Having conducted you to Rheims, I now proceed to give you an account of what I saw there, and if it should contain a few digressions as long as those in my last, I hope you will forgive me, for this Gothic Architecture offers continual temptations to lead me out of the direct road. I shall begin, not with the cathedral, but with a much more ancient building, which is supposed to have served as a cathedral before the present edifice was erected. The church of St. Remi is said by Whittington to have been dedicated in 1049. I know not whether he mean the nave or the choir, but they are certainly of different periods. The Recueil des Abbayes, &c. says it was built in the time of Charlemagne, and consecrated by Leo IX., who was pope from 1048 to 1054; the nave, except the vaulting, is much in the style of St. Germain des Prés, at Paris, but the pillars are of various shapes, and do not seem the result of one design. There are two stories of aisles to the nave, and to the straight part of the choir, and above these are the principal windows of the clerestory, with semicircular heads, and over them a range of circular openings, which I do not recollect to have seen elsewhere. The buttresses, externally, are alternately semicylindrical with a small projection, and rectangular with a very considerable one; the first are parts of the original structure, and evidently denote the roof to have been of timber, and not vaulted; indeed all the vaulting appears to be posterior to the walls and piers; the latter were probably added at the same time with the vaulting which rendered them necessary. The middle of the western front is a restoration, in which many old parts have been re-used, and the fragments of marble and granite render it probable that the spoils of some Roman building were employed in the ancient edifice. The two old towers remain, the southern doorway, and the window over it are beautiful specimens of what I have denominated in a former letter, the third style of Gothic; and are probably of the fourteenth century. The choir is of the first style, very much resembling that of Nôtre Dame at Chalons, and though this church is certainly inferior in general effect to the one just mentioned, yet some of the partial views it presents are, I think, superior to any thing there. The flying buttresses of the choir are supported, on their first separation from the building, by a little column; and a narrow gallery, which surrounds the clerestory, passes between these columns and the body of the church. The same disposition prevails at Nôtre Dame at Chalons, at Amiens, and at the cathedral at Rheims. There are some granite shafts of columns in the church, which have perhaps belonged to an ancient temple.

Under this church is a crypt, where we are shewn the tombs of Clothaire the First, and of Sigismond, king of Burgundy. The former died in 561, probably at Soissons; the latter was thrown with his wife and family into a well at Orleans, and we should not certainly expect to find him in the same chapel with one of his principal enemies at Rheims. A simple vault, or succession of vaults of small dimensions, can give us no internal evidence of the time of its construction.

And now, in order to preserve something of a chronological order, let me transport you from Rheims to Mantes, where I have since seen a church which is a puzzle for the antiquaries. Whittington says that it was built by Eudes de Montreuil, and I understand him to quote Millin for the assertion; but I cannot find the passage in that writer. On the contrary, Millin tells us that it was built by the same architect who built Royaumont. Now, Royaumont was finished in 1228, sixty-one years before the death of E. de Montreuil, which took place in 1289. The first church which was erected at Mantes is said to have been built in 865, but it was destroyed by William the Conqueror; and we certainly see at present no remains of any edifice of that date, yet Millin seems inclined to consider the northern doorway of the western portal as part of the original construction. Altogether this church has the appearance of a building that has undergone considerable alterations at different periods. In what I take to be the original work, the nave has two stories of aisles, the end of the choir is round, and the windows are without tracery. In the clerestory and in the lower aisles the windows are pointed; in the upper aisles they are entire circles; all have Saxon ornaments externally, and are quite plain within. All these circumstances indicate a style prior to that existing at the beginning of the thirteenth century, which is the date generally assigned to the edifice, under the auspices of Blanche of Castille, mother to Louis the Ninth. Two towers have been added at the west end. The southern, which is said to be the most ancient, is a very strange composition of slender Gothic. The upper story but one is surrounded by a colonnade, if the expression be admissible, of two ranges of columns, one above another, without either arch or architrave between them, but merely connected with the wall by a stone slab on each capital of the lower range. The upper range supports arches, on which rest several unconnected slabs, steeply sloping, wrought into scales, and conducing neither to the beauty, the strength, nor the shelter of the edifice. The north-west tower, built, according to the tradition of the place, three hundred years after the other, is much less light in its construction, and not much more handsome. It would rather appear more ancient than posterior to the first, if there were any difference, but the summit is comparatively modern, and this has probably given birth to the opinion that its date is so much posterior.

