Digitized by the Internet Archive in 2013 http://archive.org/details/facsimileofsketcOOvill FACSIMILE OF THE S K E T C H-B K OF WILARS DE HONECORT, AN ARCHITECT OP THE THIRTEENTH CENTURY. FACSIMILE OF THE SKETCH-BOOK OF WILARS DE HO NEC OUT, AN ARCHITECT OF THE THIRTEENTH CENTURY ; ILLUSTRATED BY COMMENTARIES AND DESCRIPTIONS, AS ARRANGED WITH VARIOUS ADDITIONS, AND PUBLISHED BY M. ALFRED DARCEL FROM THE MSS. OF M. J. B. A. LASSUS, LATE ARCHITECT OF NOTRE-DAME AND OF THE SAINTE CHAPELLE AT PARIS, &C. li? -\N SLATED, EDITED, AND AUGMENTED WITH MANY NEW ARTICLES AND NOTES, AND WITH THE REMARKS OF M. J. QUICHERAT, PROFESSOR OF ARCH.-EOLOGY AT THE ECOLE DES CHARTES AT PARIS, BY THE EEY. KOBEKT WILLIS, M.A., F.E.S., JACKSONIAN PROFESSOR OF NATURAL AND EXPERIMENTAL PHILOSOPHY IN THE UNIVERSITY OF CAMBRIDGE ; MEMBER OF THE IMPERIAL LEGION OF HONOUR J CORRESPONDING MEMBER OF THE ROYAL ACADEMY OF SCIENCES AT TURIN, &C, &C. LONDON: JOHN HENRY and JAMES PARKER, 377, STRAND. 18,59. PREFACE. rpiIE manuscript which is the subject of the present volume is a most valuable monument of the state of the art of delineation in the thir- teenth century. The actual works of painting, sculpture, and architecture which remain to us exhibit the finished results of those branches of the fine arts. This volume exemplifies the manner in which the artists carried on their studies. It proves that if they did not attain to perfection in represent- ing corporeal forms, it was not for want of perceiving that they ought to be studied from the life, or from neglecting to carry out such studies. It also shews that they were not deterred by pious prejudices from copying the antique. Wilars de Honecort has himself recorded that his lion was from nature, — many other of his animals were certainly so. Several of his human figures are evident academic studies from living models set in attitudes for the purpose : and their anatomical details are most carefully worked out, as well as the artist could manage them. One page is occupied by an unmistakeable Greek, dressed in a chlamys; another by a drawing of a Eoman sepulchral monument, with figures. In these examples the drapery was evidently the object of his admiration, for the human forms and the architecture are transformed into the styles that were familiar to him, after the manner of all the artists who attempted to delineate antiquity before the present century. The architectural drawings are especially interesting for the light they throw upon mediaeval practice. For example, "Wilars de Honecort travels to Eheims, apparently to collect materials, by which to copy portions of it for his buildings of the choir of Cambray, and preserves for us the resulting draw- ings. I have shewn that in one instance at least, where he has drawn a part vi PREFACE. of Eheims erroneously, the corresponding part of Cambray was erected as he drew it, and not as it stood at Rheims. I have also shewn that in his draw- ings of Rheims exactness in proportion and detail are neglected, and that, with few exceptions, he drew the buildings as he drew the antiques, not as they existed before his eyes, but in the fashion which they had assumed when his drawings were made, and to which his own practice had accustomed him. The drawings are expressed by very few lines, all the minor details being omitted, but their absence is compensated for by sections of moldings. No dimensions are given, and the comparison of the drawings with the build- ings they represent shews that the relative magnitudes of the parts to the whole are never preserved. But it must be remembered that the sketches were made for his own use, and not for the purpose of conveying to others the aspect of the buildings represented. Consequently, he only drew the combinations that struck his fancy as likely to be suggestive in his practice. He trusted to his own experience to supply dimensions when the occasion arose to make full-sized working drawings for the masons. He appears even to have altered parts of the buildings he was sketching, im- proving them as he thought, and giving them a more fashionable air as he went along, to save himself the trouble of doing so when he wished to engraft them upon one of his own designs. It is evident that the methods of drawing which this Sketch-book has pre- served to us are wholly insufficient to convey any ideas of the exact proportions or artistic character of an edifice. But we see that in those days there could have been none of the mechanical copying which is the reproach and mis- fortune of our own, because there was no sufficient power of delineation to enable a travelling architect to transfer a building or a detail to his sketch- book so completely as to admit of its being reproduced when its effect upon his eyes had been forgotten. He might have caught inspiration from the sight of great works in his passage, but unless he possessed a genius of the same order as that which had originated them, he would have been wholly PREFACE. vii unable to give to his imitations their beauty and spirit, and in any case, must have supplied so many details of his own to enable the building to be erected, that it would necessarily acquire an individual character. Our artist has in general furnished us with no means of determining whether his compositions and sketches are original or copies. Many of them are certainly drawn by himself from the life or reality. His geometrical me- thod of portraiture must not be considered as his own invention, but merely as a collection of examples to place the system on record for the use of his successors. Neither can his series of geometrical devices relating to masonry and construction claim to represent the ordinary practice of his period, for a regularly educated architect would not make notes of matters familiar to himself and his fellow-workmen. From the nature of these problems it is manifest that the whole is merely a chance collection of expedients to meet particular cases, or of novel methods which struck him in the course of his travels, and which he noted for his own use. The same may be said of the machines, which, however, are for the most part extremely curious, and shew the antiquity of many contrivances now familiar to us. The facsimile plates which are contained in this volume were engraved at Paris in 1851, under the direction of M. Lassus. The collection of materials for their illustration and the preparation of the commentary employed so much time, that upon his lamented death on the 15th of July, 1857, the work was left incomplete. His manuscripts were, however, placed by his family in the hands of M. Darcel, and were by him prepared for publication, and finally issued from the press in the autumn of the last year. The text of the present volume differs in many respects from that of the French edition, although based upon it in the same manner as that was based upon the original and admirable commentary published in 1849 by M. Jules Quicherat, the Pro- fessor of Archaeology at the Ecole des Chartes. To that essay I owe my first knowledge of the existence of the manuscript. In 1851, I eagerly sought the original in the Bibliotheque Imperiale, and obtained the rare privilege of viii PREFACE. tracing those of its pages which interested me as belonging to architecture and mechanism. Having thus obtained the materials for studying at leisure the interpre- tation of these selected portions, I was induced to postpone the publication of the results I had arrived at, by the prospect of a speedy appearance of the whole from the able hands of the eminent and highly qualified editor who, as I found, had undertaken the work. But as his labours have been unhappily cut short and left imperfect, I have ventured to add to them my own, and to attempt the formation of a commentary that should include the opinions of the writers who have as yet interested themselves in the question. Much matter at the beginning of the French edition relating to a con- troversy between the classical and mediseval styles in Paris has been omitted, as foreign to the illustration of our artist. In justice to M. Quicherat, I have substituted a translation of his spirited and ingenious essay for the Notice sur Villard de Honnecourt with which the French editor has prefaced the commentary. I have supplied a description of the manuscript itself, with classified tables of the subjects of the drawings and their peculiar arrange- ment, which I trust will facilitate an acquaintance with its varied contents. In accordance with the French edition, each plate is furnished with its own explanation. Those which relate to the drawings of figures and animals are literally translated from the admirable articles of M. Lassus, with a few slight abridgments. But the discussions of the architectural plates, and those which relate to geometry and mechanics, have either received large additions, or are entirely new. In the former case, each portion has been distinguished by the initial of the writer to whom it belongs a , placed at the end of it ; such initial being understood to include all the matter above it which extends either to the beginning of the article or to some previous initial. The foot-notes are a These initials are : (Q.), Quicherat ; (L.), Lassus ; omitted. Iu other cases where the initials have been (A.D.), Alfred Darcel ; (W.), Willis. In the descrip- left out, the use of the first person in the text will at tions of the first six plates, which happen to apply once shew that it is written by the present editor, wholly to figures and animals, and are entirely de- — (W.) rived from the French, this notation was accidentally - PREFACE, ix marked in the same manner. When the article has been written anew, the opinions of previous commentators worked into it have been carefully dis- tinguished by name. I cannot better close this Preface than by a concise glance at the labours of the amiable and highly cultivated artist with whose name mine has been in this volume associated. J. B. A. Lassus, born at Paris on the 19th of March, 1807, became, in 1828, a student of the Ecole des Beaux Arts at Paris, at a period when the contest between the classical and mediaeval architects had been fostered by the publication of Victor Hugo's celebrated novel, the Notre- Dame de Paris. Lassus became a devoted medievalist. His first architectural work was the restoration of the refectory of St. Martin des Champs at Paris, to fit it for the library of the Conservatoire des Arts et Metiers. It was fol- lowed by the restoration of St. Severin in 1837, St. Germain l'Auxerrois in 1838, and others, including that of the Sainte Chapelle, which was in 1849 left completely in his hands. Besides these restorations, which were the means of forming a school of sculptors, glass-painters, smiths, decorators, and workers in wood, Lassus erected many mediaeval buildings from his own designs, of which the first was St. Nicholas at Nantes in 1843, followed by the con- struction in 1848 of a nave to the Cathedral of Moulins, of which only a choir had previously existed, and of many others. He wrote much, constantly struggling in defence of his favourite style against the partisans of classical antiquity, and maintained his place in the first rank of his profession. In 1850 his merits received the distinction, so highly valued by his countrymen, of the cross of the Legion of Honour. But in the midst of his prosperity and success an insidious malady was working within, and finally removed him from his earthly labours in 1857, leaving a blank in his profession that will not easily be filled up. E. WILLIS. b CONTENTS. PAGE Preface . . . . • . • • T List of Engravings . . • • • • . xi Essay on Wilars de Honecort and his Sketch-book, by M. Jules Quicherat . . 1 Description of the Manuscript and its Contents . . . . .10 Tabular View of the various Pagings, Quires, and Subjects of the Manuscript . .16 Classified list of the Subjects of the Drawings . . . ' * .17 Explanation of the Plates . . . - . . . .21 N.B. As each Plate is placed opposite to its own explanation, the Classified List (p. 17), compared with the List of Engravings (p. xi.), will enable any required subject and its explanation to be referred to with facility. Addenda et Corrigenda .... ... 239 Glossarial Index . . . . . . . .241 LIST OF ENGRAVINGS. 1. FACSIMILE PLATES. PLATE PAGE PLATE PAGE PLATE PAGE PLATE PAGE I. . . . 21 XVII. . . 57 XXXIII. . . 105 XIIX. . . 177 n. . . . 23 XVIII. . . 59 XXXIV. . . 109 I. . . 179 in. . . . 25 XIX. . . 63 XXXV. . . 113 II. . . 181 IV. . . . 27 XX. . . 65 XXXVI. . . 115 III. . . 183 v. . . . 29 XXI. . . 67 XXXVII. . . 117 IIII. . . 185 VI. . . . 31 XXII. . . 69 XXXVIII. . . 119 IIV. . . 187 VII. . . . 33 XXIII. . . 71 XXXIX. . . 135 IV. . . 189 VIII. . . . • 35 XXIV. . . 73 XL. . . 147 IVI. . . 191 IX. . . . 37 XXV. . . 75 XLI. . . 155 IVII. . . 193 X. . . . 39 XXVI. . . 77 XIII. . . 157 IVIII. . . 195 XI. . . . 41 XXVII. . . 79 XIIII. . . 159 IIX. . . 205 XII. . . . 45 XXVIII. . 91 XIIV. . . 165 IX. . . 217 XIII. . . . 47 XXIX. . . 97 XIV. . . 169 IXI. . 219 XIV. . . . 49 XXX. . 99 XIVI. . . 171 IXII. . 227 XV. . . . 51 XXXI. . . 101 XIVII. . . 173 IXIII. . . 235 XVI. . . . 53 XXXII. . 103 XIVIII. . . 175 IXIV. . . 237 2. ILLUSTEATIVE PLATES. PLATE ^IXV. IXVI. I IXVII. IXVIII. ✓ IXIX. . IXX. LXXI. - IXXII. IXXIII. ( Cathedrale de Chartres, labyrinthe I Xotre Dame de laon, Plan de la tour du Xord Notre Dame de laon, Tour du Nord Plan of Xotre Dame de Cambray West elevation of do, South elevation of do, St. Faron de Meaux Rose window of the Cathedral at Chartres Rose window of the Cathedral at lausanne Portrait of J. B. A. lassus facing Plate XVII. s „ Plate XVIII. page 90 page 87 page 95 „ Plate XXIX.. „ Plate XXX. Title xii LIST OF ENGRAVINGS. 3. WOODCUTS. HGTTEE PAGE 1. Plan of the church of St. Elizabeth at Cassovia ■) 8 2. Plan of the church of St. Yved de Braine . j 3. * Wrestlers, from the MS. of the Roman d Alexandre . 80 4. Plan of the Cathedral of Eheims ..... . 89 5. TVl ,i 1j P 1 1 1 • 1 1* A TTT'I 1 TT J_ Plan of the vault of double aisles, according to W liars de Honecort ) . 93 6. Plan oi the vault ot double aisles at JN otre Dame de Paris ) 7, 8. How to shape the springing stones oi a vault . 124 9. How to shape a skew voussoir ..... . 126 10. »** rn i j_ j_i * i _/> i • j " lo lay out the site ot a cloister ..... . 127 11. To set the four corner-stones of a cloister .... . 129 12. To cut the voussoirs of vaulting surfaces .... . 135 13. *lo cut the key-stone ot an arch of the fourth point . 141 14. A lo cut the key-stone of an equilateral arch .... . 142 15. *To cut the key-stone of an arch of the third point . ib. 16. *The finished key-stone of the third point .... ib. 17. *To cut the key-stone of an arch of the fifth point . ib. 18. *The finished key- stone of the fifth point .... . ib. 19. Plan of a square chapter-house ..... . 148 20. *To trace a pentagon ...... . 149 21. Plan of a pentagonal tower ... . ib. 22. To draw three kinds of arches with one radius . 152 23. Diagram of an equilateral arch ..... . 153 24. To draw a spiral ...... . ib. 25. *Honecort's Lewis ...... . 163 26. *Honecort's Trebuchet .... . 197 27. *To explain the action of the Trebuchet .... . 198 28. Trebuchet, from a German miniature of the fourteenth century . 199 29. *Trebuchet, from the Romance of Alexander .... . 200 30. Trebuchet, from a French manuscript of the fourteenth century . 202 31. Interior view of an apsidal chapel of Rheims Cathedral, by M. Viollet-le-Duc . 205 32. ^Diagram of base-moldings of do. . . 212 33. *Base-mold of do. in perspective ..... . 213 34. Exterior view of an apsidal chapel of Rheims Cathedral, by M. Viollet-le-Duc . 215 35. Section of the original battlement of Rheims Cathedral \ . 218 36. Plan of do. . . . ) 37. *Tracery of the chapel windows of do. . 221 38. *Tracery of the nave windows of do. . ib. 39. *Honecort's drawing of the latter ..... . ib. 40. *Window of the eastern gable of Salisbury Cathedral . 222 41. *Plan of one of the great crossing-piers of Rheims Cathedral . 227 42. *Plan of one of the piers that separate the chapels . 229 43. * Sections of moldings of the chapel windows .... . 233 These woodcuts are the same as those of the French edition, with the exception of those marked with an asterisk, which have been engraved by Mr. Jewitt for the present edition, in additional illustration of Honecort's text. The blocks of the two views of the apsidal chapel of Rheims Cathedral (figs. 31 and 34), were obligingly lent to the publisher by the editors of M. Viollet-le-Duc's Dictionary of Architecture. ESSAY ON WILARS DE HONECORT AND HIS SKETCH-BOOK, BY M. JULES QUICHERAT. (Extracted from the Revue Archeologique for 1849, t. vi. p. 65.) ^JUIE uncertainty which prevails with respect to the practical methods employed by the mediaeval artists, and our absolute ignorance of the manner in which they were taught and trained, must create an interest in the description of a manuscript, unique of its kind, which is apparently the sketch-book of an archi- tect of the thirteenth century. This singular work, which we shall term an Album, is contained in the collection of manuscripts from the Abbey of St. Germain des Pres deposited in the Imperial Library at Paris, (S. G. Latin, 1,104). It is a small volume of thirty-three leaves of vellum stitched into a thick rough leather cover a , which wraps over the front edge of the leaves. A memorandum written in the fifteenth century on the last page records that the book then contained forty-one leaves, but the mutilations by which their number has been reduced to thirty-three are apparently of consider- able age. The leaves are not all cut to the same size, their dimensions vary- ing from 6 to 6| inches in breadth, and from 9 to 9^ in height. Each of them is occupied with pen and ink drawings that have been previously sketched with a lead point, and many of the drawings are accompanied by explanatory notes written in the Picard dialect of the thirteenth century, and in the running-hand of that period. This manuscript was known to Willemin, who selected from it a sufficient number of figures to compose a plate of costumes for his Monuments Francais Inedits h , and M. Pottier accordingly examined it, and concisely described it in the a A more detailed account of the volume is given the manuscript of De Honecort, and to this plate the in the following chapter. following description is supplied : — " Le volume qui b Willemin's Monuments Frangais Inedits, pub- a fourni ces costumes, dessines au simple trait, est lished at Paris in 1839, was commenced in 1806. un recueil extremement singulier et digne de tout l'in- The descriptive text was written by Pottier. In pi. teret des artistes, c'est V Album le calepin d'un artiste 102, vol. i., Willemin engraved several subjects from du xiii e . siecle, qui a depose sur ses pages toutes les B 2 ESSAY ON WILARS DE HONECORT explanatory text of that work. It was subsequently shewn to several experienced antiquaries, who carefully studied it, but reserved their opinions, possibly from the difficulty they found in discovering a satisfactory interpretation of the whole of its contents. The writer of these remarks has no such ambitious pretensions, for who can be expected to explain every part of a miscellaneous collection, em- bracing every branch of construction and decoration. The purpose of this essay is simply to follow up the discovery of Willemin and Pottier by making a stronger appeal to the attention of the learned, in the form of such a detailed description of the contents of this precious volume as may induce them to study and discuss it, to publish it more completely, or at least to extract from it all the valuable infor- mation that can be obtained for the advance of archaeological science. We will now endeavour to draw out, by comparing together the explanatory notes already mentioned with the subjects of the drawings, some particulars relating to the author of the manuscript, the period at which he lived, and his works. On the second page (pi. 2) we read, " Wilars de Honecort salutes you, and im- plores all who labour at the different kinds of work contained in this book to pray for his soul, and hold him in remembrance. For in this book may be found good help to the knowledge of the great powers of masonry, and of devices in carpentry. It also shews the power of the art of delineation, the outlines being regulated and taught in accordance with geometry." This may pass for the author's preface. It tells us his name, his birth-place, and the nature and purpose of his book. Wilars de Honecort having com- piled this collection, bequeaths it to future artists in the same pursuits, requiring but their grateful remembrance and their prayers. Wilars, to judge by his surname, was a Cambraisian, for Honuecourt is a village on the Scheldt, five leagues south of Cainbray. This conjecture acquires more fautaisies de son imagination toutes les acquisitions de son savoir. On y tronve des sujets pieux, des scenes domestiques, des modeles d'architecture, des problemes de geometric Voici au reste, sinon le titre, au moins le preambule exact dont l'auteur a fait preccder son ouvrage." Here follows the en- tire legend of the second page, including the " Ci poies vos trover lesagies (miracles?) des xij. Apostles en scant," which the writer has hastily confounded with Honecort's real preface that follows it. Pottier next gives a description of the subjects selected in the plate, observing that "Le style des draperies etudie, fouille, et tourmente comme dans certaines figures du xvi e . siecle, est vraiment extraordinaire pour l'epoque, si tant est qyHil rCait pas t'te rajeuni par le graveur." The last remark is enough to prove that M. Pottier had not taken the trouble to examine the original manuscript at all, but described the figures from the engravings of Willemin. The de- scriptive note of the contents of the manuscript above quoted was probably written by Willemin himself at the time when the tracings were made for his en- graving. The figures selected by Willemin are the two foliage faces of pi. 9, the group in pi. 26, and the warvior mounting his horse in pi. 45. — (W.) AND HIS SKETCH-BOOK. 3 certainty from the subjects of two of his drawings, the one a plan of the Church of Vaucelles, an abbey close to Honnecourt, the other a plan of the choir of the Cathedral of Cambray. Like most of the men of his time who pretended to knowledge or cultivation, our architect had travelled. "I have been in many lands," he writes ; adding, " as this book shews." Effectually, the book is an itinerary : his steps may be traced in it through France from north to east, and across the German empire to its extreme limits. Stopping at Laon, he sketches one of the towers of its cathedral, " the most beautiful that the world contains d ." His careful studies of the architecture of the Cathedral of Rheims shew that he remained there a long time. His passage by Meaux is attested by a plan of St. Stephen, and his visit to Chartres by a drawing of the great western rose-window of the cathedral. In the next place we find him before the west front of Lausanne Cathedral, making a hurried sketch of the rose-window there. Lastly, his Album bears evidence of a long residence in Hungary. It is to be regretted that the manuscript of Wilars de Honecort contains so little information concerning his own architectural works. In fact, there is but one composition distinctly claimed as his own, and he even shares the merit of that with a fellow-workman e . It is simply a plan for the presbytery of a church of the largest class. The choir is circumscribed by a double aisle, with nine chapels radiating outwards, and alternately square and semicircular in plan f . " Vlardus de Hunecort and Petrus de Corbeia contrived this presbiterium in a discussion together." There is no evidence to shew that it was ever actually erected. For lack of direct proofs by which to place our Cambraisian artist amongst the great masters of construction of the thirteenth century, we must have recourse to induction. One of the allusions to his journey to Hungary is made upon occasion of a sketch which he took at Rheims : — " When I drew this I was under orders to go to Hungary g ." Why under orders, unless commissioned to work as an artist in his profession ? His reputation must have been so thoroughly established as to have extended to the confines of Europe ; and as it is improbable that an archi- tect would have been fetched from a distance of four hundred leagues for a trifling c Fl. 17. not his inventions, but merely copies. — (W.) d Ibid. f PI. 28. Corbie is a village in Picardy, near Amiens. e The care with which he explains that this is his s PI. 19. own work seems to shew that the other drawings are B 2 i ESSAY ON WILARS DE HONECORT work, we can only suppose that Wilars de Honecort journeyed to Buda or Strigonium to erect some magnificent church. It has been already stated that one of the drawings in the Album h is a plan of the ancient Cathedral of Cambray. Under this is a note to the effect that it represents the plan of the eastern part, or " chevet," of that church, " as it is now rising from the ground. Farther on in this book will be found the interior and exterior elevations, the arrangement of the chapels and side walls, and the form of the flying buttresses." But instead of these promised drawings, we find at the end of the book a set i representing the analogous parts of the Cathedral of Rheims most carefully drawn. Above one of those which represent the chapels is written, " This shews the elevations of the chapels of the Church of Rheims — like them will be those of Cambray if they be built." It appears, therefore, that when this was written the east end of Cambray Cathe- dral was actually commenced but unfinished, and that the reference to the eleva- tions of Cambray written under the plan was merely a reference to those of Rheims which were to be taken as the model of the former. But, as we find our architect thus identifying in his mind these two buildings, declaring beforehand, and with the air of a master, the form which it was in- tended to give to the unfinished parts of Cambray, and at the same time study- ing and copying most minutely the portions of Rheims which were required to complete them, how can we escape the inference that Wilars de Honecort was the architect of the Cathedral of Cambray. It may be said, that supposing this to be true we have only proved him to be a plagiarist, instead of a man of originality and ability. But there may have been other reasons to make a similarity between these two churches imperative upon the architect. Cambray was not the metropolitan church, but was dependent upon the archiepiscopal Church of Rheims. Archaeology shews that this kind of ecclesiastical relation between churches was often expressed architecturally by similarity of plan or style. The partial reproduction of the Church of Rheims at Cambray may be the result of this principle, and not of a defect of originality in the architect. But if it were a copy, the sanctuary of Cambray must have had an aspect of peculiar magnificence. For there was an old saying in the North, that to make a perfect church you must unite the choir of Cambray, the nave of Arras, the tran- sept of Valenciennes, and the steeple of Antwerp. The traditions of the country " PI. 27. 1 Pis. 59, GO, &c. AND HIS SKETCH-BOOK. 5 lament its loss, for it was destroyed at the Revolution. But in 1824, when its site was levelled, the architect of the city, M. Aime Boileux, was enabled to make a complete plan, which was engraved in M. Leglay's history of the Cathedral k . This plan coincides exactly with that given in our manuscript. If the above reasoning be thought to have led us rather to probabilities than to certain information respecting the practice of Wilars do Honecort as an archi- tect, the facts appealed to will at least enable us to determine with mathematical precision the age of the manuscript, and thus the period of the author; for we have only to extract from the histories of the two buildings of Cambray and Rheims the dates of the respective portions already alluded to. The Cathedral of Cambray was originally Romanesque throughout, but its eastern portions were reconstructed on an enlarged plan in the thirteenth century. The transepts were in building in 1227; the new choir was commenced at the back of the existing one, by the foundation of the first or western chapel to the north of the sanctuary in 1230, followed by the second chapel to the south in 1239, the central chapel in 1241, and the second chapel on the north in 1243; the date of the first chapel on the south is not recorded, but was probably be- tween 1230 and 1239. Thus the radiating chapels which circumscribed the new apse of Cambray were built between 1230 and 1243. Finally, the completion of the choir itself is recorded by the fact that at Easter, 1251, the clergy took formal possession of it. Now the note attached to the plan of Cambray in the manuscript shews that the buildings were commenced, but so little advanced as to make their completion somewhat problematical, — "The chapels will be like these if ' they be ever finished" and not only were the chapels unfinished, but also the flying buttresses essential to the choir, for the form of which our author refers to those of Rheims. Hence the plan in the manuscript must have been drawn during the suspension of works between 1243 and 1251. The known dates of the works at Rheims are in perfect accordance with this conclusion. The east end, commenced in 1211 by Robert de Couci, was finished as far as the transepts when he died in 1241. The apse, or " chevet," with its circuit of chapels, was certainly finished in 1215'. As for the nave, of which Wilars has also given drawings, it was built between 1241 and 1257 ; and as the drawing only embraces a single bay, it follows that if only one compartment were k Vide plate G7. this is an oversight, for that author discredits the 1 M. Quicherat notes that the choir was conse- tradition which is the sole foundation for the date in crated October 18, 1215, according to Harlot ; but question. — (W.) 6 ESSAY ON WILARS DE HONECORT completed before 1251, our chronological inference, that the manuscript was written between 1243 and 1251, will hold goorl. But it may be possible to fix the date of the manuscript within narrower limits, and to ascertain with precision that of the principal fact in the biography of Wilars de Honecort. The sketch (in plate 19) which, as he tells us, he made at Rheims when he was on the start for Hungary, represents a window of the side aisles of the nave, and is therefore, as well as his journey, posterior to 1241. Hungarian history shews that in 1242 the Tartars invaded the Danubian pro- vinces, and drove out nearly the entire Hungarian population, who returned the next year and expelled their conquerors, but found all their towns in ruins. The city of Strigonium (now called Gran), the capital and ornament of the empire, was rased to the ground. Bela, the reigning king of the Hungarians at that epoch, applied all his resources to the restoration of that great city. He sought to restore it to its splendour, its animation, and its European character, for at the time of the invasion it was almost exclusively peopled by French and Italians. Amongst other works he constructed for the Friars Minors a sumptuous church to the Virgin, in which he had chosen his burial-place. In ignorance of the exact date of the construction of this great church, we cannot ascertain whether Wilars de Honecort was concerned with it, but it is impossible not to see a connection between his journey and the works undertaken to repair the ravages of the in- vading Tartars. We may suppose that he started for Hungary in 1244, after the complete deliverance of the country. By his oavii account he remained there a considerable time; "1 was there for many a day" (maints jours m ), an expression that may comprehend two or three years. Supposing him to have returned to France towards 1247, he may have written the memoranda in his sketch-book before the recommencement of the works at Cambray, which were completed in 1251. Probably he was then in declining years, and preparing to retire from the world, and therefore set about completing his manuscript for the use of posterity. Our author's journey to Hungary suggests some other instructive remarks. First, King Bela was the brother of Elizabeth of Hungary, a princess so devoted to our Lady of Cambray that her offerings served to pay for the work of re- construction of the transepts of that church, commenced in 1227 under the pre- sumed direction of Wilars de Honecort. Secondly, Elizabeth of Hungary died in 1231, was canonized, and became the object of peculiar devotion at Marburg, - PI. 29. AND HIS SKETCH-BOOK. 7 where she was buried. There, under her invocation, was built in 1235 a mag- nificent church ; and this church has semicircular transepts, an uncommon ar- rangement", but which happens to occur also at the Cathedral of Cambray. Thirdly, the south apsidal chapel of the latter cathedral, built in 1239, and, as we have supposed, under the direction of Wilars de Honecort, was dedicated to this very saint, Elizabeth of Hungary. The dates which we have derived in this essay have shewn that Wilars de Honecort belonged to the great school of art of the period of Philip Augustus °. They place him in the midst of that generation of remarkable men by whose labours the Gothic style and system of construction was brought to its highest perfection. This fact invests with an incalculable interest the book we are about to examine, containing, as it does, so much information with respect to the manual processes and methods of construction employed at the time it was composed. Thus far we have literally followed the able and ingenious essay of M. Quicherat, to most of whose conclusions M. Lassus declares his entire adhesion. The latter anxiously sought for information respecting the architecture of Hungary, in hopes of discovering traces of Wilars de Honecort ; for this purpose he entered into a correspondence with Doctor Emeric Henszlman, a native of Cassovia, or Kaschau. This antiquary published the results of his researches on the influence of French architects of the middle ages in Hungary in a Parisian journal p , from which the French editor of the present volume (M. Darcel) has extracted the facts and plans that follow, assisted by the rough notes and letters that M. Lassus had left amongst his papers. It appears that at Gran, the ancient Strigonium, every mediaeval building has been replaced by more modern edifices, so that no researches can be made there, and that amongst the few Hungarian monuments of architecture in which French influence can be detected, the most striking is the church dedicated to St. Elizabeth at Cassovia, or Kaschau. The plan of this church is very similar in the disposition of the chapels ° This is a very common arrangement in Germany, Philip the Second of France, who reigned from and probably travelled from thence to Cambray, which 1179 to 1223. is thus simply an instance of German influence upon p Namely, the Honitmr des Architectes. Paris, the neighbouring districts. — (W.) March, 1S57. 8 ESSAY ON WILARS DE HONECORT to that of St. Stephen of Meaux, and more especially so to that of St. Yved at Braine, near Soissons, as the subjoined plans shew -very clearly. The want of elevations, by which alone the real styles of these churches and their actual construction can be compared, makes it impossible to decide whether or no they belong to the same school of architects. But we will now follow the descriptions in the French edition as far as they go. In the two churches at Cassovia and at Braine there are the same shallow lateral chapels, the same central chapel composed of two severeys terminated by a five-sided apse, or semi-decagon, and the same column placed in the axis of each lateral chapel to receive the middle vault rib. In both plans the opposite columns which belong to the most eastern of the lateral chapels are placed in continuation of the rank of piers that separate the nave and side aisles. But the relative position of the other two chapel columns to the great tower-piers are quite different q . Also the lateral chapels at Braine r are semicircular, but those of Cassovia are polygonal, and the arrangement of the other portions of the plans are altogether dissimilar. In fact, the latter church was built in the fourteenth century, conse- quently long after the journey of Wilars de Honecort. It stands, however, upon the foundations of a church of the thirteenth century, the crypt of which still exists under the first apsidal chapel on the north side. Two documents in the archives of Cassovia mention the existence of a church of St. Elizabeth in 1263 and 1292, which, according to the conjecture of Dr. Henszlman, may have been built by Stephen, her nephew, who resided in that city in 1260, and afterwards came to the throne of Hungary as Stephen V. in 1270. Unfinished or ruined in consequence of the departure of Stephen to the seat of royalty, the church was subsequently carried forwards to completion between 1330 and 1380, by Elizabeth, the queen of Charles Robert, upon the ancient foundations, the plan of which is, as it has been shewn, essentially French ; but the disposition, and probably the number, of the piers was modified. Moreover, in this subsequent completion the German system of three aisles of equal height was adopted. M. Henszlman quotes other Hungarian churches which retain evidences of the French system; for example, at Szekesfchervar, Veszptem, H. Marton, and i The plan of the Hungarian church indicates a peculiar system of vaulting symmetrically disposed about the central tower that cannot be thoroughly understood without elevations or perspective views. -(W.) r Without attempting to assume Wilars to have been the architect of St. Yved at Braine, it may easily be granted that he, as a Picard, being ac- quainted with this church, would naturally imitate it at Cassovia. — (W.) / AND HIS SKETCH-BOOK. 9 Ley den. Lassus, or his editor, concludes that the journey of Wilars de Honecort to Hungary was made after the apse and transepts of Cambray were finished, and that he may have begun St. Elizabeth at Cassovia about 1250, and have assisted in the construction of the church of St. Elizabeth at Marburg and in the restorations at Strigonium j agreeing with M. Quicherat in the supposition that he obtained this employment by having constructed the church at Cambray, the favourite ob- ject of the devotions of St. Elizabeth. c DESCRIPTION OF THE MANUSCRIPT AND ITS CONTENTS' The general nature and dimensions of the manuscript have been described in the previous chapter, but there are many particulars relating to it which require a more minute examination. It is composed of inferior vellum, and, as Mr. Burges well describes it, re- sembles " a large pocket-book, and is bound in pig-skin, now become about the same colour as a very much used saddle. The leaves are fixed to the cover by threads going right through the back, and encircling very stout bands of leather on the outside: these bands are thus real and constructive V The cover wraps over the front edge of the leaves, and they are arranged in six quires. The sheets of vellum of which the book is composed are either folded in the middle, so that each forms two leaves, or merely doubled over at the back edge, so as to form but one leaf and a narrow slip or guard. The whole are stitched into six quires, which contain different numbers of leaves. Thus the first quire has seven leaves, in three sheets and one half-sheet. The second quire has seven leaves, and is made up of two whole sheets and three halves. The third quire has but three leaves, namely, one whole and one half-sheet. The fourth quire has six leaves, or two whole sheets and two halves. The fifth quire has eight leaves, in three whole and two halves, and the last quire is merely a single whole sheet. This is by no means the original extent of the manuscript, which has been re- duced to this condition by the abstraction of more than a third of its leaves. The slips of the half-sheets for the most part shew by the state of their edges that the leaf which originally completed them has been cut out, and some of them retain portions of the writing or drawing which occupied the missing piece. In some of the quires a double slip bears testimony to the former existence of a whole sheet of which both leaves have been removed. It is possible that other sheets or leaves may have been so withdrawn as to leave no trace behind. This kind of " I have substituted this chapter for the short dc Honnecourt," in the French edition. — (W.) notice given in this place, headed "Album de Villard b Builder, Nov. 13, 1858, p. 758. DESCRIPTION OF THE MANUSCRIPT AND ITS CONTENTS. 