t' • - 'S^.-' ->i#^ w* pii. •."-■.■•■ '■ .' ' ■i L ■* 5* ai ' ”» -Jf l ‘ ■ • 'l*-'' ■ f ^ '• ' / r ’ I*#:,:..: r' -v.^i*. -■ -/ ■■ , « '1 ♦■’j !4 -S.V* VV- ‘ t ■ -■>.-, - *:-> - ■• ,1 ■ f ■'. ■ i" ;' ' '■ >-v --•■ i.- _ '• « j' '-- - 1- , '* -•■ ■ '■ '•;«ir ^ ■ ' ■'■ ■ \-j ' tS-f ^ i ; ‘'‘'’' Z >• »“'■■ ■’, ' -'.1.= H ■ fi' 'w' VI .’- ■ <>,(• ■ . %• UrS# f . ‘- - ■ - ' «. ■> .*•.•.(( - • . ■' >' '- . • x I , - '•■1;. •’■'■s'. I ^yr '■ •-::->■ •■ mX- • J :yi. 'K -^W r-_— ^ V ,■^ ■ ^'- • • ^ • i -1 V r.rf 1 j «■ • O?'. S^**Ws*^*“ '<44^-a THE GRAMMAE OF ORNAMENT BY OWEN JONES. ILLUSTRATED BY EXAMPLES FROM VARIOUS STYLES OF ORNAMENT. ONE HUNDRED AND TWELVE PLATES. LONDON : PUBLISHED BY DAY AND SON, LIMITED, GATE STEEET, LINCOLN’S INN FIELDS. |!tinpr-TiNi'-#''r-' : V - V >; 5^21 * >. ^ ♦ ■-r^./ '•■■■^■;> .q /*Tfl-/,A U'P*' .'JJ.../,\ f'. ..-J. •.X.-./ ..iv_/.^ilA_ •. M' i i* ••■T, > ' • f . • * ^ ’ . * vVv :v • • ‘j • If’- • ^ x-c— At./ -i-iaV. • • *• t **. ■ ’ ' - : . 4 t 4 - A' ^ - M 7 ,T . . . * • • ■' •• • iV I- t -I y. o' ry • .• .1 .. 4 ' • • ■:: vsv^S3Ki'' % •■: •w *■ * •> * » #l i t f ♦- i.. • ,y 4 .^ *: • . I t 4 ■ ’ ' Xd' • ♦ • >;V., PREFACE TO THE FOLIO EDITION. It would be far beyond the limits of the powers of any one individual to attempt to gather together illustrations of the innumerable and ever-varying phases of Ornamental Art. It would he harely possible if undertaken by a government, and even then it would be too voluminous to be generally useful. All, therefore, that I have proposed to myself in forming the collection which I have ventured to call the Gramm, ar of Ornament, has been to select a few of the most prominent types in certain styles closely connected with each other, and in which certain general laws appeared to reign independently of the individual peculiarities of each. I have ventured to hope that, in thus bringing into immediate juxtaposition the many forms of beauty which every style of ornament presents, I might aid in arresting that unfortunate tendency of our time to be content with copying, whilst the fashion lasts, the forms peculiar to any bygone age, without attempting to ascertain, generally completely ignoring, the peculiar circumstances which rendered an ornament beautiful, because it was appropriate, and which as expressive of other wants, when thus transplanted, as entirely fails. It is more than probable that the first result of sending forth to the world E 1 PREFACE. this collection will be seriously to increase this dangerous tendency, and that many will be content to borrow from the past those forms of beauty which have not already been used up ad nauseam. It has been my desire to arrest this tendency, and to awaken a higher ambition. If the student will but endeavour to search out the thoughts which have been expressed in so many different languages, he may assuredly hope to find an ever-gushing fountain in place of a half-filled stagnant reservoir. In the following chapters I have endeavoured to establish these main facts, — First. That whenever any style of ornament commands universal admira- tion, it will always be found to be in accordance with the laws which regulate the distribution of form in nature. Secondly. That however varied the manifestations in accordance with these laws, the leading ideas on which they are based are very few. Thirdly. That the modifications and developments which have taken place from one style to another have been caused by a sudden throwing off of some fixed trammel, which set thought free for a time, till the new idea, like the old, became again fixed, to give birth in its turn to fresh inventions. Lastly. I have endeavoured to show, in the twentieth chapter, that the future progress of Ornamental Art may be best secured by engrafting on the experience of the past the knowledge we may obtain by a return to Nature for fresh inspiration. To attempt to build up theories of art, or to form a style, independently of the past, would be an act of supreme folly. It would be at once to reject the experiences and accumulated knowledge of thousands of years. On the contrary, we should regard as our inheritance all the successful labours of the past, not blindly following them, but employing them simply as guides to find the true path. In taking leave of the subject, and finally surrendering it to the judgment of the public, I am fully aware that the collection is very far from being complete : there are many gaps which each artist, however, may readily fill up for himself. My chief aim, to place side by side types of such styles as might best serve 2 PREFACE. as landmarks and aids to the student in his onward path, has, I trust, been fulfilled. It remains for me to offer my acknowledgment to all those friends who have kindly assisted me in the undertaking. In the formation of the Egyptian Collection I received much valuable assistance from Mr. J. Bonomi, and from Mr. James Wild, who has also con- tributed the materials for the Arabian Collection, his long residence in Cairo having afforded him the opportunity of forming a very large collection of Cairean Ornament, of which the portion contained in this work can give but an imperfect idea, and which I trust he may some day be encouraged to publish in a complete form. I am indebted to Mr. T. T. Bury for the plate of Stained Glass. From Mr. C. J. Richardson I obtained the principal portion of the materials of the Elizabethan Collection; from Mr. J. B. Waring, those of the Byzantine, and I am also indebted to him for the very valuable essays on Byzantine and Eliza- bethan Ornament. Mr. J. O. Westwood having directed especial attention to the Ornament of the Celtic races, has assisted in the Celtic Collection, and written the very remarkable history and exposition of the style. Mr. C. Dresser, of Marlborough House, has provided the interesting plate No. 8 of the twentieth chapter, exhibiting the geometrical arrangement of natural flowers. My colleague at the Crystal Palace, M. Dighy M^yatt, has enriched the work with his admirable essays on the Ornament of the Renaissance and the Italian periods. Whenever the material has been gathered from published sources, it has been acknowledged in the body of the work. The remainder of the drawings have been chiefly executed by my pupils, Mr. Albert Warren and Mr. Charles Auhert, who, with Mr. Stubbs, have reduced the whole of the original drawings, and prepared them for publication. 3 PREFACE. The drawing upon stone of the wdiole collection was entrusted to the care of Mr. Francis Bedford, who, \Yith his able assistants, Messrs. H. Fielding, W. R. Tymms, A. Warren, and S. Sedgfield, with occasional help, have executed the One Hundred Plates in less than one year. My special thanks are due to Mr. Bedford for the care and anxiety which he has evinced, quite regardless of all personal consideration, to render this work as perfect as the advanced state of chromolithography demanded ; and I feel persuaded that his valuable services will be fully recognised by all in any way acquainted with the difficulties and uncertainties of this process. Messrs. Day and Son, the enterprising publishers, and at the same time the printers of the work, have put forth all their strength ; and notwith- standing the care required, and the vast amount of printing to be performed, the resources of their establishment have enabled them, not only to deliver the work with perfect regularity to the Subscribers, but even to complete it before the appointed time. OWEN JONES. 9 Argyll Place, Dec. 15, 185(). 4 GENEKAL PEINCIPLES IN THE ARRANGEMENT OF FORM AND COLOUR, IN ARCHITECTURE AND THE DECORATIVE ARTS, WHICH ARE ADVOCATED THROUGHOUT THIS WORK. General principles. Peoposition 1. The Decorative Arts arise from, and should properly he attendant upon. Ar- chitecture. Proposition 2. Architecture is the material expression of the -wants, the faculties, and the senti- ments, of the age in -which it is created. Style in Architecture is the peculiar form that expression takes under the influence of climate and materials at command. Peoposition 3. As Architecture, so all works of the Decorative Arts ; should possess fitness, proportion, harmony, the result of all which is repose. Proposition 4. True beauty results from that repose which the mind feels when the eye, the intellect, and the affections, are satisfied from the absence of any want. Proposition 5. Construction should be decorated. Decoration should never be purposely constructed. That which is beautiful is trae ; that which is true must he beautiful. Proposition 6. Beauty of form is produced by lines growing out one from the other in gradual undulations : there are no ex- crescences ; nothing could be removed and leave the design equally good or better. Proposition 7. The general forms being first cared for, these should be subdivided and orna- mented by general lines; the interstices may then be filled in with ornament, which may again be subdivided and en- riched for closer inspection. Proposition 8. All ornament should be based upon a geometrical construction. Proposition 9. As in every perfect work of Archi- tecture a true proportion will be found to reign between all the members which compose it, so throughout the Decorative Arts every assemblage of forms should be arranged on certain definite propor- tions ; the whole and each particular member should be a multiple of some simple unit. On general form. Decoration of the sur- face. On propor- tion. PROPOSITIONS. On harmony and contrast. Distribution. Radiation. Continuity. On the con- ventionality of natural forms. On colour generally. Those proportions will be the most beautiful which it will be most difficult for the eye to detect. Thus the proportion of a double square^ or 4 to 8j will be less beautiful than the more subtle ratio of 5 to 8 ; 3 to 6, than 3 to 7 ; 3 to 9, than 3 to 8; 3 to 4; than 3 to 5. Proposition 10. Harmony of form consists in the proper balancing, and contrast of, the straight, the inclined, and the curved. Peoposition 11. In surface decoration all lines should flow out of a parent stem. Every orna- ment, however distant, should be traced to its branch and root. Orienial practice. Proposition 12. All junctions of curved lines with curved or of curved lines with straight should be tangential to each other. Natural law. Oriental practice in accord- ance with it. Proposition 13. Flowers or other natural objects should not be used as ornaments, but conven- tional representations founded upon them sufficiently suggestive to convey the in- tended image to the mind, without de- stroying the unity of the object they are employed to decorate. Universally obeyed in the best periods of Art, equally violated when Art declines. Proposition 14. Colour is used to assist in the devel- opment of form, and to distinguish objects or parts of objects one from another. Proposition 15. Colour is used to assist light and shade, helping the undulations of form by the proper distribution of the several colours. Proposition 16. These objects are best attained by the use of the primary colours on small sur- faces and in small quantities, balanced and supported by the secondary and ter- tiary colours on the larger masses. Proposition 17. The primary colours should be used on the upper portions of objects, the secondary and tertiary on the lower. Proposition 18. {Field’s Chromatic equivalents.) The primaries of equal intensities will harmonise or neutralise each other, in the proportions of 3 yellow, 5 red, and 8 blue, — integrally as 16. The secondaries in the proportions of 8 orange, 13 purple, 11 green, — integrally as 32. The tertiaries, citrine (compound of orange and green), 19; russet (orange and purple), 21 ; olive (green and purple), 24 ; — integrally as 64. It follows that, — Each secondary being a compound of two primaries is neutralised by the re- maining primary in the same proportions: thus, 8 of orange by 8 of blue, 1 1 of green by 5 of red, 13 of purple by 3 of yellow. Each tertiary being a binary com- pound of two secondaries, is neutralised by the remaining secondary: as, 24 of olive by 8 of orange, 21 of russet by 11 of green, 19 of citrine by 13 of purple. On the pro- portions by which har- mony in colouring is produced 6 PROPOSITIONS. On the con- trasts and .haimonious equivalents of tones, shades, and hues. On the posi- tions the several colours should oc- cupy. Proposition 19. The above supposes the colours to be used in their prismatic intensities, but each colour has a variety of tones when mixed with white, or of shades when mixed with grey or black. When a full colour is contrasted with another of a lower tone, the volume of the latter must be proportionally in- creased. Peopositiow 20. Each colour has a variety of hues, obtained by admixture with other colours, in addition to white, grey, or black : thus we have of yellow, — orange-yellow on the one side, and lemon- yellow on the other; so of red, — scarlet-red, and crimson-red; and of each every variety of tone and shade. When a primary tinged with another primary is contrasted with a secondary, the secondary must have a hue of the third primary. PnoposmoN 21. In using the primary colours on moulded surfaces, we should place blue, which retires, on the concave surfaces ; yellow, which advances, on the convex; and red, the intermediate colour, on the undersides ; separating the colours by white on the vertical planes. When the proportions required by Proposition 18 cannot be obtained, we may procure the balance by a change in the colours themselves : thus, if the surfaces to be coloured should give too much yellow, we should make the red more crimson and the blue more purple, — i.e. we should take the yellow out of them ; so if the surfaces should give too much blue, we should make the yellow more orange and the red more scarlet. Proposition 22. The various colours should be so blended that the objects coloured, when viewed at a distance, should present a neutralised bloom. Proposition 23. No composition can ever be perfect in which any one of the three primary colours is wanting, either in its natural state or in combination. Proposition 24. When two tones of the same colour Ontiieiaw , of simulta- are juxtaposed, the light colour will moiMoon- appear lighter, and the . dark colour colours, de- rived from darker. Mons. chev ruil. Proposition 25. When two different colours are juxta- posed, they receive a double modification; first, as to their tone (the light colour appearing lighter, and the dark colour appearing darker) ; secondly, as to their hue, each will become tinged with the complementary colour of the other. Proposition 26. Colours on white grounds appear darker ; on black grounds, lighter. Proposition 27. Black grounds suffer when opposed to colours which give a luminous comple- mentary. Proposition 28. Colours should never be allowed to impinge upon each other. Proposition 29. When ornaments in a colour are on a ground of a contrasting colour, the orna- On the means of in- creasing the harmonious effects of ment should be separated from the ground by an edging of lighter colour ; as colours. 1 Observa- 3. r6Cl ,tions derived flower on a green ground should have an"*dSatio™of edging of lighter red. 7 ground may be used without outline; but a dark ornament on a light ground requires to be outlined with a still darker tint. Pkoposition 35. Pkoposition 31. Gold ornaments on any coloured ground should be outlined with black. Proposition 32. Ornaments of any colour may be sepa- rated from grounds of any other colour by edgings of white, gold, or black. Proposition 33. Ornaments in any colour, or in gold, may be used on white or black grounds, without outline or edging. Proposition 34. In “self-tints," tones, or shades of the same colour, a light tint on Imitations, such as the graining of onimita- woods, and of the various coloured marbles, allowable only, when the em- ployment of the thing imitated would not have been inconsistent. Proposition 36. The principles discoverable in the works of the past belong to us ; not so the results. It is taking the end for the means. Proposition 37. No improvement can take place in the Art of the present generation until all classes. Artists, Manufacturers, and the Public, are better educated in Art, and the existence of general principles is more fully recognised. LIST OF PLATES. Chap. I. Ornament of Savage Tribes. Plate. No. 1 1 Ornaments from Articles belonging to various Savage Tribes, exhibited in the United Service and British Museums. 2 2 Ditto ditto ditto. 3 3 Ditto ditto ditto. Chap. II. Egyptian Ornament. 4 1 The Lotus and Papynis, types of Egyptian ornament. 5 2 Ditto ditto, with Feathers and Palm-branches. 6 3 Capitals of Columns, showing the varied applications of the Lotus and Papynis. 6* 3*Ditto ditto ditto. 7 4 Various Cornices, formed by the Pendent Lotus. 8 5 Ornaments from Mummy Cases in the British Museum and the Louvre. 9 6 Geometrical Ornaments from Ceilings of Tombs. 10 7 Ornaments with Curved Lines from Ceilings of Tombs. 11 8 Various Ornaments from Ceilings and Walls of Tombs. Chap. III. Assyrian and Persian Ornament. 12 1 Painted Ornaments from Nineveh. 13 2 Ditto ditto. 14 3 Carved Ornaments from Persepolis, and Sassanian Ornaments from Ispahan and Bi-Sutoun. Chap. IV. Greek Ornament. 15 1 The Various Forms of the Greek Fret. 16 21 17 3 18 4 Ornaments from Greek and Etruscan Vases in the British Museum and the Louvre. 19 5 20 6 21 7. 22 8 Painted Greek Ornaments from the Temples and Tombs in Greece and Sicily. D Arabian Ornament. 1 Arabian Ornaments of the Ninth Century from Cairo. 2 Ditto Thirteenth Century ditto. 3 Ditto ditto ditto. 4 Portion of an Illuminated Copy of the Koran.^^ 5 Mosaics from Walls and Pavements from Houses in Cairo. Turkish Ornament. 1 Ornaments in Relief from Mosques^ Tombs_, and Fountains at Constantinople. 2 Painted Ornaments from the Mosque of Soliman at Constantinople. 3 Decoration of the Dome of the Tomb of Soliman I. at Constantinople. Chap Moresque Ornament from the Alhambra. 1 Varieties of Interlaced Ornaments. 2 Spandrils of Arches. 3 Lozenge Diapers. 41=^= 3*Ditto ditto. 42 4 Square Diapers. 43* 4*Ditto ditto. 43t 4tDitto ditto. 43 5 Mosaics. Chap. XI. Persian Ornament. Ornaments from Persian MSS. in the British Museum. 45 2 Ditto ditto. 46 3 Ditto ditto. 47 4 Prom a Persian Manufacturer's Pattern-Book, South Kensington Museum. 47* 4*Ditto ditto ditto. 48 5 From a Persian MS. South Kensington Museum. 10 LIST OF PLATES. Chap. XII. Indian Ornament. Plate. No. 49 1 Ornaments from Metal-work from the Exhibition of 1851. 50 2 51 3 52 4 53 5 53* 5* 54 6 54* 6* 65 7 Ornaments from Woven and Embroidered Fabrics and Painted Boxes exhibited at Paris in 1855. Ditto from Embroidered and Woven Fabrics and Paintings on Vases exhibited in the Indian Collection in 1851, now at South Kensington Museum. 'i Specimens of Painted Lacquer-work from the Collection at the India-Hoase. Chap. XIII. Hiadoo Ornament. 56 1 Ornaments from a Statue at the Asiatic Society’s House. 57 2 From the Collection at the Crystal Palace, Sydenham. 58 3 From the Collection at the India House. Chap. XIV. Chinese Ornament. 59 60 61 63 t' 2 3 . 4 Chinese Ornaments painted on Porcelain, and on Wood, and from Woven Fabrics. Conventional Renderings of Fruits and Flowers. Chap. XV. Celtic Ornament. 63 1 Lapidary Ornamentation. 64 2 Interlaced Styles. 65 3 Spii’al, Diagonal, Zoomorphic, and later Anglo-Saxon Ornament. Chap. XVI. Mediaeval Ornament. 66 1 Conventional Leaves and Flowers from Illuminated MSS. 67 2 Borders from Illuminated MSS. and Paintings. 67* 2 * Ditto ditto ditto. 68 3 Diapers from Illuminated MSS. and Backgrounds of Pictures. 69 4 Stained Glass of various periods. 69* 4* Ditto ditto. 70 5 Encaustic Tiles, ditto. Illuminated MSS. 71 1 Portions of Illuminated MSS. of the Twelfth and Thirteenth Centuries. 72 2 Ditto ditto Thirteenth and Fourteenth ditto. 73 3 Ditto ditto Fourteenth and Fifteenth ditto. 11 LIST OF PLATES. Plate. 74 75 76 77 78 79 80 81 83 83 1 84 2 85 3 86 1 86 * 1 87 2 88 3 89 4 90 5 91 1 92 2 93 3 94 4 96 5 96 6 97 7 98 8 99 9 100 10 Chap. XVII. Renaissance Ornament. 0 . J Renaissance Ornaments in Relief, from Photographs taken from Casts in the Crystal Palace, Syden- ' I ham. r ; Enamels from the Louvre and H6tel Cluny. 1 Ornaments from Potteiy at South Kensington Museum, i Ditto from ditto, H6tel Cluny and the Louvre. Ditto ditto, ditto. jomaments from Stone and Wood from the Collections of the Louvre and Hotel Cluny. Chap. XVIIL Elizabethan Ornament. jvarious Ornaments in Belief from tlie Time of Henry Till, to that of Charles II. Painted Ornaments and Ornaments on Woven Eabrics, ditto. Chap. XIX. Italian Ornament. Pilasters and Ornaments from the Loggie of the Vatican, reduced from the full-size Paintings at South Kensington Museum. 'Ditto ditto ditto ditto. Ornaments from the Palazzo Ducale, Mantua. Ditto from the Palazzo Ducale and the Church of St. Andrea, Mantua. Ditto from the Palazzo del Te, Mantua. Ornaments from Printed Books. Chap. XX. Leaves and Flowers from Nature. Horse-chestnut leaves. Vine leaves. Ivy leaves. leaves of the Oak, Fig-tree, Maple, White Bryony, Laurel, and Bay-tree. Leaves of the Vine, Hollyoak, Turkey Oak, and Laburnum. Wild Rose, Ivy, and Blackberry. Hawthorn, Yew, Ivy, and Strawberry-tree. Various Flowers in Plan and Elevation. Honeysuckle and Convolvulus. Passion Flowers. 13 Chapter I. — Plates 1, 2, 3. ORNAMENT OF SAVAGE TRIBES. PLATE 1. 1. Cloth. Otaheite. — United Service Museum. 2. Matting from Tongotabu, Friendly Islands. 3. Cloth. Otaheite.— U. S. M. 4. Cloth. Sandwich Islands. — U. 3- M. 5-8. Cloths. Sandwich Islands. — British Museum. 9. Cloth Matting from Tongotabn, Friendly Islands. 10. Cloth. Otaheite.— U. S. M. 11. Cloth. Sandwich Islands. — B. M. 12. Cloth. 13. Cloth made from Paper Mulberry, Feejee Islands.— B. M. PLATE 11. 1. South America. — United Service Museum. 9, 10. Tahiti. Adze. U. S. M. 2. Sandwich Islands. U. S. M. 11, 12. Friendly Islands. Drum. U. S. M. 3. Owhyhee. U. S. M. 13, 14. Tahiti. Adze, U. S. M. 4. New Hebrides. Inlaid Shield. U. S. M. 15. Sandwich Islands. U. S. M. 5. Sandwich Islands, U. S. M. 16, 17. New Zealand. U. S. M. 6. South Sea Islands. U. S. M. 18-20. Sandwich Islands. U. S. M. 7, 8. Sandwich Islands. U. S. M. PLATE III. Owhyhee. Club. — United Service Museum. Sandwich Islands. Club. U. S. M. New Zealand. Patoo-Patoo. U. S. M. Tahiti. Adze. U. S. M. New Zealand. Paddle. U. S. M. 6. New Zealand. Pajee, or War Club. — U. S. M. 7. South Sea Isles. War Club. U. S. M. 8. Handle, full size of Fig. 5. U. S. M. 9. Feejee Islands. Club. U. S. M. From the universal testimony of travellers it would appear, that there is scarcely a people, in however early a stage of civilisation, with whom the desire for ornament is not a strong instinct. The desire is absent in none, and it grows and increases with all in the ratio of their progress in civilisation. Man appears everywhere impressed with the beauties of Nature which surround him, and seeks to imitate to the extent of his power the works of the Creator. Man’s earliest ambition is to create. To this feeling must be ascribed the tattooing of the human face and body, resorted to by the savage to increase the expression by which he seeks to strike terror on his enemies or rivals, or to create what appears to him a new beauty.* As we advance higher, from the * The tattooing on the head which we introduce from the Museum at Chester is very retoarkahle, as showing that in this 'very barbarous practice the principles of the very highest ornamental art are manifest, every line upon the face is the best adapted to develope the natural features. £ 13 OENAMENT OF SAVAGE TRIBES. decoration of the rude tent or wigwam to the sublime works of a Phidias and Praxiteles, the same feeling is everywhere apparent: the highest ambition is still to create, to stamp on this earth the impress of an individual mind. From time to time a mind stronger than those around will impress itself on a generation, and carry with it a host of others of less power following in the same track, yet never so closely as to destroy the individual ambition to create ; lienee the cause of styles, and of the modifications of styles. The efforts of a people in an eai-ly stage of civilisation are like those of children, though presenting a want of power, they possess a grace and nalveU rarely found in mid-age, and never in manhood’s decline. It is equally so in the infancy of any art. Cimabue and G-iotto have not the material charm of Raphael or the manly power of Michael Angelo, but surpass them both in grace and earnest truth. The very command of means leads to their abuse : when Art struggles, it succeeds ; when revelling in its own successes, it as signally fails. The pleasure we receive in con- templating the rude attempts at ornament of the most savage tribes arises from our appreciation of a difficulty accomplished; we are at once charmed by the evidence of the intention, and surprised at the simple and ingenious process by which the result is obtained. In fact, what we seek in every work of Art, whether it be humble or pretentious, is the evidence of mind,— the evidence of that desire to create to which we have referred, and which all, feeling a natural instinct within them, are satisfied with when they find it developed in others. It is Female Head from Kerr Zealand, to the Muaenm, Cheater. strange, but SO it IS, that this evidence of mind Will be more readily found in the rude attempts at ornament of a savage tribe than in the innumerable productions of a highly-advanced civilisation. Individuality decreases in the ratio of the power of production. When Art is manufactured by combined effort, not originated by individual effort, we fail to recognise those true instincts which constitute its greatest charm. Plate I. The ornaments on this Plate are from portions of clothing made chiefly from the bark of trees. Patterns Nos. 2 and 9 are from a dress brought by Mr. Oswald Brierly from Tongotabu, the principal of the Friendly Island group. It is made from thin sheets of the inner rind of the bark of a species of hibiscus, beaten out and united together so as to form one long parallelogram of cloth, which being wrapped many times round the body as a petticoat, and leaving the chest, arms, and shoulders bare, forms the only dress of the natives. Nothing, therefore, can be more primitive, and yet the arrangement of the pattern shows the most refined taste and skiU. No. 9 is the border on the edge of the cloth; with the same limited means of production, it would be difficult to improve upon it. The patterns are formed by small wooden stamps, and although the work is somewhat rude and irregular in execution, the intention is everywhere apparent ; and we are at once struck with the skilful balancing of the masses, and the judicious correction of the tendency of the eye to run in any one direction by opposing to them lines having an opposite tendency. 14 ORNAMENT OF SAVAGE TRIBES. WTien Mr. Brierly visited the island one woman was the designer of all the patterns in use there, aud for every new pattern she designed she received as a reward a certain number of yards of cloth. The pattern No. 2, from the same place, is equally an admirable lesson in composition which we may derive from an artist of a savage tribe. Nothing can be more judicious than the general arrange- ment of the four squares and the four red spots. Without the red spots on the yellow ground there would have been a great want of repose in the general arrangement; without the red lines round the red spots to carry the red through the yellow, it would have been still imperfect. Had the small red triangles turned outwards instead of inwards, the repose of the pattern would again have been lost, and the effect produced on the eye would have been that of squinting; as it is, the eye is centred in each square, and centred in each group by the red spots round the centre square. The stamps which form the pattern are very simple, each triangle A and each leaf how readily the possession of a simple tool, even 4Bkby the most stinctive observation of the forms in which all the works of Nature creation of all the geometrical arrangements of form with which we are acquainted. On the upper left-hand corner of pattern No. 2, the eight-pointed star is formed by eight applications of the same tool ; as also the black flower with sixteen pointing inwards f and sixteen pointing outwards. 1 The most com- plicated patterns of the Byzantine, Arabian, I and Moresque mosaics would be generated by the same means. The secret of success in all ornament is the pro- duction of a broad general effect by the repetition of a few simple elements; variety should rather be sought in the arrangement of the several portions of a design, than in the multiplicity of varied forms. The stamping of patterns on the coverings of the body, when either of skins of animals or material such as this, would be the first stage towards ornament after the tattooing of the body by an analogous process. In both there would remain a greater variety and individuality than in subse- quent processes, which would become more mechanical. The first notions of weaving which would be given by the plaiting of straws or strips of bark, instead of using them as thin sheets, would have equally the same result of gradually forming the mind to an appreciation of the proper disposition of masses : the eye of the savage, accustomed only to look upon Nature’s harmonies, would readily enter into the perception of the true balance both of form and colour ; in point of fact, we find that it is so, that in savage ornament the true balance of both is always maintained. After the formation of ornament by stamping and weaving, would naturally follow the desire of forming ornament in relief or carving. The weapons for defence or the chase would first attract attention. The most skilful and the bravest would desire to be distinguished from their fellows by the possession of weapons, not only more useful, but more beautiful. The shape best fitted Plaited straw from the sandwich islands, for the purpose having been found by experience, the enriching of the surface by carving would naturally follow ; and the eye, already accustomed to the geometrical forms produced by weaving, the hand would seek to imitate them by a similar repetition of cuts of the knife. The ornaments on Plate II. show this instinct very fully. They are executed with the utmost precision, and exhibit great taste and judgment in the distribution of the masses. Nos. 11 and 12 are interesting, as showing how much this taste and skill may exist in the formation of geometrical patterns, whilst those resulting from curved lines, and the human form more especially, remain in the very first stage. ♦ being a single stamp, we thus see uncultivated, if guided by an in- are arranged, would lead to the 15 !i3f- to ORNAMENT OF SAVAGE TRIBES. The ornaments in the woodcuts below and at the side show a far higher advance in the distribution of curved lines, the twisted rope forming the type as it naturally would be of all curved lines in ornament. The uniting of two strands for additional strength would early accustom the eye to the spiral line, and we always find this form side by side with Head of Cauoe, New Guinea. geometrical patterns formed by the interlacing of equal lines in the orna- ment of every savage tribe, and retained in the more advanced art of every civilised nation. Head of Canoe, New Guinea. The ornament of a savage tribe, being the result of a natural instinct, is necessarily always true to its purpose ; whilst in much of the ornament of civilised nations, the first impulse which generated received forms being enfeebled by constant repetition, the ornament is often- times misapplied, and instead of first seeking the most convenient form and adding beauty, all beauty is destroyed, because all fitness, by superadding ornament to iU-oontrived form. If we would return to a more healthy, condition, we must even be as little children or as savages ; we must get rid of the acquired and artificial, and return to and develope natural instincts. The beautiful New Zealand paddle. Nos. 6-8, on Plate III., would rival works of the highest civilisa- tion : there is not a line upon its surface misapplied. The general shape is most elegant, and the decoration everywhere the best adapted to develope the form. A modem manufacturer, with his 16 ORNAMENT OF SAVAGE TRIBES. stripes and plaids, would have continued the bands or rings round the handle across the blade. The New Zealander’s instinct taught him better. He desired not only that his paddle should be strong, but should appear so, and his ornament is so disposed to give an appear- ance of additional strength to what it would have had if the surface had remained undecorated. The centre band in the length of the blade Handle of a Paddle. — B. M, is continued round on the other side, binding together the border on the edge, which itself fixes all the other bands. Had these bands run out like the centre one, they would have appeared to slip off. The centre one was the only one that could do so without disturbing the repose. The swelling form of the handle where additional weight was required is most beautifully contrived, and the springing of the swell is well defined by the bolder pattern of the rings.