Many of the chapels appear to have been built in the thirteenth century, at which time the vaulting of the nave and choirs was perhaps added, and some of the windows of the upper aisle were altered from their original circular form, into that which they now bear. Later still (in 1405) the porch of the southern door of the western entrance was erected. It is very beautiful, and is the only part of the edifice which is so. Besides these, are a great many other incongruities which are probably assignable to different periods. The vaulting is with oblique groins, as at Beauvais. You know, that in oblique groining, the piers are usually alternately larger and smaller. In this church, the direct arch between the greater piers seems to be formed on an equilateral triangle, or nearly so, rising on the capitals of the shafts; but on the smaller piers the perpendicular line is continued considerably above the capital, and the direct arch between them is consequently very obtuse. Perhaps it was this whim which attracted the admiration of Sufflot, who is said to have been lost in astonishment at the hardiesse of the vaulting, although the nave is only 34 feet wide. The boldness of the architect is however sufficiently conspicuous in other respects, for the piers of the chevet are only 1 foot 11 inches in diameter, to support a vaulting which rises 102 feet 6 inches (English measure) from the ground. One of them is consequently crippled, and has been banded and supported on every side with iron. M. Gabriel, one of the companions of Soufflot, in his examination of this church, contends that the six columns of the chevet might all be cut away, and that nevertheless, by the scientific disposition of the stones, the upper part and the vaulting would remain secure: this indeed would be something wonderful. All the arches of the nave and choir are pointed.

So much for Mantes, but I have still to trespass upon your patience before I bring you back to Rheims, with an account of a building, which from its early date, its peculiar architecture, and its great magnificence, I consider the most interesting specimen of the Gothic style in France, or probably in the world. I had conceived, from what Whittington says of it, pp. 54, 55, 57, that I should find a building of the Norman taste, but this is not the case; Chartres is decidedly Gothic, of a peculiar manner indeed, but such as one would suppose posterior to all the three edifices above described. There are some additions to the original building, but these are extremely well marked, and the mass of the edifice is so clearly the result of one design, and the production of one period, and the time of its erection is so well authenticated, that it takes place of all other cathedrals in antiquarian interest, and yields to few in beauty. Let me relate to you what information I have been able to pick up on the spot, it may help you to form some idea of what one has to wade through to arrive at any satisfactory results in the history of French Gothic. I first bought a little book of the history of Chartres, in order to obtain the dates of the different parts, but I learnt from it little of what I wanted to know. The author begins by telling us, that the ancient nations of Gaul were the most religious people in the world, and that the innocence of their lives, and the holiness of their priesthood, made them worthy to participate in the most important revelations, and to have the future incarnation of the Word shewn to them, long before it was accomplished. There were, he says, three classes of people to whom this communication was made, the Magi, the Sybils, and the Druids. The first learnt it by their knowledge of astrology, the second received the gift of prophecy in recompense of their virginity, and the Druids knew by a prophetic spirit rather than by any fortuitous prediction, that a virgin would one day bear a son for the salvation of the world; and they consequently raised altars in several places, inscribed virgini parituræ, (did they write Latin?) and amongst others there was a very celebrated one at Chartres.

When afterwards Christianity was preached in these parts, there were three circumstances of similarity in the Christian and Druidical rites, which greatly facilitated its progress. The worship of the virgin, who, according to their traditions, was to bring forth a son; the offering of bread and wine, usual in their sacrifices; and the adoration of the Tau, that is, of the cross. The Christian service was performed at the ancient altar of the virgin, and crowds thronged from all parts of the universe to present their offerings. The present cathedral is built over the grotto where this altar formerly stood.