11 evidence is corroborated by allusions in the legends that accompany the drawings to others that are not now to be found, and also by the ancient pagings of the volume, by which the loss of leaves can be detected. The drawings are, with one exception, made on both sides the leaf; and there are four systems of paging, which have been applied to the book at different periods. The first is an alphabetical set, which is written on both sides of the leaf, but extends only to the letter r. This belongs to the thirteenth century, and must have been made soon after the manuscript was composed. The second system belongs to the fifteenth century, and extends from one end of the volume to the other, but is only applied to the right side of the pages. It begins with the letters of the alphabet, and proceeds with tolerable regularity till it reaches the letters t and u. But the writer, at this point, having discovered that the alphabet would not carry him to the end of the volume, has assumed the letter v to be the numeral v, and has continued the paging with Roman numerals to xxvii. which completes the book. The characters have been written by an unskilful scribe, and are much blotted, the pages having been turned before the ink was dry, so that for the most part each character leaves a trace on the oppo- site page, and some of them are so blurred as to be illegible. The third system is a modern paging in Arabic numerals, written on the right side of the leaf only, and therefore inconvenient for reference ; because, as the drawings are on both sides, the terms recto and verso must be employed to dis- tinguish them. In all the references of the present volume the fourth system has been employed, which is the one used in numbering the plates of the facsimile. These plates are arranged in the same order as the pages of the manuscript, and consequently their numbers correspond to a complete set of paging numbers applied in the usual manner to each side of the leaf c . The traces of abstracted leaves that have been just described shew that the manuscript was originally stitched together in the following order. The first quire had four sheets, the second had six, the third had three, the fourth had four, the fifth had five, and the sixth had five, making a total of twenty-seven sheets, or fifty-four leaves, supposing that no leaves have been ab- stracted excepting those of which traces or slips remain, or of which the loss can be proved by the interrupted paging. Now for the missing leaves, it appears that the first quire has lost the fourth c The only exception is, that the third leaf having even to the traces occasioned by the setting off of its recto blank, no number applies to it. In the the opposite marks. The modern Arabic paging plates, the characters which form the ancient pagings of the manuscript is not inserted, are carefully represented with their blurs and blots, c 2 12 DESCRIPTION OF THE MANUSCRIPT leaf ; the second quire, the second, fifth, sixth, seventh and tenth leaves ; the third quire, the first, fifth and sixth leaves ; the fourth quire, the first and second leaves ; the fifth quire, the second and eighth leaves ; and finally, the sixth has lost the whole of the four inner sheets. The total loss of leaves is therefore twenty-one, and the number of leaves that remain, as already mentioned, is thirty- three. Some of the leaves may have been cancelled by Wilars himself. At the end of the volume there is written, in fifteenth-century charac- ters, — "En ce livre a quarante et 1 feuillet. J. Mancel." The proprietor of the book was therefore J. Mancel, and it had then forty-one leaves. It was apparently this Mancel who wrote the paging which has been already stated to be in characters of his period ; and we learn by comparing his memorandum with the account already given of the number of leaves, that the book had lost thirteen leaves when he had it, and that eight have disappeared since. Mancel's paging shews that one of the latter was the missing second of the fifth quire, which he had numbered xi., and that the other seven were all taken from the sixth quire, of which the only two remaining leaves are numbered xix. and xxvij. Supposing, therefore, that this quire was originally of five complete sheets, one of the leaves must have been removed before the book came into the possession of Mancel. The characters employed by the paging scribe of the fifteenth century after he adopted the Roman numerals are very distinct, but those which he wrote in the previous part of the volume are often ambiguous and obscure ; their peculiarities are pointed out in the descriptions of the plates as they occur. With respect to the manner in which the drawings are executed, Mr. Burges, who has minutely examined the originals with the eye of a practised artist, states that the object was first drawn with some substance resembling the black lead of the present clay ; in all probability a lead or silver point, or even the common black stone used by the masons. The circles, drawn in pencil with compasses, were inked in by hand, and this was the case even with the straight lines. " The ink, by the strangest coincidence, exactly resembles the indelible brown of the present day, and like it where it has been used thickly, becomes a very dark rich brown ; where, on the contrary, employed more sparingly, it presents a light yellowish brown rt ." " Vide Mr. Burges, in the Builder, Nov. 13, 1858, d'abord traces a la pierre noire?' They were then p. 758. M. Quicherat states that the drawings were passed over with a pen and ink, but the black stone sketched with black-lead, "esquisses a la mine de was sometimes used to shade the depths and folds plomb." M. Lassus, or his editor, that black stone of the draperies.— (French edition, p. 57.) was employed : " Les dessins sont, pour la plupart, AND ITS CONTENTS. 13 The subjects that are chosen for the drawings may be classed under figures, architecture, with masonry, carpentry, and practical geometry, and machines. The first occupies the greatest space ; for of the sixty-three plates that remain, thirty-five are wholly devoted to it, and six shared with the others, making a total of thirty-eight plates, and leaving only twenty-five for the remaining subjects; in other words, figures take up about three-fifths of the whole. Many of them are drawn on a scale that occupies the whole page, and they are either single or in groups, and represent sacred personages, symbolical and moral figures, studies from the antique or from nature and ordinary life, elementary figures for learners, and animals of all kinds. The persons or events to which they refer are very rarely indicated by inscriptions, and must be left to conjecture. Architecture, which occupies a space equivalent to about sixteen plates, gives to us plans of churches and other buildings, mostly drawn from real ones, perspective views, if they may be so called, of the tower of Laon and the chapels of Rheims Cathedral, drawings of windows, of a clock-house, lectern, stall-work, &c, and elevations and details of one compartment of Rheims Cathedral. Three plates are exclusively devoted to masonry and practical geometry, about two to carpentry, and three, with the halves of two others, to machines. At first sight it would seem as if the subjects were mixed up in this volume without method or classification, but this is not altogether true, neither is it possible that the want of order which they exhibit is the effect of a re-binding or a rearrangement of the volume ; for it will be observed that the same subject continues over several pages, and that generally when another subject is taken up, the page, or at least the leaf, at that point is shared between the two. Thus figures occupy the first seven plates, then come two plates of machines and archi- tecture, a leaf with figures on one side and architecture on the other, a leaf with architecture on one side and architecture and figures on the other, two pages with figures, one with figures and machines, two of architecture, one architecture and figures, seven of figures, one of figures and architecture, two of architecture, and so on, as the annexed table shews. If the subjects at the two ends of the same sheet be compared, the same kind of evidence of the original mixture of dissimilar objects will be obtained. Rigid classification plainly never entered into the plan of the artist, but he went on drawing figures until he was tired, and then began drawing architecture, and so on ; or, more probably, at first assigned a few blank pages at the beginning of his book to figures, and the next few to architecture ; and thus, when the former were filled with figures, he was com- pelled to continue his figures on the pages beyond his architecture, and thus the subjects became arranged in alternate groups in a manner which happens to every 14 DESCRIPTION OF THE MANUSCRIPT writer of a commonplace book. That the sketches were made at separate times, and probably from real objects, or copied from paintings or sculptures that he encountered in his travels, is shewn by the fact that several of them are inverted, by the accident of opening the book with the wrong end upwards, which so frequently occurs in sketching. It is also evident that he, as space ran short, sometimes inserted a sketch in a spare corner of a back page that had been nearly filled up already. His intention of observing as much method as the circum- stances admitted of, is shewn by four contiguous pages devoted exclusively and completely to his elements of portraiture, and followed by three others similarly appropriated to the geometry of masonry, as he himself declares. On the whole, I conclude that the volume is a veritable sketch-book, and the drawings inserted in it from time to time, and that it is not a collection made up or rearranged in after-life by its possessor. The inscriptions, on the contrary, from the manner in which they are written between and amongst the drawings, and the dark colour of their ink, shew plainly that they are subsequent additions for the information of posterity, and not contemplated at the time the drawings were made, for no space had been reserved for them. One at least of the drawings was made upon the vellum before the sheets were bound up, for the lances of the cavaliers in plate 15, which is part of the outside sheet of the second quire, are continued across the present fold, and shew their points above the heads of the figures in plate 26. This only proves that when the book was bound this sheet of vellum was introduced into it upon which the drawing had already been made. The plates of this volume form a complete facsimile of the original manuscript, and therefore preserve its unclassified arrangement. Nevertheless, it is absolutely necessary that the detailed description or explanation of each plate should ac- company it, for the attempt to describe the subjects of the plates in a methodical series would lead to so much troublesome reference from one plate to the other, as to nullify the advantages of such a systematic mode of proceeding. It is true that in the admirable essay of M. Quicherat, of which the first part has been already presented to our readers, this method is employed, and with great success. But that essay is illustrated only by a few copies of the leading draw- ings selected from the collection, and printed in wood on the pages where their descriptions occur. Two tables are subjoined, which will enable the arrangement of the volume and the nature of its contents to be understood, and the various specimens of each subject compared together at pleasure. The first table gives a comparative view of the different systems of paging that have AND ITS CONTENTS. 15 been applied to the manuscript, the arrangement of the sheets in quires, and the distribution of the subjects. It is disposed in the form of four parallel columns. The first contains the modern paging of the MS., which is confined to the right side of its leaves ; the asterisks denote the places of the leaves that have been abstracted, and the brackets which connect the figures and asterisks in pairs shew that each pair belong to the same sheet, and thus explain the arrangement of the quires c ; the second column contains the ancient paging, the third the numbers applied to the plates of this edition, which correspond to a system of paging on both sides of each leaf ; the last column shews the distribution of the subjects under the general heads of Figures (F), Architecture (A), Machines, Carpentry, &c. The second table is a classified list of all the several drawings in the manu- script arranged under these general heads, with references to the plates that respectively contain them. e In the French edition some discrepancies occur as they occur in the latter descriptions. My own between the account of the distribution of the quires table is the result of my own notes, made from an at page 56, and the notes whicli are appended to the examination of the manuscript in 1851, and collated descriptions of the plates ; I have pointed these out with the descriptions in the French edition. — (W.) 16 DESCRIPTION OF THE MANUSCRIPT Modern Paging and Original Ar- rangement of Quires. 1= Ancient Paging-. 13th r 15th centy. I centy —4 5 6 7 -10 . * -11 -12 . * -13 -14 # — 15 .—16 '—17 * * Paging of Plates. k. n. 10 11 12 13 Subjects. 14 15 16 17 18 19 20 21 22 23 24 25 26 27 F. F. F. and Machines. A. A. A. and F. 28 29 30 31 32 33 > Figures. Machines. Architecture. F. A. A. A. and F. F. and A. A. A. A. and F. F. F. and A. Carpentry. Modern Paging and Original Ar- rangement of Quires. & O an — JS ,— 1-91 —20 — 21 — 22 -23 .2 I 25 26 —27 —28 29 -32 r Ancient Paging-. s. t. ? v.? vii. (xi.) xii. xiv. XVUl. (XX.) (xxi.) (xxii.) (xxiii.) (xxiv.) (xxv.) (xxvi.) xxvii. Paging of Plates. Subjects. 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 ! Art of J Portraiture. ) Masonry and Geometry. F. F. Machines. Mach". & Carp'. F. F. and A. F. F. A. F. Machines. A. A. A. Recipes. CLASSIFIED LIST OP THE SUBJECTS OF THE DRAWINGS. SACRED OR EMBLEMATICAL FIGURES. PL. Figures. — Christ teaching, scaled on a throne . . 31 Christ seated, in the attitude of benediction . . . .20 Christ standing, also in the act of benediction . . . .53 Christ prostrate (in Gethsemane, or on the road to Calvary?) . . .32 Sleeping figure (an apostle in Gethsemane ?) . . . .45 The Flagellation and the return to Pilate . . . .55 A crucifix . , . . . . , . . .4 A monumental cross, or rood, with the Virgin and St. John . . .14 The Descent from the Cross . . . . . .25 Virgin and Child . . . . . . 34 & 10 The Twelve Apostles, with three other figures . . . .2 Two standing personages (apostles ?) . . . .54 Large head (a study for an apostle ?) . . . .34 One of the damsels before Solomon . . . . .22 A prophet? standing . . . . . .49 Small seated figure . . . . . .30 Group of six (the Magi before Herod? or Paul before Agrippa?) . . 24 Symbolical figure of the Church . , . . .7 Martyrdom of SS. Cosmos and Damian . . . . .52 Small seated figure of a bishop . . . . .1 Pride and Humility . . . . . .5 The wheel of Fortune . . . . . .41 . SECULAR FIGURES. Two seated figures — Young man and lady (copied by Willemin) . . 26 Warrior mounting his horse (copied by Willemin) . . .45 Warrior standing . . . . . .3 Two cavaliers . . . . . . .15 Two foot-soldiers . . . . . . .49 Large, seated personage . . . . . .48 Two gamblers . : . . . • • . v 16 Two wrestlers . . . . . • .27 Two other groups of wrestlers . . . . .36 Two male figures, studied from nature = . . . . .42 , Three male figures : one seated, and drawing a sword ; another standing, mitred ; and a third standing, crowned = . . ' . . .23 D 18 CLASSIFIED LIST OE THE DRAWINGS. Female with parroquet and dog Viol-player with dancing-dog A thresher . A mower and two trumpeters Knight on horseback Head of a mendicant Various heads and faces Tomb of a Saracen (Pagan), from the antique? Antique figure of Mercury ? . Male figure standing at an altar (from the antique ?) The Art of Portraiture occupies plates 34, 35, 36, and 37. ANIMALS. "Winged Lion of St. Mark, and Bull of St. Luke . The taming of the lion Combats with lions Lion, viewed in front . • Lions, 36; horses (with riders), 5, 15, 36, 45; horse's head, 35; stag, 34; sheep, 35; greyhound, 35; dogs, 13, 46, 50; cat, 13; boar, 16; boar's head, 36; hare, 16; bear, 6; porcupine, 47 ; dragon, 20; demon, 1. Eagle, 35 ; swan, 6 ; owl, 1 ; magpie, 1 ; parroquets, 50 ; pelican, 1 ; ostriches, 35 ; fishes, 37; crawfish, 13; grasshopper, dragon-fly, and horse-fly, 13; snail, 3. Flowers, 36 ; foliage and foliage ornament, 9 ; two foliage heads, 9 ; two others, 42 ; initial letter S, 11. ARCHITECTURE AND CONSTRUCTION. PLANS. PL. H A square-ended church for the Cistercians . . . 27 The east end or presbitery of a church projected by De Honecort and Peter de Corbie 28 Ditto of Vaucelles . ..... 32 Ditto of Cambray ...... 27 Ditto of a church at Meaux . . . . . 28 A square chamber, vaulted, with a central pillar . . 40 I An apse with twelve windows ..... 38 The tower of Laon Cathedral . . . . .17 DRAWINGS. View of the tower of Laon Cathedral . . . .18 Interior of one of the apsidal chapels of Rheims Cathedral . . 59 Exterior of the same . . . 60 50 50 34 36 36 17 35, 37 10 57 21 I CLASSIFIED LIST OE THE DRAWINGS. 19 PL. Elevations, exterior and interior, of one severey of the nave of Rheims Cathedral Gl Details and sections of mouldings for the above 62 Flying-buttresses of llheims Cathedral . . 63 A window of the nave of the same . ■ 19 Rose-window at Chartres 29 Rose-window at Lausanne . . . 30 A gate-house 35 A clock-house . 11 A lectern . 12 T» Till Poupee and stall partition Do Large poupee for a stall . . « 56 Small tabernacle . 17 Symbolical city . 6 Tile pavements . 37, 29 Labyrinth of Chartres Cathedral . 13 PRACTICAL GEOMETRY. To measure the diameter of a nook-shaft . . . 39 21 To find the diameter of a column of which only half is visible . . 38 1 To find the centre of a circle from three points of its circumference . 33 2 To adjust a square . . . . . 39 22 To draw a spiral . . . . . 39 25 To cut the mold of a great arch in a small space . . . 38 3 To lay out a square cloister . . . . 38 11 To set the four corner-stones of a cloister . . . 38 14 To trace the plan of a pentagonal tower . . . 40 35 To divide a square slab into two equal squares . . 38 15 To make two vessels, the one twice as capacious as the other . . 38 17 To measure the breadth of a stream without crossing it .38 12 To measure the width of a distant opening . . 38 13 To measure the height of a tower . . . 39 32 The proportions of a spire ..... 39 28, 29 To cause a pear to fall upon an egg , . . .40 34 To set up two pillars at the same height . . . 39 31 To bring two stones to the same point . . . 38 7 To describe three kinds of arches with one opening of the compasses . 40 40 To shape the key-stone of an arch of the third point, and also of the fifth point 39 23, 24 MASONRY An oblique voussoir A window in circular masonry . An arch with the centering outwards The voussoirs of hanging arches 38 38 38 39 D 2 9 8 1 30 20 CLASSIFIED LIST OF THE DRAWINGS. A cusped voussoir . To cut the springing-stones of a vault or arch To form the springing-stones of a vault To cut the voussoirs of vaulting surfaces To find the centre of a given voussoir To trace the joints of voussoirs To cut the joints of voussoirs by scales To draw the extrados of a given voussoir The bond of a square pier The bond of a pier at Rheims . PL. 40 38 40 39 40 38 39 40 39 29, 62 39 6 38 20 36 18 27 37 26 CARPENTRY. Roof for a vaulted chapel . . . . . 33 Boarded waggon-roof . . . .33 Roof for a side aisle . . . . . 33 Floor with short timbers ..... 44 Wooden bridge with short timbers . . . . 38 10 Framing to restore a falling house . . . . 44 MACHINES. Saw-mill . ' . . . • . . 43 Saw to cut off pile-heads . . . . . 44 Screw to raise weights .... 43 Trebuchet . . . . . . 58 Heliotropic angel ...... 43 Eagle-desk ...... 43 Hand-warmer with gimbals ..... 16 Tantalus cup . . •. . . . 16 o u Perpetual motion . . . . . . 8 3 -> Method of cutting a screw . . . . . 38 16 Framing of a wheel ■ . . . . .44 Hammer wheel . . , ■ . . . .37 Crossbow, with sight . • . = . . . 43 Esconsa, or dark lantern . . . . .33 RECEIPTS. Hydraulic cement ...... 41 Depilatory paste ...... 41 A potion to cure wounds ..... 64 How to preserve flowers ..... 64 ft EXPLANATION OP THE PLATES. PLATE I. MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER a. Tins page is marked with the red stamp which in the days of the Revolution authenticated the seizure of the volume, as national property, from the library of the Abbey of Saint Germain des Pres. The stamp, in red ink, bears in its centre the two letters R F, as initials of the words Republique Frangaise, and is sur- rounded by the inscription bibliotheque nationals. The words "Sancti Germani a Pratis, N. 1104," in the middle of the page, shews where this volume came from and its distinctive number, whilst the me- morandum "S. G. Lat. 1104," placed at the top of the page, shews that it is in the present day ranked in the St. Germain Latin collection, in which it bears that number. The page is occupied by a figure of a bishop, and some drawings of animals, which were probably designed for a " Bestiary," as books of natural history were termed in the middle ages. In the first place, we find a pelican perched on its nest with wings outspread, in the act of tearing its breast to feed its young. In the Bestiaries the pelicans are said to be extremely fond of their young, but that when the latter begin to grow up, they rebel, and attack the old ones, who become enraged and kill them. On the third day after this, the mother (or father) returns, and tearing open its breast with its bill, the blood is shed upon them and restores them to life. For they receive the blood and drink it a „ The pelican is thus assumed as an emblem of the Saviour, and therefore represented during the middle ages in medallions enamelled or engraven at the extremities of crosses. The bishop, in pontifical costume, is sitting, giving a blessing in the Latin manner with his right hand, and holding his episcopal crosier in his left : he wears a low mitre, is clothed in an alb with tight sleeves, a tunic and chasuble with the border of the amice shewing round the throat, leaving the upper part of the alb exposed; the maniple has been omitted. This figure, which, from the numerous a Vide an admirable Essay on the Bestiaries, in the Melanges d'Arclteologie of Cahier and Martin, vol. ii. p. 13G. 22 EXPLANATION OF THE PLATES. folds, seems to have been a specimen of German art, was probably a sketch from a painted window rather than from a piece of sculpture. An owl, the nydicorax, is the reverse of Christ, and represents the Devil, the friend of darkness. In Bestiaries the owl is generally represented surrounded by day-birds, who are disturbed by his presence, and in a woodcut published by the Rev. P. Cahier, in his Essay on the Bestiaries b , already quoted, we find among these day-birds that threaten the impassive owl with their beaks, a magpie, drawn very much like the one in our plate. It might be inferred that this magpie, holding something cruciform in its beak, and leaning towards a monster ap- parently watching it, is a distant recollection of the fable of the fox and the crow, but this cross is a subsequent addition, as well as the tablet (somewhat re- sembling a tombstone) on which the bird now appears perched, and against which the monster seems to lean. There appears originally to have been no con- nexion between these figures. The monster is perhaps the " Centicore," an ima- ginary animal which is thus described in M. Cahier's Essay, already quoted, (t. iii. p. 223) : A beast from the deserts of India, in colour black, and very fierce, with two horns on his head, perfectly straight, and as sharp as a sword. When he fights with another beast, he lays one horn along his back, and only uses the other. His snout is round, he has the thighs of a lion, the feet and body of a horse, and the tail of an elephant. By using only one horn at a time he is said to symbolize mankind, who never put forth their whole force in combating the devil. As to the inscription on the tablet, the ink is so faded that it is impossible to decipher it : but enough can be traced to shew that it was in French, and must have been either a draught or a copy of an epitaph, but without any connexion with Wilars de Honecort. The date m.cccc.iiii xx .iii. juiex (July, 1483), that accompanies the lozenge-shaped escutcheon, is, moreover, in the writing that belongs to the close of the fifteenth century. b See also, for the explanation of the symbolism of animals, the Bestiaire de Guillaume, clerc de Nor- mamlie, published by M. Ch. Hippeau. I EXPLANATION OF THE PLATES. 23 PLATE II. MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER b. M ' Ci poics vos trover les agios des xij. apostles en seant. " ' Wuars do Ilonecort vous salve, et si proie a tos ceus qui de ces engiens ouverront con trovera en cest livre quil proient por s'arme et quil lor soviengne de iuij Car en cest livre puet on trover grand consel de le grant force de maconerie et des engiens de carpenterie, et si trovercz le force de le portraiture, les trais ensi come li ars de iometrie le command et ensaigne.' " "lei vous pouvcz trouver la figure des douze apotres assis. " Villard de Ilonnecourt vous salue, et prie tous ceux qui travaillent aux divers genres d'ouvrages contenus en ce livre de prier pour son ame, et de se souvenir de lui ; car dans ce livre on peut trouver grand secours pour s'instruire sur les principes de la maconnerie et des constructions en charpente. Vous y trouverez aussi la methode de la portraiture et du trait, ainsi que la geomctrie le commande et l'enseigne a ." " You find here the images of the twelve apostles sealed. " Wilars de Ilonecort salutes you, and implores all who labour at the different kinds of work contained in this Look to pray for his soul, and hold him in remembrance. For in this book may be found good help to the knowledge of the great powers of masonry, and of devices in carpentry. It also shews the power of the art of delineation, the outlines being regulated and taught in accordance with geometry." This inscription, the author's preface, from which we learn his name, his birth- place, and the nature and object of his work, has been fully explained and dis- cussed in the introductory chapter 6 . It only remains to consider the images, which were evidently drawn before the inscription was written, from the manner in which the lower lines interfere with the heads of the figures. It will be remarked that there are, besides the twelve, three others not alluded to by De Honecort. The twelve apostles are seated, holding labels, and seeming to converse in groups of two or more. They remind us of those of the north-east doorway of the Cathedral of Bamberg. There the apostles are seated six and six, above the capitals of the columns in the jambs of the doorway, and hold one and e M. Merimee is of opinion that the proper trans- which are given hy the French editors, unless other- lation of agie is the costume or attire of the apo- wise mentioned ; but I have occasionally endeavoured sties. (Building News, vol. iv. p. 1280.) — (W.) to keep closer to the original in the English trans* d The interpretations of the Picard inscriptions lation of these inscriptions. — (W.) into modem French in this English edition are those c See p. 2, above. 24 EXPLANATION OF THE PLATES. the same label, on which the " Credo" composed by them was intended to be inscribed. On the lintel of the west door of Chartres the apostles are also seated, conversing, two and two, and mostly holding books. The three other figures in the first row are dressed in the civil costume of the thirteenth century, while the apostles wear a robe and mantle draped in the antique fashion. The man wears a cowl over his robe ; the woman, holding a book in her hand, is dressed in a full robe without a girdle, and with loose sleeves, and has a veil over her head and neck. Behind her is a dancing-girl, perhaps the daughter of Herodias, as in the tympanum of the north-west door of the Cathedral of Rouen ; she is dancing on her hands. I EXPLANATION OF THE PLATES. 25 PLATE III. RECTO OF THE SECOND LEAF, MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER C, IN THE FIFTEENTH WITH THE LETTER b. A snail coming out of his shell, his head armed with four horns. A beardless warrior clothed in the mailed and hooded hauberk, with short sleeves ; it reaches to just above the knees, without it being possible to see whether it terminates like breeches or like a kilt. He wears an iron hat over the mailed hood, and a surcoat drawn in at the waist, open in front below the girdle, and with very short sleeves covering the hauberk, and which is put on over a tunic, the tight sleeves of which alone are visible. Mailed greaves are laced on the leg below the knee, which is left bare, and are continued downwards so as to cover the foot. The knee appears to be entirely uncovered, perhaps because it was pro- tected when on horseback by the saddle-bows, which sometimes projected con- siderably. This warrior carries his shield f on his left arm, from which a club is also suspended by a looped strap, and he holds the staff of his lance in his hand ; he points with two fingers of his right hand to his forehead. Is this Goliath ? At all events, it is not " de Honnecort, such as he appeared in Hungary g ," despite the inscription traced near him, for this inscription is an addition of the fifteenth century. ' This shield, making allowance for the perspective distortion, appears to be of the heater form, with the upper angles rounded. * " de Honnecort cil qui fut en Hongrie." K PL.IV EXPLANATION OF THE PLATES. 27 PLATE IV. VERSO OF THE SECOND LEAP, MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER d. Christ on the cross nailed with three nails, sinking from exhaustion so as to throw all the weight of the body upon the arms. A cruciferous disk is fixed a little below the intersection of the arms of the cross in the place where the head should be. The letters I. N. R. I. are inscribed on the scroll nailed to the head of the cross. This figure is excessively contorted, rather in the manner of the fourteenth century than of the age in which Wilars lived. It appears to us to have been rather a study for a painting, — perhaps an attempt to escape from the conventional forms of the age, — than the sketch of a real piece of sculp- ture. A death's head has been rudely scrawled by a later hand on the ground where the cross is planted, but is omitted in the engraving. e 2 I I EXPLANATION OF THE PLATES. 29 PLATE V. VERSO OF THE THIRD LEAF, (IT HAS NO DRAWING ON THE RECTO,) MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER /. " ' Orgieus esi cume il tribuche — Humilite.' " " L'Orgeuil trtsbuchant— l'Humilite." " Pride, and how he got a fall — Humility" The contrast of the Virtues and Vices, which has given to Prudentius the sub- ject of the Psychomachia, a work that we find ornamented with most inter- esting but very ugly miniatures in the manuscripts of the ninth century, has also supplied the sculptors of the middle ages with a fertile theme, which they have developed in the doorways of the Cathedrals of Paris, of Rheims, of Chartres, and of Amiens. Thus in the sur-base of the central doorway of Notre Dame de Paris, the series of Virtues with their respective contrasting Vices is represented. The Virtues are female figures in half relief, seated, and bearing on a circular es- cutcheon their characteristic attributes ; the Vices are shewn in circular bas-reliefs, carved in the stone beneath each Virtue. Here, as at Notre Dame, Humility is seated in an attitude of perfect calm, and holds a circular escutcheon, on which is represented a bird with outspread wings, assuredly meant for a dove. Pride here, also as at Notre Dame, is a cavalier shamefully unseated, his horse having fallen on his knees. The same contrast betwixt the repose of the lines in the figure of the Virtue, and the movement of those in the group of the Vice, exists in both these examples, namely, at the cathedral doorway and in the manuscript, so that it is very difficult not to believe that the drawing must be a memorial sketch of the sculptures at Notre Dame. We must, however, ob- serve, that the indication of the dappled marking of the horse leads to the sup- position that these sketches of Wilars de Honecort, with their decidedly marked outlines, are sketches or studies for a painting on glass or on vellum h . In the plate, Humility is seated, clothed in a loose upper robe with large sleeves, and reaching nearly to the ankle. It is confined at the waist by a folded girdle ; a long tunic which extends below the feet, and has tight sleeves ter- h The drawing may have been made from a painted horse and other details would have been inserted statue, or bas-relief, in which case the marking of the exactly as in a picture. — (W.) 30 EXPLANATION OF THE PLATES. minating at the wrist, is seen beneath it ; a veil covers her head and shoulders. Her left hand grasps the upper edge of the disk bearing a dove, which rests on her knee. Raising the right hand, she is looking at Pride, who is stumbling in the mud. The disordered state of the cavalier's dress permits us to examine some parts of the civil costume of the thirteenth century, which are seldom ex- hibited. Beneath his long mantle, which floats in the air, this cavalier wears a tunic reaching below the knee, with tight sleeves, girt about the loins with a belt, and opened behind, as it probably is in front. This opening shews the breeches and the " chausses," or long stockings which cover the whole leg and foot, and rise above the knee over the breeches ; a pointed spur is attached to each heel. The head of Pride is uncovered, and his hair is bound by a simple twist. We are not to suppose that this costume was peculiar to horsemen, for poems and minia- tures shew us similar hose, head-dresses, and mantles in ordinary use, as, for example, in the seated figure of plate xxiii. ; but the representation we find here is valuable because it furnishes information very rarely to be met with concerning the under costume. The only peculiarity worth mentioning with respect to the horse is the strongly marked nails of the shoes : the bridle with its curb, the square saddle called a la Fran^alse, and the stirrups, are of the forms usually seen. EXPLANATION OF THE PLATES. 31 PLATE VI. KECTO OF THE FOURTH LEAF, MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER i, AND IN THE FIFTEENTH WITH THE LETTER d ; THE OMISSION OF THE TWO LETTERS g AND h SHEWS THAT ONE LEAF HAD BEEN LOST SINCE THE EARLY PAGING WAS MADE. A bear. — This animal, the outline of which Wilars sketched with black stone before drawing it with a pen, is not mentioned in any of the Bestiaries \ and the drawing must be a study from nature, like that of the lion which we shall see further on, as well as the graceful swan that is placed beneath it. Were these two animals destined for coats-of-arms, or are they sketches made with no other intent than to recal a beautiful or interesting object? The swan that sings before it dies, being the image of " the soul in joy or tribulation," is always to be found in the Bestiaries. Beneath the swan is the symbolical representation of a town, indicated by an embattled enclosure with edifices of antique form rising above it, such as are sculptured above the canopies of statues to prefigure the celestial Jerusalem. 1 In the symbolical systems of the middle ages to know how to represent him. Vide Melanges the bear appears as the emblem of luxury, violence,