* Club, Eastern Archipelago. * Captain Cook and other voyagers repeatedly notice the taste and ingenuity of the islanders of the Pacific and South Seas : instancing especially cloths, painted “ in such an endless variety of figures that one might sxxppose they borrowed their patterns from a mercer’s shop in which the most elegant productions of China and Europe are collected, besides some original patterns of their own.” The “ thousand different patterns ” of their basket-work, their mats, and the fancy displayed in their rich carvings and inlaid shell-work, are, likewise, constantly mentioned. See The Three Voyages of Captain Cook, 3 vols. Lend. 1841— i3j Dumont D’Urville’s Voyage au Po/c Surf, 8vo. Paris, 1841 ; Ditto, rf’.ffisfoire, fol.; Prichard's Natural History of Man, G. W. Earle’s .Jfaijwe Races of Indian Archipelago, Lond. 1853 ; Kerr’s General History and Collection of Voyages and Travels, London, 1811-17. F 17 |r 5 k ^ ► • 1 ' .is ■*« ' * .<■0^ " ■ !«■ I'fe ''iik i'y - .,,-,r/W. j^;,c-^ ; G,. , v — C3 V4;: t;' .-■ ■ ■ \S';. , ,•' : T*- .^T '!/'■►. ••■ -T* #^• v..-> f ■ f" . ■■ t- -. ■:. -r: m V t'-'- 1 -. -i;- ' A^L. !? ■ i: ^ •#^' '/*r''i 3 j ,1 -- . £ i'- -.J i;iu ;>rv. •; :■ . .'i.'J ^ » U ' *, < -irf-i.* .Jf ii . . ’ j»’r <^T •• ■■ - V. V ai WILDE STAMME TAFEL 1, SAVAGE TRIBES NM. TRIBUS SAIIYAGES PL 1. A w- Ar 1 V 1 ^ 1 m E\ 1 mHH ►♦♦♦♦♦♦< ♦♦♦♦♦ t lllllll ■ I IIIIIIIIIKI ■ ■ 111111111111 ■ ■ lllillllll ■ ■ I MR I ^ -V'-' ' . :f^- •‘v: 'm - ■' -/■^fl Ojg^.* 0 . >■ ■ r I"..,". ' • •' ■' ' '’v'-J-vV.. '•■ .< I<|« •• • > .IV ' ■ • ''■ ^ ■'■fc'4'*«»i< ■ ■'« "'. '■■•'■■-•' ■■ , '■ ■■ ■>' *' ■/ ':" ' ., v'- 'i^ ■ ■ .'■;•'■'■■■■ ..•^.s . •■ ' ..j-- ' . .,.w' -v -^v "* * ■ « . . r' .-- ■' V. •' • . :. • • . ■' >'v ->f ~ ... •■ .i^j# .. . '. J'fk . ■•I'' ’.'. . - . i. *v • ■ 'tr'' • ■» «; ; — . ... ■7 ’:V;k ' • ■ ,' '.'■' .. ■ -.. ■ STAMM E SAVAGE TRIBES N"E TRIBUS SAUYAGE 3S!S Gold and enamelled Vases in the Form of the Lotus. 16.) 16. A Rudder Oar decorated with the Lotus and the Eye, representing the Divinity. 1 7. Ditto, another variety. 18.1 f Boats made of Papyrus Plants bound together. PLATE VI. 1. Capital of the large Columns of the Temple of Luxor, Thebes, of the time of Amunoph III., 1250 B.C., ac- cording to Sharpe. It represents the full-blown Papyrus, and around it Papyri and Lotus Buds alter- nating. 2. Capital of the smaller Columns of the Memnonium, Thebes, b.c. 1200. Represents a single Bud of the Papyrus decorated with the coloured pendent Fascise that are seen in the painted representations of Columns of Plate rV. Nos. 5, 6, 12. 3. Capital of the smaller Columns of the Temple of Luxor, B.c. 1250. Representing eight Buds of the Papyrus bound together, and adorned with pendent and coloured Fasciae. 19 EGYPTIAN ORNAMENT. Capital from the unfinished hypjethral Temple in the Island of Philae. Roman period, b.c. 140. Composed of the Papyrus Plant in three stages of growth, and arranged in three tiers : the first composed of four full-blown and four large expanding Papyri ; the second tier, of eight smaller expanding flowers ; and the third tier, of sixteen buds : making in all a bundle of thirty-two plants. The stem of each plant may be traced, by the size and colour of its stalk, down to the horizontal bands or fascise. See Plate IV. Nos. 5, 6, 12. Capital from the Temple at Koom-Ombos. The full- grown Papyrus surrounded by various flowers. Capital from the principal Temple, Philse. Representing two tiers of the Papyrus, in three stages of growith. The first tier composed of eight plants, four full-blown and four expanding ; the second tier composed of eight buds : making sixteen plants. In this capital the circular form is not disturbed, as in No. 11. Capital from the unfinished hypsethral Temple, Philge. Composed of three tiers of the Papyrus Plant in three stages of growth. The first tier has eight full- blown and eight expanding plants ; the second tier, six- teen expanding flowers ; and the third tier, thirty-two buds of the Papyrus : in all, sixty-four plants. The stem of each plant is distinguished by its size and colour, and continued down to the horizontal bands which bind them together round the shaft. Capital from the Portico of Edfu, b.c. 146. Represents the Palm-tree, with nine branches, or faces. The horizontal fasciae of the Palm-tree Capital differ from the fasciae of all the other capitals, inasmuch as there is always a pendent loop. PLATE VP. Capital from a Temple in the Oasis of Thebes. Repre- senting a collection of Aquatic Plants, with triangular Stalks tied round a single full-blown Papyrus. Capital from the Portico of Edfu, b.c. 145, of similar structure to No. 4. Capital from the principal Temple in the Island of Philse, B.C. 106. The full-blown Papyrus surrounded by the same flower in various stages of growth. Capital from a Temple in the Oasis of Thebes. Capital from the Colonnade of the Island of Philse. Representing sixteen Lotus Flowers bound together in three tiers. Shown in elevation. The Capital No. 8 seen in Perspective. 10. Capital from a Temple in the Oasis of Thebes. Re- presenting eight Lotus Flowers bound together in two tiers. 15. Capital from the unfinished hyp^thral Temple, Phil». Composed of the Papyrus in two stages of grcwth, arranged in three tiers. The first composed of four full-blown and four expanding flowers ; the second tier, of eight smaller, full-blown ; and the third tier, of sixteen, still smaller. 17. Capital of the Orseco-Egyptian form, but of tlie Roman period. Very remarkable, as showing the Egyptian and Greek elements combined, viz. the Papyrus in two stages of growth, with the Acanthus leaf and the tendrils of the Honeysuckle. PLATE VIL 7, - 8 , 9.. 10 . 11 . 12 . 13. 14. 15. 16. 17 , 18, 19. 20 . 21 . Ornament on the top of the Walls of a Tomb at Beni- hassan. Ditto ditto. Ditto, from Karnac, Thebes. Ditto, from Gouma, Thebes. Ditto, from Sakhara. Decoration of the Torus moulding of some of the early Tombs in the neighbourhood of the Pyramids of Giza. ^From a wooden Sarcophagus. From the Tombs, El E[ab. From the Tombs, Benihassan. From the Tombs, Gourna. Ditto. Ditto. From a Necklace. From the Wall of a Tomb, Gourna, immediately under the Ceiling. Portions of a Necklace. From the Wall of a Tomb. From a Necklace. 20 22. From the upper part of the Wall of a Tomb, Sakhara. 23. Ditto, at Thebes. 24. From a Necklace. 25. From the Wall of a Tomb, Gourna. 26. From a Sarcophagus. 27. From the Wall of a Tomb. 28. From a Sarcophagus. 29. From the upper part of a Picture. 30. Arrangement of Lines from dados. 31. From a Sarcophagus in the Louvre. 32. From the Wall of a Tomb, Gourna, representing the Lotus, in plan as well as in elevation. 33. From a Ceiling at Medinet Haboo. 34. Arrangement of Lines from dados, in Tombs. Nos. 1-5, 10, 11, always occur on vertical surfaces, and on the upper part of the walls of tombs and temples. Nos. 7-9, 12, 14, 18, 20, are all derived from the same elements, viz. the lotus in a pendent position, with a bunch of grapes interven- ing. This very constant Egyptian ornament in some of its forms so much resembles the Greek moulding, usually termed the egg-and-tongue, or egg-and-dart moulding, that we can hardly doubt that the Greek moulding, was derived from this source. Nos. 13, 15, 24, 32, exhibit another element of Egyptian ornamentation derived from the separated leaves of the Lotus. EGYPTIAN ORNAMENT. PLATE VIII. The whole of the Ornaments on this Plate are from Mummy- cases in the British Museum and the Louvre, and, like those of the last Plate, are mostly composed of the Lotus-flower and single leaves of the same plant. In No. 2, above the Lotus-leaves, is a white ornament on a black ground, very common in the tombs, suggested by the interwoven strands of a rope ; and in No. 7 we have the chequered pattern, one of the earliest ornaments, evidently derived from the weaving to- gether of different-coloured strands. In the lower part of No. 18 we have another very common orna- ment, derived from feathers. PLATE IX. The Ornaments on this Plate are taken from Paintings on Tombs in various parts of Egypt, from original Draw- ings. They are chiefly patterns that could be pro- duced by the loom, and a single glance will show that this is doubtless the origin of most of them. 1-8 are representations of Mats on which the kings stand. They were evidently formed of interwoven straws of different colours. The transition from this state to the formation of patterns, such as 9-12, 17-19, 21, would be very rapid, and they are most probably only repro- ductions of woven articles of daily use. Nos. 9 and 10 may have suggested the fret to the Greeks, unless they arrived at it themselves by a similar process. 20 is from a Ceiling of a Tomb at Gourna. It represents the Trellis-work of a Garden Walk, covered with a Vine. It is by no means an uncommon ornament for the curved ceilings of small tombs, and usually occupies the whole ceiling of each excavation at the period of the nineteenth dynasty, 21-23 are derived from Mummy-cases in the Louvre, of a late period. PLATE X. 1-5. From Mummy-cases in the Louvre, of a late period. Geometrical arrangements of the single Lotus-leaf. 6. From a Tomb at Thebes. Each circle is formed of four Lotus-flowers and four Buds, the inteimediate star probably intended for four Lotus-leaves. 7, From a Tomb at Thebes. 8, 9, From a Mummy-case, 10-24 are from Ceilings of Tombs in various parts of Egypt. In Nos. 10, 13-16, 18-23, are various examples of an ornament representing the unvTinding of a pile of rope, which may have given the first suggestion of the volute. In No. 24 the continuous blue line is evidently from the same type. PLATE XL 1, 4, 6, 7, are from Tombs at Thebes, and are further examples of the Rope Ornament given in the last Plate. Nos. 2 and 3 are varieties of arrangements of Stars, very common on the ceilings both of tombs and temples. No 2 is formed on squares, No, 3 on equilateral triangles. 9. From a Mummy-case. 10. From the Embroidery on a King’s Robe. 11-16 are varieties of Borders from Paintings in Tombs. 17. From the Dress of a Figure in one of the Royal Tombs of Biban el Moluk. It represents the Scales of the Armour worn by the Heroes and Gods of Egypt. 18-20 are similar, and most probably were suggested by the feathers of birds. 21. Ornament on the Dress of the god Amun, from Aboo- simbel. 22. From a Fragment in the Louvre. 23. Dado from the Tomb of Ramses, Biban el Moluk, pro- bably representing, in diagram, a Papyrus -grove, as it occupies a similar position to those dados of a later period which were formed of buds and flowers of the papyrus. 24. From a very ancient Tomb at Giza, opened by Dr. Lepsius. The upper part represents the usual Egyptian torus ; the lower portion is from the dado of the same tomb, and shows that the practice of imitating grained woods in painting is of the highest antiquity. G 21 EGYPTIAN ORNAMENT. The Architecture of Egypt has this peculiarity over all other styles, that the more ancient the monu- ment tlie more perfect is the art. All the remains with which we are acquainted exhibit Egyptian Art in a state of decline. Monuments erected two thousand years before the Christian era are formed from the ruins of still more ancient and more perfect buildings. We are thus carried back to a period too remote from our time to enable us to discover any traces of its origin ; and whilst we can trace in direct succession the G-reek, the Eoman, the Byzantine, with its offshoots, the Arabian, the Moresque, and the Gothic, from this great parent, we must believe the architecture of Egypt to be a pure original style, which arose with civilisation in Central Africa,* passed through countless ages, to the culminating point of perfection and the state of decline in which we see it. Inferior as this state doubtless is to the unknown perfection of Egyptian Art, it is far beyond all that followed after ; the Egyptians are inferior only to themselves. In all other styles we can trace a rapid ascent from infancy, founded on some bygone style, to a culminating point of perfection, when the foreign influence was modified or discarded, to a period of slow, lingering decline, feeding on its owm elements. In the Egyptian we have no traces of infancy or of any foreign influence ; and we must, therefore, believe that they went for inspiration direct from nature. This view is strengthened when we come to consider more especially the ornament of Eg3q)t ; the types are few and natural types, the representation is but slightly removed from the type. The later we descend in art, the more and more do we find original types receded from ; till, in much ornament, such as the Arabian and Moresque, it is difficult to discover the original type from which the ornament has been by successive mental efforts developed. The lotus and papyrus, growing on the banks of their river, symbolising the food for the body and mind; the feathers of rare birds, which were carried before the king as emblems of sovereignty; the palm-branch, with the twisted cord made from its stems ; these are the few types which form the basis of that immense variety of ornament with which the Egyptians decorated the temples of their gods, the palaces of their kings, the covering of their persons, their articles of luxury or of more modest daily use, from the wooden spoon which fed them to the boat which carried their similarly adorned embalmed bodies across the Nile to their last home in the valley of the dead. Following these types as they did in a manner so nearly allied to their natural form, they could hardly fail to observe the same laws which the works of nature ever display; and we find, therefore, that Egyptian ornament, however conventionalised, is always true. We are never shocked by any misapplication or violation of a natural principle. On the other hand, they never, by a too servile imitation of the type, destroyed the consistency of the representation. A lotus carved in stone, forming a graceful termination to a column, or painted on the walls as an offering to their gods, was never such a one as might be plucked, but an architectural * In the British Museum may be seen a east of a bas-relief from Kalabshee in Nubia, representing the conquests of Eamses II. over a black people, supposed to be Ethiopians. It is very remarkable, that amongst the presents which these people are represented as bringing with them as a tribute to the King, besides the leopard skins and rare auiroals, ivory, gold, and other products of the country, there are three ivory carved chairs precisely Similar to that on which the King sits to receive them ; from which it would appear that these highly.elaborated articles of luxury were derived by the Egj-ptians from the interior of Africa. 22 EGYPTIAN ORNAMENT. representation ; in eitlier case the best adapted for the purpose it had to fill, sufficiently resembling the type to call forth in the beholder the poetic idea which it was sought to supply, without shocking his feeling of consistency. Egyptian ornament is of three kinds : that which is constructive, or forming part of the monument itself, of which it is the outward and graceful covering of the skeleton within; that which is representative, but at the same time conventionally rendered ; and that which is simply decorative. In all cases it was symbolic, and, as we have observed, formed on some few types, which were but slightly changed during the whole period of Egyptian civilisation. Of the first kind, viz. constructive ornament, are the decorations of the means of support and the crowning members of the walls. The column only a few feet high, or one forty or sixty feet, as at Luxor and Karnac, was an enlarged papyrus plant : the base representing the root ; the shaft, the stalk ; and the capital, the full-blown flower, surrounded by a bouquet of smaller plants (No. 1, Plate VI.), tied together by bands. Not only did a series of columns represent a grove of papyri, but each column was in itself a grove ; and at No. 17 of Plate IV. we have a representation of a grove of papyri in various stages of growth, which would only have to be assembled as they stand, and be tied round with a string, and we should have the Egyptian shaft and its highly-ornamental capital ; and further, we have in Nos. 5, 6, 10, 11, 12, on Plate IV., painted representations of columns forming parts of temples, in which the original idea is unmistakably portrayed. We may imagine it the custom of the Egyptians in early times to decorate the wooden posts of their primitive temples with their native flowers tied round them ; and this custom, when their art took a more permanent character, became solidified in their monuments of stone. These forms, once sacred, their religious laws forbade a change; but a single glance, however, at Plates VI. and VI.*, will show how little this possession of one leading idea resulted in uniformity. The lotus and papyrus form the type of fifteen of the capitals we have selected for illustration ; yet how ingeniously varied, and what a lesson do they teach us ! From the Greeks to our own time the world has been content with the acanthus leaf arranged round a bell for the capitals of columns of all architecture called classic, differing only in the more or less perfection of the modelling of the leaves, or the graceful or otherwise proportions of the bell; a modification in plan has but rarely been attempted. And this it was that opened the way to so much development in the Egyptian capital; beginning with the circle, they surrounded it with four, eight, and sixteen other circles. If the same change were attempted with the Corinthian capital, it could not fail to produce an entirely new order of forms whilst still retaining the idea of applying the acanthus leaf to the surface of a bell-shaped vase. The shaft of the Egyptian column, when circular, was made to retain the idea of the triangular shape of the papyrus stalk, by three raised lines, which divided its circumference into three equal portions; when the column was formed by a union of four or eight shafts bound together, these had each a sharp arris on their outer face with the same intention. The crowning member or comice of an Egyptian building was decorated with feathers, which appear to ha^e been an emblem of sovereignty ; whilst in the centre was the winged globe, emblem of divinity. The second kind of Egyptian ornament results from the conventional representation of actual things on the walls of the temples and tombs ; and here again, in the representations of offerings to the gods or of the various articles of daily use, in the paintings of actual scenes of their domestic life, every flower or other object is portrayed, not as a reality, but as an ideal representation. It is at the same time the record of a fact and an architectural decoration, to which even their hieroglyphical writing, explanatory of the scene, by its symmetrical arrangement added effect. In No. 4, on Plate TV., we have an example in the representation of three papyrus plants and three lotus flowers, with two buds, in the hand of a king as an offering to the gods. The arrangement is symmetrical and graceful, and 23 EGYPTIAN ORNAMENT. we here see that the Egyptians, in thus conventionally rendering the lotus and papyrus, instiactively obeyed the law which we find everywhere in the leaves of plants, viz. the radiation of the leaves, and all veins on the leaves, in graceful curves from the parent stem ; and not only do they follow this law in the drawing of the individual flower, but also in the grouping of several flowers together, as may be seen, not only in No. 4, but also in their representation of plants growing in the desert. Nos. 16 and 18 of the same plate, and in No. 13. In Nos. 9 and 10 of Plate V. they learned the same lesson from the feather, another type of ornament (11 and 12, Plate V.); the same instinct is again at work at Nos. 4 and 5, where the type is one of the many forms of palm-trees so common in the country. The third kind of Egyptian ornament, viz. that which is simply decorative, or which appears so to our eyes, but which had doubtless its own laws and reasons for its application, although they are not so apparent to us. Plates VIII., IX., X., XI., are devoted to this class of ornament, and are from paintings on tombs, dresses, utensils, and sarcophagi. They are all distinguished by graceful symmetry and perfect distribution. The variety that can be produced by the few simple types we have referred to is very remarkable. On Plate IX. are patterns of ceilings, and appear to be reproductions of woven patterns. Side by side with the conventional rendering of actual things, the first attempts of every people to produce works of ornament take this direction. The early necessity of plaiting together straw or bark of trees, for the formation of articles of clothing, the covering of their rude dwelling, or the ground on which they reposed, induced the employment at first of straws and bark of different natural colours, to be afterwards replaced by artificial dyes, which gave the first idea, not only of ornament, but of geome- trical arrangement. Nos. 1-^, Plate IX., are from Egyptian paintings, representing mats whereon the king stands ; whilst. Nos. 6 and 7 are from the ceilings of tombs, which evidently represent tents covered by mats. Nos. 9, 10, 12, show how readily the meander or G-reek fret was produced by the same means. The universality of this ornament in every style of architecture, and to be found in some shape or other amongst the first attempts of ornament of every savage tribe, is an additional proof of their having had a similar origin. The formation of patterns by the equal division of similar lines, as by weaving, would give to a rising people the first notions of symmetry, arrangement, disposition, and the distribution of masses. The Egyptians, in their decoration of large surfaces, never appear to have gone beyond a geometrical arrangement. Flowing lines are very rare, comparatively, and never the motive of the composition, though the germ of even this mode of decoration, the volute form, exists in their rope ornament. (Nos. 10, 13-16, 18—24, on Plate X., and 1, 2, 4, 7, Plate XI.) Here the several coils of rope are subjected to a geometrical arrangement ; but the unrolling of this cord gives the very form which is the source of so much beauty in many subsequent styles. We venture, therefore, to claim for the Egyptian style, that though the oldest, it is, in all that is requisite to constitute a true style of art, the most perfect. The language in which it reveals itself to us may seem foreign, peculiar, formal, and rigid ; but the ideas and the teachings it conveys to us are of the soundest. As we proceed with other styles, we shall see that they approach perfection only so far as they followed, in common with the Egyptians, the true principles to be observed in every flower that grows. Like these favourites of Nature, every ornament should have its perfume; i.e. the reason of its application. It should endeavour to rival the grace of construction, the harmony of its varied forms, and due proportion and subordination of one part to the other found in the model. When we find any of these characteristics wanting in a work of ornament, we may be sure that it belongs to a borrowed style, where the spirit which animated the original work has been lost in the copy. The architecture of the Egyptians is thoroughly polychromatic, — they painted everything ; therefore we have much to learn from them on this head. They dealt in flat tints, and used neither shade 24 nor EGYPTIAN ORNAMENT. shadow, yet found no difficulty in poetically conveying to the mind the identity of the object they desired to represent. They used colour as they did form, conventionally. Compare the representation of the lotus (No. 3, Plate IV.) with the natural flower (No. 1); how charmingly are the characteristics of the natural flower reproduced in the representations I See how the outer leaves are distinguished by a darker green, and the inner protected leaves by a lighter green ; whilst the purple and yellow tones of the inner flower are represented by red leaves floating in a field of yellow, which most completely recalls the yellow glow of the original. We have here Art added to Nature, and derive an additional pleasure in the perception of the mental effort which has produced it. The colours used by the Egyptians were principally red, blue, and yellow, with black and white to define and give distinctiveness to the various colours ; with green used generally, though not universally, as a local colour, such as the green leaves of the lotus. These were, however, indifferently coloured green or blue; blue in the more ancient times, and green during the Ptolemaic period; at which time, also, were added both purple and brown, but with diminished effect. The red also, which is found on the tombs or mummy-cases of the Greek or Roman period, is lower in tone than that of the ancient times ; and it appears to be a universal rule that, in all archaic periods of art, the primary colours, blue, red, and yellow, are the prevailing colours, and these used most harmoniously and successfully. Whilst in periods when art is practised traditionally, and . not instinctively, there is a tendency to employ the secondary colours and hues, and shades of every variety, though rarely with equal success. We shall have many opportunities of pointing this out in subsequent chapters. H 25 *( EGYPTIAN N°E EGYPT IE NS AEGYPTISCH T\FEL VI AEGYPTISCH EGYPTIAN N°4 egyptiens TAFEL YTT FLYII, AEGYPTISCH EGYPTIAN N°5 EGYPTIENS I I ilMI: JH II tl■IJI I lit f It I O C C G C O G'G C G O O G 14 ilMlWI iPPiPiiiiiiiiiiiiilF fTTTTTTTTTTTTTTTTrV 46 aegyptisch EGYPTIAN N° 6, egyptiehs V^A/WWS^VWS^ WS^^^^\^A(WW ywwvs<^;<• W W y.< vr VN V/ i/.\ V/ y\ y,< /A w vv PEK X X X X X X X X X X X X X V ✓ \ X X X Y \ X X X X X X X X X X X X X X \/ X X X 13 AEGYPTISCH EGYPTIAN N°7 EGYPTIENS mw^mi pjpS^S^yM ?L.X EGYPTIAN N° 8. ■k 'k k ■ ^ k ?5f A Sf s ^ k k '')lt ft 'A * )Sf "5^ '* ^ lr< K ^ ^ ^ ^ ^ ?V ^ Ai,* **,**/„ ^ ^ ^ *<■ j>\ AEGYPTISCH V.X o o © © ©■ © ©©© © © © © © ©© © © © © ((% S...* (Pi ©© © © © •*..» © (?) © ©•© ©e © © © © © €> ©e C5 © o 0 © © ©© © © EGYPTIENS I utiitut rn i ^ n H H W h L^ ^ ^ ua, aai A/>7\ymA: ’!C?\/\A AAA 7 pVAA :A.AA.Ay; PLXin, !••••« • •••• !••••< 20 0 Dlsla 0 PlPln n o| D |d a p £ P 0 n _Q_ □ 0 P _b a 0 3 £ a a. n|ald B alctln o □ C [a ft- n n t; 0 tt U 0 n 11 5 0 £ £ a n n|o|n a nm|o a n|Dlcr B P_ D n 0 a a a 0 n. £ n 0 □ £ P a ninln o n. ninln ti 21 24 Chapter IV. Plates 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 . GREEK ORNAMENT. PLATE XV. A collection of the various forms of the Greek Fret from Vases and Pavements. PLATE XVI.-XXI. Ornaments from Greek and Etrascan Vases in the British Museum and the Louvre. PLATE XXII. 1 and 4. 3, 5-11. 12-17. IS. 19-21, 24-26. 22 and 27. 29. 