The most famous relic here was the shift of the virgin, which was stolen from a Jew widow by some pious patricians of Constantinople. It was taken from them by an emperor, whose piety was, I suppose, of the same sort, and presented to Charlemagne, who brought it to Aix. It was removed thence by Charles the Bald, and given to the cathedral at Chartres. This relic has of course performed abundance of miracles, but most of these are what would be called by many people in England special providences. And, if amongst us, we had no division into sects, would not these special providences soon become to be considered as miracles, and alleged as proofs of the truth of particular doctrines? We have no reason then to pride ourselves on our freedom from such superstitions, as it depends on circumstances over which we have no control, but much cause to be thankful, not that we have this or that form of worship, but that we have the liberty of thinking for ourselves on religious subjects. This book is not an antiquated work: it was printed in 1808, and may serve to prove, that whatever injury the revolution may have done to religion in the minds of the French, it has by no means rooted out superstition.

There is a public library at Chartres, containing between twenty and twenty-five thousand volumes, and I there found a history of the city deserving more attention; although even in this, the author employs 200 pages in telling us what happened before the arrival of the Romans. This folly seems as strong in France as it is said to be among the Welsh; and many of the local histories are prefaced by an account of Samothès, the son of Japhet, who first peopled Gaul, and a long series of princes, who gave successively their names to the Druids and Bards, to the Celts, the Gauls, and even to the Francs. These, instead of being a German nation, were the subjects of Francus, the son of Hector, called Astyanax and Scamander by Homer, who came to France and married the daughter of Rhemus, king of Rheims. It was only a return to the land of his forefathers, for Dardanus, the founder of the Trojan line, was a Frenchman.

But to return to this History of Chartres, which was written by a M. V. Chevard, and printed at Chartres. It assures us that the old cathedral was burnt in 1020; that Fulbert, who was then bishop, began the present edifice almost immediately, but that it was not completed at the time of his death, in 1028, although it appears to have been considerably advanced. We are even told that it was finished before the middle of the eleventh century, but this word is often used very loosely in the accounts of Gothic buildings. Matilda, wife of William the Conqueror, who died in 1083, covered the main body of the edifice with lead. The basnef, the towers, and the west front, were finished in 1145. The south porch was added in 1060, by Jean Cormier, or Jean Le Sourd, physician to Henry I. of France. The famous steeple is one of those of the western front. The place was formerly occupied by a wooden one, but this being burnt in 1507, gave rise to the present, which rose almost immediately from the ruins. I had heard so much of the height of this steeple, that the first view of it disappointed me in this respect, but the great elevation of the body of the building in the French churches, effectually prevents any such extreme impression of height as is produced by the spire of Salisbury Cathedral; or, at least, the elevation must be indeed enormous to occasion it. The vaulting of these edifices is more lofty, and the space between that and the timber roof is much greater than is usual in England; indeed, sometimes in our churches the direct tie-beam is omitted, in order to bring the vaulting absolutely into the roof. The roof itself is also higher. Altogether the ridge of the roof at Chartres must be full 50 feet higher than that at Salisbury, and 50 feet added to the mass of the building, and taken from the spire, will greatly diminish the apparent elevation of the latter. The height of the edifice is very much lost in its bulk. It even forms a base for the lighter part, and in some degree a standard by which to measure it. Afterwards, however, in walking round the town, and seeing the cathedral in different points of view, I gave to the height its full value. The impression was not that of a very high steeple, but of a very lofty church: an effect greatly enhanced by its fine situation, on the summit of a hill, with the town collected at its foot. The whole height of the new spire is 403 English feet. The upper part of the tower and the spire, are of the most light and beautiful work imaginable. The ornaments are executed with the greatest delicacy, and in entire relief, the stems of the vines, and the bunches of grapes which enrich the mouldings being entirely detached, and the work suspended merely by the extremities of the leaves; and all the veins and ribs are shown as if they were to be seen at hand, instead of at an elevation of 300 feet. Even parts, which cannot be seen at all from below, are finished with the same care. The staircase, by which one ascends this spire, forms a little tower of itself, also of open work, quite independent of that which supports the spire. The opposite spire is much more solid and simple in its form, and seems to be part of the edifice of Fulbert; its height is 365 feet, and its appearance is more like that of Norwich than any other English spire I am acquainted with, but the resemblance is not at all continued in the tower which supports it. There are several pinnacles rising above the base of the spire, and the whole composition is more Gothic than at Norwich. As the cathedral was two or three times destroyed by fire, before it was erected in its present form, i. e. before the time of Fulbert, it is possible that some of the lower part of this tower belonged to an earlier edifice; the pointed arch is however exhibited in it. The whole western front is very beautiful. The porch is ornamented with statues and columns, as at Rheims and Amiens, but not in such profusion, nor is it so deep. The execution is also more stiff and rude, and the resemblance is probably much stronger to the western doorway of St. Germain, as it existed before the revolution, than to either of these buildings. Some of the statues are merely stuck to the little columns behind them, under others there is a projection to receive the feet, but very small, and apparently insufficient. Over them are small canopies, which are likewise attached to the shaft of the column. The capitals of these columns, instead of foliage, are formed of little figures, with canopies over them, surmounted by what have the appearance of little models of large buildings. It is remarkable that, although the arches of this doorway are somewhat pointed, yet the architecture represented in the models never is so. In the south portal, on the contrary, which I shall describe by and by, where these models are tenfold more abundant, a considerable number have pointed arches. This circumstance seems very extraordinary if this western doorway was built, as is asserted, eighty-five years after the southern porch. Over the great door is a triple window, the middle division being the largest and highest, the only instance of this disposition in the building. Above this is a magnificent rose window, but of simpler arrangement, and a larger portion of solids than we find in those of a later date. The windows of the nave and choir are also terminated by a rose, but with this singular difference, that in the great roses the exterior is ornamented with mouldings, while the internal faces are plain; in the smaller ones, on the contrary, the internal faces are moulded and the outside is plain.