EXPLANATION OF THE PLATES. 35 PLATE VIII. ItECTO OF THE FIFTH LEAF, MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER /, AND IN THE FIFTEENTH WITH THE LETTER e. [On THE LEFT HAND OF THE FORMER LETTER ARE SEEN THE BLOTS PROCEEDING FROM THE SIG- NATURE OF THE LAST PLATE. — (W.) ] " ' Maint ior se sunt maistre dispute de faire torner une ruee par li seule. Ves ent ci con en puet faire par mailles non pers ou par vif argent.' " "Maint jour, se sont maitres disputes pour faire tourner une roue par elle seule. Voici comment on peut le faire par maillets non pairs ou par vif-argent." " Many a time have skilful workmen tried to contrive a wkeel that shall turn of itself: here is a way to make such a one, by means of an uneven number of mallets, or by quicksilver." Wilars de Llonecort presents to us a device for a perpetual motion ; it is not clear whether he intends to claim the contrivance of it, or whether he had met with it in the course of his travels. It differs very little from a well-known con- trivance for this purpose which has been so often published, and its fallacy so fully explained in popular books k , that it is unnecessary to dwell at length upon the mechanical principles which it involves. It is extremely curious in this place, because it shews the great antiquity of the problem, the solution of which has wasted the time, the brains, and the means of many an unhappy artisan or philosopher. In the drawing we have now before us, the two upright posts, which are framed together and skilfully braced so as to insure their steadiness, support between them a long horizontal axle, to the centre of which is fixed a wheel with four spokes. The absence of perspective in this drawing makes the wheel appear as if it were parallel to the frame, instead of being, as it is, at right angles to it. Seven mallets, or arms, each loaded with a heavy weight at the end, are jointed at equal distances to the circumference of the wheel, so that those which happen to have their joints below the diameter of the wheel will hang freely down, but k For example, in Ozanam's or Hutton's mathemati- press, as it is to be found in books of sucli easy ac- cal recreations. M. Lassus has supplied an elaborate cess as those I have referred to, and would scarcely description, with demonstrations of the fallacy of this be intelligible to persons unacquainted with mathe- class of contrivances, which I have ventured to sup- matics. — (W.) F 2 36 EXPLANATION OF THE PLATES. if the wheel be turned round by hand or otherwise, the weights of those which are on the ascending side will in succession rest on its circumference, and will in that position be carried over the highest part of the wheel, and downwards on the descending side, until the arms that bear them are brought into a vertical position and a little beyond it, and then the weight will fall suddenly over and rest on the opposite position on the circumference of the wheel, until its further descent enables it to dangle freely as before. The effect of this mechanism upon the position of the weights is not truly represented, for the upper mallet has fallen over too soon. In the modern form of this contrivance a pin, or stop, is introduced, by which the mallet when it falls over is compelled to rest, so that its arm shall point to the centre of the wheel, and thus the descending weight be held at a greater distance from the centre than when ascending. It is ex- tremely probable that this difference is a mere error of the artist, for the drawing has the appearance of having been made from a model of the wheel at rest ; a condition in which, of course, it would always be found, unless moved by some external force. The inventor seems to have thought that the action above described would always place four weights on the descending side, and leave but three on the ascending side, each weight as it rises to the top being intended to leap suddenly over to the descending side, in the manner just ex- plained : or perhaps, as M. Lassus suggests, the contriver imagined that the blows given to the wheel in succession by the falling mallets would help it for- ward. It is surprising, that although the slightest model would shew the failure of devices of this class to persons incapable of mathematical reasoning, yet such machines have been seriously proposed in books, and are continually re-contrived by ingenious workmen. The allusion to quicksilver in the manuscript shews that Wilars was acquainted with the well-known contrivance described in the books already referred to, in which portions of that metal inclosed in channels are used instead of the falling weights. M. Lassus ingeniously supposes that the first idea of this machine may have been suggested by the sight of wheels like this, with swinging hammers, which in certain churches were used instead of the bells on Good Friday \ and which keep up their motion, according to the velocity acquired, some time after the moving power has ceased its action. — (W.) 1 A Lenoir, Architecture Monastique, t. i. p. 157. I PL. IX EXPLANATION OF THE PLATES. 37 PLATE IX. RECTO OF THE FIFTH LEAF, MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER m. " Tutes de feuilles," or foliage heads, as Wilars de Honecort calls them below, in reference to other specimens of this style of ornament, (see plates xli. and xlii.), were much in use in the thirteenth century, in which they generally occupy the centre of small rosettes, or the tympanum space of a gablet above its arch. The Cathedral of Paris displays numerous examples of them. A foliage head is simply a human head in full face, the hair, eyebrows, and beard of which are trans- formed into leaves, which completely surround it. The elementary forms of these leaves, although fancifully curled and arranged, are studied from natural types, some of which Wilars has taken care to draw at the bottom of the page ; one of them is a fig-leaf. The origin of these foliage heads may have been pagan. In fact, the vase of the thirteenth century which occupies the centre of the Ecole des beaux-arts, offers, amongst the heads of the gods of antiquity, that of a Silvanus, characterized by the leaves that surround it. An ornament frequently used in goldsmiths' work is drawn below the two heads, and offers the greatest analogy with the crests and friezes of the shrine, or chasse, for the great relics at Aix-la-Chapelle m , and of that of St. Eleuthere, at Tournay. — (L.) m Vide Melanges dCArcheologie, t. i. PLX EXPLANATION OF THE PLATES. 39 PLATE X. RECTO OF THE SIXTH LEAF ; THIS HAS NO PAGING OF THE THIRTEENTH CENTURY, BUT WAS MARKED IN THE FIFTEENTH WITH THE LETTER /. " ' Dc tel maniere fu li sepouture d'un Sarrazin q' io vi une fois . ' " " De telle maniere fut la sepulture d'un Sarrasin que je vis une fois." " This is the representation of the sepulchre of a Saracen that I once saw." This tomb of a Saracen, or rather of a pagan, (for he who M as not a Christian was a Mahometan in the eyes of a contemporary of the Crusades,) is ap- parently sketched from memory, and recals by its disposition the diptychs of the Lower Empire. But the inscription leaves no room to doubt that it was a real sepulchral monument which Wilars de Honecort had in mind when he made this drawing. That he himself impressed it with its very manifest mediajval character is easily intelligible, for the faithful rendering of style in drawing is a quality entirely modern. Before Joseph Strutt published in 1789 his "Antiquities of England," with engravings in which the Archaic character was as strictly pre- served as was possible, no antiquarian had ever thought of attempting more than an approximate representation of the form of the monuments he was studying. But the drawings always possessed the character which prevailed at the time when the artist lived ; and we who look upon ourselves as being so scrupulous in this respect, may, perhaps, hereafter be accused of the same fault in a lesser degree. Our architect of the thirteenth century was not more to blame in giving so mediajval a character to a monument of antiquity, than Montfaucon, Gori, and so many others were in presenting to the public representations of Greek, Egyptian, Byzantine, Roman, or Erankish figures, with the air and attitudes of the time of Louis XIV. We are of opinion that in this picture, traced from a somewhat faded remem- brance, Wilars de Honecort has introduced unwittingly the forms of some dip- tychs which he, whose active mind examined everything, must assuredly have inspected. We may suppose that his intention was to give the likeness of one of those two-storied tombs which were more commonly employed by the Gallo- Romans than by the other nations of the Roman empire. The principal personage is seated, with a flowered sceptre in his hand, just as Philip Augustus himself 40 EXPLANATION OF THE PLATES. might have sat on his throne, and the two half-clad genii, which carry each of them a thyrsus in one hand, support with the other, high above his head, a wreath n , which is, however, composed of trilobed leaves. The bases and capitals of the columns, the vases above, transformed into the likeness of the cruets em- ployed for the service of the mass, the pax filled with holy wafers in the tympanum below, and the finial which crowns the pediment, are all Gothic, whilst the draperies recal the Byzantine or Carlovingian age . Whatever may be thought of these transformations, this drawing is very interesting, for it shews that the mediaeval artists had more respect for the works of antiquity than is generally supposed, and that its architects attempted to imitate them in their constructions as the troubadours did in their poems p . — (L.) " A Roman basso-relievo, published by Montfaucon, retraces nearly the scene of the upper part of the drawing. (V Antiquite expliquee, Supplement, t. iv. pl. 18.) ° The Memoirs of the Royal Academy of Turin give a very interesting example of the singular manner in which archaeological fidelity was under- stood in the sixteenth century. It is the repro- duction of a drawing executed at that time at Joree, and preserved in a manuscript collection of inscrip- tions, from whence it was taken by M. l'Abbe Gazzera, who has inserted it in his memoirs entitled, Del ponderario e delle antiche lapidi Eporediesi. The inscription, very faithfully copied, is as follows : — AVRELI VITALIS CENTVRIONIS LEG. IIII ELA QVI VIXIT But the Aurelius Vitalis, centurion of the fourth legion, to whom it applies, is represented, not as a Roman horseman, but as a knight in complete armour of the sixteenth century, attended by a squire dressed in the same style. The armour and attitude are such as would suit the Chevalier Bayard or the Mare- chal de la Palisse, whilst the inscription itself is very exact, and leaves nothing to desire. This fact, which might furnish a very interesting page for the future history of archaeology, should there ever be found any one to write it, is recorded in the Memorie delta reale Accademia delle scienze di Torino, (t. xiv. p. 26, n°. 12, and pl. 5,) and was pointed out to me by M. Long- perier, a member of the Institute. (A. D.) p Upon this design M. Quicherat remarks that " murs sarrasins" is a mediaeval term always meaning Roman ruins. Wilars de Honecort has given a re- presentation of what he conceived to be a tomb, but has probably mistaken its object. The subject is rather the divine honour paid to an emperor. Above, Romulus and Remus hold up a crown of foliage. The emperor is seated on a "pulvinar," and at his feet is an altar served by two Augustals. (Revue Atcheolor/ique, p. 215). — (W.) I PLXI c4V utnafimg wnttauf- ? tt<#6tef EXPLANATION OP THE PLATES. 41 PLATE XI. VERSO OF THE SIXTH LEAF, MARKED IN THE THIRTEENTH CENTURY WITH THE LETTER 0. " ' C'est li masons don orologe. " ' Ki velt faire le maizon dune ierloge ves ent ci une que io vi une fois. Li premierz estages de desos est quares a .iiij. peignonciaus. Li estages deseure est a vnj peniaus et puis covertic. et puis .iiij. peignonciaus. entre .ij. peignons .1. espasse wit. Li estages tos de seure sest quares a .iiij. peignonciaux. et li combles a .viii costes. Ves aluec le portrait.' " " C'est la maison d'une horloge. " Qui veut faire la maison d'une horloge, en voie ici une que j'ai vue une fois. Le premier etage inferieur est carre a quatre pignons ; l'etage de dessus est a huit panneaux et puis [une] couverture, et puis quatre pignons. Entre deux pignons [il y a] un espace vide. L'etage le plus eleve est carre a quatre pignons, et le comble a huit cotes. Comparez avec le portrait." " This is a clock-house. " Whoever wishes to make a clock-house may view here one that I once saw. The lowest or first story is square, and has four goblets ; the story above it has eight sides, then a roof, and •upon that four goblets, but between every two goblets is a broad space ; the highest story is square, with four goblets and an eight-sided roof. Compare the description with the portrait." This inscription, of which the title is written with a different ink, and ap- parently by a different hand from the rest, describes summarily the general appearance of the edifice or clock-case in question. It must have been of limited dimensions, to judge from the slight inclination of the perspective lines of the sketch, which seems to have been made from the existing object. It was pro- bably an interior clock-turret, like those which are still to be seen within the cathedrals of Rheims, Beauvais, and others q . The different stories are intended for the reception of bells, dials of various kinds, and automaton figures, which are set in motion at periodic times by the mechanism of the clock, to strike the bells, or perform evolutions representing scenes from Scripture, or legends. q At Lyons and Strasbourg, for example. The bably suggested by tbe general form of the previous clock-case at Lyons, in the north transept of the clock-house. The clock at Strasbourg, made in 1574, cathedral, rises in the form of a tower, about thirty- is a complex edifice with three turrets. Those of five feet high, square in the lower stories, and octa- Pheims and Beauvais have no resemblance to the one gon in the upper, gradually diminishing in diameter, represented in our manuscript. They are engraved in after the manner of the clock-house of Wilars de Gailhabaud's Architecture, t. iv., and the two former Honecort. Although its style is that of the sixteenth in Dubois' Histoire de VHorlogerie. — (W.) century, when it was made, its arrangement was pro- 42 EXPLANATION OF THE PLATES. That it was made of wood is shewn by the slender dimensions of the architec- tural members, and especially by the horizontal lintels which occur in the lower and upper story. In style, the semicircular arches indicate the Romanesque period, while the elegance and lightness of the general design place it at the latter end of that period, or even after the introduction of the Pointed arch. For although that form is not employed in the arches of this design, it must be remembered that the art of working in stone was always in advance of the working of metal and wood r . The plan of this clock-case is easy to understand. First we have a basement story, square in plan, and having a shaft at the angles ; each face is surmounted by a horizontal cornice, upon which rests a triangular gable, foliated with semi- circular foils or lobes. The lower cusps are sustained by diminutive shafts ; between each gable is a little turret or pinnacle. Above the basement the construction becomes octagonal, and according to M. Lassus the passage from the square to the octagon is effected by forming the roof out of twelve triangular pieces arranged three by three. The drawing rather appears to indicate that the octagon rises immediately from a flat floor fixed at the level of the cornice of the basement story, and that the gables rise vertically, without any connexion with the roof behind them, the contrast of the octagon and square being disguised by the pinnacles at the angles a . The second, or octagonal story, has its angles marked by a post, or style, and each face has an arcade of two arches, supported by a central shaft and two semi- shafts, or responds. The third story is also octagonal, but of less diameter than the second, and the connexion between the two is simply effected by a sloping roof of eight pieces. The vertical sides or panes of this story are alternately square and gable-shaped, the former shape having a single semicircular arch, and the latter a trefoil arch with semicircular lobes. The upper story is square, but placed in such a position that each angle coincides with the apex of the gablet below. Thus the transition from the octagon to the square is simply effected by four sloping boards, each in the form of a trapezium, which connects the upper edge of the square face of the lower story with the lower edge of the upper story, while its sides rest on those of gablets to right and left. The square story, like the basement story, is capped by a horizontal cornice-moulding, above which are triangular gables with a cross ' In the remainder of this description I have not tiles is distinctly drawn on the right-hand side, rising followed Lassus. — (W.) vertically behind the angle-turret or pinnacle. * The edge of the octagonal basement covered with EXPLANATION OF THE PLATES. 43 upon each apex. There is a shaft at each angle, the capital of which supports the cornice ; and in addition, each face of this story is ornamented with an arcade of six small semicircular arches on very slender shafts ; the whole is crowned by an octagonal spire. The roofs and the spire are carved in representation of tiles or shingles. The front face of the basement story is divided into two by a monial, but the lateral face has none ; similarly, the front face of the upper square story has six arches, and the lateral face appears to have but three. Are we to conclude that these lateral faces were really narrower, and therefore that the plan was a rect- angle, instead of a true square, or that want of perspective skill prevented the artist from introducing the missing members ? In all clock-turrets, however, a square lower story is provided to contain the principal mechanism and to exhibit the great dial. This dial could not have been placed on that face of the lower story which is divided by a monial ; I therefore infer that the left-hand face was square, and intended for the dial, and that it was really of the same width as the other. The monial may, however, be the meeting style of a pair of folding doors, provided to protect the dial. In the upper corner of the page is a rich initial S, formed by a winged dragon with a foliaged tail. This, amongst many other examples in the manuscript, serves to shew that the architects of the middle ages interested themselves in every branch of the fine arts. — (W.) / ?l: xii u&r fausgvt - tettnC -pot fttfttt*' CAsmgttt^ ucfent ate tn Votr auotv tme terete infettrtn ttv WYe tmtxxttxr /bote Auc>t*-t»t>£bct:- c\vt#gnt ouf ^luft zCttac aTb£f»t> ^^VfW tetojcre boit Avcnr EXPLANATION OF THE PLATES. 53 PLATE XVI. RECTO OF THE NINTH LEAF, MARKED IN THE FIFTEENTH CENTURY WITH THE LETTER i. " Vesci une cantepleure con puet faire en j. henap en tel maniere q'ens enmi le henap doit avoir une torete et ens enmi liu de le tourete doit avoir .1. beliot qui tiegne ens el fons del henap. Mais que li behos soit ausi Ions com li henas est parfons. Et ens en le torete doit avoir .in. travecons par sontre le fons del henap. si que li vins del henap puist aler al behot. et par deseur le torete doit avoir .1. oisiel qui doit tenir son biec si bas que, quant li henas iert plains, quil boive. Adont s'en corra li vins par mi le behot et par mi le piet del henap qui est dobles. Et sentendes bien que li oisons doit estre crues." " Voici une chantepleure qu'on peut faire dans une coupe. Pour cela il doit y avoir au milieu de la coupe une petite tour, et la tour doit etre traversee par un tube qui aille au fond de la coupe et soit aussi long que la coupe est profonde. De plus, il doit y avoir dans la tour trois petites traverses allant contre le fond de la coupe, afin que le vin de la coupe puisse entrer dans le tuyau ; et par-dessus la petite tour il doit y avoir un oiseau qui tiendra son bee assez bas pour qu'il sernble boire quand la coupe sera pleine ; alors le vin circulera par le tube et par le pied de la coupe qui est double. Entendez bien que Toiseau doit etre creux." " This is a contrivance that may he made in a drinking -cup. In the midst of the cup is fixed a little tower, and in the middle of the tower is a tube that extends to the bottom of the cup, and the length of the tube is equal to the depth of the cup. There must be also three little cross- pieces to the tower touching the bottom of the cup, so as to allow the wine in the cup to enter the tube. On the top of the tower must be a bird holding his beak so low, that he may seem to drink when the cup is filled. Then the ivine will run through the tube, and through the foot of the cup, which is double. It must be understood that the bird must be made hollow." This explanation is incomplete, and the drawing inexact in several particulars. The contrivance is a toy well known under the name of a Tantalus cup °, and the purpose of the mechanism is that, when liquor is poured into the cup, and, gradually rising in it, approaches the brim, and seems ready to offer the expected draught, it shall suddenly vanish, to the amusement of the bystanders, being in fact con- veyed into the hollow foot of the cup by the central tube, which acts upon the principle of the syphon, but is disguised in its form so as to appear merely as the ornamental pedestal of the bird. c See Hutton's " Recreations," vol. iv. p. 27, or any popular treatise on hydraulics. 54 EXPLANATION OF THE PLATES. The central tube must pass completely through the bottom of the cup, and enter deeply into the hollow foot, which must be capacious enough to contain more than the contents of the vessel, and be pierced to allow the air to escape : the upper end of the tube must not rise so high as the brim, as Honecort states, but must stop a little short of it. The tower, as it is called, is in the form of an inverted thimble of greater diameter than the tube, and must be supported upon short legs below, so as to fix it firmly in a position concentric to the tube, and yet to allow the liquor when poured into the cup to enter freely between the inside of the thimble and the outside of the tube. The two together form what is termed an annular syphon, of which the tube is the long leg and the thimble the short leg. When liquor is poured in, it rises in the cup and in the syphon equally, until its level reaches the upper end of the tube ; it then begins to flow into the tube and fill it, so that the syphonic action begins, and if the pouring be stopped the contents of the cup will instantly disappear by passing into the foot. But if the pouring be gradually continued, the surface of the liquor will remain at the same level, as if the bird was drinking it as fast as supplied. The bird has no connexion with the mechanism, as Honecort seems to think. Sometimes the hollow foot of the cup is omitted, so that the wine may run out at the end of the tube, and the spectators enjoy the additional amusement of seeing the unfortunate subject of their merriment with his clothes drenched d . This is a contrivance of considerable antiquity ; it is to be found in the twelfth problem of the " Pneumatics" of Hero of Alexandria, as " a vessel from which the contents flow when filled to a certain height." The annular syphon is employed exactly as in the example given by Wilars, and the liquor simply runs out at the foot.— (W.) "Et se vos voleis faire .1. escaufaile de mains vos fereis ausi come une pume de keuvre de .11. moities clozeice. Par dedens le pume de keuvre doit avoir .vi. ciercles de keuvre. Cascuns des ciercles a .11. toreillons, et ens enmi liu doit estre une paelete a .11. toreillons. Li toreillon doivent estre cangiet en tel maniere que li paelete al fu demeurt ades droite. Car li uns des toreillons porte lautre ; et se vos le faites adroit si com li letre le vos devise et li portraiture, torner le poes quel part que vos voleis, ia li fus ne sespandera. Cis engiens est bons a vesque. 11 Amongst the plate at Corpus Christi College in corresponding to the description in the text. The Cambridge, there is an ancient mazer-cup of the fif- lower extremity of the tube has been long plugged teenth century mounted in silver, in the centre of up, and the mischievous trickery of the machine for- which is an hexagonal tower of silver surmounted gotten. — (W.) by a swan, and having the central tube ill all respects i EXPLANATION OF THE PLATES. 55 Hardiement puet estre a grant messe, car ia tant com il tiegne cest engieng entre ses mains, froides nes ara, tant com fus puist durer. En cest engieng na plus." " Si vous voulez faire une chaufferette a mains, vous ferez comine une pomme de cuivre de deux moities qui s'emboitent. Par dedans la pomme de cuivre il doit y avoir six cercles de cuivre. Chacun des cercles est muni de deux tourillons, et au milieu il doit y avoir une petite poele a deux tourillons. Les tourillons doivent etre contraries de telle facon que la petite poele a feu reste toujours droite, car chaque cercle porte les tourillons de l'autre. Si vous faites exacternent comme la description et le dessin I'indiquent, vous pouvez tourner dans le sens que vous voudrez, jamais le feu ne se repandra. Cet engin est bon pour un eveque ; il peut hardiment assister a la grand'messe, car tant qu'il le tiendra dans ses mains il n'y aura froid aussi lougtemps que le feu pourra durer. En cet engin il n'y a rien de plus." " If you desire to make a chauferette {calefactorium), or hand-warmer, you must construct a, kind of apple of brass hi two halves which fit together, inside the apple place six brazen circles, let each circle have two pivots, and in the middle place a little brasier with two pivots. The pivots must be placed in contrary directions, so that in all jjosilmis the brasier may remain up- right, for every circle supports the pivots of the next. If you make this contrivance exactly as the description and drawing sheivs it, you may turn it about in any way you please, and the cinders will never fall out. It is excellent for a bishop, for he may boldly assist at high mass, and as long as he holds it in his hands they will be kept warm so long as the fire remains alight. This machine requires no further explanation" In the drawing the outside ring represents the section of the spherical cover, and the inner circle the hemispherical bowl which contains the burning charcoal. Each pair of pivots is placed in a direction at right angles to the next pair in suc- cession ; the hinges and pins which served to connect the two halves of the outer sphere are also shewn in the drawing. The contrivance, under the name of gimbals, is in common use, principally to support marine compasses, chrono- meters, and other philosophical instruments, so that they may maintain the hori- zontal position during the rolling of the ship; modern science, however, has shewn that one circle, intermediate between the rolling frame and the object which is to be kept level, is perfectly sufficient, so that the other five which De Honecort has so liberally provided are entirely superfluous ; and he might have suppressed them, and yet with equal truth have written, as he has done, the in- scription which appears in the centre : — " Cis engiens est fais par tel maniere quel part quil tort ades est li paelete droite." (" This machine is so made, that which- ever way you turn it, the little brasier remains upright.") All the inventories of church treasuries mention the calefactorium, or " escau- faille a, mains," which they sometimes call the apple, or "pomum." Du Cange cites two from York, one of them described as " unum calefactorium argenti deauratum cum nodis curiosis insculptis," i. e., of silver-gilt, with curiously 56 EXPLANATION OF THE PLATES. engraved knobs. A copper-gilt specimen of the thirteenth century, in the col- lection of M. Carraud, shews that these knobs on the surface of the utensil were so sculptured as to furnish the apertures necessary for the supply of air to the fuel within 6 . Wilars de Honecort, evidently before his descriptions were written, had made several sketches on the same page, which are turned the wrong way upwards. One represents a boar tracking a hare, which is couched, and nearly as big as himself. Both of the animals are very exactly drawn. There is also a group of two men playing at dice. The right-hand figure is seated cross-legged on the ground, and has a cloth loosely thrown over his chest and shoulders, but is other- wise naked to the waist. He wears loose breeches, reaching just below the knee, and twisted about his waist as if rolled over some kind of girdle. His legs, from the absence of the muscular lines which are so liberally bestowed on the other unclothed portions of his body, are evidently clad in stockings, and he has pointed shoes on his feet, slit down and laced laterally on the inside of the foot. He is employed either in placing or taking up certain coins or coun- ters from the board, which rests on the ground between the two, and is engaged in an earnest discussion or dispute with his companion, who is seated on the ground opposite to him, with one leg bent under his body and the other ex- tended. This man is clothed like the former, with the exception of the shoulder covering, but as his feet are concealed, and the portion of one leg, which is alone shewn, seems to bear a muscular line, it may be inferred that the stockings are absent, and possibly the shoes. He grasps the dice in his hand as if about to throw them on the board. The latter is conjectured by M. Lassus to be one of the shallow trays which the masons employed in the middle ages to convey their materials to the place where they were working, and he remarks that it re- sembles such a tray which he had seen in a manuscript. Accordingly he con- siders the group to represent two masons amusing themselves in their interval of rest by playing at dice ; adding, that the natural attitudes of these men, and the careful delineation of the muscles of the arms and body, shew that the sketch was made from nature. M. Quicherat, on the other hand, describes the sub- ject to be two slaves playing at dice, the one resembling a Greek, the other a barbarian f . e The French editor mentions that Lassus once A row of little holes along the border of each hemi- had in bis hands a chaufferette about five English sphere were apparently intended to stitch a cloth cover inches in diameter, of plain exterior, with hinges, and on the surface. provided with four interior circles (like those of ' Revue Archeologique, vol. vi. p. 219. our MS.), and a central receptacle for a red-hot iron. PL.KVII V ftUoUf CAw^xxt^: ft fcn&t?. «ic )\Qnc>h<; ; rbvvtv otu** ^trffitotftofcicf 3 i»t^ ,_I EXPLANATION OF THE PLATES. 91 PLATE XXVIII. RECTO OF THE FIFTEENTH LEAF, MARKED IN THE FIFTEENTH CENTURY WITH THE LETTER p. This is now the first leaf of the third quire, but when it was marked, there was already missing the outer sheet of this quire. This page contains two plans of complex east ends, or presbyteries, for large churches ; the upper one is a design, the lower one is taken from the then exist- ing Church of St. Stephen at Meaux, a town about twenty-four miles to the east of Paris. The upper one bears the Latin inscription, in pale ink, — " Istud bresbiteriu' invener't ulardus d' hunecort & petrus de corbeia ir se disputando." At the bottom of the page, in continuation of a description of the lower plan, is written, in black ink, the equivalent French inscription : — " Deseure est une glize a double charole. K vilars 1 de honecort trova & pieres de corbie." " Ci-dessus est une eglise a double collateral, que trouverent Villard de Honnecourt et Pierre de Corbie." The two inscriptions may be rendered in English thus : — " Above is (the presbytery of) a church with a double circumscribing aisle, which Wilars de Honecort and Peter de Corbie contrived together." To understand this plan it must be premised that in the early examples of radiating chapels a portion of the circular side-aisle wall was always left between each chapel, and had a window pierced in it to light the aisle. Of this system the plan at the bottom of the plate is an example, and our own Cathedral of Norwich may be added as an English specimen. In the thirteenth century these spaces were omitted, and the chapels placed close together, as at West- minster Abbey, and consequently a greater number were obtained. Their form continued for some time to be circular, as at Cambray, (Plates 67 and 27), but the polygonal form gradually superseded it during the course of the century. 1 The K and V are run together in such a manner as to make it possible that the latter letter was in- tended for a W.— ( W.) N 2 92 EXPLANATION OF THE PLATES. The arrangement of these aisles and chapels admitted of great variety, into which it is not my intention to enter in this place, but merely to shew that the compo- sition of such combinations must have been an architectural problem of great interest when the drawings in our manuscript were composed, and it was there- fore quite natural that Wilars de Honecort and his friend Peter de Corbie should exercise their ingenuity upon a new solution of it. The novelty apparently con- sists in making the radiating chapels alternately square and round. At Issoire, in Auvergne, the Romanesque church has a square chapel at the east end, placed between and in contact with two semicircular chapels, exactly as in Honecort's plan, but the two remaining radiating chapels to the westward of the former are separated from them by plain wall. The plan of Vaucelles given by Honecort in pi. 32 is of the same nature. The complete series of alternately square and round chapels proposed by our author does not appear in any known example, as M. Lassus remarks, adding that such an arrangement would probably produce an unsatisfactory effect, be- cause the square chapels, from their form and greater projection, would hide the semicircular ones. The nearest approach to this plan of Honecort's is the pres- bytery of Chartres Cathedral, erected at the very beginning of the century. This has a double circumscribing aisle, and a continuous series of seven chapels, all curvilinear in plan, but alternately deep and shallow. — (W.) The singular arrangement of the vault of the compartment of the outer side- aisle which is opposite to each semicircular chapel deserves attention. Each of these chapels has a single middle vault-rib, which rises to the summit of the arch that separates the chapel from the aisle. (This rib springs from a vaulting-shaft, and its thrust outward is sustained by an external buttress.) But the thrust of the upper extremity of the rib upon the keystone of the arch is received by a pair of ribs which diverge from the opposite side of the keystone, and crossing the outer aisle rest respectively upon the piers that separate this compartment of the outer aisle from the inner aisle. Thus the compartment of the side- aisle is covered by three vaulting cells of a triangular plan without diagonal ribs k . But the compartments of the side-aisles which are opposite to the square chapels are vaulted with diagonal ribs in the ordinary manner. The vault of the outer side-aisle presents, in consequence, a series of compart- ments alternately of different and inharmonious forms. Such arrangements be- k In fact, the semicircular chapel and the neigh- bouring compartment of the side-aisle may be con- sidered as covered by a single vault, with five ribs which diverge from the keystone of that pair which form the arch of separation between the chapel and side-aisle. — (W.) EXPLANATION OF THE PLATES. 93 long to the expedients of Romanesque masons, and a similar system is em- ployed under the same circumstances at the Cathedral of Senlis. The vaults No. 1.— Plan of the vault of one compartment of the double aii-lc-, according to Wilars de Houecort. of the inner side-aisle at the east end of Notre Dame de Paris present the same appearance in plan as this outer side-aisle, as the diagram No. 2 shews, but in reality there is a great difference between the two, for the points of intersection of the ribs in the latter are all at the same level as the imposts of the ribs ; but at Senlis and in the design of Wilars de Honecort the points of intersection are at the level of the summits of the arches. — (L.) No. 2.- Plan of the vaults o( one 1 of the double aisles of Notre Dame de Paris- 94 EXPLANATION OF THE PLATES. DESCRIPTION OF THE SECOND PLAN. The lower plan bears the inscription, in pale ink, " Istud est presbiterium S ei Pharaonis in miaus." This is written in the midst of the plan, like the Latin inscription of the upper one. But at the bottom of the page is written, in black ink, but in the same handwriting, "Vesci lesligement de le glize de Miax de saint Estienne." " This is the plan of the Church of St. Stephen at Meaux." Thus the one inscription refers the plan to St. Faron, the other to St. Stephen, and we have to enquire which of these memoranda is correct. The Church of St. Faron at Meaux is destroyed, but the plan of it, given in Plate 70, differs altogether from that of De Honecort. The Church of St. Stephen appears at first sight equally dissimilar, for it has five chapels to its apse, instead of the three shewn in the manuscript. But an attentive examination of this church (the Cathedral of Meaux) shews that the two chapels placed between the eastern one, and those on the north and south sides respectively, are interpo- lations of the fourteenth century. They must have been made since the year 1268, in which a document quoted by M. Quicherat 1 states that this beautiful and noble building was full of cracks and settlements, and on the point of falling into utter ruin. These additional chapels are nearly in the style of the fourteenth century, and their buttresses have on their faces tabernacles with pinnacles, which do not appear either on the original chapels, or in any part of the church. The sills and stringmolds of the additional chapels are also lower than those of the old ones, and the tracery of their windows different. In the interior, the piers placed between each pair of chapels have on one side bases and capitals in a more ancient style than on the other, and the vault-ribs also shew similar differences. After M. Lassus had made the above observations, the recent restoration of the cathedral under the direction of M. Danjoy gave to that architect an opportunity of examining the chapels, and led to the discovery of the foundations of the plain circular wall which connected the ancient chapels, in accordance with Honecort's plan, and which had been demolished to make way for the entrance- arch from the side-aisle to the interpolated chapel. He also found the base of 1 A notice issued by the Bishop Jean de Poincy in the documents attached to the History of Meaux by D. Toussaint du Plessis. — {Revue Archeologique, t. vi. p. 182.) PL. XXIX Pl^.LXXl |i m l u i l| I | | I I I • 25 75 EXPLANATION OF THE PLATES. 