30-33. From a Sarcophagus, in Sicily. — Hittorpf. From the Propylsea, Athens. — Hittorpf. From the Coffers of the Ceiling of the Propylaea. — Penrose. String-course oyer tlie Panathenaic Frieze. Pnblislled by Mr, Penrose in gold only, we have supplied the blue and red. Painted Ornaments. — Hittorpf. Ornaments in Terra Cotta. Painted Ornament from the Cymatium of theraking Cornice of the Parthenon. — L. Vulliamy, the blue and red supplied. Various Frets, the traces of which exist on all the Temples at Athens. The colours supplied. We have seen that Egyptian Ornament was derived direct from natural inspiration, that it was founded on a few types, and that it remained unchanged during the whole course of Egyptian civilization, except in the more or less perfection of the execution, the more ancient monuments being the most perfect. We have further expressed our belief that the Assyrian was a borrowed style, possessing none of the characteristics of original inspiration, but rather appearing to have been suggested by the Art of Egypt, already in its decline, which decline was carried still farther. Greek Art, on the contrary, though borrowed partly from the Egyptian and partly from the Assyrian, was the development of an old idea in a new direction; and, unrestrained by religious laws, as would appear 31 GEEEK ORNAMENT. to have been both the Assyrian and the Egyptian, Greek Art rose rapidly to a high state of perfection, from which it was itself able to give forth the elements of future greatness to other styles. It carried Upper part ofa Stole. L. Vctlliamy. The upperpart ofa'Stele. L. Vulliamt. the perfection of pure form to a point which has never since been reached; and from the very abundant remains we have of Greek ornament, we must believe the presence of refined taste was 32 GREEK ORNAMENT. almost universal, and that the land was overflowing with artists, whose hands and minds were so trained as to enable them to execute these beautiful ornaments with unerring truth, Greek ornament was wanting, however, in one of the great charms which should always accompany ornament, — viz. Symbolism, It was meaningless, purely decorative, never representative, and can hardly be said to be constructive; for the various members of a Greek monument rather present surfaces exquisitely designed to receive ornament, which they did, at flrst, painted, and in later times both carved and painted. The ornament was no part of the construction, as with the Egyptian : it could be removed, and the structure remained unchanged. On the Corinthian capital the ornament is applied, not constructed : it is not so on the Egyptian capital ; there we feel the whole capital is the ornament, — to remove any portion of it would destroy it. However much we admire the extreme and almost divine perfection of the Greek monumental sculpture, in its application the Greeks frequently went beyond the legitimate bounds of ornament. The frieze of the Parthenon was placed so far from the eye that it became a diagram ; the beauties which so astonish us when seen near the eye could only have been valuable so far as they evidenced the artist-worship which cared not that the eye saw the perfection of the work if conscio.us that it was to be found there : but we are bound to consider this an abuse of means, and that the Greeks were in this respect inferior to the Egyptians, whose system of incavo relievo for monumental sculpture appears to us the more perfect. The examples of representative ornament are very few, with the exception of the wave ornament and the fret used to distinguish water from land in their pictures, and some conventional renderings of trees, as at No. 12, Plate XXI., we have little that can deserve this appellation, but of decorative ornament the Greek and Etruscan vases supply us with abundant materials; and as the painted ornaments of the Temples, which have as yet been discovered in no way differ from them, we have little doubt that we are acquainted with Greek ornament in all its phases. Like the Egyptian the types are few, but the conventional rendering is much further removed from the types. In the well - known honeysuckle ornament it is difficult to recognise any attempt at imitation, but rather an appreciation of the principle on which the flower gr^ws ; and, indeed, on examining the paintings on the vases, we are rather tempted to believe that the various forms of the leaves of a Greek flower have been generated by the brush of the painter, according as the hand is turned upwards or downwards in the formation of the leaf would the character be given, and it is more likely that the slight resemblance to the honeysuckle may have been an after recognition than that the natural flower should have ever served as the model. In Plate XCIX. will be found a representation of the honeysuckle : and how faint indeed is the resemblance. What is evident is, that the Greeks in their ornament were close observers of nature, and although they did not copy, or attempt to imit^e, they worked on the same principles. The three great laws which we find everywhere in nature — radiation from the parent stem, proportionate distribution of the areas, and the tangential curvature of the lines — are always obeyed, and it is the unerring perfection with which they are, in the most humble works as in the highest, which excites our astonishment, and which is only fully realised on attempting to reproduce Greek ornament, so rarely done with success. A very characteristic feature of Greek ornament, continued by the Romans, but abandoned during the Byzantine period, is, that the various parts of a scroll grow out of each other in a continuous Hne, as the ornament from the Choragic Monument of Lysicrates. In the Byzantine, the Arabian Moresque, and Early English styles, the flowers flow off on either side from a continuous line. We have here an instance how slight a change in any K 33 GREEK ORNAMENT. generally received principle is sufficient to generate an entirely new order of forms and ideas. Koman ornament is constantly struggling against this apparently fixed law. At the head of the Eoman chapter is a fine example, which may be taken as a type of all other Roman ornament, which scarcely ever got beyond the arrangement of a volute springing from a stem fitting into From the Choragic Monument of Lysicaates, Athens. L. Vuluamt. another stem, encircling a flower. The change which took place during the Byzantine period in getting rid of this fixed law was as important in its results to the development of ornament, as was the substitution of the arch by the Eomans for the straight architrave, or the introduction of the pointed arch in Gothic architecture. These changes have the same influence in the development of a new style of ornament as the sudden discovery of a general law in science, or the lucky patented idea which in any work .of industry suddenly lets loose thousands of minds to examine and improve upon the first crude thought. Plate XXII. is devoted to the remains of coloured ornaments on the Greek monuments. It will be seen that there is no difference whatever in the character of the drawing to those found on the vases. It is now almost universally recognised, that the white marble temples of the Greeks were entirely covered with painted ornament. Whatever doubts may exist as to the more or less colouring of the sculpture, there can be none as. to the ornaments of the mouldings. The traces of colour exist everywhere so strongly, that in taking casts of the mouldings the traces of the pattern are strongly marked on the plaster cast. What the particular colours were, however, is not so certain. Different authorities give them differently : where one will see green, another finds blue, — or imagines gold where another sees brown. We may be quite certain, however, of one point, — all these ornaments on the mouldings were so high from the ground, and so small in proportion to the distance from which they were seen, that they must have been coloured in a manner to render them distinct and to bring out the pattern. It is with this consideration that we have ventured to supply the colour to 18, 29 31, 32, 33, which have hitherto been published only as gold or brown ornaments on the white marble. Plate XV. In this Plate are given a collection of the different varieties of the Greek fret, from the simple generating form No. 3, to the more complicated meander No. 13. It will be seen, that the variety of arrangement of form that can be produced by the interlacing of lines at right angles in this form is very limited. We have, first, the simple fret. No. 1, running in one direction with a single line ; 34 GEEEK ORNAMENT. the double fret, No. 11, with the second line interlacing with the first; all the others are formed by placing these frets one under the other, running in different directions, as at No. 17 ; back to back, as at Nos. 18 and 19; or enclosing squares, as at No. 20. All the other kinds are imperfect frets, — that is, not forming a continuous meander. The raking fret. No. 2, is the parent of all the other forms of interlacing ornament in styles which succeeded the Greek. From this was first derived the Arabian fret, which in its turn gave birth to that infinite variety of interlaced ornaments formed by the intersection of equidistant diagonal lines, which the Moors carried to such perfection in the Alhambra. Greek. Arabian. Arabian. Moresque. Celtic. The knotted work of the Celts differs from the Moresque interlaced patterns only in adding curved terminations to the diagonal intersecting lines. The leading idea once obtained, it gave birth to an immense variety of new forms. The knotted rope ornament of the Greeks may also have had some influence in the formation both of these and the Arabian and Moresque interlaced ornaments. The Chinese frets are less perfect than any of these. They are formed, like the Greek, by the intersection of perpendicular with horizontal lines, but they have not the same regularity, and the meander is more often elongated in the horizontal direction. They are also most frequently used fragmentally,— that is, there is a repetition of one fret after the other, or one below the other, without forming a continuous meander. The Mexican ornaments and frets, of which we here give some illustrations from Mexican pottery in the British Museum, have a remarkable affinity with the &reek fret ; and in Mr. Catherwood’s illustra- tions of the architecture of Yucatan we have several va- rieties of the Greek fret : one especially is thoroughly Greek. But they are, in general, fragmentary, like the Chinese: there is also to be found at Yucatan a fret with a diagonal line, which is peculiar. The ornaments on Plate XVI. have been selected to show the various forms of conventional leafage to be found on the Greek vases. They are all very far removed from any natural 35 E/3 E\3 From Yucatan. GREEK ORNAMENT. type^ and are rather constructed on the general principles which reign in all plants, than attempt to represent any particular one. The ornament No. 2 is the nearest approach to the honeysuckle, — that is, the leaves have the peculiar turn upwards of that flower, but it can hardly be called an attempt to represent it. Several of the ornaments on Plate XVII. are much nearer to Nature : the laurel, the ivy, and vine will be readily distinguished. Plates XVIII., XIX., XX., and XXI., present further varieties from borders, necks, and lips of vases in the British Museum and the Louvre. Being produced by one or two colours, they all depend for their effect on pure form : they have mostly this peculiarity, that the groups of leaves or flowers all spring from a curved stem, with a volute at either end, and all the lines grow out of this parent stem in tangential curves. The individual leaves all radiate from the centre of the group of leaves, each leaf diminishing in exquisite proportion as it approaches the springing of the group. When we consider that each leaf was done with a single stroke of the brush, and that from the differences which appear we may be sure no mechanical aids were employed, we must be astonished at the high state of the Arts which must have existed for artists to be found in such numbers able to execute with unerring truth what it is almost beyond the skill of modern times even to copy with the same happy result. 36 GREEK ORNAMENT. ORNAMENTS FROM MEXICAN POTTERY IN THE BRITISH MUSEUM. r 37 GRIECIHSCH GREEK N°I. TAFELXY. 151515151515151! grecs PL XV iiiiiimiiiiinmiiii 3 5 mi! nig sBsa g m E E E E E E E E o 0 o 0 D O O o o 2 El EI \b Ie Is Ie b b b b ^1^ giki3ra P ii l5| m pi m [p m iZfS iii 19 20 GREEK N° 3. GRIECIHSCH TAFELXVIL szmsEi .iMmU i] GREGS , PLX-'lf ^ •IfflllfflOl 1 iiiiiii iiiiiii II “I 36 37 3K 39 40 4t 4-; 43 4^44 .«S I * m^i A A A * A *■^AA/Wv^A/'VWWV f f f f f GRIECIHSCH. GREEK N° 4^ GREGS. TAPJ2U XV!U r-6-aOT^V>i GREEK N° 5 GRIECIHSCH TAFEL. XIX GREGS PI. XIX, \AAAAA/ AAA/V\A \AAA/\/V AAAAAA VWVW AAAAAA GREGS, PL rxi GRiBciHscH. GRE.E.K N° 7. \\\\ 7'A£^ 7 Chapter V. — Plates 23, 24, 25. POMPEIAN OKNAMENT. PLATE X X TTT. Collection of Borders from different Houses in Pompeii.— Z ahn^s Pompeii. PLATE XXIV. V arious Pilasters and Friezes from different Houses in Pompeii. — Z ahn’s Pompeii. PLATE XXV. Collection of Mosaics from Pompeii and the Museum at Naples. — From the Author’s Sketches. The ornament of Pompeii has been so ably and so fully illustrated in Zahn’s magnificent work, that we have thought it only necessary for this series to borrow from him the materials for two plates, to illustrate the two distinct styles of ornament which prevail in the decorations of the edifices of Pompeii. The first (Plate XXIII.) are evidently of Grreek origin, composed of conventional ornaments in fiat tints, either painted dark on a light ground, or light on a dark ground, but without shade or any attempt at relief; the second (Plate XXPV.) are more Roman in character, based upon the acanthus scroll, and interwoven "with ornament in direct imitation of Nature. We refer the reader to Zahn’s work* for a full appreciation of the system of ornamentation in use at Pompeii. An examination of this work will show that this system was carried to the very limit of caprice, and that almost any theory of colouring and decoration could be supported by authority from Pompeii. The general arrangement of the decoration on the walls of the interior of a Pompeian house * Les plus Beaux Ornemens et les Tableaux let plus Remarquables de Pompei, d’Herculanum, et de Stabiee, (to., par GruUlaunie Zahn : Berlin, 1828. 39 POMPEIAN ORNAMENT. consists of a dado, about one-sixtb of the height of the wall, upon which stand broad pilasters, half the width of the dado, dividing the wall into three or more panels. The pilasters are united by a frieze of varjring width, about one - fourth of the height of the wall from the top. The upper space is frequently white, and it is always sub- jected to a much less severe treatment than the parts below, generally representing the open air, and upon the ground are painted those fantastic architectural buildings which excited the ire of Vitruvius. In the best examples there is a gra- dation of colour from the ceiling downwards, ending with black in the dado, but this is very far from being a fixed law. We select from the coloured illustrations in Zahn’s work several varieties which will show how little this was the result of system: — Diagram ofthe side ofa Pompeian House. Dado. Pilasters. Yellow Green Bed Bed Black Yellow Black Yellow Blue Yellow Blue Yellow Black Green Black Grey Black Black Pands. Freize. Bed Black Black Purple Black Red Green Green Green Green Blue Blue Yellow and Bed White (alternately) Yellow and Bed Black (alternately) Green and Red White (alternately) The most effective arrangement appears to be black dado, red pilasters and frieze, with yellow, blue, or white panels, the upper part above the frieze being in white, with coloured decorations upon it. The best arrangement of colours for the ornaments on the ground, appears to be, on the black grounds, green and blue in masses, red sparingly, and yellow still more so. On the blue grounds, white in thin lines, and yellow in masses. On the red grounds, green, white, and blue in thin lines; the yellow on red is not effective unless heightened with shade. Almost every variety of shade and tone of colour may be found at Pompeii. Blue, red, and yellow are used, not only in small quantities in the ornaments, but also in large masses as grounds for the panels and pilasters. The yellow of Pompeii, however, nearly approaches orange, and the red is strongly tinged with blue. This neutral character of the colours enables them to be so violently juxtaposed without discord,— a result still further assisted by the secondary and tertiary colours by which they are surrounded. The whole style, however, of the decoration is so capricious that it is beyond the range of true art, and strict criticism cannot be applied to it. It generally pleases, but, if not absolutely vulgar, it oftentimes approaches vulgarity. It owes its greatest charm to the light, sketchy, free-hand manner of its execution, which it Is quite impossible to render in any drawing ; and 40 POMPEIAN ORNAMENT. whicl\ kas never been accomplished in any restoration of the style. The reason is obvious : the artists of Pompeii invented as they drew; every touch of their brush had an intention which no copyist can seize. Mr. Digby Wyatt’s restoration of a Pompeian house in the Crystal Palace, Sydenham, admirable and faithful as it is in all other respects, necessarily failed in this; no one could possibly have brought greater knowledge, experience, and zeal to bear upon the realisation of that accuracy in the decorations which was so much desired than did Signor Abbate. The want of his perfect success consisted in the fact, that his paintings were at the same time too well executed and not sufficiently individual. The ornaments which are given on Plate XXIII., and which have evidently a Greek character, are generally borders on the panels, and are executed with stencils. They have a thinness of character compared with Greek models, which show a marked inferiority ; we no longer find perfect radiation of lines from the parent stem, nor perfect distribution of masses and proportional areas. Their charm lies in an agreeable contrast of colour, which is still further heightened when surrounded with other colours in situ. The ornaments from pilasters and friezes on Plate XXIV., after tlie Eoman type, are shaded to give rotundity, but not sufficiently so to detach them from the ground. In this the Pompeian artists showed a judgment in not exceeding that limit of the treatment of ornament in the round, altogether lost sight of in subsequent times. We have here the acanthus - leaf scroll forming the groundwork, on which are engrafted representations of leaves and flowers interlaced with animals, precisely similar to the remains found in the Eoman baths, and which, in the time of Eaphael, became the foundation of Italian ornament. In Plate XXV. we have gathered together all the forms of mosaic pavement, which was such a feature in every home of the Eomans, wherever their dominion extended. In the attempt at relief shown in several of the examples, we have evidence that their taste was no longer so refined as that of their Greek teachers. The borders formed by a repetition of hexagons at the top and the sides of the page, are the types from which we may directly trace all that immense variety of Byzantine, Arabian, and Moresque mosaics. 4 M 41 POMPEIEHS PL XX IIl I^OMPE-IISCH POMPEIAN 3 -Mi Fragment in White Marble from the Mattel Palace, Rome.—L. Vulliamy.' Chapter VI. — Plates 26, 27. ROMAN ORNAMENT. PLATE XXVI. 1, 2. Fragments from the Forum of Trajan, Rome. 1 4. Pilaster from the Villa Medici, Rome. 3. Pilasters from the Villa Medici, Rome. ] 5, 6. Fragments from the Villa Medici, Rome. Nos. 1-5 are from Casts in the Crystal Palace ; No. 6 from a Cast at South Kensington Museum. PLATE XXVII. 1-3. Fragments of the Frieze of the Roman Temple at Brescia. 4. Fragment of the Soffits of the Architraves of the Roman Temple at Brescia. Nos. 1-4 from the Museo Bresciano ;f 6. Fragment of the Soffits of the Architraves of the Roman Temple at Brescia. 6. From the Frieze of the Arch of the Goldsmiths, Rome. No. 6 from Taylor and Crest’s Rome. Examples of Ornamentul Sculpture in Archilecture, by Lewis Vulliamy, Architect. London, f Masco Brcsciano, itliistrato, Brescia, 1833. 43 ROMAN ORNAMENT. The real greatness of tlie Romans is rather to be seen in their palaces, baths, theatres, aqueducts, and other works of public utility, than in their temple architecture, which being the expression of a religion borrowed from tlie Greeks, and in which probably they had little faith, exhibits a corre- sponding want of earnestness and art-worship. In the Greek temple it is everywhere apparent that the struggle was to arrive at a perfection worthy of the gods. In the Roman temple the aim was self-glorification. From the base of the column to the apex of the pediment every part is overloaded with ornament, tending rather to dazzle by quantity, than to excite admiration by the quality of the work. The Greek temples when painted were as ornamented as those of the Romans, but with a very different result. The ornament was so arranged that it threw a coloured bloom over the whole structure, and in no way disturbed the exquisitely designed surfaces which received it. The Romans ceased to value the general proportions of the structure and the contours of the moulded surfaces, which were entirely destroyed by the elaborate surface-modelling of the ornaments carved on them; and these ornaments do not grow naturally from the surface, but are applied on it. The acanthus leaves under the modillions, and those round the bell of the Corinthian capitals, are placed one before the other most unartistically. They are not even bound together by the necking at the top of the shaft, but rest upon it. Unlike in this the Egyptian capital, where the stems of the flowers round the bell are continued through the necking, and at tlie same time represent a beauty and express a truth. The fatal facilities which the Roman system of decoration gives for manufacturing ornament, by applying acanthus leaves to any form and in any direction, is the chief cause of the invasion of this ornament into most modern works. It requires so little thought, and is so completely a manufacture, that it has encouraged architects in an indolent neglect of one of their especial provinces, and the interior decorations of buildings have fallen into hands most unfitted to supply their place. In the use of the acanthus leaf the Romans showed but little art. They received it from the Greeks beautifully conventionalised ; they went much nearer to the general outline, but exaggerated the surface-decoration. The Greeks confined themselves to expressing the principle of the foliation of the leaf, and bestowed all their care in, the delicate undulations of its surface. The ornament engraved at the head of the chapter is typical of all Roman ornament, which consists universally of a scroll growing out of another scroll, encircling a flower or group of leaves. This example, however, is constructed on Greek principles, but is wanting in Greek refinement. In Greek ornament the scrolls grow out of each other in the same way, but they are much more delicate at the point of junction. The acanthus leaf is also seen, as it were, in side elevation. The purely Roman method of using the acanthus leaf is seen in the Corinthian capitals, and in the 44 ROMAN ORNAJIENT. examples on Plites XXVL and XXVII. The leaves are flattened out, and they lay one over the other, as in the cut. Fragment of the Frieze of the Temple of the Sun, Colomia Palace, Boroe. — L. Vdlliamy. The various capitals which we have engraved from Taylor and Cresy’s work have been placed in juxtaposition, to show how little variety the Eomans were able to produce in following out this application of the acanthus. The only difference which exists is in the proportion of the general form of the mass ; the decline in this proportion from that of Jupiter Stator may be seen readily. How different from the immense variety of Eg3rptian capitals which arose from the modification of the general plan of the capital, even the introduction of the Ionic volute in the Composite order fails to add a beauty, but rather increases the deformity. The pilasters from the Villa Medici, Xos. 3 and 4, Plate XXVL, and the fragment. No. 5, are as perfect specimens of Eoman ornament as could be found. As specimens of modelling and drawing they have strong claims to be admired, but as ornamental accessories to the architectural features of a building, they most certainly, from their excessive relief and elaborate surface treatment, are defi- cient in the first principle, viz. adaptation to the purpose they have to fill. The amount of design that can be obtained by working out this principle of leaf within leaf and leaf over leaf is very limited; and it was not till this principle of one leaf growing out of another in a continuous line was abandoned for the adoption of a continuous stem throwing off ornaments on either side, ^hat pure conventionrd ornament received any development. The earliest examples of the change are found in St. Sophia at Constantinople ; and we introduce here an example from St. Denis, where although the swelling at the stem and the turned back leaf at the junction of stem and stem have entirely disappeared, the continuous stem is not yet fully deve- loped, as it appears in the narrow border top and bottom. This principle became very common in the illuminated MSS. of the eleventh, twelfth, and thirteenth cen- turies, and is the foundation of Early English foliage. The fragments on Plate XXVII., from the Museo Bresciano, are more elegant than those from the Villa Medici; the leaves are more sharply accentuated and more conventionally treated. The frieze from the Arch of the Goldsmiths is, on the contrary, defective from the opposite cause. N 45 From the Abbey of St. Deiiis, Paris. ROMAN ORNAMENT. We have not thought it necessary to give in this series any of the painted decorations of the Romans, of which remains exist in the Roman baths. We had no reliable materials at command ; and, further, they are so similar to those at Pompeii, and show rather what to avoid than what to follow, that we have thought it sufficient to introduce the two subjects from the Forum of Trajan, in which figures terminating in scrolls may be said to be the foundation of that prominent feature in their painted decorations. Tlie Aciintiua, full si®, from a Photogrraph. 4G ROMAN ORNAMENT. Arch of Trajan, Ancoua. Aroh of Titus, Borne. V* Temple of Mars Victor, Borne. Pantheon, Rome. Portico. Pantheon, Rome. Interior of Pantheon, Home. Arch ot Septimius Sevcrua, Rome. Corinthian and Composite Capitals reduced from Taylor and Crest’s Rome.* * The Architectural Antiquities of Rome, by G. L. Taylor and Cresy, Architects. London, 1821. 47 iUliiAiCI[ii!|iU‘u f .ijin.'' vi r .^ a diljn! i;iL< L| il LU J liiJlljlMIU^ llJ.iiJiiii«J ^4«ii v iik ‘■: -^iimhiai' ^■^takW^Khi ii iiJiiMiM Jt|.llii<|iiiMiUlil ilil Chapter VII. — Plates 28, 29, 29*, 30. BYZANTINE ORNAMENT. PLATE 1, 2, 3. Stone Sculptured Ornament, Sta. Sofia, Constanti- nople. 6th Century. — Salzenberg, Alt Christliche Baudenkmale, Conatantinopel. 4, 5. From the Bronze Gates, Sta. Sofia. — Saxzenbeeg, u. a. 6, 7. Pox-tions of Ivory Diptychs, Beauvais Cathedral; apparently Anglo-Saxon work of the 11th century. — WiLLEMiN, Monuments Frangais inedits. 8. Portion of Bronze Door, Basilica of the Nativity, Beth- lehem. 3rd or 4th century. — Gailhabaud, L" Ar- chitecture et les Arts qui en dependent. 9-13. Stone Sculptures, from St. Mark’s, Venice. 11th century. — J. B. W. from Casts at Sydenham. 14, 15, 16. Portion of a Capital, St. Michael’s Church, Schwabisch Hall. 12th Centuiy. — Heideloff, OrnameniiJc des Mittelalters. 17. From a Doorway, preserved at Murrhard Monastery — Heideloff, u. a. 18. Composition of Bosses, from St. Sebald, Nuremberg, and the Church of Nosson, Saxony. — Heideloff. 19, 20. Friezes from the Church of St. John, Gmund Swabia. — Heideloff. 21. Romanesque Wood and Ivory Carving, in the Collec- tion of Heir Leven, Cologne. — Heideloff. XXVIII. 22. From the principal Bronze Door, Monreale, near Palermo.— J. B. W. 23. From the Bronze Door of the Duomo, RaveEo, near Amalfi. — J. B. W. 24. 25. From the Bronze Door of the Duomo, Trani. 12th century. — Barras et Ltjynes, Recherches sur les Monu- ments des Normands en Sidle. 26. Stone Sculpture, from the small Cloister, Huelgas Monastery, near Burgos, Spaia 12th century. — J. B. W. 27. From the Porch of Lucca Cathedral. Circa 1204 a.d. — J. B. W. 28. From St. Denis (Porch), near Paris. 12th century. — J. B. W. 29. From the Cloisters of Sant’ Ambrogia, Milan. — J. B. W. 30. From the Chapel of Heilsbronn, Bavaria. — Heideloff. 31. From St. Denis.— J. B. W. 32. From Eayeux Cathedral. 12th century. — Pugin, An- tiquities of Normandy. 33. From St Denis. — J. B. W. 34. Bayeux Cathedral. — Pugin, u. a. 35. From Lincoln Cathedral Porch. Close of 12th cen- txxry. — J. B. W. 36. From the Kilpeck Porch, Herefordshire. 12th century. — J. B. W. t ilth and 12th cen- turies. PLATE XXIX. 1-6. Mosaics from Sta. Sofia, Constantinople. 6th century. — Salzenberg, Alt Christliche Baudenkmale von Constantinopel. 7. Msu'ble Pavement, Agios Pantokrator, 'Constantinople. First half of 12th century. — Salzenberg, u. a. 8, 9. Mai’ble Pavement, Sta. Sofia. 10, 11. Mosaics, Sta. Sofia. — Salzenberg. 12-15. From Illuminated Greek MSS., British Museum. — J. B. W. 16, 17. Borders, from Illuminated Greek MSS. — Champolijon Figeac, Palceographie Universelle. 18. The centre, from St. Mark’s, Venice. — Digby Wyatt, Mosaics of the Middle Ages. O 49 BYZANTINE ORNAMENT. PLATE XXIX*. 19. From a Greek MS., British Museum. — J. B. W. The border beneath from Monreale. — Digby Wyatt’s Mosaics. 20. From the Homilies of Gregory Nazianzen. 12th century. — Champollion Figeac, u. a. 21, 22. From Greek MSS., British Museum. — J. B. W. 23. From the Acts of the Apostles, Greek MS., Vatican Library, Rome.— Digby Wyatt, u. a. 24. St. Mark’s, Venice. — Digby Wyatt, u. a. 25. Portion of a Greek Diptych. 10th century. Florence. — J. B. W. (The Jieurs-de-lys are believed to be of later workmanship.) 26. Enamel of the 13th contury (French). — Willemiy, Monuments Franqais in^dits. 27. From an Enamelled Casket (the centre from the Statue of Jean, son of St. Louis). — Du Sommerard, Les Arts du Moyen Age. 28. From the Enamelled Tomb of Jean, son of St. Louis, A.D. 1247. — WiT.LEMIN, u. a. 29. Limoges Enamel, probably of the close of 12th century. — Willemin, u. a. 30. Portion of Mastic Pavement, 12th century. Preserved at St. Denis, near Paris. — Willemin. 1 , 2 . 3. 4, 5. 6 . 7-10. 11 . 12 . 13. 14. 15, 16. 17. 18. 19. 20 . 21 . PLATE XXX. Mosaics (opus Grecanicum) from Monreale Cathe- dral, near Palermo. Close of 12th century.— J. B. W. Mosaics from the Church of Ara Cceli, Rome.— J.B.W. Monreale Cathedral. — J. B. W. Marble Pavement, St. Mark’s, Venice. — J. B. W. From San Lorenzo Fuori, Rome. Close of 12th century. — J. B. W. San Lorenzo Fuori, Rome. — J. B. W. Ara Cceli, Rome. — J. B. W. Marble Pavement, St. Mark’s, Venice. — J. B. W. San Lorenzo Fuori, Rome. — Architectural Art in Italy and Spain, hy Waring and MacQuoid. Palermo.— Digby Wyatt, Mosaics of the Middle Ages. From the Cathedral, Monreale. — J. B. W. From Ai’a Cceli, Rome. — J. B. W. Marble Pavement, S. M. Maggiore, Rome. — Hessemer, Arahische und alt Italidnische Bau Yerzierungen. Marble Pavement, San Vitale, Ravenna. — Hessemer, u. a. Marble Pavement, S. M. in Cosmedin, Rome.— Hes- SEMER, 11. a. 22, 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33-36. 36-38. 39. 40. 41. 42. Mosaic, St. Mark’s, Venice. — Specimens of the Mosaics of the Middle Ages, Digby Wyatt. Baptistery of St. Mark, Venice. — Architectural Art in Italy and Spain, Waring and MacQuoed. San Giovanni Laterano, Rome. The Duomo, Civita Castellana. From Digby Wyatt’s Mosaics of the Mid- dle Ages. Ara Cceli, Rome. — J. B. W. San Lorenzo, Rome. ) Architectural Art in Italy and Ara Cceli, Rome. I Spain, Waring and Mac- San Lorenzo, Rome, j Quoid. San Lorenzo Fuori, Rome. — J. B. W. San Giovanni Laterano, Rome. — Digby Wyatt’s Mosaics of the Middle Ages. Monreale Cathedral. — J. B. W. Marble Pavement, S. M. Maggiore, Rome. — Hesse- MER, u. a. St. Mark’s, Venice. — Mosaics of the Middle Ages, Digby Wyatt. From the Baptistery, St. Mark’s, Venice.— J. B. W. From St. Mark’s, Y qtAcq.