The southern porch is very curious on many accounts. It was built, as I have already said, in 1060, by Jean Cormier, physician to Henry I. This date is important, because it seems exceedingly well authenticated, and the addition of the porch proves the church, if not finished, yet to have been in a state of great forwardness at that period. There are openings, not arched, but square-headed, combining all the parts of this porch into a sort of open portico. It abounds with detached shafts, of which there are none within the church, with large and small figures, and with models of architecture, in some of which, as I have already remarked, the pointed arch is exhibited. The arches of the porch itself are all pointed. The footstools of the figures are usually themselves grotesque human figures, and many of them with crowns. These statues, and the canopies over them, are much better managed than in the western front. They are rudely finished, but the labour bestowed on the lace of some of the garments shews that this rudeness was the effect, not of negligence, but of want of skill. The foliage of the capital of the columns spreads over the underside of the canopy. Above the porch is a range of five windows, of equal size and height, and over these a rose window. In the transept at Amiens, the angular spaces between a similar range of arches, and a circle above them, are opened, and form additions to the rose window; here they are closed, and no attempt is made to unite them. There is, as usual, a gallery in front of the gable, and at each end of this gallery is a small octangular tower, surmounted by a spire. This seems to have been a common mode of finishing in the early French Gothic; it occurs at a church in Soissons, which bears all the marks of antiquity; and in other places. The general opening of the windows, both in the clerestory and aisles, in this church, is round-headed, but they are divided by a large plain mullion (such as I have never seen elsewhere) into two parts, each of which has a pointed arch, and over them is a rose.

It was, perhaps, part of the original design to have a tower on each side of each end of the transept, and one on each side of the choir, but the parts which are now exhibited of these towers, above the roofing of the church, seem to be of later date; none of them are finished, and what has been performed of the two latter does not correspond with the work of the other four. These are ornamented with numerous little shafts, extremely long and slender, most of which are united to the solid masonry, but those at the angles are detached, in order to give an exaggerated appearance of lightness. I imagine them to be posterior to those of Nôtre Dame, at Paris, but earlier than those at Rheims; but among the various efforts which ultimately completed the first mentioned of these churches, we cannot determine during which the towers were built, and the priority of date is left in great uncertainty, for the style of building is not decisive, or rather, I have not sufficient knowledge to be able to determine it from that character. The northern front of the cathedral at Chartres presents a similar style of ornament, but without the projecting porch, which makes so important and interesting a feature on the south.

