'3 
 
Digitized by tlie Internet Arcliive 
 in 2013 
 
 http://archive.org/details/architecturalstuOOpeti 
 
Frontispiece 
 
ARCHITECTURAL STUDIES 
 IN FRANCE. 
 
 BY THE EEV. J. L. PETIT, M.A., F.S.A. 
 
 WITH ILLUSTRATIONS FROM DRAWINGS BY THE AUTHOR 
 AND P. H. DE LA MOTTE. 
 
 LONDON: 
 GEORGE BELL, 186, FLEET STREET. 
 1854. 
 
7^0-94-4- 
 
 W£6£TTy CENTER 
 
PKEFACE. 
 
 I HAVE no doubt that a work might easily be compiled 
 from foreign architectural writers, giving whatever infor- 
 mation I am likely to convey in the following pages, and 
 with infinitely greater fulness, clearness, and accuracy. I 
 shall unquestionably be often found to have pointed out 
 features as being remarkable, which the French antiquary 
 recognises as ordinary and common-place ; or to have mis- 
 taken exceptional instances for general characteristics of 
 style ; and it is not probable that I shall avoid errors 
 whenever I attempt to fix a date. Nevertheless, my task, 
 if such it can be called, may not be altogether needless and 
 superfluous. The peculiarities of a national style strike the 
 visitor more vividly than the native ; while, on the other 
 hand, the features to which he is accustomed acquire a new 
 interest. The student may discover that many of those 
 which he has been taught to consider necessary marks of 
 Gothic architecture itself, are, in fact, no more than national 
 or local distinctions, and hence he may reduce into a nar- 
 rower circle the system from which he endeavours to work 
 out the true principles of the style. 
 
 We cannot, I think, fully enter into the character of 
 English architecture unless we give some attention also to 
 French, German, and Italian. And if we take that most 
 interesting period when the genius of the Gothic Style was 
 beginning to develop itself, namely, the end of the twelfth 
 and beginning of the thirteenth century, we shall at once 
 
iv 
 
 PREFACE. 
 
 see the high claims of French architecture. The Itahan 
 never did reach to the Northern Gothic, which I must 
 always consider as the true Gothic, however fully I may 
 admit the great beauty and interest of the Southern Pointed 
 Styles. The German transition appears to me to be scarcely 
 more than a peculiar form of the Romanesque, having little 
 or no tendency to further progress ; and likely, unless 
 under some external impulse, to pass away without deve- 
 loping any decided style. Those which have, at the time 
 I refer to, any life or movement, at least in the direction of 
 Northern Gothic, are the French and the English ; and 
 these, I think, may be shown to have an inherent and 
 independent vitality of their own ; either would have lived, 
 and advanced to perfection, without the aid of any external 
 influence. Whatever connexion may have existed between 
 them undoubtedly strengthened and advanced both, but 
 the germ of progress was in each independently, and showed 
 itself by movements in great measure independent of each 
 other. The comparison between the two styles shows this ; 
 their similarity proves the identity of their aim ; their dif- 
 ference, the independence of their action. 
 
 But the era to which I refer, you will observe, is subse- 
 quent to the Norman Conquest. That great event influ- 
 enced the architecture, as it did the whole tone and character 
 of the nation. It is, therefore, the Normanised English 
 that I speak of, and also, I may say, the Normanised 
 French, for the Normans are no more the indigenous people 
 of one country than of the other. The first development 
 of Gothic in each may be supposed to combine the genius 
 and energy of the Norman with the principles of architec- 
 ture, whatever they might be, which he found already to 
 exist. The ruder and less defined such principles, the 
 
PREFACE. 
 
 V 
 
 more readily they would be swept away, and the freer 
 scope would be given to the genius of the invader. Hence 
 we shall see many curious phenomena during the period of 
 transition, in the several localities where the Norman influ- 
 ence was felt; and these would depend not merely upon 
 the directness or indirectness of such influence, but also 
 upon the elements already existing, on which it might be 
 exerted ; so that, according as the old style was more 
 firmly established, the indications of change to a new one 
 would be fainter ; and, on the contrary, where the princi- 
 ples of the old style were vague and indefinite, the advance 
 of the new would be unimpeded and unencumbered by 
 elements foreign to its spirit and character. Now in Eng- 
 land we may almost say that there was no ante-Norman 
 style whatever ; at least all her alleged Saxon remains 
 taken together, present nothing which could stand for a 
 moment against, or in any degree influence the progress 
 of, a style that might lay claim to the slightest portion of 
 artistic merit. And I suppose the case was much the same 
 with a great part of northern France ; it is therefore in 
 these localities that we might look for a more ready, rapid, 
 and complete transition, while in places where either Nor- 
 man influence was less felt, or the native architecture more 
 firmly established, the transition would be slow and imper- 
 fect, and a frequent impulse from without might be required 
 to bring the new style to maturity. 
 
 If, therefore, I express my belief that the Gothic prin- 
 ciple was more completely developed, at an early period, 
 in England than anywhere else, (and the comparisons 
 between Amiens and Salisbury do not convince me to the 
 contrary,) I am merely declaring that the Norman influ- 
 ence was as strong as it could be, and any opposing 
 
PEEFACE. 
 
 element as weak as it could be ; that architectural art 
 in England at the time of the Conquest was not in a 
 condition to make any stand on its own ground, but could 
 easily be swept away without leaving anything that might 
 affect its successor. The province of Normandy itself 
 comes very near to England in this respect ; but still I 
 think it will be found to present Romanesque indications 
 at a time when in our own country they have as com- 
 pletely vanished as they possibly can in the system of 
 Gothic architecture. 
 
 I will not say that my purpose in writing the following 
 pages is to illustrate this, but the line I have taken will, I 
 think, enable me to do so as w^ell as any other I could 
 have proposed. For I have devoted myself principally to 
 the Ante-Gothic, the Transitional, and the early Gothic 
 styles ; those of the fourteenth and fifteenth centuries, much 
 as they contribute to the magnificence of French eccle- 
 siastical architecture, having occupied but little of my 
 attention. 
 
 I am sure a student would obtain more knowledge of 
 English architecture by noticing a limited number of Eng- 
 lish buildings, and also of foreign ones, each, of course, be- 
 ing taken in different localities, than by giving up his time 
 exclusively to the former, and examining every church or 
 old specimen in the country. If his aim is not merely his 
 own information, but the advancement of art, the necessity 
 is still greater that he should extend his views. 
 
 The remains of English mediaeval architecture, numerous 
 and important as they are, are not sufficient for its own 
 revival ; some new combination is necessary to give the 
 stimulus now so much wanted. It is evident that archi- 
 tects feel this, by the number of churches sjmnging up on 
 
PREFACE. 
 
 vii 
 
 French or German models. But we look for something 
 more than mere copies. The deficiency is not to be reme- 
 died by a few importations of transoms or capitals, or even 
 of general outlines and plans. Our object is not, or ought 
 not to be, to produce a building which might be mistaken 
 by the antiquary for a specimen of a particular period or 
 country, but one fitted in every respect, in reality as well as 
 appearance, for its purpose, and sound and correct in its 
 design, its construction, and its ornamentation. And if in 
 our attempts to attain this end we make use of ancient exam- 
 ples, we must not content ourselves with finding authorities 
 for such and such a form, but must learn carefully to discri- 
 minate between features which are simply characteristic of 
 date or country, and true architectural principles. 
 
 I am far from thinking that nomenclature is a remedy 
 for every defect in art or science ; still I cannot but 
 feel that confusion of terms generally springs from, and 
 always leads to, confusion of ideas. I know that I am 
 constantly employing such terms as the chevron mould- 
 ing," the beak-head moulding," the dog-tooth mould- 
 ing," and the like ; and I believe they are pretty generally 
 used. But the expression, in fact, confuses two things 
 which ought to be kept very distinct, as belonging to 
 different classes altogether ; I mean purely architectural 
 decoration, and sculptural, or what may be called adven- 
 titious, decoration. The former renders the architect inde- 
 pendent of other artists ; the latter enables him to press 
 into his service all the talent he can command. A very 
 rich building may be j)i*oduced by the hands of workmen 
 who profess to do no more than carry a given section to 
 any length required. Such is the case with our great 
 Cistercian abbeys ; they have little or no sculpture, yet, 
 
viii 
 
 PEEFACE. 
 
 from the use of purely architectural ornament, their effect 
 is rich and striking. By architectural ornament, I mean 
 such mouldings, or other members, as, without being ne- 
 cessary, still in a manner illustrate the principles of con- 
 struction ; show the divisions of the building into its sepa- 
 rate stages ; mark the directions of the several thrusts and 
 supports ; distinguish angles and other important lines ; 
 and give an apparent strength to the sustaining masses, 
 and lightness to those sustained ; which, in the case of the 
 pier and arch is admirably done by giving the latter a 
 greater number of mouldings and surfaces than the former. 
 It is on this class of ornament that the style must depend 
 principally for its character; because a building may be 
 very perfect, rich, and beautiful, that employs no other. 
 
 The other class of ornament, which comprises foliage of 
 capitals, statues occupying niches, much of the work that 
 occurs as a decoration of surfaces, or that enriches the 
 mouldings of arches, imposts, or strings; the carving of 
 corbels and brackets, and the like, belongs rather to the 
 sculptor than the architect ; and in discussing the merits 
 of any style, as regards decoration, we must keep in mind 
 the distinction between the two classes. If the architect 
 is thrown upon his own resources, and cannot command 
 the aid of the sculptor, he must necessarily employ archi- 
 tectural rather than sculptural ornament. This may ac- 
 count for the enrichment of surfaces with panellings, which 
 in a manner repeat in miniature the construction of the 
 building, instead of those imaginative designs which belong 
 to another branch of art. And even though we should 
 feel more gratification from the latter, yet I think we have 
 no right to condemn the style or system which admits the 
 former, as does our Perpendicular, provided it leave room 
 
PREFACE. 
 
 ix 
 
 for the exercise of artistic design and workmanship where 
 such can be commanded. 
 
 But it is one thing for an architect to prove himself 
 independent of the aid of the sculptor, and another, to be 
 unable to avail himself of it when offered. That style 
 alone can be considered as truly noble, which can admit as 
 an accessary, to the advantage of both architect and artist, 
 art in its highest character; which, while it requires a certain 
 degree of subordination, is enabled to compensate amply 
 for it by the dignity it confers on the work presented ; 
 which can alike elevate, and be elevated by, its assistant. 
 A Doric temple would be grand, even if enriched by no 
 sculpture whatever ; bat it readily admits, and, as it were, 
 advances that of the highest order ; the higher the charac- 
 ter of the sculpture, the more completely it accords with 
 the spirit of the architecture. Therefore the Doric is a 
 noble style ; and, according to this test, the Roman, or 
 Italian, whatever be its faults, is no other than a noble 
 style, for it has a character of its own, independent of 
 adventitious ornament, and yet no work of sculpture can 
 be too grand to form an appropriate accessary. The same 
 may be said of the true Mediseval Gothic ; it gave full 
 scope to the highest art that existed during the period of 
 its reign ; it eagerly appropriated the best works it could 
 obtain, and ennobled them by the positions in which it 
 placed ihem, — witness the sculptures of Wells and Lincoln, 
 and many continental churches, and the exquisite foliage 
 of which it is needless to mention any examiple. And 
 therefore the old Gothic was a noble style, and will be 
 confessed and felt to be such as long as a memorial of it 
 exists. But if the modern Gothic, instead of exalting art, 
 lowers it ; if it requires its artist to work down to its own 
 
 b 
 
X 
 
 PREFACE. 
 
 standard, instead of up to the highest standard he can 
 propose ; if it rejects all models but such as no refined 
 sculptor would set before himself as a study, however he 
 might admit certain excellences in them ; if it perforce 
 ties itself down to stiff, grotesque, or conventional forms, 
 because it is unable to adapt itself to graceful, natural, and 
 dignified ones, then it is a mean and ignoble style, and 
 will be felt to be such as long as a trace of it remains. 
 Let it affect every peculiarity, let it copy every detail of its 
 predecessor, yet the lapse of centuries will never obliterate 
 its true character ; the difference will always be felt be- 
 tween the work of men who are doing their best, and that 
 of persons who in aping their productions, are deliberately 
 lowering the standard of art. 
 
 It may be said that some mediaeval statues have a grace 
 and dignity that it would be difficult to find in later works. 
 Most probably. Their sculptors worked in earnest, and un- 
 doubtedly attained a high degree of excellence. Still, I 
 think, I shall not be contradicted when I again assert, 
 that no sculptor, anxious to advance his own reputation 
 and his art, will ever set up a mediaeval statue as his 
 model. He may acknowledge its merits, and learn nmch 
 from a careful examination of it ; but still he will not look 
 up to its designer as his master and guide. When I say a 
 mediaeval statue, I mean one fairly expressive of the cha- 
 racter of northern Gothic statuary ; such as would in the 
 present day be considered suitable to a Gothic building. 
 If this be so, I repeat that unless we can adapt our Gothic 
 to the requirements of art, according to our highest ac- 
 knowledged standard, in such manner that there shall 
 appear to be no incongruity between the architecture and 
 sculpture, we ought to abandon it altogether, as a style 
 
PREFACE. 
 
 xi 
 
 calculated to repress, rather than to encourage, advance- 
 ment to perfection.* 
 
 Any argument against retaining Gothic in the construc- 
 tion of new buildings, will apply w^ith double force against 
 the restoration, as it is called, of old ones. For if in the 
 one case, we simply encumber the ground with a building 
 of mean and ignoble character, in the other, we substitute 
 a mean for a noble building : and all the skill of the archi- 
 tect or labour of the workman, will never mend the matter. 
 Occasional repair may be necessary, both with a view to 
 the stability of the fabric, and the purposes assigned to it ; 
 but beyond this the architect who respects the work of our 
 mediaeval ancestors, ought not to go. The cases in which 
 the rebuilding of the whole, or a great part, of an old fabric, 
 is demanded, will probably be found to be more rare and 
 exceptional than we are at present disposed to imagine ; 
 and even the repair of sculptural decoration, and especially 
 the reproduction of ornaments belonging to the earlier styles, 
 is, I take it, altogether wrong in principle. But I shall 
 have occasion frequently to recur to this subject. 
 
 As my wish has been to induce the reader to make his 
 own observations, rather than to content himself with any 
 that may be set before him, I have aimed, in the following 
 pages, rather at sketches which might give the general 
 character, than at minute descriptions. If he has an op- 
 portunity of filling them up on the spot, he will thank me 
 for their slightness. Though I have not turned my atten- 
 tion so much to sculptural decorations, as those that 
 
 * I am not entering into the question, whether our recognised standard 
 of perfection is the highest and best we can propose. I only argue, that as 
 long as we profess to recognise it, we ought to work fairly up to it. 
 
xii 
 
 PREFACE. 
 
 are more strictly architectural, and to the general aspect 
 of the whole building, yet I trust, by the aid of Mr. De 
 la Motte's drawings, that I may have brought before his 
 notice some beautiful specimens of the former. He will 
 hardly fail to perceive, for instance, that S. Nicholas, of 
 Blois, and the very curious and interesting church of 
 Loches, abound with exquisite workmanship. 
 
 The cuts which Mr. De la Motte has made from his own 
 drawings, may be depended upon as accurate, so far as the 
 scale will allow ; the reader will distinguish them by the 
 initials P. D. annexed in the index. For the drawing from 
 which is engraved the cut of S. Mark's, Venice, in the 6th 
 chapter, I am indebted to Mr. T. Hill, Architect. The 
 other cuts are mostly from sketches of my own ; and as the 
 engraver has often found it necessary to supply my deficien- 
 cies by conjectural corrections, they cannot so safely be 
 taken as evidence in matters of detail, but they will, I hope, 
 be of use in illustrating the points to which they refer. With 
 regard to the Anastatics, they are professedly rough copies 
 of rough sketches, intended to convey little more than 
 general character ; and although they will seldom give the 
 reader as much information as he wishes, at least they are 
 not calculated to lead him into error. 
 
 I hope, on some future occasion, I may be able to avail 
 myself of Mr. De la Motte's well-known skill in photo- 
 graphy. The value of this beautiful art in conveying ar- 
 chitectural character, combined with the most faithful 
 accuracy of detail, is universally recognized. It is appli- 
 cable both to general outline and to minute ornament. It 
 will preserve to the student the proportions of a tower or a 
 front, and the carvings of a bunch of foliage. It will 
 never supersede the work of the sketcher, for the success 
 
PREFACE. 
 
 of its processes will often depend on circumstances beyond 
 control ; but it will work hand in hand with him, and 
 assist him ; and above all, it will keep him up to the mark 
 in his own work ; for he will scarcely venture upon inac- 
 curate representations, with such a monitor by his side to 
 call him to order. 
 
 I will now name a few of such French works as bear 
 upon the subject before me, and some of which I have 
 found extremely useful in my architectural studies. The 
 reader will observe that I do not profess to give a selection 
 from the great number of works that have appeared of 
 this description, but simply the titles of those to which 
 my own attention has been, more or less accidentally, 
 turned. 
 
 M. L'Abbe J. J. Bourasse. — Les Cathedrales de France. Tours, A'^. 
 Marni et 1843. 
 
 M. Prosper Merimee. — Notes d'lm Voyage dans Touest de la France. 
 Paris : Libraire de Fournier, Rue de Seine. No. 14, Bis. 1826. 
 
 Notes d'un Voyage dans le Midi de La France, Bruxelles : Louis Hanman 
 etcomp^ 1835. 
 
 Notes d'un Voyage en Auvergne. Paris: Libraire de H. Fournier, 18, 
 Rue de Vernueil. 1838. 
 
 These works will be studied with advantage by the antiquarian tourist 
 before he has planned out his route. 
 
 M. de Caumont. — Cours d'Antiquites Monumentales. 
 Bulletin Monumental. 1834 to 1851. 
 
 The second volume of this work is an admirable treatise on mediaeval 
 architecture, illustrated chiefly by French specimens. 
 
 Statistique Monumental de Calvados. Paris : Derache, Rue du Bouloy, 
 No. 7 ; Dumoulins, Quai des Augustins. Caen : A. Hordel, Rue Froide, 
 2. 1846. 
 
 A work of this kind ought to be, and might be, produced by every 
 architectural society in England; namely, a satisfactory, though 
 
xiv 
 
 PEEFACE. 
 
 concise account of every relic of antiquity in the district. As the 
 department, which M. de Caumont has undertaken to illustrate, is 
 one peculiarly rich in antiquities, the work, which now consists of 
 two volumes, is far from being complete ; still even in its present 
 state it is most valuable. Nothing is passed over unnoticed, though 
 objects of small interest are dismissed in very few words. Plans, 
 elevations, perspective views, details, are given really for the purpose 
 of illustration, and not for mere ornament. The drawings are evi- 
 dently correct, and their execution bold, spirited, and characteristic, 
 without being unnecessarily elaborate. The plan of the work, I 
 repeat, is one which might be adopted with advantage, though it 
 might be difficult to produce a similar work of equal merit. 
 
 M. Didron Aine. — Annales Archeologiques, 10 vols. 1844 a 1850. 
 Paris: au bureau des Annales Archeologiques, Rue d'Ulm, 7, (ancien 
 No. 1), pies du Pantheon. A la librairie Archeologique de Victor Didron. 
 Place Saint Andre des Arts, 30. 
 
 M. L. Batissier. — Histoire de I'Art Monumental dans I'antiquite et au 
 Moyen Age ; suivie d'un Traite de la Peinture sur Verre. Paris : Furne 
 et Compagnie, Libraires Editeurs. 1845. 
 
 Le Moyen Age et la Renaissance : Histoire et description des mceurs et 
 usages, du commerce, et de I'industrie, des sciences, des arts, des litteratures 
 et des beaux arts en Europe. Direction Litteraire de M. Paul Lacroix ; 
 Direction Artistique de M. Ferdinand Sere ; Dessins facsimile par M. A. 
 Rivaud. 5 vols. 1848 to 1851. Paris, Administration, 5, Pont de Lodi. 
 
 * 
 
 M. J. H. Michon. — Statistique Monumentale de la Charente. Dessins 
 et plans par M. M. Zadig Rivaud, Maire d'Angouleme, Jules Geynet, De 
 Lafarque Tauzia, Paul Abadie, Architecte, et Ed. Fabore. Paris : Chez 
 Derache, Libraire, Rue du Bouloy, 7. Angouleme: Chez les principaux 
 Libraires. 1844. 
 
 M. Jules Gailhabaud. Monuments Anciens et Modernes. Collection 
 formant une histoire de I'architecture des different Peuples a toutes les 
 epoques. Paris : Libraire de Firmin Didot, Freres, Imprimeurs de ITnstitut 
 de France, Rue Jacob, 54. 1846. 
 
 M. T. De Jolimont. — Description historique et critique et Vues Pittor- 
 esques, dessinees d'apres nature ; et lithographies des monuments les plus 
 remarquables de la ville de Dijon. Paris : Imprimerie de A. Barbin, Rue 
 des Marais S. C. No. 17. 1830. 
 
PREFACE. 
 
 XV 
 
 Description historique et critique et Vues des Monuments Religieux et 
 Civils les plus remarquables du departement du Calvados. Batis dans les 
 siecles du Moyen age et ceux de la renaissance, jusqu'au regne de Louis 
 XIV. exclusivement. Paris : Imprimerie et Libraire de Firmin Didot, Im- 
 primeur du Roi, Rue Jacob, No. 24. 1825. 
 
 M. A. F. Arnaud. — Voyage, Archeologique et Pittoresque dans le De- 
 partement de I'Aube et dans Tancien diocese de Troyes. Troyes, Impri- 
 merie de L. C. Cardon. 1837. 
 
 M. A. Couchaud. — Chois d'Eglises Bysantines en Grece. Paris : Le- 
 noir, Quay Malaquais, 5. 1842. 
 
 M. Albert Lenoir. — Architecture Monastique. Paris : Imprimerie Na- 
 tionals 1852. 
 
 Le Dr. Eug. J. Woillez. — Archeologie des Monuments Religieux de 
 Pancien Beauvoisis pendant la metamorphose Romane. Paris : Derache 
 Libraire Editeur, Rue de Bouloi, 7. 1832-1849. 
 
 I am indebted to this work for my introduction to one of the most in- 
 teresting groups of village churches that I have ever studied. 
 
 M. Chapuy. — Le Moyen Age Pittoresque. Vues et fragments d'Archi- 
 tecture, meubles et decors en Europe du 10^ au 17^ siecle : Dessines d'apres 
 nature. Paris : public par Veith et Hauser, Boulevard des Italien, 11. 
 
 M. M. Ch. Nodier, I. Taylor, et Alph^ de Cailloux, Voyages Pittoresques 
 et Romantiques dans I'ancienne France. Paris: De Timprimerie de A. 
 Firmin Didot, Rue Jacob, 24. 
 
 M. FeHx de Verneilh. — L' Architecture Byzantine en France, Saint Front 
 de Perigueux, et les Eglises a coupoles de I'Aquitaine. Paris : Librairie 
 Archeologique de Victor Didron, Rue Hautefeuille. 13. 1851. 
 
 1 have had frequent occasion to refer to this able work, which should 
 be studied by any one who wishes to pay attention to the very pecu- 
 liar class of edifices of which it treats. 
 
 M. Alex. Ducourneau. — La Guienne Historique et Monumental. 1842- 
 1844. Bordeaux : Imprimerie de P. Coudert, Rue porte dijeaux, 43. 
 
 I might notice besides these a great number of valuable local works, as 
 those which illustrate Bourbonnais, Auvergne, Velay, &c. Monographs of 
 Cathedrals, as Chartres, Noyon, &;c. Guide Books of different towns, as 
 
xvi 
 
 PEEFACE. 
 
 Poitiers et ses Monuments, Angers Pittoresque (from which I have bor- 
 rowed the Plan of a church) and many others which appear to be carefully 
 drawn up, and are full of information. 
 
 In my Appendix I take occasion to notice Mr. Ruskin's 
 second volume of " The Stones of Venice," which fell into 
 my hands just when I had concluded, as I believed, the 
 literary portion of my work. The unavoidable delays in- 
 cident to illustration, have given time for the appearance 
 of Mr. Ruskin's third volume; and as I have so often 
 taken advantage of his remarks, in the furtherance of my 
 own views, and yet the conclusions I have drawn are in 
 many respects apparently so opposite to those at which he 
 has arrived, I cannot avoid noticing them, however briefly. 
 Like him, I would look forward to an architectural style 
 that shall appeal to a deeper sense than a critical taste for 
 correctness ; and display a power beyond that of mere 
 science. That such a style will again spring up, sooner or 
 later, I have little doubt. But I cannot venture to pro- 
 nounce, what will be its constructive or decorative cha- 
 racter ; whether it will employ the arch, or the lintel, or a 
 combination of both ; whether its arches will be pointed or 
 round, complete or segmental, simple or compound ; whe- 
 ther it will have plain or clustered columns ; whether its 
 forms and proportions will be defined by strict laws and 
 limits, or vary according to the fancy of the architect. It 
 may prove to be something as little within the contempla- 
 tion of our present architects and architectural writers, as 
 the richest Gothic was in the imagination of the first build- 
 ers of the Roman basilica. 
 
 But as we must work for the attainment of such a style, 
 the question is, are we to work upon Classical or Gothic 
 materials. Mr. Ruskin condemns the classical style in toto. 
 
PREFACE. 
 
 xvii 
 
 He would not allow a vestige of it to remain in our archi- 
 tecture. He denounces, as justly as forcibly, the vices and 
 corruptions that have clung to it ; but I do not think he 
 has shown that by cutting all these away, we should alto- 
 gether destroy its character and vitality. Divest it of 
 every thing that may be attributed to Pride of Science, 
 and Pride of State, and Pride of System, or to Paganism 
 and Infidelity ; there still remains a noble, and not alto- 
 gether a meagre and unadorned style ; one at least which 
 cannot be despised as a foundation to work upon, since on 
 it was raised the superstructure which terminated in the 
 most beautiful Gothic of the mediaeval times. 
 
 I look at the Classic as a style on which we may safely 
 fall back, when we have got no other. It has nothing in 
 it to offend that refinement always attendant upon the pro- 
 gress of civilisation ; it adapts itself to the highest and most 
 perfect condition of the arts ; and to say that it cannot 
 afford scope for originality, invention, and genius, is to 
 deny the existence of Wren and Michael Angelo. 
 
 I am not speaking in favour of this style, from any 
 prejudice or partiality. I cannot feel the same interest in 
 it that I do in mediseval Gothic ; I would not willingly see 
 the poorest of our old churches replaced by a classical 
 structure. But I think as much may be said against 
 modern Gothic as against revived Italian ; it may equally 
 abound with pedantry, and not equally meet the require- 
 ments and refinements of the age. 
 
 The Gothic style, as hitherto developed, is precious as 
 the record of a departed age ; an age of immense interest 
 and importance, and one of which every record should be 
 religiously preserved ; not by imitations, which, the greater 
 their merit as such, only cause the greater confusion among 
 
 c 
 
xviii 
 
 PREFACE. 
 
 the landmarks of history ; but by clearing away all that is 
 not genuine, and presenting a system to the historian or 
 antiquary, pure and unadulterated, however imperfect and 
 mutilated. We do not want restored or revived mediseval 
 manuscripts, pictures, statues, buildings, customs, manners, 
 or literature. What we want, is a faithful memorial of a 
 most remarkable, distinct, and definite period in the pro- 
 gress of the world ; for that the middle ages are to be taken 
 as such, any reflecting mind will acknowledge. And this 
 memorial must be preserved in a distinct and definite form. 
 If possible, without loss ; but at any rate without addition 
 and corruption. I believe, in what I say I am only echoing 
 Mr. Ruskin's own sentiments ; that is, if I rightly retain 
 the impression of some beautiful passages in his Lamp of 
 Memory. 
 
 Now the foregoing observations are applicable to me- 
 diseval, but not to classical architecture. The gigantic 
 remains of Roman art are records of the power, and, in 
 many respects, of the tastes and habits of that wonderful 
 people ; but I cannot consider the style itself to be parti- 
 cularly suggestive of their character ; every stone does not 
 bear, as in Gothic, an impress of the mind of the builder 
 and designer. No doubt this deprives the style of much 
 interest; but it renders it a convenient tabula rasa on 
 which we may write our own history and character without 
 confusing them with those of other ages. As to the 
 religious view of the subject, that may be shortly disposed 
 of. Whatever style we choose, if we work in a Christian 
 spirit, we shall Christianize it ; if in a Pagan, we shall 
 Paganize it. There can be no necessary connection be- 
 tween pure doctrine and pointed arches ; and I cannot 
 conceive how the introduction of a straight lintel is to 
 undermine the principles of our faith. 
 
PREFACE. 
 
 xix 
 
 I do not think I am influenced in my views, touching 
 the comparative merits of mediseval and modern Gothic, 
 by an abstract love of antiquity. I do not estimate an 
 object simply by its age ; I have no particular taste for 
 cromlechs or druidical remains ; I would not go further to 
 visit an undoubted Saxon relic, than a good specimen of 
 decorated or perpendicular; and I believe that if the 
 period of good Gothic workmanship had fallen a century 
 or two nearer our own times, I should equally value its 
 examples. Nor do I think that at any distance of time 
 from the present the Gothic structures of this century, at 
 least of the half of it already concluded, will be held to 
 bear comparison with those of the mediseval period. 
 
 I would not be thought to disparage the age in which 
 our lot is cast. I believe we have some losses to recover, 
 such as an eye for architectural form, in which we are 
 deficient even as compared with our predecessors of the 
 last century ; and a correct taste in colour. But I can see 
 no natural impediment to the recovery of these or any 
 other gifts ; we are not indifferent to excellence ; we are 
 not without feelings, tastes, and affections which may be 
 addressed by art or by language ; we are not indisposed to 
 honour the memory of Turner, to do homage to the ex- 
 panding genius of Millais ; and that we can appreciate a 
 power of description which perhaps has never been equalled, 
 and a strain of eloquence almost new to our literature, our 
 reception of Mr. Ruskin's own works will amply testify. 
 That we fail in giving life and energy to our reproductions 
 of an architectural style which we admire, and in great 
 measure understand, I attribute to no incapacity on the 
 part of our architects, but to an error ; one very natural 
 and pardonable, but still, as long as it lasts, very fatal ; 
 
XX 
 
 PREFACE. 
 
 namely, that of endeavouring to transplant the growth of 
 one period into another widely different. And I have 
 tried to show that we are not doing this if we take the 
 classical style as our groundwork ; but at the same time I 
 would not rest content with the style as we find it. If we 
 give our care and thought to refining, purifying, and en- 
 nobling it, we shall possibly be led to the formation of a 
 style no less characteristic, no less beautiful, than our past 
 Gothic. 
 
 If the reasons on which I ground my opinion be proved 
 futile and insufficient, I most willingly retract it. The 
 subject is a grave and important one, as regards the inter- 
 ests of architecture, and it must be handled with delibera- 
 tion ; we must weigh every argument, prove every asser- 
 tion ; we must not suffer ourselves to be carried away by 
 a series of apt illustrations, or a flood of eloquence. It is 
 a question encompassed by difficulties ; but if we meet 
 them fairly and fearlessly in the exercise of an unbiassed 
 judgment, we shall undoubtedly clear our way to the 
 truth. 
 
 December^ 1853. 
 
INDEX OF ILLUSTRATIONS. 
 
 Those which are not worked into the text face the pages to which they are referred. Those marked 
 P. D. are from Mr. De la Motte's own Drawings. Those marked A. are Anastatics. 
 
 
 
 Page 
 
 
 
 Page 
 
 Chartres Cathedral . Frontispiece 
 
 
 Loches, View of Town . . . 
 
 
 26 
 
 
 
 1 
 
 LocHES, S. View of Church . . 
 
 P.D. 
 
 27 
 
 Interior of Autun Cathedral 
 
 . A. 
 
 4 
 
 Loches, Interior 
 
 P.D. 
 
 27 
 
 
 
 8 
 
 North Aisle, Loches . . . 
 
 P. D. 
 
 28 
 
 Tower of S. Front, Perigueux 
 
 
 9 
 
 Loches, N. Elevation .... 
 
 P.D. 
 
 28 
 
 Chateau d'Aste, Pyrenees 
 
 
 10 
 
 Loches, Ground Plan .... 
 
 P.D. 
 
 29 
 
 Chapel, near Maintenon . . 
 
 . A. 
 
 14 
 
 Loches, in West Porch . . . 
 
 P.D. 
 
 29 
 
 
 
 16 
 
 Loches, Capital and Corbels of the West- 
 
 
 
 
 17 
 
 ern Arch of the Central Tower 
 
 P. D. 
 
 30 
 
 
 , A. 
 
 18 
 
 Loches, Base and Capital . . . 
 
 P. D. 
 
 30 
 
 Etrechy, Details and Front 
 
 
 19 
 
 Loches, Section 
 
 P.D, 
 
 31 
 
 Notre Dame, Etampes, Exterior 
 
 P. D. 
 
 19 
 
 Loches, Hagioscope .... 
 
 P.D. 
 
 31 
 
 Notre Dame, Etampes, Interior 
 
 P.D. 
 
 19 
 
 Loches, Strings 
 
 P. D. 
 
 32 
 
 Notre Dame, Etampes, Interior 
 
 P. D. 
 
 19 
 
 Crypt, Loches 
 
 P.D. 
 
 33 
 
 
 
 20 
 
 Loches, Bracket, &c 
 
 P. D. 
 
 34 
 
 S. Jules, Etampes 
 
 
 21 
 
 Loches, Tomb of Agnes Sorel . 
 
 P. D. 
 
 38 
 
 S. Jules, Etampes, Details. (The Cut 
 
 
 
 P.D. 
 
 38 
 
 representing the Crockets is by 
 
 P. D.) 
 
 21 
 
 Loches, Sections and Details 
 
 P.D. 
 
 38 
 
 S. Martin, Etampes, Exterior . 
 
 P.D. 
 
 21 
 
 Loches, Parts of Steeples, and Part of 
 
 
 Meungs, Bay of Nave .... 
 
 P. D. 
 
 22 
 
 Interior, the lower Cut by P. D. 
 
 
 38 
 
 
 
 23 
 
 Loches, Sections 
 
 P.D. 
 
 38 
 
 Meungs, Interior 
 
 P. D. 
 
 23 
 
 Loches, N. E. View of Church 
 
 
 38 
 
 S. Nicholas, Blois. (Plan by P. D.) . 
 
 23 
 
 
 P.D. 
 
 38 
 
 S. Nicholas, Blois, Bay of Nave 
 
 P. D. 
 
 24 
 
 Beaulieu, Interior 
 
 P. D. 
 
 38 
 
 S. Nicholas, Blois, Capitals 
 
 P.D. 
 
 24 
 
 Capital of Pier, La Trinite, Angers 
 
 39 
 
 S. Nicholas, Blois, Lantern 
 
 P.D. 
 
 24 
 
 
 
 41 
 
 S. Nicholas, Blois, Details 
 
 P.D. 
 
 24 
 
 
 
 52 
 
 S. Nicholas, Blois, Capitals 
 
 P.D. 
 
 25 
 
 La Trinite, Angers, Compartment of 
 
 
 Chapel of the Castle at Amboise P. D. 
 
 25 
 
 
 
 54 
 
 West Doorway, Loches . . . 
 
 P.D. 
 
 26 
 
 La Trinite, Angers, S. E. View 
 
 
 54 
 
xxii 
 
 INDEX OF ILLUSTRATIONS. 
 
 Page 
 
 Plan of La Trinite 55 
 
 S. Martin, Angers 56 
 
 DouE, Remains of Church 57 
 
 DoTJE, Plan of Pier 57 
 
 Fontevrault, View of Church . . .58 
 
 Poitiers Cathedral,, S. E 59 
 
 Vaulting, from S. Manvieux .... 62 
 
 La Cite, Perigueux 63 
 
 Plan of a simple Domical Building . . 65 
 
 Plan of S. Front, Perigueux .... 66 
 
 S. Front, Exterior 67 
 
 S. Front, Dome in Roof 68 
 
 S. Front, Interior 69 
 
 La Cite, Perigueux, Cap supporting the 
 
 Transverse Arch, &c 72 
 
 La Cite, Exterior A. 72 
 
 La Cite, Base of Pier 73 
 
 Angouleme Cathedral, Cap of Pier . 74 
 
 Angouleme Cathedral, Exterior A. 74 
 
 Angouleme, Interior A. 75 
 
 Angouleme, Inside of Dome .... 75 
 
 Fleac, Interior 76 
 
 Fleac, Exterior A. 76 
 
 Roulet, Interior A. 77 
 
 Le Puy, Interior A. 78 
 
 Le Puy, Exterior A. 78 
 
 Model of S. Paul's A. 79 
 
 S. Mark's, Venice, Interior, T. Hill, Esq. 79 
 
 Roulet, Exterior 80 
 
 Rock Cottages, near Loches . P. D. 81 
 
 CoRMERY, Inside of Church 82 
 
 Tours, Towers of St. Martin's Abbey . 82 
 
 CoRMERY, Outside of Church . . A. 83 
 
 CoRMERY, Three Brackets . . P. D. 83 
 
 CoRMERY, Spire-Lights 84 
 
 CoRMERY, Spire of Abbey . . P. D. 84 
 
 MoNTs, Outside of Church ... A. 85 
 
 MoNTS, Dome under Tower 86 
 
 Artannes, Outside of Church . . A. 86 
 
 Artannes, Interior 87 
 
 Pont Ruan, North Door ...... 88 
 
 Valleres Church A. 89 
 
 Page 
 
 ViLLANDRY, Interior of Church ... 89 
 
 Villandry, Exterior A. 89 
 
 ViLLANDRY, Interior Construction . . 90 
 
 Savonnieres, Capital 90 
 
 S. Martin le Beau, Outside . . A. 91 
 
 S. Martin le Beau, Interior of Apse . 91 
 
 Chapel at Blere 92 
 
 La Croix, Blere A. 92 
 
 Blere, Outside of Church ... A. 92 
 CiVRAY (on the Cher), Inside of 
 
 Church A. 93 
 
 Chisseaux Church A. 93 
 
 CiVRAY, Outside of Church .... 93 
 
 AiGUEVivE Abbey 94 
 
 S. Denis, Amboise, Bases and Capitals 
 
 P. D. 95 
 
 S. Denis, Amboise, Capitals . P. D. 95 
 
 S. Denis, North Side ... P. D. 95 
 
 MoNTLouis, Part of Apse 95 
 
 S. Denis, Amboise, S. Transept, &c. . 95 
 
 S. George, near Tours 96 
 
 Rochecorbon Church .... A. 96 
 
 Cinq Mars, Outside of Church ... 97 
 
 Vernou, West End 97 
 
 Cinq Mars, Inside of Church, under the 
 
 Tower , 98 
 
 Langeais Church A. 98 
 
 Chanceaux Church and old Buildings 
 
 A. 98 
 
 S. Come, near Tours 99 
 
 RouziERES, Buttress, &c. of Apse . . 99 
 
 S. Come, Window in an old Chapel . . 100 
 
 Bassens, Central Tower 101 
 
 Montierneuf, Poitiers 102 
 
 S. HiLAiRE, Poitiers A. 103 
 
 Notre Dame, Poitiers .... A. 104 
 
 S. Jean, Poitiers A. 105 
 
 CiVRAY, Inside of Church 106 
 
 CiVRAY, Exterior 107 
 
 Courcome, Interior of Church . . . 108 
 
 Courcome, Exterior A. 108 
 
 Courcome, Part of Nave 109 
 
INDEX OF ILLUSTRATIONS. 
 
 Page 
 
 CouRCOME, Capital ... 
 
 
 110 
 
 BocHERViLLE, Normandy, Capital 
 
 of 
 
 
 
 
 110 
 
 La Palud, Inside of Church . . 
 
 
 111 
 
 La Palud, West Front .... 
 
 
 112 
 
 La Palud, General View . . . 
 
 A. 
 
 112 
 
 
 
 113 
 
 B assens, View of Church . . . 
 
 A. 
 
 114 
 
 La Sauve, View of Ruins . . . 
 
 A. 
 
 115 
 
 Bassens, Part of West End . . 
 
 
 115 
 
 La Sauve, Window in Tower 
 
 
 115 
 
 
 A. 
 
 116 
 
 
 A. 
 
 117 
 
 Agen, Interior of Church . . . 
 
 A. 
 
 117 
 
 
 
 118 
 
 
 
 118 
 
 Serignac Church 
 
 A. 
 
 118 
 
 SuRBAzoN, General View of Church 
 
 A. 
 
 119 
 
 SuRBAzoN, Part of the Interior, Apse of 
 
 
 
 
 119 
 
 Tarbes, Central Tower of Cathedral 
 
 
 120 
 
 MoRLAAS Church 
 
 A. 
 
 120 
 
 
 
 121 
 
 S. Emilion, Interior of Rock Church 
 
 
 122 
 
 
 A. 
 
 122 
 
 Pleinpied, View of Church . . 
 
 A. 
 
 124 
 
 MoRNAY, between Bourges and Nevers 
 
 
 
 A. 
 
 124 
 
 S. Etienne, Nevers, Interior . 
 
 A. 
 
 125 
 
 
 A 
 
 125 
 
 Roui, between Nevers and Autun 
 
 A. 
 
 126 
 
 
 
 127 
 
 
 A. 
 
 128 
 
 
 A. 
 
 129 
 
 
 A. 
 
 130 
 
 
 A. 
 
 131 
 
 Plan of S. Michel, le Puj . . . . 
 
 
 131 
 
 
 
 131 
 
 
 
 132 
 
 Uzerche, View of Town . . . . 
 
 
 132 
 
 S. Sernin, Toulouse, N. E. View 
 
 
 133 
 
 S. Sernin, S. Door 
 
 
 134 
 
 xxiii 
 
 Page 
 
 S. Sernin, NT. W. View ... A. 134 
 
 S. Sernin, Interior of Aisle . . A. 135 
 
 S. Sernin, Bay of Nave 135 
 
 S. Gaudens, Turret on North Side of 
 
 Church 136 
 
 S. Gaudens, General Exterior View A. 136 
 
 S. Gaudens, Interior .... A. 136 
 
 S. Gaudens, Triforium of Choir . . . 137 
 
 S. Pe, Pyrenees A. 137 
 
 S. Gaudens, Capital 137 
 
 Lescar, near Pau, Interior . . A. 138 
 
 PuY De Dome 139 
 
 Caen, S. Nicholas and Abbaye aux 
 
 Hommes 140 
 
 English Mouldings .... P. D. 140 
 
 English Mouldings .... P. D. 141 
 
 English Mouldings .... P. D. 141 
 
 English Mouldings .... P. D. 141 
 
 French Mouldings A. 141 
 
 French Mouldings A. l^^l 
 
 BouGivAL A. 142 
 
 Acheres A. 143 
 
 PoissY A. 143 
 
 Triel, Bay of Nave ....... 144 
 
 Vernouillet A. 144 
 
 Vaux A. 145 
 
 Epone A. 145 
 
 RoMiLLY, Brackets under the Tower 
 
 Arches A. 145 
 
 RoMiLLY, View of Church 145 
 
 Jumieges, Pier Arches 150 
 
 S. Saturnin's Chapel, S. Wandrille A. 150 
 
 Vaulting at Bernieres and Caen . . . 151 
 
 S. Pierre, Caen, Steeple ... A. 152 
 
 Abbaye AUX Hommes, S.W. Spire A. 153 
 
 AiRAN A. 153 
 
 BouGY . A. 153 
 
 ViLLONS 154 
 
 Thaon 154 
 
 Langrune, Exterior A. 154 
 
 Langrune, Inside of Steeple . . A. 155 
 
 Langrune, Interior of Church . A. 155 
 
xxiv 
 
 INDEX OF ILLUSTRATIONS. 
 
 
 
 Page 
 
 
 
 Page 
 
 
 A. 
 
 155 
 
 Porte d'Arroux, Autun . . . 
 
 A. 
 
 162 
 
 S. Gabriel, Belfry of Church . . 
 
 A. 
 
 156 
 
 Temple of Janus, Autun 
 
 
 162 
 
 
 A. 
 
 156 
 
 Roman Tower at Perigueux . . 
 
 
 163 
 
 
 A. 
 
 157 
 
 Palais Gallien, Bordeaux . . 
 
 
 164 
 
 
 
 158 
 
 Palais Gallien, from the Interior 
 
 A. 
 
 164 
 
 Porch of Dragey Church, and a Column 
 
 
 
 
 166 
 
 in S. Quentin, Normandy . . 
 
 
 158 
 
 Gateway at Rouen 
 
 A. 
 
 167 
 
 Details of Genets and Pontorson 
 
 
 158 
 
 
 
 168 
 
 
 A. 
 
 158 
 
 Sculpture on the Northern Doorway of 
 
 
 S. Leonard's, near Avranches 
 
 A. 
 
 158 
 
 the West Front of Rouen Cathedral 
 
 
 Pontorson, General View . . . 
 
 A. 
 
 159 
 
 
 
 169 
 
 Pontorson, Transom of Door . . 
 
 
 159 
 
 Figure from a Painted Window 
 
 
 
 
 
 159 
 
 C. Winston, Esq. 
 
 189 
 
 
 
 160 
 
 AuvERS, a Village Church on the Oise . 
 
 190 
 
 Roman Remains at Aries . . . 
 
 A. 
 
 161 
 
 A Village Church near Boulogne 
 
 
 199 
 
 
 
 161 
 
 
 
 205 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 COUTANCES. 
 
 CHAPTER I. 
 