— A rchitectural Art in Italy and Spain. From the Duomo, Monreale. — J. B. W. BYZANTINE ORNAMENT. The vagueness with wliich writers on Art have treated the Byzantine and Romanesque styles of Architecture, even to within the last few years, has extended itself also to their concomitant decoration. This vagueness has arisen chiefly from the want of examples to which the writer could refer; nor was it until the publication of Herr Salzenberg’s great work on Sta. Sofia at Constantinople, that we could obtain any complete and definite idea of what constituted pure Byzantine ornament. San Vitale at Ravenna, though thoroughly Byzantine as to its architecture, still afforded us but a very incomplete notion of Byzantine ornamentation: San Marco at Venice represented but a phase of the Byzantine school; and the Cathedral of Monreale, and other examples of the same style in Sicily, served only to show the influence, but hardly to illustrate the true nature, of pure Byzantine Art ; 50 BYZANTINE OENAMENT. fully to understand that, we required what the ravages of time and the whitewash of the Mahom- medan had deprived us of, namely, a Byzantine building on a grand scale, executed during the best period of the Byzantine epoch. Such an invaluable source of information has been opened to us through the enlightenment of the present Sultan, and been made public to the world by the liberality of the Prussian government; and we recommend all those who desire to have a graphic idea of what Byzantine decorative art truly was, to study Herr Salzenberg’s beautiful work on the churches and buildings of ancient Byzantium. In no branch of art, probably, is the observation, ex nihilo nihil Jit, more applicable than in decorative art. Thus, in the Byzantine style, we perceive that various schools have combined to form its peculiar characteristics, and we shall proceed to point out briefly what were the principal formative causes. Even before the transfer of the seat of the Koman Empire from Eome to Byzantium, at the commencement of the fourth century, we see all the arts in a state either of decline or transformation. Certain as it is that Rome had given her peculiar style of art to the numerous foreign peoples ranged beneath her sway, it is no less certain that the hybrid art of her provinces had powerfully reacted on the centre of civilisation ; and even at the close of the third century had materially affected that lavish style of decoration which characterised the magnificent baths and other public buildings of Rome. The necessity which Constantine found himself under, when newly settled in Byzantium, of employing Oriental artists and workmen, wrought a still more vital and marked change in the traditional style; and there can be little doubt but that each surrounding nation aided in giving its impress to the newly-formed school, according to the state of its civilisation and its capacity for Art, until at last the motley mass became fused into one systematic whole during the long and (for Art) prosperous reign of the first Justinian. b In this result we cannot fail to be struck with the important influence exercised by the great temples and theatres built in Asia Minor during the rule of the Caesars ; in these we already see the tendency to elliptical curved outlines, acute-pointed leaves, and thin continuous foliage without the springing-ball and flower, which characterise Byzantine ornament. On the frieze of the theatre at 51 BYZANTINE ORNAMENT. c !) c 6 Patara (a), and at the Temple of Venus at Aphrodisias (Caria), are to be seen examples of flowing foliage such as we allude to. On the doorway of the temple erected by the native rulers of Gralatia at Ancyra (b'), in honour of Augustus, is a still more characteristic t3Tpe ; and the pilaster capital of a small temple at Patara (c), ascribed by Texier to the first century of the Christian era, is almost identical with one drawn by Salzenberg at Smyrna (c2), which he believes to be of the first part of Justinian’s reign, or about the year 525 A.D. In the absence of authentic dates we cannot decide satisfactorily how far Persia influenced the Byzantine style, but it is certain that Persian workmen and artists were much employed at Byzantium ; and in the remarkable monuments at Tak-i-Bostan, Bi-Sutoun, and Tak-i-Gihero, and in several ancient capitals at Ispahan — given in Flandin and Coste’s great work on Persia — we are struck at once with their thoroughly Byzantine character; but we are inclined to believe that they are pos- terior, or at most contemporaneous with the best period of Byzantine art, that is, of the sixth century. However that may be, we find the forms of a still earlier period reproduced so late as the year 363 A.D. ; and in Jovian’s column at Ancyra (e), erected during or shortly after his retreat with Julian’s army from their Persian expedition, we recognise an application of one of the most general orna- mental forms of ancient Persepolis. At Persepolis also are to be seen the pointed and channelled leaves so characteristic of Byzantine work, as seen in the accompanying example from Sta. Sofia (/) ; and at a later period, i.e. during the rule of the Cgesars, we remark at the Boric temple of Kangovar (g) contours of moulding precisely similar to those affected in the Byzantine style. Interesting and instructive as it is to trace the derivation of these forms in the Byzantine style, it is no less so to mark the transmission of them and of others to later epochs. Thus in No. 1, Plate XXVIII., we perceive the peculiar leaf, as giveD in Texier and in Salzenberg, reappear at Sta. Sofia; at No. 3, Plate XXVIII., is the foliated St. Andrew’s cross within a circle, so co mm on as a Bomanesque and Giothic ornament. On the same frieze is a design repeated with but slight altera- tion at No. 17, from Hermany. The curved and foliated branch of No. 4 of the sixth century ^Sta. Sofia) is seen reproduced, with slight variation, at No. 11 of the eleventh century (St. Mark’s). The toothings of the leaves of No. 19 (Germany) are almost identical with those of No. 1 (Sta. Sofia) ; and be- tween all the examples on the last row but one (Plate XXVIII.) is to be remarked a generic resemblance in subjects from Germany, Italy, and Spain, founded on a Byzantine type. The last row of subjects in this plate illustrates more especially the Eomanesque style (Nos. 27 and 36), showing the interlaced ornament so afifected by the Northern nations, founded mainly on a native type; whilst at No. 35 (St. Denis) we have one instance out of numbers of the reproduction 52 BYZANTINE ORNAMENT. of Roman models; tlie type of the present subject, — a common one in the Romanesque style, — being found on the Roman column at Cussy, between Dijon and Chalons-sur-Saone. Thus we see that Rome, Syria, Persia, and other countries, all took part as formative causes in the Byzantine style of art, and its accompanying decoration, which, complete as we find it in Justinian’s time, reacted in its new and systemised form upon the Western world, undergoing certain changes in its course ; and these modifying causes, arising from the state of religion, art, and manners in the countries where it was received, frequently gave it a specific character, and produced in some cases co-relative and yet distinct styles of ornament in the Celtic, Anglo-Saxon, Lombardic, and Arabian schools. Placing on one side the question of how far Byzantine workmen or artists were employed in Europe, there can be no possible doubt that the character of the Byzantine school of ornament is very strongly impressed on aU the earlier works of central and even Western Europe, which are generically termed Romanesque. Pure Byzantine ornament is distinguished by broad-toothed and acute-pointed leaves, which in sculpture are bevelled at the edge, are deeply channelled throughout, and are drilled at the several springings of the teeth with deep holes ; the running foliage is generally thin and continuous, as at Nos. 1, 14, and 20, Plate XXIX*., Plate XXIX. The ground, whether in Mosaic or painted work, is almost universally gold; thin interlaced patterns are preferred to geometrical designs. The introduction of animal or other figures is very limited in sculpture, and in colour is confined prin- cipally to holy subjects, in a stiff, conventional style, exhibiting little variety or feeling ; sculpture is of very secondary importance. Romanesque ornament, on the other hand, depended mainly on sculpture for effect : it is rich in light and shade, deep cuttings, massive projections, and a great intermixture of figure-subjects of every kind with foHage and conventional ornament. The place of mosaic work is generally supplied by paint; in coloured ornament, animals are as freely introduced as in sculpture, vide No. 26, Plate XXIX*.; the ground is no longer gold alone, but blue, red, or green, as at Nos. 26, 28, 29, Plate XXIX*. In other respects, allowing for local differences, it retains much of the Byzantine character; and in the case of painted glass, for example, handed it down to the middle, and even the close of the thirteenth century. One style of ornament, that of geometrical mosaic work, belongs particularly to the Romanesque period, especially in Italy; numerous examples of it are given in Plate XXX. This art flourished principally in the twelfth and thirteenth centuries, and consists in the arrangement of small diamond- shaped pieces of glass into a complicated series of diagonal lines ; the direction of which is now stopped, now defined, by means of different colours. .The examples from central Italy, such as Nos. 7, 9, 11, 27, 31, are much simpler than those of the southern provinces and Sicily, where Saracenic artists introduced their innate love of intricate designs, some ordinary examples of which are to be seen in Nos. 1, 5, 33, from Monreale, near Palermo. It is to be remarked, that there are two distinct styles of design coexistent in Sicily: the one, such as we have noted, consisting of diagonal interlacings, and eminently Moresque in character, as may be seen by reference to Plate XXXIX.; the other, consisting of interlaced curves, as at Nos. 33, 34, 35, also from Monreale, in which we may recognise, if not the hand, at least the influence, of Byzantine artists. Altogether of a different character, though of about the same period, are Nos. 22, 24, 39, 40, 41, which serve as examples of the Veneto-Byzantine style ; limited in its range, being almost local, and peculiar in style. Some are more markedly Byzantine, however, as No. 23, with interlaced circles; and the step ornament, so common at Sta. Sofia, as seen at Nos. 3, 10, and 11, Plate XXIX. The o^us Alexandrinum, or marble mosaic work, differs from the ojpus Grecanicum, or glass mosaic work, chiefly from the different nature of the material ; the principal (that of complicated 53 p BYZANTINE ORNAMENT. geometric design) is still ttie same. The pavements of the Romanesque churches in Italy are rich in examples of this class ; the tradition of which was handed down from the Augustan age of Rome ; a good idea of the nature of this ornament is given in Nos. 19, 21, 36, 37, and 38. Local styles, on the system of marble inlay, existed in several parts of Italy during the Roman- esque period, which bear little relation either to Roman or Byzantine models. Such is No. 20, from San Vitale, Ravenna; such are the pavements of the Baptistery and San Miniato, Florence, of the eleventh, twelfth, and thirteenth centuries; in these the effect is produced by black and white marble only; with these exceptions, and those produced by Moresque influence in the South of Italy, the principles both of the glass and marble inlay ornament are to be found in ancient Roman inlay, in every province under Roman sway, and especially is it remarkable in the various mosaics found at Pompeii, of which striking examples are given in Plate XXV. Important as we perceive the influence of Byzantine Art to have been in Europe, from the sixth to the eleventh century, and still later, there is no people whom it affected more than the great and spreading Arab race, who propagated the creed of Mahomet, conquered the finest countries of the East, and finally obtained a footing even in Europe. In the earlier buildings executed by them at Cairo, Alexandria, Jerusalem, Cordova, and Sicily, the influence of the Byzantine style is very strongly marked. The traditions of the Byzantine school affected more or less all the adjacent countries ; in Greece they remained almost unchanged to a very late period, and they have served, in a great degree, as the basis to all decorative art in the East and in Eastern Europe. J. B. WARING. September 1856. *** For more information on this subject, see “Handbook” to Byzantine and Romanesque Court at Sydenham. — ■Wyatt and Waking. BOOKS EEFEEKED TO FOE ILLESTEATIOKS. Saxzenberg. Alt Christliche Baudenkmale von Constantinopel. Flandin et Coste. f^oynge en Perse. Texiek, Description de I’Armdnie, Perse, dc. Heideloee. Die Ornamentik des Mittelaltei's. Kkeutz. La Basilica di San Marco. Gailhabatjd. L'Architecture et les Arts qui en dependent. Du SoiffMERABD. Les Arts du Moyen Age. I Baekas et Luynes (Due de). Recherches sur les Monuments des Normands en Sidle. Chasitollion Figeac. Palaographie UniverseUe. W11J.EMIN. Monuments Frangnis inedits. Hessemer. Arabische und alt Italidnische Bau Verzierungen, Digby Wyatt. Geometrical Mosaics of the Middle Ages. Waring and MacQuoid. Architectural Arts in Italy and Spain. Waring. Architectural Studies at Burgos and its Neighbourhood. 54 V V - ' 1- " • ' . I --.I - • --T*.— 1.,^. BYZANTINISCH BYZANTINE N° Z BYZANTINS BYZAHTINISGH TAFEl, XXIX BYZANTINE N"2^ byzantihs FL xx;x m mi h * ;y pMkM^<^ i% I &m^M. ^;®^|g 85 #S. 2 SK«M 4 'sxsxnexaaai ■>Z ZiC wi^- BYZANTINISCH TAFEL XXX ■ BYZANTINE N° B BYZANTINS PL XXX m ■Aa 2»'9I^X' ■ m K: §g ^ itiill ;• i-4.''f ’ i ■iV'!-’ ^Vple. 20, 22. From a Fountain at Tophana, Constantinople. PLATE XXXVII. 1, 2, 6, 7, 8. From the Yeni D’jami, Constantinople. I 4 , 5 . Ornaments in Spandrils under the Dome of the Mosque 3. Rosace in the Centre of the Dome of the Mosque of of Soliman I., Constantinople. Soliman I., Constantinople. PLATE XXXViri. Portion of the Decoration of the Dome of the Tomb of Soliman I., Constantinople. The architecture of the Turks, as seen at Constantinople, is in all its structural features mainly based upon the early Byzantine monuments ; their system of ornamentation, however, is a modifi- cation of the Arabian, bearing about the same relation to this style as Elizabethan ornament does to Italian Renaissance. When the art of one people is adopted by another having the same religion, but differing in natural character and instincts, we should expect to find a deficiency in all those qualities in which the borrowing people are inferior to their predecessors. And thus it is with the art of the Turks as compared with the art of the Arabs : there is the same difference in the amount of elegance and refinement in the art of the two people as exists in their national character. We are, however, inclined to believe that the Turks have rarely themselves practised the arts ; but that they have rather commanded the execution than been themselves executants. All their mosques and public buildings present a mixed style. On the same building, side by side with ornaments derived from Arabian and Persian floral ornaments, we find debased Roman and Renaissance details, leading to the belief that these buildings have mostly been executed by artists differing in religion from themselves. In more recent times, the Turks have been the first of the Mohammadan 61 R i V 2 TURKISH ORNAMENT. races to abandon the traditional style of building of their forefathers, and to adopt the prevailing fashions of the day in their architecture ; the modern buildings and palaces being not only the work of European artists, but designed in the most approved European style. The productions of the Turks at the Great Exhibition of 1851 were the least perfect of all the Mohammadan exhibiting nations. In Mr. M. Digby Wyatt’s admirable record of the state of the Industrial Arts of the Nineteenth Century, will be found specimens of Turkish embroidery exhibited in 1851, and which may be compared with the many valuable specimens of Indian embroidery represented in the same work. TURKISH ORNAMENT. distribution of form, and in all the principles of ornamentation, as the most elaborate and important article of decoration. ^ The only examples we have of perfect ornamentation are to be found in Turkey carpets ; but these are chiefly executed in Asia Minor, and most probably not by Turks. The designs are thoroughly Arabian, differing from Persian carpets in being much more conventional in the treatment of foliage. By comparing Plate XXXVII. with Plates XXXII. and XXXIII, the differences of style will be readily perceived. The general principles of the distribution of form are the same, but there are a few minor differences that it will be desirable to point out. The surface of an ornament both in the Arabian and Moresque styles is only slightly rounded, and the enrichment of the surface is obtained by sinking lines on this surface ; or where the surface was left plain, the additional pattern upon pattern was obtained by painting. The Turkish ornament, on the contrary, presents a carved surface, and such ornaments as we find painted in the Arabian MSS., Plate XXXPV., in black lines on the gold flowers, are here carved on the surface, the effect being not nearly so broad as that produced by the sunk feathering of the Arabian and Moresque. Another peculiarity, and one which at once distinguishes a piece of Turkish ornament from Arabian, is the great abuse which was made of the re-entering curve A A. This is very prominent in the Arabian, but more especially in the Persian styles. See Plate XLVI. With the Moors it is no longer a feature, and appears only exceptionally. This peculiarity was adopted in the Elizabethan ornament, which, through the Benaissance of France and Italy, was derived from the East, in imitation of the damascened work which was at that period so common. It will be seen on reference to Plate XXXVI., that this swell always occurs on the inside of the spiral curve of the main stem ; with Elizabethan ornament the swell often occurs indifferently on the inside and on the outside. It is very difficult, nay, almost impossible, thoroughly to explain by words differences in style of ornament having such a strong family resemblance as the Persian, Arabian, and Turkish ; yet the eye readily detects them, much in the same way as a Roman statue is distinguished from a Greek. The general principles remaining the same in the Persian, the Arabian, and the Turkish styles of ornament, there will be found a peculiarity in the proportions of the masses, more or less grace in the flowing of the curves, a fondness for particular directions in the leading lines, and a peculiar mode of inter- weaving forms, the general form of the conventional leafage ever remaining the same. The relative degree of fancy, delicacy, or coarseness, with which these are drawn, will at once distinguish them as the works of the refined and spiritual Persian, the not less refined but reflective Arabian, or the unimagi- native Turk. Plate XXXVIII. is a portion of the decoration of the dome of the tomb of Soliman I. at Constan- tinople; it is the most perfect specimen of Turkish ornament with which we are acquainted, and nearly approaches the Arabian. One great feature of Turkish ornament is the predominance of green and black; and, in fact, in the modern decoration of Cairo the same thing is observed. Green is much more prominent than in ancient examples where blue is chiefly used. 63 u TURKISCH. T-URKIS'H N°2. turgs PL.XXXYIl TAFEL XXXVII. mmk LlRKISCK TURKISH N° a T LI R C S Chapter X. — Plates 39, 40, 41, 41*, 42, 42*, 42'f, 43. MORESQUE ORNAMENT, FROM THE ALHAMBRA. PLATE XXXIX. INTERLACED ORNAMENTS. 1-5, 16, 18, are Borders on Mosaic Dados. j 13, 16. Square Stops in the Bands of the Inscriptions. 6-12,14. Plaster Ornaments, used as upright and horizon- 17. Painted Ornament from the Great Arch in the Hall of tal Bands enclosing Panels on the walls.' 1 the Boat. PLATE XL. SPANDRILS OF ARCHES. 1. From the centre Arch of the Court of the Lions. 2. From the Entrance to the Divan Hall of the Two Sisters. 3. From the Entrance to the Court of the Lions from the Court of the Fish-ponds. 4. From the Entrance to the Court of the Fish-pond from the Hall of the Boat. 6, 6. From the Arches of the Hall of Justice. PLATE XLI. LOZENGE 1. Ornament in Panels from the Hall of the Boat. 2. „ „ from the HaU of the Ambassadors. 3. „ in Spandril of Arch, entrance to Court of Lions. 4. „ in Doorway of the Divan, Hall of the Two Sisters. PLATE 9, 10. Ornaments in Panels, Court of the Mosque. 11. Soffit of Great Arch, entrance to Court of Fish-pond. 12. Ornaments in Sides of Windows, Upper Story, Hall of Two Sisters. DIAPERS. 5. Ornament in Panels of the Hall of the Ambassadors. 6. „ in Panels of the Courts of the Mosque. 7. 5 , in Panels, Hall of the Abencerrages. 8. „ over Arches, entrance to the Court of Lions. XLI*. 13. Oraaments in Spandrils of Arches, Hall of the Abencer- rages. 14, 15. Ornaments in Panels, Hall of Ambassadors. 16. „ inSpandrils ofArches, Hall ofthe Two Sisters. PLATE XLII. SQUARE DIAPERS. 1. Frieze over Columns, Court of the Lions. 2. Panelling in Windows, Hall of the Ambassadors. PLATE XLII*. 3. Panelling of the centre Recess of the Hall of the Ambassadors. 4. Panelling on the Walls, Tower of the Captive. 65 s MORESQUE ORNAMENT. PLATE XLIIf. 6. Panelling on the Walls, House of Sanchez. 6. Part of the Ceiling of the Portico of the Court of the Fish-pond. PLATE XLIII. Pilaster, Hall of the Ambassadors. Dado, ditto. Dado, Hall of the Two Sisters. Pilaster, Hall of the Ambassadors. Dados, Hall of the Two Sisters. 7. Pilaster, Hall of Justice. 8. Dado, HaU of the Two Sisters. MOSAICS. 10 . 11 . 12, 13. 14. 15. 16. Dado in centre Window, Hall of the Ambassadors. Pilaster, Hall of the Ambassadors. Dado, Hall of Justice. Dados, Hall of the Ambassadors. From a Column, Hall of Justice. Dado in the Baths. Dado in Divan, Court of the Fish-pond. MORESQUE ORNAMENT. Our illustrations of the ornament of the Moors have been taken exclusively from the Alhambra, not only because it is the one of their works with which we are best acquainted, but also because it is the one in which their marvellous system of decoration reached its culminating point. The Alhambra is at the very summit of perfection of Moorish art, as is the Parthenon of Greek art. We can find no work so fitted to illustrate a Grammar of Ornament as that in which every ornament contains a grammar in itself. Every principle which we can derive from the study of tlie ornamental art of any other people is not only ever present here, but was by the Moors more universally and truly obeyed. We find in the Alhambra the speaking art of the Egyptians, the natural grace and refinement of the Greeks, the geometrical combinations of the Romans, the Byzantines, and the Arabs. The ornament wanted but one charm, which was the peculiar feature of the Egyptian ornament, symbolism. This the religion of the Moors forbade ; but the want was more than supplied by the inscriptions, which, addressing themselves to the eye by their outward beauty, at once excited the intellect by the difficulties of deciphering their curious and complex involutions, and delighted the imagination when read, by the beauty of the sentiments they expressed and the music of their composition. “ There is no conqueror but God.” Arabic inscription from the Alhambra. To the artist and those provided with a mind to estimate the value of the beauty to which they gave a life they repeated, Looh dud Ibcitti. To the people they proclaimed the might, majesty, and good deeds of the king. To the king himself they never ceased declaring that there was none powerful but God, that He alone was conqueror, and that to Him alone was for ever due praise and glory. 66 MORESQUE ORNAMENT. The builders of this wonderful structure were fully aware of the greatness of their work. It is asserted in the inscriptions on the walls, that this building surpassed all other buildings ; that at sight of its wonderful domes all other domes vanished and disappeared ; in the playful exaggeration of their poetry, that the stars grew pale in their light through envy of so much beauty ; and, what is more to our purpose, they declare that he who should study them with attention would reap the benefit of a commentary on decoration. We have endeavoured to obey the injunctions of the poet, and will attempt here to explain -some of the general principles which appear to have guided the Moors in the decoration of the Alhambra — principles which are not theirs alone, but common to all the best periods of art. The principles which are everywhere the same, the forms only differ. 1. * The Moors ever regai’ded what we hold to be the first principle in architecture — to decorate construction^ never to construct decoration : in Moorish architecture not only does the decoration arise naturally from the construction, but the constructive idea is carried out in every detail of the ornamentation of the surface. We believe that true beauty in architecture results from that ^‘repose which the mind feels when the eye, the intellect, and the affections are satisfied, from the absence of any want.” When an object is constructed falsely, appearing to derive or give support without doing either the one or the other, it fails to afford this repose, and therefore never can pretend to true beauty, however harmonious it may be in itself ; the Mohammadan races, and Moors especially, have constantly regarded this rule ; we never find a useless or superfluous ornament ; every ornament arises quietly and naturally from the surface decorated. They ever regard the useful as a vehicle for the beautiful ; and in this they do not stand alone : the same principle was observed in all the best periods of art ; it is only when art declines that true principles come to be disregarded; or, in an age of copying, like the present, when the works of the past are reproduced without the spirit which animated the originals. 2. All lines grow out of each other in gradual undulations ; there are no excrescences ; nothing could be removed and leave the design equally good or better. In a general sense, if construction be properly attended to, there could be no excrescences ; but we use the word here in a more limited sense : the general lines might follow truly the construction, and yet there might be excrescences, such as knobs or bosses, which would not violate the rule of construction, and yet would be fatal to beauty of form, if they did not grow out gradually from the general lines. There can be no beauty of form, no perfect proportion or arrangement of lines, which does not produce repose. All transitions of curved lines from curved, or of curved lines from straight, must be gradual. Thus the transition would cease to be agreeable if the break at A were too deep in proportion to the curves, as at B. Where two curves are separated by a break (as in this case), they must, and with the Moors always do, run parallel to an imaginary line (c) where the curves would be tangential to each other ; for were either to depart from this, as in the case at D, the eye, instead of following gradually down the curve, would run outwards, and repose would be lost.f * This essay on the general principles of the ornamentation of the Alhambra is partially reprinted from the “ Guide Book to the Alhambra Court in the Crystal Palace,” by the Author, f These transitions were managed most perfectly by the Greeks in all their mouldings, which exhibit this refinement in the highest degree ; so do also the exquisite contom*s of their vases. 67 3. The general forms were first cared for ; these were subdivided by general lines ; the interstices were then filled in with ornament, which was again subdivided and enriched for closer inspection. They carried out this principle with the greatest refinement, and the harmony and beauty of all their ornamentation derive their chief success from its observance. Their main divisions contrast and balance admirably; the greatest distinctness is obtained; the detail never interferes with the general form. When seen at a distance, the main lines strike the eye; as we approach nearer, the detail comes into the composition ; on a closer inspection, we see still further detail on the surface of the ornaments themselves. 4. Harmony of form appears to consist in the proper balancing and contrast of the straight, the inclined, and the curved. As in colour there can be no perfect composition in which either of the three primary colours is wanting, so in form, whether structural or decorative, there can be no perfect composition in which either of the three primary figures is wanting ; and the varieties and harmony in composition and design depend on the various predominance and subordination of the three.* In surface decoration, any arrangement of forms, as at A, consisting only of straight lines, is monotonous, and affords but imperfect pleasure; but introduce lines which tend to carry the eye towards the angles, as at B, and you have at once an increased pleasure. Then add lines giving a circular tendency, as at c, and you have now complete harmony. In this case the square is the leading form or tonic ; the angular and curved are subor- dinate. We may produce the same result in adopting an angular composition, as at D : add the lines as at E, and we at once correct the tendency to follow only the angular direction of the inclined lines ; but unite these by circles, as at f, ABC and we have still more perfect harmony, i.e. repose, for the eye has now no longer any want that could be supplied.^ 5. In the surface decoration of the Moors all lines flow out of a parent stem : every ornament, however distant, can be traced to its branch and root. They have the happy art of so adapting the ornament to the surface decorated, that the ornament as often appears to have suggested the general form as to have been suggested by it. In all cases, we find the foliage flowing out of a parent stem, and we are never offended, as in modern practice, by the random introduction of an ornament just dotted down, without a reason for its existence. However irregular the space they ^ \ / S / \ ;o: A a; a: a: :o: :o; A ))K o )^o) o)^( p * There can be no better example of this harmony than the Greek temple, where the straight, the angular, and the curved, are in most perfect relation to each other. Gothic architecture also offers many illustrations of this principle; every tendency of lines to run in one direction is immediately counteracted by the angular or the curved: thus, the capping of the buttress is exactly what is required to counteract the upwai'd tendency of the sti’aight lines; so the gable contrasts admirably with the ouiwed window- head and its perpendicular raullions. t It is to the neglect of this obvious rule that we find so many failures in paper-hangings, carpets, and more especially articles of costume; the lines of papers generally run through the ceiling most disagreeably, because the straight is not coixected by the angular, or the angular by the curved; so of carpets, the lines of cai'pets are constantly running in one direction only, carrying the eye right through the walls of the apai-tment. Again, to this we owe all those abominable checks and plaids which constantly disfigure the human form — a custom detrimental to the public taste, and gradually lowering the tone of the eye for form of this generation. If children were bom and bred to the sound of hurdy-gurdies grinding out of tune, their ears would no doubt suffer deterioration, and they would lose their sensibility to the harmonious in sound. This, then, is what is certainly taking place with regard to form, and it requires the most strenuous efforts to be made by all who would take an interest in the welfare of the rising generation to put a stop to it. 68 MORESQUE ORNAMENT. haye to fill, they always commence by dividing it into equal areas, and round these trunk-lines they fill in their detail, but invariably return to their parent stem. They appear in this to work by a process analogous to that of nature, as we see in the vine-leaf; the object being to distribute the sap from the parent stem to the extremities, it is evident the main stem would divide the leaf as near as may be into equal areas. So, again, of the minor divisions ; each area is again subdivided by intermediate lines, which all follow the same law of equal distribution, even to the jnost minute filling-in of the sap-feeders. 6. The Moors also follow another principle ; that of radiation from the pai’ent stem, as we may see in nature with the human hand, or in a chestnut leaf. We may see in the example how beautifully all these lines radiate from the parent stem ; how each leaf diminishes towards the extremities, and how each area is in pro- portion to the leaf. The Orientals carry out this principle with marvellous perfection ; so also did the G-reeks in their honeysuckle ornament. We have already remarked, in Chapter IV., a peculiarity of Greek ornament, which appears to follow the principle of the plants of the cactus tribe, where one leaf grows out of another. This is generally the case with G-reek ornament ; the acanthus leaf-scrolls are a series of leaves growing out one from the other in a continuous line, whilst the Arabian and Moresque ornaments always grow out of a continuous stem. 7. All junctions of curved lines with curved, or of curved with straight, should be tangential to each other; this also we consider to be a law found everywhere in nature, and the Oriental practice is always in accordance with it. Many of the Moorish ornaments are on the same principle which is observable in the lines of a feather and in the articu- lations of every leaf; and to this is due that additional charm found in all perfect ornamentation, which we call the graceful. It may be called the melody of form, as what we have before described constitutes its harmony. We shall find these laws of equal distribution, radiation from a parent stem, continuity of line, and tangential curvature, ever present in natural leaves. 