very frequent (down, indeed, to a certain period the uniform) use of dark grounds under blue, is, I think, quite clear from the fact, that the same grounds were prepared for painting the blues a secco.§
‡ In either case the intonaco is allowed to acquire considerable firmness before the blue is applied.
§ In the case of pure ultramarine, if one may judge from the description of Cennino Cennini, the shadows only were put in a secco by the more ancient painters; the whole drapery was gone over first with a uniform tint of red and black in fresco, and over that with a uniform layer of ultramarine, tempered with yolk of eggs. On this, when dry, the shadows were hatched.
To what extent this method of applying ultramarine and other pigments with tempera, while the intonaco remained moist, was employed by the masters of the 15th and 16th centuries, it is almost impossible to determine by examining their works. The appearance of a red or other dark ground under blue affords undoubted evidence that tempera has been used with the blue; but if the blue pigment was ultramarine or smalt (i. e. a colour not affected by lime), it does not decide the question whether the blue was applied before or after the intonaco had become dry. Of course if the blue carbonate of copper has been employed, there is no doubt that it was applied a secco, whatever may be the colour of the ground on which it is laid. It is not difficult, however, I think, to account for the deficiency of documentary evidence on the use of tempera with colours applied to wet plaster. Cennino Cennini, who is our only source of positive information on the early method, and Vasari and Armenini, who may be taken to represent the practice of the 16th century, were at variance on a most material question. Up to the beginning of the 16th century fresco had always been reckoned to be one part only of the process of mural painting, of which tempera was the other. The words fresco and secco, applied to painting on walls, referred not to the mode in which the picture was finished, but to the mode in which it was begun. If it was begun on wet plaster it was termed a fresco, if on dry, a secco; but in both cases it was finished a secco. Such being the usual practice, it is obvious that the possibility of applying colours with tempera on the wet plaster would not be made much account of; if it was attended in any case with the least uncertainty or inconvenience, the artist had the ready resource of waiting till the plaster was dry. It is not, therefore, surprising that Cennino alluded only incidentally to the use of tempera on the moist intonaco. Vasari and Armenini, on the other hand, lived at a time when the ideas of artists on the subject of fresco had undergone a revolution. Towards the end of the 15th century the amount of tempera used in finishing frescos had greatly diminished, especially among the more dexterous artists of the Florentine school; attempts were gradually made to do without it altogether; and at length, in Vasari’s days, the comparative success of these attempts led to an opinion (ever since current) that fresco, as a process of painting, is so complete in itself as not only not to require or be improved by the assistance of tempera, but to be spoiled by it. It came, accordingly, to be reckoned discreditable to an artist, and a proof of weakness, to employ it at all on works in fresco. This notion of the capabilities of fresco, though entertained generally at that time, and perhaps acted upon by the school to which Vasari belonged, must have been found on the whole impracticable; and I make no doubt that there were artists who consulted their convenience more than the maintenance of an opinion, and who, though they might endeavour to abstain from retouching their works a secco, nevertheless continued to employ tempera for the application of colours like ultramarine to the wet intonaco, without acknowledging that they did so. The use of tempera at all, and for any purpose, in fresco-painting was inconsistent with the popular opinion; if it must be used, the less said the better. Hence it is that we discover only by an unexplained hint in Armenini’s treatise that, notwithstanding the fuss made about pure fresco, a certain class of pigments could not be, and, in fact, were not at that time, applied with water alone; in other words, could not be and were not then painted in pure fresco.