 IT was most natural that when English antiquaries were first 
 enabled to turn their attention to the mediaeval architecture of 
 France, Normandy should attract their principal notice. They 
 would look to that province for the models from which our own 
 finest specimens are derived ; those magnificent designs which the 
 architects brought over by the Conqueror transferred to the cathe- 
 drals of England. Having studied the character of the early 
 Norman architecture, as presented in the two great abbeys at 
 Caen, the church of S. Nicholas in the same city, and the no less 
 
 B 
 
2 
 
 AKCHITECTURAL STUDIES IN FRANCE. 
 
 important structures of Jumieges and Bocherville, they would be 
 led to trace the progress of the art through a series of buildings of 
 the highest order, as Lisieux, Bayeux, Coutances, and many 
 others, to those master-pieces of the later styles, which adorn the 
 city of Rouen. The parish church of almost every town and 
 village, in some parts of the province, affords a specimen of beau- 
 tiful, and to the English antiquary, unusual, composition, or of 
 elaborate workmanship. I am not aware that we have in England 
 any thing analogous to the peculiar Norman steeple of the thirteenth 
 and fourteenth centuries, of which S. Pierre at Caen furnishes so 
 exquisite an example. We can hardly be surprised that an eccle- 
 siastical tour in Erance signified little more than a tour in Nor- 
 mandy ; with perhaps an occasional excursion to some of the best 
 known cathedrals in other provinces. 
 
 But if we confine ourselves exclusively to this interesting pro- 
 vince, we shall form a very incorrect notion of the general spirit 
 and character of the church architecture of Erance. The same 
 might perhaps be said of any other province ; for it is reasonable 
 to suppose that over so extensive a surface as that of the Erench 
 territory there must prevail numerous and strongly marked local 
 varieties. This is the case in the comparatively small area of 
 England — every county has its own peculiar architectural features, 
 and it would be necessary to examine buildings in several, before 
 we could form a fair idea of the English mediseval styles in general. 
 The remark however applies with still greater force to Normandy, 
 for this province and Brittany are in a manner insulated from the 
 rest of Erance. They present a kind of link between English and 
 French architecture, and furnish examples partaking much more 
 of the former than the latter. Eor instance, the Romanesque of 
 Normandy is very different from that of the southern and eastern 
 parts of Erance. Notwithstanding the later introduction of the 
 pointed arch, it exhibits, (in common with our own,) a transitional 
 tendency at a much earlier period. The Norman architect seems 
 always to have fixed his eye upon some future development. He 
 threw out his design, simple, and frequently even naked, in point 
 of ornament, and not seldom careless in actual workmanship, but 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 3 
 
 elaborate in construction, and carefully arranged in its proportions, 
 as matter for a succeeding generation to work upon. Perhaps he 
 foresaw that his actual work was destined to speedy demolition, that 
 it might make way for improvements suggested as it were by itself ; 
 and was willing to bestow his time, thought, and labour rather 
 upon an idea that would not only last, but grow and ripen into 
 excellence, than upon a mere fabric likely to be swept away in 
 the course of a century. 
 
 Now the southern architect of the eleventh and twelfth centuries 
 was content with his style as it existed. He neither looked for, 
 nor wished for, any further development. He gave up his atten- 
 tion altogether to refining and adorning it. I question if the 
 richest buildings in southern France of the middle of the twelfth 
 century show the slightest advance towards Gothic beyond those 
 of the tenth or eleventh — even the constant use of the pointed 
 arch, the introduction of which in the north almost instanta- 
 neously formed the style, proves, by its inefficiency in giving 
 the impulse, that the principle either did not exist, or was very 
 stagnant. 
 
 But if such a thing as a pure, beautiful, and correct round- 
 arched style lies within the bounds of possibility, (and I confess I 
 hope it may, for the sake of our own church architecture,) we can 
 nowhere better look for its elements than in the southern provinces 
 of France. We meet with specimens worked almost with the 
 classical delicacy of the Italian style. The shafts often have nearly 
 a pure Corinthian capital. The heavy cushion capital, so universal 
 in England, and so common in Normandy, is scarcely to be found 
 south of the Loire. Where an exceptional instance occurs, it may 
 possibly be attributed to English workmanship, as in the province 
 of Guienne, where I noticed its appearance in the abbey of La 
 Sauve, near Bordeaux. However the earliest Romanesque may 
 have originated in barbarous and unskilful imitations of Roman 
 architecture, the later Romanesque, when it showed an inclination 
 to imitate, was the very reverse of barbarous. We remark this in 
 places where the existence of Roman remains has influenced the 
 architecture of the twelfth century. At Autun, for instance, the 
 
4 
 
 AECHITECTURAL STUDIES IN FRANCE. 
 
 interior of the cathedral, with its fluted pilasters, and its triforium 
 arcade of round arches, evidently adopted from classical models, 
 may be easily compared with the Roman Porte d' Arroux ; and the 
 comparison will not be disadvantageous to the later work. I have 
 seldom seen an interior, of the size, of a more impressive efifect, or 
 more pleasing in its general arrangement and proportions ; and 
 the deviations from the Roman type, such as the pointing of the 
 pier arches, and the composition of the vaulting pilasters, show no 
 ordinary degree of thought. 
 
 There are three diff'erent styles or schools which the architect 
 will do well to study who wishes to revive (I do not mean, copy) 
 mediaeval architecture in the present day. Two of them I think 
 are peculiar to France, and the other is better represented in that 
 country than in any other. These are the styles of Auvergne, 
 of Perigord and Angoumois, and of Anjou. By the first of these 
 I mean that of which Notre-dame of Clermont Ferrand, and 
 S. Paul of Issoire are examples ; and of which the principal 
 characteristic is the barrel roof. This style in fact prevails through 
 a large extent of country, indeed it is the common type of the 
 French Romanesque. But in Auvergne it appears to me to be 
 worked up with a higher degree of finish than elsewhere, and to 
 have more the character of an independent style, besides which it 
 exhibits certain peculiarities in that province, which considerably 
 increase its efifect. I may also be influenced by some very bold 
 restorations at Clermont, Issoire, and Brioude, the success of 
 which shows that the style must have something beyond mere 
 antiquity to recommend it. The churches of Perigord and An- 
 goumois are without lateral aisles, and are roofed with a series of 
 domes ; those of Anjou, also without aisles, have square compart- 
 ments of cross vaulting, much raised at the apex. These styles 
 will be more fully discussed in separate chapters. 
 
 It is by visiting the churches of central and southern France 
 that we shall be able to form a better estimate of the excellencies 
 of our own styles. We can draw a comparison, if such it may be 
 called, which shall be to the advantage of both countries, and the 
 disparagement of neither. We shall see that English architects 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 5 
 
 took a prominent position in the advancement of the art, and that 
 instead of merely following in the steps of other nations, they 
 outstripped them in some respects, and this at a very important 
 period, in the progress to perfection. From what has been said, 
 it may be inferred that Romanesque elements would in most 
 parts of France linger considerably beyond the introduction of 
 Gothic. They did so through the whole of the thirteenth century. 
 Even in Normandy the square abacus continued to prevail up 
 to the very end of that century. Buildings coeval with our most 
 finished early English, would, if we met with them in this country, 
 be at once classed as transitional, and referred to the end of the 
 twelfth century. And I cannot help thinking that our antiquaries 
 must sometimes find it difficult to reconcile the apparent discre- 
 pancies, in characteristics indicating date, which they meet with 
 in French churches. 
 
 On the other hand, the squareness of the general sections, and 
 the monotony of the mouldings, while they betoken less advance- 
 ment in the elegancies of the style, give an expression of severe 
 dignity which cannot fail to produce its effect on the spectator. I 
 do not intend to enter into a comparison between the cathedrals 
 of the two countries ; this has been done by several writers with 
 much discrimination and judgment.* Perhaps in smaller churches 
 the contrast is still more forcible. The French architectural 
 
 * The comparison between Amiens and Salisbury is drawn with very great fairness 
 by Mr. Garbett, in his " Rudimentary Treatise on the Principles of Design in 
 Architecture," p. 200 and following, and illustrated by correct plans and general 
 views, though on a minute scale ; and his remarks are the more valuable, as they give 
 in fact the distinguishing characters, not only of these two particular buildings, but of 
 French and English architecture in general, after the eleventh century, as regards 
 churches of the first class in magnitude. I may here state, that I could not perceive 
 the difference of the two western towers of Amiens to be anything but intentional. I 
 may possibly have missed some indication, while I was looking for it; but I find the 
 note I made on the spot to be as follows. The towers of Amiens are pretty similar, 
 up to the string above the rose, and mostly of early pointed with the square abacus, 
 though with some introductions. Above this there is a difference, and the upper 
 stage of the north tower was evidently intended from the first to be the highest of the 
 two. The present restorers seem to think the same, as they are preserving the 
 difference. 
 
6 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 student would be as much struck, and learn as much, on visiting 
 a group of English churches, in some district where neither design 
 nor workmanship is spared, for instance, Lincolnshire or North- 
 amptonshire, as we should by a corresponding group of French 
 village churches. The former exhibit a depth, purity, and deli- 
 cacy in mouldings, a studied elegance in the tracery of windows, 
 a symmetry of outline, and refinement in proportions, with a care- 
 ful exclusion, from an early period, of every incongruous element, 
 which are not so apparent in the French specimens. But in the 
 latter, we find a massiveness and grandeur which we look for only 
 in our own large conventual churches. Most of them are either 
 wholly, or in part, vaulted with stone. Many have the main 
 features belonging to a cathedral ; the pier arch, the triforium, 
 and the clerestory, and these are so designed as to avoid giving 
 the idea of a model in miniature. The central tower is, I should 
 say, on the whole, more prevalent than in England. In some 
 districts it forms the rule rather than the exception, while in 
 others it is very rare. In Burgundy it is the property of all the 
 styles, from Romanesque to Flamboyant ; in Touraine, at least in 
 the part of that province with which I am at all acquainted, it is 
 apparently confined to the Romanesque, or early transition. As in 
 Germany, it is often octagonal, and the spire shows itself at an 
 early period. The apse is also a more common feature than in 
 England, though in many districts it by no means prevails to the 
 exclusion of the square chancel ; indeed, in the smaller Roman- 
 esque buildings in the north of France, I question if the latter is 
 not of more frequent occurrence. In southern France the apse 
 is more nearly universal. Where there are aisles or transepts, 
 the main apse is usually flanked by two smaller ones. This 
 arrangement, however, was not unusual in England, as is proved 
 by the large arch so frequently visible in the eastern wall of a 
 transept. The apsidal aisle with radiating chapels is of frequent 
 occurrence in the larger Romanesque churches ; in the succeed- 
 ing styles it became commonly the eastern finish of cathedrals. 
 Where later additions have been made, the most unexpected and 
 picturesque combinations present themselves. The large and lofty 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 7 
 
 chancel is added to the nave designed for a smaller structure, or 
 the raised clerestory and roof of the nave almost swallow up the 
 low central tower. The original nave becomes an aisle to a nave 
 of later date, and the central tower becomes a lateral one. Aisles 
 and chapels are added with but little reference to the scale of the 
 building, and frequently have the effect of a series of transepts 
 joining each other and terminating in gables. I do not say that 
 these are architectural beauties, but they increase the interest 
 both to the student and the artist. 
 
 The little I have to say upon the French cathedrals may as 
 well be introduced here as in any other place. The central 
 tower, it is well known, does not in general occur in those of the 
 first class (I put Normandy out of the question). The prominent 
 features presented by the outline at a distance are the pair of 
 western towers, and perhaps a small wooden spire at the inter- 
 section. But many of the architects, it is plain, were not quite 
 satisfied with this meagre substitution for a central tower ; they 
 designed an equivalent, which, though inferior to the western 
 towers in height, might exceed them in aggregate mass, number, 
 and extent. At Chartres we see something of the intention, though 
 it is not fully carried out. Take the outline at a distance, and we 
 have only the western steeples, and a line of roof unbroken by any- 
 thing rising above it. But within this outline are masses which if 
 raised a little higher would give amazing richness and variety to 
 the edifice, and even as it is, they are very striking in the nearer 
 views. Each transept is flanked with a square tower, as at 
 Rouen, besides which are additional towers of smaller dimensions, 
 attached to the choir aisles. None of these however, are carried 
 in height beyond the clerestory wall. I have not had an oppor- 
 tunity of visiting Laon, where the towers of the transept form a 
 striking feature in the general outline.^' The unfinished cathedral 
 of Clermont Ferrand has a tower flanking the front of one of its 
 transepts. The whole design probably comprehended six towers. 
 
 * I see from engravings that this cathedral has a low central tower ; those flanking 
 the western and transept fronts being more lofty. 
 
8 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 all of which would be seen at a distance. Bordeaux cathedral has 
 the four transept towers well carried up, and the two northern 
 ones have fine spires ; these, with the polygonal choir, w^hich 
 much exceeds the nave in height, form a remarkably fine group. 
 There are no western towers, but a detached campanile stands 
 near the east end. As it is difficult to obtain a good view of this 
 cathedral, I may mention that the tower of S. Eulalie, at no very 
 great distance to the south, commands it well. 
 
 As they appear in cathedrals, the geometrical style of the four- 
 teenth century is extremely beautiful and elegant, and the later 
 Flamboyant is rich, varied, and imposing. Perhaps its finest speci- 
 mens excel our best Perpendicular. But in smaller churches it is 
 remarkable how much the interest falls ofi* at the end of the thirteenth 
 century, and how very poor and meagre the late Gothic becomes 
 when attempted to be worked plain. In this respect we have the 
 advantage. Our plain Perpendicular is not necessarily a poor style. 
 However simply it may be worked, there is generally something, 
 perhaps not easily described or defined, which connects it with 
 the Gothic of the best period. Even when the piers are plain 
 octagonal shafts, the mouldings of the arches mere chamfers, and 
 the windows terminating in a square head, instead of a pointed 
 arch, we still feel that it is Gothic, and that it has not been 
 worked without some thought or care. We can hardly say this 
 of a plain late Gothic church in France. The meagre pointed 
 windows, often without mullions or tracery, generally without 
 foliation, and the discontinuous imposts without capitals, have an 
 effect hardly counteracted by the vaulting, the ribs and ornaments 
 of which often appear rather to be the additions of an indiscreet 
 restorer than the productions of an age coming within the Gothic 
 range. We may occasionally meet with a steeple of good outline, 
 but rarely (except in Normandy and Brittany, where the above 
 observations do not apply) shall we find any to be compared with 
 the examples that may be met with in most English counties. 
 Such a group of towers as we find in Somersetshire, is, I should 
 think unknown in France. 
 
 But if the interest ceases earlier than with us, it begins earlier. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 9 
 
 Our ante-Norman architectural remains are far too scanty to 
 enable us (without further assistance) to establish any style or 
 system, and even those of a higher date than the twelfth century 
 are by no means numerous. We have but few buildings, of which 
 the style may be said to be at all formed, earlier than the reign of 
 Henry I. The Romanesque of the eleventh century is a grand, 
 pure, decided style throughout France ; and many buildings to 
 which a higher antiquity is reasonably assigned, though rude in 
 the execution of their ornaments, are not without pretension to 
 architectural excellence. A study of the French churches known 
 or supposed to belong to a date prior to the year 1000 is very 
 necessary if we would trace the gradual development of the me- 
 diaeval styles. But as in the following chapters my attention will 
 be turned to buildings of the earlier, rather than of the later 
 periods, I need not dwell on this subject at present. 
 
 TovvEK OF St. Front, Perigueux. 
 C 
 
10 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 Chateau d'Aste, Pyrenees. 
 
 CHAPTER II. 
 
 The reader must forgive my frequent introduction of the shortest 
 and most odious monosyllable in the English language. For 
 unless I say something of my own views', as well as my oppor- 
 tunities of making observations, I may lead him to suppose I am 
 directing his attention to the most interesting specimens to be 
 found in a province or district, instead of those which suited 
 my own particular purpose, or which happened to be within my 
 reach. My object then has been to study principally, if not 
 exclusively, the styles preceding the development of Gothic in 
 the thirteenth century, and this, as I have already hinted, with a 
 view of considering whether they contain the elements of a pure, 
 fixed, and permanent round-arched style. In examining speci- 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 11 
 
 mens I have had an eye rather to composition and construction, 
 than to ornament or detail ; and consequently have often visited 
 extremely simple and unpretending structures in preference to 
 those of greater richness and magnificence. The time I have 
 devoted to the study of outline and general effect has often 
 precluded me from the examination of features which would 
 have a prior claim on most antiquaries ; and this must account for 
 my silence upon many of those accessaries to architectural splen- 
 dour which with some constitute the chief beauty and interest of a 
 church, as monuments, sculptural decoration, paintings and glass. 
 Where I have been induced to dwell for some time on the exterior 
 of a building, I have often glanced very hastily through the interior, 
 and perhaps forgotten cloisters, crypts, and a host of curiosities, 
 which a little more time and trouble would have enabled me to 
 visit. My reader then must not suppose that I am about to say 
 all that can be said upon every building I introduce to his notice, 
 or to press upon him all its claims to attention, nor yet that I am 
 pointing out every remarkable example that occurs within the 
 circle in which he finds me engaged ; he must rather look upon 
 my small harvest as an earnest of the much richer one which 
 awaits a more extended and laborious research. 
 
 As to the roughness of my sketches, I will, instead of apologising 
 for it, practically acknowledge and atone for the defect, by a 
 liberal sprinkling from the portfolio of my friend Mr. P. H. De la 
 Motte, the accuracy of whose pencil is known to most antiquaries. 
 By his assistance I am enabled to lay before the reader illustra- 
 tions, as full as can be expected in a work like the present, of the 
 very curious church of Loches, in Touraine ; a building, which, 
 as far as I can ascertain, has never had full justice done to it in 
 the way of engravings, though several antiquaries have appreciated 
 its great value and interest. It well deserves such a series of 
 drawings as have been given to Chartres, Noyon, and other 
 remarkable French cathedrals. 
 
 However carefully a tourist may have laid down his plan before- 
 hand, he will probably at his return from his tour acknowledge 
 that he has been indebted for some of its most satisfactory results 
 
12 ARCHITECTURAL STUDIES IN FRANCE. 
 
 to deviations he did not originally contemplate. For my own 
 part, I may safely say that if matters unconnected with my archi- 
 tectural pursuits had not influenced my route, I should have 
 missed, to say the least, considerable variety. I might perhaps 
 have nearly exhausted some one province, Normandy for instance, 
 and remained ignorant of much that has interested me elsewhere. 
 I find I have noticed, more or less cursorily, upwards of three 
 hundred French churches. Of these I could not easily name 
 fifty of which my previous knowledge, from descriptions or en- 
 gravings, has at all influenced my movements. I have either 
 found out by means of local works, or enquiries, or a reference on 
 the spot to Murray's Hand-book (which has given me far more 
 assistance than I could have anticipated from a work so limited 
 in its size) what objects w^ere likely to be of value to me in any 
 district through which I passed, or else I have fallen on them 
 accidentally. I need not say that I missed a considerable number ; 
 of this I was made aware whenever I compared notes with any 
 traveller over the same ground ; but I seldom failed to meet with 
 sufficient to occupy my full time and attention. And we very 
 often prize more what we pick up by a fortunate accident, than 
 w^hat we obtain with much labour and forethought. The tourist 
 has no reason to complain, who having started with the intention 
 of visiting Normandy and Brittany, finds himself unaccountably 
 in Poitou or Perigord. With the help of a tolerable map, a 
 descriptive account of the neighbourhood, such as may be procured 
 in most of the principal towns in France, and a good pocket- 
 telescope, especially if he can add to these the services of an 
 intelligent driver, he need seldom come home disappointed from 
 his morning's excursion. At Caen, M. de Caumont's Statisque 
 Monumental de Calvados* enabled me to plan my excursions with 
 great certainty. At Avranches, I found my telescope useful. At 
 Tours, I could throw myself altogether upon the judgment of my 
 voiturier, M. Charles Souillet, whom if I were merely to mention 
 as an intelligent guide, I should only say what I might of many 
 
 * See Preface. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 13 
 
 others, but I must in justice add that I found him an agreeable 
 companion, conversant in the ancient history of his country, as well 
 as the objects of curiosity within his reach, and extremely anxious 
 to make me acquainted with them. Had I remained any length of 
 time at Angouleme, I should have found the Statisque Monumental 
 de la Charente (to which I shall have occasion to refer hereafter*) 
 most valuable. At Bordeaux, I should doubtless have found 
 Guienne Monumentale " * equally useful, but it is more volumi- 
 nous, and the objects of interest lie in general at a greater distance 
 from the town. 
 
 As I have been less able to give attention to objects seen on my 
 road from one place to another, than to make excursions in 
 different directions from places where I remained for a time, 1 
 may possibly offer something new to travellers who have passed 
 through the same districts, but without any divergence except to 
 well known objects. It might be supposed that by selecting as a 
 principal object some well-known specimen of remarkable grandeur 
 and beauty, we should be sure of meeting in its neighbourhood 
 much that is worth our notice ; many minor satellites to the great 
 luminary. This is often the case, but not uniformly. I saw 
 nothing* that struck me in the immediate neighbourhood of Chartres. 
 Around Evreux and Lisieux (themselves presenting magnificent 
 examples), nearly all the village churches are small, plain, rough in 
 their masonry, with no belfry beyond a simple wooden spire on the 
 roof. Indeed there are parts of Normandy, through which one 
 might travel for a whole day, without finding any reason to name 
 it as a province celebrated for its ecclesiastical antiquities. This 
 struck me especially between Mortain and Falaise, on which journey 
 however the richness of the scenery quite compensated for the 
 want of other attractions. 
 
 A little geological knowledge w^ould considerably help the 
 tourist. M. de Caumont, in the admirable work I have already 
 referred to, shows how the churches even of adjacent districts are 
 affected by the materials found on the very spot ; though I must 
 
 * See Preface. 
 
14 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 say that the excellence of the stone, and the ease with which it is 
 worked, has been productive of evil as well as good ; for there is 
 scarcely a church near Caen that has not suffered from modern 
 barbarities or restorations. In the department of the Oise there 
 are fortunately railroads, and a considerable demand, in conse- 
 quence of works carried on at a distance, for the fine building stone 
 of that district, so that the group of country churches it presents, 
 perhaps exceeded by no other in France, is just now undergoing 
 little or no injury, though most of the older specimens have large 
 additions of the fourteenth or fifteenth century. 
 
 We ought also to know something of the history of the country. 
 The storm of the Revolution, while it swept away many of the 
 finest edifices, seems to have spared, or lightly touched, the humbler 
 structures. At Avranches, not a vestige of the magnificent cathe- 
 dral is left, but there is abundance of matter in the valleys below. 
 I suspect the religious wars of an earlier period were more 
 unsparing. In some parts of the south, churches of a mediseval 
 date are very thinly scattered, and of these many are partially 
 destroyed, and replaced by more modern structures. 
 
 I went one morning by the train from Versailles as far as 
 Maintenon, to look at the remains of the aqueduct commenced 
 and left unfinished by Louis XIV. Having given it as much 
 time as I wished, I found I had still an hour or two on my hands. 
 I had not seen any object likely to interest me from the line of 
 railroad near the station at which I had stopped, nor did the 
 country seem to promise much. However, I walked along a road 
 by the side of a small stream that falls into the Eure, and in little 
 more than ten minutes came upon a building which had I been in 
 quest of any further object, I might have passed without much 
 notice, supposing it to be a pigeon-house, or some office belonging 
 to the farm to which it was attached. On looking at it however 
 more carefully, I perceived it was a circular chapel of the twelfth 
 century ; very simple, and not in the best condition ; but of a 
 plan which I believe to be extremely rare ; having no inner con- 
 centric ar.cade, such as we see in the four English round churches. 
 The only similar specimen I am acquainted with in England is 
 
ARCHITECTUEAL STUDIES IN FBANCE. 
 
 15 
 
 the chapel in Ludlow Castle. The building in question has an 
 eastern apse, and a round-headed western doorway, enriched with 
 a few mouldings. No vaulting at present remains, and I could 
 not ascertain whether there had ever been any. The specimen 
 struck me as deserving of more attention than, from its secluded 
 situation, it is likely to receive. 
 
 But I mention it just now, first, because I am not likely at any 
 future time to refer to the same part of the country, and secondly, 
 that I may make what some will perhaps think a needless remark. 
 If we happen to have no other guide which may lead us to the 
 antiquities of a district, we may often, by following the course 
 of a stream, obtain a very fair chance of finding them. The 
 larger rivers speak for themselves ; every one of them has at inter- 
 vals its cities and important towns, provided of old with their 
 apparatus of walls, gates, castles, cathedrals, churches, and monas- 
 teries. The only question is, how much has been destroyed, and 
 how much left. Not only these, but their smaller tributaries have 
 their objects of interest ; and it might be curious to observe how 
 each seems to carry along its course some peculiar architectural 
 feature. I have fancied, though I might find it difficult to point 
 out, a slight distinction between the churches on the Loire, the 
 Cher, and the Indre, in their passage through Touraine. In one 
 respect I have certainly noticed a difference ; namely, in their 
 orientation. I remember on starting from Tours one morning, to 
 have observed that the sun was strong on the south side of the 
 cathedral and the churches of the town ; but on my arrival, a full 
 hour later, at the church of S. Martin le Beau, I found that the 
 light had not left the north side of that and other churches on the 
 banks of the Cher. Now, though the course of both rivers is, to 
 speak roughly, from east to west, yet there is necessarily a con- 
 vergence to the point where the one falls into the other, and I 
 cannot help thinking that the churches have their orientation 
 according to their respective rivers ; those on the Loire pointing 
 rather to the north, those on the Cher considerably to the south. 
 I merely throw out the suggestion, not having extended my obser- 
 vations sufficiently to establish a rule. 
 
16 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 In my excursions through this district I have unquestionably 
 met with the best specimens, and in the greatest number, on and 
 near the rivers ; and when I have traversed the dry tract of land 
 to the north of the Loire, I have found comparatively little to 
 repay me. But we need not dwell upon this point. Who is 
 there that has not in his rambles tracked the course of some stream 
 from village to village, at one turn catching a glimpse of an old 
 steeple as it rises from among the trees, at another, of a picturesque 
 manor-house crowning the slope of the bank ; here he meets with 
 a sample of nature s own architecture, with its arched recesses, or 
 gray unbroken wall roofed with moss and ivy ; there, telling a tale 
 of scarcely less remote antiquity, a shattered relic of the primseval 
 forest. Depend upon it, there is a great analogy between the 
 pursuits of the fly-fisher, the artist, and the antiquary. 
 
 -Si- 
 
 WiMILL. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 17 
 
 MONTLHERV 
 
 CHAPTER III. 
 
 Paris is naturally the most convenient point whence to diverge ; 
 and architecturally it is perhaps one of the best. That is, by 
 taking points easily reached in different directions, we find the 
 most strongly marked distinctions. The Seine, towards Rouen, 
 has a character of its own. The cylindrical pier, neither very 
 tall nor massive, is common. The Romanesque work often 
 exhibits, instead of an inner order of a square section, a large 
 torus ; such as occurs in the earlier parts of Ely cathedral ; and 
 which probably led the way to that system of mouldings which 
 characterizes the English early Gothic, in which the diagonal or 
 oblique surfaces preponderate over the cardinal ones. On the 
 
 D 
 
18 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 Oise, near Criel, tlie churches have much that prepares us for the 
 Angevine style. The piers are massive and clustered ; the 
 vaulting compartments are often nearly as wide in the direction 
 of the nave as transversely to it, and the orders, however moulded, 
 retain a general squareness of outline. To the north-east, (east- 
 ward of the Oise) I did not fall in with any Romanesque specimen ; 
 but Gonesse, apparently of the thirteenth and fourteenth centuries, 
 presents a new and beautiful arrangement of triforium ; this may 
 be a unique example. On the Marne, S. Maur near Vincennes, 
 seemed to promise characteristic features in that direction. In this 
 church, as in many others, a Romanesque belfry rests on a de- 
 cidedly Gothic substructure. 
 
 But as my business is at present chiefly with the southern pro- 
 vinces, I shall at once make for the Loire, a river which on its 
 own banks or those of its tributaries exhibits some of the most 
 interesting churches in France, and if I mistake not, in a great 
 part of its course forms an important architectural boundary. Its 
 nearest and most northern point is Orleans. By ascending the 
 river we give ourselves the opportunity of studying that phase of 
 the Romanesque which appears, as I have remarked, to most ad- 
 vantage in Auvergne, through which province runs the Allier, one 
 of the principal tributaries of the Loire ; by taking the other 
 direction, and following the course of the river towards the sea, 
 we come upon a group of great interest, as regards style, construc- 
 tion, and antiquity, in the province of Touraine, and become 
 acquainted with the Angevine style. I shall at present take the 
 latter route, and on leaving the river shall shape my course to the 
 southward, that I may study the domical churches of Angoumois 
 and Perigord. 
 
 I am always attracted by a central tower ; whether that the 
 outline simply pleases me, or that it promises to interest me 
 either in its construction or by its antiquity. Be this as it 
 may, I have seldom found cause to regret my partiality. I there- 
 fore gave up an hour or two to Etrechy, a small village near 
 Etampes, about half-way between Paris and Orleans. The church 
 has a nave with aisles, transepts, a central tower, a chancel with a 
 
Etrechy. 
 
 Page 19. 
 
Notre Dame, Etampes. 
 
Notre Dame, Etampes. 
 
 Page 19. 
 
AKCHITECTURAL STUDIES IN FRANCE. 19 
 
 t 
 
 square termination, and flanking apses. The style is principally 
 early pointed, perhaps transitional, with additions and insertions 
 of a later date. It is very plain as regards ornament, but well 
 worth study for the sake of comparison with the churches north 
 of Paris. The whole is vaulted, with ribs, and the apses are 
 groined in cells, with ribs and a boss ; the side windows of the 
 chancel are in couplets, plain and slightly pointed. Those of the 
 nave aisles are single. The nave has three bays, the piers are 
 cylindrical with a square abacus, on which rests the vaulting 
 shaft, which is single. There is no triforium or clerestory. An 
 arcade of three arches below each window (now much mutilated,) 
 enriches the aisle. The west window is a triplet of plain lancets, 
 and the west door is pointed. The vaulting abacus throughout is 
 a portion of an octagon, though the foliage in the capitals below is 
 early. In some of the pier capitals the foliage is extremely bold. 
 The leaves are put on without much arrangement in the way of 
 grouping, and seem to be true, though rough copies of nature. 
 There is a small plain crypt with unribbed cross vaulting under 
 the chancel. The tower piers have rectangular sections. To a 
 person wishing to study the character of a building rather than its 
 ornamental details, this church is valuable. 
 
 Etampes contains four churches, all of great value to the student. 
 On glancing at them, which I regret to say I did much too hastily, 
 I fancied I saw in them a sort of frontier line between the northern 
 and southern architecture of France. Notre Dame has a western 
 steeple of great beauty, consisting of a tower, with very lofty 
 pinnacles, and a spire. The composition, as regards the spire - 
 lights and pinnacles, is in effect not very unlike those of the thir- 
 teenth and fourteenth centuries in Normandy, but the detail is 
 entirely Romanesque, and all the arches in the upper stages, or 
 from the level at which the tower disengages itself from the front, 
 are round headed. The spire is ribbed and ornamented with 
 scales. Notwithstanding the details, I have little doubt, from the 
 composition, that this steeple belongs, in date, to the advanced 
 Gothic. The manner in which the upper octagonal stage of the 
 tower harmonizes with the spire lights, and is connected by the 
 
20 AECHITECTUEAL STUDIES IN FRANCE. 
 
 pinnacles both with the square base be- 
 low and the spire above, is worthy of an 
 architect that could design the western 
 towers of Coutances, or the beautiful 
 steeple of S. Pierre at Caen. And the 
 substructure, if I recollect right, is purely 
 Gothic. The church has aisles and tran- 
 septs. The bays of the nave are very 
 wide, so as to make the vaulting compart- 
 ments nearly square. There is no diago- 
 nal rib. The arches are pointed. The 
 general effect of the interior somewhat 
 prepares us for the Angevine churches. 
 S. Basil has a low central tower of early pointed. Each face 
 has a pair of belfry windows, separated by a cluster of shafts, and 
 subdivided by a single shaft with obtusely pointed arches of one 
 plain square order. The principal arches of this belfry are of two 
 orders with tori, and surmounted by a label. The west front is 
 Romanesque, with a rich door. The greater part of the church 
 has been rebuilt, or cased, with flamboyant work, and has little 
 interest. There is some late painted glass. 
 
 S. Jules, though also much altered, retains more of its ori- 
 ginal work. Its principal feature is its very curious central tower, 
 remarkable both from its construction and outline. The object of 
 the architect has been to adapt, at the intersection of the transepts, 
 a square tower, narrower than either the nave, chancel, or 
 transepts. The base is square ; visible above the roof of the 
 nave, but absorbed by the transepts and chancel. From the 
 angles rise triangular slopes, as for the support of an octagon ; on 
 these, as well as on the space left on each of the faces of the tower, 
 stand equal gables ; four cardinal, and four diagonal. The points 
 of the diagonal ones support the angles of a smaller square tower, 
 the faces of which fall behind the gables resting on the sides of 
 the base. I have not seen the internal structure, none of the 
 squinches being visible inside the church ; but it is easy to 
 imagine a system containing two tiers of squinches. The angles 
 of the tower are shafted, and the belfry windows pointed. The 
 
Page 20. 
 
S. Jules, Etampes. 
 
 Page 21. 
 
S. Jules, Etampes. 
 
 S. Jules, Etampes. ' 
 
 S. Jules, Etampes. 
 
 Page 21. 
 
AECHITECTURAL STUDIES IN FRANCE. 
 
 21 
 
 upper part terminates with a gable over each face, enriched by a 
 curious sort of crocket evidently of early character. The tower 
 piers are plain and massive, cruciform, with square sections. The 
 arches are pointed, but these may have been altered. The nave 
 has five aisles, of which the outer ones are flamboyant, and present 
 externally a series of gables ; no uncommon arrangement. The 
 west door is Romanesque, and has an arch of three orders boldly 
 moulded with torus and hollow, the two outer orders supported by 
 shafts set in re-entering angles. The label has a kind of large 
 head-nail, forming a rough hexagonal pyramid. This door forms 
 a shallow porch connected with the wall of the front by a small 
 slope, above which is a plain round-headed window with a label. 
 There are some good bases and capitals in the nave. 
 
 St. Martin's is much purer and freer from alterations. It is 
 early pointed, or transitional, and consists of a nave with aisles, 
 small transepts not extending beyond the aisles, and a semi-circu- 
 lar apse, also having an aisle, from which project three radiating 
 apses. The radiation of apsidal chapels, in the Romanesque and 
 transitional styles, we shall find to be more common in the central 
 and southern provinces of France than in the north. In Auvergne 
 it is very striking. The piers of this church are low, massive, 
 and clustered, and the abacus is square. Most of the arches 
 are pointed, but there are some round-headed windows in the 
 apsidal aisle. The triforium consists of two pointed arches in 
 each bay, which are again subdivided by a shaft. They do not 
 form an open gallery, but have sufficient projection to allow of 
 shafted jambs. The clerestory is plain. In the apse we find 
 coupled columns, like those in Canterbury cathedral. The capi- 
 tals are rich and striking, but none, I think, worked with any 
 degree of minuteness. The choir has the original ribbed vaulting. 
 That of the nave is of a later period, and of wood. There is no 
 central tower, but a flamboyant western tower has been added, 
 which stands clear of the west front. The outline of this tower is 
 picturesque, as it has four gables like that which we have already 
 noticed. 
 
 The tourist would do well to give up a few hours to this old 
 
22 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 town, which is encompassed by very pleasing scenery. The keep 
 of the castle remains ; its ground plan is something like that of 
 Clifford's Tower in York. 
 
 From the line of railroad between Etampes and Orleans we 
 observe several village churches which appear to be worth notice. 
 Some have the central tower, others a tower engaged in one of 
 the aisles. The tower with a gabled roof is still common ; it 
 becomes rarer as we proceed southward. 
 
 Orleans Cathedral is certainly an impressive structure, both 
 externally and internally, though not without the faults which 
 might be expected from the late period to which it belongs. The 
 horizontal line prevails too much in the west front, and the circu- 
 lar stage on the top of each of the western towers is not in perfect 
 accordance with the Gothic character. Nevertheless the compo- 
 sition, as seen from a distance, is well proportioned and striking. 
 The mouldings of the piers run through the architraves, and are 
 not stopped by any capitals. The triforium is very well arranged, 
 consisting of an arcade of four small trefoil arches, occupying a 
 very limited space (in point of height) as compared with both the 
 pier arches and the clerestory. The mouldings have not much 
 force or boldness, but this is a common defect in flamboyant 
 work. 
 
 The church of S. Aignan, consisting of the polygonal choir, 
 and the transepts, of a large church, is a good example of flam- 
 boyant. 
 
 In this style also is Notre Dame de Clery, about nine miles 
 from Orleans, a large and lofty church with polygonal choir and 
 short transepts. It has no central tower, but a low belfry, not 
 equalling in height the roof of the nave, is attached to the north 
 aisle. The choir has flying buttresses, and internally the vaulting 
 ribs of the apse, or rather the portions of the vault itself, which 
 in this position become mere walls near the springs, are pierced, 
 and enriched with tracery. The triforium space is blank, which 
 gives an air of plainness to the interior not common in large 
 buildings of the style. The east window has some fine painted 
 glass of the sixteenth century. 
 
Page 22. 
 
Meungs. 
 
 Page 23. 
 
Meungs. 
 
 Page 23. 
 
Page 23. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 23 
 
 The next church we notice, Meungs, on the north bank of the 
 Loire, at a short distance from Clery, is an excellent specimen of 
 the late Romanesque or Transition. This is a cross church, but 
 he tower is western instead of central. The ground plan is 
 curious, as the transepts, as well as the choir, terminate in semi- 
 circular apses, besides which the transepts have each an apse to 
 the eastward, and another apsidal chapel projects from the north 
 aisle of the nave, near the northern transept. The arches are all 
 pointed, the piers are clustered, and the roof has the cross vaulting 
 with ribs. There is no triforium. The square abacus is used 
 throughout, and the capitals are enriched with foliage. There is 
 no apsidal aisle. The tower has pinnacles and a spire, and some 
 castellated remains adjoining the church add much to its pic- 
 turesque appearance. 
 
 Still more have the southern influences prevailed in the church 
 of S. Nicholas of Blois. Externally this church has a striking 
 outline, though perhaps more picturesque than grand. It is 
 difficult to obtain a satisfactory view of it, on account of the 
 buildings by which it is enclosed. It may be seen however 
 pretty well from the bridge, and perhaps from some of the 
 higher parts of the town to the northward, and there is also 
 a tolerable view from a small court or alley near the east end 
 on the north side. It has two unequal western towers, one of 
 which has lately been well restored; indeed rebuilt; and a low 
 plain central tower with a wooden roof and small spire. From 
 the shortness of the nave these form a very compact group. The 
 nave has aisles ; the choir, which terminates in an apse, has an 
 aisle with three radiating chapels, of which the eastern one is of 
 late date, and the transepts have each an eastern apse. The nave 
 consists of four bays, independently of the compartment belonging 
 to the western towers. The choir has two bays besides the apse, 
 which is supported by seven narrow arches on single shafts. The 
 transepts have two bays, both of which are absorbed in the aisles 
 of the choir, which has two on each side ; consequently the 
 transeptal apses are separated from the transepts (properly so 
 called) by a space of two bays. The arches, as in the specimens 
 
24 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 above noticed, are mostly pointed, (the style being transitional) 
 but a few windows are round-headed. The pier arches of the 
 nave are of two orders, the inner one square and plain, the outer 
 having a hollow and torus. The eastern arch however of the 
 nave has only one plain order. The two piers on each side, 
 nearest to the western towers, are cylindrical, with four very bold 
 shafts, the front one of which runs up as a vaulting shaft, and is 
 banded at the spring of the arch, and also above and below the 
 triforium range. Above the spring of the pier arch it has two 
 smaller shafts on each side. The triforium in the three western 
 bays of the nave is an arcade of five pointed arches, each with one 
 order having a torus on the edge, resting on single shafts with 
 capitals and square abacus. The clerestory (now blocked up) has 
 two large plain pointed arches, and a large circle. The bay 
 nearest the central tower has only three arches in the triforium, 
 and a window of a single light above. The western pier of this 
 bay, as well as those of the tower, has a series of re-entering angles 
 with shafts, such as we commonly see in our own Norman build- 
 ings. The roof has cross vaulting with ribs. Most of the capitals 
 are exquisitely carved, and much resemble the Corinthian in their 
 workmanship, though not perhaps in proportion. Though I have 
 set down this church as transitional, I have now no doubt that the 
 greatest part of the nave belongs to the pointed style of the thir- 
 teenth century, and is as much developed as usual in French 
 buildings. The ornament we call the dog-tooth appears in this, 
 and other buildings, perhaps of a date as early as the twelfth 
 century, but I do not recollect ever to have seen it in France used 
 as it is in England, namely, to enrich a deep hollow. It generally 
 occurs on the lower surface of a simply chamfered label. 
 
 But the most important feature in this church is the central 
 lantern (if we can call that a lantern which is not lighted by any 
 windows) beneath the square tower at the intersection. It is 
 circular, and consequently rests on Byzantine pendentives ; that 
 is, triangular surfaces belonging to a hemispherical dome, which 
 if complete would rest upon a circular base equal in its diameter 
 to the diagonal of the square contained between the piers of the 
 
S. Nicholas, Blois. 
 
Page 24. 
 
Page 24. 
 
S. N'iCHOLAS, Blois. 
 
 S. Nicholas, Blois. 
 
 Page 24. 
 
S. Nicholas, Blois. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 25 
 
 tower. We shall have to speak more of this sort of pendentive 
 when we come to the churches of Perigord and Angoumois. The 
 lantern, or cupola, is vaulted in eight cells, the edges being ribbed. 
 Each cell has a pointed arch subdivided by a shaft. I should 
 have observed that the choir of this church has flying buttresses, 
 which, if belonging to the original work, must be early examples, 
 as the eastern part, setting aside additions and insertions, must 
 belong to the twelfth century. The student proceeding south- 
 wards ought to give careful attention to this church, as well as 
 those of Meungs and Etampes, as lying on or near the boundary 
 between districts occupied by very different schools or styles. 
 
 Amboise. 
 
 £ 
 
26 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 West Doorway, Loches. 
 
 CHAPTEE IV. 
 
 The mere admirer of the picturesque, if he should chance in his 
 travels to light upon such a town as Loches, in Touraine, will 
 assuredly be unwilling to leave it in haste. It rises up the 
 side of a steep rocky slope on the southern bank of the Indre, 
 and assumes, to the spectator approaching it from Blois and Mont- 
 richard, the form of a pyramid ; the faces of which are diversified 
 by numerous towers, turrets, and gateways, and the summit 
 
Page 26. 
 
Page 27. 
 
Page 27. 
 
 i 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 27 
 
 crowned by the massive Norman keep of the castle, and the 
 fantastic outline of the collegiate church of St. Ours. It is not 
 often that he can have noticed, within so small a compass, so 
 striking, I may say so startling a contrast between two buildings 
 of so nearly the same date. The castellated structure diflPers in 
 no respect materially from our own Norman keeps ; that, for 
 instance, of Rochester Castle. Its plan may perhaps be a little 
 more oblong, and the flat pilaster buttresses, common in Norman 
 work, have in this example massive shafts engaged, one in each, 
 forming a semi-cylinder. The subordinate group belonging to the 
 castle, though containing many features of interest, does not affect 
 the simplicity of outline as seen at a distance. The church, on 
 the contrary, is all broken into points and angles ; it does not 
 exhibit so much as the horizontal line of a nave roof ; indeed our 
 first impression is to regard the whole group as belonging to a 
 west front, the body of the church being concealed behind the 
 houses. This group consists of two steeples of nearly the same 
 height, though different detail, flanked, the one of them by an 
 apsidal projection, the other by a low rectangular porch, while 
 the intermediate space is occupied by two large octagonal pyra- 
 mids of stone, in contact, or nearly so, with each other, as well as 
 with the belfry towers. These towers might, as I remarked, very 
 well be the flanking towers of a front ; they do not, I think, stand 
 farther apart, relatively to their height and bulk, than those of 
 Rouen cathedral, or Wells, which comprehend aisles as well as 
 nave. The intermediate pyramids, however, puzzle us ; we begin 
 to consider how such a superstructure can be combined with the 
 usual arrangement of a church ; but just when we have given up the 
 problem in despair, a nearer approach solves the riddle in a 
 somewhat unexpected manner, by shewing that what we have 
 taken for the front, is the side ; that the flanking towers are in 
 reality a central and a western tower ; and the two pyramids of 
 stone form the roof of the nave, not only externally, but, as we 
 shall presently see, internally also. The transepts are short, and 
 have a lean-to roof, like aisles. The apsidal projection, which I 
 have mentioned, is the eastern apse of the church, and this is 
 
28 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 flanked by smaller ones, projecting eastward from the transepts. 
 The nave has aisles, but would be equally complete without them, 
 
 North Aisle, Loches. 
 
 for we shall find, on examining the interior, that they are merely 
 independent rooms or chapels, not communicating with the main 
 body by a regular structural arcade, but opening into it by occa- 
 sional doorways. The north and south sides of the western tower 
 are free from these excrescences, and rise clear from the ground ; 
 but on the west side is a large porch, of the same style with the 
 rest (namely, late Romanesque), of a square plan.* 
 
 * Dates from Tablettes Chronologiques de I'Histoire du Chateau et de la Ville de 
 Loches. — 962. GeofFroy Grise-Gonelle, compte d'Anjou, de Vendorae et seigneur de 
 
Page 28. 
 
ARCHITECTURAL STUDIES IN FRANCE. 29 
 
 We will begin with this porch in describing the several parts of 
 the church according to the ground plan. 
 