97 PLATE XXIX. VERSO OF THE FIFTEENTH LEAF. " Jestoie une fois en Hongrie la v ie mcs maint jor la vi io le pavement d'unc glize de si faitc raaniere." " J'ctais une fois en Hongrie, \h ou je demeurai maints jours, et j'y vis un pavement d'^glise fait de telle manic're." " I was once in Hungary, and there remained for many a 1 day. There saw I the pavement of a church made in this wise." Wilaks de Honecort gives the tracings of five different patterns contained in as many square compartments. If we consider each compartment to represent a separate paving-tile or stone, it must be composed of a stone incrusted with coloured pastes, or of terra-cotta incrusted in different colours. But if we take the whole drawing to be made up of five separate drawings, each representing a portion of a different pavement, which is the most likely interpretation, then it follows that each pattern is composed of a mosaic of marbles, or different coloured pieces of terra-cotta. The first material is the most probable, because the angles of some of the pieces in the stars of the last design appear too acute for terra-cotta. The last design but one, which is all made up of pieces of the same shape throughout, bounded by two concave and two convex arcs, is very common in Arabian constructions, and is also found in the eighth century in the borders of Carlovingian miniatures. This mosaic-work has nothing in common, as far as design, with the Italian mosaic of the middle ages which bore the name of opus Alex an drin urn. — (L . ) The patterns are all of a kind that admit of being composed of separate pieces, for it will be observed that every constituent piece has a simple and distinct outline. But in patterns formed by inserting into recesses sunk on the face of a stone or tile, clay or other pastes of a different colour, the recesses are either grooves or florid forms that shew clearly how the pattern is made. — (W.) " Chi prennes matere don piler metre a droite loisons." " Ici prenez exemple pour faire un pilier a joints caches °." " Take here an example of a pier with a correct bond, or joints." ° Joints caches in the translation given by the the joints delineated is to place them in a position French editors. It is perfectly true that the effect of where they are concealed, but droite will not bear O 98 EXPLANATION OF THE PLATES. The plan represents one of the piers of Rheims, and is repeated u^on Plate 62 in a more complete form, to which we may therefore refer for the explanation. " Ista est fenestra in templo See Marie Carnoti." " C'est la fenetre de l'eglise de sainte Marie de Chartres." " This is a window of the church of Saint Mart/ at Chartres." This sketch of the rose window of the west front of the Cathedral of Chartres is tolerably exact, as will appear by comparing it with the actual window shewn in Plate 71. But Honecort has introduced several variations which appear to be intentional, for he belonged to a generation of architects whose compositions pos- sessed greater lightness than those of their predecessors, which still retained the solid character of the Romanesque. An architect who had seen and studied the apse of Rheims would find the rose of Chartres too full of plain surfaceman d would be tempted to add openings where none existed, and to enlarge the existing ones. In Honecort's drawing the bases of the radiating columns rest on the circumference of the central circle. In the real window they spring from" a plinth, which is indented so as to form an external foliation to the central circle. The quatrefoiled openings between the arcade and the outer circles have no existence in the real window, and the external quatrefoils of the latter are changed into trefoils, which fit their places better and admit of a larger opening. The drawing is a bare outline or simple souvenir of the general form, omitting details of sculpture and construction, but sufficient for a man thoroughly acquainted with the practice of his own time. — (L.) It may be added, that Honecort has placed his great circles in contact with the heads of the arches of the central arcade, instead of which, they rest in the space between two arches in the real window. This alteration gets rid of the tri- angular blank spaces of the latter, which are too small for piercing, and sub- stitutes a large quadrilateral space that admits of a quatrefoiled opening. His great circles have all twelve foils, but in the real window the central circle has twelve foils, and the outer circles have only eight. It may be doubted whether these variations were intentional, or the mere result of the sketch having been made from memory. — (W.) that sense : it appears rather to be used by Hone- the chapels of Cambray, if they make them right /" cort in the sense of right, or correct, as in Plate 60, which the French editors render " si on les con- "d'autretel mauiere doivent estre celes de Canbrai strait," — " if they make them at all." son lor fait droit," — " In the same form ought to be EXPLANATION OF THE PLATES. 99 PLATE XXX. RECTO OF TH K SIXTEENTH LEAF, MARKED IN THE FIFTEENTH CENTURY WITH THE LETTER q. " C'est unc reonde veriere de le glise dc Lozane. — Ista est fenestra in Losana ecclesia." Without this double inscription it would have been impossible to suppose this sketch to have been intended for the magnificent rose of Lausanne, the variations from the reality, as shewn by Plate 72, being so great. The sketch must have been made from memory, if not from a mere passing glance at the window. The central square is tolerably well filled up, but is placed square in position instead of lozengewise. The semicircular spaces which rest on each side of the lozenge in the original are totally omitted. The eight small trefoil openings are undoubtedly correctly placed in the outer circumference, but the large quatrefoils between them afford but a miserable substitute for the rich quatrefoiled circles of the original. In fact, the unique principle of this remarkable composition is wholly lost. I would rather believe that the drawing was made up by its author, long after his visit to Lausanne, from a few hasty lines scratched on the spot upon his tablets, than follow M. Lassus in supposing that the window was so lighted when he saw it as to conceal the characteristic lines which he has omitted, or that he drew it from the inside of the church p . — (W.) Below the window is the figure of a bearded man in tunic and mantle, seated, and holding with his right hand the foot of his left leg, which is crossed over his right. He looks upward, apparently conversing with some one above. This may perhaps represent Moses putting the shoes off his feet at the burning bush. — (L.) '' I consign to a note tlie following remarks of detected between the churches of Lausanne and Laon, M. Lassus, which are valuable in themselves, but and which prove French influence. Moreover, as have no direct reference to the illustration of our history records - that the Bishop of Lausanne, who author, who cannot for a moment be supposed the presided over the reconstruction of his church after architect of Lausanne Cathedral. " The important its destruction by fire, finished his days in the diocese differences between the original and the drawing of of Cam bray, we may suppose, with M. Rame, that the the Lausanne window seem to prove that Wilars de same architect who had built the Picard cathedral Honecort was not the architect of the church of might have also built the Swiss cathedral. But the Lausanne, for in that case he would certainly have unfortunate testimony afforded by the inaccuracy of made a correct drawing of his own work. In de- Honecort's sketch destroys altogether, in our opinion, scribing the tower of Laon (Pis. 17, 18), we have the supposition that he was the person charged with already pointed out the resemblances which M. Rame this work." — (L.) o 2 EXPLANATION OF THE PLATES. 101 PLATE XXXI. VERSO OF THE SIXTEENTH LEAF. A personage, still young, with full drapery, but naked feet, is seated on an ornamental bench, the eyes are raised upwards, the left hand points forward with the fore-finger, but the right hand, evidently raised, is not drawn. This must be a study for a figure of Christ teaching. The grand style of the drapery, the calm serenity of the countenance, and the careful drawing of the extremities, place this sketch amongst the best in the volume. The careful perspective of the seat shews that it was intended for, or copied from, a mural painting. — (L.) PLXXXII EXPLANATION OF THE PLATES. 103 PLATE XXXII. RECTO OF THE SEVENTEENTH LEAF, MARKED IN THE FIFTEENTH CENTURY WITH THE LETTER T. " Istud est presbiterium beate Marie Vacellensis, ecclesie ordinis Cisterciensis." " C'est le chevet de la bienhcureuse Marie de Vaucelles, eglise de Tordre de Citeaux." "This is the presbytery of the Church of St. Mary at Vaucelles, of the Cistercian Order." The church in question was erected in the neighbourhood of Cambray, and dedicated in 1235 by Henry de Dreux, Archbishop of Rheims. It was destroyed long ago, but existed in 1713, when the two Benedictines, Martene and Durand, describe it as a magnificent church four hundred feet in length' 1 . The views of the abbey published in the eighteenth century give no idea of the form of its apse. The plan in our manuscript partly resembles the joint design of Wilars de Honecort and Peter de Corbie (PI. 28). There is the same square eastern chapel uniting the two circular radiating chapels, or " absidioles." There is also on each side of the choir the same rectangular chapel accompanied on its east side by a circular chapel. But the square chapel which connects, in Hone- cort's design, the two neighbouring circular chapels, has no existence at Vaucelles, where the square chapels have each two compartments in depth, and communicate by an open arch with the adjacent absidiole. As already remarked in the notice of Plate 28, this terminal square chapel is a concession to the ancient Cistercian forms ; but as the transept of Vaucelles is not shewn in this plan, we cannot tell whether the parallel chapels on the east side of the transept, which is characteristic of the order, were employed in this instance. — (L.) It may be worth remarking, that if a square chapel be inserted in this plan on each side of the apse between the two separated semicircular chapels, by treating the two neighbouring buttresses as the piers of entrance to the square chapel, we obtain the plan proposed by Honecort and his friend in Plate 28, even with respect to the peculiar vaulting-lines of their semicircular chapels. For by this change each of the semicircular chapels at Vaucelles is left with a single external buttress, and Honecort's external side-aisle arises naturally out of the inner com- ' Voyage litteraire de deux religieux Benedictins, Src. Par. 1721 — 1724. 104 EXPLANATION OF THE PLATES. partments of the square chapels, alternating with the three inner triangular vaulting-cells of the circular chapels, and thus leaving only the two outer triangular vaulting-cells for the circular chapel. — (W.) "Ce est un imaie Deiu si cuine il est cheus." " This is a figure of our Lord when He fell pros/rale." This may either represent the agony in the garden of Gethsernane, when " He fell on His face on the ground and prayed," or the fall under the weight of the cross on the road to Calvary r . This admirable figure, expressive of such utter exhaustion, is open to the criticism that the left foot is in so forced a position as to make its connection with the leg very difficult to comprehend. But on the contrary, the hands, which support the weight of the body, are drawn with remarkable truth ; the manner in which their form is given in a mere general outline, is exemplified in other parts of the manuscript s . — (L.) 1 Compare Plate 45. s Iii Plates 10, 50, 55.— (W.) PLXXXIII EXPLANATION OF THE PLATES. 105 PLATE XXXIII. VERSO OF THE SEVENTEENTH LEAF. Upon this page begins a set of drawings of carpentry which, judging from the mention of the subject at the beginning of the manuscript on Plate 2, may have extended over the succeeding pages. Unfortunately, the four following leaves are missing, including eight pages, and thus we are probably deprived of a series of many drawings of wooden framing, of which those on the present page seem to be the beginning. In Plates 43 and 44 are some wooden machines and constructions which must also have been amongst those alluded to as " engiens de carpenterie," or " devices of carpentry," in the general summary on Plate 2. " Or poes veir .i. bon conble leger 4 por hierberger deseur une chapele a volte." " You see here a good light (or simple) roof to cover a vaulted chapel" This might be described as a queen-post roof, of which the central portion of the tie-beam had been cut away, the object of the construction being to allow the upper surface of the vault to rise above the level of the walls, and thus to enable them to be made lower than if the tie-beam were carried across. The vault indicated below it is rather a plan of the chapel than a section of the vault. This frame is too weak to serve for wide spans. " Et si vos voles veir .1. bon conble legier a volte de fust prendes aluec gard." " And if you would see a good light (or simple) roof for a wooden vault, look carefully at this." This second roof, partly framed with arched or embowed pieces, is intended to be lined beneath with thin boards forming the surface of a waggon-vault, like many that still remain in England u . " Vesci le carpenterie d'une fort acainte." " Here is the frame of a strong penthouse roof." That againte means a side-aisle is shewn by the legend attached to Plate 62, ' Ltger, as M. Quicherat observes, is in old French siastical Perpendicular Roofs," for Weale's Papers on usually employed in the sense of " easy to construct Architecture. It may be compared with Little Cox- ordo." well, Berkshire, (Parker's Glossary, Plate 174.) " The roof of Old Basing Church, Hampshire, At the left side of the roof near the apex a single is the nearest to this in general appearance that I crocket between two parallel lines is sketched, as if have been able to discover, but is much later in style, it were the beginning of a drawing for the decoration It is engraved in Mr. Clutton's "Examples of Eccle- of the gable.— (W.) P 106 EXPLANATION OP THE PLATES. and the passages quoted by Du Cange under the word accincta prove that the same word was applied to any penthouse x . This roof is intended to cover the side-aisle of a church above the vault, the upper surface of which is indicated by the curve line in the drawing. Its rafter bears on the end of a hammer-beam, to which it is also connected by three vertical posts. A brace rests on a corbel in the main wall, and supports the upper part of the rafter, which is also sustained by a second brace which springs from the inner post of those that connect the rafter and hammer-beam. The square block under the inner extremity of the hammer-beam appears to represent the section of a longitudinal beam, the extremities of which may be supposed to rest upon low walls carried up over the transverse ribs of the vaults, so as to support the ends of the hammer-beams without allowing them to rest on the vaults. — (L.) " Vesci une esconce qui bone est a mones por lor candelles porter argans. Faire le poez se vous saves torner." " This is a sconce which is useful to monks to carry their lighted candles. You can make it if you know how to turn." The thing represented in the figure is frequently mentioned in mediaeval writings under the name of absconsa (see Du Cange). It is, properly speaking, a ventilated case in which a candle may be burned (without displaying its light in all directions : Anglice, a dark lantern) . The legend shews that it was made in the lathe, and was principally employed in convents where the religious had to traverse by night their cloisters and courts with lighted candles y . John de Garland, in his Dictionary, compiled at the end of the eleventh century, gives the monks two kinds of shades for protecting their candles, "crucibulum cum sepo et absconsa, et laterna," that is to say, the night-light or watch-light with tallow, in a dark-lantern, and the common horn-lantern which exhibits the light. The figure in the manuscript must be the dark-lantern, or absconsa, because in this elegant vessel no openings appear, excepting those which are necessary to introduce the candle and let out the smoke. This esconse differs from those represented in cotemporary manuscripts, and the legend which explains how it was made would have been sufficient to have 1 Note by M. Quicherat, p. 179, Revue Archceol., t. 6. * Quicherat, p. 223. EXPLANATION OP THE PLATES. 107 proved that the lathe was used in the middle ages, if so many remaining works had not themselves already taught us that fact z . — (L.) The lanterns in miniatures, especially in the nocturnal scene of the arrest of our Saviour, are nearly the same as those which are in common use at present, consisting of a frame of thin metal garnished with horn, and having a conical cover. A square handle serves to hold them by, and is so formed that it may be attached to a long staff, so as to enable the light to be raised high up, to light a company, or be seen afar off. — (A. D.) ' The lathe used in the thirteenth century was of the simplest form, with a spring pole and cord coiled round the work. The lathe with continuous circular motion is represented in the drawings of Leonardo da Vinci, and has a great wheel beneath the bench, moved by a pedal connected to a winch fixed to the axis of the wheel. The object to be turned is at- tached to an axis, or mandrel, which carries a small grooved pulley, and an endless band communicates the motion of the great wheel to this pulley, exactly as in the ordinary lathe of the present day. — (A. D.) p 2 EXPLANATION OF THE PLATES. 109 THE ELEMENTS OF POETEAITUEE. PLATE XXXIV. RECTO OF THE EIGHTEENTH LEAF, MARKED IN THE FIFTEENTH CENTURY WITH A CAPITAL S. This leaf is the third of the fourth quire, the first two leaves of which are wanting, and also the last two leaves of the third quire, making altogether a loss, anterior to the fifteenth century, of eight consecutive pages, as already remarked. On one of the slips from which these pages were cut the letters d and ba remain. " Chi conmence le mate de la portraiture." " Incipit materia porturature." "Here begin the elements of portraiture" These two inscriptions the author has written at the foot of the page in the same pale ink as the sketches. On the reverse side of this page, upon which this series of drawings is continued, a third inscription to the same purpose is added, which may be given in this place for the sake of comparison. It is written in darker ink and in a more compact character. "Ci comence li force des trais de portraiture si con li ars de iometrie les ensaigne. por legierement ourer. et en lautre fuel s'r cil d'le maconerie." The interpretation of this appears to be, — "Here begin the powers of the lines of portraiture for facilitating work, as taught by the art of geometry. On the other leaf ivill be those of masonry a ." Four pages in succession are exclusively devoted to this method, and for- tunately in a part of the manuscript which has escaped mutilation. At the bottom of the last is an inscription which shews that in these four we have all that the author recorded in illustration of this subject \ " The translation in the French edition is simply French edition. — (W.) " Ici commence la methode du trait pour dessiner la b The author has also employed his method in figure ainsi que l'art dc la geometrie l'enseigne pour Plates 41 and 61, but these were never intended to iacilement travailler." I have ventured to substitute form part of the treatise, or rather series of drawings, uew descriptions of the four plates on the Elements under consideration. — (W.) of Portraiture instead of the concise notices in the 110 EXPLANATION OF THE PLATES. " En ces .iiij. fuelles a des figures de lart de iometrie. mais al conoistre covient avoir grant esgart ki savoir velt de q' cascune doit ourer." " hi these four pages are figures of the art of geometry ; but to understand them great atten- tion must be given by any one who would comprehend the peculiar use of each." This art of geometry has been admirably characterized by M. Quicherat as follows d : — " It would be extremely difficult to give a precise definition of this method, so arbitrary is it in application. The process consists either in reducing human forms to simple lines, or in reducing the representations of human or animal figures to elementary forms, such as triangles or squares set in juxtaposition 6 . All this is done without calculation or principle, so that geometry has no other office than to furnish the forms and nomenclature of a very questionable ap- proximation. The processes in question teach, not a science of drawing, but a mere art of readily reproducing certain attitudes, by merely retaining in the memory the simple geometrical figures which are respectively associated with them. Thus, eye and hand would become the slaves of habits which, because they dispense with the study of nature, make drawing easy, according to the boast of Wilars de Honecort." " The matlere de portraiture is, in truth, a mere routine, and the drawings are a set of patterns for a certain number of selected subjects. But it is remarkable that the peculiar attitudes and aspects produced by this method are precisely those which characterize the works of the painters and sculptors of the thirteenth century." It may be inferred that Wilars de Honecort does not claim the invention of this system, but merely the composition of a sufficient number of elementary figures to place it upon record for the use of posterity. The drawings exhibit several distinct methods. First, a diagrammatical representation of a human figure, viewed in front or obliquely, which consists in substituting for the body an isosceles triangle, with its narrow base upwards ; the head and neck are supplied by a little circle on a stem placed in the centre of the base. The angles of the base are the shoulders, from which proceed the arm, fore-arm, and hand in the guise of straight lines meeting at the angles corresponding to their required positions. The lower limbs are similarly indicated by straight lines, but the thigh-lines diverge together from the apex of the triangle, ignoring the fact that their proper articulations with the body are separated by nature nearly as d Revue d'Jrcheologie, p. 211. e After the manner of the Chinese puzzle.— (W). I EXPLANATION OF THE PLATES. Ill widely as those of the arms above. Nevertheless, this artifice supplies a spirited and unmistakeable representation of human attitudes. Two examples only are given in this treatise, both of them in the plate under review. The first repre- sents the infant Jesus on the knee of the Virgin. The infant is a pure specimen of this artifice, the female figure is drawn in the same way, but lines represent- ing the drapery have been added. Yet we see the triangle inclined forwards, and the right arm represented by two lines, but plainly sustaining the leg of the child. The square below seems to represent the seat, of which the back rises in a single line terminated by a knob. One leg of the female is seen in front. The second specimen is a king seated on his throne. But in the sixty- first plate the author has delineated the angels that crown the buttresses of Rheims in this manner, and also the figures in the wheel of fortune in Plate 41. The second method consists in selecting some simple and easily recollected geometrical figure, the lines or angles of which will coincide with the leading lines or points of the natural figure, so that by drawing the first the arrangement of the second may be reproduced by merely filling it up with the necessary details. Thus a rectangle, raised above the ground to a distance equal to its height, serves as the foundation for the body of a stag. A right-angled triangle, with the longest side in front, and vertical, and of which the right-angle coincides with the upper corner of the rectangle, indicates the place of the neck. A third and smaller triangle seems to guide the outline of the face, but is imperfectly drawn. The sheep in the next plate is sketched on the same principle, and should be compared with the stag. Beneath the stag is a man thrashing with a flail. This is an example of a figure viewed sideways, and the diagram employed in this case is repeated in several other side-views of men in different positions, which by comparison serve to illustrate the method. These are in Plate 36, the mower with his scythe, the two figures blowing long horns, and standing back to back, and the sitting figure with a child on its knee. In the front, or oblique views, as already explained, a triangle represents the body, and in these side-views, as in the thresher and mower, a single line from the shoulder to the hip appears to be this triangle seen edgewise. Two lines radiating from its lower extremity represent the thighs, a third, joining the upper end of the body line with the knee of the hinder leg, forms a triangle, and governs the front of the body, or rather, perhaps, the direction of the front outline of the thigh. 112 EXPLANATION OF THE PLATES. The extremely elementary sitting figure at the bottom of Plate 36 seems to shew that the ruling principle of all the above-cited examples of side-figures is to represent the body in a side view from the hip upwards by a nearly isosceles triangle, with a short base downwards. Some additional remarks will be found in the explanations of Plate 36. At the bottom of Plate 34 is a pair of figures which appear to be a man and woman, the one in a bold and manly attitude, the other submissive. In these figures a small triangle is employed for the face, and the usual triangle for the body. But to obtain the solid form of the legs, and in some respect their direction, the sides of the triangle are continued downwards to the ground, and two lines diverging from the centre of the horizontal upper side of the triangle meet these sides so produced on the ground line. The left side of the diagram so obtained strictly governs the left half of the male figure, and his left arm has its outer outline formed of a straight line springing from the angle of the triangle, and meeting a second straight line with a curved stroke at the end to designate the fore-arm and hand. But solidity is given to the arm by the addition of a freely drawn inner outline ; and thus it is explained that the diagrammatic arm indi- cates the outer outline f . The right side of the figure is sketched in a spirited and natural manner without respect to the geometrical lines 8 . In the female figure the greater part of the diagram seems to be useless. In Plate 36 we shall find a figure in which this diagram is rigidly employed for the whole. In the upper corner of the page is drawn a grand head, which M. Lassus suggests to have been a St. Peter, and inserted in this manner by the artist by way of shewing that he could himself dispense with the elementary methods that he has taken such pains to display for the use of others h . — (W.) ' This is not always the case. In the right arm of the king two lines are added for the solid outline, which place the diagrammatic one like a bone in the middle of the real arm. — (W.) e The minstrel in Plate 50 is so similar to the man above described, that the foundation of his contour must have been laid in the same manner, although no geometrical lines are shewn. — (W.) h The first method may be supposed to proceed upon the principle that the lines of the diagram re- present the bones of the skeleton, and thus the atti- tudes of the human figure are represented by first drawing the bare bones and then clothing them with flesh. If the diagram be modified so as to bring it more into harmony with the real arrangement of the articulations, and its lines drawn with a due attention to foreshortening, this system becomes a reasonable and scientific one. It was, in fact, proposed in this form by Lautensack in his Ars Perspective?, Frank- furt, 1564.— (W.) EXPLANATION OF THE PLATES. 113 PLATE XXXV. VERSO OF THE EIGHTEENTH LEAF. In this page there are several examples of the use of the peculiar star-shaped pentagonal figure which was known as the Pentagram or Pentangle, attributed in the olden time to Pythagoras or Solomon, used as a mystic symbol, and as such employed by the Freemasons, and invested with magic powers. Wilars de Honecort has inscribed a small pentagram on his sketch of a tabernacle in Plate 17, and in this page he has used it to regulate the proportions of the front gablet of a tabernacle, and of the ridge of its lateral gablets in height and in length. The same figure is employed for the face of a bearded man. Its five points determine respectively the position of the apex of the forehead, the breadth of the face at the level of the eyebrows, and the breadth and position of the angles of the lower jaw ; the point of the nose is seated at the intersection of the two lower sides of the figure. In the spread-eagle the pentagram is drawn so irregularly as to serve no apparent purpose '. At the top of the plate is a horse's head, with a man's head beneath it, both in profile ; and an equilateral triangle, with its front side vertical, is inscribed in each in a manner as nearly similar as the dissimilarity of the two will allow. The triangle, but not equilateral, appears to be the universal foundation for the side view of an animal's head, for it occurs again in the stag, the greyhound, the sheep, and the pig. In the man's profile the head is completed by placing a semicircle on the upper side of the triangle. A man's front face is sketched by drawing a square, and dividing it by two lines into three unequal compartments. The chin is formed by a portion of the circle which would be inscribed in the square, the upper line determines the place of the eyebrows, the lower the point of the nose. Next to this is a circular face, such as children draw to represent the man in the moon : a transverse line at one-third of the diameter from the top deter- mines the level of the eyebrows, the tip of the nose is half way from this line to the chin. In the third row, a venerable countenance is characterized by a pair of triangles. In Plate 37 will be found another head of this class 3 . 1 It may be worth mentioning that the length of 5 An equilateral triangle, similarly placed to that each ray of a regular pentagram is equal to the of the head in the second row, is employed by Fra diameter of its pentagonal body. Luca Pacioli da Borgo for the demonstration of the Q 114 EXPLANATION OF THE PLATES. It seems as if the principle of Honecort's method was that each countenance or object should suggest its peculiar diagram to the artist, by means of which he might sketch it more faithfully and recollect it. Indeed, guide-lines are often either drawn, or imagined to exist, in the modern methods of sketching, and such diagrams as our author gives might have been traced upon a drawing of the middle ages which it was intended to copy, for the same purpose as the squares which are now usually ruled to guide a copyist, especially in reducing drawings or plans. A greyhound and a sheep have each the neck and head composed of two triangles set in juxtaposition. The body of the sheep is a rectangle, like that of the stag in the last plate; but the lanky greyhound has a pair of triangles ingeniously substituted for this rectangle. In the third line the outline of a human hand is formed by a square with a thumb added to it, and the general contour only of the fingers indicated by a curved line. This is so rough as scarcely to deserve attention, were it not that the hands of the figures in Plates 32, 50, and others, shew that our author employed this mode of delineating the extremities. Finally, at the bottom of the page a pair of circular arcs are used to give character to two ostriches, or, rather, to draw them alike. The inscription beneath the page has been already explained. — (W.) proportions of the human face and head in profile. equal portions, of which the first line coincides with (Divina Proportione, pars prima, p. 25, Ven. 1509.) the mouth. This is nearly the same as in Honecort's The face is divided by horizontal lines into three equal circular face. Such rules are probably of great an- parts ; the first extending from the forehead to the tiquity. Albert Durer has some similar diagrams in top of the eyelid, the second to the point of the nose, his book on Human Proportions. — (W.) the lowest part is divided by two lines into three / EXPLANATION OF THE PLATES. 115 PLATE XXXVI. RECTO OF THE NINETEENTH LEAF, MARKED IN THE FIFTEENTH CENTURY WITH THE LETTER T, (OR, ACCORDING TO M. LASSUS, A POSSIBLE V.) This page contains several examples of figures in pairs set symmetrically in opposition, as the two trumpeters back to back, and two groups of wrestlers ; besides the first figure on the page, which is evidently intended to be filled up so as to represent two men looking in opposite directions. The two trumpeters k are sketched upon an inverted pentagram ; if this pentagram be divided by a vertical line into two equal halves, it will be seen that each trumpeter is governed by the same diagram as the mower and the thresher, already described in p. 111. In the unfinished double man at the top of the page each figure is founded upon a pair of triangles resembling the letter K. The same K-shaped diagram is used for the wrestlers beneath the trumpeters, but as these men stand face to face the two K's unite, and produce a diagram resembling an upright square containing a diagonal square. The second group of wrestlers on the right hand of that just described is cir- cumscribed by a kind of beehive-shaped diagram, the lines of which appear to have been sketched merely to assist in drawing the two opposite figures alike. In the second figure of the upper row the usual triangular body is employed, with its sides continued downwards to assist in giving position to the right leg, but on the left side to that of the thigh only. On the other hand, the third figure of the second row is an example of the complete application of the same diagram, which is used in a partial manner for the two figures at the bottom of Plate 34, but which here produces a sturdy warrior standing in an attitude of defiance ; his head, perfectly circular, is drawn like that in the right hand upper corner of the last plate, and in nearly all the profiles of this page, the diagram of that in the second row of that plate is used as a short-hand mode of indicating the head and face, and in general the knee-joint, whether viewed in profile or in front, is drawn as a complete circle. k The slightly curved trumpet here represented, time angle sona sa bosine," says the text, and the which was in general use in the thirteenth century, accompanying drawing shews an angel sounding a is termed a losine in a manuscript of the Apocalypse trumpet like those represented in our manuscript. — cotemporary with Wilars dc Ilonecort : — "Et li sep- (A. D.) Q 2 116 EXPLANATION OF THE PLATES. The last figure in the upper row seems, like the first, to be incomplete. In the second row the monk's frock is whimsically obtained by reversing the usual triangle. The cavalier in the third row is, as M. Quicherat observes, the type of that which is reproduced upon so many mediaeval seals, and is a most curious example of the system on account of the ingenuity with which the leading lines are sub- jected to a star, formed by eight rays, diverging at equal angles from a point determined by the intersection of the level line of the horse's back with the front outline of the rider's body. The two crouching lions, resembling those which are so commonly found in the church portals of Italy, are each well characterized by a triangle on a horizontal base, with one acute and one obtuse angle l . The object of setting two symmetrically opposite appears to be to shew the use of these diagrams in drawing reversed figures* exactly alike, of which process the two trumpeters and the wrestlers beneath them are also such excellent examples. The pairs of figures so arranged in Plates 10, 14, 25, 27, and 50, were probably produced by diagrams of this nature. The seated figure with a child on her lap, next to the lions, has been already noticed in p. 112 ; and lastly, two flowers, the one containing a pentagram, the other a six-rayed star, shews that the method was extended to flowers. — (W.) 1 It is probable that the diagram in this case is an and similar figures is also an irregular form of the inverted pentagram with its lower ray cut off, and favourite pentagram. — (W.) that the diagram employed for the sturdy warrior \ EXPLANATION OF THE PLATES. 117 PLATE XXXVII. VERSO OF THE NINETEENTH LEAF. In this plate we find a new class of drawings, in which many figures are so grouped and entangled as to confound their members together, and make the same member serve for two or more figures. Thus four men are arranged round a centre in such a manner that each viewed separately appears to have two legs, yet there are but four legs in the whole group. Each man lifts a hammer, and seems to be driving a nail into the foot of the man next in front of him. The whole machine is probably intended to revolve about its centre for the purpose of striking a bell. Machines of this kind are not uncommon in foreign churches. The hand or arm that carries the hammer of each figure would in that case be mounted on a pivot, so as to fall on the bell, and then escape from it as the wheel by its rotation carries it past its edge. The group of three fishes at the upper corner of the plate have but one head and one eye in common m . At the bottom of the page is a curiously ingenious group of figures, not com- pletely filled up, but easy to finish from the indications given. If so completed, the design would exhibit eight figures with varied attitudes, yet symmetrically arranged about a centre n . The whole is contained in a square divided into thirty- six smaller squares, and by diagonal lines for the purpose of directing the draughtsman. By the side of this a head and face are sketched upon a re- ticulation consisting of a square group of sixteen small squares with diagonals. Apparently these squares are introduced, not so much to supply a rule of proportion, as to enable the artist to draw the two halves of the head and face alike ; just as in the previous example the squares would enable him, after drawing one figure, to place all the others symmetrically about the centre, and finally to make a copy on a larger. This, in fact, would be the modern mode of doing the same thing. The pig's head, with its triangle, belongs to the same class as the heads of the horse and man, the sheep and greyhound, on Plate 35. The final inscription of this page has been already examined under Plate 34. — ( W.) end of the art of portraiture. m M. Lassus observes that grotesque combinations longing to the middle of the thirteenth century, con- of this class occur in several sculptures of the middle firms the date of the manuscript already inferred from ages, especially at the doorway of the library at other evidence. Rouen Cathedral. He quotes also the three com- n The usual hieroglyphic head and face of Plate bined legs which are. the arms of Sicily, (and, it may 35 serves to shew which way the figures arc looking, be added, of the Isle of Man,) and remarks that the and the same diagram is employed for the hammer- cylindrical helmet, with its two slits for vision, be- men above. — (W.) ar c>)tt ftur »m k ou PLXXXVIII >uTn Snvc o>totibe que on tic uojv cme tore archw Uo(bw ui\« avc- arcbu tailor* 1 ^ I N >— *- /» p com cl pvA et au^^f pafo-rj*^ hi cfr Wf • W Vhit a flW am at* c^upaftjfom one p«m^ ^leT.ii. ***** V 5f0 m«v, f EXPLANATION OF THE PLATES. 