Chartres is very rich in painted glass; in this respect it far exceeds any other cathedral I have seen; the colours are deep, without losing their brilliancy, and the light is stronger than at Rheims, although the windows of the aisles, with only one or two exceptions, are painted, as well as those of the clerestory. The glass is said to be half an inch thick; I believe this is not much thicker than some of the old glass in York cathedral. Many of the windows contain escutcheons. This church is 461 feet long internally, and the vaulting is 113 feet high, the piers of the nave are composed alternately of octagonal pillars, with four circular shafts attached to them, and of circular pillars, with as many octagonal shafts attached to them. All the arches and the vaulting are pointed, except perhaps (and of this I am not sure), that the cross vaulting of the nave may be of circular arches. The construction of the roof has been much praised, but it is not good; the timbers are all small, and the trusses are very close together. At the point of the choir there is as usual a maître poutre of immense size, which you are told supports the whole roof, but which in fact supports nothing, being itself suspended by the converging rafters. There is a space of about six feet between the tie-beam and the top of the vaults.

The single story of side aisles, the polygonal end of the choir, the piers which support the groins behind it, and the windows of the choir single, and not disposed by threes, all unite to refer this building to the second style of French Gothic; which the greater massiveness of the work, and the presence of some circular arches in the towers, might otherwise render doubtful. The single story of aisles and the greater height of the building seem to indicate a later period than Nôtre Dame at Paris, but on the other hand the smaller windows, surmounted in the nave by a single rose, the more solid divisions of the great rose windows, and the style of finishing externally, announce an earlier stage of the art. If I had to estimate the date from the architecture, I should be very much puzzled by many peculiarities, either very rare, or not met with elsewhere; but on the whole, excepting a portion of the towers, I could not have placed it before 1150.

With good proportions, beautiful parts, and finely coloured windows, you will conclude that the whole impression produced is sublime; but I wish I had you here, where you would find some better proof of this, than the cold conviction of your reason. The people seemed very devout, and were all day long kissing the pedestals, and various parts of the decorative architecture, about a figure of the virgin, which is almost black. In this part of France the virgin is usually represented with a very dark complexion; and such is, I believe, the case with the most popular images of her in all Catholic countries. There is a labyrinth in the pavement which is said to be a league, measured along all its folds; a countryman applied to me to know if this was true. I told him it was impossible, and shewed him that the number of turns, multiplied by the length of the middle one, only gave 1320 feet, but he was determined to believe as his fathers had believed before him.

I must not quit the cathedral without mentioning the beautiful shrine-work which surrounds the choir, to see which is alone well worth a journey to Chartres. It consists of forty-five compartments, forming a sort of continued gallery, and contains in all about two hundred and fifty figures, each of three feet high. It is a very curious specimen, both for the extreme delicacy of the workmanship, and as a model of the last period of Gothic architecture in France. It is complete point lace in stone, and some of the threads are not thicker than the blade of a penknife. The style is rich and beautiful, or at least many parts are beautiful; but as a whole, it wants simplicity, and is inferior in design to the architecture of King’s College chapel at Cambridge, and perhaps even to Henry VII. chapel at Westminster; but the extreme intricacy of the multiplied ornaments in the last-mentioned building does not please me. In the work at Chartres the disposition of the masses is much more simple and intelligible, but the tracery and detail of the ornaments are even more confused. It is worthy of notice that the vaulting continues entirely simple, and without any trace of the palm-tree branching, exhibited in that of King’s College chapel, or of the still more complicated arrangement of that of Redcliff church at Bristol. This fine work is in two series, the first of which is said to have been executed with the surplus of the money raised for erecting the spire. It is precisely of the same style as that erection, if we make allowance for its greater delicacy, adapted to the different nature of the work; but no dates are marked on it: this forms the largest part. The second series exhibits some traces of the knowledge of Roman architecture, and has dates from 1523 to 1530. This is ornamented with arabesques in imitation of the Italian cinque cento. I observed two dates of a later period, T. Bovdin Mil vi^c xi, and a similar inscription of 1612, but there is no difference of style to account for them.