 (I Western porch. Each face has an enriched Romanesque 
 
 LocHES, IN West Porch. 
 
 door, with a round head. The parapet has a modern balustrade. 
 The vaulting of the interior is in one square compartment, pointed, 
 
 Loches, fils de Foulques le Bor, lui succeda; il fit construire a son retour de Rome 
 I'eglise collegiale de Loches, sous Piiivocation de Notre Dame. Le roi Lothaire donna, 
 le 18 Septembre 962 a Laon, les lettres patentes pour la construction de cette eglise. 
 Hardouin, cinquante-cinquieme eveque de Tours, consacre la nouvelle collegiale en 
 963.— Maan. Hist. Eccl. Tur. 74. N° 2 Dom Martenne. 1180. Thomas Pactius ou 
 Paccius, auteur d'une Chronique manuscrite des Comtes d'Anjou, que j'ai vue a la 
 BibHotheque du roi, etait prieur de I'eglise collegiale de Loches depuis plus de qua- 
 rante ans, lorsqu'il fit construire presque entierement et a ses frais Veglise actuelle du 
 chateau, en 1180, on voit son buste sculpte dans le mur au-dessus de la porte de 
 I'eglise, du cote du midi. 
 
30 
 
 AKCHITECTURAL STUDIES IN FRANCE. 
 
 and much domed, with very bold ribs. A fine Romanesque 
 door, with a semicircular head, leads into the church. 
 
 h Western tower. Externally square to about the level of the 
 roof-ridge (supposing the nave to have been finished in the ordi- 
 nary manner), above which it becomes octagonal, having a round 
 headed belfry window in each face. The cardinal ones consist of 
 
 Capital and Corbels of the Western Arch of the Central Tower. 
 
 three plain orders, without shafts ; the diagonal ones of two orders, 
 set in a gabled projection, like that of a spire-light. The whole is 
 crowned by a spire resting upon the cornice. The interior com- 
 partment beneath the tower has a cylindrical barrel vault (the 
 axis longitudinal), with ribs of a square section, externally there 
 
LOCHES. 
 
 Page 30. 
 
Page 31. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 31 
 
 are flat pilaster buttresses to the square part. The arch into the 
 nave corresponds with this vault in shape. 
 
 c c The nave, divided into two square compartments by a 
 pointed arch of one square order ; each of these compartments is 
 covered by an octagonal pyramid, the same shell of stone forming 
 both the exterior and interior roof. The diagonal faces are 
 brought down into the angles of the square, by flat slopes, resting 
 on enriched brackets, supported by shafts, with capital and abacus. 
 The pointed arch line between the nave and central tower is 
 preserved ; but the actual arch opening into the tower is a fine 
 semicircular one, resting on shafted piers, which project sufficiently 
 to allow an arch to be pierced through, on each side of the main 
 arch, to form hagioscopes. 
 
 Hagioscope, North Side cfr Central Tower. 
 
 d Central tower. It rests on four round-headed arches, of three 
 square orders (as seen from inwards), the inner one of which is 
 supported by a bold shaft and capital. The tower compartment 
 is roofed with an octagonal cupola, on Romanesque squinches, 
 
32 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 that is, the round arch and concave surface over each angle. 
 Externally this tower is square, and has in each face three bold 
 shafts, with capital and abacus supporting the cornice ; two of 
 these are near the edges, the other in the middle of each face, 
 dividing it in two portions. Each compartment has a round- 
 headed window of three square orders ; the two outer ones shafted, 
 with a slight label. At the angles are somewhat massive octagonal 
 pinnacles, with round arches in their faces, and the whole is 
 crowned with an octagonal spire without windows. This tower 
 has been much restored, but, it appears, correctly, as far as the 
 main architectural lines are concerned. 
 
 e Apsidal chancel. The part next the tower has a cylindrical 
 vault, the apse itself semi-domical. The windows are round- 
 headed. 
 
 f f Small apses flanking the principal one. 
 g g Transepts, with cylindrical vaults. 
 
 This eastern part is probably the oldest portion of the build- 
 ing, and anterior to the date generally given as that of the erection 
 of the church. 
 
 h h Nave aisles. I think these are Eomanesque, as well as a 
 chapel on the south side of the choir ; one to the north of the 
 nave is Flamboyant. But, as we have observed, the aisles, what- 
 ever their date, can hardly be considered as comprehended in the 
 design of the nave, into which they open irregularly. 
 
 There is much excellent work upon the capitals ; but instead of 
 describing this, or the very curious sculptured font, or the crypt, 
 I must refer the reader to Mr. De la Motte's sketches.* My very 
 
 * See note at the end of the chapter. 
 
ARCHITECTURAL STUDIES IN FRANCIE. 
 
 .33 
 
 limited time was fully occupied in taking such drawings of the 
 exterior, and rough notes, as might enable me to preserve the 
 general outline and the most striking features. 
 
 When we meet with a unique building, such as that before us 
 assuredly is, we are tempted to speculate how it came to be built 
 at all, and why it has not been taken as a pattern for other build- 
 ings in the neighbourhood. 
 
 Now if we draw a line from north to south through Loches, 
 extending it northwards as far as the Loire, and southwards pretty 
 nearly to the Pyrenees, we shall find that westw^ard of this line 
 
 Crypt, Loches, under South Apse. 
 
 are a considerable number of important churches which have the 
 same peculiarity of plan, viz. a wide nave, independent of aisles ; 
 I mean, either without aisles altogether, or with such as might 
 
34 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 have been omitted without in the least affecting the character of 
 the church. I should say that I am now speaking of buildings of 
 the Romanesque or transitional period ; that is, earlier than the 
 beginning of the thirteenth century. We shall see presently that 
 this is the case with some of the most interesting village churches 
 
 LOCHES. 
 
 near Tours, and with larger structures on and near the Loire to 
 the eastward of that city. It is the plan of Angouleme Cathedral 
 and the group of domical churches derived from Perigueux, and of 
 Bordeaux Cathedral. There may be a few eastward of the line 1 
 have drawn, but at no great distance from it, and I think most of 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 35 
 
 the large churches beyond will be found to exhibit the usual plan 
 of a regular nave with aisles. 
 
 The roofing of a large undivided nave would naturally exercise 
 the ingenuity of architects. We know that the cross vault, cover- 
 ing very large spaces, was constructed by the Romans ; and it 
 also appears in large Komanesque buildings in Germany, where it 
 is worked in square compartments. But in the south of France 
 it is not, I think, used generally, except in the aisles of churches, 
 before the beginning of the twelfth century. I have seen too little 
 of French architecture to be able to say that the cross vaulting sel- 
 dom or never occurs in early Romanesque naves (south of the 
 Loire), but most of the roofs I have seen are barrel roofs ; that is, 
 where, the nave being furnished with aisles, the span is not very 
 great. We shall see that a barrel roof, supported by the abutment 
 of a quadrantal roof over the triforium, with a cross vaulting to 
 the aisle itself, is a common arrangement. But when the nave 
 exceeded a certain width, it is possible that the plain barrel roof 
 might be felt to be inconvenient, insecure, or ungraceful. In 
 most of the village churches which I have adverted to, near Tours, 
 the roof is a mere timber ceiling. But this expedient was not 
 likely to satisfy the architect of a large, and, as he might hope, a 
 permanent structure. Conical stone roofs to towers larger in 
 area than mere turrets, had probably shown themselves for some 
 time previous to the date of the building under consideration ; 
 they are, I believe, common in Poitou. Notre Dame, of Poitiers, 
 offers an example. A square pyramidical roof, also of stone, is 
 probably an early feature ; of this some churches in Normandy, 
 of the Romanesque period, afford good specimens. It is likely 
 that the octagonal spire, set on the square base, made its appear- 
 ance later than these ; but still, if we can judge from architectural 
 features, it appeared at least as early as the middle of the twelfth 
 century. One at Cormery, a few miles from Loches (of which we 
 shall have to speak shortly), must be of that date, if not earlier. 
 Both the Romanesque towers of Loches itself have stone octagonal 
 spires, one on a square, the other on an octagonal base. The 
 
36 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 steeple of Beaulieu church, hardly a mile from Loches,* also 
 Romanesque, has a spire ; indeed I believe some of the earliest 
 types of the common church spire, so constant a feature in the 
 later styles, are to be found in this province. The architect, 
 then, of the nave of Loches divided it into two square compart- 
 ments, and applied to each of them the spire then coming into 
 use, with no alteration beyond a mere expansion in form. Indeed 
 spires crowning towers are not uncommonly found in precisely 
 the same proportion. The octagonal spire would be an early 
 improvement on the square pyramid, on the mere score of light- 
 ness, since the diagonal face is less than the parts of the cardinal 
 faces for which it is substituted. Also a greater part of the thrust 
 might be transferred to the angles, which are more easily and 
 conveniently strengthened than the sides. The squinch here used 
 bears some analogy to the Byzantine pendentive, and, like it, 
 accommodates itself in a certain degree to the arch necessarily 
 introduced in two at least of the cardinal sides. The construction 
 therefore seems to be good mechanically, and also convenient, as 
 regards its accessaries ; and it is economical, as not requiring a 
 second shell of roof internally or externally, though whether an 
 internal cupola with a curved surface may have been originally 
 intended, admits, I think, of a doubt. I do not know what effect 
 this roof has as regards sound. 
 
 If we consider, with M. Verneilh, that there is a connection 
 between the roofs of Perigord, which consist of a series of domes, 
 and the domical cross vaultings of Anjou, this church (which he 
 adverts to in his work) is interesting as a kind of link between the 
 two. It certainly occupies an intermediate position in point of 
 date, for it is later than the earliest specimens of the Perigueux 
 dome, and earlier than the Angevine vaultings. To what extent 
 it was influenced by the one, or exercised influence on the other, 
 it is impossible to say, but it cannot be passed over by any one 
 
 * See note at the end of the Chapter. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 37 
 
 who wishes to enter fully into the examination of the two different 
 styles. 
 
 It may be asked, why a form certainly not deficient in simplicity 
 of construction, or in durability, should not have been repeated 
 for the purpose of roofing large areas, as it has been for towers 
 without number. I have remarked upon a former occasion, that of 
 the two kinds of support required in a Gothic building, namely, the 
 vertical support, and the lateral abutment, the former belongs 
 principally to the interior, the latter to the exterior of a building. 
 An internal buttress,* exhibited as such, is manifestly out of place ; 
 and although a series of arches is necessarily a system of abut- 
 ments, yet the eye ought rather to be drawn to the vertical than 
 the lateral means of support. The springs of the arches should 
 generally be vertical; sloping straight lines, such as those of angular 
 canopies, should be avoided, at least on a large scale. I do not 
 recollect ever seeing them over pier arches ; if they do occur in 
 this position, I am sure it must be very rarely. I hardly like 
 them over the triforial arches at Amiens and Evreux. The seg- 
 mental arch of the triforium at Salisbury has always struck me as 
 a defect ; nor do I like the Tudor pier arches at S. George's, 
 Windsor. In both cases the idea of abutment is too strongly 
 forced upon the mind. On this principle the sloping sides of a 
 pyramid, and the decided angles formed at its junction with the 
 walls, appear unsuitable for an interior, though quite in character 
 with the exterior of a building, which demands no concealment of 
 buttress. That this distinction between the proper treatment of 
 the exterior and of the interior was felt by ancient architects is, I 
 think, plain, from the rarity of instances in which the rule is 
 disregarded. The spire of a steeple does not constitute an in- 
 ternal feature ; it is always hidden by a roof. The arch, vault, 
 and dome are generally, and properly, made to rest, in all appear- 
 
 * The internal buttresses of large cathedrals are, to the eye, no more than divisions 
 between the several chapels of the aisles. 
 
38 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 ance, as much as possible, on vertical supports, the lines being 
 brought, by a gentle sweep, to a eoincidence in direction with 
 shafts and imposts perpendicular to the horizon. Therefore the 
 roof of Loches, where this is not done, forms an exceptional case ; 
 distinct, in an important principle, from the domes which precede 
 it in date, and the vaults which follow it. 
 
 Notes referred to at pages 32, 36. 
 
 Loches — The Crypt is situated under the south apse; it is small, but has the 
 appearance of being original. 
 
 The Sculpture throughout the building is excellent, and the construction of the 
 masonry is of a very interesting character, particularly the corbelling of the nave roof. 
 
 The Font is of a cylindrical form, its ornaments are of an unusual character. 
 
 Beaulieu. — This Church is situated about one mile east of Loches. It is a beau- 
 tiful specimen of the architecture of the period. The proportions of the tower and 
 spire are extremely good. 
 
Font, Lochks. 
 
 Tage 38. 
 
LocHES, W. Porch. 
 
 Page 38. 
 
LocHES, Central Steeple. 
 
 LocHES, Western Steeple. 
 
 LoCHES. 
 
LocHES, Pier of Central Tower. 
 
 Page 38. 
 
i 
 
 V 
 
 Beaulieu. 
 
 Page 38. 
 
Beaulieu. 
 
 Page 38. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 39 
 
 Cap in La Trinitk, Angers. 
 
 CHAPTER V. 
 
 Take a straight horizontal line a b, and with centre a, radius a b, 
 describe the circular quadrant b i. 
 Draw the straight line a i at right 
 angles to a b, meeting the circle in 
 I. At B and i draw the tangents 
 b l and i l, meeting at l. These 
 will be respectively perpendicular 
 
 to A B and A I, and a b l i will be / 
 
 / 
 
 a square. Join a l, and with cen- 
 tre A, radius a l, describe the quad- 
 rant M L K, meeting a b produced I 
 in M, and a i produced in k. m b a 
 
 Now this figure is a projection, on a plane surface, o a section 
 
40 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 of a sphere, of which a k is the 
 radius ; b i represents the quad- 
 rant of a small circle, formed by 
 cutting the sphere with a plane 
 parallel to the plane of the paper, 
 and distant from the plane of its 
 parallel great circle m k by a line 
 equal to b a. 
 
 L will also represent a point 
 corresponding with i, in a small 
 circle formed by a plane cutting 
 the sphere at right angles to the paper, and passing through b and 
 L. Also a quadrant of a great circle passing through k and b 
 will be represented (on the plane of the paper) by a quadrant of 
 an ellipse b k, of which the semiaxes are respectively a b and 
 
 A K 
 
 which, 
 
 :a B^ + B l" or 2 A b", are in the ratio of 
 
 1:^/2. This drawing will therefore represent in elevation part 
 of a hemispherical dome, cut by four equal semicircular arches, 
 which touch at their springs ; the half of one of these is given in 
 piano by the quadrant b i ; and the half of the adjacent arch is 
 projected into the straight line b l. These may be considered as 
 the transverse and longitudinal semi-arches of a square vaulting 
 compartment, and the quadrant represented in projection by the 
 elliptical quadrant b k, a diagonal rib, taken from the spring to 
 the apex of the vault, and circular, like the transverse and longi- 
 tudinal arches, instead of being elliptical, which must be the case 
 when the ridges of the intersecting vaults are horizontal. So that 
 if to a square vaulting compartment we make both the longitudinal 
 and transverse arches semicircular, and the diagonal also semi- 
 circular, we may construct a vault which shall be a continuous 
 spherical surface, without edge, nook, or change of curvature ; and 
 if the diagonal rib be introduced, it will lie along such surface, 
 and not occupy any edge or nook. 
 
 This vaulting, without the diagonal ribs, occurs in the aisles of 
 the very curious abbey church of Bernay in Normandy, a work 
 belonging to the early part of the eleventh century, and it is 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 41 
 
 Vault, Bernay. 
 
 strictly domical in its construction, 
 the stones being laid regularly in 
 horizontal courses. 
 
 It also appears with diagonal ribs, 
 as well as ridge ribs, that is, coincid- 
 ing with portions of great circles 
 drawn through the highest point of 
 each arch, at the intersection of the 
 transept in S. Pierre, Saumur. I take 
 for my authority a work of great value 
 to the antiquary, to which I shall 
 have occasion to make frequent refer- 
 ence L' Architecture Byzantine en 
 France, par M. Felix de Verneilh, Paris, Libraire Archeologique de 
 Victor Didron, 13, Rue Hautefeuille, 1852"). The example at 
 Saumur is figured as " Coupole a nervures." That of Fontevrault 
 (similar to the Bernay vault) Coupole sans pendentifs distincts." 
 
 But this construction does not satisfy one condition that we might 
 look for in cross vaults ; viz. that each vaulting cell be formed, so to 
 speak, by the progress of the smne arch from the end or side, to 
 the centre of the vaulting compartment ; so that wherever we may 
 take a vertical section, parallel to the arch, it should exhibit 
 exactly the upper portion of that arch ; that is, if a semicircle, part 
 of a circle of the same diameter, if pointed (of circular arcs) an 
 intersection of equal circles, giving the same angle. 
 
 If in the diagram already 
 
 used, we take a point o in b k, 
 the representative of the diago- 
 nal rib, and with centre a and 
 radius a o describe the circular 
 arc o N, cutting a k in n, o n 
 will be half the arch formed by 
 a vertical section of the cell be- 
 longing to the arch b i, at the 
 point o (if we treat the dome as 
 a vault of four cells) and it will 
 
 G 
 
 .A 
 
42 
 
 AECHITECTURAL STUDIES IN FRANCE. 
 
 M 
 
 B 
 
 K not be a part of any arch equal 
 ^ and similar to that of which the 
 quadrant b i is the half, for the 
 ^ arc o N is part of a circle having 
 a longer radius than a i ; whereas 
 to satisfy the above condition it 
 ought to have been described 
 with a radius exactly equal to it. 
 And the ridge of a vault having 
 the same longitudinal, transverse, 
 and diagonal ribs, but satisfying 
 
 the aforesaid condition, will not coincide with an arc of the great 
 circle of the sphere, but will lie somewhat above it. Now, in- 
 stead of the arc o n, described as before, let us consider an arc o p,* 
 described with a radius o q or q p, equal to a i. This it is clear 
 satisfies the condition, for o p is an exact reproduction of the 
 upper part of the quadrant b i, being part of an equal circle. 
 But the line o q may be conceived as drawn from the point o in 
 the circular quadrant represented by the elliptic quadrant b k, 
 in a plane parallel to the paper, to a point q in the plane of the 
 great circle vertical to the paper, and of which a k represents 
 merely the edge. And the locus of q upon that plane, that is the 
 line in which q will be found wherever we may take the point o 
 in the line b k, so that o q shall always equal a i, will exactly 
 give the form of the line required for the ridge of the vault, 
 between i and k, and which, equally with the circular arc, is 
 represented in projection by the straight line i k. In short, that 
 the point p must be found in a similar line, commencing from i, 
 with that of q, commencing from a, as every point in the one is 
 determined by a vertical line of a constant length from a corre- 
 sponding point in the other. 
 
 In the diagram we have used, draw o v parallel to k a, and q v 
 parallel to a b or perpendicular to o v. Now, since the plane in 
 which we suppose q to be taken, and the plane of the paper, or of 
 the quadrant m k, are at right angles to each other, the point v 
 
 * See the Diagram in the next page. 
 
AECHITECTURAL STUDIES IN FRANCE. 
 
 43 
 
 in the latter corresponds with the point Q in the former, and the 
 locus of Q in the one will be de- 
 signated by the locus of v in the 
 
 other (for o v would be the pro- 
 jection or representation of o q 
 if the plane now vertical to the 
 paper was to be taken in eleva- 
 tion flat-ways). We may there- 
 fore now treat the problem as 
 one of plane geometry, consider- 
 ing the line b k as an actual 
 ellipse, and so find the curve 
 B V w, in which v, corresponding to q, is always to be found. 
 
 Join A Y and produce o v to x. 
 
 Because o qzza b=a y, 
 
 And Q vzza X, and q v o and a x y are both right angles, 
 It is clear that the triangles q v o and a x y are equal and similar. 
 Therefore o vzzx y, and o yzzx v, 
 Hence x y : x v : : x y : o y. 
 
 But by the property of the ellipse, xy:oy:: ai:ik. 
 Consequently, x y : x v : : a i : i k, and therefore v lies in an 
 ellipse, of which the semi- axes equal a i and i k. Or if a w be 
 cut ofi* equal to i k, and a quadrant of an ellipse be drawn, having 
 for its semi-axes a and a w, the point v will be found in that 
 ellipse, and a similar curve, formed by erecting parallel vertical 
 lines, equal to a b, from each point between b and w, will give 
 the outer curve l k, the required ridge of the vault. And a 
 vaulting so constructed will have its diagonal rib situated (as 
 usual) on an edge. 
 
 The above is an easy problem enough, and I fear I shall be 
 thought to have worked it out in a clumsy and tedious manner, 
 but I thought it important to show how it may be done without 
 confusing the eye by perspective representations of oblique planes, 
 which constitute the great difficulty in comprehending problems of 
 solid geometry. 
 
 We will now consider the case of a vault with pointed arches. — 
 In the quadrant i b take a point f, draw f d perpendicular to A b, 
 
44 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 
 / F 
 
 
 it 
 
 
 
 M G 
 
 B G perpendicular to a b, and g f per- 
 pendicular to F D. With centre a and 
 radius a g describe the circle c g e. 
 Produce d f to e and g f to l and join 
 A F. Also describe the circular quad- 
 rant M K as before, making 
 A M : A B : : ^2 : 1 . 
 
 Now by the property of the circle, 
 
 E FX(f D-fE d) = F GX(f L + L g) 
 
 Or (e D — F d) (e d + f d)=:(l g — l f) (l g + l f) 
 
 =(a b — a d) (a b+a d) 
 Therefore e — f d^=: a b^ — a 
 But A B=A F and a f^zza d'^+f 
 Consequently a b^ — a d^zzf 
 Therefore e d^ — f d^=f or e f d- 
 
 And ED : F D : : y/2 : 1 
 Hence the point e will be found in the quadrant of an ellipse, of 
 which the semi-axis major is a k and the semi-axis minor a b. 
 
 And the elliptic arc b e will be the projection of a diagonal 
 vaulting rib to a pointed vault, of which the halves of the longi- 
 tudinal and transverse arches are represented by f b in piano, and 
 B G in projection. 
 
 According to the present construction the diagonal rib b e will 
 lie along a continuous spherical surface, b g e f, but the lines e g 
 and E F will lie in pointed ridges or nooks. 
 Produce now a f to h and join he. 
 
 F D : F E : : A I : I K : : A F : F H. 
 
 Therefore e h is parallel to a b, 
 and the arc m h is evidently simi- 
 lar and equal to that of which 
 B e is a projection, and the figure 
 M H N, seen in piano, is the same 
 as that seen in projection in the 
 form of B e D ; and consequently 
 the angle at h, between the arc 
 and the vertical line, is the same 
 s£ ^ f5 ^ with that of the diagonal rib at 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 45 
 
 E. But as the arcs h m and f b are concentric, and cut off by the 
 same radii, the arc f b is similar to h m, and the angle at h, 
 between the arc and vertical line, is the same as that at f between 
 the arc b f and the vertical line f d. The arch therefore formed 
 by the diagonal ribs at the apex of the vault is a pointed arch 
 similar to that of the cells. 
 
 To find the shape of the ridge, so that the point of the arch 
 formed by a vertical section may not vary in its angle and the 
 curve of the arcs by which it is formed, we proceed much as before. 
 Let Q R be drawn equal and parallel to a f, and describe the cir- 
 cular arc R p, cutting the elliptic curve that represents the diagonal 
 rib in p, and it is plain that the angle at r, between the arc and 
 vertical, is exactly the same as that at f, and both arcs (belonging to 
 equal circles) of the same curvature ; therefore p r is the section 
 of the cell which meets the conditions, and the locus of q in the 
 plane perpendicular to the paper 
 must be found. If o p be drawn 
 perpendicular to a b, cutting the 
 circle b i in v, and q t to o p, t 
 will be the point in the plane of 
 the paper corresponding to q. 
 Join a v. 
 
 Now in the figure p q is equal 
 to A V, Q T to A o, and the tri- 
 angles p Q T and V A Y equal 
 and similar ; hence, as before, 
 p vzzo t. 
 
 But pv:ov::ik:ia (by property of the ellipse), therefore 
 ot:ov::ik:ia. Consequently t lies in an ellipse whose 
 semi-axes are respectively equal to i k and i a or a b ; and if we 
 draw the elliptic quadrant b s, having taken a szzi k, the portion 
 B w, cut ofi* by the vertical line f d, will give the form of the ridge 
 required. 
 
 If the point r be taken at e the two parallelograms will coin- 
 cide, and A Q will become equal to e f, consequently d w will be 
 equal to e f, and the curve will pass, as it should do, through the 
 apex of the pointed vault. 
 
 
 R 
 
 
 
 
 
 K 
 
 
 
 T 
 
 
46 
 
 ARCHITECTUEAL STUDIES IN FRANCE. 
 
 An oblong compartment may be covered by a spherical dome 
 as well as a square one, and if we convert it, as above, to cellular 
 vaulting, preserving the same conditions, we shall arrive at much 
 the same result. 
 
 Let M K be a quadrant of the great circle of the sphere, repre- 
 sented in piano, c i half the arch cutting it on the plane of the 
 paper, and c n the projection of half the transverse arch, whose 
 direct representation would be b h. It is clear that the tangents 
 can be drawn as in the figure annexed, viz. n h equal and perpen- 
 
 dicular to l b, and i l equal to 
 ^ c N, and perpendicular to it, if 
 both be produced. Then if the 
 larger arch corresponds with the 
 plane of the paper, the elliptic 
 quadrant c k is the projection 
 of the diagonal rib ; but if the 
 smaller arch be placed in the 
 same plane, b l will represent 
 o the larger one, and b k the di- 
 ^ c X B O -A agonal. 
 
 To find the necessary form of the ridge we take a point p in the 
 ellipse c k, draw p v x vertical, join v a, and draw p q parallel to 
 A V — we must determine the curve in which the point q is found in 
 the plane vertical to the paper. In the figure above, draw p r 
 parallel to m a, cutting the ellipse b k in r, draw r o vertical, 
 and Q T horizontal. Then, supposing the smaller arch to be set 
 on the plane of the paper, r will represent the point P, and t the 
 point Q. p V is equal to a q and o t. 
 Now by property of ellipse 
 
 p V : X p : : I K : A K 
 Therefore ot:or::ik:ak 
 Or o R : A K : : o T : I K 
 
 But o R : o Y : : A K : a H 
 Or or:ak::oy:ah 
 Therefore ot: ik:: oy:ah 
 Or o T : o Y : : I K : A H 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 47 
 
 Consequently t is in an ellipse whose semi-axes are a b and a line 
 1=1 K ; or the ridge itself will be represented by the quadrant of 
 an ellipse whose semi-axes are l i and i k. In the same manner 
 the other ridge will have for its semi-axes n h and h k. 
 
 Where the arches are unequal, but similar, pointed ones, the 
 result is of the same nature, and the working out the proposition 
 involves no difficulty, though the diagram may be somewhat more 
 complicated. I will therefore leave it to the ingenuity of the 
 reader. 
 
 You may ask why I have taken the trouble of going into this 
 matter minutely, knowing as I must that the conditions I have 
 stated are rarely, if ever, observed accurately, or the line, for the 
 finding of which I have given a rule, carefully drawn. 1 will 
 try if I can explain myself. I consider the refinement of a style 
 to depend in great measure either upon its observance, or recog- 
 nition, of particular rules. Some of these may be somewhat arbi- 
 trary, though of course none contrary to reason. And where the 
 designer provides for the strict mathematical observance of certain 
 laws, expressed or understood, he shows a higher degree of refine- 
 ment than when he professes to be content with a mere approxi- 
 mate observance, even though the latter should neither compromise 
 safe construction, nor produce unsightly forms. For instance, 
 suppose it is required to adapt a cylindrical tower or dome to a 
 square substructure. One method would be, to construct Roman- 
 esque squinches as for an octagon, across the four angles, and 
 again across the obtuse angles of the octagon, and so on. Having 
 obtained a polygonal base of sixteen, or at most thirty-two sides, 
 we may easily fit on the circle, without loss of strength, and 
 perhaps without forcing on the eye any want of adjustment. Still 
 any one would consider the design coarse and inartificial, as 
 compared with the Byzantine pendentive,* which provides for an 
 
 * It may be observed that the Byzantine pendentive has the same lines and surfaces 
 as the cushion capital, of which Mr. Ruskin shows the rationale (Stones of Venice, 
 vol. i. p. 104,) and would nearly serve, in miniature, as a mould for it ; the only differ- 
 ence being-, that in the one case the horizontal circle touches the vertical semicircle, in 
 the other, it is removed to a short distance, leaving a small band of the sphere unbroken. 
 
48 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 accurate adaptation of the two forms to each other. And it would 
 also be felt, that however careful and accurate might be the 
 workmanship in the former construction, or however rough and 
 incorrect in the latter, the relative merit of the two is unchanged ; 
 the latter, with all its imperfections, belongs to a higher class of 
 buildings than the former. On the same ground I look upon any 
 system of vaulting which is reducible to a strict mathematical 
 rule, providing for the exact observance of certain laws, that are 
 in some cases indispensable, in all productive of beauty and regu- 
 larity,* as a mark of great refinement in architecture, even though 
 in practice they may not be strictly observed ; as a fault in execu- 
 tion is far more pardonable than a coarseness in design. 
 
 In the construction of the common horizontally ridged vaulting, 
 Mr. Garbett shows that the elliptical diagonal rib, necessarily 
 occurring in designs when the transverse and longitudinal arches are 
 composed of circular arcs, was really used for a very short time ; 
 that the false ellipse, made up of circles of different curvature was 
 soon substituted, and that this substitution was actually an improve- 
 ment. It may have been so in strength, it certainly was so in 
 economy of labour, and it probably was so even in beauty, as 
 harmonizing better with actually circular lines than the true 
 ellipse. But this does not alter my view of the case. The imita- 
 tion of the true curve proves that its propriety was recognised. 
 Had the original designer of a cross vault with horizontal ridges 
 made the diagonal rib a mere segment of a circle, he would have 
 been guilty of a barbarism, and the attempt to do this gave rise to 
 those ploughshare surfaces which Mr. Garbett notices.! The use 
 of the true elliptic arc in its proper place was a great refinement in 
 that department of architecture, and one all the more valuable 
 from the ease with which a substitute, equally effective in every 
 respect, is worked. One great excellence of exact design is, that 
 
 * The condition provided for in the above problems, namely, that the section of the 
 cell should always be a part of the same arch, is one that insures a continuous curva- 
 ture, pleasing to the eye, in the surfaces of the vault ; though a departure from it, to 
 a slight extent, may not produce any noticeable effect. 
 
 t See note at the end of the chapter. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 49 
 
 it often provides for a wide license as regards inexact workman- 
 ship. An architect building a tower would take care that the 
 centre of gravity of the whole mass should, in his calculations, 
 stand directly over that of the section at the base, or as nearly so 
 as possible ; because this arrangement allows the greatest possible 
 latitude for accidental errors and disturbances ; and the quantity 
 of these which may be admitted without destroying equilibrium is 
 the fair measure of the designed stability. The mathematician 
 designs a cycloidal arc as fit for the oscillation of a pendulum, a 
 parabolic surface as the best for a concave reflector, and he will 
 perhaps enable the workman to produce these accurately ; but he 
 also shows that the more easily formed circular arcs and spherical 
 surfaces, are, where small portions are used, so nearly identical 
 in form and effect, that for all practical purposes the substitution 
 may safely be made. 
 
 But in the works of nature, the highest, as well as those the 
 nearest to our reach, we perceive the law of exact design, inexact 
 execution. The latter term I must be understood to use in no irre- 
 verent sense, but to apply it to deviations from the more prominent 
 plan, themselves also being designed by a wise Providence, and 
 for great and salutary purposes. Thus we may conceive the orbit 
 of the earth to be designed as a circle ; this by an excess or defi- 
 ciency in the force of projection, or by the operation of certain 
 disturbing forces, drawn into an ellipse, this again, from agencies 
 not beyond the calculation of the astronomer, turned by the pro- 
 gression of the apsides into a curve of a diflferent form, while 
 every movement is the result of a law of marvellous and beautiful 
 simplicity, calculated by its own nature to correct the very irregu- 
 larities which itself occasions.* So we may consider the earth to 
 be designed spherical ; to be sulFered by its revolution to attain 
 the form of a spheroid, while the surface of this is so varied by 
 
 * A force varying inversely as the square of the distance, as does that of gravity, 
 keeps the revolving body in an orbit forming an ellipse, instead of continually 
 approaching to or receding from the centre in a spiral. This ellipse, however it may 
 be changed by disturbing actions, can never become a curve of the latter description. 
 
 H 
 
50 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 inequalities, that it would be hard to find an actual portion of it 
 coinciding with the surface of the smooth and perfect figure. In 
 saying that art must correct nature, Mr. Garbett has uttered a 
 profound and important truth, and a person of Mr. Ruskin's genius 
 and philosophical turn of mind ought not to have endeavoured to 
 cast ridicule upon it, even if he disliked the terms in which it is 
 expressed. It is because art has so narrow a space for her work, 
 compared with nature, that she must select her typical, rather 
 than her real forms ; and the narrower the space, the more strict 
 and careful must be this selection. If you wish to paint a flower- 
 piece in colours, you will probably find some artificial grouping 
 and arrangement necessary, and some filling up of accidental 
 defects ; if you are restricted to mere light and shade, still 
 more adaptation may be wanted ; if you are confined to simple 
 outline, you must endeavour to. approach still nearer the typical 
 form, that you may fairly express the truth of nature. If you 
 are still further limited by the trammels of decorative design 
 or architectural ornament, you must mark out the typical form 
 with yet more decision, and even exaggerate some of its charac- 
 teristics. By searching for typical forms, and laying them down 
 to the best of our power from nature's own data, instead of con- 
 tenting ourselves with the real forms presented us, we are in fact 
 paying homage to the excellence of her design, which will bear 
 the strongest light that can be thrown upon it by perfect exe- 
 cution. And where will you find such forms? Walk along a 
 laurel hedge, and try how soon you can find a leaf which you 
 would copy accurately if you wished to give the perfection of a 
 laurel leaf. One arrived at its full growth, without a flaw, distor- 
 tion at any point, discoloration, or defect. You will very soon 
 see what is the sort of leaf to look for ; you would learn that in a 
 few minutes if you had never seen a laurel hedge before, but I 
 suspect you will not quite so soon meet with the object of your 
 search. And why ? Because nature is exact in design, inexact in 
 execution. Exact design, though it will allow inexact execution, 
 still does not fear the most exact. But where the design is inexact, 
 then the finer the execution, the more certainly will it betray its 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 51 
 
 defects. In the cases I have taken above, of the adaptation of a 
 cylinder to a square, if it is done by intermediate polygons, then 
 the more accurate the workmanship, the more plainly will be 
 shown the imperfect junction ; on the contrary, if the spherical 
 pendentive be employed, the truer the work, the more perfectly 
 will be developed the excellence of the design. The inexactnesses 
 of nature, if properly considered, are themselves the result of 
 consummate design, and the source of incalculable beauty. Still 
 they must be introduced cautiously in the limited fields of art. 
 Her deviations are of another description. A few strokes of the 
 pencil or chisel must perform the part of the infinitely elaborate 
 workmanship of nature. Accident in one will cause as much 
 deflexion from an original type, as the complex and wonderful 
 machinery of the other. That freedom of hand, without which 
 any imitation of nature would be painfully defective, must in itself 
 involve many incorrectnesses, which are quite compensated by the 
 indication of vigour and originality which accompanies them."^ 
 Inaccurate execution may even serve as a foil to excellent design. 
 But the artist must show his acquaintance with types of perfection, 
 and this he must do, by expressing them simply and broadly ; 
 conscious of the inability of art to convey them, as does nature, 
 by an aggregate of inexact specimens, rather than by a few exact 
 ones. 
 
 I have been betrayed into a pretty long digression by looking 
 into the roof of an Angevine church ; a thing by the way, which I 
 have not yet described to you. It is divided in general into square 
 vaulting compartments. The longitudinal and transverse arches 
 are both pointed, the former being strongly marked by one or 
 more orders on an impost projecting from the wall, for there are no 
 aisles. The diagonal of the vault is ribbed, its arch being pointed, 
 and similar, or nearly so, to the transverse and longitudinal ones, 
 consequently the roof is much domed. The ridge of the vault 
 cannot easily be determined, as to the form of its curve, by the 
 spectator standing below ; but it is probably a combination of 
 
 * See Ruskin, Lamp of Life. 
 
52 
 
 ARCHITECTURAL STUDIES IN ERANCE. 
 
 circular arcs nearly equivalent to the elliptical arc of which we have 
 spoken. Be this as it may, it was considered to be an important 
 line, for it is usually marked by a rib, a feature which hardly ever 
 occurs in the common cross vaults of the period (the end of the 
 twelfth century). It might perhaps be possible to design the masonry 
 of the vault in such a manner as to avoid nearly altogether, the thrust 
 of the arched cellular ridge against the side wall at the crown of 
 the longitudinal arch ; but if not, a buttress might easily be applied 
 in the centre of each bay, between the two windows by which it is 
 frequently occupied. But a certain abutment is obtained, not 
 only by the thickness of the wall, but by the depth of the arch 
 itself, which projects, as we have seen, very boldly from the wall. 
 The main buttresses are necessarily very large and heavy, owing 
 to the width of the arch ; and the imposts stand out well, forming 
 a sort of interior abutment. The orders of the arches are square, 
 with a small roll or torus at the edge. The abacus is also square, 
 and often delicately moulded, and the capitals enriched with 
 foliage and other sculpture. 
 
 The finest example of this style is the cathedral at Angers, 
 which Mr. Inkersley* in his valuable work on French architecture, 
 pronounces the most graceful conception of the twelfth century. 
 The west front presents the usual appearance of a central com.- 
 partment flanked by towers. These, instead of tapering up by 
 means of buttresses, spring vertically. Indeed the upper work 
 appears slightly to overhang the base, an arrangement which, not- 
 withstanding the narrowness of the towers, the lightness, I may 
 almost say flimsiness, of the spires, and the revived Italian cupola 
 placed between them, gives this part of the exterior an imposing 
 effect, which in some degree prepares us for the majestic appear- 
 ance of the interior. This front is adapted to a building undi- 
 vided into aisles, yet of the same breadth, including the towers ; 
 
 * An Inquiry into the Chronological Succession of the Styles of Romanesque and 
 Pointed Architecture in France, with Notices of some of the principal Buildings on 
 which it is founded. By Thomas Inkersley. London, John Murray, Albemarle 
 Street, 1850. 
 
Angers Cathedral. 
 
 Page 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 53 
 
 at least the northern and southern faces of these do not project 
 beyond the line of buttresses of the nave. The front has only 
 one western door, which is much enriched ; the jambs have some 
 large and very finely sculptured figures. According to a plan 
 given by M. de Verneilh, the lower part of the towers seems 
 to be nearly solid. The nave, exclusive of a small part corres- 
 ponding with the towers, has three bays, each forming a square 
 compartment ; the transepts, and compartment of the intersection 
 are also squares, as is one eastward, beyond which is a narrow 
 intermediate compartment, and a semicircular apse, which how- 
 ever becomes polygonal in its upper stage. The nave is, I believe, 
 pretty well known to belong to the latter half of the twelfth cen- 
 tury ; the choir is somewhat later, and the transepts were com- 
 pleted by the middle of the thirteenth century. By the way, the 
 quotation given by Mr. Inkersley is valuable, as it states the 
 painted windows of the nave to be put in before the erection of 
 the vaulting. A building of that width, upwards of fifty feet in 
 the span of the arches, would require some time to allow the work 
 to settle, before a stone vault could have been ventured upon. 
 During this time a temporary roof may have been used, perhaps 
 the great arches themselves may have been built, and the inter- 
 vening spaces ceiled with carpentry, or left open to the outer 
 roof. 
 
 The different parts, while they show the progression of style, 
 still harmonize together perfectly. The windows of the nave are 
 round headed, but tall and graceful in their proportion. Those 
 of the choir and transept are pointed. Foliated circles are also 
 introduced. The fronts of the transept have fine roses, formed by 
 radiating shafts and trefoil arches. 
 
 The lower stage of each bay of the nave (internally) has a fine 
 pointed wall arch, the springs of which are about the height of the 
 base moulding. The arrangement is an uncommon one, and the 
 effect very bold. The prevalence of the massive round engaged 
 shaft in the piers, (a decided Angevine feature) adds much to the 
 grandeur. All the vaults are highly domed, the longitudinal, 
 transverse, and diagonal arches being pointed; the nave is without 
 
54 
 
 ARCHITECTUKAL STUDIES IN FRANCE. 
 
 the ridge rib ; the transepts, choir, and compartment of intersec- 
 tion, have it. On the whole I was more struck with the interior 
 of this cathedral than even Chartres or Amiens. 
 
 To account for the priority of the nave to the choir, we may 
 suppose the former to have been added to the eastern part of an 
 earlier church, as is the case with S. Radegonde of Poitiers, whose 
 nave is a fine specimen of Angevine, with some local modifica- 
 tions. And I think this must be the case with La Trinite at 
 Angers, the nave of which appears of the same date with that of 
 the cathedral. The eastern part of this church consists of a 
 central tower with an apse flanked by smaller ones projecting from 
 very short transepts, which, as at Loches, make scarce any show 
 
 externally. The arches un- 
 der the tower are pointed. 
 
 and the compartment groin- 
 ed with a diagonal rib ; the 
 apse windows are round 
 headed. The tower is square, 
 crowned with an octagon 
 and a stone cupola. The 
 lower stage has each face 
 divided into four by three 
 engaged shafts, there being 
 none at the angles. On each 
 side of the central shaft is a 
 tall round-headed window. 
 The faces of the octagonal 
 part are also shafted, and 
 have round-headed win- 
 dows. If I was not deceived, 
 the architraves seemed to 
 have been touched by cinque 
 cento architects, but I take 
 the whole of the tower, with such exceptions, to belong to the Ro- 
 manesque or transitional period. The nave, of which the outer wall 
 pretty nearly ranges with the front of the transept, has seven bays, 
 
 La Trinite, Angers. 
 
Trinite, Angers. 
 
 Page 54. 
 
AKCHITECTURAL STUDIES IN FRANCE. 
 
 55 
 
 six of which form three square compartments worked with sexpartite 
 vaulting. There are no aisles, but each bay has an apsidal recess 
 under a rich pointed arch. Above this is a string course sup- 
 porting a tall round-headed window with shafts. The western 
 bay has no apse on the north side, and exhibits some earlier Ro- 
 manesque work, as do also some remains connected with the west 
 end of the church. The piers have massive coupled shafts, with 
 enriched capitals and square abacus. Those supporting the large 
 transverse arches which separate the square compartments, rise 
 but little higher than the string course above the apsidal arches ; 
 but the intermediate shafts, which bear the transverse rib of the 
 sexpartite vault, are much higher, ranging nearly with the springs 
 of the window arches of the upper stage. Consequently the vault 
 
 La Trinite, Angers. 
 
 differs from the common sexpartite vault in which the central 
 transverse rib is similar in form to the main transverse arches, 
 separating the compartments, with its spring at the same height, 
 so as to leave each of the two longitudinal arches symmetrical, 
 which they are not in the example before us. The vault is 
 highly domed, and has longitudinal ridge ribs, so that eight 
 diverge from the apex, namely the two ridge ribs, the two trans- 
 verse ribs dividing the cells of the sexpartite vault, and in fact, 
 forming a transverse arch, and the four diagonal ribs marking the 
 
 The above Plan is taken from " Angers Pittoresque." 
 
56 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 external edge of the same cells, and forming diagonal arches over 
 the square compartment. A curious peculiarity has been noticed 
 in this church ; viz. the increase in height of the nave to the 
 westward. Some of the capitals and bands of foliage are of ex- 
 quisite workmanship ; both in its detail and general composition, 
 this church will be found to be one of the most curious and inter- 
 esting in France. 
 
 Parts of St. Martin's, in Angers, are said to be of very high 
 antiquity, but these are rapidly disappearing, and any features of 
 architectural interest that it retains, are probably due to the ele- 
 
 St. Martin, Angers. 
 
 venth and twelfth centuries. The central tower compartment is 
 covered by a dome, whose pendentives rest on small engaged 
 shafts, themselves supported on the abacus of massive cylindrical 
 
Page 57. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 57 
 
 ones, engaged in re-entering angles, and bearing blank round 
 arches of one square order. The original arches of the intersection 
 are plain and round-headed, of one square order, on a square 
 impost with a string by way of capital. The crowns of these are 
 not so high as the capitals of the engaged columns above men- 
 tioned, and the interval between the two arches is occupied by 
 a couplet of round-headed windows. A few pier arches on each 
 side of the nave still remain, perfectly plain and round-headed. 
 The choir, which is rather large, is of late twelfth or early 
 thirteenth century work, and ends in a polygonal apse. It is 
 somewhat difficult to find this church, and impossible to obtain 
 a satisfactory outside view. 
 
 The ruined church of Done, near Saumur, if carefully studied, 
 will give the idea of an Angevine church in perfection. It may 
 be considered as divided in its length into five square equal bays and 
 an apse, the separations between the bays being boldly marked by 
 massive engaged piers and transverse arches. Three of these 
 bays constitute the nave ; the fourth belongs to the central tower ; 
 the fifth, with the apse, forms the choir. The transepts are also 
 square, each consisting of one bay, similar to the others, and have 
 eastern chapels not terminating with apses. There are no aisles. 
 Consequently the plan is very like that of Angers Cathedral ; but 
 instead of western towers it has a fine central one. The composition 
 of the engaged piers of the 
 
 nave is remarkably fine — a 
 large cylindrical shaft sunk 
 to half its diameter in the 
 flat face of the pier, a small 
 shaft in a re-entering angle 
 on each side, and a large PlanofPieh. 
 one in the ansrle between 
 
 the rectangular side of the pier and the wall. Externally, are 
 buttresses corresponding with the piers. Each bay has a single tall 
 window of one light ; these, with the bearing arches throughout, 
 and the tower windows, are pointed. The apse windows are 
 round-headed. Such of the vaults as remain are exceedingly do- 
 
 1 
 
58 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 mical. I sincerely trust the French antiquaries may give their 
 attention to the preserving of this monument, which otherwise 
 cannot last much longer. A reproduction of the building, in any 
 spot where a large church is wanted, and where sufficient funds 
 could be obtained for the purpose, would not be difficult, and 
 the undertaking would be a judicious one, in a utilitarian as 
 well as artistic point of view. 
 