8. We would call attention to the nature of the exquisite curves in use by the Arabs and Moors. As with proportion, we think that those proportions will be the most beautiful which it will be most difficult for the eye to detect;* so we think that those compositions of curves will be most agreeable, where the mechanical process of describing them shall be least apparent; and we shall fiod it to be universally the case, that in the best periods of art, all mouldings and ornaments were founded on curves of the higher order, such as the conic, sections ; whilst, when art declined, circles and compass-work were much more dominant. The researches of Mr. Penrose have shown that the mouldings and curved lines in the Parthenon are all portions of curves of a very high order, and that segments of circles were very rarely used. The exquisite curves of the G-reek vases are well known, and here we never find portions of circles. In Roman architecture, on the contrary, this refinement is lost ; the Romans were probably as little able to describe as to appreciate curves of a high order ; and we find, therefore, their mouldings mostly parts of circles, which could be struck with compasses. * All compositions of squares or of circles will lie monotonous, and afford lut little pleasure, because the means whereby they are produced are very apparent. So we think that compositions distiibuted in equal lines or divisions will be less beautiful than those which require a higher mental effort to appreciate them. T 69 In the early works of the Gothic period, the tracery would appear to have been much less the offspring of compass-work than in the later period, which has most appropriately been termed the Geometrical, from the immoderate use of compass-work. Here is a curve (a) common to Greek Art, to the Gothic period, and so much delighted in by the Mohammadan races. This becomes graceful the more it departs from the curve which the union of two parts of circles would give. 9. A still further charm is found in the works of the Arabs and Moors from their conventional treatment of ornament, which, forbidden as they were by their creed to represent living forms, they carried to the highest perfection. They ever worked as nature worked, but always avoided a direct transcript; they took her principles, but did not, as we do, attempt to copy her works. In this, again, they do not stand alone; in every period of faith in art, all ornamentation was ennobled by the ideal ; never was the sense of propriety violated by a too faithful representation of nature. Thus, in Egypt, a lotus carved in stone was never such an one as you might have plucked, but a conventional representation perfectly in keeping with the architectural members of which it formed a part ; it was a symbol of the power of the king over countries where the lotus grew, and added poetry to what would otherwise have been a rude support. The colossal statues of the Egyptians were not little men carved on a large scale, but architectural representations of Majesty, in which were symbolised the power of the monarch, and his abiding love of his people. In Greek art, the ornaments, no longer symbols, as in Egypt, were still further conventionalised ; and in their sculpture applied to architecture, they adopted a conventional treatment both oi pose and relief very different to that of their isolated works. In the best periods of Gothic art the floral ornaments are treated conventionally, and a direct imitation of nature is never attempted ; but as art declined, they became less idealised, and more direct in imitation. The same decline may be traced in stained glass, where both figures and ornaments were treated at first conventionally ; but as the art declined, figures and draperies, through which light was to be transmitted, had their own shades and shadows. In the early illuminated MSS. the ornaments were conventional, and the illuminations were in flat tints, with little shade and no shadow; whilst in those of a later period highly-finished representations of natural flowers were used as ornament, casting their shadows on the page. ON THE COLOURING OF MORESQUE ORNAMENT. When we examine the system of colouring adopted by the Moors, we shall find, that as with form, so with colour, they followed certain fixed principles, founded on observations of nature’s laws, and which they held in common with all those nations who have practised the arts with success. In all archaic styles of art, practised during periods of faith, the same true principles prevail ; and although we find in all somewhat of a local or temporary character, we yet discern in all much that is eternal and immutable ; the same grand ideas embodied in different forms, and expressed, so to speak, in a different language. 10. The ancients always used colour to assist in the development of fo'inn, always employed it as a further means of bringing out the constructive features of a building. 70 MORESQUE ORNAMENT. Thus, iu the Egyptian column, the base of which represented the root — the shaft, the stalk — the capital, the buds and flowers of the lotus or papyrus, the several colours were so applied that the appearance of strength in the column was increased, and the contours of the various lines more fully developed. In Gothic architecture, also, colour was always employed to assist in developing the forms of the panel-work and tracery ; and this it effected to an extent of which it is difficult to form an idea, in the present colourless condition of the buildings. In the slender shafts of their lofty edifices, the idea of elevation was still further increased by upward-running spiral lines of colour, which, while adding to the apparent height of the column, also helped to define its form. In Oriental art, again, we always find the constructive lines of the building well defined by colour; an apparent additional height, length, breadth, or bulk always results from its judicious application ; and with the ornaments in relief it developes constantly new forms which would have been altogether lost without it. The artists have in this but followed the guiding inspiration of Nature, in whose works every transition of form is accompanied by a modification of colour, so disposed as to assist in producing distinctness of expression. For example, flowers are separated by colour from their leaves and stalks, and these again from the earth in which they grow. So also in the human figure every change of form is marked by a change of colour; thus the colour of the hair, the eyes, the eyelids, and lashes, the sanguine complexion of the lips, the rosy bloom of the cheek, all assist in producing distinctness, and in more visibly bringing out the form. We all know how much the absence or im- pairment of these colours, as in sickness, contributes to deprive the features of their proper meaning and expression. Had nature applied but one colour to all objects, they would have been indistinct in form as well as monotonous in aspect. It is the boundless variety of her tints that perfects the modelling and defines the outline of each ; detaching equally the modest lily from the grass from which it springs, and the glorious sun, parent of all colour, from the firmament in which it shines. 11. The colours employed by the Moors on their stucco-work were, in all cases, the primaHes, blue, red, and yellow {gold). The secondary colowrs, purple, green, and orange, occur only in the Mosaic dados, which, being’ near the eye, formed a point of repose from the more brilliant colouring above. It is true that, at the present day, the grounds of many of the ornaments are found to be green ; it will always be found, however, on a minute examination, that the colour originally employed was blue, which, being a metallic pigment, has become green from the effects of time. This is proved by the presence of the particles of blue colour, which occur everywhere in the crevices : in the restora- tions, also, which were made by the Catholic kings, the grounds of the ornaments were repainted both green and purple. It may be remarked that, among the Egyptians and the Greeks, the Arabs and the Moors, the primary colours were almost entirely, if not exclusively, employed, during the early periods of art ; whilst, during the decadence, the secondary colours became of more importance. Thus, in Egypt, in Pharaonic temples, we find the primary colours predominating ; in the Ptolemaic temples, the secondary; so also on the early Greek temples are found the primary colours, whilst at Pompeii every variety of shade and tone was employed. In modern Cairo, and in the East generally, we have green constantly appearing side by side with red, where blue would have been used in earlier times. This is equally true of the works of the Middle Ages. In the early manuscripts and in stained glass, though other colours were not excluded, the primaries were chiefly used; whilst in later times we have every variety of shade and tint, but rarely used with equal success. 12. With the Moors, as a general rule, the primary colour's were used on the upper portions 71 MORESQUE ORNAMENT. of objects, the secondary and tertiary on the lower. This also appears to he in accordance with a natural law ; we have the primary blue in the sky, the secondary green in the trees and fields, ending with the tertiaries on the earth ; as also in flowers, where we generally find the primaries on the buds and flowers, and the secondaries on the leaves and stalks. The ancients always observed this rule in the best periods of art. In Egypt, however, we do see occasionally the secondary green used in the upper portions of the temples, but this arises from the fact, that ornaments in Egypt were symbolical; and if a lotus leaf were used on the upper part of a building, it would necessarily be coloured green; but the law is true in the main; the general aspect of an Egyptian temple of the Pharaonic period gives the primaries above and the secondaries below ; but in the buildings of the Ptolemaic and Eoman periods more especially, this order was inverted, and the palm and lotus leaf capitals give a superabundance of green in the upper portions of the temples. In Pompeii we find sometimes in the interior of the houses a gradual gradation of colour downwards from the roof, from light to dark, ending with black; but this is by no means so universal as to convince us that they felt it as a law. We have already shown in Chapter V. that there are many examples of black immediately under the ceiling. 13. Although the ornaments which are found in the Alhambra, and in the Court of the Lions especially, are at the present day covered with several thin coats of the whitewash which has at various periods been applied to them, we may be said to have authority for the whole of the colouring of our reproduction ; for not only may the colours be seen in the interstices of the ornaments in many places by scaling off the whitewash, but the colouring of the Alhambra was carried out on so perfect a system, that any one who will make this a study, can with almost absolute certainty, on being shown for the first time a piece of Moorish ornament in white, define at once the manner in which it was coloured. So completely were all the architectural forms designed with reference to their subsequent colouring, that the surface alone will indicate the colours they were destined to receive. Thus, in using the colours blue, red, and gold, they took care to place them in such positions that they should be best seen in themselves, and add most to the general effect. On moulded surfaces they placed red, the strongest colour of the three in the depths, where it might be softened by shadow, never on the surface; blue in the shade, and gold on all surfaces exposed to light; for it is evident that by this arrangement alone could their true value be obtained. The several colours are either separated by white bands, or by the shadow caused by the relief of the ornament itself — and this appears to be an absolute principle required in colouring — colours should never be allowed to impinge upon each other. 14. In colouring the grounds of the various diapers the blue always occupies the largest area; and this is in accordance with the theory of optics, and the experiments which have been made with the prismatic spectrum. The rays of light are said to neutralise each other in the proportions of 3 yellow, 5 red, and 8 blue; thus, it requires a quantity of blue equal to the red and yellow put together to produce a harmonious effect, and prevent the predominance of any one colour over the others. As in the “Alhambra,” yellow is replaced by gold, which tends towards a reddish- yellow, the blue is still further increased, to counteract the tendency of the red to overpower the other colours. INTEKLACED PATTERNS. We have already suggested, in Chapter IV., the probability that the immense variety of Moorish ornaments, which are formed by the intersection of equidistant lines, could be traced through the 72 MORESQUE ORNAMENT. Arabian to the Greek fret. The ornaments on Plate XXXIX. are constructed on two general principles; Nos. 1—12, 16—18, are constructed on one principle (Diagram No, 1), No. 14 on the other (Diagram No. 2). In the first series the lines are equidistant diagonally crossed by horizontal and ilgpSiiPHSiliKgpl! iiiliiliiisisSstili! KiKSliSiiSnSIKSWSKii ■ ■ BTB a B'B B k'BBKM ■ m ■ W B W'B ssasasasHsasliiifSiK llSilswi! lisisKW laiisisi iililil'!* iilliiiiS lliirSiiii isisiiSsisSsSiisS SliiiaiiiaiiijKwiiSS iiiiiiizii •IB la^B B v.B B^«B »;« 1 ikTBBBraaKBBa.^aKBBWBaacBiKaav.B asa!a!ai»8ai5»!aisa iaTBBK«B».;«BaMB»iaBR:«a»:«aKBBK« BtlB B BTB BBIB a KB a KBI Diagram No. 1. Diagram No. 2 . perpendicular lines on each square. But the system on which No. 14 is constructed, the perpendicular and horizontal lines are equidistant, and the diagonal lines cross only each alternate square. The number of patterns that can be produced by these two systems would appear to be infinite ; and it will be seen, on reference to Plate XXXIX., that the variety may be still further increased by the mode of colouring the ground or the surface lines. Any one of these patterns which we have engraved might be made to change its aspect, by bringing into prominence different chains or other general masses. LOZENGE DIAPERS. The general effect of Plate XLI. and XLI*. will, we think, at once justify the superiority we have claimed for the ornament of the Moors. Composed of but three colours, they are more harmonious and effective than any others in our collection, and possesses a peculiar charm which all the others fail to approach. The various principles for which we have contended, the constructive idea whereby each leading line rests upon another, the gradual transitions from curve to curve, the tangential curvatures of the lines, the flowing off of the ornaments from a parent stem, the tracing of each flower to its branch and root, the division and subdivision of general lines, will readily be perceived in every ornament on the page. SQUARE DIAPERS. The ornament No. 1, on Plate XLII., is a good example of the principle we cofitend for, that to produce repose the lines of a composition should contain in equilibrium the straight, the inclined, and the curved. We have lines running horizontally, perpendicularly, and diagonally, again contrasted by circles in opposite directions. So that the most perfect repose is obtained, the tendency of the eye to run in any direction is immediately corrected by lines giving an opposite tendency, and wherever the eye strikes upon the patterns it is inclined to dwell. The blue ground of the inscriptions and ornamental panels and centres, being carried over the red ground by the blue feathers, produces a most cheerful and brilliant effect. The leading lines of the ornaments Nos. 2-4, Plates XLII. and XLII*., are produced in the same way as the interlaced ornaments on Plate XXXIX. In Nos. 2 and 4 it will be seen how the repose 73 u MORESQUE ORNAMENT. of tke pattern is obtained by the arrangement of tbe coloured grounds ; and bow, also, by this means an additional pattern besides that produced by form results from the arrangement of the colours. Pattern No. 6, Plate XLIIf., is a portion of a ceiling, of which there are immense varieties in the Alhambra, produced by divisions of the circle crossed by intersecting squares. It is the same principle which exists in the copy from the illuminated Koran, Plate XXXIV., and is also very common on the ceilings of Arabian houses. The ornament No. 5, Plate XLIIf., is of extreme delicacy, and is remarkable for the ingenious system on which it is constructed. All the pieces being similar, it illustrates one of the most important principles in Moorish design, — one which more perhaps than any other contributed to the general happy result, viz., that by the repetition of a few simple elements the most beautiful and complicated effects were produced. However much disguised, the whole of the ornamentation of the Moors is constructed geometrically. Their fondness for geometrical forms is evidenced by the great use they made of mosaics, in which their imagination had full play. However complicated the patterns on Plate XLIII. may appear, they are all very simple when the principle of setting them out is once understood. They all arise from the intersection of equidistant lines round fixed centres. No. 8 is constructed on the principle of Diagram No. 2, cited on the other side, and is the principle which produces the greatest variety; in fact, geometrical combinations on this system may be said to be infinite. 74 MORESQUE N° I MAURIS CH MAURESQUES PL XXXIX TAEEL XXXIX. tmm 16 6 •V'M MORESQUE MALI R I S C H MAUR tSQUE'? • r Wi\jitWir?inMi' JliiiiiiilyMillllii il j jiil i iiiiiii^a,iiiyd MAURisGH MORESQUE. N° 5 mauresqurs TAfEL^CUIl PLXLIII. ®*S*z*S H ^ ^ 'm I fur fw fT'i, 000m mm^m mmmm Chapter XI.— Plates 44, 45, 46, 47, 47*, 48. PERSIAN ORNAMENT. PLATES XLIV., XLY., XLVI. Ornaments from Persian MSS. in the British Museum. PLATES XLYII., XLYII*. From a Persian Manufacturer’s Pattern-Book, South Kensington Museum. PLATE XLYIII. From a Persian MS. South Kensington Museum. The Mohammadan architecture of Persia, if we may judge from the representations published in Flandin and Coste’s ^‘Voyages en Perse,” does not appear to have ever reached the perfection of the Arabian buildings of Cairo. Although presenting considerable grandeur in the main features, the general outlines are much less pure, and there would appear to be a great want of elegance in all the constructive features as compared with those of Cairo. Their system of ornamentation also appears to us much less pure than the Arabian and Moresque. The Persians, unhke the Arabs and the Moors, were free to introduce animal life, and this mixing up of subjects drawn from real life in their decoration led to a much less pure style of ornament. With the Arabs and Moors, ornaments with their inscriptions had to supply every want, and therefore it became of more importance in their structures, and reached a higher point of elaboration. Persian ornament is a mixed style; combining the conventional, which is similar to the Arabian, and probably derived from a common origin, with an attempt at the natural which sometimes has influenced both the Arabian and Turkish styles, and is even felt in portions of the Alhambra. The great attention paid to the illuminating of manuscripts 75 PERSIAN ORNAMENT. in Persia, which, doubtless, were widely disseminated in Mohammadan countries, would readily spread the influence of this mixed style. The decorations of the houses of Cairo and Damascus, the mosques and fountains of Constantinople more especially, exhibit this mixed style; groups of natural flowers are constantly found growing from a vase and enclosed in panels of conventional Arabian ornament. The ornament of modern India also feels this ever-present influence of the Persian mixed style. In a book-cover fropa the India House (Plates LIII. and LIV.) is an example of this; the outside is treated in the ^ure Arabian manner, whilst the inside (Plate LIV.) is quite Persian in character. The ornaments on Plate XLIV., from illuminated MSS. in the British Museum, present also the mixed character we have referred to. The geometrical patterns are purely conventional ornament, and have great aflSnity with the Arabian, but are less perfect in distribution. Nos. 1-10, on the contrary, are from backgrounds of pictures, representing tapestry on the walls ; they possess great elegance, and the masses are well contrasted with the grounds. The patterns on Plate XLV. are chiefly representations of pavements and dados, and probably were intended for glazed tiles so abundantly used by the Persians. Compared with the Arabian and Moresque mosaics, they exhibit a marked inferiority, both in the distribution of form and in the arrangement of colour; it will be observed that, throughout our Persian subjects, the secondary and tertiary colours are much more dominant than in the Arabian (Plate XXXIV.), or in the Moresque, where blue, red, and gold, are the prevailing harmonies, and, as may be seen at a glance, with much- increased effect. The ornaments on Plate XLVI. have a much greater affinity with the Arabian: Nos. 7, 16, 17, 21, 23-25, are very common ornaments for the heads of chapters in Persian MSS., indeed there is but little variety to be found in these, numerous as they are. Compared with the Arabian MSS. (Plate XXIV.), a great similarity will be found in all the leading lines of the construction of the ornaments and also in the surface decoration of the ornaments themselves ; but the masses are much less evenly distributed. However the same general principles prevail. Plates XLVII. and XLVII*. are arranged from a very curious Persian book at South Kensington Museum, which appears to be a manufacturer’s pattern-book. The designs exhibit much elegance, and there is great simplicity and ingenuity displayed in the conventional rendering of natural flowers. Both these Plates and Plate XLVIII. are very valuable, as showing the extreme limit of this conven- tional rendering, reached, but not exceeded. When natural flowers are used as decoration, and subjected to a geometrical arrangement, they can have neither shade nor shadow, as was the case with the later MSS. of the Mediaeval School, see Plate LXXIII. ; without falling under that reproach so justly due to the floral papers and floral carpets of modern times. The ornament at the top of Plate XLVIII., which forms the title-page to the book as well as the borders throughout, present that mixed character of pure ornament, arranged in conjunction with the ornamental rendering of naturai forms, which we have considered as characteristic of the Persian style, and which, we think, renders it so much inferior to the Arabian and the Moresque. 7 ? TAFEL XLIY IT <8 19 PERSISCH. PERSIAN N° 3 PERSE S. TAFEL XLYI PL XLYi. PER SI'S CH PERSIAN N°4* PER S £ ' PLXLYIi' TAFEL XlVn* PERSISCH. PERSIAN N° 5 TAFEL XLYIU PERSE S. PL XLYIU Chapter XII.— Plates 49, 50, 51, 52, 53, 53% 54, 54*, 55. INDIAN ORNAMENT, FROM THE EXHIBITIONS OF 1851 AND 1855. ^ r-9iiir38pii< PLATE XLIX. Ornaments from Works in Metal, exlubited in the Indian Collection in 1851. PLATES L., LL, LII. Ornaments from Embroidered and Woven Eabrics, and Paintings on Vases, exhibited in the Indian Collection in 1851, and now at South Kensington Museum. PLATES LIIL, LIII*., LIV., LIV*. Specimens of Painted Lacquer-work, from the Collection at the India House. PLATE LV. Ornaments from Woven and Embroidered Fabrics, and Painted Boxes, exhibited in the Indian Collection at Paris in 1855. C 8- =■ The Exhibition of the Works of Industry of all Nations in 1851 was barely opened to the public ere attention was directed to the gorgeous contributions of India. Amid the general disorder everywhere apparent in the application of Art to manufactures, the presence of so much unity of design, so much skill and judgment in its application, with so much of elegance and refinement in the execution, as was observable in all the works, not only of India, but of all the other Mohammadan contributing countries, — Tunis, Egypt, and Turkey, — excited a degree of a.ttention from artists, manufacturers, and the public, which has not been without its fruits. Whilst in the works contributed by the various nations of Europe, there was everywhere to be observed an entire absence of any common principle in the application of Art to manufactures, — whilst from one end to the other of the vast structure there could he found but a fruitless struggle after novelty, irrespective of fitness, that all design was based upon a system of copying and misapplying 77 X INDIAN ORNAMENT. the received forms of beauty of every bygone style of Art, without one single attempt to produce an Art in harmony with our present wants and means of production — the carver in stone, the worker in metal, the weaver and the painter, borrowing from each other, and alternately misapplying the forms peculiarly appropriate to each— there were to be found in isolated collections at the four corners of the transepts all the principles, all the unity, all the truth, for which we had looked elsewhere in vain, and this because we were amongst a people practising an art which had grown up with their civilisation, and strengthened with their growth. United by a common faith, their art had necessarily a common expression, this expression varying in each according to the influence to which each nation was subject. The Tunisian still retaining the art of the Moors, who created the Alhambra ; the Turk exhibiting the same art, but modified by the character of the mixed population over which they rule ; the Indian uniting the severe forms of Arabian art with the graces of Persian refinement. All the laws of the distribution of form which we have already observed in the Arabian and Moresque Ornament are equally to be found in the productions of India. From the highest work of embroidery, or most elaborate work of the loom, to the constructing and decorating of a child’s toy or earthen vessel, we find everywhere at work the same guiding principles, — there is always the same care for the general form, the same absence of all excrescences or superfluous ornament ; we find nothing that has been added without purpose, nor that could be removed without disadvantage. The same division and subdivision of their general lines, which form the charms of Moresque ornament, is equally to be found here ; the difference which creates the style is not one of principle, but of individual expression. In the Indian style ornaments are somewhat more flowing and less convention- alised, and have, doubtless, been more subjected to direct Persian influence. The ornaments on Plate XLIX., are chiefly taken from Hookhas, of which there was an immense variety exhibited in 1851, and all remarkable for great elegance of outline, and for such a judicious treatment of the surface decoration that every ornament tended to further develope the general form. It will be seen that there are two kinds of ornament, — the one strictly architectural and conven- tional: such as Nos, 1, 4, 5, 6, 8, which are treated as diagrams ; and the other, such as Nos. 13, 14, 15, in which a more direct imitation of nature is attempted : these latter are to us very valuable lessons, showing how unnecessary it is for any work of decoration to more than indicate the general idea of a flower. The ingenious way in which the full-blown flower is shown in No. 15, in three positions in Nos. 14 and 15, the folding back of the leaf in No. 20, are very suggestive. The intention of the artist is fully expressed by means as simple as elegant. The unity of the surface of the object decorated is not destroyed, as it would be by the European method of making the flower as near like a natural flower as possible, with its own light and shade and shadow, tempting you to pluck it from the surface. On the Persian, Plate XLVII., will be seen a similar treatroent of natural flowers ; the comparison shows how much of Persian influence there is in this floral style of India. In the apphcation of the various ornaments to the different portions of the objects the greatest judgment is always shown. The ornament is invariably in perfect scale with the position it occupies ; on the narrow necks of the Hookhas are the small pendent flowers, the swelling forms of the base are occupied by the larger patterns ; at the lower edge, again, appear ornaments having an upward tendency, and, at the same time, forming a continuous line round the form to prevent the eye running out of it. Whenever narrow flowing borders are used, as in No. 24, they are contrasted by others with lines flowing in an opposite direction ; the general repose of the decoration is never for a moment lost sight of. In the equal distribution of the surface ornament over the grounds, the Indians exhibit an instinct 78 INDIAN OENAMENT. and perfection of drawing perfectly marvellous. The ornament No. 1, on Plate L., from an embroidered saddle-cloth, excited universal admiration in 1851. The exact balance obtained by the gold embroidery on the green and red grounds, was so perfect that it was beyond the power of a European hand to copy it with the same complete balance of form and colour. The way in which the colours are fused in all their woven fabrics, so as to obtain what they always appear to seek, viz., that coloured objects when viewed at a distance should present a neutralised bloom, is very remarkable, A due regard to economy in the production of our Plates has necessarily limited the number of printings, and we have not always therefore been able to obtain the proper balance of colour. The Indian collection at South Kensington Museum should be visited and studied by all in any way connected with the production of woven fabrics. In this collection will be found the most brilliant colours perfectly harmonised — it is impossible to find there a discord. All the examples show the nicest adjustment of the massing of the ornament to the colour of the ground ; every colour or tint from the palest and most delicate to the deepest and richest shades, receiving just the amount of ornament that it is adapted to bear. The following general rules, which are applicable to all woven fabrics, may be observed : — 1. When gold ornaments are used on a coloured ground, where gold is used in large masses, there the ground is darkest. Where the gold is used more thinly, there the ground is lighter and more delicate. 2. When a gold ornament alone is used on a coloured ground, the colour of the ground is carried into it by ornaments or hatchings worked in the ground-colours on the gold itself. 3. When ornaments in one colour are on a ground of a contrasting colour, the ornament is separated from the ground by an edging of a lighter colour, to prevent all harshness of contrast. 4. When, on the contrary, ornaments in a colour are on a gold ground, the ornaments are separated from the gold ground by an edging of a darker colour, to prevent the gold overpowering the ornament. — See No. 10, Plate L. 5. In other cases, where varieties of colour are used on a coloured ground, a general outline of gold, of silver, or of white or yellow silk, separates the ornament from the ground, giving a general tone throughout. The carpets and low-toned combinations of colour, a black general outline is used for this purpose. The object always appears to be, in the woven fabrics especially, that each ornament should be softly, not harshly, defined; that coloured objects viewed at a distance should present a neutralised bloom; that each step nearer should exhibit fresh beauties ; and a close inspection, the means whereby these effects are produced. In this they do but carry out the same principles of surface decoration which we find in the architecture of the Arabs and Moors. The spandrU of a Moorish arch, and an Indian shawl, are constructed precisely on the same principles. The ornament on Plate LIII., from a book-cover at the India House, is a very brilliant example of painted decoration. The general proportions of the leading lines of the pattern, the skilful distribu- tion of the flowers over the surface, and, notwithstanding the intricacy, the perfect continuity of the Unes of the stalks, place it far before any European effort of this class. On the inside of the same cover, Plate LIV., the ornaments are less conventional in their treatment ; but how charmingly is observed the limit of the treatment of flowers on a flat surface ! This book-cover offers in itself a specimen of two marked styles, the outside Plate LIII., being after the Arabian manner, and the inside after the Persian. 79 INDISCH INDIAN N" I INDIEN S INDISCH INDIAN 4 INDIEN S . PL U) INDIAN 5 INDISGH N D I E N S TA-FEL UII IKD.SCH INDIAN N° 7 I N D I £ N S Chapter XIII.— Plates 56, 57, 58. HINDOO ORNAMENT. PLATE LVI. Ornameuts from a Statue in Basalt at tbe House of the Eoyal Asiatic Society. PLATE 1. Burmese, of Glass. — Crystal Palace. 