119 THE GEOMETKY OF MASONEY. PLATE XXXVIII. RECTO OF THE TWENTIETH LEAF. The paging added in the fifteenth century becomes confused at this point, for the book having been closed before the ink was dry, the characters written on this and the next page are so blotted as to make them illegible. This is, however, evident, that the literal paging is henceforward changed into an arithmetical one in Roman characters. Apparently the writer finding, when his alphabet had reached v, that it was nearly exhausted, determined to continue his paging in Roman numerals, by considering v to represent the number five instead of an alphabetical character, but has carried out his intention in a bungling manner. This page and the two following are exclusively devoted to a series of diagrams representing various geometrical devices relating to construction, and for the most part to masonry. The inscription terms them geometrical, — " Totes ces figures sunt estraites de geometrie." Under every diagram is a short and gene- rally enigmatical legend which indicates its purpose, but in no case explains the artifice, and may consequently be as " a word to the wise," for the learner may seek its meaning in vain. They appear to have no pretension to the constitution of a body of instruction, neither can they claim to shew the ordinary practice of the period j they seem rather to be a chance collection of expedients to meet particular cases which the author picked up here and there in the course of his travels, and noted, primarily for his own use, like the other drawings in his note-book, and lastly, has consigned them with the other contents of the volume to posterity. The diagrams, with very feAV exceptions, are regularly ranged in rows upon the page, each row being bounded below by a border of two parallel lines, between which the legends descriptive of the purpose are written in compartments placed below their respective diagrams . Notwithstanding this appearance of method, This table, which exhibits the distinguishing numbers attached to the explanations which follow, being disposed in the same order as the figures to which they re- spectively apply in the plate, will assist in finding the diagram that belongs to each. Plate 39 contains the continuation of the series from 20 to 32, and Plate 40 its con- clusion from 33 to 40. I have found it necessary to substitute new descriptions for all those which relate to the three plates on this subject in the French edition. — (W.) 1 . 2 . 3 4 . 3 . 7 . 8 9 10 11 12 13 14 15 16 17 18 19 120 EXPLANATION OF THE PLATES. there is no classification in the arrangement of the collection. Devices of the most puerile character for pastime alternate with stone - cutting, mensuration, and carpentry, as chance, the size of the diagram, or the order of acquisition may have determined. 1. " Par cu prenum la grosse done colonbe que on ne voit mie tote." "Mow to take the diameter of a column oftvldclt only half is visible" This is a simple device, contrived to obtain the relative position of three points of the circumference of a horizontal section of the column. The two points of the ordinary mason's compasses are placed in contact with the surface of the column, and a piece of wire, applied to the curved bar by which the legs are kept in position, is pushed into contact with an intermediate point of the surface, and held by the fingers or tied fast, so that the compasses may be placed flat on a drawing-board, and the position of the three points accurately laid down. 2. " Ar chu trovom le point en mi on canpe a conpas p ." " Ainsi trouve-t-on le point au milieu (Tun champ decrit au coin pas." — (Quicherat.) "How to find the point in the centre of a circular area." This, which is a continuation of the first, shews the three points laid down on the board, and the circular section of the column as duly described from the centre point obtained from them. M. Quicherat remarks that this figure " merely shews the solution obtained, without indicating the method, as it represents merely a circle, on the circumference of which are marked the three points, by means of which it was obtained. The problem was well known to practical masons q under the name of the trois points perdus." It appears in the earliest written books on the subject, such as Albert Durer's " Geometry," and Philibert de Lorme's " Architecture," (1. iii. c. 4). The two points which are marked below the diagram seem to shew that a rougher method was employed at this earlier period. The lowest of these may be the intersecting point of the compass-legs, marked at the same time with three p compas is used for a segment of a circle at p. 45 equivalent problem, " To describe a circle about a above. given triangle," (bk. iv. p. 5,) tlie points of the tri- 11 The construction is the same as thai uf Euclid's angle being the three points given. EXPLANATION OF THE PLATES. 121 circumference points ; the other is perhaps the place of the wire which is set in the middle of the iron bar. The intermediate point of the original three seems also to be carefully taken midway. Thus we are led to suppose that a diametrical line was drawn by help of two of these last-mentioned points r , and the required centre found upon this line by trying different openings of the compasses, until one was hit upon that would draw the arc through the three points. This pro- cess, inelegant and uncertain as it may appear to be to a geometer, is, after all, sufficiently rapid in practice, and more consistent with the coarse methods em- ployed even by modern artizans than the exact construction. — (W.) 3. " Ar chu tail om le mole don grant arc dedens. in. pies de tere." " Par ce moyen taille-t-on le modele d'un grand arc dans trois pieds de terre." "How to cut the mold of a great arch in a space of three feet." The mold, or pattern, which is used for shaping the faces of the voussoirs of an arch has necessarily the upper and lower edges formed of portions of the cir- cular arcs that bound it above and below (technically termed the extrados and intrados) ; the sides of the mold must converge to the centre of the arc. This mold can be readily laid down by means of a long ruler, or even a stretched string, of the length of the radius of the circle, and attached to a pin in the place of the centre. But this supposes that the floor or place in which the drawing is, is large enough to contain the length of the radius. The problem enunciated above shews how to perform the operation in a very small workshop, and would in modern phraseology be termed a method of describing an arc of a circle or its radii when its centre is inaccessible s . In this diagram the large arc appears to be obtained by scribing a series of arcs one from the other in succession. A complete semicircular mold, or templet, as it is called, is first made as large as the space allows of, and a guage-rod, or scribing-stick, is provided, with a broad-faced notch so formed as to enable it to rest or travel upon any part of the circumference of the semicircle with its edge always in the direction of its radius. If, for example, a tracing-point be attached to this rod, at a foot distance from the notch which touches the semicircle, and the rod be made to travel along the edge of it, the tracing-point will describe ' Perhaps the two lower points were obtained by Centrolinead. Vide Trans. Soc. Arts, vols. xxxiL, setting off two pairs of intersections from the extreme xxxiii., and xxxix., for the description of these con- cireumference points, with equal radii respectively. trivances by', Messrs. Nicholson, Farey, and Rotch ; 5 For this purpose instruments are constructed also Peter Nicholson's "New Practical Builder," which bear the names of Arcograph, Cyclograph, and 1823, p. 562, &c. R 122 EXPLANATION OF THE PLATES. an arc of a circle whose radius is a foot longer than that of the semicircle ; and if the edge of this new arc be cut out so as to form a second templet, and the rod again travelled along its edge, an arc of larger radius will be obtained, and so on. The same guage-rod may be used, as shewn in the diagram, for obtaining the direction of the lateral lines or joints of the voussoirs. The case in question would rarely, if ever, occur in practice, but, as I have already said, these problems must not be considered as shewing the ordinary methods in use by the cotem- poraries of Wilars de Honecort, but rather expedients, or tours de force, for the exercise of ingenuity. M. Quicherat 3 explains the legend by supposing that it relates the execution of a clay model of the full size of one voussoir, which is to be used as a pattern for all the rest. His explanation is headed, — " Execution du modele en terre avant de construire un arc." And the legend is then translated and interpreted thus : — " Tailler le moule" is to model in solid from the elevations and profiles, and thence to carve out a voussoir which, in accordance with the known properties of the semicircular arch, (which the author distinguishes by the name " grand arc"), may serve as a pattern for all the other voussoirs of the same arch. "Dedans trois pieds de terre" shews either the surface of ground, or the bulk required for the work. " It signifies little which of the two. Neither does it matter whether the trois pieds be an accurate measure, or an indefinite expression to signify a small quantity, or whether the three complete semicircles shewn below the segments which are produced by the operation are drawn to explain it, or for some other purpose, still the fact of the execution of a model in relief is put out of doubt." Thus far M. Quicherat. M. Lassus, on the other hand, is of opinion that the workshop must be sup- posed too small to contain a drawing of the whole arch on the full size. A drawing on a small scale is therefore made and divided into the convenient number of voussoirs ; and then, in the remote corner of the workshop, a portion of the circumference, as large as the space will allow, of the full-sized arch, is drawn concentric to the small drawing. The radii of the small arch that repre- sent the joints being produced to meet the large arch, will give the width of the voussoir, and the inclination of its sides. The guage applied to the arch in the drawing is, according to this explanation, employed to transfer the dimensions of the voussoir from the drawing to the stone. 8 Revue d'Arclieologie, p. 169. By some strange marque q'un architecte anglais tres verse dans la misunderstanding he refers to my own memoir on connaissance du gothique. M. Willis, en est venu, the vaults of the middle ages, (Transaction of British avec la seule ressource de ses observations, a con- A reinfects, and Daly, Revue d 'Architecture, 1S43,) as jecturer le meme fait," p. 1G9. corroborating his opinion : — " II est digne de re- EXPLANATION OF THE PLATES. 123 4. " Ar chu vosom une arc lc cintrcel de vers le cicl." " Voici un arc, le cintre tourne vers le ciel." The meaning appears to be, — " This shews an arch, the centering of which is on the outer side" The purpose of this diagram is very obscure. As mediaeval arches are for the most part built in ranges of concentric voussoirs forming successive orders, a case may be conceived to occur in practice in which one of these arches may be required to be set beneath one already constructed, which would therefore serve as an outside centering to determine its form. But the voussoirs would require to be wedged up from below to sustain them'. The drawing appears intended as a memorandum of the process, whatever it be. — (W.) 5. " Ar chu fait om on cavece a xn vesrires." " Par ce moyen fait-on un chevet a denize verrieres." " How to make an apse with twelve windows." In this singular diagram there is a polygonal apse with five sides in addition to two parallel sides ; four buttresses are represented, and the walls of the five sides indicated, but not those of the parallel sides ; neither is the junction of this apse with the remainder of the building indicated. Vault-lines are shewn, but in a manner incompatible with the plan, because they spring from the wall at points which are not opposite to the buttresses. In fact, the five-sided apse is divided into three vaulting compartments. Two dots in each compartment near the wall may stand for pillars or window- jambs, or for the columns of an arcade below the windows. It is impossible to gather from this hasty figure the arrangement of twelve windows which the legend mentions as the characteristic of the plan. Lassus suggests the plausible interpretation that we should read vn. instead of xn., in which case a window in each of the five inclined faces, added to one in each of the side walls, would make up the number ; the vault-lines must be supposed to be entirely wrong. — (W.) ' The engraving represents the arch as completely fitted with voussoirs, but in the manuscript one of the joint-liues is left out, namely, the fourth on the right side, so as to shew an unfinished arch, of which five voussoirs are set on one side, and three on the other, and two are wanting. M. Quicherat's woodcut re- presents it this way, and agrees with my own tracing, which I should otherwise have suspected of error. M. Lassus thinks that the complete arch represents the one formed of the voussoirs traced by the previous process. The guage-stick is to be set in coincidence with the joints of this arch in order to describe those of the outer one. — (W.) 2 124 EXPLANATION OF THE PLATES. 6. " Ar chu tail om erracemnens." " Par ce moyen taille-t-on les sominiers (arrachemens) de voute u ." " Thus are shaped the first, or springing stones of a vault, or arch" The first stones of an arch-vault are to this day called "arrachemens" by French masons, and the purpose of this diagram is merely to shew how to trace the soffit. In fig. 7, A B C A'" is the head, or vertical side, of such a stone, in which B C is the edge of the soffit. The lower bed A"'C is necessarily horizontal, and in mediasval vaults, and sometimes in the arches, the upper bed is also horizontal. Manifestly the upper extremity B of the soffit overhangs the lower extremity C, and if a perpendicular B B' be let fall from the top, B' C will measure the overhang. If a mason's square be placed on the upper bed of the stone, and a straight- edge be employed to continue the perpendicular B B' from its upright leg, the point C can be marked by setting off the distance B' C obtained from the working drawings. A camber-slip, or sweep, whose edge is shaped to the curvature of the soffit, being placed with that edge in contact with the two points B and C, enables the curve to be traced. This explanation differs from that given by Wilars only in the position of the stone, which in our diagram is placed in its true position, as easier to understand, but in his is inverted (as in fig. 8), and therefore enables the process to be performed, whether the upper bed be hori- zontal, or like that of a modern voussoir, inclined towards the centre of the arch, as indeed is the case in most of the mediaeval arches. As he has not drawn this upper bed in its inverted position, it is clear he thought its form Figure 8. of no consequence. — (W.) 7. " Ar chu fait om cheir deus pires a un point si Ions ne seront." " Par ce moyen fait-on arriver deux pierres a un point, si elles ne sont pas eloigners." "By this means two stones can be brought to the same point, if they are not far distant." Both Quicherat and Lassus declare this diagram to be wholly inexplicable. The ° The diagrams are given by Lassus, and his explanation is essentially the same as that which I have given. — (W.) EXPLANATION OF THE PLATES. 125 former, by classing it with one in Plate 40 which shews how to make a pear fall on an egg, evidently imagines that this relates to a similar trick. It seems to me to belong to masonry, and to be some artifice for guiding the stones of a large column into their proper places as they are being lowered for setting. Con- sidering the very simple devices that are thought worth recording in this collection, I should even venture to suggest that the whole affair relates to making corre- sponding marks upon the edges of two stones that are to be in contact, so as to guide the setters in placing them. The figure, in this view of the question, represents the plan of a circular shaft built of radiating stones, of which the joints tend to the centre-point. To set the stones so that they shall all tend to this point is the meaning of the legend. Three of the joint lines are crossed with a short line which represents the two guide-marks in contact. The short radial line on the left must be a perspective view of an upright pin stuck in the centre : the two characters below are masons' marks. — (W.) 8. " Ar chu tail om vosure destor T de machonerie roonde." "Par ce inoyen taille-t-on une voussure*, de fenetre en maconnerie ronde." " Thus is cut the voussoir x (of a window) in a building of circular masonry." M. Quicherat understands by vosure d'estor the rib of a vault garnished with mouldings, and by machonerie roonde the curvature of the rib, and thus interprets the figure to represent two sections of mouldings placed in opposite directions y . This view can hardly be accepted, for the drawing clearly represents the plan of the window of a round tower, as Lassus interprets it. A straight-edge is placed horizontally across the opening, and at equal distances from the jambs on each side are set-off lines with divisions upon them, which in these drawings usually indicate measurements, and here serve to shew the deviation of the curved face of the stone from that of a plane wall. Lassus suggests that the two stones represented must have been wrought in the workshops, and that the marks in question are intended to guide the setters. The above explanation applies only to the jamb-stones, for the voussoirs of the arched head of a window in a circular wall have each of them a different deviation according to their distance from the keystone, and this of so complex a ' Perhaps from estorer, creer, construire, batir, &c, for translating it voussure, or vaulting, in the present instaurare (Roquefort), or from tor, a tower. — (W.) instance. Vide, for example, fig. 4 in Plate -10, where 1 It must be observed that the word voussoir is in the vosure is an unmistakeable voussoir, or arch- trie manuscript spelt vosoir, vosor and vosure, in- stone. — (W.) differently, and that there is consequently no reason ? Revue, p. 170. 126 EXPLANATION OF THE PLATES. nature, that considering the small stones employed in the middle ages, we must conclude that the heads of the voussoirs were first worked as if they were intended for an arch in a plane wall, and that after being set in position their surfaces were wrought into coincidence with the general curvature of the wall. It may be fairly conjectured that the word vosure included jamb-stones, as well as its modern sense of arch-stones ; yet the figures at the bottom of this plate, which evidently belong to the same class as the one under consideration, incon- testably appertain to arch-stones.— (W.) 9. " Ar chu tail om vosure besloge z ." " Par ce moyen taille-t-on voussure oblique." " In this way is cut a slceiv voussoir." From this figure and others it is evident that the useful instrument termed a bevel, or jointed square, which can be set to any angle, was not known to the cotemporaries of Wilars de Honecort. No trace of it is to be found in his sketches, and in all cases that appear to call for its application, as in the present figure, the angle is indicated by measuring the triangle by which it differs from a right angle, or by cutting a triangular board to serve as a mold or pattern a . The present figure shews a plan of an oblique passage through a wall. The acute angle on the left hand, which would now be taken with a bevel for the guidance of a mason in working the jamb-stones, is ascertained by applying the inside of a square, and measuring the length of the perpendicular distance from its extremity to the surface. The obtuse angle on the opposite side is ascertained by applying the square with one leg in the direction of the face of the wall, and the other con- sequently perpendicular to it, and then measuring the distance from the inner angle of the square to the obtuse edge of the jamb b . By these measures the angles can be transferred to the stone. — (W.) 1 M. Quiclierat observes that the word besloge by the method here given of obtaining this parallelism its analogy with lalonge, or berlonge, which are dif- c' fercnt forms of the old adjective larlong, must bear the same meaning. a Vide No. 28 below. b Quiclierat suggests that the figure belongs to the voussoirs of a vault-rib in an oblique direction, " taille des voussoirs d'une nervurc biaise," without farther explanation. Lassus, on the other hand, gives a totally different interpretation, in which I cannot concur, but which, as others may be of another opinion, I subjoin A. in the words of the writer : — " As the two outer or F,guie visible surfaces of the wall or vault must be parallel, is the most simple conceivable. A right-angled tri- EXPLANATION OF THE PLATES. 127 10. " Par chu fait om on pont de sor one aive de fas de xx pies de lone." " Par ce fait-on un pont sur une eau de bois de vingt pieds de long." " This is the way to make a bridge over a river of wood of twenty feet in length." This ambiguous description, which leaves us in doubt whether the twenty feet belong to the river, the bridge, or the individual pieces of timber, having been taken by Quicherat in the first sense, he exclaims in astonishment that it is in- credible to behold the number of pieces employed in a bridge of so small a span. But Lassus more practically adopts the latter appropriation, which gives the problem the form of " How to make a bridge over a river of about fifty feet in breadth, with timbers of only twenty feet in length?" Two lofty piers of masonry are erected, between which the frame of carpentry is set up. It is probably ex- aggerated in its proportional height. The framing is indicated by lines only, but explains itself. The roadway has apparently a gateway-arch at each end to bar the passage at pleasure. — (W.) 11. " Ar chu fait om on clostre, autretant es voies com el prael." " Par ce moyen trace-t-on un cloitre avec ses galeries et son preau." " Thus a cloister may be laid out, with its deambnlatories and garth." The dimensions of the proposed cloister must be supposed to be given, and the position of one of the sides. The diagram appears to correspond to the folloAving well-known process. Set a stake C in the ground nearly opposite the centre of the side A B, of which the position and length are given, A, and about half-way across ; the exact position of the stake is of no consequence. The diagonal A C, drawn from C to the left hand upper corner of the square, represents a rod or a string stretched to the exact distance. Carry the upper extremity of the string round, so as to trace a circle DAE, intersect- ing the given side A B in a point D. Draw a straight line through D C, and angle ABC, whose base B C includes a certain number of divisions, is described upon the right hand voussoir. If now a square be applied against the left face (of the oblique opening, in the manner shewn in Honecort's figure), the angle o which it makes with the bare wall will be equal to the angle a at A, the apex of the first triangle. Thus by measuring be- tween the wall and the leg of the square a distance B' C equal to the base B of the first triangle, the leg which touches the soffit of the voussoir will be in a direction parallel to the other face. But the distance B' C must be exactly normal to the wall, and taken at a distance A' B', from the arris A' of the oblique opening equal to the thickness A B of the wall or voussoir, so as to make the two triangles ABC, A' B' C, equal." The figure can hardly be supposed to apply to an oblique arch, for the angle which the face of the wall makes with the soffit of such an arch varies at every point, unless it be taken in a horizontal plane. — ( W.) 128 EXPLANATION OF THE PLATES. produce it to meet the circle in E. The line AEF will be perpendicular to D A, and the length of the side A F must be set off upon it. The other sides are easily set off by lines F G, B G, respectively equal to their opposite sides, meeting at G. This elementary process, which is one of those given by De Lorme for this very purpose, is so well known that I should scarcely have ventured to detail it, except to shew its perfect accordance with the diagram c . — (W.) 1 o "Ar chu prent on la largece done aive. sens paseir." " Par ce moyen prend-on la largeur d'une eau sans la passer." "Thus is measured the breadth of a stream without crossing it." " Some object being chosen on the far bank of the stream, the operation is performed on the near bank by an instrument composed of two rulers, and two transverse and parallel bars which lie beneath them ; each ruler is in turn fixed to the bars with its edge directed towards the chosen object, by taking a sight along that edge. " The whole frame is then transported to the nearest convenient field, and the position of the point at which the directions of the edges intersect ascertained and marked. This point will be at the same distance from the instrument as the chosen object was in the first position of the apparatus, therefore by direct measure- ment that distance will be obtained. Apart from the rudeness of the operation, it has the disadvantage of deducing long distances from a very short base." — (L.) 13. " Ar chu prent om la largece done fenestre ki est Ions." " Par ce moyen, prend-on la largeur d'une fenetre qui est eloignee." " Thus is measured the breadth of a distant window, or other opening." The same instrument is employed which was used in the previous operation, * Quicherat abandons this diagram as inexplicable. Lassus says that "the method indicated consists simply in verifying the fact that the angles of the quadrilateral figure which forms the trace of the plan are at equal distances from its centre, in which case the quadrilateral will be a square. After having fixed the length and the position of one of the sides of the cloister, its centre must be determined, if not already fixed. In the next place, after tracing the three other sides at right angles, the process is veri- fied by trying if the four semidiagonals from the centre are equal. The determination of the directions of the sides and position of the centre are obtained by laying down perpendiculars in a manner which Wilars de Honecort explains below. — (L.)" EXPLANATION OF THE PLATES. 129 but the two rulers are fixed parallel to each other. The edge of one being directed to the left-hand side of the distant opening to be measured, and the edge of the other to its right-hand side, the distance between the rulers will be equal to the distance between the two sides of the opening. To prevent mistakes con- cerning the sides of the rulers that have been employed in taking the sights, Wilars de Honecort is careful to shew that each ruler has but one straight edge.— (L.) 14. " Ar chu assiet am les .tiii. coens don clostre sens plonc es sens linel." " Par ce moyen asseoit-on les quatre coins d'un cloitre sans plomb et sans ligne." " Thus may be set the four corner-stones of a cloister without plummet or level*." The four objects delineated seem to represent the four stones on a greatly exaggerated scale. Each stone is dressed on its upper and lower face, and worked square on the two sides only which are turned towards the ambulatory, the other two sides are left rough to be built into the thickness of the wall. A diagonal line is marked on the upper surface of each, in which the whole artifice consists. Four posts are indicated by the black points on the outside of the diagram. These must have been set in position at the outer angles of the square traced by the previous operation described in No. 11, and the centre of the square is marked by a cross. The stones having been set down with their angles coinciding with the previously marked points at the corners of the ambulatory, are adjusted in position by placing the diagonal lines traced on their surface in coincidence with strings stretched diagonally across the area from the corner- posts, or perhaps by taking a sight from each stone along its diagonal line to the opposite corner - post. The operation may be verified by stretching lines or taking sights along the lateral faces of the stones from each corner to the next. The operation of placing the stones with their surfaces all at the same level, cannot be performed without some kind of levelling instrument, although if one of them were set with its surface truly level to begin with, the others might be roughly set at the same level by strings or sights from the first. Such a process seems to be hinted at in the legend. I have suggested the taking of sights, which of course would be done by the help of a straight-edge, because this diagram immediately follows one in which this method is employed. d M. Quicherat reads the legend, "assiet om veau." In my copy of the manuscript I find, "assiet sens plonc et sens livel sans plomb et sans ni- om sens plons et sens livel." S 130 EXPLANATION OF THE PLATES. Both Quicherat and Lassus suppose that the objects at each corner, which I have viewed as the actual corner-stones, are boards carefully squared, and having the diagonal line drawn upon their surfaces ; the latter ingeniously explains the object of the process to be to determine the four corner-points of the square area without any other instruments than these boards : — " A line representing one side of the required area being laid down upon the ground, one of these boards is placed at each end, with its angle A coinciding with \/ the extremity, and its side A C with the direction, of the line. The other two boards are carried to the opposite side, and each one is placed in such a position that its side A B will coincide with the prolongation of the side A B of the previously fixed opposite board, E and also that its diagonal line A D will coincide with the similar line of the other previously fixed board at the opposite end of the diagonal. The rectangle thus obtained will be a true square ; the four points marked as the prolongation of the diagonals at equal distances from the corners of the boards merely mark the angles of the ambulatories. — (L.)" Quicherat well observes that such proceed- ings as we are now considering must lead to inaccurate results, but that the want of precision in the directions and positions of the parts of ancient build- ings shew that such loose methods must have been employed in setting them out e .— (W.) Figure 11. 15. " Ar chu partis om one pirre que les ji. moities sont a queres." " Par ce moyen divise-t-on une pierre de telles facon que ses deux moities soient carrees." " How to divide a stone so that its two halves shall each be square." The figure itself shews the real meaning of the problem to be, " How to divide one square slab into two square slabs f ." The corners of the original slab being cut off in the direction of lines which join diagonally the middle points of its sides, a smaller square is obtained ; if the four triangular pieces which were cut off be put together with their right angles e Re VUe> p . 1G7. mode of doubling a square, >hich being contained ' This is a deduction from the problem given by in this author, is shewn to belong to the archi- Vitruvius, b. ix. c. 1, under the name of Plato's tects.— (W.) EXPLANATION OF THE PLATES. 131 in contact, a second square will be produced exactly equal to the first. In the figure, the diagonals of the inner square are merely drawn to make this result evident. — (L.) 16. "Ar chu tor torn le vis don persoir." "Par ce moyen tourne-t-on la vis d'un pressoir." " Thus is turned (or carved) the screw of a press." The circle is the plan of a wooden cylinder which has been prepared in the turning lathe, and the diagram is intended to explain the method of tracing the spiral line which is to guide the workman in hollowing out the groove which is to convert the cylinder into a screw. A ruler is notched at equal distances corre- sponding to the pitch of the intended screw, that is to say, to the distance between the turns of the spiral. Three lines are drawn on the surface of the cylinder parallel to its axis, and at three equidistant points of its circumference indicated in the figure. These lines are to be divided into equal spaces by the help of the saw-shaped ruler, observing that the divisions on the lines are not all at the same level, but on each raised a space of one-third of the pitch higher than on the one next behind it. The points thus obtained on the surface of the cylinder lie in the path of a regular spiral, and serve to guide the workman in winding the string shewn in the figure regularly round its surface. By help of this string a continuous line is then traced, and a groove subsequently sunk by means of carving tools so as to complete the screw. This is a well-known process, and is even now partly resorted to in some cases for the production of original screws g . The figure, like all the others in this page, is a mere memorandum for the use of a person who had seen the process, and shews how much remains to be supplied in the attempt to explain the other diagrams. 17. " Ar chu fait om .n. vassias. que li ons tient .n. tans que li atres." " Par ce, fait on deux vaisseaux tels que Tun tienne deux fois autant que 1' autre." " HoiD to make two vessels so that one shall hold twice as much as the other" The vessels must be supposed cylindrical, and the circle in the diagram to be the plan of the smaller of the two ; a mason's square is carried round the outside * Vide Holtzapffel's "Mechanical Manipulation," vol. ii. p. 580. s 2 132 EXPLANATION OF THE PLATES. of the vessel, with its two branches in contact with it, while a pencil in the angle describes a larger circle. The area of this large circle can be easily shewn to be double that of the small one h , because its radius is the hypothenuse of a right- angled triangle whose smaller sides are each equal to the radius of the small circle. If the two vessels are the same height, the one will have double the capacity of the other 1 . This principle will not apply to vessels of other forms, as spherical or hemispherical, because the capacities of similar figures vary as the cubes of their similar lines. The explanations given by Quicherat, and after him by Lassus, are in principle the same as the above, but neither of them have noticed the marks on the area of the circle, and beyond it to the left of its centre, the explanation of which appears to be as follows. If the two vessels are to be exactly similar in form, instead of being of the same height, the cube of the radius of the large vessel must be double the cube of the smaller radius ; therefore the large radius is to the small radius in the pro- portion of the cube root of two to unity. This is very nearly in the proportion of one and a-quarter to one j . If, therefore, we increase the radius of the small vessel by a quarter, we obtain the large radius ; and to do this is the object of the marks on the left of the centre. Of the two marks within the circle, the first bisects the radius, the second bisects the half so obtained, and gives the quarter. This quarter is transferred to the outside, and the arc drawn concentric to the smaller circle is the resulting circumference of the larger vessel. — (W.) IS, 19. " Ar chu tail on vosure riuleie k ." " Par ce moyen taille-t-on une voussure reglee." " How to cut a voussoir according to rule." The plan (No. 18) is that of an ordinary mediaeval window with the jambs splayed inwards. The figure (No. 19) on the right hand shews the heads of two arch-stones or voussoirs, of which possibly the left-hand one may belong to the external arch of the window-head, the other to the inward or scoinson arch, which is of greater span. These are laid on a flat surface at distances from a h For as the areas of circles are in the proportion proportions, of the squares of their radii, the square of the large i The real ratio is 1.26 to 1, which differs from radius must be double the square of the small one. one and a-quarter only by one-hundredth part of the 1 This will also be the case if the vessels be frusta radius, of cones and of the same height, provided their bases k Rieule : exact, soumis a la regie, regulier. and their tops bear respectively the above-described (Roquefort). EXPLANATION OF THE PLATES. 133 common point (not marked in the drawing), respectively equal to the radii of the arches to which they belong. Thus the two upper sides or joint lines in the drawing are in one straight line, shewn by the cord stretched from one to the other. The same cord if fixed to the centre point, and applied to the lower joint-lines, at distances from the upper, respectively equal to the breadth of the soffit of a voussoir, will enable those lines to be drawn at the proper angle ; and a tracing-point attached to it would enable the curves of the upper and lower surfaces to be drawn. It must be observed that whether an arch be large or small, the breadth of the voussoir will be the same in mediaeval architecture. In these drawings the breadth taken on the soffit-line is marked at four divisions in each, to shew that the two soffits are equal. This is in substance the explanation given by Lassus, excepting that he has represented the divisions on the soffit-lines to be unequal in the two figures, and proportional to the radii of the respective soffits. If the process were intended for a conical vault constructed of long voussoirs extending from the outer to the inner face of the wall, the same disposition would give the tracing of the outer and inner head of each voussoir. But the mold in fig. 19 which is farthest from the centre would be proportionally larger than the other, and the two lower joints would coincide in direction as well as the upper. — (VV.) t PLXXXIX lel>4 4 Vur * agUrfw aggfl* o^fo^ par chuvvX cm cnr ckf dd quint yojtif ffflchu fait otv on p pa« clnv T4.it oft vofovf par eftbovVW Wvr to ox" tattle kf mc\ef\ p/u 4 cW t^ul om tow lv 'ijAJite EXPLANATION OF THE PLATES. 135 PLATE XXXIX. VERSO OF THE TWENTIETH LEAF. As this leaf and the following contain diagrams in continuation of the same subject as those on the last leaf, I have continued the numbering of them for the sake of reference '. 20. " Par chu tail om pendans riules, metes le bas el haut." "Par ce moyen taille-t-on pendants regies : mettez le haut en bas." "In this tea?/ are cut pendants according to rule. Turn the figure upside down." Upon this diagram M. Quicherat remarks that " pendant" is the name given to the voussoirs that compose the vaulting surfaces that rest upon the ribs of a me- diaeval vault, and that these being so small and so thin, there is no need to make them with curved soffits like the voussoirs of a rib or arch, but their soffits may be flat, (and, still more, their extrados, or outer surface, which is out of sight™). The angles of their joints require as little care, for the mason can provide for them at the moment of setting by a few strokes of the axe or the addition of a little mortar. But Lassus sees in the diagram an elaborate method of cutting the surfaces of the joints to their required angles, which I shall give in his own words in a note n . 