 S. Pierre, Saumur, has a fine Angevine nave without aisles, 
 considerably wider than the choir and compartment of the central 
 tower. The nave is vaulted in three square compartments, the 
 arches being pointed, and the crown of the vault, as in the other 
 cases, much raised. The ridges, as well as the diagonal edges, are 
 marked by ribs. The windows are round-headed, one to each 
 compartment, between two blank arches on shafts. 
 
 The eastern part of Fontevrault, though it exhibits slightly 
 pointed arches, has a pure Romanesque rather than a transitional 
 character. The choir is apsidal, with an aisle of the same shape, 
 and radiating semicircular chapels, which also occur eastward of 
 the transepts. Internally, the apse is separated from its aisle by 
 eleven very slightly pointed arches of one plain square order, 
 resting on tall columns with the square abacus. Above is a small 
 triforium arcade of round arches on shafts, and the clerestory has 
 round-headed windows. The roofs are in general cylindrical ; 
 that of the apse semidomical ; that under the tower is nearly the 
 Bernay vault. 
 
 As I could not obtain admission into the nave, which is turned 
 into apartments for prisoners, I speculated on its arrangement 
 from external appearances. As the buttresses are alternately 
 large and small, the spaces between the former corresponding to 
 square vaulting compartments, I supposed it might be Angevine, 
 either with the domical cross vault, or the sexpartite vault of La 
 Trinite. But M. de Verneilh notices it as one of the domical 
 churches of which the cathedral at Angouleme is a type, and 
 which sprang from S. Front, Perigueux. The nave has been 
 roofed with four domes, now destroyed, but the pendentives re- 
 main, with sufficient indications to assure us without doubt of 
 
Page 58. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 59 
 
 the original construction. The example is therefore valuable, 
 occurring as it does among the Angevine churches, since it 
 suggests a sort of connection between the two styles, which, how- 
 ever striking the difference may appear at the first aspect, still 
 are not without considerable analogy. The restrictions attendant 
 on a visit to the monastic buildings belonging to this church, 
 together with my wish to obtain a satisfactory view of the exterior, 
 prevented me from noticing several objects of interest, especially 
 a very curious circular structure of the twelfth century. 
 
 S. Radegonde, Poitiers, has an Angevine nave inserted between 
 a Romanesque chancel and Romanesque western tower. It has 
 no aisles ; the vaulting compartments are rather oblong trans- 
 versely, but very domical, and I think furnished with ridge ribs. 
 We have observed that the oblong compartment admits of just as 
 much accuracy as the square, on the Angevine system, and this 
 whether the arches are pointed or semicircular. 
 
 The cathedral of Poitiers furnishes an instance of this. It has 
 aisles, and the span of the pier arches is less than that of the 
 nave, consequently the compartments are oblong. They are 
 much domed, and have the ridge ribs. There is no triforium nor 
 clerestory range of windows, but the cellular arch of each vaulting 
 compartment is considerably above the pier arch, leaving an inter- 
 mediate space. The range of windows along the walls of the aisle 
 is extremely elegant ; they mostly stand in couplets, except where 
 they are replaced by a large tracery window ; those near the east 
 are round-headed; those to the westward are pointed. They rest on 
 a string, below which is a blank arcade. The aisle compartments, 
 if I remember, are oblong longitudinally. They are contained 
 under the same external roof with the nave ; it is consequently 
 of great expanse, but not very high-pitched. The transept is 
 without aisles, and about the width of one of the pier arches, 
 consequently, though its walls are of the same height as the 
 external walls of the nave and choir, that is, of their aisles, the 
 ridge of its roof falls below that of the principal roof, which is 
 therefore unbroken from one end to the other. This arrangement 
 gives the building, at a distance, the character of an enormous 
 
60 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 hall, rather than a cathedral, especially as the western towers have 
 not sufficient height to add much variety to the outline ; yet its 
 very simplicity produces much grandeur, which is increased by 
 the flat elevation of the east end, the apses being worked internally 
 in the thickness of the wall. This church, and S. Eadegonde, as 
 well as the cathedral of Angers, contain some very rich specimens 
 of painted glass, the latter, as we have seen, as early as the twelfth 
 century. 
 
 If I notice any other churches in Poitiers, it will be in a subse- 
 quent chapter, for the two I have mentioned are the only ones 
 that I know of in any way referable to the style now under our 
 consideration. But I was much hurried in my visits to that most 
 interesting city, a careful examination of which would occupy 
 a greater number of weeks than I was able to bestow upon it 
 of hours. 
 
 The Angevine style, I must repeat, is one we should study well, 
 if we wish to revive mediaeval church architecture. It will perhaps 
 be found to admit more originality, and to be less exposed to 
 conventionalism, than any other. Its vaulting system is admirable 
 in effect ; it admits, as we have seen, of the most accurate mathe- 
 matical design ; and, if Mr. Garbett's remarks upon the ventilation 
 of ceilings be of the value they appear to be, it must be the best in 
 a sanitary point of view.* For an aperture at the crown of each 
 compartment is always practicable ; in fact, I should suppose one 
 is usually found there, as in the bosses of our own cathedrals, 
 through which I have often dropped a plumb-line to obtain the 
 height. The thrust of the sloping ridge on the wall of each com- 
 partment, would in most cases be met by the solidity necessary to 
 produce depth and effect, or by a buttress and pinnacle. We have 
 seen that the vault may be truly constructed, whether its compart- 
 ment be square or oblong, whether its arches be pointed or 
 semicircular, and that the only curve differing from the simple 
 circular arc is the elliptical curve of the ridge, which it is difficult 
 
 * Student's Guide to the Practice of Designing, Measuring, and Valuing Artificers' 
 Works, &c. London, John Weale, 59, High Holborn, 1852-1833. Seethe Chapter 
 on Plasterer's Work, p. 194. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 61 
 
 to detect from below, and for which a more easily drawn line 
 might be substituted. In flat ridged vaulting, where the ridges 
 are of the same height, the adoption of the pointed arch does not 
 solve one single problem, or meet one single difficulty in compo- 
 sition. It may be useful as a stronger form, in construction, but 
 in arrangement it does no more than slightly conceal or soften 
 down errors or harshnesses. It will not enable us to avoid the 
 introduction of the elliptical form in the narrower arches of oblong 
 compartments, supposing the wider ones to contain circular forms, 
 though it may help us to disguise the change of curve ; nor, if the 
 elliptical form be neglected, will it remove, though it may soften, 
 irregularities in the surfaces. It does, in fact, no more than 
 introduce a system of approximation, instead of accuracy in design. 
 But the pointed arch enabled a vaulting cell to be carried into 
 the main vault at a point lower than its ridge, and this simple 
 change, I think, may be shown to be the germ from which arose 
 ultimately that beautiful and refined system, the fan-roof. But I 
 will not now pursue the subject further. 
 
 It was not till I had written the above chapter that I had an opportunity of becoming 
 acquainted with Professor WiUis's able and instructive paper on the Construction of 
 Mediaeval Vaults. Though I have looked at the subject from a different point of 
 view, I trust I shall not be found to have advanced any thing, either in my present 
 remarks, or in former observations on vaulting, really at variance with the principles 
 which he has laid down, and confirmed by striking examples. He clearly condemns 
 the practice of constructing a Gothic cross vaulting on the supposition that it is formed 
 by the crossing at right angles of two uniform vaults, I mean vaults of which every 
 transverse arch, taken from its vertical spring, shall be the same. This construction 
 would evidently involve some arches, whether longitudinal, diagonal, or transverse, of 
 a variable curvature, as the ellipse. 
 
 " A common method of executing the modern Gothic vaulting, especially in plaster, 
 has been to obtain the form of the diagonal ribs, by ordinates from those of the 
 transverse ribs. The effect of this is, that the entire vault appears as if it consisted 
 of two waggon vaults, crossing at right angles ; and a horizontal rod moved from top 
 to bottom, in contact with any pair of opposite arches, will touch the diagonal ribs and 
 the connecting vaulting surfaces throughout. This is the genuine principle of the 
 Roman and Italian groined vault, but is altogether foreign to the principles of Gothic 
 architecture, in which every rib should spring as a separate and independent arch, and 
 in which the elliptic curves produced by this method are totally at variance with the 
 characteristic forms of the style." — Professor WilHs on the Construction of the Vaults 
 of the Middle Ages, p. 21. 
 
62 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 I have considered this regular vault to be the first contemplated by the Gothic 
 architect, who, however, lost no time in making such deviations as he judged expedient, 
 whether on the score of convenience or beauty. Thus, although the vaulting surface 
 itself be really the matter of first consideration, the form of the edges or ribs being 
 determined by that, he soon found it less troublesome to determine first the form of 
 the ribs, and dispose of the surfaces according to them. It was easier to make the ribs 
 circular than elliptical ; and here the use of the pointed arch would serve his purpose ; 
 for, take even a square vaulting compartment, if the ridges of the vault must be in the 
 same horizontal plane, and the diagonal rib, springing vertically, form a circular arc 
 not larger than a quadrant, from spring to apex, then the longitudinal and transverse 
 arches must be pointed, as at Sinzig, Worms, and in many German churches. A 
 segmental arch is sometimes used as a diagonal rib, but I cannot bring myself to 
 admire it. When the transverse vault is narrower than the longitudinal one, the 
 stilted arch (see Willis) is often substituted for the elliptic, or acutely pointed arch. 
 This occasions that receding of the vault which he notices, forming a kind of nook 
 above the impost ; and I quite agree with him, that it gives amazing force and character 
 to a Gothic vault. These deviations from a primary type are, in Gothic art, as in 
 Nature, the product of deep thought and careful design, and, in both cases, the source 
 of much that is grand and striking. I conceive the observance, or recognition of strict 
 mathematical rules, to be productive of refinement, grace, and elegance, while the 
 deviations from them give force, character, and picturesqueness. 
 
 A'adlting, from S. Manvieux, Norm ANT) v. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 63 
 
 CHAPTEE VI. 
 
 We will begin again, as in the last chapter, with the Bernay vault 
 which we took for the basis of the Angevine system of vaulting, 
 but treat it in a different manner ; and I beg to assure the reader 
 that I do not intend again to trouble him with any geometrical 
 problems. 
 
 The quarter of the Bernay roof is 
 represented in the diagram by the area 
 B L K I ; the projection of a curved sur- 
 face, being part of a hemispherical 
 dome of which the radius is a k. Sup- 
 posing the whole hemispherical dome 
 to stand perfectly well, if its base rests 
 on the ground or the top of a cylin- 
 drical tower or drum, then the portion 
 of the dome forming the Bernay roof 
 will also be firm, provided the vertical 
 support given by the arches of which 
 
64 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 parts are represented by the lines b i 
 and B L, and also a lateral abutment 
 in a direction perpendicular to the 
 plane of the respective arches, viz. in 
 the several directions of l o, and that 
 of a line vertical to the plane of the 
 paper, are sufficient to compensate for 
 the segments cut away by the several 
 arch-planes. But a dome need not, 
 like an arch, be complete at the top ; 
 the upper part may be cut off, leaving 
 an open, horizontal circle ; and on 
 the circumference of this circle, as a 
 base, may be erected another dome. This may render necessary 
 some alterations in the lower part of the constructions, but a firm 
 well-balanced structure may be designed without any difficulty ; 
 and by taking away the top of the Bernay dome above the arches, 
 that is the surface of which l k i represents a quarter, and substi- 
 tuting in its place the hemispherical dome, of which l n i is the 
 projection, we come to the dome on Byzantine pendentives, in- 
 vented and used in eastern churches at a very early period of the 
 christian era. 
 
 Now though it is possible that the change of the low fiat domical 
 roof L K for the higher and more vertically pressing dome l n, 
 may so far vary the conditions of equilibrium as to admit of a less 
 lateral abutment at l and i, and the edges of the pendentives, and 
 to require a greater vertical strength in the arches and their piers ; 
 still it does not altogether dispense with lateral abutment. And 
 this it is easily seen will be satisfactorily provided by a cylindrical 
 vault running horizontally from each of the arches in a direction 
 perpendicular to their planes. This vault must be long enough 
 to supply the necessary abutment, unless it be superseded by 
 something else ; but it need not be limited there, as it has the 
 nature of an immoveable obstacle, preventing the dome from being 
 forced outwards, while it exercises no tendency to force it inwards. 
 We have therefore, if we take a ground plan, the square compart- 
 
ARCHITECTUEAL STUDIES IN FRANCE. 
 
 65 
 
 ment covered by the dome a, and its pendentives d, d, d, d, which 
 we will call the dome compartment ; and four adjacent compart- 
 ments, covered by cylindrical roofs corresponding with the arches 
 under the dome ; these we will call the arch compartments. But 
 the arches themselves supporting the dome must have their piers 
 and abutments ; which might be given by a wall in the direction 
 of each arch, which wall acts also as the impost of the cylindrical 
 vault. 
 
 The vault itself, however, may require some further abutment ; 
 either at each end, or at intervals during its length, or through- 
 out. Now if the square be completed at each angle of the figure, 
 of which two sides are necessary, the other two sides, combined 
 with the vaulting of the area enclosed, 
 will form a very good abutment to the c 
 cylindrical vault; we may also find 
 it convenient to gather the walls of 
 these four compartments into strong 
 piers at the angles (see Ruskin), the 
 openings being arched over ; and thus 
 we arrive at a well constructed build- 
 ing in the form of a square, consist- 
 ing of one dome compartment, four 
 arch compartments, and four others ; which, including the space 
 occupied by the masonry, as well as that within it, I will call pier 
 compartments. Any walls employed for filling up the ends of the 
 building may be treated as mere screens, in no way contributing 
 to the support of the fabric. 
 
 The relative size of the compartments surrounding that of the 
 dome, is of course arbitrary, so long as they ensure sufficient 
 abutment. But a certain limit will be seen to be expedient, since, 
 if the square pier compartment be made too large, either its own 
 roof will require further abutment, or an unnecessary mass of 
 material will be used in the solid parts of its structure. If there- 
 fore the pier compartments exhibit domes, externally or internally, 
 they should at least be considerably smaller than the central one. 
 
 Now this, I believe, with the addition of apse and vestibule, is a 
 
 K 
 
66 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 usual plan of ancient eastern churches, and it has been adopted 
 since the revival of the classical style. The beautiful church of 
 the Carignana in Genoa is an example. The plan we have given 
 evidently admits of indefinite multiplication, so as to cover any 
 area we may wish. A dome compartment may be added to any 
 or all of the arch compartments in the plan given, and the neces- 
 sary arch and pier compartments added to these. The architect 
 of S. Front, Perigueux, has done this ; not for the first time, for 
 M. de Verneilh shows that he has copied the nearly contempora- 
 neous church of S. Mark, of Venice ; and, by a passage quoted 
 from Procopius, proves that a church of much the same plan was 
 built about the time of S. Sophia at Constantinople. Still, I 
 suppose no building now in existence shows the principle of com- 
 position and construction so plainly and undisguisedly as this 
 cathedral of Perigueux ; a building which gives the idea of being 
 thrown off as a rough and grand sketch, the filling of it up being 
 left as a problem to future architects. I will take the liberty of 
 copying M. de Verneilh's plan, at least as much of it as I require 
 for my purpose ; and I shall also draw from his most able, accurate, 
 and interesting work " L'Architecture Byzantine en France," to 
 which I have already referred, such information respecting this 
 fabric and others derived from it, and the histories and dates con- 
 nected with them, as I may find necessary. 
 
 The cathedral, composed as you see of exactly the elements I 
 have mentioned, namely, dome compartment, arch compartment, 
 and pier compartment, and exactly in the positions which have 
 been assigned to them, forms a Greek cross, having at its west 
 end a vestibule, the remains of an older church, above which is a 
 fine Romanesque tower ; and at the east end an apsidal chapel of 
 the fourteenth or fifteenth century, which replaces the smaller 
 semicircular one. The transepts had also eastern apsides, of 
 which that on the south remains. There are five equal dome 
 compartments, viz. one over the nave, one over the choir, one 
 over each transept, and one over the intersection. All of these 
 are surrounded by their respective arch and pier compartments, 
 those round the centre doing double duty. Indeed these com- 
 
Plan or S. Front, Perigueux. 
 
 Page 66. 
 
AKCHITECTURAL STUDIES IN FRANCE. 67 
 
 partments might have been reduced in size, and simple piers used, 
 since one dome compartment would act as abutment to its neigh- 
 bour. This was done in later edifices, where the domes, covered 
 from the first by a long roof, did not require to stand so far apart. 
 
 Outside View of S. Front, Perigueux. 
 
 The church does not at present exhibit the external outline in- 
 tended by the architect. There appear to be quite sufficient data 
 for ascertaining what this was, and for effecting a very accurate 
 restoration. M. de Verneilh's frontispiece gives probably the 
 
68 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 least conjectural restoration that has ever yet been proposed for 
 any building. Most of what he represents actually exists, and 
 where renewal would be necessary, the building seems rather to 
 give definitely, than to suggest vaguely, the forms required. Ac- 
 cording to his engraving, the arch compartments were covered 
 with low pitched stone roofs, forming gables, of which there would 
 be twelve in all, unless one was absorbed by the western tower. 
 The domes stand out boldly, and are divided by a cornice into 
 drum and dome, the former battering considerably. This may 
 
 be seen by going up into 
 the roof. The pier com- 
 partments, except the four 
 central ones, were finish- 
 ed with four-sided spires 
 or pyramids, not very 
 acute ; these have dis- 
 appeared in consequence 
 of the several fronts be- 
 ing raised, at an early 
 period, to a horizontal 
 line above the gables. But 
 a small portion of one is 
 said to be discernible, from which the form may be determined. 
 The gables are still strongly marked out by their bracketed 
 angular cornices. The domes are now completely hidden, in the 
 external view, by a structure of stonework, forming a sort of cle- 
 restory, and carrying a roof ; this is quite modern. 
 
 It is an extraordinary fact, in the history of the pointed arch, 
 that it should have made its earliest appearance in places where 
 the Gothic system was the latest developed ; and under conditions, 
 which, mathematically speaking, the most strictly required a 
 round arch. Now in going with me through the preceding re- 
 marks, if you have been disposed to take one thing for granted 
 more than another, it is probably, that the arches of the dome 
 compartment are semi-circular, and the corresponding roofs semi- 
 cylindrical. Yet in nearly all the examples that I have seen. 
 
ARCHITECTURAL STUDIES IN FRANCE. G9 
 
 where the roofing is composed of a series of domes, and in 
 all French examples that I know from engravings, the arch is 
 pointed; in this cathedral of S. Front, where it is supposed to 
 appear for the first time in France, very obtusely ; in others, 
 
 S. Front, Pekigueux. 
 
 very decidedly. How it comes here, for it does not make its 
 appearance in S. Mark's of Venice, is the question. Perhaps 
 in the first instance it may have been accidental, as requiring 
 less accurate construction ; perhaps it was considered a stronger 
 
70 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 form for the support of so much superincumbent weight ; per- 
 haps a reason may be assigned which has occurred to French 
 antiquaries, though I was not aware of this when I threw out the 
 suggestion as one which had struck myself ; viz. that a pointed 
 barrel roof was used as being more easily fitted with an external 
 stone roof, which necessarily required an angular ridge. Un- 
 questionably the pointed barrel roof occurs in buildings of the 
 eleventh, perhaps- the tenth century, in the south of France. Now 
 we have seen that the arch compartments of Perigueux are real 
 roofs, and not simple bearing arches, and have been covered with 
 an external pitched roof of stone, corresponding with gables which 
 still appear in all the fronts and sides ; and the adaptation of the 
 internal to the external roof would naturally suggest a pointed 
 arch. It is true, the arch is still more decidedly pointed in later 
 examples, in which it does not become a real roof ; but the form 
 having been introduced, and not proving unsightly, motives of 
 convenience, and the very novelty of the thing, may have ren- 
 dered its adoption permanent. A small quantity of upright wall, 
 between the line of the arch and that of the pendentive, will 
 prevent the necessity of disturbing the correct circular line formed 
 by the latter ; supposing it of consequence to preserve it strictly, 
 which I presume was not always done. The pier compartments? 
 plain as they are, and devoid of all ornament except a simple 
 string at the top by way of cornice or capital, still give the idea 
 of complete architectural compositions in themselves, rather than 
 masses designed for the support of a superstructure. This would 
 be a fault, if the pier compartment and superstructure were made 
 to have the same prominence. But they are not so in this instance, 
 though the fact is owing rather to accident than original design. 
 The springs of the arches stand back from the faces of the pier, 
 instead of being even with it ; or, as is frequently the case, pro- 
 jecting slightly forward. The cause of this unusual arrangement 
 M. de Verneilh has, with singular acuteness and sagacity, ascer- 
 tained and proved. The entire surfaces of the pier compartments 
 have been coated with a layer of smooth masonry, to counteract a 
 sinking of the upper structure, which began to show itself before 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 71 
 
 its completion. The layers of stone, it may be observed, are 
 alternately wide and narrow. It is supposed that the wider course 
 consists of slabs of stone placed on their narrowest sides or edges, 
 touching the face of the old masonry ; while the narrow courses 
 are stones of the same shape laid flat, and consequently pene- 
 trating into the heart of the pier, in order to ensure the firm 
 connection of the whole. These stones therefore, however dif- 
 ferent in external effect, are really analogous to our Saxon long 
 and short work ; to which they are compared by the writer to 
 whom I refer. I am not aware how much ornament has been 
 concealed or destroyed by this operation, which sufficiently ac- 
 counts for the heaviness of the pier-compartments, and the nar- 
 rowness of their arches, as also the flatness of their vault or roof. 
 It may not in this respect be considered an improvement ; still it 
 has developed a correct artistic principle, as I have said, by 
 throwing back the roof, and disconnecting it in a manner from a 
 structure of sufficient importance to assume a certain indepen- 
 dence of its own. We shall have to recur to this in a future part 
 of the chapter. 
 
 We have seen that Loches, although sound in its construction, 
 picturesque in its outline, and by no means deficient in grandeur 
 of effect, yet stands alone in its peculiarities. And I have at- 
 tempted to explain this on the score of a certain artistic incon- 
 gruity. Now Perigueux, the structure of which must be more 
 difficult and complicated, presents no such incongruity ; and ac- 
 cordingly we find it is the type of many other buildings. Kot 
 indeed on the same ground plan, and all differing in some impor- 
 tant particulars, but still evidently derived from the same source. 
 M. de Verneilh, with the assistance of his fellow labourer in the 
 same field, M. I'abbe Michon, the author of a work to which I 
 have already alluded, " Statistique Monumentale de la Charente," 
 is enabled to state that to his knowledge there exist at least forty 
 churches in the province of Aquitaine which are roofed with a 
 series of domes. Most of these lie in groups round Perigueux and 
 Angouleme, but some at a considerable distance from either. One 
 of the earliest is the church of S. Etienne de la cite at Perigueux, 
 
72 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 formerly the cathedral, though always a smaller church than its 
 neighbour. Only half the present church, however, belongs to 
 the original structure, which is supposed to have been commenced 
 soon after, if not actually before, the completion of S. Front. As 
 it now stands, the church consists of two square compartments, of 
 which the western one, part of the original nave, is the oldest, 
 and differs in few particulars from any of those of S. Front. The 
 most important point of difference is, that the pier compartment is 
 here compressed into a solid pier, and consequently the arch 
 compartment is limited in its breadth. In this instance the pier 
 is very massive, and the arch broad ; it consists of a single square 
 order, and is pointed. The cupola forms an external feature, 
 as at S. Front before the erection of the modern roof. It has 
 a conical tiled roof, and is crowned with an elegant lantern, 
 somewhat similar to that on the western tower of S. Front, formed 
 of a circle of shafts very close to each other, supporting a small 
 cupola, without the intervention of any arches. Westward of this 
 compartment are the remains of another, showing the pendentives 
 of its dome. The eastern square of the church is a careful resto- 
 ration (after the ravages caused by the religious wars of the six- 
 teenth century, when this building and many others were ruined 
 
 by the Protestants,) of 
 an extremely elegant 
 structure of the twelfth 
 century, on nearly the 
 same model with the 
 other, but higher, and 
 much enriched. The 
 imposts of the arches 
 are loftier, and fur- 
 nished with shafts ; the 
 arches themselves are 
 mere pointed ; though 
 La Cite Perigdeux. the picrs are solid, the 
 
 arch compartment is 
 
 recognized, forming a narrow barrel vault between the orders of 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 73 
 
 the arch and the external wall. The windows, a triplet of round- 
 headed arches on the three disengaged faces of the square, are 
 shafted, and have ornamented mouldings ; the lower stage forms a 
 blank arcade. The cupola is of the same size with the other, but 
 being placed at a greater height, gives variety of outline to the 
 exterior, and a very impressive effect to the interior of the choir. 
 The lantern at the top is similar to the other. The western piers 
 are carried up and form stair-turrets, adding much to the general 
 outline ; which from the want of gables might have been heavy. I 
 have sketched the base of one of the shafts in the choir, but cannot 
 tell whether it belongs to the origi- 
 nal work of the twelfth century, or 
 to the Restoration. I am inclined 
 to believe, the former ; but if to 
 the latter, it speaks for the excel- 
 lence of the work. It struck me 
 as being remarkably pure and clas- 
 sical ; which will be found to be 
 the character of the late Roman- 
 esque in the district we are con- 
 sidering. 
 
 I could only, for want of time, 
 obtain a very hasty glance at the 
 interior of Cahors cathedral, but 
 it was sufficient to impress me with 
 the grandeur of its composition. 
 The piers with their arches are square and plain, and the cupola 
 of great span. Two of these cover the nave, and though they do 
 not form a decided feature externally, are still discernible. There 
 is no central tower, and the west front has, if I remember, rather 
 a heterogeneous appearance. 
 
 The church of S. Emilion, from the want of towers, is externally 
 rather heavy. The nave has two domes, which only appear inter- 
 nally. The structure is lighter than in the former examples, the 
 arch compartment being reduced to a mere arch. 
 
 The cathedral of Angouleme, before the destruction of its 
 
 L 
 
 La Cite, Perigueux. 
 
74 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 southern tower in the sixteenth century, must have been a magni- 
 ficent building, and worthy of the fine position it occupies. Its 
 massive nave, the lofty towers rising from its transepts, and the 
 lower octagon of the intersection standing between them, would 
 form a group rarely equalled even in that age of fine architectural 
 combinations. The north tower and the octagon still remain. 
 The former is not equal in its area to the whole of the transept 
 from which it springs, but falls within it, in fact corresponds to 
 the central cupola of the structure we have given as the basis of 
 this system of composition. The transept consequently has its 
 pier compartments (compressed as before into solid piers), its arch 
 compartments, and its central dome, which, though spherical, has 
 pendentives rather of a Eomanesque than Byzantine character. 
 
 Externally the tower over this dome 
 presents a fine specimen of Koman- 
 esque enrichment. It has several 
 stages of shafted arcades variously 
 arranged, resting on strings, the 
 arches being round, and many of 
 them pierced for windows. The 
 proportion of the tower is lofty ra- 
 ther than massive, as is the case 
 with many belonging to the same 
 period. The central octagon is very 
 plain ; the diagonal sides are nar- 
 rower than the cardinal ones. The 
 windows in the former, throwing 
 light into the dome, are original; 
 those in the latter were pierced at 
 the time of the restoration of the 
 church after its injuries by the Pro- 
 testants. The west front, of which 
 the elevation is pretty nearly a square, is filled with beautiful 
 sculpture of the twelfth century, in round-headed arches; the 
 whole of very delicate execution ; indeed the exterior altogether 
 shows the great refinement of the style of that period in the 
 
 Angouleme. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 75 
 
 south of France. This is especially remarkable in such of the 
 windows, both in the sides of the nave and fronts of the transept, 
 as remain unmutilated. The nave has three compartments, each 
 roofed with a hemispherical dome on Byzantine pendentives. Of 
 these the western one is evidently earlier than the rest, and than 
 the front which precedes it. It exhibits the plain solid pier, and 
 the wide archivolt of a single square order. The other two are 
 later, and separated by an arch with a narrower archivolt, of two 
 orders. The piers project more into the nave than their corre- 
 sponding buttresses do externally. The compartment of the inter- 
 section has a dome, the plan of which is a square with the angles 
 boldly rounded off ; it is lighted as we have seen by windows in 
 
 Interior of Dome, Angoui.eme. 
 
 the drum or octagon supporting it. The choir has a semicircular 
 apse, its vaulting being cylindrical and semi-domical. Much late 
 work is added in the way of chapels. The north transept has a 
 short cylindrical vault between the intersection and the area of the 
 tower, which, as I have said, is domed, as was the corresponding 
 
76 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 area on the south side, which has also the intermediate compart- 
 ment. The bearino^ arches of the nave and intersection are 
 pointed. 
 
 Gerard de Blaye, who became bishop of Angouleme in 1101, 
 and held the see thirty-five years, built this cathedral ; according 
 to records, a prime lapide." This must apply to a construc- 
 tion, which, as was usual, did not involve the demolition of the 
 whole of the preceding structure, and forms no argument against 
 the earlier date of the western bay ; of which its manner of con- 
 nection with the others, to go no further, seems sufficient proof. A 
 full account of the cathedral is given both by M. de Verneilh, and 
 M. Michon. In the work of the latter, (Statistique Monumentale 
 de la Charente) is a good engraving of the extremely interesting 
 west front. 
 
 Fleac, near Angouleme, is an example of this system of con- 
 struction as applied to a small and very simple village church. 
 The nave is roofed by two domes on Byzantine pendentives, sepa- 
 rated by a round arch, with not a very wide archivolt. The 
 longitudinal arches in the wall are also round, but those under the 
 tower are pointed. The tower, which stands between the nave 
 and chancel, has its compartment roofed with what we have called 
 the Bernay vault, that is, dome and pendentive belonging to one 
 sphere. The chancel is apsidal, with semi-domical vault. The 
 tower is square, and there are no transepts. The church is plea- 
 santly situated on the bank of the Charente, about four miles from 
 Angouleme. 
 
 Roulet, near Angouleme, on the Bordeaux road, is somewhat 
 richer in its ornament, and probably later. It has a nave, central 
 tower, and chancel longer than mere apse. There are no well- 
 developed transepts. The nave has three domes, the circle sepa- 
 rating them from the pendentives having a sort of tooth ornament. 
 The piers present a cluster of five engaged shafts and columns, the 
 longitudinal arches against the wall, as well as the transverse arches, 
 having two square orders ; both are pointed. The tower com- 
 partment, which is older, has an octagonal cupola on Romanesque 
 squinches. The tower is square, with fine Romanesque belfry 
 
Page 76. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 77 
 
 windows, (two double ones in each face) and a conical spire with 
 the fir-apple ornament. The parapet, if original, has a very 
 peculiar style of ornament, namely, a sort of bead or torus, set 
 vertically. This church is characterized by M. Michon as one of 
 the most elegant of those with the series of domes. 
 
 The two works I have mentioned will enable the reader to 
 study these domical churches both in an antiquarian and archi- 
 tectural point of view, and will be most valuable to him if he 
 wishes to make his observations on the spot. They will assist him 
 in forming a most interesting series, according to date and pro- 
 gression of style, a series extending over a period of little less 
 than two centuries ; for the part of S. Front we have been consi- 
 dering, is supposed with good reason to have been commenced by 
 Bishop Froterius or Frotaire, about 990, and finished by Geraud 
 de Gourdon in 1047 ; other buildings begun later may have been 
 finished earlier. The nave of Roulet, which we last noticed, must, 
 from the style of its decoration, belong to the very end of the 
 twelfth century, if not to the following. 
 
 I shall not enter into any question as to the propriety of calling 
 these buildings Byzantine. The hemispherical dome, with its 
 peculiar pendentive, is no doubt derived from the east ; but the 
 churches w^hich present this feature do not generally dififer in 
 other respects from the Romanesque structures of the country. 
 The enrichments are of exactly the same character ; and it may be 
 remarked that the cupola of the intersection is very frequently 
 octagonal, and precisely the same which is found through the 
 whole of the south of France, whatever be the roof to the body of 
 the church. S. Front itself is the only specimen that, as it were, 
 stands aloof from the rest. 
 
 Before we quit the subject of domical churches, we should turn 
 our attention to a very remarkable building, the cathedral of Puy 
 in Velay. From the precipitous nature of the ground, the eleva- 
 tion of the west front does not correspond with the transverse 
 section of the church itself. The floor of the nave is about the 
 level of the string above the great western door, which opens into 
 a kind of crypt, from which we reach the church by a staircase. 
 
78 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 Now as this front is not immoderately lofty in its proportion, the 
 interior itself must be somewhat deficient in height. The archi- 
 tect had, to the eye and imagination, to correct this deficiency. 
 He did this by throwing across, at each division of the bays, an 
 arch supporting a wall pierced with arched openings ; thus giving 
 to each bay a rectangular lantern covered with a domical roof. 
 It is clear that this arrangement, by which almost the whole of 
 the real roof is concealed from the spectator, whatever be his 
 point of view, adds an amazing imaginary height ; and the light 
 through the clerestory gives to each compartment the effect of an 
 open central tower. 
 
 We have seen what the derivatives of S. Front actually are. 
 Is it idle to speculate what they might have been ; what perhaps 
 they still may be ? On entering the church we are first struck 
 with the pier compartments. Very simple and plain they are in 
 their appearance, but to an architect, I should think, very sug- 
 gestive. Might they not, instead of being compressed into mere 
 pilasters, be developed into grand masses, afibrding a marvellous 
 breadth of effect, and capable of receiving the most varied enrich- 
 ment ? Our own architect. Sir Christopher Wren, shows that they 
 may. He may not have known S. Front, but he must have 
 known its model, S. Mark, when he conceived a design, which, 
 had it been carried out, would have given his cathedral the 
 noblest interior in the world. In the model of his intended work, 
 fortunately on a sufficient scale to allow us to appreciate the 
 beauties he aimed at, we see the series of domes with their pen 
 dentives, the arch compartment expanded into a wide cylindrical 
 roof, and the pier compartment of sufficient width to exhibit a fine 
 arched recess between columns crowned with a rich entablature. 
 The recurrence of these masses, few in number, but at wide inter- 
 vals, and their reappearance at a distance beyond the immense 
 area of the central dome, gives a perspective of length not sur- 
 passed by the finest Gothic buildings. We cannot help feeling 
 and being sure of this, while we stand within that wooden model, 
 less than twenty feet in length. But the lighting of the building 
 must be considered. I have attempted to give the efi^ect of S. 
 
S Mark 
 
 Page 79. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 79 
 
 Front as I saw it. The windows in the domes being stopped up, 
 the domes themselves are dark, of course with considerable grada- 
 tion. The principal light, admitted through the windows of the 
 fronts, rests upon the pier masses, but is gently diffused through 
 the whole, without any strong contrasts. The accidental throwing 
 back of the springs of the arches which we remarked, puts these 
 masses into a somewhat bolder relief than would otherwise be the 
 case. I have not seen S. Mark's, but give a sketch for which I 
 am indebted to a friend, which shows the actual effect of light and 
 shade. Here the light is introduced in great measure through 
 windows at the base of the dome, as originally at Perigueux. 
 These small openings must rather catch the eye, and disturb the 
 harmony. But Sir Christopher Wren brings the light through 
 the top of the domes over his nave, and the effect is inconceivably 
 fine. The domes are now light instead of dark, the arch com- 
 partment dividing them, in deep shade. One or other of the 
 opposite pier masses, catches a strong light, and is insulated from 
 the roof it supports by a sparkling cornice. The alternations of 
 light and shade lead us to the central dome, uniformly pervaded 
 by a delicate aerial tint, and the choir and apse beyond are massed 
 in a faint shadow. I have little doubt that the building, if carried 
 out on the scale intended, would have shown in its interior the 
 same distribution of light and shade (one certainly unattainable 
 in Gothic architecture) which we perceive in the model. 
 
 Perigueux could easily be put into revived Italian, but could S. 
 Paul's be put into E-omanesque ? Can an interior, similar in its 
 composition, in its distribution of light and shade, and in the 
 grandeur of its architectural masses, be produced in a pure, con- 
 sistent, and refined round-arched style ? If so, it is surely worth 
 while to try and solve the problem. The admission of light through 
 the top is easy enough. Where the principal dome is covered with 
 a tower, the opening of the former might be simply glazed, and 
 the windows of the latter placed at a sufficient height to throw the 
 light well down upon it. The other domes might be covered 
 with glazed cupolas of stone or wood, or else lighted by dormer 
 windows in the roof. 
 
80 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 I am convinced that the Mosaics of S. Mark's have, and were 
 intended to have, the effect of isolating the roofs from the pier 
 compartments below, so as to allow, without impropriety, an ar- 
 chitectural completeness to be given to these masses. 
 
 How to give this completeness, and to attain to, or exceed the 
 dignity of the piers in the model of S. Paul's, is the difficulty, and 
 this is the problem I throw out for the consideration of architects. 
 
 ROULKT. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 81 
 
 CHAPTER VII. 
 
 We will now resume our wanderings, with a less definite purpose, 
 but still in the hope of occasionally meeting with something that 
 may please or interest us. We return then once more to the 
 
 M 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 83 
 
 the capitals of the jambs to the belfry windows. Possibly a spire 
 was intended, as at Loches. The nave has a pointed barrel roof, 
 but as it has also some transverse timber beams, I cannot be 
 certain that a part of it at least is not plaster. The roofs of the 
 transepts and chancel are unquestionably of masonry, and it is 
 difficult to pronounce whether they are pointed or cylindrical. 
 Those of the apsides are semi-domical. The tower compartment 
 has a hemispherical dome with pendentives, precisely similar to 
 those of Perigord, and resting on pointed arches. All the win- 
 dows are round-headed ; but the western door is pointed, and 
 consists of two plain square orders and a label. There is con- 
 siderable enrichment externally towards the east end ; a cor- 
 nice of considerable projection is sup- 
 ported on rather massive engaged 
 shafts, with capitals of foliage and 
 square abacus. In the intervals are 
 some curious brackets, one of them 
 is very like a coffee-pot. The situa- 
 tion of this church, overlooking the 
 village and river, is extremely pleas- 
 ing. 
 
 But the principal feature of the 
 place is the steeple of the ruined 
 
 abbey, which must have been a building of considerable size and 
 
84 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 importance. It is a fine tower and spire, situated evidently at the 
 west end, but not having a central door for a principal entrance. 
 It has stages of Romanesque windows, and some sculpture in the 
 west wall externally ; and also some curious masonry, which, though 
 in stone, seems to have been imitated from the brick masonry in 
 the pile of Cinq Mars. The spire, which is octagonal, is so 
 arranged that an angle corresponds with the middle of each face 
 of the square tower ; and one, consequently, with each angle. I 
 did not see what the pendentives were. Considerably above its 
 base is a stage of spire lights, one on each face, eight in all, of 
 pure Romanesque, exhibiting the character rather of the eleventh 
 than twelfth century. I am not acquainted with any octagonal 
 
 spire exhibiting earlier decided characteristics. There are several 
 large Romanesque fragments among the remains of the church, of 
 which it would not be very difficult to trace the plan. The 
 monastic buildings are on the north side of the church. A long 
 range, attached to the tower, appears to belong to the thirteenth, 
 or early part of the fourteenth century ; and the cloisters have a 
 beautiful arcade of plain pointed arches with shafts and capitals, 
 very like our early English. The effect of some windows of a single 
 light, with a free trefoil feathering in the head, is extremely good. 
 Near the east end are the remains of a polygonal chapel of later 
 date ; apparently Flamboyant. 
 
Page 84. 
 
ARCHITECTURAL STUDIES IN FRANCE 
 
 85 
 
 Following the course of the Indre downwards, we pass Evres 
 and Veignes, both of which have spires set diagonally, as at 
 Cormery. The tower, in each, stands on the south side, near the 
 chancel ; the churches are picturesque, but afford little to interest 
 the antiquary. 
 
 S. Branch, south of the Indre, at a few miles distance, has a 
 cross church, of which the central tower and chancel were partly 
 rebuilt in the course of the last century, in consequence of injuries 
 caused by the fall of the spire. Still the arches under the tower 
 appear old ; and if plainness be any proof of antiquity, they must 
 be very old. A single square order, on a massive impost of the 
 same section, the arch resting on a very simple string, constitutes 
 the whole. There is no vaulting to the compartment. The north 
 transept is vaulted with ribs, apparently of twelfth-century work. 
 The nave is plain, without aisles, and unvaulted ; somewhat 
 wider than the chancel, and lighted by narrow round-headed 
 windows. The round-headed western door has an appearance of 
 great antiquity. This, and some other naves of churches which 
 we shall have to mention, should be carefully examined, and com- 
 pared together ; as it is more than probable that some of them date 
 from an earlier period than the eleventh century. 
 
 Passing the ancient keep of Montbazon, which, like Loches, is 
 perfectly Norman in its character, though it has the cylindrical 
 buttress, we next arrive at the little village of Monts, whose 
 picturesque church ranges over a period of five or six centuries ; 
 for I should be quite as much inclined to attribute to the tenth 
 as to the eleventh century the lower part of the south tower. 
 The principal apse evidently belongs to the twelfth ; and the north 
 side presents the row of gables so common in additions of the 
 fifteenth and sixteenth. The tower compartment, though not cen- 
 tral, has a hemispherical dome, the pendentives of which are little 
 more than flat projections. The arches are semicircular, and 
 perfectly plain. It is not improbable that this tower was originally 
 central, and that the present south aisle of the nave and chancel 
 occupies the place of the original nave and chancel themselves. 
 This conversion of a small church into the aisle of a much larger 
 
86 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 one, seems to have been by no means uncommon, and that at an 
 early period. The principal apse is semicircular, and has late 
 Romanesque work, tolerably rich. The bearing arches of this 
 portion are pointed. 
 
 MONTS. 
 
 The next village is Artannes, the church of which deserves a 
 very careful examination. To its east end is attached a castellated 
 mansion, apparently of the fifteenth century ; but most of the 
 church is earlier. The chancel, which is not apsidal, though it 
 probably replaces one of that form, seems a work of the twelfth 
 century, and has some round-headed windows, though the arches 
 are mostly pointed. The vaulting is ribbed, and somewhat domi- 
 cal. The central tower is low, and covered with a wooden spire. 
 It has a pair of round-headed windows in each face, of two plain 
 square orders. It is often difficult to decide whether work of this 
 character belongs to an early Romanesque period, or to a com- 
 paratively modern restoration ; and on looking at the interior, I 
 somewhat hesitated in assigning the earlier date in the present 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 87 
 
 instance. The transepts are Romanesque, but the northern one 
 has a large eastern aisle, which forms a double transept ; this is of 
 a late date. The nave, I am convinced, is very old. It is much 
 
 Artannes. 
 
 wider than the area of the tower, so much so that it migbt be 
 divided into aisles ; and, in fact, opens into the tower and transepts 
 by three arches. These are plain and round-headed,*' as also are 
 the other arches of the tower. The vaulting of its compartment 
 is evidently later than the piers. The nave is not vaulted ; its 
 windows are round-headed, narrow, and with deep jambs ; but not 
 
 * The northern one is pointed, but appears to have been altered. 
 
88 
 
 ARCHITECTUEAL STUDIES IN FRANCE. 
 
 much splayed ; very different from any windows in the chancel. 
 The west end has been buttressed up at a later period ; nor, I 
 think, does its door exhibit early features. 
 
 Pont Ruan, as my guide assured me, is a very old church, 
 founded in the sixth century. Now, in a small village church we 
 are perhaps more likely to meet with some of the original structure 
 than in fabrics of greater size and importance. I believe, French 
 antiquaries lay great stress on the petit appareil," the masonry 
 of small, nearly square stones, pretty regular, and not very closely 
 jointed ; and consider that where it appears very decidedly, at 
 least in certain districts, it aro-ues a date more remote than the 
 
 Pont Ruan. 
 
 tenth or eleventh century. The nave of this church certainly 
 presents the masonry in question more clearly than either S. 
 Branch or Artannes ; but not quite so clearly as Chanceaux, a 
 church on the other side of the Loire, which I shall mention in its 
 place. The windows are round-headed and narrow, and there is 
 an early Romanesque north door, though supposed to be much 
 
VlLL/VDRY. 
 
 Page 89. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 89 
 
 later than the building to which it is attached. As in the last 
 example, the nave is much wider than the chancel ; but as there 
 are no transepts, a window (of the thirteenth century) is inserted 
 on each side of the chancel arch. This part of the church also 
 shows the commencement of a division into aisles, which however 
 does not seem to have been carried further. The chancel is early 
 pointed, and loftier, though narrower, than the nave ; its east end 
 is flat, and over the chancel arch is a small bell gable. 
 
 Near this, also upon the Indre, is Sache ; the church of which 
 consists of a nave, and angular apsidal chancel. On the south side 
 is the tower, of early pointed work ; on the north an aisle has been 
 added presenting a series of gables. The old part of the church, viz. 
 the nave and chancel, belongs to the end of the twelfth century, and 
 is decidedly Angevine ; the compartments being square, the vault- 
 ing very domical, and the engaged piers massive and projecting. 
 The bearing arches are pointed, those of the windows mostly 
 round. 
 
 There is also something Angevine in the church of Azay le 
 Rideau ; but the oldest part, now the north aisle, tower and its 
 eastern apse, seem, as in other instances, to have constituted 
 formerly the whole church. 
 
 Such is also the case with Valleres, which has a massive southern 
 tower with an eastern apse ; the tower compartment has a cylin- 
 drical vault, the apse semi-domical. The actual chancel is poly- 
 gonal, and seems to belong to the end of the thirteenth or 
 beginning of the fourteenth century. The rest of the church is 
 restored. 
 