2. Burmese Shrine. C. P. 3. Burmese Standard. C. P. 4-6. From Burmese Shrine. C. P. LVII. 7-10, 12-17. Ornaments from the Copies of the Paintings on the walls of the Caves at Ajunta. — Crystal Palace. 11. Burmese, from a Monastery near Prome. — C. P. 1. Burmese. — East India House. 2, 3. Burmese Shrine. — Crystal Palace. 4. Burmese Gilt Chest. — C. P. 5. Hindoo. — United Service Museum 6-9. Hindoo Ornaments. — E. I. H. 10. Burmese. — C. P. 11. Hindoo. — U. S. M. We have not been able, with the materials at command in this country, to procure sufficient illustrations for a fair appreciation of the nature of Hindoo ornament. In the works hitherto published on the ancient architecture of India, sufficient attention has not been directed to the ornamental portions of the buildings to enable us to recognise the true character of Hindoo ornament. In early publications on the art of Egypt all the works of sculpture and ornament were so falsely rendered, that it has taken considerable time for the European public to become persuaded that there existed so much grace and ' refinement in the works of the Egyptians. The Egyptian remains, however, which have been transported to this country, the casts of others existing in Egypt, and the more trustworthy representations which have of late been published, have placed this beyond doubt, and Egyptian art is taking its true place in the estimation of the public. 81 PLATE LVIIL 12. Burmese.— Britkh Museum. 13. Hindoo.— E. I. H. 14. Hindoo.— U.S.M. 15. Hindoo. — E. I. H. 16-19, 21. Burmese. — C. P. 20, 22-25. Burmese. — U. S. M. 26. Burmese. — 0. P. Y HINDOO ORNAMENT. When the same thing shall have been done for the ancient architecture of India, we shall be in a better position than we are at present to form an opinion how far it is entitled to take rank as a really fine art, or whether the Hindoos are only heapers of stones, one over the other, adorned with grotesque and barbaric sculpture. Had we possessed only picturesque views of the Parthenon and the Temples of Balbeck and Palmyra, we should unhesitatingly have said that the Romans were far greater architects than the Grreeks. But the contour of a single moulding from the Parthenon would at once reverse the judgment, and proclaim loudly that we were viewing the works of a people who had reached the highest point in civilisation and refinement. Although ornament is most properly an accessory to architecture, and should never be allowed to usurp the place of proper structural features, or to overload or to disguise them, it is in all cases the very soul of an architectural monument; and by the ornament alone can we judge truly of the amount of care and mind which has been devoted to the work. All else in any building may be the result of rule and compass, but by the ornament of a building we can best discover how far the architect was at the same time an artist. No one can peruse the Essay on Hindoo Architecture by Ram Raz * without feeling that a higher state of architectural perfection has been reached than the works published up to the present time would lead us to believe. In this work not only are precise rules laid down for the general arrangement of structures, but also minute directions are given for the divisions and subdivisions of each ornament. One of the precepts quoted by Ram Raz deserves to be cited, as showing how much the general perfection was cared for : ” Woe to them who dwell in a house not built according to the proportions of symmetry. In building an edifice, therefore, let aU its parts, from the basement to the roof, be duly considered.” Among the directions for the various proportions of columns, bases, and capitals, is a rule for finding the proper diminution of the upper diameter of a column in proportion to the lower. Ram Raz says, that the general rule adopted by the Hindoo architects was to divide the diameter of the column at the base by as many parts as there were diameters in the whole height of the column, and that one of these parts was invariably deducted to form the upper diameter. From which it is apparent that the higher the column the less it will diminish; and that this was done because the apparent diminution of the diameter in columns of the same proportion is always greater according to the height. The best specimens of Hindoo ornament we have been able to procure are represented in Plate LVI., from a statue of Surga, or the Sun, in basalt, at the house of the Asiatic Society, and supposed to belong to a period between the fifth and ninth century A.D. The ornaments are very beautifully executed, and evidently betray Greek influence. The ornament No. 8 represents the lotus, seen as it were in plan, with the buds in side elevation : it is held in the hand of the god. In the sacred books quoted by Ram Raz are several directions to ornament the various archi- tectural members with lotuses and jewels ; which seem to be the chief types of the decorations on the mouldings. The architectural features of Hindoo buildings consist chiefly of mouldings heaped up one over the other. Definite instructions are quoted by Ram Raz for the varying proportions of each, and it is evident that the whole value of the style will consist in the more or less perfection with which these transitions are effected ; but, as we said before, we have no opportunity of judging how far this is the case. * “ Histoiy of the Architecture of the Hindoos.” By Ram Raz. London, 1834. 82 HINDOO ORNAMENT. On Plate LVII. we have gathered together all the examples of decorative ornament that we could find on the copies of the paintings from the Caves of Ajunta, exhibited by the East India Company at the Crystal Palace. As these copies, notwithstanding that they are said to be faithful, are yet by a European hand, it is difficult to say how far they may be relied upon. In the subordinate portions, such as the ornaments, at all events, there is so little marked character, that they might belong to any style. It is very singular, that in these paintings there should be so little ornament ; a peculiarity that we have observed in several ancient paintings in the possession of the Asiatic Society. There is a remarkable absence of ornament even on the dresses of the figures. 83 HINDOO N ° I HiNDons oooooo ooooooo HINDOO N° 2 HINDOUS. PL. mi HINDUS. TAFEL u;i] 10 HiNDOUS HINDOO N °3 HINDHS TAFEl. LYU; jiimoimooooooT , '^V>> vr' 0 PL.LYIII ^ /:'|V /j'N.1 /'ivv /■ |\N /' v. /'X //i\, /-y f vD \y/) y-D ^ a v /' ya' \ AiV) Y/i'Y. //,M a7v jMALEriEL STAINED GLASS^ yitraux pei-nts 24 lUUMINIRTE MANUSCRIFTE. I LLU M I NATED M.S.S. N° E MANUSCRIPTS ENLUMINE TAPEL IXXll PU LXXIl umHINIRTE MAHnSCHIPTE ILLUMINATED ML.S N°S. MANUSCRIPTS ENLIIMINES TAFEl, WXIU. o- Chapter XVII.— Plates 74, 75, 76, 77, 78, 79, 80, EENAISSANCE OKNAMENT. 81, 82. PLATE LXXIV. 1, 8, 9. Bas-reliefs from the Church of Sta. Maria dei Mira- coli, Venice. 2. Bas-relief from the Scuola di Sau Marco, Venice. 3. Bas-relief forming the continuation upwards of Fig. 2. 4, 6. Bas-reliefs from the Church of San Michele in Murano Venice. 5, 7. Bas-reliefs from the Scala dei Giganti, Venice. PLATE LXXV. 1, 2. From a Collection of Casta taken under the super- intendence of Professor Varny, from the principal Cinque-cento Monuments of Genoa. 3. From the first Ghiberti Gate of the Baptistery, Florence. 4, 5, 8, 9, 11. From Genoa. 6. From Venice. 7. From the Church of Santi Giovanni e Paolo, Venice. 10. From the H6tel Bourgtheroulde, Eouen. PLATE LXXVI. Bas-relief by Andrea Sansovino, from the Church of Sta. Maria del Popolo, Eome. Bas-relief from the Church of Sta. Maria dei Miracoli, Venice. Bas-relief from the H6tel Bourgtheroulde, Rouen. Bas-relief from a Collection of Casts of the best Cinque- cento Ornaments of Genoa, taken under the superin- tendence of Professor Varny, 5, 7, 8, 10. Bas-reliefs from Genoa. 6. Bas-relief from the Martinengo Tomb, Brescia. 9. Bas-relief from the Base of the “ Trois Graces of Germain Pilon, in the Louvre. PLATE LXXVII. 1-3. Ornaments enamelled on Copper in the early Li- moges Champlev6 style, from the HOtel Cluny Museum, Paris. 4-8. Ditto, of a later period. 9. Ornaments from the background of a Picture, in the Hotel Cluny. 10, 11. Enamels on Gold Ground, from the Louvre. 12. Silver Inlay in Ivory, of the Sixteenth Century, from the Hotel Cluny. 13. From a Casket in the HOtel Cluny. 14. From a Powder-horn in Iron of the Sixteenth Cen- tury, in the HOtel Cluny. 15-17. Similar objects in Boxwood, from the same Museum. 18-20. From Sixteenth Century Limoges Enamels, in the same Museum. 21. From ditto, in the Louvre. 22-24. Enamels on Gold Ground, Sixteenth Century, Louvre. 25. Portion of an Ebony Cabinet of the Sixteenth Cen- tury, in the HOtel Cluny. 26. Inlaid Ornament on a Dagger Sheath of the Sixteenth Century, in the HOtel Cluny. 27, 28. From Pottery of the Sixteenth Century, in the Louvre. 29. Limoges ChamplevO Enamel on Copper, from the Hotel Cluny. 30. Painted Ornaments, HOtel Cluny. 31. From the Armour of Henri III., in the Louvre. 32. A Metal Plate in the same Museum. 33-35. From Metal Work, in the Louvre. 36. From the Armour of Fran§ois II., in the Louvre. 37-39. Repousse Ornaments in Copper, from the HOtel Cluny. 40, 41. Limoges ChamplevO Enamel, from the same Museum. 42-44. From Goldsmiths’ Work of the Sixteenth Century, in the Louvre. 45, 46. From a Picture in Limoges Painted Enamel, Six- teenth Century, in the HOtel Cluny. 47. Ornament in Copper, from the above. 48- Ivory Inlay in Ebony, from the above. 49. Painted Ornament, from the above. 50-63. Limoges ChamplevO Enamel, from the above. 54-56. From Accessories to Pictures, from the above. 57-61. Limoges Champleve Enamel. 107 RENAISSANCE ORNAMENT. PLATE LXXYIIL 1-36. Ornaments taken from Specimens of Hispano-Arabic, French, and Italian Earthenware, preserved in the South Kensington Museum, and principally from the Majolican Wares of Pesaro, Gubbio, Urbino, Castel Durante, and other Italian towns of the Fifteenth, Sixteenth, and Seventeenth Centuries. PLATE LXXIX. 1-3. Ornaments selected from the faience, or enamelled Earthenware, of Bernard de Palissy, in the Hotel Cluny. 4-10. From Specimens of Majolica, in the H6tel Cluny. 11-13. From faience of the Fifteenth Century, in the H6tel Cluny. 14-18, 21. From faience of the Sixteenth Century, in the Louvre. PLATE 1, 2. Ornaments ivom faience. 3-6. Ornaments from faience of the Sixteenth Century. 7-10. Ornaments from faience of the Seventeenth Cen- tury. 11, 12. From faience with Metallic Lustre. 13. From a Vase in Venetian Glass of the Sixteenth Century. 14-21. 'Svo'ca. faience of the Sixteenth Century. 22, 23. From faience of an Earlier Date. 19, 20. From Porcelain of the Seventeenth Century, in the Louvre. 22, 23. From the German Pottery, m grhs, with Painted Glaze of the Sixteenth Century, in the H6tel Cluny. 24-33. From Earthenware, French, Spanish, and Italian, in the Hotel Cluny. 34. From the Louvre. LXXX. 24-27. From Qrh Flamand, or Earthenware. 28-32, From faience of the Sixteenth Century. 33. From a Carved Wood Panel of the Seventeenth Century. 34-38. From Enamelled Earthenware. 39—42. From Silk Embroidery on Velvet. N.B. — The whole of the Specimens on this Plate have been derived from the H6tel Cluny, Paris. 1. From a Sideboard carved in wood, dated 1554, in the H6tel Cluny. 2. Wood Panels of the Sixteenth Century, in the H6tel Cluny. 3. From an Oak Chair-back, in the H6tel Cluny. 4-6. From Carved Wood-stalls of the Fifteenth Century, in the HOtel Cluny. 7-10, 25, 26, 35, 36. From Furniture, in the HOtel Cluny. 11. End of a Beam of the end of the Fifteenth Century, in the H6tel Cluny. 12, 13, 20, 21, 39, 40. From Furniture of the Sixteenth Cen- tury, in the H6tel Cluny. 14, 15. From Furniture of the Fifteenth Century, in the Hotel Cluny. 16. From a Sideboard, in the HOtel Cluny, Shutter Panels of the end of the Fifteenth Century, in the H6tel Cluny. 18. Carved Ornament, from the Louvre. 19. From a Boxwood Comb, in the Hotel Cluny. 22. Stone Balustrading, from the Chateau d’Anet, 23. Stone Carving, from the Louvre. 24. From a Chimney-piece, in the H6tel Cluny. 27-30. Carving in Marble from the celebrated Basin of the Fountain of the Chateau Gaillon, now in the Louvre. 31, 32. Stone carving. Seventeenth Century, in the Louvre. 33. Wood-carving from the H6tel Cluny. 34, 38. From the Fountain of the Chateau Gaillon, Louvre. 37. From the Stock of an Arquebuss of the Sixteenth Century, in the Hdtel Climy. PLATE LXXXL 17. PLATE LXXXII. 1-9. Carved Ornament, from Oak Furniture of the Six- teenth Century, in the Hotel Cluny. 10, 11, 19, 34. From the Bed of Franqois I., in the Hotel Cluny. 12 , 13, 14, 32, 33. From Oak Furniture of the Sixteenth Century, in the H6tel Cluny. 108 15-17. From a Sideboard of the Fifteenth Century. 18. From an Oak Sideboard, dated 1524, in the H6tel Cluny. 20-29. From Furniture of the Sixteenth Century, in the Hotel Cluny. 30, 31. Panels of Shutters of the end of the Fifteenth Cen- tury, in the Hotel Cluny. EENAISSANCE ORNAMENT. If two intelligent students of Italian Art and Literature diligently set themselves to trace, the one the latest date at which the direct, though lingering, light of Roman greatness waned to its feeblest glimmer in the land over which it had once shed its dazzling rays, and the other the earliest effort made to excite a veneration for what most historians declare to have almost utterly died out in the lapse of ages — classical beauty — there is little doubt that they would not only meet, but cross one another, in the progress of their researches. The truth is, that the material monuments of the ancient Romans, scattered thickly over the soil of Italy, were so substantial and majestic, that it was impossible to live under their shadow and to forget them. Fragments of exquisite beauty, in stone, bronze, and marble, were to be had for the trouble of turning up the soil that scarcely covered them; and thus they were, from time to time, pressed into service for tombs, and as accessories in buildings, in the construction of which the principles of Art to which those fragments owed their beauty had been entirely lost sight of. Hence, the G-othic style was at once slow to take root in Italy, and destined to bloom brilliantly, but for a short season. Almost concurrently with the introduction of the pointed arch into Northern Italy by an Englishman, in the construction of St. Andrea, at Vercelli, early in the thirteenth century, and with the German works of Magister Jacobus, at Assisi, a protest was commenced in favour of the ancients and their arts, by that great reviver of antique sculpture, Nicola Pisano. The close of the thirteenth century was further marked by a complete revolution in the world of letters. Dante, in his time, was scarcely less known as a Christian poet than as an emulator of the great Mantuan, and a profound student in classical learning. In the fourteenth century, Petrarch and Boccaccio, intimate friends, spent long and laborious lives, not in writing Italian poetry or prose, as is often fancied, but in labouring incessantly in the preservation and restoration to the world of the long-lost texts of the Roman and G-recian authors. Cino da Pistoia and other learned commentators and jurists, brought into fashion the study of the great “ Corpus ” of ancient law and maintained academies in which it was adopted as a text. Boccaccio it was who first gave to Italy a lucid account of Heathen Mythology, and who first instituted a chair for the study of the Grecian language at Florence, bringing over Leontius Pilatus, a learned Greek, from Constantinople, to be the first professor. These efforts at a revival of classical learning were seconded by a numerous band of notables, among whom the names of John of Ravenna (Petrarch’s pupil), Lionardo Aretino, Poggio Bracciolini, iEneas Sylvius (ultimately Pope Pius II., 1458-1464), and Cosmo, the father of the Medici, are most popularly and familiarly known. It was at a moment when the labours of such men as these had accumulated in public and private libraries all that could be recovered of classical learning, that about the middle of the fifteenth century the art of printing was introduced into Italy. Under the auspices of the Benedictines of Subiaco, the Germans Sweynheim and Pannartz set up their press in the celebrated Monastery of Santa Scholastica, from which issued, in the year 1465, their edition of Lactantius. Removing to Rome in 1467, the first-fruits of their labour was ‘^Cicero de Oratore.” Thus, while in Germany and France biblical and ecclesiastical literature, and in England popular, first gave employment to the printer; in Italy, classical, for a time, almost exclusively F F 109 EENAISSA2^CE OENAMENT. engaged his attention. Nicholas Jenson, the Frenchman, who was sent by Louis XL to the ateliers of Fust and Scheffer, to learn “ le nouvel art par lequel on faisait des livres,” carried his acquired knowledge from Mayence to Venice, where he invented the Italic character, subsequently adopted by the learned Aldus Manutius. This remarkable man, who was a no less learned editor than he was zealous printer, from about the year 1490, gave to the world in rapid succession editions of the G-reek and Latin Classics. Among his earliest works is one ever memorable in the history of Art, the Hypnerotomachia,” or dream of Poliphilus, written by the learned ecclesiastic Fra Colonna. It is profusely illustrated with engravings on wood, the design of which has been frequently ascribed to no less great an artist than Andrea Mantegna. Through those illustrations, which display a profound study of ancient ornament, types of form diametrically opposed to those of the middle ages were disseminated over the Continent of Europe. The publication of Vitruvius at Rome, about 1486, at Florence in 1496, and at Venice, with illustrations, in 1511, as well as of Alberti’s great work, ‘'^De Re vEdificatoria,” at Florence, in 1485, set the seal upon the classical tendency of the age in matters of Art, and afforded the means of speedily transmitting to other countries the details of ancient design, so warmly taken up throughout Italy. The successors of the first Aldus at Venice, the G-ioliti in the same city, and the Giunti at Florence, rapidly multiplied the standard classics; and thus the art of printing speedily caused a movement of revival to become cosmopolitan, which, had that noble art remained undiscovered, would very probably have been limited, to a great extent, to the soil of Italy. Long, however, as we have already asserted, before the aspirations of the first labourers in the mine of antiquity had been thus brought to fruition, indications had been given in the world of Art of an almost inherent antagonism on the part of the Italians to Gothic forms. In the ornaments which surround the ceilings of the Church of Assisi, ascribed to Cimabue, the father of painting, the acanthus had been drawn with considerable accuracy; while Nicola Pisano and other masters of the trecento, or thirteenth century, had derived many important elements of design from a study of antique remains. It was scarcely, however, until the beginning of the fifteenth century that the movement can be said to have borne really valuable fruit. In its earliest stage the Renaissance of Art in Italy was unquestionably a revival of principles, and it was scarcely until the middle of the fifteenth century that it came to be in anywise a literal revival. Conscious as we may be, that in some productions of this earlier stage, when Nature was recurred to for suggestion, and the actual details of classic forms were comparatively unknown and unimitated, there may exist occasional deficiencies, supplied at a later period, and under a more regular system of education; we are yet free to confess a preference for the freshness and naiveU with which the pioneers worked, over the more complete but more easily obtained graces of an almost direct reproduction of the antique. The first great step in advance was taken by the celebrated Jacopo della Quercia, who having been driven from his birth-place, Sienna, to Lucca, executed about the year 1413, in the Cathedral of that city, a monument to Ilaria di Caretto, wife of Giunigi di Caretto, Lord of the City. In this interesting work (of which a good cast may be seen in the Crystal Palace) Jacopo exhibited a careful recourse to nature, both in the surrounding festoons of the upper part of the pedestal and the “puttini,” or chubby boys supporting them; the simplicity of his imitation being revealed by the little bandy legs of one of the “ puttini.” His great work, however, was the fountain in the Piazza del Mercato Siena, which was completed at an expense of two thousand two hundred gold ducats, and even in its present sad state of decay offers unmistakable evidence of his rare ability. After his execution of this capo d’opera, he was known as Jacopo della Fonte; this work brought him much distinction, and he was made Warden of the Cathedral in that city, where, after a life of much labour and many vicissitudes, he died in the year 1424, aged sixty-four. Although one of the unsuccessful 110 RENAISSANCE ORNAMENT. candidates for the second bronze door of the Florence Baptistery, as we shall presently see, he was much esteemed during his life, and exercised a great and salutary influence on sculpture after his death. G-reat, however, as were his merits, he was far surpassed in the correct imitation of nature, and in grace, dexterity, and facility in ornamental combination, by Lorenzo G-hiberti, who was one of his immediate contemporaries. In the year 1401, Florence, under an essentially democratic form of government, had risen to be one of the most flourishing cities of Europe. In this civic democracy the trades were distinguished as guilds, called ‘^Arti,” represented by deputies (consoli). The Consuls resolved in the above- mentioned year to raise another gate of bronze to the Baptistery, as a pendant to that of Andrea Pisano, which had been previously executed in a very noble, but still G-othic style. The SignoHa, or executive government, made known this resolve to the best artists of Italy, and a public competition was opened. Lorenzo Ghiberti, a native of Florence, at that time very young (twenty-two), ventured on the trial, and with two others, Brunelleschi and Donatello, was pronounced worthy. These two last-named artists appear to have voluntarily retired in his favour ; and in twenty- three years from that date the gate was finished, and put up. The. beauty of its design and work- manship induced the Signoria to order another of him, which was ultimately finished about the year 1444. It would be impossible to overrate the importance of this work, either as regards its historical influence on art or its intrinsic merit, — standing, as it does, unrivalled by any similar specimen in any age for excellence of design and workmanship. The ornament (for a portion of which see Plate LXXV., Fig. 3), which encloses and surrounds the panels, is worthy of the most careful study. Lorenzo Ghiberti belonged to no school, neither can it be said he founded one, he received his education from his father-in-law, a goldsmith ; and his influence on Art is to be seen rather in the homage and study his works received from men such as Buonarotti and Eaffaelle, than from his formation of any school of pupils. He died in his native city at a good old age, in the year 1455. One of his immediate followers, Donatello, imparted a life and masculine vigour to the art, which, in spite of all their beauty, were often wanting in the compositions of Ghiberti ; and the qualities of both these artists were happily united in the person of Luca della Kobbia, who, during his long life (which extended from 1400 to 1480), executed an infinity of works, the ornamental details of which were carried out in a style of the freest and most graceful analogy with the antique. In the person of Filippo Brunelleschi the talents of the sculptor and the architect were combined. The former are sufficiently evinced by the excellence of the trial-piece in which he competed with Ghiberti for the execution of the celebrated gates of San Giovanni Battista ; and the latter, by his magnificent Cathedral of Sta. Maria delle Fiore at Florence. This combination of architectural and sculptur- esque ability was, indeed, a distinguishing feature of the period. Figures, foliage, and conventional ornaments, were so happily blended with mouldings and other structural forms, as to convey the idea that the whole sprang to life in one perfect form in the mind of the artist by whom the work was executed. A development of taste coincident with that noticeable in Tuscany took place at Naples, Rome, Milan, and Venice. At Naples, the torch that was lit by Massuccio was handed on by Andrea Ciccione, Bamboccio, Monaco, and Amillo Fiore. At Rome, the opulence of the princes, and the great works undertaken by the successive pontiffs, attracted to the Imperial city the highest procurable ability; and hence it is, that in the various palaces and churches fragments of exquisite decorative sculpture are still to be met with. Bramante, Baldassare Peruzzi, and Baccio Pintelli (of whose arabesques on the exterior of the Church of Sant’ Agostino, one of the earliest buildings of the pure revival executed in the Imperial, our woodcuts give some elegant examples), and even the great Raffaelle himself, did not disdain to design ornaments 111 Arabesques designed by Baccio Pintelli, for the Church of Sant' Agostino, Rome. 112 RENAISSANCE ORNAMENT. for carvers, of the purest taste and most exquisite fancy. Of the perfection attained in this depart- ment of art by the last-named artist, the celebrated wooden stalls of the choir of San Pietro dei Casinensi, at Perugia, will long remain unquestionable evidence. The carrying out of these carvings by Stefano da Bergamo does full justice to the admirable composi- tions of Raffaelle. At Milan, the important works of the Duomo, and the Certosa at Pavia, created a truly remark- able school of art; among the most celebrated mas- Panel from the Piscina of the High Altar of the Certosa, Pavia. ters of which may be noticed, Fusina, Solari, Agrati, Amadeo, and Sacchi. The sculptor’s talent had long been traditional in that locality, and there can be no doubt that these artists embodied in the highest forms the lingering traditions of the Maestri Goma- schi, or Freemasons, of Como ; from whose genius many of the most celebrated buildings of the middle ages derived their highest graces of adornment. Of Panels from the Piscina of the High Altar of the Certosa, Pavia. Arabesques designed by Baccio Pintelli, for the Church of Sant’ Agostino, Rome HENAISSANCE ORNAMENT. all the Lombard Cinque-centists, however, the highest admiration m\ist he reserved for Agostino Buati, better known as Bambaja, and his pupil Brambilla, whose exquisite works in arabesque at the Certosa must ever remain marvels of execution. Our woodcuts, selected from the Piscina of the High Altar, furnish some idea of the general style of the Pavian arabesques. At Venice, the first great names which call for notice are those of the Lombardi (Pietro, Tullio, Giulio, Sante, and Antonio), through whose talents that city was adorned with its most famous monu- ments. They were followed by Eiccio, Bernardo, and Domenico di Mantua, and many other sculptors ; but their lesser glories are altogether eclipsed by those of the great Jacopo Sansovino. At Lucca, Matteo Civitale (born 1435, died 1501) fully maintained the reputation of the period. Eeturning to Tuscany, we find, towards the close of the fifteenth century, the greatest perfection of ornamental sculp- ture, the leading characteristic of which, however, we now no longer find to be the sedulous and simple imitation of nature, but rather a conventional rendering of the antique. The names of Mino da Fiesole the greatest of the celebrated school of the Fiesolani — Benedetto da Majano, and Bernardo Eossellini, bring to our recollection many exquisite monuments which abound in the churches of Florence, and the other principal towns of the Grand Duchy. These artists excelled alike in wood, in stone, and in marble, and their works have been surpassed in this style of art only by those of their predecessors we have already named, and by some few others, their contemporaries. Of these, Andrea Contucci, better known as the elder Sansovino, was pre-eminent in his art; and it would appear impossible to carry ornamental modelling to greater perfection than he has exhibited in the wonderful monuments which form the pride of the Church of Sta. Maria del Popolo, at Eome. His pupil, Jacopo Tatti, who subsequently took his master’s name, may be regarded as his only rival. Of him, however, more hereafter. Ornameate the Pis- Portions of Pilasters from the Church of Sta. Maria doi Mii-acoli, Venice, cina of the High Altar of the Certosa, Pavia. G G 113 RENAISSANCE ORNAMENT. Having thus succinctly traced the historical succession of the great sculptors of Italy, all of whom, it must constantly be borne in mind, were ornamentists also, we proceed io point out some few of those lessons which may, as we conceive, be derived from a study of their works by the artist and art- workman. One of the most peculiar and most fascinating qualities of the best Cinque-cento ornament in relief is the skill with which those by whom it was wrought availed themselves of the play of light and shade produced by infinite variations of plane, not only in surfaces parallel to the grounds from which the ornament was raised, but brought to a tangent with it at ever-varying angles of impact. The difference in effect between a scroll of the volute form, in which the relief gradually diminishes from the starting of the volute to its eye, and one in which the relief is uniform throughout, is very great; and it is to their undeviating preference for the former over the latter, that the Cinque-cento artists are indebted for the infallibly pleasing results they attained alike, in their simplest and most complicated combinations of spiral forms. This refined appreciation of delicate shades of relief in sculpture was carried to its greatest per- fection by Donatello, whose authority in matters of taste was held in the highest possible esteem by the contemporary Florentines, and whose example was followed with respect and devotion by all classes of artists. Not only was he the first to practise the hassissimo relievo^ in which the effect of pro- jection and of rounded modelling is obtained within apparently impracticable limits of relief, but he was the first to combine that style of work with mezzo and alto relievo ; thus maintaining an almost pictorial division of his subject into several planes. Too good a master of his craft to ever overstep the special conventions of sculpture, Donatello enriched the Florentine practice of the Cinque-centisti with many elements derived from the sister art of Painting. These inven- tions — for they are almost worthy of the name, though arrived at only through a sedulous study of the Antique — were adopted and imi- tated with the greatest avidity by the ornamentists of the period ; and hence we may trace some of the most peculiar and striking technical excellence of the best Kenaissance carving and modelling. Ultimately, and at its acme of perfection, this system of regular arrangement of ornament in planes was so ingeniously managed in rela- tion to light and shade, that, viewed from a distance, the relievo pre- sented only certain points symmetrically disposed with reference to some dominant geometrical figures. An approach of a few paces served to bring to the sense of vision the lines and figures connecting the points of greatest salience. A yet nearer approach revealed the leafage and delicate tendrils necessary to convey a tangible idea of the type of nature selected for convention, while no inspection could be too close to test the artist’s perfect appreciation of the refinements of sur- face texture. The “ cisellatura,” or ‘‘ chasing,” of the best Italian Cinque- cento ornament, such as may be seen in the Church of the Miracoli, Venice (Figs. 1, 8, 9, Plate LXXIV.), by the Lombardi ; in the Church of Sta. Maria del Popolo (Fig. 1, Plate LXXVI.), Rome, by Sansovino; in the gates of the Baptistery, Florence (Fig. 3, Plate LXXV.), by G-hiberti ; in the carvings of San Michele di Murano (Figs. 4, 6, Plate LXXIV.) ; the Scuola di San Marco (Fig. 2, Plate LXXIV.) ; Small PilasteraofMar- del Miracoli, Venice, ble Staircase in the Tullio Lombardo, a.d. Church of Sta. Maria 1485, about. 