It deserves also to be noticed, that Vasari’s strictures on the use of tempera apply solely to retouchings, properly so called, – that is to say, to corrections of defects in pictures already completed in fresco; and it is probable that his objections to the practice were at that time well founded. The use of tempera on the dry plaster had then ceased to be regarded as an integral part of mural painting a fresco; it was looked upon solely as a means of concealing failures and correcting defects of work already quasi completed in fresco. If a shadow dried too light, for instance, tempera was employed to deepen it; if a tint dried of unequal depth, it was stippled up in tempera to uniformity; its use, in short, became matter of accident; the artist could not tell beforehand where he might have occasion to employ it. It is quite obvious, then, that if special precautions were necessary under the old system (as I believe they really were) to ensure, in point of effect, the joint durability of tempera and fresco, – if to ensure this it was necessary to determine beforehand what portions of the work were to be done in fresco, what in secco, and to provide that individual objects in a picture should be wholly executed either by the one process or the other, in every case where a change of tint, supposing it took place, would be perceptible – if these precautions were indispensable to the safe use of tempera over fresco, the practice of retouching in Vasari’s days cannot at times have been attended with very happy results. It is not likely, I think, that Vasari would have willingly admitted the legitimacy of using tempera for any purpose in fresco painting; but his objections, as they are stated by him, are undoubtedly special. He did not object to every use of tempera, but to a special use of it; and even this only with respect to the practice of his own times.*
* Still I am not sure that Vasari’s strong objection to retouching was not a mere crotchet of his. There is no reason to suppose that Domenichino finished his frescos a secco in the ancient manner; yet though they are retouched (in Vasari’s sense) to a much greater extent than the works of any of his cotemporaries, they have proved not only the most durable on the whole, but the most devoid of blackness. I quite admit that instances may be found where certain colours have become black: as, for example, in one of the frescos by Pinturicchio, in S. Maria d’Ara Cœli, in Rome, in which the dresses of some little figures in the back ground have become quite black; but this is not the effect of the tempera, but of the light and air on certain colours, such as Naples yellow, orpiment, impure vermilion, &c., which become discoloured or black by exposure. If flesh painted in fresco were retouched with colour a secco, in which Naples yellow was an ingredient, nothing is more likely than the appearance of grayness, and ultimate blackness, wherever such colour has been laid.
That he made no reference to the ancient and then obsolete method, seems to be quite certain, from the fact that it was not, as he perfectly well knew, attended with the consequences which he attributes to retouching a secco. Whatever be the merits of the old mixed method of execution, there is no doubt, first, that it is not chargeable with the want of durability spoken of by Vasari; and secondly, that until the method was abandoned, fresco painting never exhibited the weak points (so to speak) which it has always, more or less, shown since. It is not difficult to understand why the latter should be true. Tempera was employed by the old painters, not to correct actual failures in fresco, but to make amends for the inadequacy of the very process itself. In fresco, those pigments which possess the qualifications necessary for that mode of painting, are brighter, more intense, and clearer than they are when used in distemper; on the other hand, several pigments proper to the latter, are more brilliant than any that withstand the action of lime. The early masters, then, availing themselves of the advantages offered by each of the two methods, so united them as to obtain results which neither by itself was capable of exhibiting. When they used tempera, it was not so much for a negative as a positive purpose; not so much to cover defects in work already executed in fresco, as to impart beauties beyond the reach of that mode of painting. Each process, in short, was applied to the purpose for which it was most suited;* and this adaptation of the means to the end served no less to impart beauty and variety to the work, than to render it durable in point of effect. Nothing, indeed, can be farther from the truth than that the tempera of the early masters has (to use Vasari’s words) either “become black,” has “tarnished the fresco colours,” “prevented them from acquiring their usual brightness,” or has “shortened the duration of the pictures.” If they took the precautions, which I have supposed they did, none of these results were likely to follow, unless the tempera has of itself a tendency to become black; but this is not true. When the early masters painted on walls entirely a secco, the walls were prepared in the very same manner as they were for fresco, with the sole difference that the intonaco was allowed to become quite dry before the painting was begun. A coating of size was then applied to it, and on this the picture was executed. Now, supposing the picture begun a fresco, carried on to a certain point, allowed to become quite dry, and the coating of size applied to such parts as were to be executed a secco, what is there here to affect the distemper colours more than in the other case? It is plain that if they were durable in the one case, they must have been so in the other. If, in the one, there was no tendency to blackness, and nobody has ever affirmed there was, there could be none in the other, since the circumstances were precisely the same. There is, however, as I have observed, no doubt that Vasari’s remarks were not intended to apply to the old method, which was then obsolete, and which it is most likely he despised.
* I think it likely that, as a general rule, the older masters endeavoured to avoid mere retouchings or mendings of defects, and to execute particular objects either entirely in fresco or entirely in secco, so far as the outward surface was concerned; and this to prevent partial changes of tint, supposing any change were to be expected. This rule would apply both to opaque and transparent tints; but particularly the former. A good example of pure fresco and pure distemper in union exists in a picture by Melozzo da Forlì, now in the Vatican gallery. By looking in a slanting direction, so as to make the surface reflect the light of a window, one can see where fresco has been used and where tempera, e. g. the white, dull red, and slate-coloured draperies, the hands and heads are in fresco, and exhibit the polished surface peculiar to it; on the other hand, the lake-coloured and purple draperies, the ornaments of the Pope’s chair, his scarlet cappa and shoes, and the details of the back ground, are in tempera. In this last only there appears to be a mixture of the two processes.