 This church lies near the Cher, at a short distance from its 
 junction with the Loire ; and we will now trace upwards the 
 course of the former river. 
 
 Villandry has a Romanesque church, combining in itself the chief 
 characteristics, one or other of which is found in most churches of 
 this district. It has a nave, transepts, central tower, and three 
 apsides, viz. one to the chancel, and one east of each transept. 
 Except a few insertions the whole is of one style, and, I should 
 think, belongs tp the eleventh century. The nave is wider than 
 
 N 
 
90 
 
 AKCHITECTURAL STUDIES IN FRANCE. 
 
 the chancel, without aisles, very plain, and unvaulted. It has a 
 
 rather handsome round-headed 
 western door. The arches under 
 the tower are pointed, of two 
 plain square orders ; their im- 
 posts have shafts and capitals. 
 The transepts and intermediate 
 compartment of the chancel (that 
 between the intersection and 
 apse,) have a pointed barrel 
 vault ; the apses themselves, 
 which are semicircular, a semi- 
 domical one. The tower com- 
 partment has the circular dome 
 (as in Perigord) on Byzantine 
 pendentives. The tower is square 
 and massive, and the sides which 
 are perfect have round-headed 
 arches in the belfry stage, of 
 which the orders and imposts have square sections. On each side of 
 the western tower arch is an opening from the nave to the transept, 
 '/ , forming a semi-arch; and at this 
 
 part of the nave there is a small 
 lateral recess, which I do not re- 
 member to have noticed in other 
 churches. A handsome chateau of 
 revived Italian, with some earlier 
 portions, adjoins the village. 
 
 Savonnieres. — A nave with 
 aisles, a short polygonal chancel, 
 and a tower with a lofty wooden 
 spire, between the nave and apse. 
 The arches are mostly pointed ; 
 still the style, at least in the earlier 
 parts, is decidedly Romanesque. 
 This applies to most of the piers 
 
 ViLLANDKY. 
 
 Savonnieres. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 91 
 
 and arches, and to a beautiful south doorway, which is pointed, but 
 has its architrave enriched with good Romanesque sculptures. 
 
 The drive from Tours through Savonnieres and Villandry to 
 Valleres is a very agreeable one ; and by crossing the river at 
 Langeais, and returning to Tours on the opposite side, you would 
 make a very satisfactory day's excursion. 
 
 S. Martin le Beau (de Bello, so called from a victory obtained 
 over the Normans in the ninth century by the aid of the relics of 
 that saint) is a Romanesque church of the twelfth century, with 
 
 St. Martin le Beau. 
 
 late additions. The west door is rich, and exhibits some very 
 delicate workmanship ; it is round-headed. The apse is semi- 
 circular, and is remarkable for not having a central east window, 
 the number being eight, and consequently the eastern extremity 
 presents a compartment between two windows ; these are narrow, 
 round-headed, and of good workmanship. The tower, a low 
 plain one with a round-headed arcade, stands on the south side. 
 
 La Croix, Blere. — A small cross church, with semicircular apsides 
 to the chancel and east sides of the transepts. These are Roman- 
 esque ; but the rest seems to have been considerably repaired, or 
 rebuilt, in the fifteenth or sixteenth century. The little central 
 
92 ARCHITECTURAL STUDIES IN FRANCE. 
 
 octagon is elegant in its proportion, but does not contain anything 
 remarkable in detail. 
 
 Blere. — An early pointed octagonal tower and stone spire, the 
 diagonal faces being narrower than the cardinal, and all having 
 fine lancet windows with shafted jambs. The abacus is square. 
 On the north side this appears to be the central tower of the 
 church, having a small transept, a semicircular eastern apse, and 
 a Flamboyant nave. But the greatest part of the present church 
 is attached to this on the south side, and yet presents some early 
 work at the south-west corner. With the exception of this, the apse 
 I have mentioned, and the steeple, the work is chiefly Flamboyant. 
 
 Outside the town is a beautiful little chapel of debased Gothic. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 93 
 
 It is square, with a polygonal apse, and crowned with an octangu- 
 lar dome. This, however, does not appear internally, the roof being 
 groined with ribs. 
 
 Civray must not be confounded with a town of the same name 
 having a fine church, which we shall notice in a subsequent 
 chapter. The church is plain externally, with a central tower. 
 
 This has much of that ambiguous character we remarked at 
 Artannes, indeed the masonry gives a very strong evidence of 
 late restoration. The internal arrangement is very similar to 
 Artannes, the western arch of the tower compartment standing 
 between two others which open into the nave. All are very plain. 
 The church has no transepts, but the tower being narrower than 
 the nave and chancel has a compartment on each side. 
 
 Passing near the well-known chateau of Chenonceaux, we 
 arrive at Chisseaux, which has an octagonal tower between the 
 nave and chancel. The latter is polygonal, and evidently belongs 
 to the twelfth century. The nave has every appearance of being 
 much earlier. 
 
 There is much to induce the artist to linger at Montrichard, 
 which has a castle with a fine keep of Norman character; a 
 
94 
 
 ARCHITECTUEAL STUDIES IN FRANCE. 
 
 Romanesque church with an apse, and southern tower ; and a 
 cross church of rather later style, with a central tower, much 
 restored. The town itself is of an ancient character, and affords 
 some extremely picturesque street views. But when I last went 
 through it I had not much time to spare, as I was anxious to visit 
 the interesting ruin of Aiguevive, a little to the left of the road 
 to Loches. It stands in private grounds, but thanks to the liberality 
 of the proprietor, no obstacle is presented to the tourist who wishes 
 to examine it. An account is given of this building in a very 
 pleasing article in the Proceedings of the Touraine Archaeological 
 Society. It is a pure specimen of the transitional style. The 
 church consists of a nave with aisles, central tower and transepts, 
 
 Aiguevive. 
 
 and the usual triapsal east end. All the bearing arches are 
 pointed, those of the doors and windows round. The upper part 
 of the tower forms an octagon of unequal sides, the cardinal faces 
 having two belfry windows, the diagonals one. It is crowned with 
 a lofty stone spire. The piers of the nave are cylindrical columns, 
 not very massive ; the capitals are elegant in their design, but 
 have not much minute or delicate workmanship. The roof, which 
 has been removed, was a pointed barrel vault, having for abutment 
 a half-barrel vault over the aisles. The west end has a large 
 porch or narthex. A few of the monastic buildings remain, forming 
 
I 
 
St. Denis, Amboise. 
 
 Page 95. 
 
St. Denis, Amboisk. 
 
 St. Denis, Amboise. 
 
 Tagk 95. 
 
Page 95. 
 
ARCHITECTUKAL STUDIES IN FRANCE. 
 
 with the church a picturesque group, 
 which is further set off by the delightful 
 scenery which surrounds it. The simi- 
 larity in composition of the steeple to 
 that of Blere will be noticed by the 
 architect. 
 
 On the south bank of the Loire, above 
 Tours, is Montlouis, a church indebted 
 for its picturesqueness to its situation, 
 and additions of latest Gothic period. 
 The chancel and apse are Romanesque. 
 The exterior of the latter has a bold 
 cornice on brackets, and massive en- 
 gaged shafts with capitals. 
 
 St. Denis, Amboise, is a cross church, 
 with a massive central tower. The style 
 is Romanesque and transitional ; one of 
 the bays has something of the Ange- 
 vine vault. 
 
 I 
 
 9- ^^^^8?^ " ' 
 
 St. Denis, Amboim-;, 
 
 MOMTLOUIS. 
 
96 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 On the north side of the Loire, opposite Tours, is S. Sympho- 
 rien, a Romanesque cross church, swamped by a Flamboyant 
 nave. 
 
 Following the course of the river upwards, we come to S. Rade- 
 gonde, a Romanesque church built against the rock, on the top 
 of which, north of the nave, stands the tower, which communicates 
 with a chapel beneath, cut in the rock. 
 
 St. George, 
 
 The chapel of S. George is also in part subterranean. The 
 tower compartment, between the actual chancel and a chamber 
 cut in the rock, has a circular dome on round arches, which seem 
 of considerable antiquity. 
 
 Rochecorbon, distinguished by the slender square turret crown- 
 ing its perpendicular face of rock, has, in the valley below, a 
 Romanesque church, with central tower, transepts, and apse. 
 
Cinq Mars. 
 
 I'AGli 97. 
 
ARCHITECTUKAL STUDIES IN FRANCE. 
 
 97 
 
 The nave is without aisles, and has narrow round-headed windows. 
 The belfry stage of the tower is later, and has pointed arches. I 
 should not suppose any part of this building to be earlier than the 
 twelfth century. The rocky banks here, as in other parts of the 
 Loire, as well as the Cher, and the Indre at Loches, are scooped 
 out into numberless habitations, sometimes rising above each other 
 in terraces, the roof forming a vineyard. 
 
 Vernou. 
 
 Vernou has an early Romanesque west front of rather curious 
 composition ; the rest of the church is transitional, or early pointed. 
 On the same side of the river, below Tours, is the chateau of 
 Luynes, near which are the remains of an aqueduct. 
 
 The very curious pile of Cinq Mars is a solid tower of brick- 
 work, crowned with four pinnacles ; it is supposed to have had a 
 fifth in the centre. The south face has some ornamental work, 
 which seems, as we have remarked, to have found imitators among 
 the architects of the eleventh and twelfth centuries. A gateway 
 remains of the chateau of Cinq Mars. The church has a central 
 tow^er with a stone spire, transepts, and apse. The nave is without 
 aisles. The tower compartment has a cylindrical vault, and its 
 arches are round, of a single square order, with a plain impost 
 
 o 
 
98 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 and string. The intermediate compartment has a cylindrical 
 vault ; the apse, and that of the south transept, semi-domical. 
 
 Cinq Mars. 
 
 Neither the nave nor transepts are vaulted. On the north side of 
 the chancel is a later structure. 
 
 Langeais has the remains of a Romanesque keep, and a very 
 fine chateau of the fourteenth century, with massive round machi- 
 colated towers. The church appears to have some early work 
 near the east end, which is triapsal ; and some old shafts and 
 arches are arranged on the south side, so as to form a covered 
 walk externally. The tower is at the west end, and has some 
 good Romanesque round-headed windows, enriched with the orna- 
 ments of the period ; it has a handsome stone spire. 
 
 I have already alluded to Chanceaux, a few miles north of 
 Tours ; the church consists merely of a nave and eastern apse. 
 The former is rather high in its proportions, and has plain 
 round-headed windows ; the masonry is small, and very distinctly 
 
Pack 99. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 99 
 
 marked ; in fact, it is the most decided specimen of the petit 
 appareil" that I have seen. The roof is not vaulted, and supports 
 the small wooden belfry, common in this district. The domestic, 
 or conventual, buildings adjoining; have good Romanesque work, 
 probably much later than the church. 
 
 Rouzieres has a cross church, of which 
 the central tower is destroyed. The style 
 is Romanesque, though the arches of con- 
 struction are pointed. The apse has good 
 work externally. 
 
 On the south side of the river, near 
 Plessis le Tours, are the remains of the 
 Abbey of S. Come. These are not very 
 extensive. The church, of which part of 
 the apsidal aisle, with its radiating apses, 
 remains, must have been a large one. An 
 adjoining chapel, of earlier date, has some 
 highly ornamented windows. 
 
 The Romanesque and earliest pointed 
 are the only styles which afford much in- 
 terest in the churches near Tours. The 
 town itself has good specimens of the later 
 styles. For instance, much of the cathe- 
 dral, the towers of which are finished in 
 cinque cento ; Notre Dame la Riche ; and 
 
 S. Clement. The Gothic village churches are generally plain, 
 and have a wooden belfry on their roof. Almost every central 
 tower (of stone) is Romanesque or transitional, or on foundations 
 belonging to those periods. I think Mettray was the only case I 
 had any doubt about in this respect, and here there is nothing 
 to assure us of the contrary. The capitals are often delicately 
 worked with foliage. The cushion capital does not make its 
 appearance. The billet moulding is very common, as also a 
 chequered moulding, not unlike the billet in composition, but 
 sharper in its effect, and more powerful than our hatched mould- 
 ing, to which it has some affinity. 
 
 Rouzieres. 
 
100 AECHITECTURAL STUDIES IN FRANCE. 
 
 I have no doubt that had I extended my excursions in this 
 province, especially along the course of the rivers,- and made 
 careful notes and sketches, I might have collected a vast amount 
 of useful and interesting matter. 
 
 S. Come. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 101 
 
 Central Tower, Bassens. 
 
 CHAPTER VIIL 
 
 If, in travelling from Tours to Poitiers we give ourselves time 
 to examine the objects on and near our route, we shall probably 
 find many churches similar in character to those we noticed in the 
 last chapter. But the style of Poitou has, I believe, certain pecu- 
 liarities of its own, which would make an architectural tour through 
 the province extremely interesting. This I suppose to be the case 
 from one or two articles that I have read in the Bulletin Monumen- 
 tal, for I have not had an opportunity of forming any observations 
 personally. From a railroad we can see no more than the out- 
 lines of a few churches, and those imperfectly, but they seemed to 
 promise well. I may here, however, observe that a French rail- 
 way, generally speaking, shows you more of the country than an 
 English one ; and being usually carried, where possible, along the 
 bank of some river, brings into view a series of villages which a 
 
102 
 
 ARCHITECTUKAL STUDIES IN FRANCE. 
 
 line, laid down according to the principle of many of our own, 
 would have avoided or hidden by cuttings and tunnellings. The 
 difference will be appreciated by any one who has emerged from 
 the London and Birmingham line into the None valley through 
 Northamptonshire. Still, the only impression that I could receive 
 during my rapid journey between Tours and Poitiers was, a wish 
 that I could have travelled over the same ground more leisurely, 
 and given more time to Poitiers itself. 
 
 If there were no buildings of interest in the town, its picturesque- 
 ness, as regards situation, would repay the artist. By putting 
 a tea-cup upside down in a saucer, you may form to yourself an 
 idea of its position. The valley of the Clain almost encircles the 
 high ground on which the town is built, the opposite bank also being 
 precipitous, in many parts showing faces of gray, bare, perpendicu- 
 lar rock, which here and there forms lofty detached pinnacles. 
 The town is connected with the high ground beyond its circum- 
 scribing valley by a natural causeway of no great width, over 
 which runs the road to Angouleme. On every other side the streets 
 and roads into the valley are very steep. It will be seen at once 
 that this position must in ancient times have been a very strong 
 one, and it was no doubt occupied as such from a very remote 
 period. The present remains of its defences, however, appear to 
 be principally mediaeval ; the Roman relics being the amphitheatre, 
 the area of which seems pretty well preserved, but does not present 
 many architectural features ; and an aqueduct outside the town. 
 The Cathedral and the church of S. Radegonde, which we have 
 noticed as illustrating the Angevine style, stand near the base of 
 the hill on the eastern side, fronting some precipitous rocks on the 
 opposite bank of the river. On the northern side, near an old 
 bridge, and also commanded by some high rocky ground on the 
 further bank, is the church of Montierneuf, which notwithstanding 
 some partial destruction, much alteration in the 14th century, and 
 some modern restoration, is still a very fine specimen of Roman- 
 esque. It is a cross church with central tower. The nave has aisles, 
 and a semi-cylindrical roof. The compartment of the tower is 
 roofed by a hemispherical cupola. Most of the arches are semi- 
 
Page 102, 
 
5 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 103 
 
 circular, with plain square orders. The choir has a lofty clerestory 
 of the 14th century, with flying buttresses ; the lower part and its 
 aisle still retain their Romanesque features. The tower is square 
 and massive ; indeed the heightened clerestory of the choir ranges 
 pretty nearly with its upper line, but above it are the remains of a 
 circular stage, also of Romanesque work, flanked at the four cor- 
 ners of the tower below with circular turrets having conical tops ; 
 of these only the two eastern ones remain. It is difiicult to obtain 
 a satisfactory view of this church externally ; the best are from the 
 opposite side of the river. 
 
 On higher ground than this building stands the church of S. Hi- 
 laire, still on the outskirts of the town, on the western slope of the 
 hill. Here the valley is somewhat wider, and the stream which 
 traverses it falls into the Clain a little below Montierneuf. The 
 church is mentioned by M. de Verneilh as having some connection 
 with those of Perigueux, &c. ; he thinks it not improbable that 
 when the nave was perfect it had the same description of spherical 
 dome. This part, however, is very incomplete both as regards length 
 and breadth, the arrangement having comprehended a double aisle 
 on each side. The transepts are long, the choir is apsidal, with 
 an aisle and chapels. The central tower is low, and its compart- 
 ment internally has a dome of eight sides. There is also a tower 
 on the north side of the nave, the lower part of which seems 
 ancient. The church is said to have been commenced about the 
 end of the 10th century, and completed at an early period in 
 the following. Its interior is still very striking ; but it is almost 
 impossible to obtain a satisfactory external view. 
 
 The central tower, choir, and transepts of another church are 
 seen on the eastern slope of the town, from the opposite bank of 
 the river. They appear to be Romanesque, and the nave is de- 
 stroyed. I was not able to see the interior. I believe it belongs 
 at present to some religious house. 
 
 Near the amphitheatre is the apsidal end of a church, the rest 
 of which has been destroyed. It has some beautiful columns and 
 capitals of Romanesque work, and is evidently part of a large and 
 important structure. The artist would find it a delightful study. 
 
104 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 On nearly the highest part of the town, and occupying almost a 
 central position, is the fine church of Notre Dame. Fortunately 
 an open space gives a clear view of its fine western front, as well 
 as its south side. The front is well known by engravings ; I, how- 
 ever, give my rough sketch of it, as it will enable me to point out 
 some peculiar features in Poitevin architecture. The part corre- 
 sponding with the central aisle or passage of the nave is crowned 
 with a gable, which is ornamented with a vesica. The sloping 
 compartments of the aisles are flanked externally by large round 
 turrets with conical tops. The upper stage of these has a rich 
 Romanesque arcade resting on a bracketed string ; there are also 
 brackets to the cornice below the spire. The lower stage, which 
 is not much inferior in height to the wall of the aisle, has a combi- 
 nation of tall and massive engaged columns, ranged in couplets, 
 with rich capitals of foliage, and buttress-slopes resting on the 
 abacus. * In the intervals are vertical strings or lines of foliage. 
 The principal door has a semi-circular arch, and is recessed in 
 several orders. On each side is a recessed pointed arch com- 
 prising a pair of round ones. The chief window, which has a semi- 
 circular arch, is flanked by two tiers of round-headed niches with 
 figures ; every part, not excepting the spaces of masonry, is covered 
 with elaborate work. There are no regular transepts, but the tower 
 stands between the nave and choir. Its lower stage is square, 
 with a stair-turret on the south-east angle. The next stage is also 
 square and narrower, but with wide square buttresses compre- 
 hending the angles. The upper stage is circular, with a conical 
 roof. All the arches are round-headed. The roof of the nave is 
 cylindrical, and there is no clerestory. The aisles are cross vaulted. 
 The piers are rectangular, with engaged columns or shafts. 
 
 S. Porchaire has a Romanesque western tower, exhibiting some 
 very bold grouping of columns. The church itself is principally 
 of late Gothic. 
 
 S. Jean, now the museum, is unquestionably of a date much 
 earlier than the year 1000, and perhaps nearer Roman than Me- 
 diseval. Its plan is an oblong, standing north and south ; but to its 
 west and east sides are attached a porch and a chancel, to give it the 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 105 
 
 character of a church. The south wall has some curious work, in 
 which the straight- sided arch appears, also some round arches of 
 very Roman character ; brick, as well as stone, is used in the con- 
 struction. 
 
 There are other churches and buildings of great interest and 
 picturesque beauty to be found in the ancient streets of Poitiers, 
 which contain specimens of every style, including some very fine 
 ones of the Revival. But the notes I have made might be very 
 much expanded, and, I have no doubt, corrected by any one able 
 to examine this fine old city with proper care and attention. 
 
 The west front of Ruffec church is a rich specimen of Romanesque 
 work, with a few of the Poitevin peculiarities. It has a central 
 door between two others (now blocked up) all round-headed and 
 much ornamented. The compartments in which these stand are 
 separated by massive columns tapering in three stages to the line 
 of the gable over the front, and they are also flanked by groups 
 of similar shafts in two stages. Above the door is an arcade, en- 
 riched with figures ; with a central window, originally also round- 
 headed ; and above this is some work in a pointed recess, perhaps 
 corresponding to an older gable. To this front are attached on 
 each side later compartments, corresponding with the aisles of the 
 enlarged church, the whole of which, with the exception of a little 
 of the old work retained, is Flamboyant. The tower, low and 
 massive, is engaged in the present north aisle, and has its interior 
 compartment vaulted with a barrel roof of Romanesque date, appa- 
 rently earlier than the front. 
 
 About ten miles north-east of Ruffec is Civray, which has a fine 
 cross church of a late Romanesque or Transitional character. 
 The elevation of the west front is nearly square, and is divided 
 by a string into two stages. Each of these has three large semi- 
 circular arches. The central one of the lower stage forms the 
 principal doorway, and that on each side comprises a pair of 
 round-headed arches on a shaft. The upper stage is divided 
 into three compartments by engaged shafts running up into a 
 corbelled string under the parapet. At the north and south 
 angles of the front are groups of large shafts, running nearly 
 
 p 
 
106 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 up to the parapet, and showing in the intervals the vertical band 
 of foliage. The central arch in the upper stage contains the west 
 window of the nave, the others have sculptured figures. The cen- 
 tral tower is octagonal, in two stages, of which the lower one is 
 
 CiVRAY. 
 
 open as a lantern. The windows in these have the round head, 
 and the string and cornice are bracketed. At the angles are bold 
 groups of columns, which, as well as those on the wall of the apsi- 
 dal chancel, have the telescope form that we see in Ruffec. The 
 nave has four bays, the pier arches being pointed, with two square 
 orders. The roof is the pointed barrel vault, without triforium or 
 clerestory. The aisles have a cylindrical vault, with transverse 
 ribs of one square order, each resting on an engaged column. The 
 arches from the aisles into the transepts are round. Those sup- 
 porting the tower are pointed. The squinches under the diagonal 
 
ARCHITECTURAL STUDIES IN FRANCE. 107 
 
 sides are triangular. The west front may be studied well from the 
 principal square of the town, but I have preferred sketching the 
 east end (which in Romanesque churches always affords a fine 
 outline) from a garden. Some good distant views might be ob- 
 
 tained, as the scenery round the town, which stands in a valley, is 
 very pleasing. 
 
 Courcome, about four or five miles to the south-west of Ruffec, 
 has a very interesting old church, which is mentioned in the Sta- 
 tistique Monumental de la Charente. It consists of a nave, origi- 
 nally without aisles ; an apsidal chancel ; short transepts with 
 eastern apsides ; and a central tower. All this is Romanesque, but 
 of different periods. A south aisle has been added to the nave 
 about the 15th century. M. Michon considers the nave and the 
 central cupola (as seen inside) to belong to a date earlier than the 
 
108 ARCHITECTURAL STUDIES IN FRANCE. 
 
 year 1000, but these two again to differ in date, the nave being the 
 earliest. The apse, the transepts, the west front, and the upper 
 part of the tower, belong to the 12th century ; the apse (accord- 
 ing to the opinion of the writer I have quoted) being the oldest. 
 
 COURCOME. 
 
 It is in such buildings as the present, where the difference in cha- 
 racter between different parts of a Romanesque work is stronger 
 and more evident than even between the latest Romanesque and 
 Gothic, that we are induced to assign a remote date to certain 
 portions, and to seek for the characteristics of a style belonging to 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 109 
 
 an earlier period than we feel justified in confidently fixing as the 
 date of any of our own buildings. In the example before us the 
 rude and barbarous carving of the central piers distinguishes them 
 from the elegant and refined work that prevailed in the 12th and 
 
 COUECOME. 
 
 much of the 1 1 th century ; and again this part appears to be an 
 evident addition to the plan of which the nave forms a part, if we 
 may judge both from the manner in which they are connected to- 
 gether, and from the character of the work. The north side of 
 the nave consists of four bays, each having a round arch of two 
 plain square orders resting on a string, the impost being of the 
 same section of the arch. In the wall is a plain round-headed 
 window. Between the bays is a massive engaged shaft, with a 
 rudely, but not carelessly, sculptured capital, and a square abacus 
 enriched with the billet, a band of which runs at the same height 
 forming a cornice at the spring of the barrel-roof. The shaft I 
 have spoken of carries a round arch of one square order, forming 
 a transverse rib to the roof, which is semi-cylindrical, perhaps a 
 
110 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 little stilted. The easternmost of these 
 ribs does not coincide with the tower 
 arch, but stands to the westward of it, the 
 interval being filled up by a broad seg- 
 mental arch introduced for strength at a 
 later period. The tower pier is a cluster 
 of five, set close together and not sepa- 
 rated, as in most Norman work, by any 
 salient angles ; I give a cut from a tower 
 pier at Bocherville to show the differ- 
 ence. The outer shafts, which are very 
 massive, bear the inner order of the 
 arches ; those adjoining them, the outer 
 order ; and the central shaft (which would 
 
 Bocherville. 
 
 have borne the diagonal rib had the area been cross vaulted) 
 merely occupies a place under the re-entering angle of the square 
 part of the tower. The cupola is octangular, the diagonal being 
 
 COURCOME. 
 
Page 111. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 Ill 
 
 narrower than the cardinal sides ; the former are supported on 
 Romanesque squinches, above which runs a band of billets round 
 the whole. In each of the cardinal sides is a round-headed 
 arch, not at present opened for light. The roof of the cupola 
 is part of a spherical dome. Hitherto we have seen only the 
 round arch ; but the intermediate compartment of the chancel 
 has a pointed barrel-roof, and its arch into the apse is pointed. 
 The windows of the apse are round-headed. The upper stage 
 of the tower is a very elegant composition of late Romanesque. 
 It is square, and each face is divided by shafts into four com- 
 partments, a single shaft occupying each angle, and a single 
 one also forming the division. These shafts run up to the 
 cornice under the parapet, which rests on their capitals, and in 
 every interval is a bracket. Each compartment has a round- 
 headed tall arch on shafts, with enriched label. The two central 
 ones have pierced windows, corresponding in shape. This stage, 
 though of massive appearance in the general outline, is narrower 
 than the lower one, to which it is attached by a bold slope of stone- 
 work. The west front of the 12th century has the character com- 
 mon to village churches of the same class and date, and of which 
 I may be able to give an idea in describing the next church I shall 
 bring before your notice. 
 
 La Palud, though dissimilar in outline, is in many respects very 
 like Courcome. This church stands near the beautiful ruins of 
 the Abbey of la Couronne, six or seven miles from Angouleme on 
 the Bordeaux road. It belongs (I still keep M. Michon's work 
 open before me) to two very distinct epochs of Romanesque. The 
 central part of the nave is of the most remote period of that style. 
 The front part of the nave, the cupola, the transepts and apse are 
 of the end of the eleventh century. The composition of the nave 
 is very like that of Courcome, having the arched bays on plain im- 
 posts, the semi-cylindrical roof, and the shaft supporting the trans- 
 verse arch of one square order. But it has also clerestory win- 
 dows pierced through the vaulting, and consequently forming 
 curves of a double curvature. The nave has no aisles. The cu- 
 pola rests on Romanesque pendentives, the apse is semicircular, and 
 
112 
 
 AKCHITECTURAL STUDIES IN FRANCE. 
 
 the transepts have eastern apses. On the intersection is a very 
 elegant octagon, tapering in stages, and crowned with a conical 
 spire. The belfry story has in each face a couplet of round arches 
 on shafts, the central one bisecting the real opening, a single round- 
 
 headed window . The spire has the fir-apple ornament which we 
 see at Roulet, Perigueux, &c. I think the upper part of this tower 
 must belong, as at Courcome, to the 12th century. The west front 
 is a good composition, and a very fair specimen of the prevailing 
 Romanesque fronts of the district. 
 
ARCHITECTUEAL STUDIES IN FRANCE. 
 
 113 
 
 Barbezieux, on the road to Bordeaux, has the western part of a 
 large church. It seems to have been late Romanesque, but much 
 altered. The piers are formed of shafts clustered in a remarkable 
 manner. The tower stands at the north-west angle. 
 
 Barbezieux. 
 
 Between this and Bordeaux I noticed near the road several small 
 churches that seemed as if they would repay the visitor, but I was 
 not able to examine any object except S. Andre, Cubsac, a very 
 fine, though simple church — chiefly Romanesque. 
 
 The tower, which has some good round arches, but has been 
 increased in height in modern times, stands on the north side. 
 The apse is polygonal. The west end has some large wide but- 
 tresses. The plain and massive character of this church, as seen 
 from the open space on the north side, is extremely striking. The 
 nave is without aisles, and of great breadth. 
 
 We have already noticed the general outline of Bordeaux Cathe- 
 
114 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 dral. The nave, which is very wide, still exhibits some portions 
 of late Romanesque or Transitional work, but at present its features 
 are principally Flamboyant. The choir has the polygonal apse, 
 aisle, and radiating chapels common in cathedrals of the 13th and 
 14th centuries ; it is much loftier than the nave. The clerestory 
 has some curious Flamboyant tracery. The northern spires, though 
 lofty, are not, I think, elegant ; the towers supporting them want 
 height. I rather prefer the south front, which has the towers 
 without spires. The campanile, near the east end, is a square sup- 
 porting an octagon, of good proportions, but not equal in design 
 or workmanship to some of the towers in Rouen, which it rather 
 resembles in composition. S. Michel, a fine Flamboyant church, 
 has a detached campanile near its west end ; this is a tall hexago- 
 nal tower, and forms the principal feature in the town as we ap- 
 proach it by the river. S. Croix has a fine Romanesque west front 
 with a massive square southern tower. S. Seurin has a picturesque 
 outline with two towers, a western one of Romanesque work, and 
 a small later octagon on the south side. The belfry gateway is a 
 grand and striking specimen of castellated architecture. The 
 Roman remains we shall have to notice hereafter. 
 
 A few miles from Bordeaux, to the west of the Angouleme road, 
 is the little village of Bassens, the church of which is a pretty spe- 
 cimen of Romanesque. The tower is central, and does not exceed 
 in height the ridge of the nave roof, that part of the building with 
 its aisles, being Flamboyant, and as usual built with an utter un- 
 concern as to the proportions of the earlier structure. The plan 
 of the tower is oblong from north to south. Each face has two 
 round-headed belfry windows with shafted orders. There are no 
 transepts, but the lower stage of the tower, below the belfry, pro- 
 jects, having a tiled set off, or small sloping roof. This stage has 
 two pairs of tall massive shafts with capitals, running up to the 
 cornice, and having good bases, rather classical. The engaged 
 columns of the tower piers have also a very classical appearance ; 
 the bearing arches of the tower are pointed. The west end has a 
 round-headed door of one square order. This front seems Roman- 
 esque ; it is very plain, but has some brackets at about the level 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 115 
 
 of the spring of the gable. The east end is apsidal, and has some 
 
 engaged columns running 
 up to a bracketed cornice. 
 Above this is added a mo- 
 dern stage pierced with very 
 plain windows or open- 
 ings ; a common addition 
 throughout the southern 
 provinces. The interior is 
 much disfigured and dis- 
 guised ; still this church is 
 worth study in a district not 
 abounding in objects of 
 architectural value. 
 
 About fourteen miles 
 south-east of Bordeaux are 
 the ruins of the abbey of La 
 Sauve, which must have 
 been a large church, chiefly 
 late Komanesque or Tran- 
 sitional. The tower, part of which is of a later date, is engaged 
 on the south side, and gives out- 
 line to a ruin which otherwise 
 would not repay the artist. The 
 apse is semi-circular. It was in 
 this building that I observed the 
 only instance of the cushion capital 
 I remember to have seen in the 
 south. The parish church, at a 
 short distance to the westward, is 
 of much the same date. Its east 
 end is flat, and of good composition. 
 A view of it is given in Guyenne 
 Monumental." 
 
 The bell-gable turret (generally 
 at the west end) is here a common feature. The annexed example 
 
 Bassens, 
 
 La bAuvE. 
 
116 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 is on the road between Bordeaux and La Sauve ; I think the name 
 of the village is Cenac. The church contains Romanesque work, 
 but the belfry is probably late. 
 
 When I start upon a morning's excursion, in perfect ignorance of 
 the churches in a neighbourhood, I choose a direction where I know 
 of a castle, or some other object worth notice, so that I may be sure 
 of meeting something, even if it should not be exactly what I want. 
 And therefore, as the handbook mentions a chateau at La Brede, 
 (about fourteen miles south-west of Bordeaux,) famous as the birth- 
 place or residence of Montesquieu, I went to it in the hope of 
 picking up something in the way. I was pleased at finding there 
 a picturesque and valuable old church, principally Romanesque, 
 though with later additions. It has a polygonal apsidal chancel, 
 transepts with eastern apses, central tower, the upper part of which 
 is late Gothic, a nave and aisles of Flamboyant, and a Romanesque 
 west front. The tower arches are round, of two square orders, the 
 inner one resting on an engaged shaft. The sculpture of the capi- 
 tal is carried round the square parts of the pier as well as the cy- 
 lindrical. The roof of the tower compartment is somewhat domed, 
 and has ribs of a square section. The transept vault and that of 
 the intermediate compartment of the chancel are cylindrical, that 
 of the apse is semi -domical. The massive external shaft, sup- 
 porting strings or cornices, occurs here, as in other examples. If 
 I remember, the apse, though externally polygonal, is internally 
 semi-circular, an arrangement perhaps worth consideration, as 
 giving a kind of invisible buttress at each angle. The west end 
 has a fine specimen of a projecting doorway in three stages, the 
 whole being nearly the height of the gabled front to which it is 
 attached. The lower stage comprises a wide semi-circular arch, 
 corresponding with the main entrance door ; above is a bracketed 
 horizontal string. Over this is a circle between two round-headed 
 shafted arches, and the upper part consists of a gable between two 
 slopes, (like that of the end of a church with clerestory and aisles,) 
 containing a shafted round arch pierced as a window. On each 
 side of this projection, in the main wall of the front, are round 
 arches flanking the principal one : all this is comprehended in 
 the width of the central aisle of the nave, which was in all proba- 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 117 
 
 bility single before the Flamboyant additions. A few Gothic win- 
 dows are inserted in the Romanesque portions of the building. 
 
 Between this and Bordeaux, at some distance from the road, is 
 Martillac, the church of which consists of an apsidal chancel and 
 transepts, also with the eastern apse. There is no nave at present. 
 On the west wall of the intersection is a pierced bell-turret. The 
 style of this church, where it retains its original character, is Ro- 
 manesque. 
 
 I was not so fortunate on my visit to the Chateau of Blanquefort, 
 as I did not find much besides the castle itself. This, however, 
 has the remains of some fine round towers, and will furnish the 
 artist with one or two sketches. It is on low ground. The church 
 is chiefly modern, but has some indications of Romanesque work 
 at its east end. 
 
 Agen. 
 
 At Agen, betw^een Bordeaux and Toulouse, is S. Caprais, now 
 the cathedral, a cross church of great width. It has no central 
 
118 
 
 AECHITECTURAL STUDIES IN FRANCE. 
 
 tower, but one of Flamboyant character, of which at least the up- 
 per part has been very lately built, occupies the angle between the 
 
 nave and south transept. There 
 are no aisles either to the nave, 
 chancel, or transepts. The east 
 end is semi- circular, and has ra- 
 diating chapels like those we usu- 
 ally see round an apsidal aisle. 
 The transepts also have eastern 
 apses. The arches of the inter- 
 section are pointed, with very deep 
 archivolts like those at Perigueux, 
 which rest on massive square piers, 
 ornamented by engaged columns 
 and shafts, and a round blank arch 
 in the face. They are not pierced 
 by arches as in S. Front. The roof 
 of the apse is semi-domical, with 
 ribs, which rest on shafts supported 
 on the capitals of others bearing 
 round blank arches. These divide 
 the apse into compartments, which have alternately a window and 
 a radiating apse, also lighted. 
 
 The nave is cross-vaulted with ribs ; the vaulting compartments 
 are oblong. The width of the nave is about forty feet. The arches 
 of construction are mostly pointed. The central compartment has 
 a cross vaulting. The north transept has a triforial gallery, appa- 
 rently of 1 4th century work ; the arches are subdivided and en- 
 riched with knobs. 
 
 Serignac, near Agen, has a small church without aisles ; the 
 central tower forms an octagon externally, and an octagonal cupola 
 internally. The style is late Romanesque, with pointed arches. 
 
 My journey from Bordeaux to Roquefort, on my way to Pau, 
 was performed chiefly by night, so I cannot tell what this route 
 affords. It passes, however, through Bazas, which has a large and 
 fine church of late Gothic, if I may judge from engravings. 
 
 Agen. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 119 
 
 Roquefort is a picturesque little town, with the remains of a 
 castle. These stand on the precipitous bank of a small and rapid 
 stream, over which is an ancient bridge of one pointed arch, the 
 whole forming a very striking scene. The church is Romanesque, 
 and has an eastern apse, to the north of which is attached a small 
 square tower. 
 
 A pleasant walk of about two miles takes us to the village 
 church of Surbazon, which consists of a nave and aisles, a lofty 
 chancel with a south tower, 
 to which is attached an 
 eastern semi-circular apse. 
 These portions are Roman- 
 esque ; the rest, where not 
 modernized, is Gothic of 
 about the 14th century. 
 The arch of the tower apse 
 is semi-circular and has 
 shafts, the abacus of which 
 is enriched with several 
 rows of billet moulding. 
 Near this church are the 
 remains of a tesselated 
 pavement, broken up by 
 attempts to remove it. 
 
 Pau and its neighbourhood afford but little to the architectural 
 student. In the absence of his favourite subjects he must content 
 himself with the scenery of the Pyrenees. 
 
 Morlaas, however, will please him. This is a Romanesque church 
 with a low central tower, swamped by the Flamboyant additions to 
 the nave. These consist, however, rather in the high-pitched 
 roof and the lofty gable of the west front, than in any alteration of 
 the clerestory itself. The west door is an extremely fine Roman- 
 esque one of several orders. It is much mutilated, but I do not 
 recollect to have seen any in France of greater richness. It must 
 belong to the 12th century. The transepts and chancel, which is 
 apsidal, are pure Romanesque, and exhibit some good work of that 
 
120 
 
 AKCHITECTURAL STUDIES IN FRANCE. 
 
 style I tliink the west end must have had, or been intended 
 to have, a tower. Lescar, near Pau, I will notice in my next chap- 
 ter, which I mean to devote to the consideration of churches of a 
 particular class. 
 
 The cathedral of Tarbes, though much modernized, has some 
 late Romanesque and early Gothic work of excellent character. 
 It is a cross church with apsidal chancel, transepts having the 
 eastern apse, and a massive central octagon. This latter has 
 pointed windows of two lights, and forms a fine lantern to the in- 
 terior. The squinches are of the Romanesque kind. The bearing 
 arches are all pointed, but the choir windows are round-headed, 
 and enriched with numerous mouldings, as in elaborate specimens 
 of our latest Norman. The north transept front has a curious cir- 
 cular window with early tracery. Much of this church is brick. 
 The western part of the nave is of poor modern work. 
 
 Ibos, near Tarbes, is mostly of brick, and of the 14th century 
 or later. The nave is without aisles, but has chapels between the 
 buttresses. The chancel is nearly as high as the western tower. 
 
 Tarhes. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 121 
 
 BoURGES. 
 
 CHAPTER IX. 
 
 If architecture, as applied to structures raised above the surface 
 of the ground, borrowed any hints either from natural caverns, or 
 those formed by art, its first roofs (supposing them to be constructed 
 of small pieces of stone) must have been cylindrical and semi-do- 
 mical. The former, or some approach to it or modification of it, 
 will be the section of the tunnel which an engineer would drive 
 horizontally through a hill for any purpose ; he might perhaps 
 make it a parabola or catenary, or whatever curve he thought 
 most conducive to strength, or ensuring most economy of labour, 
 still it w^ill essentially be the barrel roof which we find cut out of 
 the rock in the very curious under-ground church of S. Emilion 
 on the Dordogne, and built in a great number of the large Roman- 
 esque churches south of the Loire. 
 
 I do not know if the date of the rock church a S. Emilion has 
 
122 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 been ascertained. There are some imperfect sculptures which are 
 considered to be very early, and as one of the impost strings has the 
 Moulure en echiquier," which is common in early Romanesque 
 buildings, I conceive the date of the building itself, or I should 
 say the excavation, to be very remote. For this little portion of 
 moulding gives one the idea that Romanesque architects began 
 upon the decoration of what they already found ; but the stone 
 proving unsuitable, or the effect likely to be produced not commen- 
 surate with the labour, they abandoned the project. The church 
 
 S. Emilion. 
 
 consists of a nave and aisles ; the piers beitig immense square im- 
 posts, left in the solid rock. The roof is cylindrical, or perhaps 
 rather parabolical. The pier arches, which are round, run into 
 the curved surface of the roof, and have consequently a double cur- 
 vature. In the roof, near the vertex of one of the arches, is an 
 aperture leading to the surface, now covered by a fine steeple of 
 Transitional and early Gothic character. Attached to the perpen- 
 dicular face of the rock is also some work of the 14th century, 
 especially a very beautiful porch. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 123 
 
 I shall devote my present chapter to those districts where the 
 barrel roof is principally used in large buildings up to the 12th 
 century, though I may also have to notice some of different con- 
 struction. 
 
 On starting from Orleans in a south-easterly direction, I did not 
 rejoin the Loire till I reached Nevers, so that I missed the fine old 
 church of La Charite, which is well described in M. Merimee's 
 work. By the way I may remark, that his descriptions, though 
 far from lengthy, and unaccompanied by illustrations, give me a 
 clearer idea of the general aspect of a church, than those of almost 
 any writer I have studied. 
 
 But my route took me through Bourges, the cathedral of which 
 is the more impressive on account of its deviations from the 
 usual type. It has no transepts, and consequently its internal 
 length is unbroken throughout. Two aisles run on each side, and 
 round the apsidal choir ; these, as at Milan, are difi*erent in height, 
 but (as is not the case at Milan) the most is made of this difference. 
 Above the enormously lofty pier arches of the nave is a well-deve- 
 loped triforium and a large clerestory. The adjacent aisle, corre- 
 sponding in height to these pier arches, and itself loftier than many 
 of our English cathedrals, has its own pier arches (which corre- 
 spond with the outer aisle), triforium and clerestory. The outer 
 aisle has the usual range of windows. Such of these as are 
 original consist of single pointed lights. The clerestory of the 
 intermediate aisle has a two-light geometrical window, without 
 foliation, but complete tracery. The main clerestory is a window 
 of three unfoliated lights with a foliated circle in the head, and 
 incomplete tracery, that is, unpierced spandrels. The apse is semi- 
 circular, and its windows are smaller. Most of the windows are 
 filled with the finest painted glass. It will be observed in this, 
 as in most French cathedrals which have much early glass, that 
 the lower windows contain subjects in medallions or patterns ; and 
 the clerestories single figures in the lights. There is some very 
 fine cinque-cento glass in such of the windows as have been in- 
 serted at a. late period. 
 
 The lower apsidal aisle has small radiating chapels crowned 
 
124 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 with spires, some of them being supported in a curious manner by 
 brackets, shafts, and masses of masonry. The west front is very 
 grand, and the difference in size between the two western towers 
 gives much variety to the outline. 
 
 A few miles from Bourges, in the direction of Orleans, is the 
 Romanesque church of Mehun . This has a narrow engaged western 
 tower, the lower part of which forms a kind of porch ; its arches 
 are pointed, with square orders. The nave is wide and unvaulted ; 
 it has round-headed windows ornamented externally with shafts. 
 The chancel arch is round, and the chancel has the cylindrical and 
 semi-domical vault. It is apsidal with an aisle, from which branch 
 out apsidal chapels, besides which there is a north apse, forming a 
 kind of transept, but diverging obliquely. The south transept is 
 Flamboyant. The remains of the castle consist of two fine towers, 
 apparently of the 14th century. 
 
 A short drive from Bourges, in the direction of Moulins, takes 
 us within sight of the old abbey church of Pleinpied, a work chiefly 
 belonging to the 12th century. It is a cross church, consisting of 
 nave, aisles, transepts, apsidal chancel, with aisles which terminate 
 in apses. A low square tower crowns the intersection. The tran- 
 septs and choir have the pointed barrel roof ; that of the apse is 
 semi-domical. The compartment of the intersection has an octa- 
 gonal cupola, not pierced for light. The choir is divided from its 
 aisles by two round arches on each side. They have two orders, 
 the outer one enriched with the torus. The south aisle of the 
 nave has a plain cross vault without ribs. The nave is not vaulted, 
 or at least does not show any vault at present. The pier arches are 
 pointed, and have two plain square orders. Two of the piers on 
 the north side are square, (perhaps repaired,) those on the south 
 side are all clustered. The nave has no triforium or clerestory, and 
 has its aisles under the same external roof. The piers of the tower 
 differ very little from those of the nave either in size or composi- 
 tion. The east side of the tower, externally, exhibits pointed 
 arches. 
 
 On the road to Nevers we pass, among other objects of interest, 
 one of those simple village churches, which I suppose are found in 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 125 
 
 most districts where original Romanesque work is permitted to re- 
 main untouched. It consists of a nave, central tower, and eastern 
 apse ; no aisles or transepts. The tower is hut little elevated 
 above the roof of the church ; it is perfectly plain and has but- 
 tresses. It may have been higher ; it is now covered with a pictu- 
 resque slated roof. The western door is pointed, though evidently 
 Romanesque. The windows of the apse are round-headed. 
 
 The cathedral of Nevers has good early pointed work, in which 
 we notice some curious peculiarities, especially the form of the 
 clerestory window. The piers are clustered, the pier arches (which 
 are pointed) have effective mouldings, and there is a good trifo- 
 rium. The apse has its aisle and chapels. The tower, which 
 stands on the south side, is Flamboyant ; to which style also many 
 of the windows belong. The transept is western, and there is a 
 Romanesque western apse. 
 