I was led by the accounts of Chartres to suppose I should find some vestiges of very high antiquity in the crypt under the cathedral, but I was disappointed; there seems to be nothing but what is coeval with the building, and the vaults do not extend under the whole edifice, but only under the chapels and side aisles. The people in this neighbourhood are more unfavourably disposed towards the Bourbons than those who live to the east of Paris. A woman observed that I was one of those who had brought back Louis XVIII. She had nothing to say against them or him, but the tones of her voice did not promise that she would say any thing for either. The conducteur of the diligence perhaps was not a Napoleonite. “Whether God or the devil, Napoleon or Louis XVIII. be on the throne,” he observed, “the laws should be obeyed. There were revolutions in France before this, of which they talk so much; for instance, in the times of Charles V., who drove the English out of France; and if the French were now as devoted to their country as they were then, these things could never have happened.” I could not be displeased with any Frenchman for a feeling of soreness at the interference of foreigners in the affairs of his country, however political circumstances may have required it.

The two churches last described, and that of Nôtre Dame at Paris, may be considered as belonging to a style of Gothic, intermediate between the first and second of those I have enumerated; and as I wish to give you a sort of historical series elucidating the progress of architecture, I shall here introduce some account of the French metropolitan edifice. This is said to have been originally founded by Childebert, in 522. It had 30 marble columns, and very large windows, according to the account left us of it by Fortunatus, a cotemporary poet.^[11] This description, however, has nothing to do with the present building, which was commenced in 1010 by Robert the Pious. After his death it was neglected, and little was done till 1165, when Maurice de Sully, a liberal and munificent prelate, filled the see of Paris, and to him we seem to have been indebted for the greater part of the edifice. He destroyed the old church of Childebert, which had existed till this period; and in the year 1181 the eastern part was so far advanced, that it was consecrated by Henry, the Pope’s legate, and by the bishop himself, who died the next year. Odo de Sully succeeded, and prosecuted the work with great zeal till his death in 1208, so that for forty-three years from the resumption of the work, it was carried on with spirit, and we must suppose a large portion of it was completed. Pierre de Nemours, who died in 1220, is thought to have finished the nave and western front. The last figure of a king exhibited in its galleries is that of Philip Augustus, who died in 1223, and this is one reason for supposing it finished in his reign, but it is not a very strong one. The south transept was not, however, begun till 1257, as we are informed by a Gothic inscription on the porch, and an ancient church of St. Stephen was then destroyed to make room for it. The present rose window was renewed on the model of the ancient one in 1726. The date of the north transept is unknown; it probably preceded the south; but its porch and chapels are assigned by Le Grand to the fourteenth century.

The front is heavy, but not so heavy as usually represented in engravings; I think this appearance arises in part from the square solidity of the towers, and in part from the horizontal lines being marked too strongly, a circumstance which always produces a bad effect in Gothic architecture. I have not been able to determine whether it was intended to crown these towers with spires: I am inclined to answer in the affirmative, but rather from analogy than from direct proof. According to Landon there were twenty-five statues of kings in the arches over the western porch, viz. thirteen of the first race, nine of the second, and seven of the Capetian. They entirely filled one range of arches and no more. Now there are two ranges of arches above the doorways in this front, the lower of which, according to the elevation given by the same author, presents twenty-four niches, and the upper twenty-six. Query, how many statues were there, and where did they stand? Felibien, in his plate of the elevation, which is much better than Landon’s, figures twenty-eight niches in the lower arcade, viz. nine in the middle, seven on one side, and eight on the other, and four on the buttresses. The upper arcade is a gallery not intended for statues, the middle part of which is open on both sides. The arches of the lower range have trefoil heads, and appear from below to be entirely composed of models of architecture. The canopies of the portal abound also with models of architecture, resembling in this, and in the style of sculpture, the south portal at Chartres. Perhaps the design of these, though not the execution, may be attributed to the time of Maurice de Sully, in 1165, but this brings the date a century later than that of Chartres. I wish very much to discover that the south porch in that cathedral was of 1160, instead of 1060, but I cannot persuade myself that the physician of Henry I. lived to build it a hundred years after his sovereign’s death. The Matilda mentioned as having contributed to the church, may be the widow of the emperor.