20 21 24 25 26 27 28 29 30 31 32 Tliis table shews the order of the numbers of reference employed in the explana- tions of the diagram, in this plate. m Frezier (Traite de Stereotomie, 1. iv. part ii. c. 1) has some remarks on the "Voutes d' Aretes Go- thiques" to the same purpose, in which he gives the term Pandantif to their vault surfaces, and Pandans to the voussoirs of which they are built : — " On se con- tente ordinairement de les faire de petites pierres, sans coupe, qu'ou appelle Pandans, pour lesquelles le mortier mis un peu plus epais a lextrados qu'a la doelle fait l'office de la coupe d'un voussoir." — (W.) n "When the figure is inverted, in accordance with the legend, we recognise the plan of the head of a voussoir, on which has been traced parallel to each of the joints (A C, A' C) a line (A' B and A' B') from the junction of the opposite joint with the soffit. These lines intersect, and leave between the points B B' where they meet the upper surface a distance B B', which is greater or less according as the angle of the joints is more or less obtuse. (Throughout this explanation the width A A' of the soffit is supposed to be the same in all 136 EXPLANATION OF THE PLATES. The drawing may possibly represent the head of a voussoir, upon which is placed a triangular pattern which serves as a simple bevel to trace the angles which the joints make with the soffit. This pattern is shewn in two positions, namely, as applied first to one joint, and, secondly, to the other. As in these small voussoirs the upper surface or extrados is flat, the use of such an instrument is practicable, and the curvature of each course is so small that the angle of the joints would not vary sensibly, although the width of the soffit might be dif- ferent in the different voussoirs ; for this kind of vaulting does not require precision, from its thinness, and the fact that the mortar fills up the errors of the joints °.— (W.) the voussoirs, and also the thickness (M X) of them all to be constant.) Now the several courses of voussoirs in this class of vaults differ in curvature, one representing an arch of greater radius, and an- other of lesser radius, according to the shape given to the vaulting surface by the ribs upon which it rests. As the two joints of each voussoir converge to the centre of the circular arc to which it belongs, it fol- lows that the angle between the joints, and conse- quently the distance B B', will be the same throughout eacli course, but will varyin thedifferent courses. More- over, the breadth of the lower soffit will be the same in all the courses, as well as the height of the voussoir, but the breadth of the upper surface, or extrados, will vary from one course to another. The master- mason will, therefore, after having traced the work- ing drawing of his vault, and the section of each course of the voussoirs of the vaulting surface, deduce the distance BB' from each course, and give it to the stone-cutter expressed, for example, in inches. The latter will mark, on a block of stone already roughed out to the proper dimensions, the soffit A A' and the centre M of the extrados, and set off the given dis- tance B B', join B A' and B'A, and, finally, draw C A and C'A' respectively parallel to the latter." The only objection to this elaborate process appears to be its total uselessness, for as the width of the soffit is assumed to be the same throughout each course, and the width of the extrados C C also the same in any one course, it is much simpler to set off the latter width at once and join CA C'A', than to begin with the small interval BB'. But as the curvature of these courses is very small, the angle of the joints would not vary sensibly, suppos- ing the soffit not to be exactly of the same breadth in every voussoir. The method indicated by Lassus might have been employed to enable the angle to be set off by workmen who had no knowledge of the bevel, but in such cases the base C B' of the triangle would probably have been the distance chosen, and not the interval B B', which is too small for prac- tical use. The late M. von Lassaulx, of Coblentz, was the first to direct attention to the construction of these pendentive surfaces of the mediaeval vaults. His memoir in Crelles' Baujournal, Berlin, 1829, was translated and inserted in the Journal of the Royal Institution, (vol. i. p. 224, London, 1831.) He shewed that the slight curvature of their surfaces and their irregularity, proved them to have been laid free-handed, without centering, as each course forms a small arch of itself, and will stand as soon as it is completed. The adhesion of each small voussoir to the layer of mortar is sufficient to prevent the sliding of the stone before the termination of the course, or if not, may be assisted by some rough contrivance. M. Viollet-le-Duc has lately given, in his excellent "Dictionary of Architecture," (t. iv. p. 105,) the re- sults of his observations upon this subject, and ex- plained the simple method which appears to have been used in the Isle of France and Champagne, and which he has restored and employed. He, like M. von Lassaulx, states that he has found no difficulty in introducing the method into practice. An adroit mason, assisted by a boy to supply him with the stones and mortar, is able to complete the vaulting surface without centres or any other apparatus than his axe and a wooden sweep cut to the required curvature. — (W.) I EXPLANATION OF THE PLATES. 137 21. " En si prendes one roonde. en on aglc sen arez le grose." " Ainsi prenez une rondeur dans un angle et vous en aurez la dimension." "By thus placing any round thing in a nook, we obtain its diameter." The figure explains itself. It may be supposed to represent a column placed in a nook, but the same mode of measuring diameters is applicable to any loose cylinder set up for the occasion in a square corner of a room, or laid down between the floor and the wall. A mason's square has one leg in contact with one of the flat surfaces, and the other with the cylinder. The perpendicular distance between the extremity of the latter leg and the flat surface opposite to it, and to which it is necessarily parallel, is equal to the diameter of the cylinder. 22, 23. " Par chu fait on one clef del tiirc. et justice one scere." "Par ce moyen fait-on une clef de tiers-point, et vciifie-t-on un trait d'equerre." " Thus may be traced the key-stone of an arch of the third point, and a square be adjusted." These two clauses I consider to be independent of each other; the last belongs to the upper figure, No. 22, the first to the lower figure, No. 23. The adjustment of the square must be first explained, and the discussion of the other figure then taken in conjunction with the similar figure, No. 24. The first appears to be intended for the following method of describing a right angle, and is here introduced for the purpose of verifying a mason's square. Fix the compasses to a convenient distance, and with that first trace a circle, and then upon its circumference set off four points in succession, of which only the first, second, and fourth should be strongly marked. Manifestly the first and fourth are at the extremities of a diameter, and as every angle in a semicircle is a right angle, it follows that if the square be laid down on the board with its angle in contact with the third or intermediate point, and one of its legs in contact with the first point, the other leg will be found in contact with the fourth point if the square be a true one. In the diagram the three points are carelessly marked at nearly equal distances, but this kind of inaccuracy is visible in all the geometrical figures of the manuscript. — (W.) T 138 EXPLANATION OF THE PLATES. 24. " Par elm tail on one clef del quint point." " Par ce moyen taille-t-on une clef de quint-point." " Thus mat/ be traced the key-stone of an arch of the fifth point!' Before these diagrams can be understood, it is necessary to determine the exact meaning of the terms ' arch of the third point,' and of the ' fifth point.' The terms tiers-point, quint-point, &c, in French, and the similar ones, 'arch of the third point,' or ' fourth point' in English, and terzo acuto and quarto acuto in Italian, belong to the ages when pointed architecture was practised, and have descended to us with no very clear definitions of their meaning. In France, the term tiers-point is now used by workmen, as Qnicherat says, for the apex of an equilateral triangle, and hence is also applied to the equilateral Gothic arch, namely, that in which the centre of each side coincides with the opposite spring- ing ; and he goes on to conclude that in the middle ages this term was a general one for pointed arches. We shall also see that Lassus was in doubt as to the original sense of the word. In England, Sir Henry Wotton is the earliest writer who alludes to these terms. In his "Elements of Architecture," 1624, he says : — " As for those Arches which our Artizans call of the third and fourth point ; And the Tuscan writers cli terzo and di quarto acuto, because they alwayes concurre in an acute Angle, and doe spring from diuision of the Diameter, into three, foure, or more parts at pleasure ; I say, such as these, both for the naturall imbecility of the sharpe Angle it selfe, and likewise for their very Vncomelinesse, ought to bee exiled from judicious eyes, and left to their first inuentors, the Got/tes or Lumbards, amongst other Reliques of that barbarous Age." — (p. 51.) This passage leaves us in ambiguity with respect to the exact application of the division of the diameter, although it clearly shews the origin of the names. The most explicit account of these constructions that I have met with is in Philibert de Lorme's Nuuvefles inventions pour bien bastir, Paris, 1578. The object of this book is to explain the construction of a roof of his invention, the frame of which is wholly constructed of very short pieces of timber. Its outer surface is cylindrical, and therefore difficult to cover with flat tiles or slates, especially in the case of small spans. To facilitate the employment of these materials, he recommends a pointed arch to be used instead of the semicircular arch, and teaches " comme Ton peut faire couvertures de diverses montees taut de l'hemicycle que du tiers poind et autres . . ." In the chapter so headed, he first shews how to manage the semi- EXPLANATION OF THE PLATES. 139 circular covering by employing half slates, but states that those who prefer whole slates or tiles may reduce the curvature by using the tiers-point, — which may be seen in the window-frames of " modern" churches, — to describe which the span must be divided into three equal parts, of which two are to be taken for the radius of the arch, and the compasses set with one point on one of the divisions, and the other at the extremity of the span-line, and so on. But if the roof is to be made higher and with less curvature, then the span may be divided into four parts, and three of these taken for the radius. Or if it is to be made as high as some carpenters are accustomed to form their roofs, then the radius must be made equal to the span, so that if straight lines be drawn from the extremities of the span to the apex of the roof, they will form, together with the span-line, an equilateral triangle. The original chapter is so diffuse, that I have been compelled to abridge it °. This passage, which has escaped the notice of the French editors of Wilars de Ilonecort, supplies a clear description of the arc en tiers-point. As in the title of the chapter he mentions in order, the semicircle, the third point, and others, we may fairly assume that by others he means fourth point, fifth point, &c, and the equilateral, confining, that is to say, the term tiers-point to the arch in which the centre is obtained by a division into three, and not using it as a general term for pointed arches. The four examples which he has given explicitly, are the semi- circle, third point, fourth point, and equilateral, which form a series of the same span, but gradually rising in height p . It is curious to find that Lassus, in commenting upon the diagrams in question ° " Qui voudra ne faut que tirer la montee au lieu plus droict, il ne faut que diviser la largeur de l'ceuvre d'un hemicycle ou demy rond, et la en quatre parts, et en prendre les trois pour tirer la Bon conseil et c ■ ■ _ , . . di S ne de noter a lair e en tiers poinct amsi que vous montee Ou si vous voulez encores faire vostre fngenieux! 12 *' voiez les formes des vitres aux Eglises ceuvre d'aussi grande hauteur comme out de cou- modernes. Comme quoy, au lieu que stume aucuns charpentiers, ainsi le pouuez faire. Faut l'liemicycle se prend d'un centre, ces fapons icy se prendre avec le compas la largeur de tout le basti- prennent de deux : ainsi que pouuez coignoistre par merit et vous voirrez la forme d'vne haute la figure ensuiuante, en laquelle le lieu inarque C, de couuerture, qui est aussi large que haute par ses toute sa largeur se diuise en trois parties egales, des- courbes : et seroit uu triangle equilateral, qui le quelles faut prendre les deux, et mettre la poincte du voudroit tirer a ligne droicte par les costez."— (p. 33.) compas sur vn des centres, et l'autre sur l'extremite p Yet Delorme, in liis "Architecture," p. 110, ap- ae la largeur, et en faire la circonference. Apres vous pears to use, in the following passage, the term tiers remucrez ledict compas et le mettrez en l'autre centre, poinct in its general sense for all pointed arches : — et en ferez autant pour l'autre coste, et voirrez la " .... on faict . . lesdictes branches d'ogives plus montee qui sera beaucoup plus haute que le demy haute que l'hemicycle et d'une circonference que les rond. Mais il faudroit auoir deux centres (ainsi que ouvriers appellent a tiers poinct, et de hauteur plus nous auons diet) pour changer la poincte dudict compas ou moins, a la volonte de l'ouvrier : elles se t ii cut a faire telle circonference des deux costez: Si de deux centres, au lieu que riiemicycle ne se tire voulez les couuertures plus hautes et que le comble soit que d'vn." 140 EXPLANATION OF THE PLATES. (p. 156), lias decided that the term tiers-point belongs to the equilateral arch. A few pages farther on, however, (at p. 165,) having in the meantime met with a passage" 1 in Dubreuil's Cours de perspective, 1642, which defines this arch in the same manner as Delorme, he declares himself convinced that his former appli- cation of the term must be abandoned. The exact meaning of the terms in question was soon forgotten after the disuse of Gothic architecture. In Italy, Viola (in 1629, p. 234) defines the terzo acuto to be the equilateral arch, because it contains an equilateral triangle, and his quarto acuto and quinto acuto are pointed arches, which are respectively described about a square and a pentagon. In the former the lower side of the square coincides with the span-line, the lower angles of it are the centre points of the arch, whose radius is the diagonal. Thus, his quarto acuto becomes very nearly our fourth point, but is rather higher in proportion to its span. In his quinto acuto the lower side of the pentagon coincides with the span-line, and the centre points are, as in the former cases, at the extremities of the side, and the radius is the line which joins the centre point to one of the two opposite angles. This arch is also very nearly the fourth point, but is rather lower than it. I have had frequent occasion to observe, in measuring English examples of arches, that the centre points are, in many cases, really so placed as to corre- spond with a division of the diameter into equal parts. At Cambridge, for example, the pier-arches of the nave of Jesus College Chapel and of Cherry Hinton Church are exact arches of the third point, and the tower-arches of the above chapel are of the fourth point. The pier-arches of St. Michael's Church are of the sixth point, and the narrow arch into the tower is equilateral. I have also met with divisions that do not admit of the application of this nomenclature, such as the division of the span into seven or eight parts, of which five are taken for the radius. The ancient roof of Gonvile Hall has an arch that belongs to the latter, and the northern arches of the choir of Jesus Chapel to the former, proportion. Arches are also met with which, being more acute than the equilateral, have their centres on the outside of the arch. For . example, the centres of the soffits of the pier-arches in St. Edward's Church, Cambridge, are at a distance of exactly half the span i The passage (p. Gl, Second Edition, 1079,) re- ties egales," &c. ; proceeding so as to describe the lates merely to the arch en tiers-poinct, and is not, arch of the third point as explained iu the text. He like that of Delorme, explanatory of the other kinds concludes by saying that this arch is en tiers-point of arches, to which it does not even allude. After as well as the other, and that the one or the other having shewn the mode of tracing the arc en tiers- may be employed, but that the ancient churches poinct in the form which is more properly called the approach more to the first kind than to the second, equilateral arch, Dubreuil adds, " Le vray tiers poinct and that they have arches even more acute than est hi figure. ... On divise le diametre en trois par- the first. I EXPLANATION OF THE PLATES. 141 beyond the springing points, so that the span is two-thirds of the radius. These methods of defining the proportions of pointed arches are manifestly convenient for giving instructions to workmen or writing descriptions. In all the cases quoted above, the arch measured is that of the soffit. But it must be observed that mediaeval arches are made up of a number of concentric arches, and that every one of these is necessarily of a different proportion to the soffit. Some- times the simple form of arch that belongs to the third or fourth point will be found in one of the superior orders, or even in the hoodmold. But that the nomenclature was employed in the middle ages in the manner above described is perfectly cer- tain ; and the manuscript of Wilars de Honecort shews that its origin dates from a very early period, probably soon after the introduction of pointed forms. In his own drawings of buildings, as at Kheims and Laon, he uses the equilateral arch. The explanation of the diagrams No. 23 and 24, which shew how to cut the key-stones of an arch of the third point and of other points, appears to be as follows : let E A F be a pointed arch, of which the voussoirs of one side are shewn on an ex- aggerated scale for the sake of distinctness, and let B and D be the centres from which the halves of the arch are struck respectively. The ordinary voussoirs, as K L, have their joints directed to the centre B. But the key- Hsur.is.-A** of thorourth point, stone has only its lower joint, n m, directed to that centre; its upper joint, cA, is vertical and directed to the middle point, C, of the span. The dotted line b A, directed to the centre B, shews that the difference between a common voussoir, nmbk, and the key-stone, consists in the additional triangular-piece A b c. If, therefore, the head of a common voussoir be first delineated, and the additional triangle added, the form of the key-stone will be obtained. This triangle can be constructed as follows. If from any point s on A b, we draw s t perpendicular to Ac, the triangle Ast will be similar to the triangle ABC, for the angle at the apex A is the same in both. Now A B, the radius of the arch, is equal to B E, which is the distance of the centre point measured on the span-line, and B C is one of the parts into which the span-line is divided in order to obtain the places of the centres. The proportion of A B to B C varies according to the position of the centre points, but is always known when the name of the arch is known ; and the ratio of As to st being the same as that of A B to B C, the small triangle can be laid down when the name of the arch is given, without reference 142 EXPLANATION OF THE PLATES. to its actual dimensions. The figure represents an arch of the fourth point, in which the span is divided into four parts, and three of them taken for the radius. To describe the key-stone, first draw the head of a common voussoir, mn Ad, then set off with any convenient distance three equal parts from A to s, and from s as a centre, with the same distance strike a small arc of a circle by which a line A tc maybe drawn, which will complete the keystone kmnc. For the other denominations of arches the proportions of the triangle Ast are as easily obtained, but require some explanation. In the equilateral arch A B or B E is equal to twice B C, therefore s t is one-half of A s. In accordance with the nomenclature of the other arches, the equilateral arch might be named of the second point. In the arch of the third point, which is the first of Wilars de Honecort's ex- amples, the distance B C is half of one of the three spaces into which the span-line is divided, and C B is consequently a quarter of A B, and s t a quarter of A s. t vs. \ \ E C 14.— Equilateral arch. B E D C B Figure 15 — Arch of the Third Point. In Honecort's figure he has indicated this proportion by bi- secting As in p, and again bisecting p s in r, thus produc- ing only three points of division ; Ii « uKls - but, either by error or intentional mystification, has placed these points ai ap- parently equal distances as he has equally done in the circular diagram above, already explained. His figure appears intended for a perspective view of the key- stone when finished, as in fig. 16. In the arch of the fifth point cond example of Honecort, the B C is one and a-half of the divisions of the span, therefore A s being divided into four equal parts, the second part must be bisected to obtain the the se- distance e c Figure ];._Arch of the Fifth Poii length of s t. Thus the five strokes on the diagram given by lionecort may EXPLANATION OF THE PLATES. 143 be accounted for, or, which is more likely, the line A * in his diagram is divided into four equal parts by the marks, and the bisection meant to be left to the eye, when the compasses are opened to the distance of one and a-half r . — (W.) "Par chu tailon one clef del quint point." " Par ce moyen taille-t-on une clef de quint-point." This is a little spiral traced with the compasses in the simple and well-known manner of describing a series of increasing semicircles on the upper and under sides of a line alternately. No connection is discoverable between the spiral and the key-stone of the quint-point, and it seems merely to be an example to shew how spirals may be drawn 8 . — (W.) 26. "Par chu fait on on piler de quatre cuins venir a loison." "Par ce moyen dispose-t-on liaisons d'un pilier quadrangulaire." " TItns may be arranged the bond of a square pier." The four cuins may be either translated to mean the quoins or corners of the pier, or the four stones of which each course consists. The mode of giving bond is by disposing the joints in a slanting direction across the pier, as the plan clearly shews. The next course is manifestly intended to have its joints similarly arranged, but in the reverse direction. Thus the joints of the one course are situated over the solid parts of the other course, (or, as English workmen term it, the courses break joint.) — (L.) Otherwise, the drawing may be supposed to indicate that each course consists only of two oblong stones, which are laid so that the joint of one course shall lie across the joint of the next. — (W.) r With respect to these two diagrams of the key- stones of the tiers-point and quint-point, Quicherat merely suggests that they represent eacli key-stone developed with respect to three of its faces ; and Lassus supposes it to be seen with respect to two of its faces, its head and its joint, with which my explanation concurs. As to the marks of division, Quicherat leaves them unnoticed, and Lassus thinks that they are intended to shew that the two lines close to- gether (A s and At) are meant for one only. s M. Quicherat supposes the figure to be an ex- ample of the junction of arcs of circles of different radii, which may have been used in the description of four-centered arches, and Lassus suggests that it was a method of dividing the span of a given arch into five equal parts, in order to find the centres of the arch of the fifth point. For this purpose, however, it is useless, for it would require the span to be first divided into ten parts by some other method to find the two centre points of the semicircles. (Vide No. 40, below.)— (W.) 144 EXPLANATION OF THE PLATES. 27. " Par chu tail on vosors par esscandelon." "Par ce moyen monte-t-on des voussoirs par echelons." " How to cut voussoirs by a scale." Two lines are drawn at right angles to one of the joints of the voussoir to meet the opposite joint. The upper line is therefore necessarily longer than the lower line, yet each of them is divided into eight equal parts. It follows, therefore, that if any other line be drawn from one of the divisions of the lower scale to the corresponding division of the upper scale, this line will converge to the same point as the joints of the voussoir. If the head of a given voussoir happen to be too broad to fill up the space intended for it, which may sometimes be the case, this geometrical construction will enable a new joint to be correctly drawn at the proper width. According to this view, the object of the rule may be stated to be, " How to reduce the width of a ready-made voussoir 1 ." — (W.) 28, 29. " Par ceste raison u montom laguile done toor. et taille les moles." " Par ce moyen inonte-t-on faiguille cl'une tour et en taille-t-on les modeles." "By this rule (or proportion) we set up the spire of a tower and cut the molds, (or patterns of the stones)." The drawing (No. 28) is a diagram of a spire upon which a central line is drawn, divided into four parts, each of which is equal to the base of the spire. This indicates that the height of the spire is four times its breadth at the base, which must therefore be the rule in question. In fact, Wilars must be con- sidered as recommending this proportion as the best for a spire, and it is exactly that of the Cathedral of Bayeux, amongst others. This example (like No. 9) shews that the masons of this period did not ' The two lines may meet the joints at any convenient angle provided they be parallel. Q.ui- cherat concisely states this to be a method of " cutting voussoirs by scales, that is to say, by means of a scale of proportion established between the extrados and the soffit." This is not the case, because the scale-lines are drawn at right angles to one joint, at arbitrary points, and therefore are not in the same ratio as that of the soffit to the extrados. Lassus, on the other hand, conceives that the scales are used to enable the extrados of a voussoir to be described when the soffit and joint lines only are given. A ruler being cut equal to the height of the voussoir, and laid in turn upon each pair of corre- sponding points of the respective scales, keeping one end of it in contact with the soffit, a series of points may be set off at the other end through which the extrados may be described. — (W.) u Neither Quicherat nor Lassus have remarked upon the use of the new expression, "Par ceste raz'scm," instead of the usual " Par chu .-" they have translated the legend in the manner of the others, — "Par ce moyen, J)-c." and interpreted the process as merely shewing how to make a mold or pattern of the same inclination as that of any given spire. I EXPLANATION OF THE PLATES. 145 employ the jointed bevel, and that their angles were measured and designated not by degrees, the divisions of a circle, but by gradients, the proportion of height to length, which defines the inclination of a sloping line to a horizontal one. We should now express this as the proportion of the tangent to the radius. For the molds (No. 29) a right-angled triangle is constructed, of which one side is divided into eight parts, and one of these parts set up for the other side, and thus the inclination of the third side, or hypothenuse, becomes the same as that of the side of the spire to its base. — (W.) The figures in the lower part of the plate are somewhat entangled together, and require a little explanation. An arcade of semicircular arches resting on five single pillars is represented in an unfinished state. Two of the arches only are turned, and the last pillar on the right hand has not yet received its capital. This arcade serves to illustrate two processes, which are designated below as Nos. 30 and 31. No. 32 represents a tower, and beneath No. 30 is a figure of a man in a kneeling position, occupied in observing the height of the tower by means of an in- strument. We may now examine the nature of these three different subjects. 30. " Par chu tail ora vosure pendant." " Par ce moven taille-t-on les voussures pendantes." " How to cut the voussoirs of hanging arches" In an arcade of the kind represented in the figure it sometimes occurs that one wider opening than the rest may be required, and that this is obtained by dis- pensing with the pillar at that place, and so constructing the two arches which would have rested on it that they appear as if suspended in the air ; generally a hanging boss is added in the place where the capital of the pillar would have been, and such a one is represented in the drawing, for the rough pillar is merely a scaffold-pole temporarily placed under the boss to assist in setting the arch- stones, but intended to be removed when that operation has been completed. The artifice which enables these two arches to stand without the pillar consists in cutting their voussoirs in such a form that their joints all tend to a single point beneath the boss : thus the two arches form really only one large one, which rests upon the two lateral pillars, but has its soffit cut into the form of two arches, — a curious example of discrepancy between the mechanical and decorative con- struction. It is extremely well explained in the diagram by the string tied to a nail which is driven into the place of the centre of the real arch in the teni- u 146 EXPLANATION OF THE PLATES. porary prop. Lassus, whose explanation is substantially the same as the above, is unable to cite an example of this artifice of construction so early as the thirteenth century, but states it to be more common in the succeeding ones, especially in the fifteenth, in which the jube of St. Madeleine at Troyes offers a magnificent specimen 3 . — (W.) 31. " Par chu montom dous pilers done hautece sens plom. et sens livel." " Ainsi l'on monte deux piliers de meine hauteur sans fil - a-plomb et sans niveau." " Thus two pillars may be set up at the same height without the help of plummet or level" A most rude and primitive contrivance, which simply consists in driving a stake into the ground midway between the two pillars. The stake is provided with a rod which swings on a pin in the stake, so that it can be brought to rest against each of the pillars in turn. If it be cut to such a length as to reach to the top of the shaft already fixed, it will, when brought into contact with the second shaft, touch it at a point at the same level as the top of the former. It is hardly necessary to add that the least deviation from the exact middle point in the position of the pin upon which the rod swings would entirely vitiate the process 7 . — (W.) 32. " Par chu prentom le hautece done toor." "Par ce moyen prend-on la hauteur d'une tour." " How to take the height of a tower." A board cut in the form of a right-angled triangle of which the two sides are equal is set up upon two equal legs in the direction of the tower : the observer shifts this rude instrument along the level ground until he has succeeded in placing it at such a point that by taking a sight along its slant side, or hypothenuse, he hits the top of the tower ; he then measures the horizontal distance from the lower extremity of the hypothenuse to the tower, which will be equal to its height. The line drawn in prolongation of the base of the instrument to the tower is apparently intended to point out that the height so obtained must be measured from the point at which a sight taken along the base strikes the tower. This, taken in conjunction with the methods of measuring the breadth of a river and the width of a distant opening, already given z , forms a very curious illustration of the extreme poverty of the art of measuring heights and distances in the thirteenth century. — (W.) * Quicherat imagines this diagram to relate to the } The above interpretation is the same as that ordinary vaulting surfaces of mediaeval rib-vaults, but given by both Quicherat and Lassus. is in this instance undoubtedly mistaken. — (W.) 2 See p. 128, above. V PLXL fa^cVtt pernor cm ott T. iii. p. 331. k This discrepancy was pointed out by M. Gilbert in his history of the cathedral, p. 5 ; but he, assum- ing the tradition that Robert de Coucy was the first architect to be a fact, explains the difficulty by sup- posing that there were two of the same name, father and son, or uncle and nephew. M. Quicherat (p. 5, above) has assigned his death to the year 1241, but I know not upon what authority. — (W.) EXPLANATION OF THE PLATES. 209 order in which the parts of the cathedral were carried on ; for this we must have recourse to the building itself. From the exact descriptions given by M. Viollet- le-Duc in his admirable Dictionary of French Architecture, I gather the following summary of the structural history. The works extending from the choir to the middle of the nave 1 appear to have been continuously carried on to the height of the level of the top of the side aisles. The works above this level exhibit an abrupt change of system, shewn by a more advanced style of ornamentation, and a sudden diminution in the thickness of the walls, producing economy of materials. But, notwithstanding these changes, which indicate a pause in the works and a new architect, the original designs seem to have been respected in this upper story. Yet, although the lower part was continuously carried on, several differences of style may be detected in it, as might be expected, for it must have occupied, accord- ing to M. Viollet-le-Duc, at least eighteen years in its building. The foundations themselves must have cost many years' work, for the original soil is neither level nor firm for from four to eight yards below the surface. The superstructure is also of more than ordinary massiveness. The choir-chapels are circular in plan up to the level of the window-sills, and the entire ground-story of both the transept- gables more ancient than the upper parts of the choir-chapels. For the windows in the transepts have no monials or tracery, and are bordered with rich moldings and ornaments in the early French style. But the chapel windows, on the contrary, have early tracery, the same as that of Amiens, (c. 1230) ; and, to accom- modate this tracery, the plan of the chapels is abruptly changed from circular to polygonal at the level of the sills, it being impossible, as M. Viollet-le-Duc has pointed out, to construct tracery on a circular plan m . He states that the ornamentation of all the lower parts above defined, up to and including the cornice of the radiating chapels, denotes the work of an artist who belonged to the school which arose at the end of the twelfth century. Above this level, including the clerestory with the pinnacles and flying buttresses, the ornamentation possesses all the distinctive characteristics of the middle of the thirteenth century. The whole of the four western severeys of the nave from the ground must also have been included in this second portion of the works, the style of which shews that it could scarcely have been commenced before 1240", 1 Or, more exactly, exclusive of the four western n Diet. cCArch. Frang., t. ii. pp. 320, 321 : — "Cene compartments of the nave. fut guere qu'en 1240 que Ton continua les parties su- "' This transitional combination also occurs at Tours. perieures du choeur, que Ton commenca les premieres Diet. a" Arch. Frang., t. ii. p. 470. travees de la nef et la facade." It must be remembered e e 210 EXPLANATION OF THE PLATES. according to M. Viollet-le-Duc. Yet, as there seems no reason to doubt that the canons entered the choir in 1241, I would rather place the second period a few years earlier, for the clerestory of the choir must have been finished and roofed in, although not vaulted over the centre, at the latter date, and this agrees with the state of the works, as shewn by Honecort's drawings described below. For M. Quicherat's ingenious biography has placed the visit of our artist to Rheims about 1244, when he received his orders to go to Hungary . Comparing the documentary and structural histories, I conclude that the build- ing, commenced by the original architect in 1211, was carried on by himself and his successors until, in 1241, it had arrived at such a state as to admit of its being employed for service, the eastern portion of the choir being at that time covered in, but not vaulted over. The remainder of the work, namely, the pinnacles from the side-aisle walls upwards, with the flying buttresses and the central vaults, the clerestory of part of the nave and transepts, and the entire building of the four western compartments, with the great west front, towers, and transept gables, being, as usual, carried on piecemeal as funds could be col- lected, lingered for two centuries. The diminution of dimensions and setting back of the pinnacle-shafts, which is so clearly described by M. Viollet-le-Duc p , appears to me to have been the natural effect of the increased knowledge and ad- vanced taste of the later architects, who lived at a period when experience had shewn how to build with greater lightness, as well of material as of appearance. I would rather suppose this, than that it was the mere result of want of funds, which compelled the builders to economise materials. We may now proceed to the examination of Plate 59, which is a drawing of the interior of one of the radiating chapels of the choir. Its present state is beauti- fully exhibited by M. Viollet-le-Duc's sketch, placed opposite to it. Recollecting the fact, that in the thirteenth century no rules or principles for perspective representation had been discovered, it is really surprising to trace the fidelity of this drawing by comparing it with the leading points of the existing building, which I shall proceed to enumerate for that purpose. that although a skilful architect can derive the order and generally difficult to identify with the especial in which the parts of a building were erected and the part of the building to which they allude. The dates changes they have undergone from an examination of given in the above dictionary are very often conjec- the structure alone, and even the time probably con- turally assigned from style alone, by the author's sinned in the work, and can also shew the cotem- confession in the note at p. 292, t. ii. — (W.) porary buildings, yet the actual years in which the ° P- 6, and pi. 19 above. work was carried on must be derived from written >' Bid., vol. ii. p. 317. See pp. 224, 225, below, documents, which unfortunately are exceedingly rare, EXPLANATION OF THE PLATES. 