 But the chief part of my time at Nevers was given up to the very 
 pure Romanesque church of S. Etienne, which, according to Ba- 
 tissier, altogether belongs to at least as early a period as the 11th 
 century. It is a large cross church, with aisles, long transepts, ap- 
 sidal choir, having an aisle and branching chapels, and eastern apses 
 to the transepts. The west front has nearly a square elevation, com- 
 prehending the aisles, and rising somewhat higher than the roof 
 of the nave ; in fact, it is a sort of western tower of very oblong plan. 
 This is a feature to be noticed, as we shall soon see it adopted in 
 rather a remarkable manner. The west door is round-headed, of 
 three orders, with the torus ; all the orders are shafted, the outer 
 one having a very large shaft or column. In the upper part of the 
 front is a triplet of round-headed windows, each under a trefoil arch 
 supported on massive shafts. The outer line of the voussoirs is a 
 semi-circle, the inner one forming the trefoil head. The nave has 
 round pier arches, a triforium of a plain round arch in each bay, 
 including two smaller ones, now walled up, and a clerestory of one 
 round-headed window. The roof is cylindrical, and between the 
 bays has an arch of one square order, resting on a tall shaft or 
 column with a square abacus. The pier arch is of two square 
 orders, the inner one resting on a shaft or column engaged in the 
 
126 
 
 ARCHITECTUEAL STUDIES IN FRANCE. 
 
 side of the square pier, as is the vaulting shaft in its front. Batis- 
 sier gives a view of the transepts from a point near that which I 
 have chosen for mine, which shows the screen of round arches and 
 shafts crossing them ; being in fact a continuation of the external 
 wall of the aisle. This feature also prepares us for what we shall 
 meet with in Auvergne. Over the main crossing is an octagonal 
 cupola on Romanesque squinches, not pierced for light. This is 
 covered by a low octagon on a square base, very similar to those in 
 Lombardy. At the transept ends we observe, both externally and 
 internally, the straight-sided pointed arch (such as in England we 
 attribute to the Saxon period). All the other arches, I think, are 
 round ; those of the apse much stilted. The abacus is square 
 throughout. I did not see any instance of our cushion capital, 
 though some of the capitals are very plain. The piers of the apse 
 much resemble the Tuscan column. This fine old church ought 
 to be carefully studied. 
 
 The road from Nevers to Autun presents some very beautiful 
 scenery. At Chateau-Chinon it crosses a lofty range of hills, 
 commanding an extensive tract of much variety. Beyond this are 
 some fine rocky valleys, and on emerging into the open country 
 we obtain a view of Autun, situated on the slope of a wooded hill, 
 the highest part of the town being occupied by the cathedral. The 
 road is not altogether devoid of architectural interest ; but the 
 only object I shall notice at present is the church of Eoui, a village 
 about twenty miles from Nevers. This building is Romanesque 
 and Transitional ; the upper part of the tower (which is central) 
 is more advanced. The arcade below the belfry stage has a curious 
 sort of ornament consisting of round notches on the edge of the 
 archivolt ; it is very effective. I was much pleased with the gene- 
 ral outline of the church. 
 
 I have already referred to Autun cathedral. It has pier arch, 
 triforium, clerestory, and barrel roof, like the church at Nevers we 
 have described; but the pier arch and vault are pointed. The 
 central piers are not more massive than the others, and yet the 
 later architects have crowned the intersection with a lofty Flam- 
 boyant steeple, which is remarkable from the very deep buttresses 
 
ARCHITECTURAL STUDIES IN FRANCE. 127 
 
 by which it is flanked (two at each angle) ; the projecting cornice 
 which supports the parapet of the tower, and the elegant range of 
 canopies round the base of the spire. Each of these canopies has 
 a sept-foiled arch under a gable of open work. The west porch is 
 
 AUTUN. 
 
 flanked by towers, and is about double their depth. It is cylindri- 
 cally vaulted, with a rib on shafts. The doorway is of three orders, 
 and divided by a pilaster pier. The transom is finely sculptured. 
 The shafts are banded. This porch is the height of the aisles as 
 taken externally. I shall have to refer to the Roman antiquities 
 in a future chapter. 
 
 I have seen so little of Auvergne, and passed so hastily through 
 that picturesque province, that I shall be able to do no more than 
 point out one or two of the specimens with which it abounds. And 
 I will begin with that which, if we take both its ancient features 
 and those for which it is indebted to an admirable restoration, gives 
 almost the typical form of an Auvergnat church — namely, S. Paul 
 of Issoire. The west front, with its square tower, and the light 
 
128 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 octagonal lantern over the centre, built in the Romanesque style, 
 are modern, and in excellent character. I have no doubt there is 
 authority for both ; the central octagon seems to belong to several 
 old churches in the province, of which there are engravings. The 
 rest of the church, though it may have undergone much repair, 
 seems as if it can be depended upon. — The nave, between the west- 
 ern tower and the intersection, has seven bays. It is divided from 
 the aisles by round arches of one order. The piers have engaged 
 in them massive columns, with well sculptured capitals of a very 
 Corinthian character. In some the nucleus is cylindrical. — The 
 roof of the nave is a round barrel vault, with some transverse 
 arches of one order, which do not however occur at every pier. 
 There is a triforium, but no clerestory, the aisles being under the 
 same roof with the nave. The aisles have a cross vaulting without 
 diagonal ribs. — The cupola, at the intersection internally, is of the 
 plan of a square with the angles rounded off. It has the Roman- 
 esque pendentive. The square compartment on which it rests has 
 an open arcade on each side above the main supporting arches. 
 
 We have noticed at Nevers the western tower of an oblong plan, 
 and also the transept screens corresponding with the aisles, but not 
 supporting any external wall. Here, however, the transept screen 
 arch is as high as the tower arches ; and the wall it supports is an 
 external one, pierced by windows : in fact we have a tower of an 
 oblong plan comprehending nave and aisles, placed over the centre 
 instead of the west end. The central part of this probably supported 
 an octagonal tower, not very dissimilar to the present modern one ; 
 the part over the aisle is covered with a sloping roof. It will easily 
 be seen what a scope this arrangement gives for variety of outline. 
 
 The apsidal aisle has four apses, each springing from a gabled 
 face ; this with a peculiar mosaic ornament of dark stone on the 
 lighter building stone, gives great finish to the building. The pat- 
 terns are geometrical, not architectural. — From the central space 
 at the east end and between two apsidal chapels, projects a square 
 one. — There is a crypt beneath the chancel which I did not visit. 
 The plan is given by M. Mallay ; it comprehends the apsidal aisle 
 and its radiating chapels, but gives the westernmost chapels a square 
 
AKCHITECTURAL STUDIES IN FRANCE. 
 
 129 
 
 plan. The roof is supported by piers corresponding with those 
 above, and four additional ones under the floor of the choir. 
 
 The aisle roof above the triforium is quadrantal, and forms an 
 abutment to the barrel roof of the nave. There are one or two 
 instances of the straight sided arch which we noticed at Nevers, 
 and, if I am not mistaken, some trefoil arcades occur in the trifo- 
 rium. 
 
 We see that the main point, in which this church differs from 
 S. Etienne at Nevers, is the absence of the clerestory, in conse- 
 quence of which the aisles are comprehended under the same roof 
 with the nave. The lateral additions to the tower are therefore 
 brought out and shown externally, (the transept walls not being 
 higher than those of the aisles,) but at Nevers they are absorbed in 
 transepts equal in height to the clerestory. At Issoire we observe 
 an arcade between the tower arches and the roof ; at Nevers the 
 arches under the tower are the full height. 
 
 Notre Dame, Clermont Ferrand, might be described almost in 
 the same words with the church we have just examined. It cor- 
 responds nearly in every part, except that it wants the square 
 eastern chapel, and has only five bays to the nave instead of seven. 
 It is rather a smaller church, and not so elaborate in its workman- 
 ship. Here also is the central tower comprehending the aisles, 
 now crowned by a modern octagon of good character ; and a 
 modern tower in the Romanesque style occupies the central com- 
 partment of the west front. The trefoil arcade occurs in the tran- 
 septs, and there is a curious Romanesque south door to the nave. 
 This church is so shut in by houses that it is almost impossible to 
 obtain a view of the exterior. 
 
 Riom, on the road from Clermont Ferrand to Moulins, has a 
 church which though much altered and modernized, retains some of 
 its Auvergnese features. The nave is Romanesque, but has the 
 pointed arch. Like the two last described churches it has the tri- 
 forium, but no clerestory. The central tower comprises the aisles, 
 and is crowned with an octagon and truncated spire of the latest 
 Gothic. The choir is a good specimen of the fourteenth century. 
 
 Brioude, on the road between Issoire and Le Puy, has a fine 
 
 s 
 
130 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 church in which the pointed arch is freely used, indeed part of it 
 appears to belong to the thirteenth century. Like Issoire it has a 
 modern western tower. The central tower does not in this case 
 comprehend the aisles, but is square, surmounted by an octagon. 
 The apse is circular, and has an aisle garnished with five apsidal 
 chapels, each springing from a gabled front. The round arch pre- 
 vails in this part. The main apse is beautifully ornamented with 
 patterns in black and light- coloured stone. The chevron is used 
 in some of the arches of the choir, and the shaft and torus prevail. 
 The piers are lofty, of a square plan, with engaged columns. The 
 triforium seems to have consisted originally of three plain round 
 arches with a circle above. The vaulting of the nave is of early 
 pointed character, and ribbed. The aisles have the cross- vaulting, 
 without ribs. The cupola of the intersection is of square plan, and 
 cross- vaulted with diagonal ribs. The tower arches are pointed. 
 The main apse has a plain semidomical vault, but the eastern 
 apsidal chapel has a ribbed roof; the others plain, and the arches 
 opening into them are pointed. There is a fine south portal with 
 massive engaged columns. 
 
 We have already noticed the cathedral of Puy on account of the 
 curious composition of its roof. The whole building is full of in- 
 terest, and has remarkable portions in styles later than the Roman- 
 esque. Of these we may notice a porch, eastward of the south 
 transept, and much exceeding in size the actual choir, or at least 
 that part of it eastward of the dome, which consists of a mere apse. 
 This porch has a singular kind of detached arch, hanging as it 
 were to an outer order by small bands of stone ; the date I should 
 think about the end of the thirteenth or the fourteenth century. — 
 The campanile, at the east end of the church, is an elegant structure 
 tapering in stages and terminating in a point. Its style is Transi- 
 tional. 
 
 But the most extraordinary object in this romantic town is the 
 chapel of S. Michel, situated on the top of a rocky pinnacle, which 
 springs abruptly from the valley. A steep flight of steps cut in the 
 rock conducts us past some of the buildings or excavations con- 
 nected with the church, to the principal entrance facing to the 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 131 
 
 south-east. This is a rich porch of Romanesque work, of which 
 
 the doorway (a round-headed 
 one) stands beneath a bold foU- 
 ated order, above which is a se- 
 micircular label. The cornice 
 has an arched corbel table, and 
 the whole front is adorned with 
 mosaic work of black, red, and 
 light-coloured stone. The plan 
 of the building is adapted to the 
 area which it occupies. The an- 
 nexed sketch of it, though not 
 very correct, will give an idea. 
 It has a square compartment, 
 forming a low central tower, 
 with a cupola internally, from which branch out northern and 
 eastern apses of small projection. The nave forms a kind of 
 oblique apse with an aisle, from which it is separated by shafts with 
 good capitals ; their pier arches are 
 round. — On the south side is the 
 staircase to the entrance door of the 
 porch, which is below the level of 
 the church. The nave has a cylin- 
 drical roof with semidomical termi- 
 nation. The aisle is cross vaulted. 
 The belfry, which forms rather an 
 elegant steeple, stands to the north 
 west, attached to the aisle, and ranges 
 obliquely as regards the other tower. 
 
 In one of the most beautiful spots in 
 the province of Limousin (see Mur- 
 ray for the description of its scenery,) 
 on the road between Limoges and 
 Toulouse, stands the very picturesque 
 old town of Uzerche, on a rocky hill 
 surrounded by a small river. It pre- Uzbrche. 
 
132 
 
 AECHITECTURAL STUDIES IN FRANCE. 
 
 sents several remains of military and domestic architecture of medi- 
 aeval date, and its highest point is crowned by a fine Romanesque or 
 Transitional church. This has a nave and aisles, transepts, and 
 apsidal choir with aisle, from which probably radiated a series of 
 chapels, now mostly destroyed. Underneath is a curious crypt, 
 which I did not visit. At the intersection of the cross is a low 
 square tower, its compartment internally being occupied by the 
 
 UZERCHE. 
 
 usual cupola. Another loftier tower, of good composition, stands 
 over the nave, at some little distance from the west end, and 
 having also an internal octagonal cupola. The part to the west- 
 ward of this differs in some few respects from the rest of the nave, 
 but does not appear to be the addition of a later period. The 
 work throughout is very plain ; the arches are mostly pointed, 
 with square orders. There is no triforium or clerestory. The 
 
UZERCHE. 
 
 Page \:V2. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 133 
 
 nave has the barrel roof, with a pointed ridge ; that of the aisles 
 forms an abutting semi- arch. The roof of the apse is semi- 
 domical. I cannot help thinking that an artist might very well 
 make this little town his head quarters for a few weeks in the 
 summer. 
 
 St. Sernin, Toulouse. 
 
 S. Sernin, Toulouse. This is one of the finest and largest Ro- 
 manesque churches in the south of France. It consists of a long 
 nave with two aisles on each side, north and south transepts of 
 
134 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 great length, with eastern and western aisles, and two apsidal 
 chapels to each transept looking eastward, a choir with polygonal 
 apse, aisle, and radiating apsidal chapels. At the west is a kind 
 of transept, about the depth of a bay of the nave, but not project- 
 ing beyond the walls of the external aisles. In the centre is a 
 lofty octagonal tower, tapering in five stages. The three lower 
 
 Sr. Seunim, Toulousk. 
 
 ones are Romanesque, (in style, whatever may be their date,) each 
 face has two windows, of two square orders and a label. The 
 outer order is shafted, a single shaft, in the middle, being common 
 to both windows ; consequently each face has three shafts. . At 
 the angle is also a shaft with a square abacus, projecting fi'om the 
 cornice of the stage, after the manner of revived Italian work. — 
 The two upper stages have two straight sided pointed windows also 
 
AKCHITECTURAL STUDIES IN FRANCE. 
 
 135 
 
 of two orders, the outer one shafted, as before, and under a single 
 angular label. The angles also have shafts, as in the lower stages. 
 A parapet of plain round open arches, 
 with a small turret at each angle, crowns 
 the upper stage, and a plain ribbed spire 
 surmounts the whole. It is chiefly con- 
 structed of brick covered with plaster ; 
 indeed there is a great deal of brick- 
 work throughout the church. 
 
 The nave has the cylindrical vault, 
 with transverse arches of one square or- 
 der resting on engaged vaulting shafts. 
 The triforium consists of a large round 
 arch, comprising two smaller ones di- 
 vided by a doubled shaft, the archivolt 
 being of some depth. In some of the 
 transept arches the archivolt of this part 
 has a kind of bold fret, or reticulation, 
 by way of ornament. The triforium 
 forms a gallery, lighted by a round 
 headed window. Its roof is quadrantal. 
 There is no clerestory. The pier arches 
 have two square orders on imposts of the same section, having a string 
 or abacus at the top. The aisle adjoining the nave, or, as I shall 
 call it, the intermediate aisle, is higher than the outer one, and has a 
 round arch in the clerestory space. The external aisle has a round 
 headed window in each bay. Both aisles are cross vaulted without 
 ribs. The intermediate one is very Roman in its character, exhi- 
 biting only square-edged imposts, with an abacus of rather classical 
 profile. The ends of the transepts have pier arches and triforium, 
 so that the arcade is continued round the whole church. The 
 choir has a clerestory. The tower arches are round, and have 
 three square orders. Their piers, I think, must have been cased ; 
 they are now octagonal, and fill up one bay of the nave. The 
 central cupola is octagonal with ribs. The choir aisle has five 
 radiating chapels. The western and transept fronts have each a 
 
 Toulouse. 
 
136 AECHITECTURAL STUDIES IN FRANCE. 
 
 pair of fine Komanesque doorways, enriched with shafts, sculptured 
 capitals, the torus and billet. Many of the capitals are extremely 
 rich ; some with groups of figures, others with foliage after the 
 manner of the Corinthian capital. The round arch prevails 
 throughout, except the few later insertions. 
 
 There is some good Gothic brickwork in Toulouse ; the straight 
 sided pointed arch, that we have noticed in the upper part of the 
 steeple of S. Sernin, is frequent. The cloisters of the church now 
 used as the museum are very beautiful. A double shaft, set trans- 
 versely, with an oblong square abacus on a rich capital, supports 
 arches which have a double foliation ; they must belong to the early 
 part of the fourteenth century. 
 
 S. Gaudens, separated from the first ranges of the Pyrenees only 
 
 S. Gaudeni 
 
 by the valley of the Garonne, has an oblong Romanesque church 
 with three eastern apsides, and a low western tower. The nave 
 
S. Gaudens. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 137 
 
 has aisles, and a cylindrical roof, against which abuts a quadrantal 
 one. The chancel has a triforium, the aisle below being cross 
 vaulted, without diagonal ribs. The eastern face shews the section 
 of the cylindrical and quadrantal roofs ; the space between the 
 
 S. Gaudens. 
 
 former and the roof of the apse is occupied by three round-headed 
 windows. — The round arch prevails, and there are some capitals of 
 great elegance ; one especially, in which the attitude of the figures 
 gives them, at a distance, almost the appearance of Corinthian 
 foliage. 
 
 S. Pe, in the Pyrenean valley watered by the brook which runs 
 through Pau, has a Romanesque church with a massive tower at the 
 south west angle. It is a good deal modernised and disfigured, but 
 the south aisle has, from the absence of shafts, a very Roman cha- 
 racter. 
 
 Three or four miles from Pau, on the Bayonne road, is Lescar. 
 Its church, though a large one, makes no great show externally ; as 
 the same roof covers nave and aisles, and there is no tower, a small 
 
 T 
 
138 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 wooden belfry marking the intersection of the transepts. Internally 
 the church presents fine Romanesque work of an early period. 
 The roof of the nave is cylindrical, and that of the aisles consists 
 of a series of transverse cylindrical vaults, their partitions resting 
 upon a round arch ; in fact, the roof is similar to that of the nave 
 at Tournus on the Saone. The four shafts therefore that surround 
 each pier are of three different heights— the vaulting shaft of the 
 nave being the highest ; (this supports, as usual, a transverse arch 
 of square section ;) next, the shafts supporting the archivoUs of 
 the pier arches, and lower than these are the shafts sustaining the 
 transverse arch of the aisle. The pier rests on a cylindrical base. 
 The east end of the church is triapsal. 
 
 Many of the churches I have noticed in the present chapter, be- 
 sides many more of similar character, have been described by M. 
 Merimee and other antiquaries ; in fact it is a style with which the 
 English architect may easily make himself acquainted through the 
 medium of published works and engravings. Auvergne and Velay 
 have been well illustrated, both as regards their antiquarian and 
 picturesque attractions. The same class of buildings, as I have re- 
 marked, occupies a large portion of the south of France ; I suppose 
 the whole of Languedoc, Provence, Dauphiny, &c. The church of 
 Ainay at Lyons, the cathedrals of Valence, Avignon, Aries, Mar- 
 seilles, and many other large churches, belong to it. In the absence 
 of the clerestory, it assimilates with the Lombard Romanesque. The 
 omission of this member is what we might naturally expect where 
 the dimension of height is not considered the one of greatest im- 
 portance, and where strong and abundant light is not required. 
 And we must remember that the barrel roof itself occupies much 
 of the space filled up by the northern clerestory, which is often 
 confined to the mere transverse cells, the springs of the vaulting 
 arch falling as low as the foot of the clerestory windows. This 
 is especially the case in England. At Lincoln the vaulting spring 
 falls as low as a point in the triforium range. 
 
 The omission of the clerestory is, I think, no defect in a Roman- 
 esque church, whatever it might be in a Gothic one. The play of 
 light produced by an upper range of windows is not so necessary in 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 139 
 
 this style, and the division of the height into only two stages, if these 
 be well-proportioned to each other, is always satisfactory. The naves 
 of Clermont, Issoire, and Toulouse have much grandeur, and 
 should be studied by any architect who has adopted the revived 
 Italian. The double tier seems more suitable to this style than 
 the triple ; and the triforium, w hich, if necessary, may be lighted by 
 a window, gives more character than a plain clerestory. I ought 
 to have mentioned that where the quadrantal roof occurs in the 
 aisle or triforium, it is generally crossed at each bay by a partition 
 pierced with a complete arch. 
 
 PuY DE Dome. 
 
140 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 Caen. 
 
 CHAPTER X. 
 
 I SHALL in my present chapter make a few cursory remarks on 
 some leading features in the churches of Normandy, for the sake 
 of noticing the points in which they resemble or differ from our 
 own churches, as well as the churches of other districts in France. 
 
 We will suppose ourselves on a route from Paris to Rouen along 
 the Seine, (the churches on which river, before we enter Normandy, 
 are worthy of notice,) thence to Lisieux, Caen, Bayeux, Coutances, 
 and Avranches ; making such divergences from the several points 
 as may show us the prevailing character of each district. Though 
 the route I have given will introduce us to but a small part of this 
 
English Mouldings, Lincoln Cathedral. 
 
 Page 140. 
 
English Mouldings, Piek-arch 
 OF Choir, Southwell Minster. 
 
English Mouldings, Lincoln Caihedral. 
 
ARCHITECTUKAL STUDIES IN FRANCE. 
 
 141 
 
 interesting province, we shall find it sufficient both to mark decided 
 local peculiarities, and to afford us very considerable variety in 
 style and design. 
 
 That I may dispose of one part of my subject at once, and in 
 such a manner as to give a comprehensive view of it with the least 
 difficulty, I will draw roughly a series of mouldings,^ taking the 
 examples much in the order they occur in the above route ; and I 
 will also annex the sections of a few English mouldings, which 
 have been more carefully drawn by some of my friends ; indeed it 
 will be seen that their greater complexity demands the skill of a 
 practised artist. 
 
 In the south the Romanesque or Transitional arch, whether round 
 or pointed, is generally composed of square orders, which are oc- 
 casionally varied by a torus on the edge, but seldom have any 
 other continuous moulding, though such Romanesque ornaments as 
 the billet, chevron, beak head, and the like, are not uncommon. I 
 think even the hollow, which so constantly occurs in the face of the 
 arch in northern examples, is not very frequent in the south. As 
 the style advanced, the system of mouldings became still more 
 simple and uniform, the arches very generally consisting of square 
 orders, for the most part two, with a torus at the edge, strongly 
 marked by a bold concave cutting. This prevailed through the 
 
 * Reference to the sketches of French Mouldings: — 1. Vernouillet, S. Door, late 
 Romanesque. 2. Ditto, Tower Arch, same date. 3. Triel, Tower Pier, about 14th 
 cent. 4. Lisieux, Door of South Aisle, 13th cent. 5. Ditto, Pier Arch of Nave, 
 13th cent. 6. Thaon, Tower Windows, Romanesque. 7. Ouistreham, Pier Arch, 
 Romanesque. 8. S. Etienne le vieux, Caen, 14th cent, or later. 9. Rots, Transitional. 
 10. Norrey, late 13th or 14th cent. 11. Verson, Transitional, or early 13th cent. 
 12. Carpiquet, 14th cent. . 13. Ditto, Romanesque. 14. Bernieres, Romanesque. 
 15. Ditto, Transitional. 16. Ditto, Vaulting Abacus, of both dates, 17. Langrune, 
 from 12th to 14th cent. 18. Hubert Folic, 14th cent. 19. Bayeux Cathedral, 13th 
 cent. (?) 20. Coutances, South Porch, 13th cent. (?) 21. Mont S. Michel, Cloisters. 
 13th cent. 22. Pontorson, Window, 12th cent. 23. S. Quentin, West Door, 12th 
 cent. 24. Ditto, West Porch, 14th cent. (?) 25. Montviron, West Door, 14th cent.(?) 
 26. Grande Couronne, Window, 14th cent. 27. Rouen Cathedral, Pier Arch, 13th 
 cent. 28. Ditto, Nave Pier, 13th cent. 29. S. Vivien, Rouen, Jamb of Window, 
 14th cent. 30. Jumieges, Chapel, 14th cent. 31. Bocherville, 12th cent. 32. S. 
 Ouen, Rouen, Pier, 14th cent. 
 
142 
 
 ARCHITECTUKAL STUDIES IN FRANCE. 
 
 whole of the thirteenth century, and was hardly discontinued in the 
 fourteenth ; the abacus, with not many exceptions, being square. 
 
 Now as in leaving Paris we travel northward along the Seine, we 
 very soon observe the inner order of a Romanesque or Transitional 
 church to consist of a large torus, having for its section a semicircle. 
 Other mouldings also appear, which are equally calculated to in- 
 fluence the succeeding style, and to lay the foundation of a variety 
 of section approaching what we see in the early English. Accord- 
 ingly at Rouen and Lisieux we find the uniformity of the French 
 moulding broken in upon by the introduction of another member, a 
 kind of torus between two hollows, which however is scarcely sufli- 
 cient to detract from the severity of the general design. The 
 square abacus, or one with the angles taken off, still prevails ; but 
 as we proceed westward the round or octagonal abacus occasionally 
 presents itself, very commonly in the same clustered pier with the 
 square. At Langrune, a church we shall presently notice, the 
 round abacus comes in rather curiously, for it occurs in a part of 
 the triforial arcade which has every appearance of being earlier 
 than other parts where the square abacus is used. In a door at 
 Lisieux, and again at Coutances, and in the cloisters of Mont S. 
 Michel, is a very effective moulding, which I do not remember to 
 have noticed in England, though it may possibly be found. It is a 
 large torus from which a deep bold hollow is scooped, so forming 
 two very sharp edges, having an efifect similar to the keeled torus so 
 common among us. We often find a certain complexity of mould- 
 ing combined with a general squareness of section, and a square 
 abacus ; but on the whole, as we proceed westward, the round or 
 octagonal abacus is more freely used, the squareness is softened off, 
 and the mouldings assume the English character to a greater degree ; 
 though I believe they will never be found so delicate or elaborate 
 as in the finest early English examples. In this branch English 
 architects have decidedl}^ taken a line of their own. No buildings 
 of the thirteenth century can compete with ours, in the richness, 
 variety, purity, and careful execution, of arch mouldings. 
 
 On leaving Paris we pass Reuil, the church of which, in great 
 part rebuilt in the cinque-cento style, retains an elegant Roman- 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 143 
 
 esque central octagon, a feature which we shall lose altogether, or 
 meet with very rarely, when we have entered Normandy. — As an 
 example of the Renaissance, the body of the church is well deserv- 
 ing of notice, though it will probably be condemned by the ad- 
 mirer of Gothic architecture. 
 
 Bougival, on the Seine, a few miles from Versailles, has a tower 
 of Romanesque detail, supported by arches which appear to belong 
 to the thirteenth century. This is by no means a singular in- 
 stance. 
 
 Sartrouville has also a central octagon of Romanesque, with a 
 tall stone spire of later date. Its nave piers are cylindrical, not 
 very massive ; the arches are pointed, of one plain square order ; 
 the style is Transitional. 
 
 Herblay has also Transitional work of similar character in the 
 nave. It has a central tower of early pointed — the belfry window 
 double, with rich shafted jambs. The shafted buttresses at the 
 angles add to the character of this tower. There is some good 
 glass, very rich in colour, in the chancel. The date of this would 
 be in the sixteenth century. 
 
 Acheres has a Romanesque central octagon and stone spire ; it 
 has an eastern semicircular apse. 
 
 The church of Poissy, which has lately been well restored, has a 
 picturesque and imposing outline. The central tower is octagonal, 
 crowned with a w^ooden spire. Besides this it has a western octa- 
 gon, on a square base, and surmounted by a stone spire. The east 
 end is apsidal, with an aisle, which is also flanked by two semicir- 
 cular apsides. The transepts are absorbed in the nave aisles. The 
 building generally is late Romanesque, with additions of the Flam- 
 boyant style. 
 
 Triel has a church which appears to belong principally to the 
 fourteenth century, much of the chancel being rebuilt in debased 
 Gothic. It is a cross church with a plain central tower, apparently 
 of late date. The nave has aisles. The arrangement of the bay 
 is remarkably good, having the pier arch, triforial arcade, (of t;refoil 
 arches,) and a circle by way of clerestory window. The piers are 
 cylindrical, not very massive, with octagonal abacus, or rather 
 
144 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 square, chamfered off at the edges. 
 The octagonal abacus is also used 
 in the capitals of the smaller 
 shafts. The whole is vaulted. 
 The tower arches have continu- 
 ous imposts, without capitals, 
 though there is one to the shaft 
 of the diagonal rib in the tower 
 compartment. This church stands 
 well, and has a striking appear- 
 ance as approached by the river. 
 It contains some good late paint- 
 ed glass. 
 
 Vernouillet has a steeple of 
 very beautiful outline : it rises 
 from a short central tower of Ro- 
 manesque work, and itself ex- 
 hibits purely Romanesque detail, 
 though in the proportions of the 
 most finished Gothic. The spire- 
 lights are very lofty, their gables 
 resting on shafts ; there is also 
 an arrangement of shafts at the 
 spring of the diagonal sides of 
 the spire, probably the remains 
 of lofty pinnacles. The number 
 of open arches at the lower part 
 of the spire gives it a remark- 
 ably light, lantern-like appear- 
 ance ; and, as is the case with 
 most of the stone spires, the masonry is enriched with a sort of 
 scale or net-work. The church is principally late Romanesque 
 and early pointed. The large inferior order of semicircular section 
 appears ; and a round-headed south door has some curious mould- 
 ings. The nave is evidently shortened, or left incomplete. The 
 east end is flat. 
 
 Triel. 
 
ARCHITECTUEAL STUDIES IN FRANCE. 
 
 145 
 
 Verneuil has a central tower, Romanesque, with a plahi late ad- 
 dition, finished with two gables. Internally the arrangement is 
 very similar to the last chmxh we have mentioned. 
 
 Vaux has an elegant semicircular apse, in the early pointed style. 
 Its tower and transepts appear to be Romanesque ; its nave, which 
 has aisles but no clerestory, may belong to the fourteenth century. 
 
 Epone has a fine Romanesque tower, terminating in an octagon 
 and stone spire, on the north side of the chancel. The nave is of 
 great width, and has a timber roof. The chancel is vaulted, and 
 of thirteenth century work. There is a fine round-headed north 
 door. 
 
 Mantes was formerly conspicuous by its two western towers ; one 
 of them has been taken down. 
 
 Vernon has a central tower almost swallowed up by a large 
 Flamboyant nave. The eastern apse is semicircular. 
 
 ROMILLY. 
 
 Ilomilly has some curious construction under its central tower ; 
 the arches, which are very plain, rest on brackets of a considerable 
 projection. 
 
 u 
 
146 
 
 AKCHITECTURAL STUDIES IN FRANCE. 
 
 Rouen abounds more with the later than the earlier styles. The 
 crypt, however, of S. Gervais is said to be of very remote antiquity, 
 but has not any decided architectural indication of date, beyond 
 simplicity and rough workmanship. 
 
 Some Romanesque work appears in the north-west tower of the 
 cathedral ; but the greatest part of that magnificent structure be- 
 longs to the thirteenth and fourteenth centuries. I should say that 
 previously to the fifteenth, the central tower was the prominent 
 feature, and the north-western one formed almost a detached cam- 
 panile ranging with the front. During the fifteenth and at the 
 beginning of the following century both these towers were raised 
 with Flamboyant work, and the south-west tower erected, more 
 lofty, in point of masonry, than either of the others ; and the west 
 front encrusted with rich Flamboyant. The windows of the towers 
 flanking the transepts seem of earlier character than the fronts 
 themselves, but this may be the result of adaptation. It should be 
 noticed that in the beautiful south front, which may be considered 
 as a transition from Geometrical to Flamboyant, the waved line, 
 having a point of contrary flexure, (which in England appears at 
 an early stage of the Decorated,) is not to be found ; though in the 
 window of S. Vivien, in the same city, is a perfect flowing Deco- 
 rated window, connected with that style not only by the forms of 
 its tracery, but by its mouldings. In the Lady chapel of the Ca- 
 thedral, as in the eastern parts of S. Ouen, geometrical tracery 
 alone is found ; in the latter, not very graceful in its composition. 
 
 I believe that in every dimension usually measured to give the 
 comparative size of churches, S. Ouen exceeds the cathedral, and 
 yet the latter appears decidedly the principal building of the place. 
 This is owing to the great spread of the west front, and the dis- 
 tance between the flanking towers, combined with their height. 
 Undoubtedly a spire on the central tower was essential to the 
 whole composition, but I cannot cease to regret that a cast iron one 
 was ever thought of. The wooden steeple of Cardinal Amboise, 
 which tapered in successive stages, must, notwithstanding incon- 
 gruities of style, have been very grand in its outline ; and some- 
 thing of similar design, whatever might be the material, would 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 147 
 
 have been the proper termination. The following strikes me as 
 the reason why the present structure is so unpleasing to the eye. 
 It consists mainly of a certain number of long iron bars, of appa- 
 rently equal thickness throughout, (if they diminish towards the top, 
 they do it imperceptibly,) the lower ends of which rest on the tower, 
 the upper being gathered together at the highest point. The 
 light tracery which fills the openings, when seen from any distance, 
 can hardly be taken into account in considering the substantial 
 structure. Consequently, every horizontal section of the spire 
 appears to be accurately, or nearly, equal ; there is no diminution 
 in weight as we ascend ; a foot taken vertically, near the bottom, 
 would be little heavier than a foot so taken at the very top ; the 
 only difference being in the connecting tracery, which, as I said, 
 the eye scarcely takes into account. Now, in an ordinary stone 
 spire, even if the walls themselves are continued of the same thick- 
 ness, there is a manifest difference between the sections, or the 
 weights of portions measured on equal vertical lines, near the 
 bottom and the top, and this even when the spire is so much 
 pierced as to become a mass of open-work, as in the cathedral of 
 Freyburg in Germany, where the masonry evidently becomes 
 lighter in proportion as it ascends. In fact, in the iron steeple of 
 Rouen, this increasing lightness, the distinctive feature of a spire, 
 is altogether lost ; nothing would give it, except substituting a 
 pierced plating of metal for the light tracery. 
 
 To give a base to this unsightly structure, much of the thickness 
 of the wall in the upper stage of the tower has been cut away ; and 
 the lower part has shown symptoms of insecurity from the weight 
 imposed upon it. 
 
 Of the new west front of S. Ouen I am not disposed to speak 
 harshly, as I learn* that the architect wished to build it on a de- 
 sign more conformable with the original, which is preserved in the 
 
 * " L'Art et TArcheologie au XIX^ Siecle — Achevement de Saint Ouen De Rouen 
 Paris. Librairie Archeologique de Victor Didron, Rue Hautefeuille, 13. 1851.'* 
 (pp. 64). A spirited translation of this able and instructive essay would be ex- 
 tremely useful to many of our Architectural students. 
 
148 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 library. In fact, there are two drawings, one, imperfect in per- 
 spective, but of good late Flamboyant architecture ; another, more 
 correctly drawn, but more debased in style. From the two a very 
 satisfactory composition could unquestionably have been designed ; 
 one harmonising with the rest of the building, and carrying out 
 the original conception of the architect. This would have con- 
 sisted of the front of the nave (already existing with the exception 
 of the crowning gable) flanked by square towers set diagonally, the 
 lower part of which, to about the height of the aisles, had been com- 
 pleted during the fifteenth or following century. These towers 
 would have been crowned with octagonal lanterns, similar to that 
 of the central tower, which would have been still left pre-eminent. 
 The space between the projecting angles of the western towers would 
 have been filled up by a porch. Had the architect entrusted with 
 the work carried out this design, he would have had the honour of 
 putting the finishing touch to a most beautiful and unique building. 
 And we cannot fairly blame him for the defects of a design, which 
 he was compelled to form upon the abandonment of that which his 
 correct judgment originally suggested. Still we need not abstain 
 from criticism. The projecting bases of the towers have been 
 swept away, and those which are substituted made to range with 
 the front in the usual manner. The lower stage of the front has 
 the usual arrangement of three deeply recessed doors, much en- 
 riched ; these are good both in proportion and detail, indeed the 
 whole of the workmanship is excellent. The flanking towers be- 
 come octagonal before they have cleared the height of the nave, 
 and are finished with spires which are about equal in height with 
 the central tower. Now this is a decided error. The front should 
 either have been kept subordinate to the central tower, by being fur- 
 nished with lower towers, having no spires, as in the original design, 
 or by forming a flat mass, relieved by turrets and pinnacles ; or 
 else the steeples should have been carried boldly up, out-topping, 
 as at Coutances, every part of the building. The present steeples, 
 had they been carried a stage or two higher, would have given the 
 front a noble and beautiful proportion. As it is, it appears cur- 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 149 
 
 tailed in its height in deference to the central tower, the effect of 
 which however is in reality more diminished by western spires 
 exactly equalling it in height, than it would have been had they 
 far exceeded it — a spire, if it be anything beyond a mere pinnacle, 
 should always exceed in height any tower in the same building. 
 
 We are naturally led to compare the north-western tower of the 
 cathedral with the central tower of S. Ouen ; both being finished 
 with that beautiful addition, the octagonal lantern. I am not going 
 to give the preference to either, but simply to observe that each is 
 admirably adapted to its position. The Tour de Beurre, rising 
 directly from the ground, tapers upwards in stages, and has its 
 deep and massive buttresses in the sides as well as at the angles, 
 forming a good abutment for the numerous flying buttresses which 
 surround the octagon, and which answer a threefold purpose, to 
 enrich it, to strengthen it, and to render the junction with the 
 square part of the structure pleasing and harmonious. The cardi- 
 nal sides of the octagon fall somewhat within those of the tower, 
 which enables each angle to throw out, and each pinnacle of the 
 tower to receive, two buttresses. This arrangement gives a won- 
 derful play of light and shade, and an appearance of great depth 
 and strength. The octagon at S. Ouen is larger in proportion to 
 the square tower, and simpler in construction, throwing out its 
 flying buttresses only to the turrets at the angles. The size of the 
 windows also gives it an air of lightness suitable to the superstructure 
 of a tower limited in the depth, massiveness, and number of its 
 buttresses, as must be the case with a tower supported in the centre 
 of a wide and lofty church. 
 
 The church of S. Laurent has a tower (rising from the ground) 
 finished with a very curious lantern, being in fact a mass of but- 
 tresses most ingeniously grouped together. Its outline is very 
 striking. 
 
 S. Maclou is a fine specimen of pure Flamboyant, having most 
 of the beauties of the style, and few of its faults. The pier arches 
 have continuous imposts without capitals. The central tower has 
 had a lofty open spire of woodwork. 
 
150 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 Bocherville and Jumieges are 
 so well known, that it is needless 
 for me to offer any remarks upon 
 them. The plainness with which 
 the latter is worked gives it a 
 character somewhat different 
 from the generality of Roman- 
 esque churches in Normandy. 
 
 A very small portion now re- 
 mains of the once beautiful abbey 
 of S. Wandrille near Caudebec ; 
 it must have been a fine speci- 
 men of the early pointed of the 
 thirteenth century. The parish 
 church is partly Romanesque, 
 and has a central tower. The 
 little chapel of S. Saturnin is a 
 very curious relic of the round 
 arched style. Some antiquaries 
 have given it a very early date, others, with more probability, 
 consider it to belong to the eleventh century. It is a perfect 
 cross church of the transverse triapsal form, and has a central 
 tower. The semicircular apses spring directly from the tower, 
 and internally have a semidomical roof. The masonry is rough, 
 some having a herring-bone appearance. The windows, where 
 they have not been altered, are narrow and round-headed. The 
 nave is without aisles. The whole interior is perfectly plain. It 
 is somewhat difficult to obtain a view of this building on account of 
 the trees which surround it. The whole width, taken at the tran- 
 sept, and including the thickness of the walls, is not above fourteen 
 feet.* 
 
 Jumieges. 
 
 * Histoire de I'abbaye royale de Jumieges. C. A. Deshayes. Rouen — F. Baudry, 
 Imprimeur du Roi, Rue des Carmes, No. 20. 1829. This work also contains an ac- 
 count of S. George Bocherville, and the abbey of S. Wandrille, with views of the 
 latter as it appeared before its destruction. The chapel of S. Saturnin is described 
 and figured. 
 
Abbaye adx Hommes. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 151 
 
 I had not an opportunity of taking any notes between Rouen 
 and Lisieux. Some of the churches seemed as if they would repay 
 examination, and at a short distance from Brionne is visible a hand- 
 some tower, the remains of the abbey of Bee. 
 
 Lisieux, with its three towers, exhibits a striking outline. The 
 two western ones are dissimilar in detail, the northern having tall 
 pointed windows, while the southern has in their place three tiers 
 of round-headed windows, and is finished with a spire. Never- 
 theless it does not appear at all certain that the two differ very 
 much in date. The central tower is low and massive, and forms a 
 lantern to the interior. The arches are mostly pointed throughout. 
 The piers are short cylindrical columns ; their abacus, a square, 
 chamfered at the angles. The same style, that of the early part of 
 the thirteenth century, is preserved with much uniformity, and has 
 been little altered or interrupted. 
 
 The vaulting of the nave in the Abbaye aux Hommes, at Caen, 
 has been considered to belong to a later period than the rest of the 
 work. I regret that I had not an opportunity of visiting the build- 
 ing in company with any one who could have pointed out to me 
 the grounds of such supposition. I thought I perceived a slight 
 difference of tint in the stone, visible through the whitewash in the 
 upper part of the vaulting shafts, but I could discover no break of 
 design in the composition. The clerestory, which offers a very 
 uncommon arrangement, two unequal arches in each bay, is evi- 
 dently adapted to such a roof as the present ; and the cellular 
 arches are round, not pointed, as in the case of most of the German 
 sexpartite roofs. This vault should be studied in connection with 
 that of the Abbaye aux Dames, which is in square compartments 
 comprising two bays, but bisected by an intermediate transverse 
 arch, a composition which might either suggest, or be suggested by, 
 the sexpartite vault, to which it bears some affinity. I am rather 
 inclined to think that it is derived from the sexpartite vault, as it 
 gives nearly the same effect, at a less expense of mechanical design 
 and contrivance. In Creuilly Church, about twelve miles from Caen, 
 the Romanesque nave has the sexpartite, and a bay of the chancel, 
 evidently later, the bisected vault. Of the latter the neighbouring 
 
152 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 priory of S. Gabriel, which is carefully illustrated in the Statistique 
 Monumental, and the church of Bernieres, afford fine examples. 
 
 We will now pass to a later period, and examine some of those 
 beautiful steeples, which, built upon one general principle, seem to 
 have been dispersed through a considerable district round Caen ; 
 though not a great number remain without some mutilation, and 
 several are still incomplete. Of these, the church of S. Pierre 
 offers the best known example ; this is evidently of the fourteenth 
 century, but some (that, for instance, of Bernieres) have charac- 
 teristics of the thirteenth. We may thus give a general description 
 of these steeples. The tower, which is square, whether central or 
 rising from the ground, has, resting upon a lower stage of less or- 
 nament, a tall belfry story, also square, without buttresses, or at 
 least any projecting beyond the slope which finishes the cornice of 
 the stage beneath. This belfry has four lofty and deeply moulded 
 arches in each face, of which the outer ones are narrower than the 
 others, and unpierced ; the two in the middle being open as win- 
 dows. These are often divided by a mullion, and sometimes have 
 small plain transoms, without arch or foliation. Above is a rich 
 cornice. From the tower rises an octagonal spire, flanked by four 
 lofty pinnacles of open work, which vary in their plan, some being 
 hexagonal, others octagonal, but are alwa3'S finished with spires. 
 On the cardinal sides are spire-lights, rising to the same height 
 with the pinnacles, and often finished at the top with a quadrangular 
 pyramid. The spire in many cases is pierced with foliated open- 
 ings, such as might be described in a circle, the number of cusps 
 decreasing according to the size of the aperture ; the lowest range 
 perhaps consisting of septfoils, and the highest of trefoils. The 
 masonry is also, as usual, worked in scales. The proportions of all 
 are graceful, though some have a more massive character than others. 
 S. Pierre has pinnacles of an hexagonal form, but not quite equi- 
 lateral, in fact one of the angles very nearly coincides with a right 
 angle. Notre Dame, also in Caen, has the hexagonal pinnacle, but 
 set on differently, the face corresponding with the angle of the tower. 
 But the spire with its lights and pinnacles was often added to an 
 earlier tower. This was done with the western towers of the Abbaye 
 
AKCHITECTURAL STUDIES IN FRANCE. 
 
 153 
 
 aux Homines. The details of the northern one are assimilated in 
 some respects to the Komanesque work below, and may in fact belong 
 to the end of the twelfth, or an early year of the following century. 
 The pinnacles are octagonal. The plan of this spire has a some- 
 what oblong form from east to west, and it is not quite equilateral. 
 The southern spire is evidently later, and is remarkable from the 
 plan of the pinnacles, which forms an equilateral triangle ; their 
 effect as regards outline, as well as light and shade, is extremely 
 striking. The spire-lights, as usual, are lofty, and exhibit much 
 lightness of construction. 
 