Whittington says, “The eastern end, which is triagonal and very plain, was probably one of the first Gothic structures in France (1168). This plainness, from a proper regard to uniformity, was maintained in the subsequent part of the building, excepting in the chapels, which are of later date;” this I do not comprehend; the eastern end is semi-circular, and is richly ornamented externally with slender shafts, and spires of different heights, which may perhaps have been added at the same time with the chapels, if these are indeed posterior, but assuredly they do not make part of them. It seems to me that those parts which remain without ornament have never been completed, for they exhibit abrupt terminations, which were not in the taste of the Gothic architects at any period. All the flying buttresses are exceedingly slender, and altogether the construction of Nôtre Dame may be considered as among the boldest, and most successful, existing in Gothic architecture; although even here we find some traces of the too great operation of the thrust of the arches of the side aisles.

On entering Nôtre Dame one is struck with the double range of side aisles and open chapels besides, making an entire width of seven divisions, instead of five, as at Amiens, or three, as in our churches. It is generally supposed, that if two dimensions of a building are great, they will appear of less magnitude if the third be great also. For instance, in a very large building, great height will diminish the apparent extent in the plan, great length will diminish the apparent width, and a narrow room will look higher than a wide one of the same height and length. Yet certainly the impression of space is much less at Nôtre Dame, than in the narrower and loftier edifice at Amiens. One of our travellers has estimated the size of Nôtre Dame as about half that of Westminster Abbey; and some non architectural friends with whom I have talked on the subject, thought that he perhaps underrated it, but that certainly the French building was much smaller than the English. Nôtre Dame is 416 feet long internally, and 153 wide: the length of the transept hardly surpassing the width of the nave and side aisles. Westminster Abbey is 360 feet long and 72 wide. The transept, indeed, is 195 feet long, but the whole internal area of the French building must be at least twice as much as that of the English. Whence is this very false estimate of its size? Does it depend merely on the injudicious arrangement of the parts, or is it in some degree to be attributed to a patriotic determination to find every thing best in our own country? Here are two stories of side aisles, and this double range, and the very slender columns which divide the openings of the upper, are in some points of view very pleasing. There are three arches over each of the larger openings below, united into one common arch; but the space included between the three smaller arches and the larger one is a blank wall. This has a very bad effect, especially as it is a part in which we are accustomed to expect ornament; indeed the arrangement of this gallery is inferior to that before noticed in Nôtre Dame at Chalons. The vaulting of the nave and choir is with oblique groins, as at Beauvais and Mantes. The vaulting itself, according to Millin, is only 6 inches thick.

With the cathedral of Nôtre Dame I conclude what I had in my mind to say to you of the progress of Gothic architecture, previous to its full development in the cathedrals of Amiens and Beauvais. Another proud specimen of architecture of that period is found in the cathedral of Rheims. It was founded, we are told, in 818, but I have some doubt whether the early structure thus spoken of was not the church of St. Remi, and not one occupying the site of the present cathedral. It was burnt in 1210, together with great part of the city of Rheims. A new cathedral was immediately begun, but the ancient crypt was left; now we are not shewn any ancient crypt at the cathedral, but there is one at St. Remi. The work went on with great rapidity, for the altar was dedicated on the 18th October, 1215, and the body of the church was finished in 1241. It appears probable that this finishing does not include the famous western front, which however was completed before 1295. Thus you see the bulk of the building was erected in thirty-one years; while at Paris two active bishops could not bring theirs to so forward a state by forty-three years of persevering exertion, although the foundations were previously laid, and probably a considerable quantity of materials prepared, and although the transept was not included. The size was not much greater, and the expense must have been decidedly less, on account of the inferior richness of the latter building. This difference is rather surprising, especially when we take into consideration, that the Parisian bishops had the support of the monarch. Of the portal, or west front, the plate in Whittington is the best I have seen, though it retains many errors of a large but very bad engraving, published in 1625. It must have been partly copied from this, or from some other, which may be traced to a common origin, but not without a reference to the building, because several mistakes are rectified, and the details are better given, though the drawing is on a much smaller scale. One important error is not to be attributed to the old plate; the octagonal turrets placed at each angle of the western towers, are not closed, but entirely open, consisting merely of the slender shafts, which are kept in their upright position by numerous iron ties. The part above the arches supported by these shafts is the base of an unfinished spire, and the whole summit of the towers is evidently of a temporary nature; even as a temporary finish, however, there neither are, nor apparently ever were, any fleurs-de-lys.