It must be admitted that there is a general hue of blackness or grayness over the majority of the early frescos, especially of those executed between the middle and end of the 15th century; but this is simply the effect of smoke and dirt on pictures originally painted of a low tone, or with a preponderance of dark and rich colours. Comparing, for instance, the works of Perugino, Botticelli, Ghirlandaio, and others in the Capella Sistina, with those of Michael Angelo, in the same place, it will be found that the former, though (with one or two exceptions) they contain a larger proportion of blackened colours (i. e., of colours originally dark which the smoke has blackened), have not on the whole changed more, in point of tone, from their original state than the works of Michael Angelo.* There is no appearance of partial blackening; every part of the work is equally affected; but, of course, a film of smoke and dust, which converts white into brownish gray, must turn almost to blackness shadows originally dark, or tints originally of great depth.
* In other respects they have changed less. Vasari has admitted that Perugino, whatever his merits as a designer were, was a skilful workman; an admission fully borne out by the present condition of his principal fresco in the Capella Sistina. It is absolutely without a blemish; and having been originally painted of a light tone, it is less changed by smoke than the works surrounding it. The smoke of incense and candles seems to settle more equally on the old than on the more recent frescos, probably because the surface of the former is more equally and highly polished.
All conjecture, however, on this point is set at rest by the frescos of Pinturicchio, at Siena, in which an excess of tempera has been employed. The works I refer to are the series representing the life of Pius II., of which ten are painted on the walls of the cathedral library, and an eleventh over the door of it on the outside, i. e., on the aisle-wall of the cathedral. Those in the library are in the most perfect state of preservation, except at one or two insulated spots to which I shall presently advert. That on the outside has been exposed to the usual risks to which frescos are liable in large churches; viz., to injury from dust, condensation of vapour, and smoke from incense and candles. The library has never been used as a vestry or sacristy, and being shut up at night, neither candles nor torches have ever been burnt in it; in truth, except as a place of deposit for the magnificent choral books belonging to the cathedral, it has seldom been used for any purpose; so that Pinturicchio’s works have had every advantage. The result is very apparent: – the frescos are as bright, fresh, gay, and as devoid of dimness or blackness, as it is possible to imagine they ever can have been. Except at the spots I have mentioned, they are clean and brilliant as the unsullied page of an illuminated missal. Assuming, then, that the works within the library exhibit the original aspect of Pinturicchio’s frescos, we have the means of estimating the effect of exposure to smoke and dust on the picture outside. Now, the latter, while it corresponds entirely with the other pictures of the series† in point of style, of design, of colouring, and of ornament, differs from them altogether in its general tone or hue. There does not appear to be the smallest doubt that originally it corresponded with the others in that respect also; but it has, by exposure, acquired the blackish appearance of the nearly cotemporary works in the Capella Sistina. The smoke and dust which have turned the originally almost white pavement in the picture (judging by the others of the series) into a brownish gray, have throughout lowered the tone of the colouring to its present sombre aspect.
† It appears to have been the last executed. It represents the coronation of Pius III., which fixes the date of it posterior to the year 1503.