 The English architect ought to study this class of steeples care- 
 fully. He cannot fail to remark the beauty which arises from the 
 fine cluster round the spring of the spire, as well as the simplicity 
 of outline above, there being no spire-lights besides the principal 
 ones at the base. Our own finest steeples, Salisbury, and S. Mary's, 
 Oxford, resemble them in these respects. He will also observe, 
 that even the truncated spires are not without beauty. That of 
 Norrey, in its present imperfect condition, is a very fine object ; as 
 also the incomplete steeple of Audrieu. In some cases the tower, 
 evidently intended for a spire, has been finished with two gables, 
 as at Tilly sur Seniles. The towers of S. Pierre and Notre Dame 
 in Caen, have pierced parapets, but these in no way affect the gene- 
 ral aspect of the composition. In these examples the effect of the 
 spire is improved by a low octagonal base from which it springs. 
 
 For an account of the churches in general round Caen, I shall 
 refer the reader to the work of M. de Caumont I have so often 
 mentioned ; many of them are also made familiar to us by Mr. 
 Cotman's etchings. I will content myself with noticing a very few 
 of them at present. 
 
 Airan, near the road to Lisieux, attracted my attention by 
 a massive central tower. This however does not appear earlier 
 than the fourteenth or fifteenth century. The chancel is a fine 
 specimen of early pointed — the vaulting abacus is round, but that 
 of the shafts which ornament the lancet windows externally, is 
 square. 
 
 Villons, near Thaon, and Bougy, near Evrecy, have the little 
 
154 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 central spire turret that we meet with in Gloucestershire and Wilt- 
 shire, of which Harscomb, Boxwell, Acton Turville, and West 
 
 1 
 
 ViLLONS. 
 
 Littleton, furnish examples. The construction is on the same 
 principle. 
 
 The purely Romanesque church of Thaon is well known. I 
 need only add that its secluded situation in a wooded valley adds 
 to the interest of this beautiful relic. It is not now used, a new 
 church having been erected in the village, which is at some dis- 
 tance. 
 
 The four-sided pyramid which crowns its central tower may 
 have suggested the loftier spires, of the same plan, which are seen 
 at Rozel, Beny sur Mer, Colombieres sur Seniles, and other villages 
 in the neighbourhood. 
 
 Langrune, near the sea-coast north of Caen, has a fine church, 
 with a lofty central steeple. The earliest parts of this church may 
 belong to the twelfth century ; much of it is evidently of the thir- 
 teenth, and the upper part of the tower and spire belong to the 
 
Page 154. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 155 
 
 fourteenth. The composition of these is what I have already 
 described as characterising the steeples of this district, but the 
 pinnacles and spire-lights have disappeared, and the upper part of 
 the spire is slightly truncated. Notwithstanding these mutilations, 
 the general effect is very fine. The jambs of the belfry arches are 
 clustered with shafts, the architraves richly moulded, and the 
 capitals and cornices delicately sculptured with foliage. The spire 
 also is pierced with foliated openings, giving it an air of great 
 lightness at a distance. The interior of this belfry is worth notice, 
 as it appears to combine great strength with thinness of material ; 
 a sort of buttress, resting on the thicker walls of the tower, runs up 
 to each of the angles and sides of the spire ; these are pierced below, 
 so as to afford a free passage all round. The interior of the spire 
 is visible to the top. The lower stage of the tower forms a lantern 
 to the interior of the church. The nave has aisles, from which it 
 is divided by massive columnar piers, supporting richly moulded 
 arches, some of which are round, but most pointed. Above is a 
 triforial arcade, not however sufficiently open to form a gallery. 
 The clerestory is a single pointed light in each bay. The whole 
 church is vaulted. The tower piers are clustered with shafts, and 
 exhibit the round and square abacus. The chancel is short, and 
 forms a polygonal apse, having windows of a single light. The 
 vaulting of the nave is strengthened by flying buttresses externally. 
 The north transept has a shallow porch with a high pitched gable ; 
 the mouldings of the door-way are very minute and delicate. 
 
 The steeple of Bernieres is one of the loftiest in the department ; 
 it stands at the west end of the church, and has a western porch of 
 the thirteenth century. The nave, the vaulting of which has been 
 noticed, clearly presents two dates of Romanesque . 
 
 The priory of S. Gabriel has no more of the conventual church 
 remaining than the choir, which terminates in a semicircular apse. 
 The vaulting, as we have observed, is divided by the transverse 
 arch ; the arches are principally semicircular, and much ornamented 
 with the chevron. One or two pointed arches occur, having the 
 same ornament. 
 
 The parish church of S. Gabriel is remarkable for the manner 
 
156 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 in which the central tower is supported ; namely, by a single arch 
 springing segmentally at a very short distance from the floor, and 
 grooved or ribbed, like the arches of some of our old bridges. The 
 tower is slender, square, and has a good decorated belfry window 
 in each face. 
 
 The western steeples of Bayeux Cathedral are extremely fine ; 
 they taper from the very ground, in consequence of the buttresses, 
 which are massive, and project in many stages. A considerable 
 part of the towers is Romanesque. The central tower is finished 
 with a cinque-cento or Italian cupola, good in its way, but scarcely 
 harmonizing with the rest ; and the outline would have been finer, 
 if the tower had been kept below the elevation of the western spires, 
 as at Coutances, and in the Abbaye aux Hommes. The nave, the 
 lower stage of which is Romanesque, has a rich appearance, owing 
 to the work on the masonry. The exterior of the choir, with its 
 projecting chapels, presents some remarkable arrangement. 
 
 Tour le Bessin has a massive tower and spire at the intersection ; 
 of a composition similar to that we have described. On each side 
 of the chancel, which belongs to the fourteenth century, are curious 
 polygonal chapels. 
 
 The abbey church of Cerisy has had some modern touches ex- 
 ternally, which somewhat disguise its early Romanesque character ; 
 the interior is more in its original state. Much of the nave has been 
 destroyed. Of the central tower, the lower stage seems untouched, 
 and has a Romanesque arcade. The upper stage is a modern 
 addition, unless I am mistaken ; but still the tower is low and 
 massive. The round arch prevails in the nave, which has triforium 
 and clerestory ; the former consists of two arches under a larger one, 
 the whole being much enriched. There appears to have been no 
 vaulting ; the shafts dividing the bays run up to the flat ceiling. 
 Westward of the present nave are some early pointed arches, the 
 remains of the part which has been destroyed ; there are also some 
 remains of the monastic buildings. 
 
 The western spires of S. Lo have a fine outline, but the church 
 itself seems late, and is not of much interest. The pulpit affixed 
 to it externally is curious. 
 
ARCHITECTUEAL STUDIES IN FRANCE. 
 
 157 
 
 Coutances Cathedral has evidently work of two periods within 
 the thirteenth century, and some later additions. The composition 
 of the western steeples is remarkably good. An octagon, rising 
 from a square tower, supports the spire. The corner pinnacles, 
 however, of the tower, which are carried up beyond the spring of 
 the spire, preserve the square effect, but the outline is varied by 
 the spire-lights rising above these pinnacles. A projecting turret, 
 at the external angle of the tower, completes the whole. The cen- 
 tral lantern, which is octagonal, is also surrounded by pinnacles. 
 Its height is somewhat more than that of the spring of the western 
 spires. 
 
 'Between this and Avranches, the use of granite as a building 
 material begins to give a new character. The mouldings are care- 
 fully worked, and all those which are usual in a softer stone are 
 evidently recognised, but at the same time are necessarily modified. 
 It is impossible to avoid a certain degree of roundness ; details are 
 more generalised, their effect is obtained by as shallow cutting as 
 possible, the degree of finish required being given by the careful 
 manner in which the surface is worked. Some of the clusters of 
 foliage round the capitals are admirable studies on this account, the 
 proper forms being strictly preserved, and the effect of the masses 
 secured, with very little under-cutting. It is plain the workman 
 felt he had only the command of rather faint shadows, and yet, by 
 his arrangement of them, he has left nothing to desire. This will 
 be noticed in the indications of Romanesque ornaments, (usually, 
 when the material is more tractable, cut very clear and sharp,) 
 which we find in the abbey of Luzerne and other churches of the 
 twelfth century ; the billet, chevron, &c. being just as it were 
 indicated, and still perfectly sufficient for their purpose. 
 
 Externally these churches are, as may be supposed, plain and 
 severe. The tower, which in many cases is central, often tapers 
 upwards. An open parapet is frequent, sometimes consisting 
 merely of a horizontal course supported by upright bars of stone, 
 sometimes exhibiting a flat trefoil head. The tower has often a 
 roof between two gables. 
 
 Of these granite churches, I may name the abbey of Luzerne, 
 
158 
 
 AECHITECTURAL STUDIES IN FRANCE. 
 
 late Romanesque and early pointed. It is a ruin, but enough re- 
 mains to show what it was when complete. The piers of the nave 
 are square, with edge shafts ; the arches slightly pointed. There 
 is a clerestory, but no triforium ; the west door is round-headed, 
 and has those indications of ornament which I have noticed. The 
 central tower is a fine one, with three tall pointed windows, of a 
 single light, in each face. There are transepts, and a choir or 
 chancel with a flat east end. The situation is extremely beautiful, 
 in a wooded valley ; but the trees render it difficult to obtain a 
 good general view. 
 
 Sartilly, Dragey, Genets, the desecrated church of S. Leonard, 
 S. Jean, Fonts, Montviron, S. Quentin, S. Pair le Val, and others 
 
 near Avranches, are good specimens of the granite church, and 
 most of them present some well executed detail. 
 
 Villedieu has a central tower of Flamboyant work, of consider- 
 able richness, also in granite. 
 
 Fontorson, at the very corner of the province, on the frontier of 
 Brittany, has a church which ought to be examined, and compared 
 with some in other parts of France, as it has a little of the Ange- 
 
Page 158. 
 
Genets. 
 
 I'agk 158. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 159 
 
 vine character. The nave is without aisles, and vaulted in nearly 
 square compartments, with dia- 
 gonal ribs ; the transverse arches 
 are pointed. The capitals, and 
 the transoms of some of the doors, 
 (which are round-headed,) are 
 enriched with grotesque sculp- 
 ture. The west front forms a 
 sort of porch, with a large open 
 archway ; in the real west wall 
 of the nave is a round-headed 
 
 door with a window above. The front is flanked with square turrets. 
 The central tower is massive, and appears to belong to a later 
 style, as does also the chancel, which has aisles. The church has 
 transepts. 
 
 PONTORSON. 
 
 PoNTORSON, W. End. 
 
 For the general appearance, as well as the details of the in- 
 teresting buildings on Mont S. Michel, I refer you to Mr. Bouet's 
 excellent illustrations. The Flamboyant of the choir, chastened 
 as it were by the hardness of the material, has a very pleasing effect 
 both externally and internally. The nave is grand and simple, pro- 
 
160 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 bably of the eleventh century ; it is not vaulted. The cloisters, the 
 inner arcades of which are worked in soft stone, are an exquisite 
 specimen of early pointed, and the granite work of the thirteenth and 
 fourteenth centuries is very fine. But my architectural visit was a 
 hurried one. It is enough to know, without examining it critically, 
 that the work of ages is heaped upon that gigantic isolated rock, 
 and we had better contemplate this wonder of nature and art at our 
 leisure from a distance, than be carried over its details in the com- 
 pany of a crowd of strangers and an impatient guide. Would you 
 see and feel what Mont S. Michel really is ? Go by yourself to a 
 distance from the beaten track, (a few hundred yards will be suffi- 
 cient,) and look at that rugged pyramid, where the work of man 
 is scarce distinguished from the bare granite rock on which it is 
 reared, rising from the unbroken waste of dull colourless sand — no 
 other object in your sight, but the faint outline of a low distant 
 coast, no sound but a mysterious murmur as the wind sweeps over 
 the samphire-like weed which at intervals just emerges above the 
 sandy level : if external influence can impress a sense of solitude, 
 it is here. 
 
ARCHITECTURAL STUDIES IN FRANCE. 161 
 
 Aqueduct at Luynes. 
 
 CHAPTER XI. 
 
 If we would properly study the mediseval architecture of a country, 
 we must not neglect such remains of an earlier period as may have 
 exercised an influence on the growing style. It cannot be doubted 
 that the splendid monuments preserved in Aries, Nismes, and 
 Orange gave something of a classical character to the Romanesque 
 in their own immediate neighbourhood ; possibly through the whole 
 of the south of France. And wherever these early relics occur, 
 there will generally be found some attempt at imitation, even in 
 works belonging to a period in which the rules of the prevailing 
 style appear to have been definitely settled. I will therefore briefly 
 
 Y 
 
162 
 
 ARCHITECTURAL STUDIES IN" FRANCE. 
 
 notice a few specimens of a date previous to the mediseval era, 
 which I visited in the tour which supplied me with the materials 
 for my present volume. 
 
 We have already observed the peculiarities of Autun cathedral, 
 traceable to the neighbouring Porte d'Arroux. This beautiful 
 gate has two wide semicircular archways, of a square order, quite 
 plain, with the exception of an architrave moulding or label ; on 
 each side of this couplet is a smaller round arch, as if intended for 
 foot-passengers. Above these, resting on a string, is a series of 
 small round arches on imposts of a simple character, but separated 
 by fluted Corinthian pilasters, which run up to an entablature, 
 finished by a rich cornice. This part is in great measure copied 
 in the triforium. 
 
 The Porte S. Andre is similar in many respects, but is of the 
 Ionic order, and has projecting wings on each side ; one of these 
 has been rebuilt, and the whole of the structure much restored. 
 
 The building called the temple of Janus might at a distance be 
 taken for the keep of a Norman fortress. Its dimensions are, 
 Internal breadth from N. to S. — about 39 feet. 
 
 ,, „ from E. to W. — about 36 feet. 
 
 Thickness of wall — about 6 feet. 
 Height — apparently about 70 feet. 
 The walls are of rubble, faced with a more regular masonry of not 
 very large stones, having a stronger resemblance to the work which 
 French Antiquaries attribute to the seventh and eighth centuries 
 than to that of the eleventh or twelfth. In the west wall (inter- 
 nally) is a semicircular niche, or apsidal recess, not affecting the 
 outer surface of the wall. The higher windows are square-headed, 
 but with arches turned over them. Some of the lower openings 
 are grouped so as to resemble a large trefoil headed arch. This 
 curious relic stands on low ground, at a short distance from the 
 town, near the river. Only two of the four sides remain. 
 
 Not dissimilar in character, though of a different shape, is the 
 round tower at Perigueux. This also stands on low flat ground 
 near a river ; it is at no great distance from the church of La Cite, 
 which has been described. It appears unconnected with any other 
 
Page 162. 
 
ARCHITECTURAL STUDIES IN FRANCE. 163 
 
 building, and even now forms an imposing object, though probably 
 diminished in height. It is built of rough stone-work, faced as 
 usual with more regular masonry, and is divided by horizontal 
 
 Perigueux, Rojvian Tower. 
 
 brick bands at various intervals. At an elevation of more than 
 half the present height of the structure, is a row of small square- 
 headed windows with arches above them, the voussoirs clearly 
 marked by divisions of thin brick or tile. As there is a breach 
 comprising nearly a fourth of the circumference, it is difficult to 
 conjecture where the door, if any, was placed, or what was its 
 form. 
 
 Near Luynes, between Tours and Langeais, are the remains of 
 
164 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 a Roman aqueduct consisting of a range of arches, some complete, 
 others broken, of considerable height. The piers taper upwards, 
 and the masonry of the surface is tolerably regular ; the stones 
 used in the upper tiers are smaller than those below. Some thin 
 bricks appear among the voussoirs of the arch. 
 
 The relic of antiquity preserved at Bordeaux, called the Palais 
 Gallien, but in reality the entrance of an amphitheatre, of which 
 
 Bordeaux. 
 
 several arches remain, and of which the area may very easily be 
 traced, is worthy of attentive examination as a specimen of stone 
 masonry ornamented by the introduction of bricks. These appear 
 in the voussoirs, both of the semicircular arches, and of the hori- 
 zontal ones which are here and there inserted in the wall, whether 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 165 
 
 for strength or ornament. And they also form horizontal strings, 
 varying both in their own depth, and the intervals at which they 
 occur, which run across the pilasters, and along the surfaces of the 
 wall. But I was most struck with this gateway from its forming a 
 perfect Romanesque front, admirable in its proportions, and of an 
 arrangement combining both beauty and convenience. The sur- 
 face, it is true, is not entirely flat, partaking somewhat of the curve 
 of the amphitheatre, but I do not think it has all its curvature ; it 
 is in fact scarcely distinguishable from a flat front. This example 
 is extremely valuable if we would attempt a pure round arched 
 style, as the arch itself is the predominant feature, the elements 
 derived from the Greek style occupying a very subordinate posi- 
 tion. The pilasters, indeed, are not so prominent as the flat but- 
 tresses in most of our Norman buildings. I fancied, in looking 
 over a series of engravings illustrating the mediaeval architecture 
 of this part of the south of France, that I could recognise this type 
 or model in some of the principal churches of the eleventh and 
 twelfth century. 
 
 As the requirements of vaulting tended so materially towards 
 the development of the Gothic styles, every ancient example of 
 vaulting must be instructive and interesting. Such occurs in the 
 Roman remains now within the precincts of the Hotel de Cluny in 
 Paris. A large oblong hall is covered with a roof formed by the 
 intersection of two semicylindrical vaults which are about equal to 
 each other in diameter. The part therefore corresponding with the 
 groined or cross vault is square, but the barrel roof is continued 
 over the remainder of the area on each side of the centre ; also 
 the transverse vault is prolonged to cover a chamber or recess at 
 the side. The construction seems very accurate, and the work 
 strong ; but there are no ribs, either longitudinal, transverse, or 
 diagonal. In the walls, the brick or tile is freely used, as in other 
 cases, for horizontal bond-courses, and between the voussoirs of 
 the arches. In one of the sides is a concave apsidal recess or 
 niche under a round arch, between two smaller arches of a square 
 order, sunk in the wall. The room, which is used as a museum 
 for fragments of a considerable size, is grand in its scale and pro- 
 
166 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 portions, and in its present state exhibits no feature inconsistent 
 with a pure round arched style. 
 
 Les Therm es, Paris. 
 
 I have not been looking at these ancient remains with the eye of 
 an antiquary. There can be no doubt that they are all of a much 
 more remote date than any of the Romanesque buildings we have 
 been considering, and some of them were doubtless taken as models, 
 up to the period when the mediaeval pointed architecture became 
 fully established. In some respects the early Romanesque is more 
 rude and barbarous ; its masonry is generally less regular, and its 
 construction is often more defective ; but as a style it is more con- 
 sistent. The Roman style was seldom ornamented without admit- 
 ting some incongruous features ; the Romanesque admitted of great 
 enrichment without once recognising any architectural principle in 
 variance with those resulting from the constant use of the arch and 
 vault whenever openings had to be made, or spaces covered. If 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 167 
 
 therefore we can combine the refinement of the one style with the 
 purity of the other, we shall arrive at a style that may, if properly 
 treated, stand high in architectural merit. But we will consider 
 this subject more fully in my concluding chapter. 
 
 If we take any interest in the revival of classical architecture, 
 we shall find a great number of churches in France which show 
 the change from the Gothic. Many of these are extremely beauti- 
 ful in their way, and we might, within a very moderate space, say 
 a circle of twenty miles radius round Paris, collect a series, which 
 would carry us gradually, and almost imperceptibly, from the 
 Flamboyant to the revived Italian. The most remarkable example 
 is the church of S. Eustache, in Paris ; of lofty Gothic proportions, 
 and with tall shafts ; but having all the arches semicircular, and 
 all the details of cinque-cento. 
 
 A little gateway at Rouen struck me as being very elegant in 
 its proportions ; it probably belongs to an early period of the re- 
 vival. It tapers upwards in stages, and is pierced by round arches, 
 and ornamented with engaged columns and pilasters. 
 
 Most of the old French towns abound in good specimens of the 
 early revival. I remember to have noticed some beautiful examples 
 in Poitiers. I have not studied, or taken particular notes of any 
 of these, as my business is rather with ecclesiastical than domestic 
 architecture. Still, as the two have always had a close connection 
 in all ages when architecture did not depend upon mere caprice, 
 secular buildings ought to receive their full share of attention, and 
 those of the period I am now speaking of exhibit much variety and 
 many points of interest. I will therefore recommend the tourist 
 who visits Caen, if his whole time is not absorbed by the churches 
 in that noble city, to take notice of the ancient Hotel de Ville, 
 nearly opposite the south front of the church of S. Pierre ; and I 
 would more especially direct his attention to the Maison des Gen- 
 darmes, at the extremity of the town nearly due east of the Abbaye 
 aux Dames. The principal feature in this is a low round tower, 
 ornamented with medallions of heads. The workmanship is very 
 good. This example probably belongs to the first half of the six- 
 teenth century. 
 
168 
 
 AECHITECTURAL STUDIES IN FRANCE. 
 
 The revived Italian in France, as elsewhere, has not the massive 
 grandeur and simplicity that we find in the Roman style of the 
 first centuries ; I do not remember to have seen a west front in 
 this style so good as the Palais Gallien at Bordeaux would have 
 made. But there is more variety ; and its adaptation to ecclesi- 
 astical purposes gives it a character of its own, independent of the 
 ancient classical style. Still I think there are not many examples 
 that we shall choose for direct imitation ; my reasons for noticing 
 the style at all, as well as that from which it is derived, will 
 shortly, I hope, become evident to the reader. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 100 
 
 CHAPTER XII. 
 
 The architectural writer, like the architect, works for a future 
 generation. I do not mean by this, that either his name, or his 
 writings, are more likely than those of other authors, to descend to 
 posterity ; hut, that if he is instrumental in developing, or trans- 
 mitting sound principles or theories, it is more probable that the 
 fruits of them will be reaped by another age than by his own. 
 He may possibly live to see them adopted with more haste than 
 judgment, and to hear his own name quoted in defence of absur- 
 dities which he would be the first to condemn. I can fancy nothing 
 
 z 
 
170 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 more likely than that architectural productions will appear, pro- 
 fessedly carrying out Mr. Ruskin s or Mr. Garbett's principles, 
 and yet so outrageous as to make those writers almost wish to 
 retract their own theories. Even if such works be judiciously 
 designed, it cannot be supposed that the first efforts will be per- 
 fectly successful, and every imperfection will give force to whatever 
 popular prejudices may be opposed to them. Still the principles 
 themselves will be kept in constant agitation, questioned, discussed, 
 sifted, till, upon some happy turn of fashion, or by the intervention 
 of some master-spirit, capable not only of appreciating them, but of 
 identifying them with his own views and feelings, and working 
 upon them as if they were instinctively his own, they shall come 
 fairly into light, and establish their rightful ascendancy. 
 
 It is now upwards of a quarter of a century since Gothic archi- 
 tecture has been decidedly fashionable, not merely as a study for 
 the artist and antiquary, but as a style to be revived by the archi- 
 tect. Is it too much to say that the result of this fashion (hitherto) 
 is, that we have spoiled our old buildings by making them look 
 new, and our new ones by trying to make them look old ? Have we 
 not in our restorations overlooked the dignity and value connected 
 with real antiquity, and with the genuineness of work known to 
 belong to a good period of art ? Have we not forgotten that a 
 mere imitation, however correct, may be as worthless in compari- 
 son with even a defaced original, as a glass bead in comparison 
 with the precious stone it imitates ? Would a collector of pictures 
 put a fine though faded original behind the fire, and substitute for 
 it a brilliant copy ? But every restoration, beyond what is abso- 
 lutely necessary to ensure stability, or to obviate an appearance of 
 ruin or neglect, does as much as this. And neither our increasing 
 knowledge of Gothic detail, nor the expertness of our workmen, 
 makes it otherwise. Ours is not the age of Gothic art ; but only 
 of imitation. 
 
 I am not about to throw any blame upon architects of the pre- 
 sent day. It was right and natural that when new churches were 
 called for, old churches should in the first instance be looked to as 
 models. It was natural the architects should wish to perpetuate a 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 171 
 
 style they were beginning to appreciate and admire. And it is no 
 fault of theirs that they were born in the eighteenth or nineteenth 
 century, instead of the thirteenth and fourteenth, or that, from 
 their necessary intercourse with the world, they should be somewhat 
 imbued with the spirit of the age in which they live. It is not 
 their fault that they have not been brought up in cells and cloisters, 
 instructed in their faith through the medium of symbolical repre- 
 sentations, and excluded from the sight and knowledge of every 
 work of art and literature later than the fifteenth century. The 
 new Gothic churches of the day, are, or promise to be, all that 
 science or knowledge can make them ; but mediaeval Gothic ex- 
 hibited these and much more. It expressed a certain tone of feeling 
 which does not now exist, and is not likely to exist. The mediaeval 
 cathedral springs up like a natural production of the soil. We can 
 hardly look at it as a work of manual labour and calculation. It 
 is long before we think of examining the mechanical structure ; and 
 when we do, we are no more surprised at its perfections than we 
 are in the case of any other natural object. The architects seem to 
 have built their cathedrals, " because they could not help it," and to 
 have exhibited an intuitive, unerring knowledge both of construction 
 and decoration. We criticise any irregularity we may observe, much 
 as we should those of Nature herself ; and, as with her, look for the 
 typical form, not in one example, but from the examination of 
 many. 
 
 Now this is a growth we shall not obtain, unless we can procure 
 the soil, and that is impossible. The spirit and influence of reli- 
 gion acts, we may trust, not the less powerfully, but in a different 
 manner and direction. We attach a very different value to cere- 
 monial services, and consequently to the works connected with 
 them. We are not likely to look on the building of churches and 
 cathedrals as an atonement for sins, or a means of salvation. We 
 cannot but feel that there are many works of piety and charity 
 as acceptable, or more so, than raising costly churches ; and we 
 are, I may surely say, more deeply impressed than our mediaeval 
 ancestors, with the truth, that it is not on these things we must 
 depend for the hope of pardon and mercy. Instead of the enthusi- 
 
172 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 astic zeal which prompted the erection of these noble edifices 
 through every quarter of Christendom, we have no more than that 
 just sense of propriety which dictates the appropriation of the best 
 buildings we can raise, to the highest purposes ; which demands a 
 fitting dignity and decency both in the performance of religious rites 
 and the edifices in which they are celebrated. 
 
 It is said that Gothic is especially Christian architecture. We 
 will discuss this matter presently ; but if it be, it is the offspring of 
 a peculiar phase of Christianity, which perhaps can never again 
 occur. The Christian feeling that would prompt a person to de- 
 mand absolution from the Pope for a crime he was about to com- 
 mit, and would also prompt the same Pope to intimate that his 
 absolution was not necessary, seeing that the crime was, the murder 
 of a person who was an enemy of God,* is not one that we should 
 
 * I am happily enabled to show, what I here most positively assert, that my re- 
 marks apply to different ages of history, not to different churches. — Whether the 
 fact in question be true or not, the monk of the thirteenth century appears to have 
 related it as though it were by no means alien to the moral feeling of the day ; while 
 the writer of the nineteenth century, after giving the narrative of the monkish historian, 
 repudiates it as a calumny upon religion. I am most anxious, in such a subject as I 
 am at present engaged upon, to avoid any religious controversy ; and if I should 
 anywhere allude to differences of ceremonial, it is only that I may point out the 
 absurdity of imitating buildings adapted to one form of worship, in those which we 
 want for another. 
 
 I subjoin the passage 1 refer to, taken from the " Tablettes Chronologiques de 
 I'Histoire du chateau et de la Ville de Loches :" — " 1012 — En cette annee, Foulques 
 Nerra, pour remplir la promesse qu'il avait faite au pape de se venger des mechants 
 et de Crescentius, patrice romain, ennemi de Dieu et de lui, de le delivrer enfin. ainsi 
 que le peuple romain qui souffrait sous lui, partit du chateau de Loches et se mit en 
 chemin pour aller a Rome; il y fut en vingt jours depuis son depart de Loches, et se 
 presenta au Pape Sergius 4. II lui demanda I'absolution du crime qu'il allait com- 
 mettre, et, sous le pretexte du grand desir qu'il avait de voir Crescentius, il se rendit 
 au pied de la tour ou il etait; Crescentius lui avait fait dire qu'il le pourrait voir, et 
 meme lui parler, s'il voulait se trouver de bon matin devant la tour. Foulques fut 
 ravi de I'expedient, et ayant fait appeler les quatre freres archers, surnommes les 
 promps, il leur donna ses ordres. Apres avoir echange quelques paroles avec le 
 gouverneur de Rome, le comte donna le signal, et un instant apres Crescentius tomba 
 mort de la fenetre en has devant ses pieds ; puis le comte s'en alia de la avec les siens 
 au palais de Latran, ou etait le pape avec tout son clerge et le peuple romain. Foul- 
 ques demanda I'absolution pour lui et les siens ; le pape declara qu'il n'en avait pas 
 besoin, qu'il fallait seulement prier Dieu pour lui de ce qu'il avait abattu I'ennemi de 
 Dieu du sommet de I'iniquite ; il lui remit les reliques des bienheureux Daire et 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 173 
 
 wish to see revived, even if its revival might be accompanied by 
 that of the purest Church architecture. We must be content with 
 a somewhat advanced state of civilization, though it be less favour- 
 able than one in its infancy to the highest imaginative performances, 
 whether they be in poetry or architecture. 
 
 The Gothic style is not a bad style, nor are the architects of the 
 present day bad architects, yet the two do not agree together, and 
 the result is unfavourable to both. Either the architect feels he 
 must build a Gothic church, because he is not allowed to build any 
 other, and sets about his task in the best way he can, collecting 
 authorities for his sculptures and mouldings, taking for his general 
 model some old church that he may most fancy, or can most con- 
 veniently study, and endeavouring to adapt its plan to the wants of 
 a modern congregation ; the result perhaps is a respectable build- 
 ing, that you may look at with tolerable complacency, feel no in- 
 clination to find fault with it, and still less, ever to see it again. 
 Or else he tries to mediaevalise himself and his ideas ; recals anti- 
 quated symbolisms and ritualisms, provides carefully for the possible 
 re-establishment of exploded forms ; is very particular about his 
 vulne-windows, his sedilia and piscina ; will give great depth to 
 the piers of his chancel arch for the sake of piercing them with 
 hagioscopes ; introduces in the tracery of his windows every possi- 
 ble combination of circle and triangle ; covers his walls with texts. 
 
 Crysant, et avec son clerge et le peuple remain I'accompagna a un mille hors de la 
 ville. Le comte s'etant embarque, apporta les corps des saints martyrs jusqu'au 
 chateau de Loches, ou ils furent re9us en grande reverence par le clerge et le peuple, 
 et par I'abbe et les moines du Saint Sepulcre de Beaulieu, lesquels les resserrerent 
 honorablement au meme lieu ou ils sont encore a present." — L'Anonyme de Mar- 
 moutiers. 
 
 " Nous avons dii, en iraitant Dufour, mettre le recit attribue au moine anonyme de 
 Marmoutiers ; nous ajouterons comme lui ; qu'il est constant, d'apres le moine de Cluny, 
 Pierre Damien et Leon d'Ostie, dans leurs epitres et chaoniques, que ce fut Othon 3, 
 Empereur d'Occident, qui fit mettre a Mort Crescentius, le 29 Avril, 998 ; que cet 
 evenement eut lieu sous Gregoire 5, et non sous le pontificat de Sergius, qui ne tint le 
 siege de Rome qu'en 1009 et Mourut en 1012. Le recit de Tanonyme n'est done 
 qu'une fable que la sainte critique doit rejeter et la religion desavouer." — Tablettes 
 Chronologiques de I'Histoire du Chateau et de la Ville de Loches. Par M. le Chev. 
 Adolphe de Pierres, Membre de la Societe Archeologique de Touraine. Paris : Typo- 
 graphic de Firmin Didot Freres, Rue Jacob, 56, 1843. 
 
174 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 if not in a language unknown to the vulgar, at least in a character 
 utterly illegible to them ; and the result is a general effect rather 
 calculated to excite curiosity than to impress the mind with any 
 devotional feeling. Now I cannot blame the architect who does 
 this, if the standard of merit is correctness of imitation, or rather 
 closeness of copyism. And do we recognise any other ? Certainly 
 convenience and adaptation to its purpose is not recognised as a 
 merit in a churchy for if it has an open unbroken area, fit for a 
 large congregation assembled in sight of the minister, it is at once 
 condemned as a meeting-house, and any architect ordered to give 
 such a building a church-like appearance would begin by dividing 
 it into aisles with arcades and screens, just as the mediaeval archi- 
 tects began to do in Touraine with the wide naves of their prede- 
 cessors. 
 
 I am far from denying that many modern buildings of great 
 beauty and excellence of design have been erected in mediseval 
 styles. There is a cross church in the outskirts of Northampton 
 very satisfactory in its proportion, and, as far as I could judge -on 
 a hasty view, in its detail ; it is of the decorated style, and as the 
 dark stone of the country soon assumes the tint of age, it may 
 shortly be mistaken for a genuine church of the fourteenth cen- 
 tury. 
 
 The talented architect of Peckforton Castle in Cheshire, imbued 
 with a perfect knowledge of the subject, and inspired by the mag- 
 nificent site placed at his disposal, has produced what at a distance 
 would be taken for an Edwardian castle. And this would be a 
 really great work, if we could divest ourselves of the idea that the 
 highest standard to which it could be referred is, correctness of 
 imitation. True, it has great and intrinsic merits of its own ; its 
 masses are grand, its outline is fine and varied, it harmonizes ad- 
 mirably with the scenery ; everything speaks it to be the work of 
 a superior mind ; and yet in general estimation all these merits 
 will be absorbed in the one pre-eminent merit, namely, that it 
 might be taken for a mediaeval structure. In fact the highest 
 praise is, that a thing should be taken for what it is not. 
 
 Now this is not the revival of a style. When we see a butterfly 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 175 
 
 fresh from its chrysalis, we do not exclaim, " what a wonderful 
 likeness to the caterpillar we remember a few weeks ago," but, 
 " what a wonderful difference." A revived style must show 
 changes, and those not for the worse, accommodating it to a new 
 state of existence. The revivers of the classical style never thought 
 of confusing their works with those of the ancients. Michel Angelo, 
 or Palladio, never dreamt of producing structures that might be 
 mistaken for specimens of old Eoman art. They took what they 
 wanted from the magazines of antiquity, moulded it into new com- 
 binations, and enriched it with new additions, so as to make it a 
 real, living style, suited to the exigencies of the day, and likely to 
 receive vigour and refinement from the natural growth of taste and 
 talent which might be looked for in the existing state of society. 
 The artist had not to throw himself altogether into the past, but to 
 gain strength and nourishment from the present. 
 
 The revived classical style has unquestionably grave faults, but 
 it is one that cannot reasonably be despised. Great architects 
 have from time to time chosen it as the vehicle of their thoughts, 
 and produced great works, and will continue to do so ; * and even 
 if certain rules have been laid down, the conventional observance 
 of which is demanded, it is not on the observance of these that the 
 value of a building, I mean one of high pretension, rests ; still less 
 does it rest on its imitation of the manner of any other building, or 
 class of buildings, but on its own intrinsic worth. The date of an 
 edifice of this style is not an element in the calculation of its value, 
 except so far as it may denote a good period of art ; and this may 
 return at intervals ; for there is nothing, in the present state of 
 society, to prevent such return under favourable circumstances. 
 
 The Gothic, like the Egyptian, bears the impress of a peculiar 
 tone of feeling, and a peculiar frame of society, which now, as far 
 
 * We are too much in the habit of considering a work poor, and unworthy of notice 
 or preservation, because it belongs to what we call a debased style. We forget that it 
 may, notwithstanding, be the work of a great mind, and bear the impress of both genius 
 and feeling. For these, how often are the results of mere technical knowledge of a 
 good style substituted. There is scarcely a restored church but will furnish an 
 instance. 
 
176 
 
 AECIIITECTURAL STUDIES IN FRANCE. 
 
 as most civilized nations are concerned, belong altogether to the 
 past. The Greek architecture is the production of an age, the 
 spirit of which is in many respects not unlike that of our own ; but 
 it is also the production of a people of the most refined taste, and 
 an ardent love of abstract beauty, and therefore when transplanted 
 into other nations is apt to degenerate. The Roman architecture, 
 while it is still more extensively adapted to ordinary wants and 
 purposes, has not that extreme delicacy which prevents it from 
 thriving as it were in a rougher climate. 
 
 The essential fault of Italian architecture is considered to be, an 
 attempt to combine two systems which are incongruous with each 
 other, that of the beam, and that of the arch. Now is this attempt 
 necessarily an error ? Is a mixed construction to be unreservedly 
 condemned ? I think not ; and in this point I agree with the ad- 
 mirers of the beautiful church of S. Stephen's, Walbrook. This is 
 a perfect example of mixed construction. The colonnade with its 
 entablature and flat ceiling is used just where it is wanted Where 
 these would have been inappropriate or impracticable, we find the 
 arch, the vault, and the dome. An arcade could not be substituted 
 for the colonnade below without great loss of beauty and conve- 
 nience. There can be no rational law to debar an architect from 
 taking the fullest advantage of the materials which nature puts in 
 his hands. If he can command pieces of stone fit to be used as 
 beams and lintels, why should he not so use them, as the Greeks 
 did ? If he can command science and mechanical skill to vault or 
 arch over larger spaces, why should he be restricted from doing it 
 in the same building ? I do not know how much of the present 
 structure may, from motives of economy, or rather parsimony, have 
 been executed in wood ; but I believe it is admitted that the whole 
 design might have been completed in stone, or indestructible ma- 
 terial,* and if so it contains nothing essentially wrong in principle. 
 The system of the beam or lintel is not unknown in mediaeval art. 
 The aisles of Roslin chapel are roofed with a series of pointed 
 
 * Garbett. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 177 
 
 vaults resting on lintels between the columns and walls. The 
 porches and some of the ornamental parts of Chartres cathedral 
 present masses supported as entablatures, on shafts ; and the masses 
 of stone occupying the head of the large doorways of cathedrals 
 are in the same category. I cannot conceive that these, or such 
 buildings as S. Stephen's, Walbrook, All Saints, Northampton, or the 
 roofing of S. James's Church in London, where the two systems are 
 really combined, violate any general fundamental law of archi- 
 tecture. 
 
 But an uncalled for, an unnecessary breach of consistency is a 
 fault. I do not like the addition of a Grecian portico to a build- 
 ing constructed entirely upon an arcuated system. This is not a 
 legitimate combination, but an incongruous mixture. It is, I think, 
 the least pleasing feature in the model of S. Paul's. In the actual 
 cathedral it is made less prominent by its division into stages, and 
 besides is only part of a system that pervades the whole exterior, 
 of which almost every member belongs to trabeal architecture, 
 while the whole internal construction is arcuated. The great fault 
 of the style is that the ornamental system is not only inconsistent 
 with the constructive system, but assumes a more prominent posi- 
 tion. It does not conceal it, but it unduly predominates over it. 
 It does not actually profess or pretend to be the constructive system, 
 but it rather shows what construction would be preferred, and has 
 been abandoned from necessity. And this is not an agreeable 
 idea to be forced upon the mind, or one likely to conduce to the 
 impressiveness of the building. The superficial entablatures and 
 pilasters affixed to walls and piers are in their place when there is 
 a real construction, on corresponding scale, of columns and entab- 
 latures, but not where the construction is wholly or prin^cipally 
 arcuated. Still many subordinate ornaments that are usually, but 
 not necessarily, connected with trabeal construction may be ad- 
 missible. The style might be cleared of much of its inconsistency, 
 and most of those defects pointed out by Mr. Ruskin, without alto- 
 gether abandoning its classical character. 
 
 But in defence of the Italian style as it stands, it may be said 
 that it is a comprehensive one, which provides for two different 
 
 A A 
 
178 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 modes of construction, but does not render it necessary that they 
 should both be employed in every fabric, belonging to it ; at the 
 same time it fixes upon a uniform system of ornamentation that 
 may connect the whole into one family. I should certainly in this 
 manner defend the use of buttresses, &c. in many Gothic churches 
 where they are not constructively necessary. Still, as I have 
 hinted, we want buildings of an intrinsic perfection, and not owing 
 their value solely, or chiefly, to their being connected with any 
 nominal style. A great work should bear to be criticised, as if it 
 were the only specimen of its kind in the world, by the rules of 
 nature and common sense. Mr. Ruskin and Mr. Garbett have 
 taught us to employ this kind of criticism, and in so doing they 
 have conferred a benefit on true architecture which another gene- 
 ration will acknowledge. 
 
 The wooden model of Sir Christopher Wren's first design for 
 S. Paul's ought doubtless to be restored ; but in its present muti- 
 lated condition it is perhaps quite as instructive as if it were per- 
 fect. As in many ancient buildings, the solid construction remains 
 nearly entire, the adventitious ornaments have dropped away. 
 Does not this show the sort of ornaments that ought to be rejected 
 as foreign to the design ? In the sketch I have already given, I 
 have not attempted any restoration whatever, I have even repre- 
 sented the dome over part of the nave as broken through, and 
 showing the exterior of the large central dome. This accidental 
 mutilation may perhaps have somewhat enhanced the efi'ect of light 
 and shade, but I do not suppose it has materially altered it. 
 
 In the sketch, we recognise at once, the pier compartment, the 
 arch compartment, and the dome compartment, which I have al- 
 ready spoken of as belonging to the simplest perfect arrangement 
 of a domical building on a rectangular plan. Of these, the pier 
 compartment is evidently the portion that calls for a careful, and 
 somewhat severe system of architectural ornament. The arch and 
 dome compartments may be left to themselves. A bold cornice at 
 the base of the dome, and, perhaps, an architrave moulding round the 
 arches, are all that is really required ; I should say the surfaces 
 were better unbroken, even by raised scroll-work or panellings, but 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 179 
 
 they afford good scope for colour. The pier compartment, (as de- 
 signed by the Architect,) at least that one which is nearest the eye, 
 has an arched semidomical recess, flanked by two pilasters, which 
 are raised by pedestals, and support an entablature and cornice 
 boldly projecting from the wall. The impost of the arch has a 
 string, which runs horizontally along the face of the pier till stopped 
 by the pilaster, and also under the semidome which vaults the 
 recess. 
 
 Now the pilasters and entablature, though they may add to the 
 severity of the composition, are still out of place in a wholly ar- 
 cuated structure. The pilaster may indeed seem here to be legi- 
 timately introduced, corresponding as it does with a transverse rib 
 of the roof ; but this requires a break in the horizontal line of the 
 cornice, which, unbroken, would form a very grand finish to the 
 pier. 
 
 If the pilaster be omitted, we could not gracefully retain the full 
 entablature, but we might still retain a rich cornice, and a string 
 at a short distance below it.* Also the string on the impost of the 
 arch might Ibe carried round the whole pier. Supposing more 
 architectural enrichment is wanted on the front, might we not act 
 upon Mr. Ruskin s suggestion of making the voussoirs of the arch 
 vehicles of ornament ? f If the pier were pierced by openings of 
 a rectangular section, as at Perigueux, I think we might, with 
 great advantage. A semicircle of sculptured voussoirs, bounded 
 by a label at some distance from the arch itself (by no means an 
 uncommon Romanesque arrangement) would have a good effect ; 
 but I do not think they would suit quite so well with the concave 
 surface of the recess, except by the intervention of another square 
 order. 
 
 We might also obtain variety by the introduction of a shaft or 
 column under the edge of the arch, an arrangement of which the 
 Romanesque affords a multitude of examples. Let us apply here 
 the rationale of the torus at the edge of an order, between two hol- 
 
 See Garbett. 
 
 f " The Stones of Venice," vol. i. 
 
180 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 lows ; which Mr. Ruskin has given so clearly. The edge of the 
 pier is most exposed to injury, therefore it is expedient to round it 
 off. To avoid the appearance of this being done by accident in- 
 stead of design, the hollows are cut, which give this rounded angle 
 the character of an engaged shaft. But we may wish to construct 
 this part with a harder, or more beautiful and costly material than 
 the rest of the pier, and in this case it is better to detach it alto- 
 gether, placing it free, in the nook of a re-entering angle, as an 
 actual shaft, than to build it up in the solid mass of the pier. Be- 
 sides, even if it be of the same material with the pier, its exposure 
 to injury renders it expedient that it should be easily restored or 
 replaced, without disturbing the principal support of the weight. 
 
 All Saints Church, Northampton, presents the elements contained 
 in domical churches ; the pier compartment, the arch compartment, 
 and the dome. But in this case the former constitutes an impor- 
 tant part of the area of the church, and consists of one column, and 
 the space between that and the external walls. The roof of this 
 compartment is flat, fronted in two directions by entablatures, from 
 which spring the barrel roofs of the arch compartments. These 
 are segmental. The central compartment of the church has the 
 pendentives and dome. The superstructure is of wood ; but unless 
 the lintels demanded for the entablature are too long, it might 
 have been of stone. This opening of the pier compartment to the 
 area of the church adapts the design to parochial purposes. By 
 turning the parts of S. Stephen s, Walbrook, which correspond with 
 the pier compartments, and which are known by their flat roofs, 
 into massive piers, merely pierced or recessed by arches, we should 
 obtain a very grand cathedral form. 
 
 By studying examples of this description we see the peculiar 
 excellences of which the Revived Italian is capable, and which we 
 could not attain in Gothic. I know of no Gothic church which has 
 such a variety of perspectives as the Model of S. Paul's. One, 
 taking the length of the nave and choir. Another of the transepts. 
 A third, diagonal, looking across the dome. A fourth, looking in 
 an oblique direction behind the dome ; this is a most picturesque 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 181 
 
 interior. To these may be added views taken within the area of 
 the dome, and of different parts varying in ornament or compo- 
 sition. 
 
 I think it possible that a cathedral built according to this design 
 might admit the generally incompatible decorations of mural and 
 glass painting, by making the piers recipients of the former. 
 Every window below those in the domes might be, indeed ought to 
 be, filled with richly coloured glass. The ornaments should of 
 course be designed with a view to hypsethral light. 
 