114 RENAISSANCE ORNAMENT. the Scala dei Giganti (Figs. 5, 7, Plate LXXIV.) ; and other buildings at Venice, is beyond al) praise. The fibres of a leaf or tendril are never misdirected, nor is Nature’s tendency to grace in growth per- verted or misapprehended. Smoothness and detail are never added excepting where they have some specific function to perform ; and while labour is so prodigally bestowed as to show that every additional touch was a labour of love, it is never thrown away, as is too often the case in the present day, in converting those portions of a design which should l)e secondaries or tertiaries in point of interest into primaries. In the hands of artists less profoundly impressed than was Donatello with a sense of the just limit of convention in sculpture, the importation of pictorial elements into bas-relief soon degenerated into confusion. Even the great Ghiberti marred the effect of many of his most graceful compositions by the introduction of perspective, and accessories copied too directly from nature. In many of the orna- mental sculptures of the Certosa the fault is exaggerated until monuments, which should impress the spectator with grave admiration at their beauty and dignity, serve only to amuse him — resembling dolls’ houses peopled by fairies, decked with garlands, hung with tablets, and fancifully overgrown with foliage, rather than serious works of Art commemorating the dead, or dedicated to sacred uses. Another reproach which may with justice be addressed to many such monuments is the incongruity of the association of ideas con- nected with their purport, and those suggested by the ornaments displayed in their friezes, pilasters, panels, spandrils, and other en- riched features. Tragic and comic masques, musical instruments, semi-Priapic terminals, antique altars, tripods, and vessels of libation, dancing amorini, and hybrid marine monsters, and chimeras, har- monise but ill with monuments reared in consecrated edifices or dedi- cated to religious rites. This fault, of the confusion of things sacred and profane, may not, however, be altogether justly laid upon the shoulders of the artists of the Renaissance, whose works served but to reflect the dominant spirit of an age in which the revival of mytho- logic symbolism was but a protest against the hampering trammels of ascetic tradition erected into dogmatism under the rulers of the East, and endorsed by the Church during those centuries when its as- cendancy over an ignorant and turbulent population was at its greatest height. The minds of even the most religious men were imbued with such incongruous associations in the fourteenth century ; and it is not necessary to go farther than the “ Commedia” of Dante, which all the world of literature has designated as the Divine Epic, to re- cognise the tangled skeins of Gothic and classical inspiration with which the whole texture of contemporary literature was interwoven. To the architect, the study of Italian Cinque-cento ornament in relief is of no less utility than it can possibly be to the sculptor, since in no style has ornament ever been better spaced out, or Small Pilaster of arranged to contrast more agreeably with the direction of the ad- Jviarble Staircase Small Pilaster of the Giant's ... the Church of laceut architectural lines by which it is bounded and kept in subor- staircase. Ducai Palace, Sta. Maria, dei ^ Venice, _hy Bcndetto and Miracoh, Venice. d.ination. Rarely, if ever, is an ornament suitable for a horizontal Domemco da Mantua, position placed in a vertical one, or vice versa; and rarely, if ever, are the proportions of the orna- ments and the mouldings, or the styles and rails, by which regularity and symmetry are given to the whole, at variance with one another. In Plates LXXIV., LXXV., and LXXVL, are collected a series 115 EENAISSANCE ORNAMENT. of specimens, in the majority of which gracefulness of line, and a highly artificial, though apparently natural, distribution of the ornament upon its field, are the prevailing characteristics. The Lombardi, in their works at the Church of Sta. Maria dei Miracoli, Venice (Plate LXXIV., Figs. 1, 8, 9 ; Plate LXXVI., Fig. 2); Andrea Sansovino at Eome (Plate LXXVI., Fig. 1); and Domenico and Bernardino di Mantua, at Venice (Plate LXXIV., Figs. 5 and 7), attained the highest perfection in these respects. At a subsequent period to that in which they flourished, the ornaments were generally wrought in more uniformly high relief, and the stems and tendrils were thickened, and not so uniformly tapered, the accidental growth and play of nature were less sedulously imitated, the field of the panel was more fully covered with enrichments, and its whole aspect made more bustling and less refined. The sculptor’s work as- serted itself in competition with the architect’s ; the latter in self-defence, and to keep the sculpture down, soon be- gan to make his mouldings heavy : and a more ponderous style altogether crept into fashion. Of this tendency to 'plethora in ornament we already perceive indications in much of the G-enoese work represented in Plate LXXV,, Figs. 1, 2, 4, 5, 8, 9, and 11 ; and in Plate LXXVI., Figs. 4, 5, 7, 8, and 10. Fig. 6 in the last-mentioned plate, from the celebrated Martinengo Tomb, at Brescia, also clearly exhibits this tendency to filling up. In the art of painting, a move- ment took place concurrent with that we have thus briefly noticed in sculp- ture. Giotto, the pupil of Cimabue, threw off the shackles of Greek tra- dition, and gave his whole heart to nature. His ornament, like that of his master, consisted of a combination painted mosaic work, interlacing Maria dei Miracoli, Venice. , , . p , . „ bends, and tree rendering of the acan- thus. In his work at Assisi, Naples, Florence, and Padua, he has invariably shown a graceful appre- hension of the balance essential to be maintained between mural pictures and mural ornaments, both in quantity, distribution, and relative colour. These right principles of balance were very generally understood and adopted during the fourteenth century; and Simone Memmi, Taddeo Bartolo, the Orcagnas, Pietro di Lorenzo, Spinello Aretino, and many others, were admitted masters of mural em- bellishmeut. That rare student of nature in the succeeding century, Benozzo Gozzoli, was a no less diligent student of antiquity, as may be recognised in the architectural backgrounds to his pictures in the Campo Santo, and in the noble arabesques which divide his pictures at San Gimignano. Andrea Mantegna, however, it was who moved painting as Donatello had moved sculpture, and that not in figures alone, hut in every variety of ornament borrowed from the antique. The magnificent cartoons we are so fortunate as to possess of his at Hampton Court, even to their minutest decorative details might have been drawn by an ancient Roman. Towards the close of the fifteenth century, the style ' 116 Portion of a Doorway in ono of the Palaces of the Dorias near the Church of San Mattoo, Genoa. RENAISSANCE ORNAMENT. of polychromy took a fresk and marked turn, the peculiarities of which, in connexion with arabesque and grotesque ornament, we reserve for a subsequent notice. Turning from Italy to France, which was the first of the European nations to light its torch at the fire of Renaissance Art, which had been kindled in Italy, we find that the warlike expeditions of Charles VIIT, and Louis XII. infected the nobility of France with an admiration for the splendours of Art met with by them at Florence, Rome, and Milan. The first clear indication of the coming change might have been seen (for it was unfortunately destroyed in 1793) in the monument erected in 1499 to the memory of the first-named monarch, around which female figures, in gilt bronze, of the Virtues, were grouped completely in the Italian manner. In the same year, the latter sovereign invited the celebrated Fra Griocondo, architect, of Verona, friend and fellow - student of the elder Aldus, and first good editor of Vitruvius, to visit France. He remained there from 1499 to 1506, and designed for his royal master two bridges over the Seine, and probably many minor works which have now perished. The magnificent Chateau de Graillon, begun by Cardinal d’Amboise in the year 1502, has been frequently ascribed to him, but, according to Emeric David and other French archgeo- logists, upon insufficient grounds. The internal evidence is entirely in favour of a French origin, and against Griocondo, who was more of an engineer and student than an ornamental artist. Moreover, intermingled with much that is very fairly classical, is so much Burgundian work, that it would be almost as unjust to Giocondo to ascribe it to him, as to France to deprive her of the credit of having produced, by a French artist, her first great Renaissance monument. The whole of the accounts which were published by M. Deville in 1850, set the question almost entirely at rest; for from them we learn that Guillaume Senault was architect and master mason. It is, however, just possible that Giocondo may have been consulted by the Cardinal upon the general plan, and that Senault and his companions, for the most part French, may have carried out the details. The principal Italian by whom, if we may judge from the style some of the most classical of the arabesques were wrought, was Bertrand de Meynal, who had been commissioned to carry from Genoa the beautiful Venetian fountain, so well known as the Vasque du Chateau de Gaillon, now in the Louvre, and from which (Plate LXXXI., Figs. 27, 30, 34, 38) we have engraved some elegant ornaments. Colin Castille, who especially figures in the list of art-workmen as “ tailleur a 1’ antique,” may very possibly have been a Spaniard who had studied in Rome. In all essential particulars, the portions of Renaissance work not Burgundian in style are very pure, and differ scarcely at all from good Italian examples. It was, however, in the monument of Louis XIL, now at St. Denis, near Paris, and one of the richest of the sixteenth century, that symmetry of architectural disposition was for the first time united to masterly execution of detail in France. This beautiful work of Art was executed between 1518 and 1530, under the orders of Francis I., by Jean Juste of Tours. Twelve semicirculaj arches inclose the bodies of the royal pair, represented naked ; under every arch is placed an apostle ; and at the four corners are four large statues of Justice, Strength, Prudence, and Wisdom ; the whole being surmounted by statues of the King and Queen on their knees. The bas-reliefs represent the triumphal entry of Louis into Genoa, and the battle of Aguadel, where he signalised himself by his personal valour. The monument of Louis XII. has been often ascribed to Trebatti (Paul Ponee), but it was finished before he came to France, as the following extract from the royal records proves. Francis I. addresses the Cardinal Duprat : — “ H est deu a Jehan Juste mon sculteur ordinaire, porteur de ceste la somme de 400 escus, restans des 1200 que je lui avoie pardevant or donnez pour la menage et conduite de la ville de Tours au lieu de St. Denis en France, de la sculpture de marbre de feuz Roy Loys et Royne Anne, &c. Novembre 1531.” Not less worthy of study than the tomb of Louis XIL, and executed at the same period, are the H H 117 Portions of the Tomb of Francis II., Duke of Brittany, and his wife, Mai-guerite de Foix, erected by Anne of Brittany in the Carmelite Church, at Nantes, by Michel Colombe, a.d. 1607. and beauty, the figures animated and natural, the drapery free and graceful, and the heads full of life; but the arabesque ornaments, which almost entirely cover the projecting parts of the pilasters, friezes, and mouldings of the base, are, perhaps, the most beautiful portions ; they are very diminutive 118 RENAISSANCE ORNAMENT. in size ; the largest of the groups, which are those which cover the pilasters, being only eight or nine inches in breadth. Though so minute, the spirit of the carving, and variety of devices in these orna- ments, are marvellous. Masses of foliage, branches of trees, birds, fountains, bundles of arms, satyrs, military ensigns, and tools belonging to various arts, are arranged with much taste. The F. crowned the monogram of Francis I. — is conspicuous in these arabesques, and the dates of the years 1525, 1527, and 1529, are traced upon the draperies. The tomb which Anne of Brittany caused to be erected to the memory of her father and mother was. finished and placed in the choir of the Carmelite Church at Nantes on the 1st of January, 1507. It is the master-piece of an artist of great ability and riaivete — Michel Colombe. The ornamental details are peculiarly elegant. The monument to Cardinal d’Amboise, in the Cathedral at Kouen, was begun in the year 1515, under Eoulant le Boux, master mason of the Cathedral. No Italian appears to have assisted in its execution, and we may, therefore, fairly regard it as an expression of the vigour with which the Benaissance virus had indoctrinated the native artists. It was in 1530 and 1531 that F’rancis I. invited Bosso and Primaticcio into France, and those distinguished artists were speedily followed by Nicolo del’ Abbate, Luca, Penni, Cellini, Trebatti, and Girolamo della Bobbia. With their advent, and the foundation of the school at Fontainebleau, new elements were introduced into the French Benaissance, to which we shall subsequently advert. It would exceed the limits of our present sketch to enter fully into the historical details con- nected with the art of wood-carving. It may suffice to point out that every ornamental feature available for stone, marble, or bronze, was rapidly transferred also to wood -work, and that at no period of the history of Industrial Art has the talent of the sculptor been more gracefully brought to bear upon the enrichment of sumptuous furniture. Our Plates, Nos. LXXXI. and LXXXII., furnish brilliant evidence of the justice of our remarks on this head. The attentive student, how- ever, as he goes over them, will be unable to avoid perceiving a gradual withdrawing from the original foliated ornament which formed the stock in trade of the early Benaissance artists. He will next notice a heaping up of various objects and “ capricci,” derived from the antique, accompanied by a fulness of projection and slight tendency to heaviness; and then, finally, he will recognise the general adoption of a particular set of forms differing from the Italian, and altogether national, such as the conventional volute incised with small square or oblong indentations (Plate LXXXI. Figs. 17 and 20), and the medallion heads (Plate LXXXI. Figs. 1 and 17). The dawning rays of the coming revival of Art in France can scarcely be traced in the painted glass of the fifteenth century. The ornaments, canopies, foliage, and inscriptions, are generally flamboyant and angular in character, although freely and crisply made out, and the figures are influenced by the prevailing style of drawing. The glass, although producing a pleasing effect, is much thinner — especially the blue — than that of the thirteenth century. An immense number of windows were executed during this epoch, and specimens are to be found more or less perfect in almost every large church in France. St. Ouen, at Bouen, has some fine figures upon a white quarry ground in the clerstory windows ; and good examples of the glass of the century will be found in St. Gervais at Paris, and Notre Dame at Chalons-sur-Mame. Many improvements were introduced into the art at the epoch of the Benaissance. The first masters were employed to make cartoons ; enamel was used to give depth to the colours without losing the richness, and much more white was employed. Many of the windows are very little more than grisailles, as those designed by Jean Cousin for the Sainte Chapelle at Vincennes ; one of those representing the angel sounding the fourth trumpet is admirable both in composition and drawing. The Cathedral of Auch also contains some exceedingly fine examples of the work of Arneaud Demole ; Beauvais also possesses a great deal of the glass of this period, especially a very fine Jesse window, 119 RENAISSANCE ORNAMENT. the work of Enguerand le Prince ; the heads are grand, and the foses of the figures call to mind the works of Albert Purer. The grisailles, which ornamented the windows in the houses of the nobility, and even of the bourgeoisie, although small, were executed with an admirable delicacy, and in drawing and grouping leave little to be desired. Toward the end of the sixteenth century the art began to decline, the numerous glass-painters found themselves without employment, and the celebrated Bernard de Palissy, who had been brought up to the trade, left it to engage in another presenting greater difficulties, but which eventually secured him the highest reputation. To him, however, we are indebted for the charming grisailles epresenting the story of Cupid and Psyche, from the designs of Raffaelle, which formerly decorated the Chateau of Ecouen, the residence of his great patron the Constable Montmorency. Renaissance ornament penetrated into Germany at an early period, but was absorbed into the hearts of the people but slowly, until the spread of books and engravings quickened its general acceptation. From an early period there had been a steady current of artists leaving Germany and Flanders to study in the great Italian ateliers. Among them, men like Roger of Bruges, who spent much of his life in Italy, and died in 1464, — Hemskerk, and Albert Purer, more especially influenced their countrymen. The latter, who in many of his engravings showed a perfect apprehension of the conditions of Italian design, leaning now to the Gothic manner of his master Wohlgemuth, and now to the Eaflaellesque simplicity of Marc’ Antonio. The spread of the engravings of the latter, however, in Germany, unquestionably conduced to the formation of the taste of men who like Peter Vischer first brought Italian plastic art into fashion in Germany. Even at its best the Renaissance of Germany is impure — her industrious affection for difficulties of the hand, rather than of the head, soon led her into crinkum-crankums ; and strap- work, jewelled forms, and complicated monsters, rather animated than graceful, took the place of the refined elegance of the early Italian and French arabesques. Araboaqyc by Theodor de Bry, one of the “Pe:ita Maltros” of Germany (160SX in imitation of Italian work, but introducing strap-work, caricature, and iewelled forms. It may be well now to turn from the Fine to the Industrial Arts, and to trace the manifestation of the revival in the designs of contemporary manufactures. From the unchanging and unchangeable nature of vitreous and ceramic products, no historical evidence of style can be more complete and 120 RENAISSANCE ORNAMENT. satisfactory than that which they afford, and hence we have devoted three entire Plates (Nos. LXXVIII., LXXIX., and LXXX.) to their illustration. The majority of the specimens thereon represented have been selected from the ‘‘Majolica” of Italy, on which interesting ware and its ornamentation we proceed to offer a few remarks. The art of glazing pottery appears to have been introduced into Spain and the Balearic Isles by the Moors, by whom it had long been known and used in the form of coloured tiles for the decoration of their buildings. The earthenware called “majolica” is believed to derive its name from the Island of Majorca, whence the manufacture of glazed pottery is supposed to have found its way into Central Italy; and this belief is strengthened by the fact of the earliest Italian ware being ornamented with geometrical patterns and trefoil-shaped “foliations” of Saracenic character (Plates LXXIX. and LXXX., Figs. 31 and 13). It was first used by introducing coloured concave tiles among brickwork, and later in the form of encaustic flooring. The manufacture of this ware was extensively carried on between 1450 and 1700, in the towns of Nocera, Arezzo, Citta de Castillo, Forli, Faenza (whence comes fayence), Florence, Spello, Perugia, Deruta, Bologna, Eimini, Ferrara, Pesaro, Fermignano, Castel Durante, Gubbio, Urbino, and Kavenna, and also at many places in the Abruzzi ; but Pesaro is admitted to be the first town in which it attained any celebrity. It was at first called or “half” majolica, and was usually made in the form of thick clumsy plates, many of large size. They are of a dingy grey colour, and often have a dull yellow varnish at the back. The texture is coarse and gritty, but the golden and prismatic lustre is now and then seen, though they are more frequently of a pearly hue. This “half” majolica is believed by Passeri and others to have been made in the fifteenth century; and it was not until after that time that the manufacture of “fine” majolica almost entirely superseded it. A mode of glazing pottery was also discovered by Lucca della Robbia, who was born at Florence in 1399. It is said that he used for this purpose a mixture of antimony, tin, and other mineral substances, applied as a varnish to the surface of the beautiful terra-cotta statues and bas-reliefs modelled by him. The secret of this varnish remained in the inventor’s family till about 1550, when it was lost at the death of the last member of it. «Attempts have been made at Florence to revive the manufacture of the Robbian ware, but with small success, owing to the great difliculties attending it. The subjects of the bas-reliefs of Della Robbia are chiefly religious, to which the pure glistening white of the figures is well adapted; the eyes are blackened to heighten the expression, and the white figures well relieved by the deep blue ground. Wreaths of flowers and fruits in their natural tints were introduced by the followers of Della Robbia, by some of whom the costumes were coloured, whilst the flesh parts were allowed to remain unglazed. Passeri claims the discovery of this coloured glaze at a still earlier date for Pesaro, where the manufacture of earthenware was carried on in the fourteenth century ; but though the art of combining it with colour may have been known at that early time, it had not attained much celebrity until 1462, -when Matteo di Raniere of Gagli and Ventura di Maestro Simone dei Piccolomini of Siena established themselves at Pesaro, for the purpose of carrying on the manufacture of earthenware already existing there; and it is not improbable that their attention was attracted by the works of Della Robbia, who had been employed by Sigismond Pandolfo Malatesta at Rimini. Some confusion appears to have arisen with respect to the precise process invented by Della Robbia, and looked upon by himself and his family as the really valuable secret. We feel little doubt that it consisted rather in the tempering and firing of the clay to enable it to burn large masses truly and thoroughly than in the protecting glaze, about which there appears to have been very little novelty or necessity for concealment. Prismatic lustre and a brilliant and transparent white glaze were the qualities chiefly sought for in the “fine’ majolica and Gubbian ware; the metallic lustre was given by preparations of lead, 1 1 121 RENAISSANCE ORNAMENT. silver, copper, and gold, and in this the Guhbian ware surpassed all others. The dazzling white glaze was obtained by a varnish made from tin, into which, when half-baked, the pottery was plunged ; the designs were painted before this was dry, and, as it immediately absorbed the colours, it is not to be wondered at that we so frequently find inaccuracies in the drawings. A plate of the early Pesaro ware in the Museum at the Hague bears a cipher, the letters of which appear to be C. H. 0. N.” Another, mentioned by Pungileoni, has “ G. A. T.” interlaced, forming a mark. These instances are rare, as the artists of these plates seldom signed their works. The subjects generally chosen were saints and historical events from Scripture; but the former were preferred, and continued in favour till the sixteenth century, when they were displaced by scenes from Ovid and Virgil, though designs from Scripture were still in use. The subject was generally briefly described with a reference to the text in blue letters at the back of the plate. The fashion of ornamenting the ware with the portraits of historical, classical, and living persons, with the names attached to each, was of rather later date than the sacred themes. All these subjects are painted in a flat, tame manner, with little attempt at shading, and are surrounded by a kind of rude Saracenic ornament, differing completely from the Rafifaellesque arabesques, which, in the latter years of Guidobaldo’s reign, were so much in fashion. The plates full of coloured fruits in relief were probably taken from the Robbian ware. The decline of this manufacture caused by the Duke’s impaired income and the want of interest in the manufacture felt by his successor, was hastened by the introduction of Oriental china and the increased use of plate in the higher and more wealthy classes; still, though historical subjects were laid aside, the majolica was ornamented with well-executed designs of birds, trophies, flowers, musical instruments, sea monsters, &c., but these became gradually more and more feeble in colouring and execution till, at last, their place was taken by engravings from Sadeler and other Flemings. From all these causes the manufacture fell rapidly to decay in spite of the endeavours made to revive it by Cardinal Legate Stoppani. The “fine” majolica of Pesaro attained its greatest perfection during the reign of Guidobaldo II., who held his court in that city, and g»eatly patronised its potteries. From that time, the majolica of Pesaro so closely resembled that of Urbino, that it is not possible to distinguish the manufacture of the two places from each other, the texture of the ware being alike, and the same artists being often employed in both potteries. As early as 1486 the Pesaro ware was considered so superior to all other Italian ware, that a protection was granted to it by the lord of Pesaro of that date, not only forbidding, under penalty of fine and confiscation, the importation of any kind of foreign pottery, but ordering that all foreign vases should be sent out of the state within eight days. This protection was confirmed, in 1532, by Francesco Maria I. In 1569, a patent for twenty-five years, with a penalty of 500 scudi for infringing it, was granted by Guidobaldo II, to Giacomo Lanfranco of Pesaro, for his inventions in the construction of vases wrought in relief, of great size and antique forms, and his application of gold to them. In addition to this, his father and himself were freed from all taxes and imposts. From its variety and novelty, majolica was generally chosen by the lords of the Duchy for their presents to foreign princes. In 1478, Costanza Sforza sent to Sixtus IV. certain “vasa fictilia;” and in a letter from Lorenzo the Magnificent to Robert Malatista, he returns thanks for a present of a similar kind. A service painted by Orazio Fontana from designs by Taddeo Zuccaro, was presented by Guidobaldo to Philip II. of Spain. A double service was also given by him to Charles V. The set of jars presented to the Treasury of Loreto by Francesco Maria II., were made by the order of Guidobaldo II., for the use of his own laboratory ; some of them are ornamented with a portrait, or subject of some other description, and all are labelled with the name of a drug or mixture. The 122 RENAISSANCE ORNAMENT. colours of these jars are blue, green, and yeUow; about 380 of them still remain in the Treasury of Loreto. Passeri gives an interesting classification of ornamental pottery, with the terms made use of by the workmen to distinguish the various kinds of paintings used in ornamenting the plates, and also the sums paid to the artists by whom they were painted. He gives a curious extract from a manuscript in the handwriting of Piccolpasso, a “majolicaro” of the middle of the sixteenth century, who wrote upon his art; to understand which it is necessary to remember that the holognino was equivalent to the ninth part, and the gros to the third part, of a paul (5J pence); the livre was a third, and the florin two-thirds of a petit ecu ; and the petit ecu, or ecu ducal, two-thirds of a Roman crown (now value four shi llin gs and threepence one farthing). Trophies. — This style of ornament consisted of ancient and modern arms, musical and ma- thematical instruments, and open books ; they are generally painted in yellow cameo on a blue ground. These plates were chiefly sold in the province (Castel Durante) in which they were manufactured, one ducal crown a hundred being the sum paid to the painters of them. This style was much affected by the Cinque-centisti in marble and stone : witness the monument to Gian Galeazzo Visconti, in the Certosa, Pavia, and portions of the Genoese doorway, we en- grave. Arabesques were ornaments consisting of a sort of cipher, loosely tied and interlacing knots and bouquets. Work thus ornamented was sent to Venice and Genoa, and obtained one ducal florin the hundred. Cerquate was a name given to the interlacing of oak-branches, painted in a deep yellow upon a blue ground ; it was called the “ Urbino painting,” from the oak being one of the bearings of the ducal arms. This kind of decoration received fifteen gros the hundred; and when, in addition, the bottom of the plate was ornamented, by having some little story painted upon it, the artist received one petit ecu. Grotesques were the interlacing of winged male and female monsters, with their bodies terminated by foliations or branches. These fanciful decorations were generally painted in white cameo upon a Pedestal forming part of a Doorway of the Palace, presented by the Genoese to Andrea Doria. blue ground ; the payment for them being two ecus the hundred, unless they were painted on commission from Venice, when the price was eight ducal livres. Leaves. This ornament consisted of a few branches of leaves, small in size, and sprinkled over the ground. Their price was three livres. Flowers and Fruits — These very pleasing groups were sent to Venice, and the artists received for them five livres the hundred. Another variety of the same style merely consisted in three or four large leaves, painted in one colour upon a different-coloured ground. Their price -was half a florin the hundred. 