211 The walls beneath the window-sills are on a circular plan, and decorated with blank arcades of two arches to each severey. Above the string-molding, or tablement, which surmounts the arcade, the plan of the walls is changed into a polygonal form, as already mentioned. The string-molding over the arcade is the edge of a level surface, which forms a gallery or passage in front of the sill walls of the windows ; for the continuity of which, openings are pierced through the projecting piers. These openings are indicated in the drawing by a thick black vertical stripe. The inscription, " Vesci les voies dedens et les orbes arkes," " Here is the interior passage and the blank arches," refers to the gallery, and to the arcade below ; for I have elsewhere shewn q that the term orb was applied in the middle ages to blank or blind arches and panels. It is derived from the Latin orbas. The windows are of the primitive form of two pointed lights, upon the arch-heads of which a circle rests. The jambs and the central monial have each a shaft, from the capital of which the arch-head springs without stilting, the arch itself being nearly of the form termed equilateral. The arch-head of the whole window springs high above the lights at the level of the centre of the circle. The circle has six cusps, of which two are placed on the vertical diameter. A vaulting-shaft rises from the pavement in front of the orbate arcade, and is continued on the narrow face of the pier which separates the windows. Its capital lies below the level of the capitals of the window-shafts, and bears a vault-rib. From the pavement of the gallery upwards this vault-shaft is flanked on each side by a small shaft corresponding to the window-shafts, and having its capital on their level, and consequently above that of the vaulting-shaft ; these bear the wall-ribs, and each of their capitals is connected with that of the corresponding window-shaft by a horizontal prolongation of the neck and abacus molding and of the foliage. The base-moldings of the window-shafts, however, lie rather higher than those of the wall-rib shafts, being placed on the sill-wall. Every one of the particulars above enumerated are shewn in the sketch of Wilars de Honecort", although obscured by his imperfect mode of drawing. Even the general proportions are not very different from the truth. In criticising i Architectural Nomenclature of the Middle du pourtoir" p. 207. Ages, p. 53; and Parker's Glossary, 5th ed. art. r The chapels are delineated in Gailhabaud's-ircAi- Orb. M. Lassus, unaware of this use of the word lecture, and in M. Viollet-le-Duc's Dictionary, t. ii. orb, and supposing it to refer to the circular plan p- 472, — from which copies of the two woodcuts of of the passage, has translated the above memoran- the exterior and interior have been obligingly sup- dum, "Voici les couloirs interieurs et les arches plied by the publishers. e e 2 EXPLANATION OF THE PLATES. Honecort's ingenious struggle to represent the horizontal bands which connect the window-shafts with their corresponding wall-rib shafts, it must be remarked that the piers between the windows are slightly wedge-shaped in plan, with the narrow end inwards ; so, in reality, two sides of each pier might just be seen at once by a person standing on the centre of the polygon. The differences between Wilars' sketch and the real building are in detail. The vault appears not to have been made at the time of his visit : he has shewn the lower portion only, or springing-stones, which are always built at the same time as the wall. The upper extremity appears as if three ribs were intended to spring, but as this is impossible, from the plan of the chapel, it is probably only an attempt to represent the spreading outwards of the springing at the top. The bases of the shafts are too large, and the complex base which rises from the pave- ment is altogether different in distribution and proportion from the present one, although it has the same number of members. E <1 e D c d J c c B A b A a h. ft MiHUS M. Actual base. Honecort's bine. Fig. 52. In this woodcut the profile of the actual base and Honecort's representation of it are placed side by side. Although at first sight they appear altogether different, I believe that the difference is solely produced by the coarseness of the drawing. Reckoning downwards, we have in both, first a base-mold, E, e, with its plinth, D, d. Beneath these are a second base-mold, C, c, with a plinth, B, b, resting on a sub-plinth, A, a. Honecort has given too much space to his base-molds, at the expense of his plinths, and has drawn the profiles very roughly, but yet not so as to make it impossible that they were intended to represent the existing ones. The molding c is an undulated one, and the intermediate parallel lines were not required. But as the mediaeval base-mold of the thirteenth century is made up of EXPLANATION OF THE PLATES. 213 one concave between two convex forms, we may suppose that Honecort's rule for sketching it would naturally be to draw four parallel guide lines at the propor- tional distance corresponding to the widths of the three members, and then to mark the profile across them. Honecort's mode of distributing the base is quite contrary not only to that of the chapels in question, but to the ordinary practice of his time. The compound base is employed for the great vaulting-shafts of the interior, and also for the smaller shafts of the arcade beneath the windows. Every member of it is there- fore carried horizontally round the whole interior. But Honecort has made the whole pile of moldings (with the exception of the sub-plinth A) mitre about every separate shaft. Fig. 33 — Bate-mold of the chapels at Rheims. The above sketch shews the real distribution. It represents a fragment of the lower part of the wall, including one of the compound piers in front of which the vaulting-shaft is placed, and on each side of it one of the small shafts of the arcade : the single arcade-shaft, which stands in the middle of each compartment, is also seen. It will be perceived that the upper base-mold, E, and the upper half of its plinth, D, do really mitre round each shaft and run along the wall, in the way shewn by Honecort. The lower half of D, and the second basemold and plinth, C, B, only mitre round the vaulting-shaft and run straight from that to the next, thus affording a stylobate for the arcade and giving distinctness to the vaulting-shaft. The sub-plinth, A, is carried without break beneath the whole. This is in truth the universal principle upon which base-molds are arranged, the lower members presenting fewer breaks than the upper ones. As Honecort's mode of representing the breaks consists simply in drawing the profiles across a series of parallel lines 214 EXPLANATION OF THE PLATES. previously drawn, without any attempt to shew the returns in perspective, we may suppose that he had forgotten the real arrangement when he introduced these profiles into his drawing. Such a basement as he has represented is very unlikely to have been ever actually executed s . This interior view, compared with the corresponding exterior in the following plate, curiously illustrates the conventional methods of representation, and the difficulties they involved, before the true theories of projection and perspective were worked out. All the horizontal concave lines of the interior are drawn concave upwards on the paper, as if the artist began his drawing from the top. They become less and less concave as they descend, but never horizontal, so that if the artist had stationed himself in a pit, with his head below the level of the pavement, this part of his perspective would have been true. In the exterior, on the contrary, all the horizontal convexities are drawn convex upwards, but more so at the bottom than at the top, which is not true on any supposition l . The lateral windows with their tracery, standing obliquely to the spectator, exhibit a strange confusion of lines. The circle is a true one, instead of an ellipse, as it should be ; the arch-heads above are inclined to right and left, away from the middle of the drawing in the interior, and towards it in the exterior view ; the light-heads below awkwardly distorted to fill the space between the circle and the inclined range of capitals ; and, lastly, the sides of the jambs in the interior are both of them seen in perspective, as if the spectator were opposite to each win- dow in turn, or rather as if he stood on that point of the pavement which is the centre of the polygonal plan of the apse. — (W.) s M. Lassus supposes that the base has been altered, but I have endeavoured to shew that the drawing is in fault. He also is of opinion that Wilars has omitted the vault-ribs because they would have concealed the tracery of the windows in his sketch. » It would be true for a representation of a concave, or interior, supposing the artist's eyes to be above the top of the paper. EXPLANATION OF THE PLATES. 215 APSIDAL CHAPEL OF BHEIMS CATHEDRAL- Fig. 34 — EXTERIOR VIEW. From the Diclioriaaiie de 1'ArchitecLure Francaise, par M VioUet>le-Duc, t ii. p. 4?3. EXPLANATION OF THE PLATES. 217 PLATE LX. RECTO OF THE THIRTY-FIRST LEAF, MARKED IN THE FIFTEENTH CENTURY WITH THE NUMERAL XVIII. " Et en cele autre pagene poes vus veir les raontees des capieles de le glise de Rains par de hois, tres le comencement desci en le fin ensi com eles sunt, dautretel maniere doivent estre celes de Canbrai son lor fait droit, li daerrains entauleruens doit I'aire cretiaus." " En cette autre page vous pouvez voir les elevations exterieures des chapelles de l'eglise de Reims, ainsi qu'elles sont depuis la base jusqu'au sommet. De cette maniere doivent etre celles de Cambrai si on les construit. Le dernier entablement doit former des creteaux." "In the next page you may see the elevations of the chapels of the Church of Rheims on the outside, from the beginning to the end, just as they are. In the same manner will be those of Cambray if they are rightly made. The upper tablement (or entablature) must have merlons." By comparing this exterior view with M. Viollet-le-Duc's accurate sketch of the chapel, it will be seen that the principal features are delineated with tolerable fidelity. It shews the circular form below the window-sills and the polygonal above. In the reality, a huge buttress for the support of the flying buttresses of the clerestory is introduced between each apsidal chapel, and this buttress is omitted by our artist on both sides of his drawing. Had he not inserted the base and capital of a lateral window-shaft beyond each of his outer buttresses, the double outline of the latter might very well have been intended to shew the face of the great buttress, so that it is probable that the facings of the great buttresses were not completed at the time of his visit, and that he finished them subsequently in his sketch in imitation of the others 1 . In the tracery of the window the circle is truly represented as having its bowtell molding completely detached from that which circumscribes the arch-head of the window, but mitred with those of the light-heads. He has also marked the joints of the masonry in the tracery of one of these windows. The angels with outspread wings still stand on the pentagonal abacus of a short pentagonal pedestal, as in the drawing, and over their heads is an insignificant canopy, not very different from that represented". But these angels are clothed in a long robe, of which t Vide Plate 63 and its explanation. the different chapels. The lower stage of the but- These little canopies are variously arranged in tresses below the windows is proportionally too high, F f 218 EXPLANATION OF THE PLATES. Wilars has completely denuded them, and has made their wings much too long. He has also omitted the hoodmold of the windows with its flower knobs, and the carved flowers in the casement-molding of the tablement which tops the wall, and is itself crowned with a battlement of cretiaus or merlons v , as Wilars has carefully directed. In providing similar members for the walls of the nave and side-aisle, (in Plates 61 and 62,) he says, " There must be merlons on the tablement to provide a passage round about it in case of fire." M. Lassus states that this contrivance still remains on the apsidal chapels and transepts, but is surmounted by a high stone balustrade, erected when the works were resumed about 1240 x . A leaden gutter conceals the original finish of the walls of the older compart- ments of the nave. He explains that the merlons were flat on the top, so as to furnish stepping-stones along the wall, and especially in front of the buttresses, while the intermediate embrasures of the battlement were sloped downwards, so as to throw off the water. To do this more completely, the hinder part of the merlon was rounded off, as the annexed plan (fig. 36) shews. By comparing it with the accompanying section, (fig. 35,) the arrangement will be completely understood 7 . 0,24 «-o,iS-H o,6o > l'ig. 36.— Fl.iu ..f the Wilars' drawing represents the chapel with a high-pitched pyramidal roof, the angles of which are garnished with a crest of crockets. They are at present roofed with low isolated pyramids, covered with lead and concealed by the high balustrade. — (W.) and the base-mold very coarsely drawn, the interval between the projecting plinths being less than their width, instead of being one-half greater. I presume the molding which caps the plinth to vary in dif- ferent chapels, for Gailhabaud's artist draws it in the same form as Wilars, and M. Viollet-le-Duc sub- stitutes a double chamfer in the annexed view. T The rising parts of battlements are in English termed merlons, and the intermediate spaces em- * It is well shewn in the plan of an apsidal chapel with part of the transept, in Gailhabaud. i This construction of the merlons of Rheims is described by M. Lassus as above, and also by M. Viollet-le-Duc in his Dictionary, t. ii. p. 317, where he inadvertently states that " Villart de Honnecourt terms them carmaux." In t. iv. p. 33, art. Comiche, he gives them their proper name of cretiaus, referring to the French edition of the Album which had just appeared. EXPLANATION OP THE PLATES. 219 PLATE LXI. VERSO OF THE THIRTY-FIRST LEAF. The legend which properly belongs to this page is written at the bottom of the page which faces it in the manuscript, and is the recto of the thirty-second leaf. But for the better illus- tration of the drawings I shall place it here, together with the lateral inscription on the margin of the present page, which is also partly written on the opposite one. " Vesci les montees de le glise dc Rains et del plain pen. dedens et dehors. Li premiers entaulemens des acaintes doit faire cretians si quil puist avoir voie devant le covertic. encontre ce covertic sunt les voies dedens. Et quant ces voies sunt volses et entaulees. adont revienent les voies dehors con puet aler devant les suels des veneres. En lentaulenient daerrain z doit avoir cretiaus con puist aler devant le covertic. Ves aluec les manieres de totes les montees." " Entendez bien a ces montees. devaunt le covertiz des acaintes doit aver voie. sur lentaulenient. et de sur le combe des acaintes redoit aver voie. devant les verreres et un bas creteus si cume vos veez. en le purtraiture devant vos. et sur le mors de vos piliers doit aver angeles. et devant ars buteret. Par devant le grant comble en haut redoit aver voies. et creteus desur lentaulenient ken i puit aler pur peril de fiu. et en lentaulenient ait des nokeres por leve getir. — pur les capeles le vos di." " Voici les elevations de l'eglise de Eeims et des murailles en dedans et en dehors. Le premier entablement des bas cotes doit faire creteaux, afin qu'il puisse exister une voie devant la cou- verture. Au niveau de cette couverture sont les galeries interieures. Quand ces galeries sont voutees et entablees, on retrouve la galerie exterieure qui permet de circuler devant le seuil des verrieres. Le dernier entablement doit etre a creneaux pour que l'on puisse aller devant la couverture. Voyez la la facon de toutes les elevations." " Remarquez bien ces elevations. Devant la couverture des bas cotes il doit y avoir une voie sur l'entablement, et il doit y en avoir une nouvelle sur le comble de ces bas cotes devant les verrieres, avec des creteaux bas, coinme vous le voyez en l'image devant vous. A l'amortisse- ment de vos contre-forts il doit y avoir des anges et par devant des arcs-boutants. Devant le grand comble du haut il doit y avoir des voies et des creteaux sur l'entablement, pour circuler lorsqu'il y a danger du feu. II doit y avoir aussi sur l'entablement des cheneaux pour deverser l'eau. Je vous le dis encore pour les chapelles." 1 Daarrain, Dernier, qui est apres tous les autres. — {Roquefort.) F I 2 220 EXPLANATION OF THE PLATES. "These are the elevations of the Church of Rheims, and of the fiat (aisle) wall within and without. The first tablement of the side aisles must have merlons, so that there maybe a gangway in front of the roof. Over against this roof are the inner passages (or triforium gallery). And above the vault and tablement of these passages we find the outer gangway in front of the window sills. The upper tablement must have merlons, so that there may be a passage in front of the roof. Behold the fashion of all the elevations." " Consider these elevations carefully. In front of the roof of the side aisles there must be a gangivay on the tablement, and on the top of the roof of the side aisles another in front of the windows, with low merlons, as you see them in the picture before you. On the caps (mors a ) of the pilasters (piliers) must be angels, and in front flying buttresses (ars boteret). In front of the great roof above there must be gangways arid merlons on the entablature to provide a passage in case of fire, also gutters or spouts (nokeres) to throw off the water. — I say the same for the chapels b ." This pair of elevations represent one severey, externally and internally, of the nave, or choir, of Rheims, and are extremely interesting for the evidence of the antiquity of the similar method so commonly followed by ourselves in representing such buildings. They shew also the extreme carelessness or neglect of true proportion in the delineation of architecture in this volume. For although the two drawings shew two sides of the same wall, the outer view is five and a-half times higher than the breadth of the severey, and the inner view seven times higher than its breadth. The former is in fact correct, the latter consequently much too narrow. The width of the piers, which in reality is half the distance between them, is shewn rather too small ; the height of the triforium compartment greatly ex- aggerated from one-seventh to little more than one-fiftl] of the total height of the severey. Notwithstanding this, the width of the severey is so much contracted, that the windows remain far too narrow, and the sill-wall of the side-aisle too high. The latter is ornamented with an arcade, of which no traces are to be found in the real building. The pier-arches are most incorrectly drawn in the simple equi- lateral form, and consequently ridiculously disproportionate to the piers upon which they rest, which are made four times as high as the arches. The real pier-arches are actually of the equilateral form, but they are stilted so as to raise their springing line so high above the abacus as to make the total decorative height of the arch about half that of the pier, which is a very agree- able proportion °. » mors, the sloping cap of a buttress which, in the c The word pier here includes the capital and English phrase, dies against the wall. base. I have derived these measures from the en- b This latter remark is written on the side of gravings of Gailhabaud. the opposite page in continuation of the inscription. EXPLANATION OF THE PLATES. 221 The central shaft of the triforial arcade is represented of greater diameter than the lateral ones. In Gailhabaud's elevation the same peculiarity is shewn, but in a much smaller degree. The vaults of the interior were evidently not built when the drawing was made. This also appears from Plate G3. The indifference of the draughtsman to the real proportions of the parts of the edifice is the most strikingly shewn by the tracery of the windows. Side by side we see the inside and the outside of the same clerestory window and of the same side-aisle window. The scale of heights in the two elevations is evidently the same. Yet the arch-heads of the window and of the lights are all drawn of the equilateral form, although the narrower proportion of the horizontal scale of the interior produces the most glaring discrepancy between the artistic effect of the two representations of the same object. I shall proceed to shew that these representations are both of them unlike the original in two essential respects. Kg. 37. Fig. 36. Fig. 39. These three diagrams exhibit the proportions of three specimens of that early tracery pattern which simply consists of a circle or hoop over two arches. Fig. 37 is the primitive form, which was employed at Amiens, in the chapels of Rheims, and in several other French cathedrals, about the year 1230. The thick black line is the bowtell or roll molding, which is the most prominent member of the tracery, and the fine lines which run parallel to its course on either side shew the forms of the glazed openings. The hoop alone is ornamented with six cusps. In this early tracery the hoop rests upon the arch-heads be- low, so that their respective roll-moldings unite and are mitred together. But its bowtell stands completely free from that of the outer arch of the window-head, which encloses the whole tracery. The lateral moldings of the hoop just touch those of the outer arch for the sake of obtaining mechanical 222 EXPLANATION OF THE PLATES. support, but the decorative effect is that of a hoop simply resting on the light- heads. In the earliest remaining specimen at Amiens, which is at the south-eastern extremity of the nave, a little crocket projects from the jamb-mold of compartment B (fig. 37) to touch and support the hoop, which is more completely free from the jambs than at Rheims. In the tracery of Salisbury cloister and chapter-house (c. 1256) the same detachment of the hoop is to be found, but the moldings are much richer and more multiplied than in the French specimens. The idea of a hoop resting on two arches is also dis- tinctly exhibited in several of the earlier portions of Salisbury Cathedral c . When the tracery principle was fully developed, the hoop was mitred to the outer arch-molds, in the same way as to the small arches below. The Sainte Chnpelle at Paris, begun 1242, and finished in 1247, is a complete example of this. Iti fig. 38 the two methods are shewn, namely, the primitive detachment of the hoop on the left-hand side of the drawing, and the arrangement in which it is united to the roll-molding of the lateral arch on the right-hand side. These two methods are seen together in the same window in the clerestory of the nave and choir at Amiens d . But at Rheims the older method is retained in all the lateral windows of the choir and nave, exactly as shewn on the left-hand side of fig. 38 ; and probably out of respect to the original designs. Fig. 39 is the representation of the lateral windows of the nave at Rheims, as given by De Honecort in the plate we are now examining, and also in the sketch he made in 1244 (PI. 19), when he received orders to go to Hungary. In this drawing the hoop mitres completely with the lateral arch-molds, agreeably to the later practice introduced into the Sainte Chapelle and other cotemporary works, but by no means in accordance with the example he had before his eyes, as shewn on the left-hand side of fig. 38. Yet the characteristic which I have pointed out is one that must have been known and observed by a professional architect. In his interior view of the chapel windows (PI. 59) he has committed Fig, 40.— Eastern gable, Salisbury Cathedral. c Oxford Glossary, Plates 231, 237- d These windows are of four lights, and of two orders of tracery. The first is the hoop on two arches, and its hoop is mitred with the arch-mold. The two arches are filled with a second order of tracery, on the same pattern, but having the hoop detached. A similar example is the chapel of St. Germain en Laye, c. 1240, as appears from the engraving in Viollet-le- Duc's Dictionary, t. ii. p. 433. EXPLANATION OF THE PLATES. 223 the same mistake, yet in his exterior view he has drawn the tracery with the hoop detached in the most exact accordance with the truth. But there is a point of the same kind which must be next examined in relation to the proportions of the arches of the tracery. In the chapel windows of Rheims, (fig. 37,) the arch of the window-head springs from the level F G of the diameter of the hoop, and therefore greatly above the impost of the small arches D, C, which form the heads of the lights. Thus a very large hoop is obtained. In the lateral windows of the nave, fig. 38, the impost line, H K, of the window- head is still considerably above that of the light-heads. But Wilars de Honecort, although he has drawn the chapel-windows in Plates 59 and 60 with perfect correctness in this respect, has in his representations of the nave-windows made the window-head spring from the same level as the light-heads, and has described its arch from the same centre points as the latter. The small circles in these figures are the places of the centre points, and in fig. 39 it will be seen that the inner arches of the light-heads and the inner line of the window-head are all equilateral. Here, again, our artist has given the arrangement of the tracery of the " Sainte Chapelle," and not that of Rheims. Its effect is to make the hoop, as well as the lateral spaces b, much smaller, and to reduce the proportional height of the tracery. It thus alters the phy- siognomy of the window in a manner that would scarcely have escaped the eye of a practical architect. These remarks shew that our artist drew details in accord- ance with the newer habits of his own time, unless the older forms were so fully developed, as in the chapel windows, that he was perforce compelled to study the exact relation of their parts. The proportions which he has given to the tracery of the nave-windows are those which were generally adopted in England. This pair of sketches of the interior and exterior compartment could not have been made from the working drawings instead of the real building, for they shew the structure in a consistently incomplete state in several places, exactly as it would naturally have been seen while in progress, but as it would never have been represented in designs. Thus, the vault in the interior is not yet built, and we see the lower or solid portion of the vault-ribs carried up, but stopping short of the abacus of the window-shaft and wall-rib, or formeret, in an unfinished manner, with the rough wall above. On the exterior, the flying buttresses and the pinnacles of the side-aisle buttresses, not being required until the great vault was in progress, are also not built, but the first course of the pinnacle is set upon the parapet, and the commencement of the narrow opening by which the water was to be conveyed from the roof is shewn, with the bases of its lateral shafts. 224 EXPLANATION OF THE PLATES. In the clerestory walls the capitals of the shafts which were to support the flying buttresses are seen, they of course being built corbel-wise into the thickness of the wall, but the shafts themselves, being of superficial masonry, are not yet erected. Statues of angels are indicated on the top, with outspread wings, like those of the apsidal chapels. But they are shewn in diagram lines only, as hieroglyphics to shew that statues were intended, but not yet set up. The battlement of the side-aisles with its stepping-stones is mitred round the buttresses, to afford a passage in front of the great pinnacle shafts. But in the existing building the arrangement is different ; for when the pinnacle shafts were built, they were made smaller than in the original design, and were narrowed and set inwards to such an extent as to leave a free passage round their outer faces at the level of the parapet. The stumps of the original pinnacles shewn in the drawing were of course taken down, as well as the embattled parapet below, when the present pinnacles were carried up after the visit of Wilars de Honecort. It is necessary to state that the view taken of this subject by M. Lassus varies materially from that which I have endeavoured to develope. His entire com- mentary upon the plate in question runs thus : — " This drawing differs in so many respects from the real building, that we can only conclude that it must have been traced, before the resumption of the works in 1241, from plans that were never executed. On the exterior, besides the merlons of the side-aisle walls, whose existence the modern leaden covering prevents us from verifying, Wilars indicates a similar crest under the clerestory windows. Of this there is now no trace, and it would have been useless as stepping-stones, because the passage itself is carried through openings in the pilaster buttresses of the clerestory wall. Honecort's drawing shews this passage, but suppresses the flying buttresses and upper termination of the side-aisle buttresses, to shew the form of the pilaster buttresses of the clerestory wall. On the caps of these buttresses, which Honecort calls piliers, he has indicated — in the drawing and in its legend — angels, where in the real building are placed human figures, like cariatides, supporting the piers of the high parapet above. Yet the inscription gives in this place a gutter and merlons to furnish a passage in case of fire, consequently no high parapet. Were the figures which Wilars took for angels a mere ornament ? or were they intended to support gurgoyles, the nofrers which were to throw off the water from the high roof to that of the side aisles? Although the present high balustraded parapet is manifestly a work of the fifteenth century, yet as the traces of a different balustrade may be detected upon the towers, which belong to the end of the twelfth century, it may be granted that a similar one was destined for / EXPLANATION OF THE PLATES. 225 the clerestory of the nave by its architect. It follows, therefore, that subse- quently to the visit of the architect of Cambray to Rheiras, the plans of the latter church were changed in several respects. One of these changes is evident in the pinnacles that rise above the side-aisle wall. They are set back so far as to leave a free passage round their outer faces, so as to render useless the merlons which still remain on the oldest part of the nave, but which would have been necessary had the pinnacles been carried up on the original scale of mag- nitude. We may direct attention to the earnest and repeated importance which Honecort attaches to the means of access and circulation round about every story of the edifice, mainly to guard against fire, like that fatal one which had, at the beginning of the century, destroyed the very cathedral which he was studying. " The principal difference which we remark in the interior elevation is certainly an important one : it consists in the arcade shewn below the side-aisle windows, of which not the slightest trace can be found either in the nave or transepts, although it exists in the apsidal chapels." — (L.) It will be seen that M. Lassus accounts for the differences between Wilars' drawing and the existing building by supposing that the artist drew from plans which were never executed, while, on the contrary, I have endeavoured to shew that some of these differences are the result of his imperfect mode of drawing, and that others relate to mere parapets and stumps of pinnacles, which were naturally removed or obliterated when the works were carried on after Hone- cort's visit. The setting of the mass of the pinnacles inwards, attributed by M. Viollet-le- Duc to economy of materials and want of funds, appears to me to have been dictated by an increased acquaintance with the action of the diagonal thrusts of the flying buttresses, which are better resisted by placing the centre of gravity of the outer buttress and pinnacle at a greater distance from the front face of the lower stages of the side-aisle buttresses. In English examples I have met with similar cases of re-erecting the pinnacle shafts in a more inward position, of which there is a very striking example on the south side of the choir at Ely. — (W.) g g / \ « t ^ t? uat*1im VC^ttfctf to22utTce1e 1ottcm£t$itf com PL..LXII . 7 c(?mr c*tf uatef fuur ttolfef ^ mtauttcf - tottf mneasesit <£^t*r Stee Want MHuctf btf tmtfiaifc ttikntaulaumt Wmt Wt atietr EXPLANATION OF THE PLATES. 227 PLATE LXII. RECTO OF THE THIRTY-SECOND LEAF, MARKED IN THE FIFTEENTH CENTURY W ITH THE NUMERAL XVI III. " Ci poes wis veir lun dcs pilers toraus de le glise de Rains, et .1. de ceus dentre .ij. capieles. et sen i a .1. del plain pen. et .1. de ceus de le nef del moustier. par tos ces pilers sunt les loizons teles coin eles doivent estre." " Ici vous pouvez voir l'un des piliers de la tour de l'eglise de Reims, et l'un de ceux d'entre deux chapelles, et il y en a un des murs de cloture et Pun de ceux de la nef de l'eglise. Les liaisons de tous ces piliers sont telles qu'elles doivent etre." " Here you see one of the great piers e of the church of Eheims, and one of those which are between every two chapels; and there is one from the plain wall, and one from the nave of the church. In all these piers the bond is as it must be." Ld CO of ui Q. Figure 41 The first plan represents one of the four great crossing- piers, in which the arrangement of the shafts is shewn with great fidelity. In fig. 41 I have given Pilers toraus ; vide Glossarial Index at the end of this volume. Gg2 228 EXPLANATION OF THE PLATES. the plan of this pier, with the addition of lines shewing the relation of the shafts to the arches and ribs above. It is a compound pier of a very complex kind, and requires examination to explain its manifest asymmetry. In the earlier mediaeval styles the adjustment of the parts of a compound pier to those of the super- structure was very closely maintained, every shaft having its corresponding rib or arch above. A crossing pier, by its position, has pier-arches springing from two of its adjacent angles, and the transverse ribs of the lofty central vaults from the other two adjacent angles. In Rheims Cathedral the transverse arches of the central vaults are of the same mold as the pier-arches, and, like them, are of two orders f . Each of the four angles of the pier is therefore provided with a sub-shaft between a pair of edge- shafts to carry the two pier-arches and two transverse ribs g , as the plan shews. The diagonal ribs (D) that descend from each of the vaults are received each upon a shaft placed between the groups already described. But, in addition to these, it must be observed that the clerestory walls which rise above the pier- arches, and against which the central vaults abut, have wall-ribs to receive those vaults, and these wall-ribs rest on a shaft which rises from the ground, together with those already described. Thus it happens that each of the two faces of the compound pier which receive the clerestory walls have four intermediate shafts, while those which receive the high vault of the crossing and the low vault of the side-aisle respectively, have only three intermediate shafts. Hence the unsymmetrical form of the pier, which has been exactly delineated by Honecort. The three marks on the vertical mid-line of his pier are apparently intended as a guide in drawing the pier edges, as shewn in my diagram ; three similar ones are placed on his horizontal line opposite the left-hand lower group. His diagonal shafts are drawn larger than in reality, but in the diagram I have followed his proportions 11 . The detailed drawings of M. Viollet-le-Duc and Gailhabaud, compared with ' The section of these arches is correctly given by oversight, for the small plan engraved by Gailhabaud, Honecort in this plate, and is the second figure in and his partial plan of the transept, concur in giving the second row of the moldings ; the third figure is the four intermediate shafts to the opposite faces, in the diagonal rib. the same way as in Honecort's drawing, and the e Vide my " Remarks on the Architecture of the analysis which I have given is sufficient to shew that Middle Ages," 1835, pp. 86, et seq. if a fourth shaft be employed to carry the wall-ribs, h Honecort's plan of the great crossing-pier is said it will naturally produce the arrangement in question, by M. Lassus to vary from the real ones only in one In the nave the wall-rib shafts spring from the point, which is, that those faces of the pier that are abacus of the pier, and not from its base, as in the garnished with four intermediate shafts are placed crossing-piers, and therefore do not enter into its opposite to each other in the drawing, and are ad- plan. — (W.) jacent in the existing piers. But this must be an EXPLANATION OF THE PLATES. 