 The interior of the actual cathedral presents only one kind of 
 perspective that I consider satisfactory to the eye, namely, the side 
 aisle, which is very beautiful. In the main avenue of the nave we 
 cannot but feel that the piers, which are of necessity deficient in 
 the lightness and spring of the Gothic, have not the solidity which 
 their own style requires. They seem to vacillate between the two 
 styles, and miss the beauties of both. In the same way the arches 
 are either too wide, or too narrow. Too wide to give the effect 
 derived from multiplication, too narrow for that given by magni- 
 tude and positive distance. These defects are not discernible in 
 the sharper perspective of the aisles, the roof of which has great 
 purity and elegance. 
 
 In the model the central dome has no pendentives. Its cylin- 
 drical drum or tower springs from the ground, and is pierced by 
 eight arches, those towards the cardinal points being wider and 
 loftier than the intermediate ones looking diagonally. The arches 
 have consequently a double curvature, which is certainly a defect, 
 if a slight one ; it may however be compensated by the appearance 
 of strength given by carrying up the surface vertically to the cor- 
 nice below the dome. 
 
 Of the different systems of vaulting used in the eleventh century, 
 the domical combined with the cylindrical of small depth seems 
 the best adapted to a permanent round arched style. The simple 
 cylindrical vault limits the span of the roof considerably ; except 
 in small churches it is ill adapted for a nave without aisles. It 
 requires a very great thickness of solid wall, which itself must act 
 
182 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 as buttress. The quadrantal vault often used as an abutment is 
 unsightly. And, the whole of the roof lying above the horizon- 
 tal line over the clerestory, the apparent height does not fairly 
 represent the real height, especially if there is nothing to mark 
 the ridge of the roof. The cross vault was early introduced as 
 a manifest improvement upon the plain barrel roof, and to stick 
 to the original unimproved form would be a barbarism. Never- 
 theless this roof is good for the intermediate compartment of the 
 chancel, or short transepts, the intersection having a cupola. And 
 it forms a safe abutment to the Byzantine pendentive. 
 
 The Bernay roof, or dome belonging to the same sphere with its 
 pendentives, whether on square or oblong compartments, will I 
 think generally be found pleasing to the eye, while it offers a good 
 surface for pictorial or unprojecting decoration. 
 
 The cross vaulting, used on a large scale in Germany, and on 
 a smaller scale both in France and England, leads to the diagonal 
 rib, and this to the clustered pier, and ultimately the whole system 
 of Gothic architecture. Still many positions might occur authoriz- 
 ing its partial introduction. 
 
 Some of the eastern churches might suggest very valuable hints. 
 I have studied with much interest M. Couchaud's work on the 
 Byzantine churches of Athens,* a locality in which we might rea- 
 sonably look for some remains of architectural taste, even when 
 the styles which formed its principal glory had passed away. They 
 appear mostly to be of the simple domical plan on the square base, 
 which we have considered, with the addition of porch and apse. 
 Some are more extended. One or two have the octagon compre- 
 hending nave and aisles, like Ely, and S. Paul's, but if I under- 
 stand the plans rightly, the dome must be raised on a large square 
 base by pendentives, like S. Maria delle Grazie at Milan. The 
 domes appear to be extremely elegant, and they give an idea how 
 the double curvature in the windows may be avoided, by cutting 
 the head of the arched opening out of the hemispherical surface. 
 
 * See Preface. 
 
ARCHITECTURAL STUDIES IN FRANCE. 183 
 
 The columns immediately supporting the dome seem to be very 
 slender, but I do not observe any entablatures or lintels, except in 
 the doors. Some of the plans might suggest very good designs 
 for parish churches. 
 
 I by no means wish to advocate the reproduction of an anti- 
 quated style ; one superseded, as we may say, by its own energy 
 and progression, as was the Northern Romanesque ; but simply to 
 enquire whether the only style which has any permanent vitality, 
 though it be at present in a languishing state, and which seems 
 likely to exist and advance under the ordinary phases of society, 
 owing its support, not to periods of overstrained excitement, but to 
 a sound, refined and rational taste, and a sense of the dignity of 
 art, enhanced by a due feeling of reverence, when art is employed 
 for sacred purposes ; — I say I would enquire if such a style can be, 
 however gradually, cleared from its blemishes, without loss of the 
 majesty of which it is capable. It will not be the work of one 
 mind, or of one generation ; but whoever feels a conviction, or a 
 hope, or even a wish, that it may at some time be carried out, may 
 assist in its commencement. 
 
 In the domical plan, where the pier compartment is reduced to 
 a simple pier, as in the cathedral of Angouleme, and other churches 
 built on the same principle, the classical forms in present use are less 
 in need of correction. For the pilaster will occupy its place with 
 propriety under a rib or the order of an arch, and will admit of the 
 pure Corinthian capital, or any variety that a judicious decorative 
 sculptor will think fit to substitute for it. Nor does there seem 
 any objection against fluting, which we meet with in Romanesque 
 work at Autun and Beaune. The entablature, however, ought to 
 be reduced. In a building of chiefly arcuated construction, it 
 seems better not to introduce the full Grecian entablature. That 
 is necessarily broad and heavy, that the weight sustained may ap- 
 pear proportioned to the columns sustaining it, but when an arch 
 or roof of masonry appears above the column or pilaster, no such 
 breadth is needed ; the whole entablature is properly compressed 
 into a member corresponding with the Romanesque abacus ; or it 
 might even be omitted altogether. 
 
184 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 I do not suppose I am suggesting anything that has not been 
 done in hundreds of instances. I am not acquainted with the 
 churches of Italy, except a few in Genoa and Milan, but from the 
 engravings I have seen, I am assured that very many examples, 
 both of Romanesque and Revival, might be brought to bear on the 
 subject. The thing to be desired is, that a right system should be 
 carried out regularly and on principle ; that all which is false and 
 incongruous should be rejected, and nothing admitted but what we 
 can defend upon other grounds than that of accordance with a 
 different system, or one which is imperfect. 
 
 It is easy enough to purify the Roman or revived Italian, by 
 divesting it altogether of ornament. This is done in many plain 
 or small buildings both ancient and modern. The Pont du Gard 
 has no incongruities ; but then it has no ornament, beyond the 
 simplest impost strings. Many small village churches of the last 
 two centuries, such as we find in the Isle of Man, in Wales, in 
 the south of France near the Pyrenees, and elsewhere, are perfectly 
 pure, because they are perfectly plain, and I certainly like them 
 better than modern Gothic churches of the same standard. But 
 we do not want a naked style, one which would check the inven- 
 tive genius of the designer ; not a meagre, but a true and con- 
 sistent system of decoration. To give the voussoirs their due 
 importance, and make them vehicles of rich ornament, where the 
 surface admits it, would be going a great way, and here, even our 
 own Norman will help us considerably. Again, the sculptural 
 decoration of surfaces, as Whewell has shown, was gradually aban- 
 doned as the arcuated system advanced towards the perfect Gothic. 
 But if the actual mass of wall, as well as the framework of the 
 building, as marked by lines, be restored to its due importance, 
 ^it is surely no barbarism to revert to the old plan, and ornament 
 the flat or continuous surfaces with sculpture, instead of confound- 
 ing them with the framework by arcades and panelling. 
 • Another subject to be considered is the buttress. I shall still 
 keep to the domical arrangement, because I believe that, whether 
 simple, or multiplied, or in combination, it is best calculated 
 to develop the capabilities of the arcuated classical style. I 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 185 
 
 m 
 
 A 
 
 
 
 
 
 
 
 
 i 
 
 
 will repeat the figure, that it may not be lost sight of. Now 
 this it is clear no external buttress is 
 wanted. The arches bearing weight 
 abut against walls placed in the right 
 direction to resist the pressure, and 
 the force of the abutment may be in- 
 creased if necessary by raising towers 
 on the pier compartments. The bar- 
 rel roof, where it has merely to sup- 
 port itself and the external roof, (even 
 should that be of stone,) has sufficient 
 
 abutment in the massive wall and roof of that compartment ; which 
 latter, if a right proportion be observed, will itself require no fur- 
 ther abutment. But if in consequence of the reduction of the pier 
 compartment into a simple pier, or any extension of the plan without 
 introducing additional pier compartments, an external buttress be- 
 comes necessary, it ought to be in its character solid, immovable, and 
 self-supported. Not appearing, as in Gothic work, to lean itself 
 for support on that which it is built to sustain ; for the classical sys- 
 tem is one, not of artificial and ingenious balancings, but of simple 
 and apparently unstudied stability and repose. All weak waving 
 lines in the buttress, whatever be the authority for them, should be 
 rejected as utterly inappropriate. 
 
 The beautiful church of Wilton must, undoubtedly, at some time 
 or other, have its influence upon English architecture. It goes a 
 great way towards a pure and refined round arched style. I can- 
 not at present call to mind any decided faults or inconsistencies 
 either in its construction or decoration. It shows that such a style 
 will bear any amount of costly and elaborate enrichment ; perhaps 
 it may lead some to imagine (erroneously) that it requires it. If 
 its effect upon the art has not been instantaneous, I think it is partly 
 because it appears rather an attempt to imitate some particular 
 building, or class of buildings, than to develope general principles ; 
 and partly because it is on a plan which has been carried out to 
 greater perfection in another style. The basilican method of com- 
 position, I mean that which comprehends the nave, aisles, and cle- 
 
 B B 
 
18G 
 
 AKCHITECTUKAL STUDIES IN FRANCE. 
 
 restory, has been appropriated by Gothic architecture. It belongs 
 to it by right of conquest. However it may have originated in 
 Romanesque or Roman, the Gothic architect seized upon it, and 
 gave it a beauty, vigour, and energy of his own. He saw that it 
 was expressly calculated for the display of his own genius ; and the 
 naves of Chartres, Amiens, and York attest his sound judgment 
 as well as his boldness. If the plan he found ready to work upon 
 had not suited him, he would have rejected it. Now if we must 
 abandon Gothic, we should not do it in such a manner as to impress 
 every one with the inferiority of the style we select in its stead ; 
 we should not always be endeavouring to cope with it on unequal 
 terms, on its own ground ; and this we do, if we lay aside the 
 Pointed style, and yet cling to the basilican composition. This is 
 the great defect of the actual cathedral of S. Paul's ; and the oppo- 
 site treatment is the great beauty of the model. That does indeed 
 cope with any mediaeval building ; even if there be defects in the 
 style, we feel that in no other, at present known, could the con- 
 ception of the architect be developed with equal grandeur. 
 
 I do not suppose that such views as I have professed will just now 
 be popular. The admirers of classical architecture will not admit 
 that it has any errors to correct, and the medisevalist will hear of 
 no compromise. And I am also aware that it will be long before 
 the sacrifices we make to purity of style will be compensated by 
 the acquisition of equal beauties in exchange ; still if the principle 
 be sound, it will " bide its time," and ultimately triumph. 
 
 Though I fear that Gothic architecture, under any of its phases, 
 will prove to be unsuitable to the spirit of the present age, it is 
 impossible to help feeling the beauty, the solemnity, the construc- 
 tional excellence of the style, and its fitness for religious purposes. 
 If we can appropriate it, without injustice to itself, let us do so. I 
 am anxious that we may avail ourselves of every chance. I have 
 not seen much Italian Gothic, and therefore do not properly ap- 
 preciate it. Mr. Ruskin appears to prefer it to the Northern 
 Gothic, and Professor- Willis's able analysis of the style shows it to 
 be possessed of many true architectural merits. I think it must 
 have many points of resemblance to the Angevine style, though the 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 187 
 
 latter has more of the Northern character. The Angevine has 
 much to recommend it for present use in preference to the more 
 energetic and exuberant style which adapts itself to the basilican 
 composition. It admits of great refinement of line, mechanical 
 propriety of construction, and breadth of treatment. This will 
 suggest delicacy of execution in ornament. Some statues at the 
 western entrance of Angers cathedral are of very beautiful work- 
 manship ; and we look for such in a building of that design. In 
 an Angevine church, height is not the prominent dimension ; the 
 great width of the area spanned by the roof strikes the spectator. 
 So does the simplicity and size of the compartments, which are 
 generally square, or nearly so ; and this, with a proper management 
 of the pier-masses, admits of a perspective quite equal in efifect to 
 that produced by a multitude of slender pillars at short distances 
 from each other. The convenience of giving the congregation a 
 wide unbroken space need not surely be dwelt upon. As the im- 
 posts necessarily project considerably from the wall, to ensure a 
 deep longitudinal arch, which is wanted to perform the part of a 
 buttress, recesses are obtained in which galleries might be placed 
 to advantage, if necessary. We need not confine ourselves to the 
 Romanesque or Transitional French style, in which the churches 
 of Anjou are worked ; some of our early Perpendicular has a soli- 
 dity and massiveness which qualifies it well for an Angevine com- 
 position. The Winchester window,* or a two-light window of the 
 same character, doubled, with no intervening buttress, would be 
 suitable for each bay. The buttresses being at large intervals, 
 should be bold, massive, and projecting. The arrangement of the 
 fronts might be difiicult. Circular windows would be well intro- 
 duced. The width of the church would suggest a polygonal apse. 
 If cruciform, a fine octagonal dome, or even a massive square tower, 
 would give great dignity. The belfry tower might be lateral, or 
 detached. The pitch of the external roof and gable need not be 
 sharper than a right angle. 
 
 I have all along taken for granted that the object has been, the 
 
 * See Garbett. 
 
188 
 
 ARCHITECTUEAL STUDIES IN FRANCE. 
 
 support of an indestructible roof of masonry. But though to the 
 pursuit and attainment of this object we owe the character of 
 Gothic, and indeed of almost all arcuated architecture, still I am 
 not prepared to consider the use of the wooden roof as a blemish. 
 It was used honestly, without disguise, and constructed according 
 to its material ; it was capable of much appropriate ornament, and 
 admitted arrangements of very great convenience, for instance the 
 range of slight piers supporting light open clerestories, as in the 
 late Perpendicular. We cannot deny that churches of this descrip- 
 tion exhibit much beauty and elegance, are not without dignity, 
 and are well adapted to the uses of a large congregation. 
 
 No doubt, the less of destructible material the architect employs, 
 the better ; but it is almost impossible to debar him altogether from 
 its use, and it is also difficult to draw the limit. Still he should 
 always take care, first, that the decay of the destructible parts should 
 not compromise the safety of those which ought to be indestructible, 
 as the masonry ; and, secondly, that every destructible part should be 
 easily accessible for the purpose of inspection and repair. The fires 
 at York Minster tell their story in two ways. The roofs themselves 
 indeed were utterly destroyed ; but the masonry, and even the 
 glass, received little material injury, none, that I am aware of, but 
 what could be repaired by superficial coatings, or the restoration 
 of minor ornaments. However I do not defend the replacing of 
 the burnt roof with timber instead of stone vaulting, since the con- 
 struction has provided for, and demanded the latter. 
 
 I do not know to what extent iron may be rendered indestruc- 
 tible (of course I look to centuries and not years) by coating, or 
 burying in masonry, but I can conceive its admissibility in the 
 form of ties, when co-operating with a cement which hardens, in- 
 stead of becoming weaker by the action of time. Still the truest 
 construction (and this Mr. Garbett strongly insists upon) is that 
 which requires neither tie nor cement at all. 
 
 I cannot bring myself to agree with this able writer in his depre- 
 ciation of our plain village churches. There is a certain charm 
 about them ; and if this be not owing to any artistic merit, it is all 
 the more difficult to restore. They are genuine, though humble, 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 189 
 
 parts of a great and noble system. They show life and energy, 
 and honesty of intention on the part of their builders, no less than 
 the more elaborate structures. If their imitations of these in some 
 cases prove rudeness of skill, or want of knowledge, at all events 
 they are not intended for deceptions ; they are simply links of con- 
 nection with other buildings belonging to the same age, erected in 
 the same spirit, and destined for the same purposes. A village 
 church is really more difficult to deal with than a cathedral, since 
 its interest depends on what is less tangible. Here, association is 
 everything, and every body knows how easy it is to destroy this, 
 and how impossible to recover it. 
 
 I hope I shall not be accused of attempting to recommend the in- 
 troduction of a new style for the mere sake of novelty, or of foreign 
 fashions simply because they are foreign. I want to see that ef- 
 fected in architecture which my friend Mr. Winston is exerting 
 himself to effect in glass painting ; a substitution of art for " bogie 
 work ;" a recurrence to the principles of true taste and common 
 sense, instead of unmeaning copyism ; and the adoption or forma- 
 tion of a style suited to the spirit of the age, (which, if somewhat 
 practical and utilitarian, is not altogether base and despicable,) in- 
 stead of one which has either been exhausted by its own energy, or 
 which finds the atmosphere in which it is placed unable to support 
 it in a state of healthy existence. 
 
190 
 
 ARCHITECTURAL STUDIES IN ERANCE. 
 
 AUVERS. 
 
 CONCLUSION. 
 
 In reading over the late Mr. Kerrich's Observations on Gothic 
 Buildings, printed in the Archseologia, I was much struck with the 
 following passage. 
 
 " In later times it has been the custom to restrain the term 
 ' Gothic ' to this light style only, and it has long been so called ; and 
 that name was received all over Europe ; we find it continually 
 used by all the travel writers, and in the guide books of the diffe- 
 rent cities upon the continent, as well as by writers on the arts 
 themselves, during the whole of the two last centuries ; and it was 
 so well established, and every body understood, and knew so ex- 
 actly, what it meant, that it really does appear to be a great pity 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 191 
 
 people would not rest contented with it. It answered completely all 
 the purposes of language ; and much confusion has been caused of 
 late by the introduction and unsteady use of new and dubious 
 names ; and a vast deal has been written, which might have been 
 well spared." * 
 
 Most antiquaries of the present day have either rejected alto- 
 gether the term " Gothic," as applied to the mediaeval architecture 
 of western Europe, or have admitted it only on sufferance, with 
 some kind of apology for using a word which they considered false 
 and unmeaning. Yet after all it may be difficult to find one more 
 philosophical, or better calculated to convey a true meaning. If, 
 indeed, the word " Gothic" be taken as simply synonymous with 
 
 barbarous," (and this was doubtless the intention of those who so 
 named the style,) we might fairly reject the epithet as unjust ; but 
 if we look upon the Goths as a tribe or nation of surpassing impor- 
 tance in the history of Europe, whose influence, whether as regards 
 literature, arts, institutions, or national character, has been more 
 striking, more permanent, more progressive, than that of almost 
 any other within the records of history, we shall not be so solicitous 
 to rescue a style we admire, from the name, as if it were a stigma 
 of reproach ; especially as a little consideration may enable us to 
 trace in the characteristics of the style the very spirit of the nation 
 whose name it bears. It is true that nation had ceased to exist, 
 in mere name, long before the style appeared in its infancy ; but 
 the national element was strong and active, ever displaying itself, 
 and ready to display itself, in various developments ; a wild, but 
 not unmeaning system of mythology, a romantic vein of poetry, 
 and the chivalry of the mediaeval ages, the rudiments of which may 
 be clearly traced to the Gothic character. 
 
 I have, unfortunately, not given much attention to the study of 
 ethnology ; and therefore if I venture beyond mere generalities I shall 
 infallibly be convicted of frequent blunders ; still I will endeavour to 
 
 * Vol. 16, of Archaeologia, p. 292. Some observations on the Gothic buildings 
 abroad, particularly those in Italy ; and on Gothic Architecture in general. By T. 
 Kerrich, M. A., F. S. A., Principal Librarian to the University of Cambridge. Read 
 on the llth and I8th of May and 1st of June, 1809. 
 
192 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 state what I conceive to be the character of the Goth ; premising 
 that I am probably giving the term a more extended meaning than 
 would be admitted by the acute critic. I should describe him as 
 active, dauntless, energetic, enterprising, quick in invention, per- 
 severing under difficulties, gifted with an indomitable love of li- 
 berty, and a strong sense of justice ; with an ardent, fertile, and 
 vigorous imagination, tending, under the influence of climate and 
 scenery, to the wild and romantic ; little apt to be restrained by 
 rules, as such, but with a keen intuition of many of the true prin- 
 ciples, both of science, and of taste and art. For it does not of ne- 
 cessity argue ignorance or incapacity of art, if he sometimes roughly 
 handled those works of art which he was led to connect with de- 
 cline and degeneracy. I may further add, that the ambition of 
 this people was not so much to preserve its name or individuality, 
 as to diffuse itself abroad, and become as it were an animating and 
 governing principle, among other nations. This I conceive to be 
 the great difference between the Gothic and the Celtic tribes, so 
 similar on many points ; namely, that the one willingly mixes with 
 other nations, losing its name, and its unessential marks, though 
 preserving its influence, while the other is exclusive, and more or 
 less jealous of foreign admixture. No Gothic tribe would willingly 
 preserve, to its own inconvenience, a language different from that 
 of the people with which it might be incorporated. 
 
 At the time of the first developments of Gothic architecture, the 
 Normans perhaps were the best representatives of the ancient 
 Goths ; nearest them in lineal descent, and inheriting most fully 
 their spirit and character. It cannot be denied that they were 
 mainly instrumental in advancing the style to its full perfection. 
 In Normandy, and in England, where the Normans incorporating 
 with the people lost their name, but maintained their influence and 
 ascendancy, the purest and most advanced specimens belonging to 
 the thirteenth century are to be found ; and in these countries the 
 works of the preceding century gave the earliest and clearest in- 
 dications of the coming change. 
 
 It may be said that it is easy to trace the derivation of the Gothic 
 from the Roman, and thence from the Greek. Very true ; but 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 193 
 
 this derivation was not worked out by Greek or Roman minds. 
 Had none others employed themselves in the matter, we might 
 have advanced nearly as far as the Italian Gothic, but we should 
 never have had the opportunity of studying the peculiar character- 
 istics of our Northern Architecture. 
 
 It may also be true that the pointed arch, and many other 
 Gothic features, appeared in the east before their general adoption 
 in Europe. Yet these did not expand into the Gothic style till 
 they became the possession of the Gothic architect ; he saw their 
 value, and combined them into a system adequate to the expression 
 of his own lofty and daring imaginations. 
 
 If the name of any tribe or nation is to be employed in reference 
 to this architecture, I cannot, as I said before, think of an appel- 
 lation really more to the point than Gothic." If we called it 
 Norman, German, Lombardic, or English, or after any other 
 country recognized by an established name at the time of its rise, 
 or at the present day, we should be supposed to speak of it merely 
 with reference to that country ; we cannot, for instance, use Rick- 
 man's English nomenclature in speaking of French or German 
 buildings. If we spoke of Northern architecture, we might be sup- 
 posed to exclude Spain and Italy, where the influence of the Goths 
 penetrated ; if of Teutonic, that is a term we are apt to associate 
 with a certain class of languages, which would exclude the above 
 mentioned countries, as well as France. Celtic would be mani- 
 festly incorrect ; Scandinavian we might use, as nearly synony- 
 mous with Gothic, but should gain nothing by the exchange. In- 
 deed, even this term might give an erroneous idea ; for the part of 
 Europe which answers to the ancient Scandinavia, is, I understand, 
 rather meagre in Gothic buildings. It is to those who migrated, 
 not to those who remained in their own country, that we must look 
 chiefly for the elements of progress. The old wooden churches, 
 however, of Norway must be a subject of great interest to the anti- 
 quary. 
 
 But the term " Christian architecture" has lately been much in 
 vogue. Let us examine whether it gives accurately the real state 
 of the case. 
 
 c c 
 
194 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 Religion, or the want or perversion of it, acts upon the national 
 character, and consequently upon the arts, pursuits, institutions, 
 and habits dependent on that character, much as good or bad 
 passions act upon the features of the human face ; changing it 
 pi'o tanto ; that is, giving it an amiable or malignant expression, 
 softening and refining, animating, inflaming, exalting, debasing, or 
 brutalizing. But they do not, generally speaking, change those 
 lineaments which mark identity ; which distinguish one face from 
 another ; nor does Religion change those traits of national cha- 
 racter which can neither be placed in the category of vices or vir- 
 tues. The reception of Christianity must undoubtedly do much in 
 correcting and purifying the character of any people, but unless 
 this is altogether bad and faulty, there must be many points in 
 which it will leave it altogether unchanged. Say that the temper- 
 ament of one is ardent and imaginative, of another timid and cau- 
 tious, religion, though it may check the one and stimulate the other, 
 in such a manner as to guard against the evils of extremes, will not 
 yet fuse the two into one character ; there will still exist suflicient 
 distinction to cause essential difference both in habits, works of art, 
 feelings and pursuits. It is not altogether idle to speculate upon what 
 persons or nations would have done under different circumstances 
 from those which have actually occurred ; we must do this, if we 
 would penetrate beyond the surface of bare apparent fact ; and we 
 can in no better way learn to realize the effects which have been 
 brought about by circumstances. We may therefore figure to our- 
 selves what would have been the state of the arts in ancient Greece 
 had it been blessed with the revelations of a pure religion ; or 
 what among the successors and representatives of the Goths, had 
 Christianity never expanded itself through Western Europe. Now 
 I conceive it would require a vast amount of prejudice to affirm 
 that the rock of the Acropolis would have bristled, like S. Michel, 
 with Gothic spires and pinnacles, or that the shores of the Rhine 
 or Seine would have exhibited a front of Grecian colonnades. The 
 requirements of a different ceremonial would doubtless have sug- 
 gested different arrangement ; details, wherever they had a sym- 
 bolical character, would have been derived from the sources of 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 195 
 
 religion, and unquestionably a pure religion would have tended to 
 throw an air of purity, refinement, and sublimity over the works 
 connected with it. I have a notion, therefore, that Grecian archi- 
 tecture, instead of being changed to a totally opposite system, 
 would have had its own distinctive character even more strongly 
 marked, if possible ; that it would have become, if I may use the 
 expression, still more Grecian. On the other hand, we cannot 
 deny that a Gothic cathedral quite as strongly impresses us with 
 the ardent and restless energy of the persons who reared it, as with 
 the meek simplicity of the religion in whose service it has been 
 reared. Nor can we deny that we must look for the origin of 
 many of the wild and grotesque details with which it abounds, 
 rather in the legends of Northern Mythology than in the doctrines 
 of the New Testament. Christianity has, in fact, tended to un- 
 gothicise the style ; to subdue and restrain its luxuriance. In so 
 doing it has probably been the cause of an inconceivable degree of 
 grace and beauty ; one that a Pagan Gothic architecture would 
 never have attained ; but the latter would probably have had the 
 real characteristics of the style in stronger development ; would 
 have been more imaginative, more romantic, more picturesque. 
 Hence, though I believe Christianity has been instrumental in de- 
 veloping a grander and more beautiful style than would have 
 evolved itself without her agency, yet the elements she had to work 
 upon, and which give the style its distinctive and individual cha- 
 racter, like the features of a human face, these, I affirm, are purely 
 national, and emanate, not from the revelations of Christianity, but 
 from the imaginative Gothic mind. 
 
 This is not a mere question upon w^ords. If we take it for 
 granted that the Gothic style is in the fullest sense entitled to be 
 called " Christian architecture," we shall be apt to invest it with 
 an undue degree of sanctity ; to consider the attempt to introduce 
 any different style of ecclesiastical architecture as something anti- 
 Christian and Pagan. We shall judge it by other standards rather 
 than its own intrinsic merits ; we shall content ourselves with sym- 
 bolisms to the neglect of construction, composition, and proportion. 
 We shall look to Christianity for principles which it is not her 
 
196 
 
 ARCHITECTURAL STUDIES IN FRANCE. 
 
 design to furnish to us ; just as some have thought to find a per- 
 fect system of natural philosophy in the Mosaic account of the 
 Creation. 
 
 But if we consider Gothic architecture to be one of the creations 
 of a great national spirit, we shall at once account for the difficulty 
 of reviving or reproducing it. The Gothic mind aims at constant 
 progression ; is not satisfied with retracing steps already trodden ; 
 it will not work readily in the trammels of imitation ; it will rather 
 attain new glories than strive to recover those that have passed 
 away. We shall find that we have great architects, as well as 
 great masters in every other branch of art, if we do not suffer them 
 to be cramped with needless restrictions. The old spirit is not 
 dead in us ; nay, it never was more active. Research scarcely 
 owns a limit ; science has unveiled marvels surpassing the fictions 
 of Northern Mythology ; a summary of its triumphs would now 
 be a puerile declamation on familiar facts, as it would formerly 
 have been thought to be, upon impossibilities and absurdities. 
 The higher our aims, the fairer is the prospect of success. In 
 every other matter, while we honour and value the works of our 
 predecessors, we make use of them as the groundwork of further 
 acquisitions ; we continually build upon them, instead of merely 
 striving to attain to the same elevation. If a system becomes 
 obsolete, we do not attempt to revive it, unless we find upon exa- 
 mination, that a recurrence to it will give us a fairer start in the 
 pursuit of truth. Our reverence for great names does not make 
 us look at any perfection hitherto attained as a standard or limit ; 
 it rather urges us on to a higher perfection ; we feel that all we 
 have done in science and art is but an advance towards the 
 truth, not a realization of it. We may indeed start anew from a 
 given point, if by so doing we may hope for new and great results ; 
 but this is a very different thing from taking as our standard of 
 excellence some point that has been already reached. It is true 
 that the desire of improvement and progress has often led to de- 
 cline and fall ; the movement may for a time be downwards, but 
 not backwards ; the Gothic mind cannot, in its very nature, be 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 197 
 
 either stationary or retrograde. It is because we have Gothic 
 blood in us, that we cannot revive Gothic architecture. 
 
 We are not, generally speaking, apt to confound a language 
 with the thoughts or truths which are expressed in it ; or to give 
 to one the honour which is due to the other. We do not think it 
 necessary to make Hebrew or Greek our vernacular tongue, because 
 the great truths of our religion have been set forth in them ; we 
 merely require that they should be preserved, studied, understood, 
 and critically examined by the learned, not for the purpose of re- 
 vival, but that nothing conveyed to our minds in the sacred records 
 may escape us. Nor does our admiration of the great poets, his- 
 torians, or philosophers of antiquity stimulate us to the revival of 
 dead languages, though the scholar may exercise himself in classi- 
 cal compositions to his own profit, and by doing so, cannot fail of 
 obtaining a clearer comprehension of those writers for whose sake 
 principally he studies the language. On the same ground we 
 ought to study Gothic architecture ; we cannot fail of learning 
 much from it, and expanding our own minds, but our compositions 
 in it should be looked upon rather as exercises, undertaken for the 
 sake of giving us a clearer comprehension of the system, than as 
 attempts at revival. 
 
 We must always look upon Gothic architecture as one of the 
 noblest developments that the world has ever witnessed ; we can- 
 not but regard its productions with an admiration almost amount- 
 ing to reverence ; we cannot consider its principles without being 
 struck by the skill, science, and art, by which they were established 
 and carried out. But if we are to revive the style, we must come 
 into the field as bold inventors, not as tame imitators ; ready in- 
 deed to avail ourselves of all that has gone before us, but anxious 
 to improve and correct it, wherever this is possible. We must 
 add new elements, new combinations ; we must refine, reform, 
 reconstruct, according to our present wants, ideas, or appliances. 
 We must not be content with expressing our thoughts in a decayed 
 or antiquated language, but either find or make a new one, or so 
 reinforce an old one with new life and vigour, as to render it the 
 
198 
 
 AECHITECTURAL STUDIES IN FRANCE. 
 
 best exponent of our ideas. If all our imitations hitherto are only 
 a step to further advancement, the time and labour bestowed upon 
 them are not perhaps thrown away ; but I much fear that correct 
 imitation is too much looked to as an end, and that we are becom- 
 ing content with a mere copied perfection. 
 
 I will mention one point, out of many, in which, if we are to re- 
 vive Gothic architecture, we ought to improve upon the models of 
 our ancestors ; I do not mean to say we shall find it easy to do so. 
 We have before us three different styles of window-tracery, strongly 
 marked and distinct, though at certain transitional periods running 
 into each other ; — the geometrical, the flowing, and the perpendicu- 
 lar. Now if we can strike out a system which will avoid the harsh 
 and often ungeometrical adaptations of geometrical forms, and the 
 ungraceful subordinate openings, which occur in the first of these ; 
 the apparent weakness and indecision which we too often have to 
 condemn in the second, and the stiffs and meagre repetitions of the 
 third, we shall have gained a step. But we shall not effect this 
 by merely copying those transitional instances in which the faults 
 of each of the three different systems are softened away, though 
 these are often extremely beautiful and instructive, and our thanks 
 are due to Mr. Sharpe^' for pressing them upon our attention ; we 
 must work out some new and decided system, combining such 
 beauties of the old ones, as bear combination, and guarding against 
 their defects, taking care also not to fall into others still greater. 
 Whether this is possible or not, I cannot say ; perhaps time will 
 determine. 
 
 In any case, whether we revive or invent anew, let us endeavour 
 to go beyond our predecessors, let us aim at some standard of per- 
 fection above any which they reached ; we may not succeed, but 
 the endeavour, made earnestly and honestly, must advance us. 
 Let us learn to combine, in the highest possible degree, mechanical 
 
 * " Rise and Progress of Decorated Tracery in England." I must here express ray 
 acknowledgments to Mr. Sharpe for much information relative to the subject of my 
 present volume ; in fact, for my introduction to some of the most interesting examples 
 I have noticed. 
 
ARCHITECTURAL STUDIES IN FRANCE. 
 
 199 
 
 excellence, grandeur of composition, propriety of ornament, and 
 convenience of arrangement. Let us study how we can refine 
 without enfeebling ; how we can secure a style like the Gothic 
 spirit, at once permanent and progressive ; permanent in its prin- 
 ciples, yet progressive, in the constant development of new excel- 
 lencies ; ready to adapt itself to all contingencies ; endued with life, 
 growth, and movement, yet containing no hidden element of de- 
 cline or debasement. I may be describing an impossibility ; but 
 if we try to reach it, we shall be sure to do something. 
 
 Near Boulogne. 
 
APPENDIX. 
 
 At the time I wrote my concluding observations in this volume, I had not 
 read Mr. Ruskin's admirable chapter on the Nature of Gothic, in his second 
 volume of " the Stones of Venice," to which I could not have failed to make 
 frequent reference. In this he points out the different elements of character 
 in Gothic architecture, one of which, to go no farther, appears to be an in- 
 superable bar to its revival in the present day — namely, Savageness." This 
 is an element that can neither be recalled nor imitated. It can only belong 
 to an art, indeed "a state of society, just emerging from infancy. As long as 
 it stamps the work of a rude and uninformed, but vigorous and imaginative 
 mind, it is grand, noble, and full of interest ; but an elaborate reproduction 
 of it is a manifest absurdity. No style has ever had the value of the real 
 mediaeval Gothic, as bearing the impress of original genius, and the peculiar 
 character of an age. Its savageness is, as we see, a mark of reality, life, 
 and energy. It shows the barbarism from which Art was emerging, and 
 also her resolute determination to emerge from it. If we imitate or attempt 
 to give the effect of this savageness in the present day, we show merely our 
 determination to disfigure our works with patches of the barbarism that 
 clung to this or that century or phase of art, just as a pretended sportsman 
 soils his clothes with specimens of mud from the country over which he 
 would make you believe he has been hunting. — And while we condemn, as 
 unrealities, imitations of stone and marble, we tolerate this aping of anti- 
 quity in a modern church. 
 
 But the different phases of art are not to be revived, exhibited, obscured, 
 transposed, or changed, at pleasure. It is not an easy thing to recal what 
 is past, when we fancy we like it better than what we have among us at 
 present. 
 
 Art, if it be a thing of life, must be in a constant state of advance, whether 
 it be to maturity and excellence, or to decrepitude and decay. We cannot 
 give it a retrograde movement; the attempt will only be attended with 
 ridicule and failure. We may possibly retard its fate for a time ; we may 
 ensure it an honourable old age and a glorious death. We may embalm it 
 with our regrets, and do reverence to its memory ; but we cannot recal it to 
 life, reproduce it, as it was, with the same features, nerves and sinews. The 
 Phoenix must die and be consumed. A more beautiful offspring may arise 
 from her ashes, and may resemble and remind us of her lost parent ; we 
 may trace some of her lineaments, we may witness some of her transmitted 
 energies, but the identity is gone ; the new life implies a new individuality. 
 The lost is not restored, but succeeded. The generations of man may see 
 
 D D 
 
202 
 
 APPENDIX. 
 
 hundreds of successive Gothic styles spring up and decay, but none of them 
 will be the Gothic of the thirteenth and fourteenth centuries ; and they will 
 be attained, not by retracing our steps, and professedly imitating what is 
 past away, but by constant progress. Any thing that is gained by a retro- 
 grade movement will prove to be without life, value, or reality. There is 
 a vast difference between the imperfections of a man who is doing his best, 
 and constantly striving to do better, and those of one who purposely sets 
 before himself a lower standard than has been reached, and that he knows 
 can be reached. We look upon the former with indulgence and respect ; 
 on the latter with contempt. — What shows earnestness in the one is a proof 
 of affectation in the other. And therefore the savageness which Mr. Ruskin 
 points to as the first and most characteristic feature of genuine Gothic, and 
 as giving it the true stamp of nobility, is just the one that we cannot, and 
 must not attempt to, attain, unless it be by setting up a higher standard than 
 any hitherto reached, and then the same causes may produce the same 
 effects, and still more noble and powerful. 
 
 If art is progressive, an increasing eagerness for perfection must be a 
 necessary and natural condition. An attempt to check it is an attempt 
 to check nature. Unquestionably it contains not only the nourishment of 
 excellence, but the germs of decay. We all know that there is a point be- 
 yond which refinement will detract from beauty and power ; but if art and 
 society are in a constant state of progress, they will not stop at that point, 
 they will either proceed till over-refinement has weakened them beyond the 
 power of self-support, or else they will transfer their care to some branch 
 which has not yet received it to the same extent. And when these changes 
 have taken place just at the right time, and been made just in the right di- 
 rection, an excellence has been reached little short of supernatural. And 
 observe, what has been already gained is not given up ; no step is retraced ; 
 refinement of touch has not been abandoned in the superior aim at refine- 
 ment in composition, nor this again in refinement of sentiment. Greatness has 
 been achieved, not by falling back, but by pressing forwards ; by gaining one 
 position in advance of another, and losing none. And we shall not achieve 
 it by retreating upon the positions occupied by our predecessors, high as 
 they stood above those we now hold; but rather by pushing on to others 
 in front of us, perhaps higher and stronger than have been yet attained. — • 
 We have a rich inheritance of achievements, of instruction, and example ; 
 let us turn them to account as we best may, but never forget that the scene 
 of our exertions and successes lies in the future, not in the past. 
 
 It is not impossible that our present attention to science, in which (always 
 excepting the article of nomenclature) over-refinement is not very easily 
 reached, may leave Art to recover a more healthy tone, or at least draw off 
 to other quarters much of that demand for nicety which has begun to operate 
 injuriously. But she must begin by thinking up to her work, not by working 
 down to her thought. She must embody conceptions worthy of the elabo- 
 
APPENDIX. 
 
 203 
 
 rate workmanship forced upon her. This latter may indeed subside into a 
 certain degree of simplicity, but never rudeness ; she may more and more 
 dispense with ornament, but its rarity will never be allowed to suggest 
 either poverty or want of care. We must give up the hope of recovering 
 the element of " savageness," till the revolution of art and society renders 
 its introduction proper and natural. 
 
 But you may ask, are we to dismiss, as unfit for the present age, an archi- 
 tecture at once noble and impressive in its aspect, and admirable in its mecha- 
 nical properties ? I cannot help fancying that Mr. Raskin, in recommending 
 the Venetian Gothic, is influenced by much the same feeling that has induced 
 me to point to the Southern Romanesque and the Angevine ; an impression 
 namely that these may be adopted and brought into use on the score of their 
 architectural merit and convenience ; while the Northern Gothic, though by 
 no means deficient in these qualities, must always be looked upon as the ex- 
 pression of a particular period of history and a certain national character. 
 And therefore, in adopting the one, we are simply without affectation availing 
 ourselves of the taste and science of former generations; in imitating the 
 other, we are aping their manner and spirit, and assuming a character that 
 does not belong to us. Mr. Ruskin assures us, that the Gothic houses are the 
 most convenient and comfortable of any in Venice ; and from his description I 
 take it they admit of the very highest refinement of decoration ; they seem in 
 fact to have very little of the element of savageness. Much in their construc- 
 tion might unquestionably be introduced with advantage in our street archi- 
 tecture, for it is better that a house should be, and appear to be, supported by 
 columns and arches of stone, than by sheets of glass, with, at rare intervals, 
 an iron rod, as we see in the whole frontage of Oxford Street. And though 
 I have no doubt a palace in Venice is most thoroughly Venetian, still I can 
 perfectly conceive its being translatable into excellent English, and unme- 
 diaeval English too. I look upon an Angevine church in much the same 
 point of view; namely, as being capable of an aspect not suggesting the 
 imitation of another age or people ; as admitting of great refinement in de- 
 coration ; as combining dignity, solemnity, and convenience ; and as mani- 
 festing sound construction and durability. It will still have to be translated 
 into English, good modern English ; and I question whether the occasional 
 introduction of a foreign type may not do good service, in throwing the 
 architect more upon his own imagination and invention than would be the 
 case if he always had his model at hand to apply to on an emergency. 
 
 The constant aim at perfection leads to the love of high finish. This is a 
 merit that every one can appreciate, and there are many who can appre- 
 ciate no other. An increasing demand for it is, as I have said, a natural and 
 necessary condition in the progress of art, even if it tends to hasten its 
 decay. We may be sorry for this, but we cannot help it. I confess I do 
 not regard its effect upon the working classes with exactly the same feelings 
 as Mr. Ruskin. I happened to be reading his chapter, on the Nature of 
 
204 
 
 APPENDIX. 
 
 Gothic architecture," at a railway station, (let me assure him that his works 
 suggest far too great an abundance of ideas to allow the reader to feel in 
 the slightest degree such interruptions as are incidental in travelling), when 
 an excursion train came in with a throng of passengers, eager, bustling, and 
 happy, bent upon enjoyment; honest, hearty, healthy enjoyment, which 
 might refresh them after their day's work, whatever it might have been, and 
 brace them up for the morrow's. Now, thought I, if these are the people 
 who are made to work like machines for the sake of that accuracy and care- 
 ful finish which the spirit of the age demands, at least they have some sort 
 of remuneration. For though the production of glass beads of an exact size 
 and shape, or the polishing of a precious stone to its highest brilliancy, be 
 not necessary to happiness, yet they are only part of a system to which we 
 owe that accurate adjustment of machinery which brings under our control, 
 and modifies for our use, the highest powers in nature. The steam-engine 
 calls into play not only bold invention, but careful mechanical execution. 
 Without this, the work of the designer would be useless. 
 
 It may be said that art and science are different things. Very true ; but 
 I doubt if it would be easy to make a whole people see and feel the differ- 
 ence. When the value of precision is constantly set before our eyes in 
 matters of science, when the lens or reflector must be polished to the nicest 
 surface to enable us to detect a group of stars in a blank space of the hea- 
 vens, when the most accurate proportions must be observed to ensure the 
 success of a chemical experiment, when the imperfect action of a crank or 
 wheel might endanger the lives of hundreds, when, I say, we are every 
 minute impressed with the extreme value of care and refinement in scientific 
 works, it will not be easy to satisfy us with a rude and savage condition of 
 art; we shall rather tolerate what is feeble and ineffective, than what is 
 rough and imperfect. 
 
 But is the workman necessarily turned into a mere machine ? His hands 
 may be, but it does not follow that his mind is also. Nay, the more purely 
 mechanical the work, the more completely may the mind be emancipated 
 from it. A lady employed in hemming a pocket-handkerchief is not engaged 
 in a task that calls forth much mental energy. For that very reason her 
 mind is unconstrained by the work, free to seek amusement or instruction 
 elsewhere. She may join in conversation, or listen to reading, better pro- 
 bably than if her hands were unemployed. Even when a certain degree of 
 thought and attention is required, it does not of necessity engross the mind 
 altogether. If a man's work is unintellectual, his mind will find some means 
 of escape from it ; it will not suffer itself to be degraded into a machine ; it 
 will leave the hands to their own work, and they will, in most cases, do it 
 all the better. 
 
 And therefore, if the workman, in compensation for work calculated in 
 itself to expand and invigorate the mind, receives the means of employing 
 it profitably during its absence from the workshop, in more accessible re- 
 
APPENDIX. 
 
 205 
 
 sources of art, literature, and variety of scene, I cannot think the balance 
 altogether to his disadvantage. 
 
 But this natural and necessary condition in the progress of art may bo 
 .shown to play no unimportant part in the economy of the world. We de- 
 mand high finish and accurate reproduction of ornament ; we require at the 
 same time a large and rapid supply ; and we must have it at the lowest rate. 
 We therefore discard the human machine, (for such I admit him to be as to 
 his hands, though not his mind,) and we adopt the inanimate machine. The 
 steam-engine does the work of hundreds of men ; multitudes are deprived 
 of the means of finding employment and earning their bread in their own 
 country. Meanwhile Providence opens other fields for activity and exer- 
 tion. A stimulus to the lagging emigrant is afforded in the discovered 
 wealth of a California or Australia. Fraught with the Gothic spirit of their 
 fathers, the swarm rushes forth to conquer their new territory, as in former 
 ages upon the wild and uncivilised tracts of Europe ; the young ofF-shoot, 
 as well as the parent stock, each thrown upon its own energies, receives 
 strength from the separation. The laws of society are spread into distant 
 lands, and with them the higher laws and revelations of Christianity. A 
 light is kindled, and gradually spreads around, in the regions of darkness ; 
 and the course of events ever points to the completion of the glorious Pro- 
 phecy, " The earth shall be fidl of the knowledge of the Lord, as the waters 
 cover the sea." 
 
 La Palud. 
 
HINTED BY C. WlIITTINGHAM. 
 
I 
 
GETTY RESEARCH INSTITUTE 
 
 
 
 
 
 
 3 3125 01360 0834 
 
 r