123 RENAISSANCE ORNAMENT. Porcelain was the name of a style of work which consisted of the most delicate blue flowers, with small leaves and buds painted upon a white ground. This kind of work obtained two or more livres the hundred. It was, in all probability, an imitation of Portuguese importations. Tratti were wide bands, knotted in different ways, with small branches issuing from them. Their price was also two livres the hundred. So'prahianco was a painting in white upon a white-lead ground, with green or blue borders round the margin of the plate. These obtained a demi-ecu the hundred. Quai'tieri . — In this pattern, the artist divided the bottom oi the plate into six or eight rays diverging from the centre to the circumference ; each space was of a particular colour, upon which were painted bouquets of different tints. The painters received for this kind of ornament two livres the hundred. Gh'uppi . — These were broad bands interwoven with small flowers. This pattern was larger than the tratti,” and was sometimes embellished by a little picture in the centre of the plate, in that case the price was a demi-ecu, but without it only two jules. Portions of tbo Pilaster of a Doorway in the Palace at Genoa, presented by the Genoese to Andrea Doria. Candelabri . — This ornament was an upright bouquet extending from one side of the plate to the other, the space on each side being filled up with scattered leaves and flowers. The price of the Candelabni was two florins the hundred. The adjoining woodcut shows how common, how early, and how favourite a subject this was with the best artists of the Cinque-cento. To dwell in detail upon the merits and particular works of artists, such as Maestro G-iorgio Andreoli, Orazio Fontana, and Francesco Xanto of Eovigo, would be beyond the scope of this notice, and is the less necessary as Mr. Robinson, in his Catalogue of the Soulages Collection, has so recently thrown out some new and highly interesting speculations upon various difficult questions connected with the subject. Neither will it be desirable here to do more than to point out the interesting modifications of ceramic design and practice carried out in France through the indomitable perseverance of Bernard de Palissy, master-potter to Francis I. In Plate LXXIX. Figs. 1, 3, we have engraved several specimens of the decorations of his elegant ware, which occupy as to design, in reference to other monuments of the French Renaissance, much the same position that the design of the early majolica does to the monuments of the Italian revival. Although that style began to make its appearance in 124 KENAISSANCE ORNAMENT. the -works of the French jewellers in the reign of Louis XII., when the extensive patronage of the powerful Cardinal d’Amboise gave considerable impetus to the art, it was under Francis I. who invited to his Court the great master of the Eenaissance — Cellini — that the jeweller’s art reached its highest perfection. To rightly appreciate, however, the precise condition and nature of the precious metal-work, it is necessary to pass in rapid review the leading characteristics of the admirable school of enamellers, whose productions in the fifteenth century, and much more in the sixteenth, served to disseminate far and wide some of the most elegant ornaments which have ever been applied to metal- work. About the end of the fourteenth century, the artists of Li- moges found not only that the old champleve enamels, — of which, in Plate LXXVIL, Figs. 1, 3, 4, 8, 29, 40, 41, 50, 53, 57, 61, we have given, for the sake of contrast, numerous examples, — had entirely gone out of fashion, but that almost every gold- smith either imported the translucid enamels from Italy, or ex- ecuted them himself with more or less skill, according to his talents. In this state of things, instead of attempting competi- tion, they invented a new manufacture, the processes of which belonged solely to the enameller, and enabled him to dispense entirely with the burin of the goldsmith. The first attempts were exceedingly rude, and very few of them now remain; but that the art progressed slowly is evident from the fact, that it is not until the middle of the fifteenth centuiy that specimens are to be found in any quantity, or possessing any degree of merit. The process was this : — The design was traced with a sharp point upon an unpolished plate of copper, which was IlSil then covered with a thin coat of transparent enamel. The artist, after going over his tracing with a thick black line, filled in the intervals with the various colours, which were, for the most part, transparent, the black lines performing the office of the gold strips of the cloissonne work. The carnations pre- ’-p'^ J j \ Y sented the greatest difficulty, and were, first of all, covered over with the black colour, and the high lights and half-tints were then modelled upon that with opaque white, which occasionally received a few touches of light transparent red. The last opera- tion was to apply the gilding, and to affix the imitations of / precious stones,— almost the last trace of the Byzantine school, Lowerportioushowingthesprinringofeuroll-workofasmall , Pilaster, by the Lombardi, in the Church of Sta. which had formerly exercised so much influence in Aquitaine. iiiracou, Venice. The appearance of the finished works was very similar to that of a large and coarse translucid enamel, — a resemblance not unlikely to have been intentional, more especially as specimens of the latter were never made of any considerable size, and were therefore fit to supply the place of ivory in the construction of those small triptyohs which were so necessary an appendage to the chambers and oratories of the rich in the middle ages. Accordingly, we find nearly all the early painted enamels are either in the form of triptychs or diptychs, or have originally formed parts of them ; ' and a great KK 125 RENAISSANCE ORNAMENT. number preserve their original brass frames, and are supposed by antiquaries to have been produced in the atelier of Monvearni, as tbe name or initials of that master are generally found upon them. As to the other artists, they followed, unfortunately, the but too common practice of most of the workmen of the middle ages, and, with the exceptions of Monvearni and P. E. Nicholat, or, as the inscriptions have been more correctly read, Penicaud, their names are buried in oblivion. At the commencement of the sixteenth century, the Penaissance had made great progress ; and among other changes, a great taste for paintings in ^‘camaieu,” or ^‘grisaille,” had sprung up. The ateliers of Limoges at once adopted the new fashion, and what may be called the second series of painted enamels was the result. The process was very nearly the same as that employed with regard to the carnations of the earlier specimens, and consisted in, firstly, covering the whole plate of copper over with a black enamel, and then modelling the lights and half-tints with opaque white ; those parts requiring to be coloured, such as the faces and the foliage, receiving glazes of their appropriate tints touches of gold are almost always used to complete the picture ; and, occasionally, when more than ordinary brilliancy was wanted, a thin gold or silver leaf, called a “ pallion,” was applied upon the black ground, and the glaze afterwards superposed. All these processes are to be seen in the two pictures of Francis I. and Henry II., executed by Leonard Limousin, for the decoration of the Sainte Chapelle, but which have now been removed to the Museum of the Louvre. Limoges, indeed, owed no small debt of gratitude to the former monarch, who not only established a manufactoiy in the town, but made its director Leonard, peintre, emailleur, valet-de-chambre du Roi,” giving him, at the same time, the appellation of le Limousin,” to distinguish him from the other and still more famous Leonardo da Vinci. And, indeed, the Limousin was no mean artist, whether we regard his copies of the early German and Italian masters, or the original portraits of the more celebrated of his contemporaries, such as those of the Duke of Guise, the Constable Montmorency, Catherine de Medicis, and many others — executed, we must remember, in the most difficult material which has ever yet been employed for the purposes of art. The works of Leonardo extend from 1532 to 1574, and contemporaneously with him flourished a large school of artist-enamellers, many of whose works quite equalled, if they did not surpass, his own. Among them we may mention Pierre Raymond and the families of the Penicauds, and the Courteys, Jean and Susanna Court, and M. D. Pape. The eldest of the family of the Courteys, Pierre, was not only a good artist, but has the reputation of having made the largest-sized enamels which have ever been executed (nine of these are preserved in the Museum of the Hotel de Cluny — the other three, M. Labarte informs us, are in England) for decorating the fapade of the Chateau de Madrid, upon which building large sums were lavished by Francis I. and Henry 11, We should observe that this last phase of Limoges enamelling was not confined, like its predecessor, to sacred subjects ; but, on the contrary, the most distinguished artists did not disdain to design vases, caskets, basins, ewers, cups, salvers, and a variety of other articles of every-day life, which were afterwards entirely covered with the black enamel, and then decorated with medallions, &c., in the opaque white. At the commencement of the new manufacture, the subjects of most of the enamels were furnished from the prints of the German artists, such as Martin Schoen, Israel van Mecken, &c. These were afterwards supplanted by those of Marc’ Antonio Raimondi and other Italians, which, in their turn, gave way about the middle of the sixteenth century to the works of Virgilius Solis, Theodore de Bry, Etienne de I’Aulne, and others of the 'petits^aitres. The production of the painted enamels was carried on with great activity at Limoges, during the whole of the fifteenth, sixteenth, and seventeenth centuries, and far into the eighteenth, when it finally expired. The last artists were the families of the Nouaillers and Laudins, whose best works are remarkable for the absence of the paillons, and a somewhat undecided style of drawing. In conclusion, it remains for us only to invite the student to cultivate the beauties, as sedulously 126 EENAISSANCE ORNAMENT. as he should eschew the extravagancies, of the Eenaissanoe style. Where great liberty is afforded m Art no less than in Polity, great responsibility is incurred. In those styles in which the imagina- tion of the designer can be checked only from within, he is especially bound to set a rein upon his fancy. Ornament let him have in abundance ; but in its composition let him be modest and decorous, avoid- ing over-finery as he would nakedness. If he has no story to tell, let him be content with floriated forms and conventional elements in his enrichments, which please the eye without making any serious call upon the intellect; then, where he really wishes to arrest observation by the comparatively direct representation of material objects, he may be the more sure of attaining his purpose. In a style which, like the Renaissance, allows of, and indeed demands, the association of the Sister Arts, let the artist never lose sight of the unities and specialties of each. Keep them as a well-ordered family, on the closest and most harmonious relations, but never permit one to assume the prerogatives of another, or even to issue from its own, to invade its Sister’s province. So ordered and maintained, those styles are noblest, richest, and best adapted to the complicated requirements of a highly artificial social system, m which, as in that of the Renaissance, Architecture, Painting, Sculpture, and the highest technical excellence in Industry, must unite before its essential and indispensable conditions of effect can be efiSciently realised. M. DIGBY WYATT. BOOKS EEFEERED TO FOE ILLUSTRATIONS, LITERARY AND PICTORIAL. Alciati (A.) Embiemata D. A. Alciati, denuo ab ipso Antore recog- nita ; ac, qute desiderabnnitir, imaginibus locupletata. Accessmtnt vo^la nJiquot ab Auiorc Emhlemata svis quoque eiconibus itisigniin. Small 8vo., Lyons, 1551. Antonelli (G.) Collezione dti migliori Ornnmenti antichi, sparsi nelln citta di Venezia, coll' agginnta di ulcuni frammenti di Goticu archilettura e di varie invenzioni di un Giovnrie Alunno di questa I. R. Accadcmia. Oblong 4to., Venezia, 18J11. Baltard. Paris, ei ses Alonumens, mesvres, dessines, et graves, avec des Descnptions Historiques, par le Ciioyen Amaury Duval; Louvre, St. Cloud, Fontainbleau, Ghdleau d'Ecouen, Ac, S vols. lai-ge folio. Paris, 1808-5. C. Becker and J. von Hefner. Kunslxverhe und Ger'dthschaften des Mittelalters und der Renaissance. 3 vols. 4to. Frankfurt 1853. ’ Bergamo Stefano Da. Wood-Carvings from the Choir of the Mon- aslery of San Pietro at Perugia, IbZb. {Cinque-cento.) Said to be from Designs by Raffaelle. Bernard (A.) Reciieil d’Ornements de la Renaissance. Dessines et graves a I’eau-forte. 4to. Paris, n. d. Ceaput. Le Moyen-Age Pilforesqtie. Manxmens ft Fx-agmens d'Architectxire, Mexibles, Ax-mes, Armures, et Ohjeis de Cxxriosite X° au XI IP Siecle. Dessine d'upres Nature, qxar Chapuy, Ac. Avec un texte archdologique, descriptif et hisiorique, par M. Maret, 5 vols. small folio. Paris, 1838-40. Oleroet et George. Collection portative d’ Ornements de la Renais- sance, recueillis ct choisis par Ch. Ernest Clergct. Graves siir cuivre d’apr'es Ics originaux par C. E. Clerget ei Mme.E. George. 8vo. Paris, 1851. D’Agincodrt, J. B. L. G. S. Histoire de VAx't par ses Monximens, depuis .sa Decadence au IV^. siecle jusqu’a son Renouvellcment XVP. Oxixrrage enrichi de ^2^ planches. 6 vols. folio, Paris, 1833. Dennistoun (J.) Memoirs of the Dukes ofUrbinu, illxistrating the Arms, Arts, and Literature of Italy from 1440 to 1C30. 3 vols. 8vo. London, 1851. Deatlle (A). Unedited Documents on the Histox'xy of France. Oompies de Depexxses de la Construction du- Ch&ieau de Gaillon, publics d'apres les Registres Manvseriis des Trdsoriexs du Car- dinal AAmboise. With an Atlas of Philes. 4to. Paris, 1 850. ' Tombeaux de la Calh^drale de Roxien ; avec doxize planches, gravdes. 8vo. Eouen, 1837. Durelli (G. & F.) La Cei-tosa di Pavia, descriiia ed illustx'ata con tavole, incise daifratelli Gaetano e Francesco Durelli, 62 plates. Folio, Milan, 1853. Dussieux (L.) Essai sur VSistoire de la Pelniure sur Email. 8vo. Paris, 1839. Gailhabaud (J.) L' Architecture du F«. au XVP. Siecle et les Arts qui en dependent, le Sculpture, la Peintxxre Murah, la Peintxxre sur Verre, la Mosu'ique, la Ferronnexne, Ac., puhlies d'aprh les tra- vuux inedits des PrincipauX Archilectes Frangais et Elrangers. 4to. Paris, 1851, et seq. Ghiberti (Lorenzo). Le tre Porte del Battisterio di San Giovanni di Firenze. 46 plates engraved in outline by Lasinio, with description in French and Italian. Folio, half -morocco, Firenze, 1821. Hoffer. Collection of Ornaments in the Grotesque Style, by Hopfer. 127 RENAISSANCE ORNAMENT. Iotbard. Tombean dc Louis XII. ft de Franfoi.^ I., dessinds et gravds au trait, par E. F. Imbard, d'aprh des Marbres du Mnsdc des Petits Augustins. Small folio, Paris, 1823. Jt'BlNAL (A.) Edcherches svr I'Usage et I'Origine des Tapisseries a Personnagcs, dites Histondes, depuis PAntiquite jusqu'au XVP. Si^le iuclnsivement. 8vo. ph. Paris, 184(1. De Laborde (Le Comte Alexandre). Lis Momtmens de la France, classes chronologiqvement, et considdrds sous le Rappoi't des Fails historiques ei de I'Etude des Arts. 2 vols. folio, Paris, 1818-36. De Laborde. Eolice des Emaux exposes dans les Gaieties du Musee du Louvre, Premiere partie, Histoire et Descriptions, 8vo. Paris, 1852. LaBARTE (J.) Description des Ohjeis d'Art qui composent la Collec- tion Debruge-Dvmenil, precedes d’une Introduction Historique. Svo. Paris, 1847. Lacroix et Serb. Le Moycn Age et la Renaissance, Histoire et Description des Meeurs et Zhoges, du Commerce et de I'Industrie, des Sciences, des Arts, des Lilteratvres, ei des Beaux Arts en Europe. Direction Lilteraire de M. Patil Lacroix. Direction Artistique de M. Ferdinand Sere. Dessins facsimiles par M. A, Rivatid. 5 vols. 4to. Paris, 1848-51. Lenoir (Alex.) Atlas des Monumens des Arts liberaux, mdcaniques, et industriels de la France, depuis les Gaulois jusqu'au rhgne de Frangois I. Folio, Paris, 1823. — Musee des Monumens Franpais: ou Description historique et chronologiqiie des Statues en Marbre et en Bronze, Bas-reliefs et Tombeaux des Hommes et des Femmes c^lebres, pour servir d I'Histoire de France et a celle de II Art. Ornee de gravures et augmentie d!iine DisserlaHon sur les Costumes de chaque siecle. 6 vols. 8vo., Paris, 1800-6. Marryat (J.) Collections towards a History of Pottery and Porcelain in the Fifteenth, Sixteenth, Seventeenth, and Eighteenth Centuries, with a Description of the Manufacture ; o Glossary, and a List of Monograms. Illustrated with Colmircd Plates and Woodcuts. 8vo. London, 1850. Morley (H.) Palissy the Potter. The Life of Bernard Paltssy, of Saintes, his Labours and Discoveries in Art and Science, with an outline of his Philosophical Doctrines, and a Translation of Illus- trative Selections from his Works. 2 vols. 8vo., London, 1852. Passeri (J. B.) Histoire des Peintures sur Majoliques failes a Pesari el dans les lieux circonvoisins, decrite par Giambattista Passeri {de Pesaro). Traduite de I’ltalien et suivie d'un Appendice par Henri Delange. 8vo. Palis, 1853. Queriere (E. de la). Essai sur les Girouettes, Epis, Crgtes, Ac., des Anciens Combles et Pignons. Numerous plates of Ancient Vanes and Terminations of Roofs, Paris, 1846. Renaissance. La Fleur de la Science de Pourtraicture et Patrons de Brodcrie. Fapon Arabicque et Ytalique. Cum PrivUegio Regis. 4to. Paris. Reynard (0.) Ornemens des Anciens Mattres des XV., XVI., XVII. et XVIII. Siecles. 30 plates comprising copies of some of the most ancient and rare prints of Ornaments, Alphabets, Silverwork. Folio, Paris, 1844. Serb (F.) Les Arts Somptunires de V^. au XVIP Siecle. Histoire du Costume et de V Ameuhlement en Europe, et des Arts que en de- pendent. Small 4to. Paris, 1853. Sommerard (A. Du.) Les Arts au Moyen Age. (Collection of the HStel de Cluny.) Text, 5 vols. 8vo. ; Plates, 6 vols. folio. Paris, 1838-46. Verdier et Cattois. Architecture Civile et Domestique au Moyen Age et a la Renaissance. 4to. Paris, 1852. Waring and MacQuoed. Examples of Architectural Art in Italy and Spain, chiefly of the 13fA and 16-. ITALIENISCH ITALIAN N° P ITALIEHS TAFEL UXXYI f PL.UXXYI* 1 ^ f4 i6 ITALIEHI3CH ITALIAN ITALilEHS ITALIENISCH ITALIAN 5^ italiehs TAFfc o xr PI. y.c iri7ii'~fi R R 153 LEAVES AND FLOWEES FEOM NATUKE. 1. Isis. 2. mite Lily. 3. Daffodil. 4. Narcissus. 5. Onion. 6. Dog-Eose. PLATE XCVIII. Plans and Elevations of Flowers. 7. Mouse-ear. 8. Honeysuckle. 9. Mallow. 10. Ladies’smock. 11. Speedwell. 12. Harebell. 13. Glossocomia clematidca. 14. Convolvulus. 15. Primrose. 16. Periwinckle. 17. Clarkia. IS. Leycesteria forraosa. PLATE XCIX. 1. Honeysuckle. 2. Convolvulus. Full size. PLATE C. Passion Flowers. Full size. LEAVES AND FLOWEES EKOM NATUEE. We have endeayoured to show in the preceding chapters, that in the best periods of art, all ornament was rather based upon an observation of the principles which regulate the arrangement of form in nature, than on an attempt to imitate the absolute forms of those works; and that whenever this limit was exceeded in any art, it was one of the strongest symptoms of decline : true art consisting in idealizing, and not copying, the forms of nature. We think it desirable to insist rather strongly on this point, as in the present uncertain state in which we are, there seems a general disposition arising to reproduce, as faithfully as may be possible, natural form as works of ornament. The world has become weary of the eternal repetition of the same conventional forms which have been borrowed from styles which have passed away, and there- fore can excite in us but little sympathy. There has risen, we say, a universal cry of “ Gro back to nature, as the ancients did ; ” we should be amongst the first to echo that cry, but it will depend much on what we go to seek, how far we may succeed. If we go to Nature as the Egyptians and the Greeks went, we may hope; but if we go there like the Chinese, or even as the Gothic artists of the fourteenth and fifteenth centuries, we should gain but little. We have already, in the floral carpets, floral papers, and floral carvings of the present day, sufficient evidence to show that no art can be produced by such means; and that the more closely nature is copied, the farther we are re- moved- from producing a work of art. Although ornament is most properly only an accessory to architecture, and should never be allowed to usurp the place of structural features, or to overload or to disguise them, it is in all cases the very soul of an architectural monument. 154 LEAVES AND FLOWEES FROM NATURE. By the ornament of a building, we can judge more truly of the creative power which the artist has brought to bear upon the work. The general proportions of the building may be good, the mouldings may be more or less accurately copied from the most approved models; but the very instant that ornament is attempted, we see how far the architect is at the same time the artist. It is the best measure of the care and refinement bestowed upon the work. To put ornament in the right place is not easy; to render that ornament at the same time a superadded beauty and an expression of the intention of the whole work, is still more difficult. Unfortunately it has been too much the practice in our time to abandon to hands most unfitted for the task the adornment of the structural features of buildings, and more especially their interior decorations. The fatal facility of manufacturing ornament which the revived use of the acanthus leaf has given, has tended very much to this result, and deadened the creative instinct in artists’ minds. What could so readily be done by another, they have left that other to do ; and so far have abdicated their high position of the architect, the head and chief. How, then, is this universal desire for progress to be satisfied — how is any new style of ornament to be invented or developed ? Some will probably say, A new style of architecture must first be found, and we should be beginning at the wrong end to commence with ornament. We do not think so. We have already shown that the desire for works of ornament is co-existent with the earliest attempts of civilisation of every people; and that architecture adopts ornament, does not create it. The Corinthian order of architecture is said to have been suggested by an acanthus leaf found growing round an earthen pot ; but the acanthus leaf existed as an ornament long before, or, at all events, the principle of its growth was observed in the conventional ornaments. It was the peculiar application of this leaf to the formation of the capital of a column which was the sudden invention that created the Corinthian order. The principle of the foliation, and even the general form of the leaves, which predominate in the architecture of the thirteenth century, existed long before in the illuminated MSS. ; and derived as they were, most probably, from the East, have given an almost Eastern character to Early English ornament. The architects of the thirteenth century were, therefore, very familiar with this system of ornamentation ; and we cannot doubt, that one cause of the adoption so universally of this style during the thirteenth century arose from the great familiarity with its leading forms which already existed. The floral style, in direct imitation of nature, which succeeded, was also preceded by the same style in works of ornament. The facility of painting flowers in direct imitation of nature in the pages of a missal, induced an attempt to rival them in stone in the buildings of the time. The architectural ornament of the Elizabethan period is mostly a reproduction of the works of the loom, the painter, and the engraver. In any borrowed style, more especially, this would be so. The artists in the Elizabethan period were necessarily much more familiar with the paintings, hangings, furniture, metal-work, and other articles of luxury, which England received from the Continent, than they would be with the architectural monuments ; and it is this familiarity with the ornamentation of the period, but imperfect knowledge of the architecture, which led to the development of those peculiarities which distinguish Elizabethan architecture from the purer architecture of the Eevival. We therefore think we are justified in the belief, that a new style of ornament may be produced independently of a new style of architecture ; and, moreover, that it would be one of the readiest means of arriving at a new style ; for instance, if we could only arrive at the invention of a new termi- nation to a means of support, one of the most difficult points would be accomplished. 155 LEAVES AND FLOWERS FROM NATURE. The chief features of a building which form a style, are, first, the means of support ; secondly, the means of spanning space between the supports ; and, thirdly, the formation of the roof. It is the decoration of these structural features which gives the characteristics of style, and they all follow so naturally one from the other, that the invention of one will command the rest. It would appear, at first sight, that the means of varying these structural features had been ex- hausted, and that we have nothing left but to use either one or the other of the systems which have already run their course. If we reject the use of the column and horizontal beam of the Greeks and Egyptians, the round arch of the Eomans, the pointed arch and vault of the Middle Ages, and the domes of the Mohammedans, it will be asked — WTiat is left? We shall perhaps be told that all the means of covering space have already been exhausted, and that it were vain to look for other forms. But could not this have been said in all time ? Could the Egyptian have ever imagined that any other mode of spanning space would ever be found than his huge blocks of stone ? Could the Mediseval architect have ever dreamed that his airy vaults could be surpassed, and that gulfs could be crossed by hollow tubes of iron? Let us not despair; the world has not seen, most assuredly, the last of the architectural systems. If we are now passing through an age of copying, and architecture with us exhibits a want of vitality, the world has passed through similar periods before. From the present chaos there will arise, undoubtedly (it may not be in our time), an architecture which shall be worthy of the high advance which man has made in every other direction towards the possession of the tree of knowledge. To return to our subject, how is any new style of art or new style of ornament to be formed, or even attempted to be formed? In the first place, we have little hope that we are destined to see more than the commencement of a change; the architectural profession is at the present time too much under the influence of past education on the one hand, and too much influenced by an ill- informed public on the other ; but the rising generation in both classes are born under happier auspices, and it is to them we must look for hope in the future. It is for their use that we have gathered together this collection of the works of the past; not that they should be slavishly copied, but that artists should, by an attentive examination of the principles which pervade all the works of the past, and which have excited universal admiration, be led to the creation of new forms equally beautiful. VI e believe that if a student in the arts, earnest in liis search after knowledge, will only lay aside all temptation to indolence, will examine for himself the works of the past, compare them with the works of nature, bend his mind to a thorough appreciation of the principles which reign in each, he cannot fail to be himself a creator, and to individualise new forms, instead of reproducing the forms of the past. We think it impossible that a student fully impressed with the law of the universal fit- ness of things in nature, with the wonderful variety of form, yet all arranged around some few fixed laws, the proportionate distribution of areas, the tangential curvatures of lines, and the radiation from a parent stem, whatever type he may borrow from Nature, if he will dismiss from his mind the desire to imitate it, but will only seek to follow still the path which it so plainly shows him, we doubt not that new forms of beauty will more readily arise under his hand, than can ever follow from a continuation in the prevailing fashion of resting only on the works of the past for present inspiration. It will require but a few minds to give the first impulse ; the way once pointed out, others will follow, readily improving, refining upon each other’s efforts, till another culminating point of Art shall be again reached to subside into decline and disorder. For the present, however, we are far enough removed from either stage. We have been desirous to aid this movement to the extent of our power; and in the ten plates of leaves and flowers which accompany this chapter, we have gathered together many of those natural types which we thought best calculated to awaken a recognition of the natural laws which prevail in 156 LEAVES AND FLOWERS FROM NATURE. the distribution of form. But, indeed, these laws will be found to be so universal, that they are as well seen in one leaf as in a thousand. The single example of the chestnut leaf, Plate XCI., contains the whole of the laws which are to be found in Nature : no art can rival the perfect grace of its form, the perfect proportional distribution of the areas, the radiation from the parent stem, the tangential curvatures of the lines, or the even distribution of the surface decoration. We may gather this from a single leaf. But if we further study the law of their growth, we may see in an assemblage of leaves of the vine or the ivy, that the same law which prevails in the formation of the single leaf prevails also in the assemblage of leaves. As in the chestnut leaf, Plate XCI., the area of each lobe diminishes in equal proportion as it approaches the stem, so in any combination of leaves each leaf is everywhere in harmony with the group: as in one leaf the areas are so perfectly distributed that the repose of the eye is maintained, it is equally so in the group: we never find a disproportionate leaf interfering to destroy the repose of the group. This universal law of equilibrium is everywhere apparent in Plates XCVIII., XCIX., C. The same laws prevail in the distribution of lines on the surface of flowers; not a line upon the surfaces but tends more surely to develope the form, — not a line which could be removed, and leave the form more perfect; and this why? Because the beauty arises naturally from the law of the growth of each plant. The life-blood, — ^the sap, as it leaves the stem, takes the readiest way of reaching the confines of the surface, however varied that surface may he; the greater the distance it has to travel, or the weight it has to support, the thicker will be its substance. (See Convolvulus, XCVIII., XCIX.) On Plate XCVIII. we have shown several varieties of flowers, in plan and elevation, from which it will be seen that the basis of all form is geometry, the impulse which forms the surface, starting from the centre with equal force, necessarily stops at equal distances; the result is symmetry and regularity. Who then will dare say that there is nothing left for us but to copy the five or seven-lobed flowers of the thirteenth century ; the Honeysuckle of the Greeks or the Acanthus of the Romans, — that this alone can produce art, — is Nature so tied ? See how various the forms, and how unvarying the principles. We. feel persuaded that there is yet a future open to us; we have but to arouse from our slumbers. The Creator has not made all things beautiful, that we should thus set a limit to our admiration; on the contrary, as all His works are offered for our enjoyment, so are they offered for our study. They are there to awaken a natural instinct implanted in us,— a desire to emulate in the works of our hands, the order, the symmetry, the grace, the fitness, which the Creator has sown broadcast over the earth. London Day and Son, Limited, Gate Street, Lincoln’s Inn Fields. 157 ii F ; =’iC' '■ ■ m *•, ■,.• '.'•i'’('./^ •" ■■■ ■ - ■■ ., •'. Iim^ '--- . - ^.r ■ ■ T: Si'.’^' .'■■ * '■; . . '.V JHji^,, ;- . ■ ■" ■''Wy*- 'f- ■/: ' ■'"■'■’■ .■'V'X' - - K # H, ,,^:J-t-J.,,,. Y. ■:••. :■. *l}rU ■ .' V "rT. ^cry'''tW;4£*^.»- - '^’V r ■' ■■*)■' ^ ^ *1 L^ n** **^*'»' ■-’ sSK^K ' m Nfy.'., r*'^' ijc^ ■ i'M&i a. ; h;.ATTKH NACH OER HATUR LEAVES FROM NATURE N°2 feuilles dapreb nature TAFEL XCIl PL XCL SLCTSH SACH DER NATUR LEAVES FROM NATURE N°3: feuilles d'apres nature ■AFfc;L PT.ynii BLATTER SACK DER NATUR LEAYE.S FROM NATURE N°4, FEUILLES D'APRES NATURE iujmil4 BLATTER KACH DER HATUR LEAVES FROM NATURE N®7. FEUILLES D'APRES NATURE. 'lAFE.UXCyU. ' PL. XCYII