229 ITonecort's, shew that his plans of the other three piers are by no means minutely exact. The second of the first row is a respond of the side-aisle, and is true in the number of shafts, but the lesser shafts are much too small in com- parison with the larger. The plan omits the upper base-molding E (fig. 32, p. 212), probably because it is merely concentric with the shafts, but inserts two parallel zigzag lines, which are the plans of the plinths D and B, and represent the course of the intermediate molding, C. In reality, the plan-line of the upper plinth, D, is the only correct one, for the molding C and the plinth below it do not mitre round the plinth of the wall-shaft, but run parallel to the wall, so as to form a stylobate for the two wall-shafts that bound the severey on each side, as the lower plinth A is correctly shewn to do by the outer line in Honecort's drawing. The first plan in the second row is a pier of the nave, in which the attached shafts are drawn too large, and the plinths in the intermediate portions are curvilinear, and concentric with the body of the pier, instead of running straight in a diagonal direction from one shaft to the next, as in the reality. The next plan of the second row is one of the piers that separate the chapels Figure VI. from each other. It differs essentially from the real piers, as will be seen by comparing it with the above plan of one of them. 230 EXPLANATION OF THE PLATES. Seen from the side-aisle of the real edifice, as in M. Viollet-le-Duc's sketch (p. 205), the front of the pier is exactly like the nave-piers, shewing a large cylindrical body and three appended shafts. But Honecort has formed the front of his pier of a group of five shafts, the difference consisting in the substitution of two intermediate shafts in square-edged nooks for the cylindrical body of the actual pier. All this portion of the pier belongs to the vaulting of the side-aisles. The front shaft carries the transverse rib, the lateral shafts in Wilars' drawing, like the cylindrical body in the real building, carry the diagonal ribs, and the next pair of shafts the arches of communication between the aisle and chapels h . Within the chapels there are two other shafts on each side the pier, which belong to the diagonal rib of the chapel vault and to its wall arcade respectively, and are correctly shewn by the artist. The most curious part of this affair is, that the separating piers of the chapels, as shewn in the modern plan of Cambray Cathedral (p. 90, above), agree with Honecort's plan, and therefore not with those of Rheims. He may have purposely made this change in the arrangement of the pier, or he may have fallen into a mistake by trusting to slight memoranda when finishing his drawings 1 , but the reproduction of this remarkable variation, in the actual Cathedral of Cambray, greatly strengthens the conjecture that Honecort's drawings of Rheims were employed in the works of the former edifice. The system employed for the chapel piers at Rheims is not common. It is to be seen at Beauvais (1240-50) and at Westminster Abbey with some differences. But the system which Honecort has substituted is the ordinary one, and may be found in early examples, as at Noyon, Vezelay, and Bayeux, all early French buildings of the beginning of the century, and is also employed at Amienc and Cologne, and in many other cases. It was familiar to him, and this is probably another instance of his habit of trusting to memory for the completion of his drawings rather than to the object before him. The lines which represent the bonds of all these piers may or may not be cor- rect. Honecort's expression that they are — "teles com eles doivent estre" — as they must be — is a confession that he guessed at them from the joints, and implies that he did not see the working drawings. The French commentators truly state, h Marked chapel-arch in fig. 42. base-mold with only the lower plinth and sub-plinth ' It will be seen that the plaus of the plinths are upon the left-hand half; the word arcade is inad- as nearly as possible alike in Honecort's drawing and vertently written upon the sub-plinth that runs in in the real building. To avoid confusion of lines in front of the arches, instead of being placed between fig. 42, the complete plan of the plinths is drawn the arcade-shafts, as it ought to have been. — (W.) upon the right-hand half, and the plan of the upper EXPLANATION OF THE PLATES. 231 that as this part of the structure could only be observed in the course of repairs or demolition, they have no means of verifying them. Lassus, however, informs us that the joints of the real piers of the earlier portion of the nave are placed in the same position superficially as in Honecort's plan, so as not to be seen on the plain surfaces. The western compartments, which are later in date, are built on a different system. Each course consists of two stones, each of which carries two portions of the neighbouring attached shafts. The joint falls halfway between these, and is therefore visible on the surface of the body. The courses are set with their joints crossing each other in succession at right angles. Below the piers is placed a series of profiles of the moldings of the church, under which is written the following inscription : — " Vesci le s molle 8 des chapieles de cele pagne la devant. des formes z des veneres, des ogives z des doubliaus. z des sorvols p' de seure." " Voici les patrons des chapelles de la page la-devant, des fenetres, des meneaux, des ogives, des arcs doubleaux et des formerets par-dessus." "Here are the molds of the chapels of the former page, of the tracery and of the window lights, of the diagonal ribs, and also of the transverse ribs, and the super-arches above them." In this inscription several technical terms occur, which must be examined. In Plate 19, Honecort has described his drawing of the tracery of the side-aisle win- dows as " une des formes de Rains k ." The same word is used in English mediaeval documents for tracery, and also in French. M. Lassus follows the same interpretation, but translates verieres by meneaux. Judging by English mediaeval documents, I should rather translate this word by lights, which is applied solely to the large principal openings between the monials, and in this case would refer to the jamb-molds and monial-molds. Philibert de Lorme has recorded that the ogives are the diagonal ribs of a vault which, as they cross each other in each vaulting compartment, are also termed la croisee d 'ogives. Also, " There are other arches called doubleaux, which separate the vaults and are thicker than the others." These are the transverse ribs. In Rheims Cathedral they are unusually prominent, being formed throughout of two orders of voussoirs like the pier-arches. This is very rarely the case in France, especially so late as the thirteenth century, and never, as far as I know, in England. Now the term Sorvols is derived from Sor, super, and vols ™, or volsure, an k Architectural Nomenclature, pp. 48 and 50 ; Par- Vols is evidently a form of volsure. Thus in ker's Glossary, art. Form-pieces. the inscription at the foot of the page, we find volses 1 Architecture de Philibert del' Orme, p. 107. 1568. for vaulted. In mediaeval documents vault is often 232 EXPLANATION OF THE PLATES. arch or vault. In this case it must refer to the super-arch, or order of voussoirs which covers the arc doubleau. In the second figure of the second row we accordingly find an exact section of the transverse rib, and of the super-arch in combination with it, which also represents the section of the pier-arch. This figure may fairly be supposed to be the one described as "les molles des doubliaus et des sorvols par de seure." The mold of the ogive stands next to it. The moldings in these figures are not those of the chapel alone, for the very first compartment on the left hand is the plan of one of the windows of the side- aisles taken across the middle of the monials, or rather a plan of the jamb-molds and of the monial, for the jambs are placed too close together in proportion to the scale of the sections to form a connected plan. Every one of the sections in this table is distinguished by a peculiar mason's mark, exactly similar to those which are found on the stones of mediaeval build- ings. That these are in the present case employed as letters of reference is evident by comparing them with Plates 59 and 61, where the same characters occur, and in most instances coincide with the places assigned to the sections by comparing them with the real building. Thus in the exterior elevation of the side-aisle, Plate 61, the monial has a mark corresponding to that on the section we have just described, and the mark on the section of the right jamb is placed on the left jamb of the elevation. It is carelessly engraved in the latter, but in my own tracing is exactly the same as that of the section. The second compartment contains five separate sections, which belong to the windows of the chapel, and require a separate examination. The first figure repre- sents one of the cusps of the hoop, A, (vide fig. 37, p. 221). These large cusps in the early specimens of tracery were always constructed of separate stones of this form inserted in a groove formed in the circumscribing hoop. The remaining sections are drawn on a larger scale in fig. 43 in their real proportions, and are marked with letters of reference corresponding to those in fig. 37, to shew to which compartments they belong. A-B is a section taken from the compartment A to the compartment B, and accordingly, the groove for the reception of the cusp-piece appears on the left hand, and the rebate for the glass on the right. C-B is taken from the compartment C to the compartment B, apparently to used for an arch. M. Lassus has translated sorvol arch, and the epithet exactly represents its function, by formeret, or wall-rib. But he has made no attempt of lining or strengthening the vault or arch under to shew any connection between the term and his appli- which it stands. When placed under another arch, cation. Arc doubleau is, literally, a doubling or lining it would be translated sub-arch. EXPLANATION OF THE PLATES. 233 shew the different arrangement of the interior profiles, for the left-hand side has a fillet next to the glass rebate, and a chamfer between that and the roll-molding. But on the right-hand side the fillet is omitted ; in this respect it agrees with A-B. ArB H. WILLIS. Figure 43. D-C is taken from D to C, immediately above the capital of the monial-shaft. Both sides of this section are therefore alike, they have the fillet, and are the same as the left half of C-B. C-E is the plan of the outer arch-mold, and E P is the soffit of the inner vault of the window above the gangways. The above sections are corrected from my own tracings of the original manu- script, in which the difference between the profiles of C and B are more distinctly shewn than in the engraving. It is not unusual to give a simpler profile to the smaller openings of tracery (such as B and the other triangular spaces above and below the hoop) than to the principal compartments A, D, and C. But I am ignorant whether or no such difference exists in the moldings of the real window. It appears to me to account completely for the three sections A-B, C-B, and D-C, for otherwise the second section would be unnecessary " The marks upon these sections correspond very nearly with those upon the external elevation of the side-aisle window in Plate 61, although it represents a different window. The double cross of the first section will be found between the hoop and the triangular compartment B. The single cross upon the second section is written between the compart- ments C and B, in accordance with my explanation, but is also written above the capital of the monial, as if the section there were the same. On the other hand, the mark applied to the third section, which I have placed in the latter position, coincides with that which is given to the monial of the elevation, of which monial with the same mark a different section has already been given in the first compartment. These discrepancies can be explained by supposing that the Hh 234 EXPLANATION OF THE PLATES. The third compartment of the upper line has a section which, Lassus tells us, belongs to a gallery in the transept, and that the fourth and last is the former et, or wall-rib, of the side-aisle vaults, and the first of the second row the string-course beneath the windows of the side-aisles of the nave °. The second figure of this row, as already explained, is the transverse rib (or arc doubleau) of the whole of the vaults, and also the mold of the pier- arches, and agrees exactly with Gailhabaud's detailed sections. The third figure is a correct section of the diagonal vault-rib (or croisee d' 'ogive p ). The fourth figure is a plan of the front of the pier which separates the chapel windows from each other* 1 . The front shaft carries the vault-ribs, and the lateral shafts the wall-ribs, of which one is shewn at P, fig. 43, above. This plan is drawn looking upwards, and thus shewing the peculiar angular position and foliage-knobs of the capital of the shaft. The last figure may be intended for a plan of the same pier, looking doivnwards and shewing the plinths. The two lateral ones rest on the gangway which, as already described (p. 211 above), is carried under the window and behind the piers in question, which are thereby isolated, as in the drawing. The plinth of the front shaft is on the pavement below. The mark upon this section is placed in Plate 59 upon this plinth. — (W.) sections we are considering really apply to a lost elevation of the cliapel windows. Lastly, the mark on the fourth section, which re- presents the jamb-arch mold of the window-head, is to be found on the left-hand side of the drawing, as well as the mutilated cross which designates the outer rank of its voussoirs. Yet this in the section is ap- plied to the inner, or escoinson-arch, and in the elevation to the hood-mold, the section of which would be quite different. This also shews that the marks on the sections apply to another elevation. ° The first of these three sections has a mark very nearly the same as that upon the single shaft of the arcade in Plate 59, but reversed; those of the other two cannot be traced upon the other drawings. p The section of the transverse rib-mold has two mason's marks separated by a vertical stroke, which cannot be intended for a joint. They probably apply to its double function of a vault-rib and pier-arch mold. The cross on the left hand is marked on the front of the unfinished springing of the vault-ribs in the in- terior elevation in Plate 01. A small line drawn above it, which indicates a joint or breaking off of the masonry, makes it resemble the character which is applied to the diagonal section. The righi-hand mark probably refers to a lost drawing. i Its mark is placed in Plate 59 upon the front shaft, just above the basement, but the section cannot have been taken at this level, because it represents a pier detached at the sides, as this really is, above the stringcourse. — (W.) PL.LXIII EXPLANATION OF THE PLATES. 235 PLATE LXIII. VERSO OF THE THIRTY-SECOND LEAF. This page gives the section of the apsidal wall of the Cathedral of Rheims above the level of the chapels, with a lateral view of the double flying-buttresses, and of the great isolated piers which receive them. Although Wilars de Ilone- cort has not led us to expect minute accuracy in his drawings, we are obliged to confess that there are great discrepancies between the details which are given in this page and the reality. This is especially the case with the outer pinnacle shaft. In the drawing this is furnished with an appended buttress facing in- wards towards the church. The purpose of this is to receive the imposts of the two flying-buttresses, each of which springs from a molding. The shaft of the pinnacle is in two stories, but the flying-buttresses both abut against the lower one, which is very lofty. In the real building the shaft is also in two stories, but the two flying-buttresses abut one upon each. The lower story has on each face a blank arch 1 ', and the upper one a square tabernacle, in which is the figure of an angel. It is this upper story which in the drawing erroneously stands clear above the upper flying-buttress, and is surmounted by an octagon spire with a little square shaft and spirelet at each angle. The real upper story is capped by an octagon spire, with square angle spirelets, but they, having no shafts beneath, rise from the same level as the central octagon. The great intermediate pinnacle-shaft has none of the stringmolds and arches which are exhibited in the drawing, and its upper termination consists of a plain, heavy quadrangular spire, which is probably due to the restorations that were made after the burning of the roof in 1481. Wilars de Honecort has carefully traced the battlemented parapet which bor- ders the gutters of the chapels, of the great roof and of the gangway under the clerestory windows, but in his section has omitted the triforium gallery s . r This blank arch is indicated in Honecort's draw- molding beneath the gangway of the window-sills is ing, but being filled up with black resembles the shewn, with its return mitring round the base of the pierced openings, which are in this way designated in pilaster buttress, and from the unfinished air of this other parts of these drawings. The lower story of part of the sketch, it may be conjectured that in the pinnacle-shaft is greatly too lofty, the upper one inking his lead-lines he omitted those of the gallery, and the spire as much too short. which had been accidentally rubbed out. — (W.) ' The space for it is left blank in his drawing. The h h 2 236 EXPLANATION OP THE PLATES. The upper termination of the clerestory wall was entirely changed after the fire of 1481 ; therefore the representation of the tabernacle which originally covered the angels shewn in Plate 61, is valuable, but we find no trace of the gurgoyles (or nokers) which now exist, sustained by the cariatides which have been substi- tuted for the angels. Lastly, the crockets shewn at the angles of the great roof inform us that such ornaments ought to be attached to the rolls that separate the lead plates of the roof. We have already seen that a similar ornament must have existed at the angles of the chapels roofs, which were not lean-to roofs, like those shewn in this section, but pyramidal, as in Plate 60. This page, the last but one of the manuscript, bears, like the first, the red stamp of the Bibliotheque Nationale. — (L.) Thus far M. Lassus, whose excellent article upon this plate I have given entire. The lean-to roof of the side-aisles appears to me to shew that this section is taken across the narrow severey of the choir which is adjacent to the apse, looking east- ward, and not across the apsidal wall. Notwithstanding its numerous errors, it is extremely interesting, as shewing how early the idea of representing the ar- rangement of a building by a section arose. It is, in fact, the earliest section that has been discovered. This is the only drawing of the five which belong to Rheims that I am dis- posed to consider as a copy from a working drawing, and not from the reality. The objects delineated in the previous pages were in existence at the time of Honecort's visit, as, for example, the chapels, and the piers and moldings of which he gives plans and details, as well as the severeys of the choir and nave, which were certainly carried up halfway, and the former, as I believe, even to the roof. But the discrepancies between this sketch and the reality are so great, that we may well admit that the pinnacles and buttresses were inserted, partly from the stumps and indications exhibited by the toothings of the building, and partly from descriptions or drawings which he obtained on the spot. — (W.) PL, .LXIV P-ereuCiS cd cpz to l? Wt •^mftef ftidfcf be cot togef. ? ^wnontr cej> tmecAe con efomme g^fihrr Fcnfees we rfamtne trtSe vvutuuwr £ ^mibeis^nratw' u - cms q tettmOfntu^ tffef m^ r ^ ft ntem* t>teic tim atMknprc^ ttefoteut cf F % ftcSTcf^frW W^mte uvp cnunerefem^ue We max* j> ^WtmtSmvmmf eft fkttxmf ft paemt tt taouf wee M^xlxf- ? t? Tcf teftttqwef br utcf Sun latflkf tcfnvto^ wcots oo Cjatkcue* J tttucr tie tottce alautre* p^nWf tme: trmmet*^ EXPLANATION OF THE PLATES. 237 PLATE LXIV. RECTO OF THE THIRTY-THIRD LEAF, MARKED IN THE FIFTEENTH CENTURY WITH THE NUMERAL XXVII. The numeral proves that seven leaves between tins and the preceding have been abstracted since the fifteenth century. But as in the rest of the Album every sheet of parchment is folded into two leaves forming four pages, we may conclude that, prior to this paging by J. Mancel, one of the original pages had been removed. The reverse side of the thirty-third leaf is blank, and therefore this engraving completes the facsimile of the manuscript. " Retencis co que io vus dirai. Prendez fuelles de col roges. et sanemonde [cest une erbe con clainme galion filate] prendes une erbe con clainme tanesie et caneuvize [cest semence de canvre]. Estanpes ces .11 ij. erbes si quil ni ait nient plus de lune que de lautre. Apres si prendeis warance .rj. tans que de lune des. mi. erbes. et puis si lestanpes puis si meteis ces .v. erbes en .1. pot et si meteis blanc vin al destenprer, le meillor que vus poes avoir, auques tenpreement que les puizons ne soient trop espessez si con les puist boire. Nen beveiz mie trop ; en une escargne duef en arez vus aseiz, por quele soit plainne. Quel plaie que vus aies vus en garires. Tergies vo plaie dun poi destoupes. metes sus une fuelle de col roge. puis si beveis des puizons al matin et al vespre .11. fois le ior. Eles valent miex destemprees de moust douc que dautre vin. Mais quil soit bons ; si paerra li mous avec les erbes. et se vus les destenpres de vies vin laissies les .11. iors ancois con en boive. " Cuellies vos flors au matin de diverses colors ke lune ne touce a lautre. prendes une maniere de piere con taille a ciziel. quele soit blance molue et deliie. Puis si meteis vos flors en ceste poure. Cascune maniere par li. si duerront vos flors en lor colors." " Eetenez ce que je vous dirai. Prenez des feuilles de chou rouge et de la sanemonde (c'est une herbe qu'on appelle chanvre batard). Prenez une herbe qu'on appelle tanesie et du chenevis (c'est la semence du chanvre). Ecrasez ces quatre herbes, de sorte qu'il n'y en ait pas plus de l'une que de l'autre. Ensuite vous prendrez de la garance deux fois autant que de l'une des quatre herbes et l'ecraserez. Puis vous mettrez ces cinq herbes en un pot et les ferez infuser dans du vin blanc, le meilleur que vous puissiez avoir, avec cette precaution que la potion ne soit point trop epaisse et qu'on la puisse boire. N'en buvez pas trop ; dans une coquille d'cauf vous en aurez assez, pourvu qu'elle soit pleine. Quelle plaie que vous ayez, vous en guerirez. Essuycz vos 238 EXPLANATION OF THE PLATES. plaies d'un peu d'etoupe, mettez dessus une feuille de chou rouge et buvez de la potion matin et soir, deux fois le jour. Elle vaut mieux infusee dans du mout doux que dans d'autre vin, pourvu qu'il soit bon. Le mout fermentera avec les herbes. Si vous faites infuser dans du vin vieux, laissez deux jours avant que d'en boire. " Cueillez vos fleurs au matin de diverses couleurs : que l'une ne touche point a. l'autre. Prenez une espece de pierre que Ton taille au ciseau et qu'elle soit blanche, moulue et fine, puis mettez vos fleurs en cette poudre, chacune suivant son espece. Par ce moyen se con- serveront vos fleurs avec leurs couleurs." " Remember well what I have to tell you. Take the leaves of reel colewort and of avens. Take also an herb called tansy and some hempseed. Crush these four herbs together in equal quantities. Take of madder twice as much as of each of the others, and crush it. Then put these five herbs in a pot, infusing them in white wine, the best yon can procure, and taking care that the mixture is not too thick to drink. Bo not drink too much at a time, an eggshell will contain enough for a dose, provided it be full. It will cure any wound that you may have. Clean your wound with a little tow, put upon it a leaf of the red colewort, and drink the potion morning and night, tioice a-day. It is best prepared toith new sweet wine, if it be of good quality. It will ferment with the herbs. If your infusion be made with old wine, leave it two days before you drink it. " Gather flowers of different colours in the morning, and do not let them touch each other. Take a sort of stone which can be cut with a chisel, and see that it be lohite, and reduced to a fine powder. lay your flowers in this powder, each according to their species. In this toay the flowers will preserve their colours." The use of the first of these receipts may have been familiarized to De Honecort by the bruises which all workmen are liable to in their workshops. The second seems the same as that given in the following verses of the poem De Artibus Bomanorum, by Heraclius, who lived about the tenth century : — "Mores in varios qui vult mutare colores, Causa scribendi quos libri pagina poscit, Est opus ut segetes in summo mane pererret Et tunc diversos flores ortuque recentes Inveniat, properetque sibi decerpere eosdem. Cum que domi fuerint caveat ne ponat in unum Illos, sed faciat quod talis res sibi poscit vel quserit. Dum super sequalem petram contriveris istos Flores, incoctum pariter congere gypsum. Sic tibi siccatos poteris servare colores." These verses teach us to gather the flowers in the morning, to keep them asunder, and to cover them with unbaked gypsum after having bruised them on a stone. But Heraclius gives this as a method of obtaining vegetable colours, while Honecort appears to intend the preservation of the flowers. — (A. D.) ADDENDA ET COEEIGENDA. Page 8. The church of St. Yvcd do Brainc has lately heen made the subject of a complete mono- graph, by M. Stanislas Prioux, published at Paris in folio, with twenty-seven plates of archi- tecture and monuments. Page 13, line 3. As thirty-three leaves remain in the manuscript, and therefore sixty-six pages, there is an apparent discrepancy in the statement that only sixty-three plates are devoted to the drawings. It should have been mentioned that the recto of the third leaf and the verso of the last leaf are blank, and that the recto of the thirty-third leaf is wholly occupied by recipes. P. 17, 1. 21. For Cosmos, read Cosmas. P. 20, 1. 7. For To trace the joints of voussoirs, read How to cut a voussoir according to rule, PI. 38, No. 18, 19. P. 23, 1. 1. Agies (of the Twelve Apostles in Plate 2). M. P. Meriniee gives the following remarks in the Moniteur Universel, Dec. 20, 1858 : — " M. Littre" has communicated to me several passages, unfortunately none of them anterior to the fifteenth century, in which agiaux or agios has the sense of dress or ornaments. Menage remarks that at Paris they speak of les agios of the village bride, that is to say, her parure. Trevoux in his Dictionary gives the same phrase and the same interpretation. Honccort's legend must therefore be translated "the costume" (or, in the language of the studio, the ajustement) " of the Twelve Apostles." P. 49. In the Dictionnaire d? Architecture of M. Viollet-le-Duc, t. iv. p. 441, is a description of this cross, with a drawing shewing that author's conception of its true proportions. He is of opinion that it must have been cast in bronze, and that the shaft of the column was loftier and more slender than in Honccort's sketch. P. 57 and 58, note j. Vide Col and Forties (pilers) in the Glossarial Index. P. 80. A pair of wrestlers differing in some respects from these examples is in the Luttrell Psalter. (Vetusta Monumenta, vol. vi. pi. 24, fig. 4.) P. 93. For No. 1. and No. 2. beneath the woodcuts, read Pig. 5. and Pig. 6. P. 140, 1. 8. The quarto acuto of Viola is rounded at the apex, so that it is not in reality a pointed arch, although the mode of its description is based upon the media3val nomenclature and methods. P. 161. The angel above the apse of Rheims Cathedral also turns upon a pivot, according to M. Gilbert, and was therefore probably the subject of Honecort's description in this place. P. 162. The engine here described may be one of those to which the name feme or verne is applied in mediaeval documents. Por example, Lefeme occurs in the Ely Pabric Rolls, 16 Edw. II. ; " ij. gynes voc' femes" in a Roll of works at Westminster, t. Edw. I., and verne in another similar roll, 2 Edw. III.; "grece for the veryn," &c, Churchwardens' Accounts, Vfalden, Essex, 6 Edw. IV.; "Feme and Fergn,'" in the Pabric Rolls of York Minster, just admirably pub- lished by the Surtees Society, under the able superintendence of the Rev. J. Raine, jun. In all these cases, as well as in others, there are accompanying details which shew the thing so named to have been a mason's or carpenter's machine for raising heavy materials. On the 240 ADDENDA ET CORRIGENDA. other hand, a windlass in Lincolnshire is still called a fearn, (Halliwell's "Archaic Dic- tionary") ; and verin, in the glossaries of French practical writers of the last century, is the name of a kind of screw-jack for raising and supporting loads. (Vide Daviler, Felibien, and the Dictionnaire de Marine, Amst. 1702.) Lastly, Cotgrave gives, — " Chevre, the engine called by architects, &c, a fearne ;" and the French clievre is the English " triangle, or three legs," which has a windlass attached it whose rope acts upon the load by means of a pulley at the apex of the triangle. P. 196, note b. Windas. In his glossary the French editor quotes, as his authority for interpreting this word as a spring, the following passage from the Statutes of the " Serruriers" of Abbe- ville, a.d. 1480 : — "Kuls ne porra faire windas, cri, poullietz, et aultres engins a bender arba- lestres. . . . Item que les dits windas soient bien et souffisanment fais sans brasures sinon es lieux a ce convenables et necessaires." This passage, which may be translated, "No one shall make windlasses, levers, pulleys, or other machines to bend cross-bows," appears to relate wholly to the well-known contrivances necessary to enable the archer to bend the steel bow, which was too powerful to be overcome without mechanical assistance. It cannot be so rendered as to make windas signify a spring. In fact, Grose, amongst other writers, enumerates in his "Ancient Armour," (p. 59, and note x,) the devices supplied for bending cross-bows as above. The win- lasse, or ivindelaise, was also called inoulinet ; and the lever, a bender, goafs-foot, (or pied de bicJie). Roquefort gives " Croc de fer ... instrument pour bander une arbalete ;" and the cri of the passage above is probably another name of the same lever. — (W.) P. 227, 1. 1. Toral, piler toral, or pilers torausz, applied to the great crossing-piers of Rheims; probably tower-piers, from "Tor, a tower, (Roquefort)." M. Quicherat refers to Arcus toralis in Ducange, where the phrase in the acts of a Spanish council in 1582 is applied to the arch of separation between the nave and choir, or perhaps to the arched doorway in the choir-screen, for the Benedictine editors render the phrase in question, " Cancelli qui separant sanctuarium vel chorum a navi in basilicis, sic dicti, ut opinor, quod pars janua? superior arcus speciem refert." They suppose toralis to be an error for choralis. I also find in the Dictionnaire de Trevoax, 1771 : — "Toral, or Totjral, f. m. Terme de coutume. Elevation de terre, ordinaire- ment couverte de gazon, que Ton fait entre deux heritages, qui appartiennent a deux diffe- rens maitres pour servir de separation, Agger, aggestus toralium." This sense is also given in Ducange, under Torale, No. 2, and justifies the interpretation of Arcus Toralis as a boundary. But the Spanish Dictionary of the Academy of Madrid, quoted by Hcnsehel, gives, "Toral, adj., Lo principal o que tiene mas fuerza, y vigor en qualguier especie, como Arco toral, fundamento toral. Viene del Latino. Torus Toralis." (Diccionario de la lengna Cas- tellana. Madrid, 1739.) This would translate our phrase to mean "the great or principal piers."— (W.) GLOSSAKIAL INDEX. The Arabic numerals refer to the pages, a number in brackets, thus (No. 28), to the article upon the page that follows it, and the small letters in brackets, thus (a), to the foot-notes. Acainte, side-aisle, 105. Agies, Addenda, and 23. Aguile, a spire, (No. 2S) 144. Arbre, the axis of a wheel, 166. Arret, small arch, 57. Arkiere, a loop-hole, 57. "Trous qu'on faisoit dans les nmrs d'unc forteresse pour tirer des flechcs aux ennemis ; en bas Lat. archesia, en Prov. arkieiro, archeiro. — Lucarne pour recevoir du jour par une cour on un jardin." — (Roquefort.) Ars boteres, Ars buteret, flying buttresses, 86. B. Behot, Beiios, a tube, 53. — Beet, lit., canal. (Alle- mand.) — Bedum, the bed of a river or of a mill- course. (Ducange.) Besloge, skew," or oblique, 126. Bevum, wc bevel, 151. Bosine, a trumpet, (k) 115. C. Cantepleure, the siphon of a Tantalus cup, 53. Chan- tepleure, robinet, Er. : a waste-pipe or channel of a reservoir, to let off the water when it rises above its proper level. {Dictionnaire d'Aviler). A cock or tap, to cause water to flow from a vessel at pleasure. Capitel, capital of a column, 147. Roquefort gives, " Capitele : Chapitre, lieu oil s'assembloient les chanoines et les moines ;" and, " On : ou ubi ; au, ad; &c." Therefore on capitel in this place may be translated ' in a chapter-house. 5 — (W.) Cavece, Chavec, the chevet or east end of a church, 123, 86. Centicore, an imaginary animal, 22. Charole, the circumscribing aisle of an apse, 91. (See Carol in "Oxford Glossary," 5th ed.) Cintreel, the centering of an arch, (No. 4) 123. Clef, the key-stone of an arch, (No. 22) 137, (No. 24) 138. the capping and elbows of stall-work, 185. Coen, quoin, or corner, (No. 14) 129, (No. 26) 143. Col, projection of a buttress, 57, (j) 58. " Item ondit coste entre lesdis pilliers a deux autres pil- liers espassez portans chacun iii. piez de col et deux piez despoisse." — (Account Roll, 1399, Bull, du comite historique, p. 53. 1849). — (L.) Comble a viij. costes, an octagon spire, 41. Conble a viij. crestes, an octagon spire with a crest of crockets on each angle, 57, (h) 58. Conpas, a pair of compasses, (No. 40) 152. Or the arc described by them, as " a iij. conpas,'" a trefoil, 45. Also " on canpe a conpas," i. e. " champ decrit au compas," Er., a circular area, 120. Copresse, a shore, 166. Covertic, a sloping roof, 41, 219. Creste, a row of crockets, 57. Cretiatjs, merlon of a battlement, 217, 218, 219. D. Dotjbliaus, transverse vault-rib, 231. E. Enbracement, framing or bracing of a wheel, 166. Engieng, Engiens, any mechanical contrivance, 23, 54, 162, 165, 195. (Vide Renouvier, Des maitres de pierre de Montpellier, p. 212.) Entaulehens, entablature or stringcourse, the table- ment of our mediaeval nomenclature, 57, 217, 219. Entreclos, partition (i.e. of stall-work), 185. Erracenmens, springing stones of a vault or arch, 124, (No. 38) 150. Escaufaile, calefactorium, or hand-warmer, 54. Escaufiole, bassinoire de lit. escaufet, rechaud de feu, poelon. — (Lacombe.) Esconse, dark lantern, 106. Esligement, plan, or floor, (g) 57, 86, 94. Espases de le nef, severies of the nave, 63. Esscandelon, a graduated scale, (Mo. 27) 144. Estaces, piles in water, 165. "EsTAC.pieu poteau," &c. — (Roquefort.) Estage, story, or floor, 41, 205. I 1 242 GLOSSAETAL INDEX. Estancon, post, or stanchion, 195, 203. Estatjs, church-stall, 185, 191. F. Ferne, or Verne (Eng.), an engine for raising ma- terials, Addenda. Filloles, turrets grouped round a tower, (h) 57. Fleke, the detent of the trebuchet, 195, (b) 196, 202, (e) 202. Forkies (filers), angle buttresses arched below, so as to resemble a fork, 57. Fourke, en patois picard et rouchi. Vorkehrt, turned, or set in front, (Ger- man.) — (L.) Forme, frame or tracery of a window, 63, (o) 139, 231. Fus, Fust, timber, (Roquefort,) 105, 127. H. Henap, a cup, 53. " Pateras dicuntur cuppas hanaps. (Jehan de Garlande, c. 1220)." — (L.) J. Jagiis, gaged, (No. 38) 150. Jerloge, Orologe, a clock, 41. L. Legiere, in the sense of easy to make or do, 105, 191. Letris, a lectern, 45. " Lectrinum, lectricium, a church desk." — (Ducange.) Lewis (Eng.), Louve (Fr.), an instrument to grasp stones, 163. Linel, Livel, a mason's line or level, (No. 14) 129, (No. 31) 146. Loisons, bond or joints of masonry, (o) 97, (No. 26) 143, 227. M. Maille, a mallet, 35. Masons don orologe, Maizon dune jerloge, a clock-house, or clock-case, 41. Mole, Molle, mason's mold or pattern, 121, (No. 29) 144, (No. 37, 38) 150, 231. Montee, Droite montee, elevation or view, 57, 205, 217, 219. Mors, sloping cap of a pilaster or buttress, (a) 220. N. Nokeres, spouts to throw off water from a roof, 219 line 16. ' "Noc, gouttiere, plomb qu'on met en avance sur les toits, pour faire ecouler l'eau." — (Roquefort.) Also Tarbe, Glossaire de Champagne. The word noquet is now employed in France for the strips of lead or zinc which are laid on the ridges and other parts of roofs. Angl., flashings. O. Ogive, the diagonal rib of a vault, 231. Okbes arkes, arches against a blank wall, 211, (q) 211. P. Paelete {Petite poele, Roquefort), a little brazier, 54. Peignons, Peignonciaus, gables, 41. Pen, Pan, plain pan, flat walls (or curtain walls) of a church between buttresses, 86, 219, 227. Vide " Oxford Glossary," 5th ed., art. Pane. Pendans, the voussoirs of vaulting surfaces or pen- deutives, 135. (Vide Vosure pendant). Peniaus (panneaux, Fr.), panes, or pannels, 41. Pentagram, the star-shaped pentagon, 113. Pile r, pier of a church, 97, (No. 26) 143, 227. Pilaster, or buttress, 57, 63. ("Pilar, pilare, pilier exterieur," Renouvier, p. 216.) Plonc, Plom, plummet, or plumb-rule, (No. 14) 129, (No. 31) 146, (No. 34) 148, 165. Poupee, the terminating standard of a range of stalls, 185, 191. From poupee, a child's, doll, a bundle of hemp, or other commodities. This derivation would suit our " popit-head," better than the ap- plication of the word to the terminating standard. It may be derived from poupe, Fr., (puppis, or popa, Ducange,) the poop or stern of a ship. For the standard, usually covered with rich carving and imagery, terminates the range of stalls in the same manner as the elaborately carved stern ter- minates the hull of a ship. — (W.) Prael (pre'au, Fr.), a cloister-garth, (No. 11) 127. Presbiterium, Bresbiterium, the apse or eastern extremity of a church, 91, 94, 103. R. Reonde veriere, a rose- window, 99. Rieule, according to rule, regular, (k) 132, 135. "Rieule, regie de magon, a Lille. Rieulet, Diet. rouchi-francais." — (L.) S. Saint graal, 33, 75. Sarpens, dragons, 45. Scere, a mason's square, (No. 23) 137. Sole, platform, or base-framing, 165, 195. GLOSSARIAL INDEX. 243 Soore, a saw, 159. Sorvols, super-arch, the upper order of voussoirs in the transverse ribs of Rheims, 232. Suel, a window-sill, 210, line 9. T. Testes de fuelles, foliaged heads, 37, 155, 157. Tunc, arch of the third point, (No. 22) 137. Tor, Toor, a tower, 57, (No. 28) 144, (No. 32) 146, (No. 35) 149, 166. Toraus, (filers) the crossing or tower-piers of a church, 227, and Addenda. Torete, Tourete, turret, 53. Travecons, cross-pieces or legs connecting the siphon tube with the bottom of the Tantalus cup, 53. V. Verge, the beam of the trcbuchet, 105, &c. Veriere, Verreres, Vesrires, window, or lights of a window, 99, (No. 5) 123, 231. Voie, the deambulatory of a cloister, (No. 11) 127 ; gangway under the windows of the chapels and triforium gallery of Rheims, 211, 219. Volte, a vault, and Volte de fust, a wooden vault, 105. Vosor, Vosoir, voussoirs, (No. 27) 144, (No. 37) 150. Vosure, vault, arch, or vaulting-surface, (No. 8) 125. Vosure besloge, a skew arch, (No. 9) 126. Vosure engenolie, cusped voussoir, (No. 39) 152. Vosure pendant, hanging-arch or voussoir, (No. 30) 145. Vosure riuleie, vault or arch made to rule, (No. 18) 132. Vosure taillie, a finished voussoir, (No. 36) 149. W. Windas, a capstan, (b) 196, 201, (d) 201, Addenda. printct) bit ?ftcssrs. Jhtlser, (Eoramarhrt, 0*fort>. 1 * GETTY CENTER LIBRARY 3 3125 00762 6316