The injuries sustained at certain spots by the pictures in the library were occasioned by the percolation of moisture from the roof at the time an upper story was added to the building; and as these injuries are partial and well defined, they serve to show the effect of damp on Pinturicchio’s method of execution. So far as I could judge, the effect has been the reverse of that generally said to result from such a cause. The damp seems only to have destroyed the adhesive power of the tempera and loosened the colour, which has crumbled and fallen off as the moisture dried up.‡ This is the case, in particular, with a crimson-coloured drapery, evidently painted a secco on the dry intonaco with common lake. The colour has more or less fallen off; but judging by the portions which are uninjured, the loosening of the colour has been accompanied by no change of tint.* The only parts of these frescos on which I observed an appearance that seemed to indicate the blackening effect of damp on mere retouchings, were the skies of one or two of them, where the artist has deepened by an infinity of small horizontal touches or streaks of ultramarine. It so happens that wherever the damp from the ceiling has come near enough to be supposed to have affected the upper part of those pictures which have skies, the retouching on them is more or less visible. My glass, however, showed that these retouchings were not black but dark and brilliant blue; and as it is likely that the skies which have thus been deepened in tint were not painted in fresco but in secco, I conclude that the retouching still exhibits its original appearance.†
‡ I have observed the same result to a large extent in the Campo Santo at Pisa, especially in the works of Benozzo Gozzoli. At first I imagined that the portions thus denuded of colour had been washed; but a trial convinced me that damp from the wall was the cause. Benozzo seems to have painted the heads and hands in the following manner: – the outline and dark markings are put in with a reddish brown on an intonaco of a gray colour, and the local tints and high lights are then added. On this, when dry, he hatched the shadows and details with transparent tints a secco, the gray colour of the intonaco serving as a ground for the middle tint. Now these hatchings a secco have come off in many instances so completely as to give the heads an appearance of having been washed and reduced to the state in which they were left in fresco. Being, however, able to reach the uninjured hand of a figure, of which the head seemed to have been washed, I found that the hatchings adhered so loosely to the smooth soapy-like surface of the intonaco, that if the least force had been used the whole would have come off, and the painting reduced to the condition of the heads referred to.
* The brilliancy and beauty of the lake in Pinturicchio’s works not only in Siena but in Rome (in the Borgia apartments of the Vatican and in S. Maria al Popolo) seem to indicate that the current notion about the durability of this pigment is a mistake. When a colour remains to all appearance unchanged for three or four centuries, it may safely be reckoned a durable one. There are lake-coloured draperies in the frescos of Benozzo Gozzoli in the chapel of the Ricardi Palace at Florence, painted in 1467, which are perfectly well preserved. The fresco by Melozzo da Forli, painted previously to 1480, has already been noticed. In this the lake-coloured drapery has been at one part injured and blotted by damp; but the tint is not changed.
† This conclusion is confirmed by the fact that the picture in which this retouching of the sky is most visible is not only nearest the light, but the most perfectly preserved of the whole series. It has not a single blemish throughout; and, moreover, a cypress tree which runs up nearly to the top of the picture has not the very least symptom of mildewing or blackness, though from its colour it certainly was executed a secco. It is of a fresh and brilliant dark green with brownish shadows. The blue sea and distant landscape (high up in the picture) are also well preserved.
One of the greatest peculiarities in the mode of finishing frescos adopted by the quattro-centisti, especially by Pinturicchio, consisted in the use of stuccos gilt and painted for the imitation of certain objects. I am not now referring to the disc or corona behind the head of a saint, or to draperies of gold, and gilt and diapered grounds, in use among the painters of the 13th and 14th centuries, the effect of which was purely conventional, but to the use of gilding and relief as means of producing pictorial effect. Whether an expedient of this kind be legitimate in painting, need not now be considered; but it certainly deserves notice as one of the means by which, in certain cases, those painters who followed the old method obtained a surprising degree of force and reality of effect. The stuccos, which are gilt, have very little relief, only sufficient to indicate the form of the object by the shining of the gold; and as the outline of the stuccos follows the perspective of the picture, all incongruity between parts painted and those in low relief is avoided. In most instances, Pinturicchio has united the two means of imitation with surprising skill: the deception is so perfect that one can hardly tell, at a certain distance, what is merely painted and what is relieved and gilt or painted.‡ Sometimes he is less successful, and overpowers the painting by the wonderful reality of the stucco-work; but on the whole the effect of his frescos is enhanced and rendered highly decorative by this contrivance, whatever may be thought of its legitimacy as a means of pictorial imitation.
I have the honour to be, Sir,
Your very obedient Servant,
W. DYCE.
C. L. Eastlake, Esq.,
&c. &c.
‡ This is remarkably the case in the fresco representing the coronation of Pius III., in which the whole figure of the Pope seems to be in low relief, painted and gilt. But the deception is so complete, that even with the aid of a glass I was unable to determine precisely where the stucco work ended and the fresco began. The dress of the Pope is gilt; and as the figure is placed at the top of a lunette, which is imperfectly seen by a reflected light, I imagine that Pinturicchio adopted the expedient of giving relief to the drapery, in order to make the gold reflect light at the angles and edges of the folds.