In the richness and magnificence of the external architecture, Rheims is superior to any other cathedral I have seen, and probably to any which has ever been erected. Whittington’s plate above cited will give a tolerably correct idea of the western front, but none of the effect produced by the same profusion, extended over the whole surface of a great building. I do not know whether the view of the back of the choir is not even more striking than that of the great entrance, the buttresses all terminating in little spires, all the parts running up into pinnacles, all subordinate to a spire, 256 feet in height, which crowns the rond point, and is surmounted by an angel of gilt bronze. I do not know, by the bye, whether this angel be of gilt bronze, but I know that such a piece of magnificence existed at Chartres, and my imagination, rather perhaps than my memory, pictures it here. Nothing but an angel would do in such a place, the situation is far too dangerous for that of any human being. This spire on the chevet, perhaps rather hurts than assists the general effect of the church, when seen from a distance, but after passing near the back of the choir, no one could wish it away, and if the spires in front, and whatever was intended in the centre, were completed, it would probably form an agreeable accessory from every point of view. All these spires and pinnacles are richly decorated, and what is more, the ornaments are highly beautiful, both in design and execution; the sort of plume which finishes some of the pinnacles, is one of the most graceful terminations I have met with. There are some trifling differences of detail in the corresponding parts, but the general form is always similar, and the character is uniformly preserved. None of these differences are distinguishable without examination.

Passing from the outside to the interior, the first circumstance which struck me was the obscurity of the nave, contrasted with the light of the aisles. The coloured glass of the former has been preserved, while that of the lower windows has very little colour. The opposite disposition of white glass in the clerestory, and coloured in the nave, would be preferable, yet this has a better effect than I should have expected à priori, and I conceive would even find advocates. It is probably owing to this arrangement that the coloured glass at Rheims seems to have little brilliancy. The whole length of the building is 466 feet, that of the clerestory 386, the width of the latter 47. The nave is 121 feet in height; the aisles I suppose about 54 feet, or something less than half: all the parts are well finished, but the interior has by no means the predominating beauty of the exterior. We may judge of details by rule, but the only true method of estimating the excellence of an architectural composition is by the sentiment it produces. I must acknowledge that this is in some respects, an uncertain criterion, as the impression produced depends in part upon the temper of the mind at the moment, and even on the feelings of the body. However, we may make allowances, and we may repeat the trial under different circumstances. It is on this ground that I pronounce the inside of the cathedral at Rheims to be inferior to that of Amiens or of Chartres. The capitals of the columns of the nave in this cathedral are of very full and deep relief; the foliage runs round the capital, and is often very gracefully disposed; this is a step towards the third style of French Gothic, the first and second having in general only detached leaves or figures. The construction of the roof is very curious: the architect seems to have intended to gain double strength by applying a king-post truss on each side of the timbers of a queen-post truss. The latter rises on the outside of the walls, the former on the inside, and its principal rafters meet in a point considerably lower than if they followed the direction of those of the other truss. The tie-beams are about 12 feet above the point of the vaulting. All the timbers are said to be of chesnut, and the proof is, that no spiders are found upon it. Over the great arches of the intersection are four semicircular arches, evidently intended to discharge the weight of a central tower or spire, from the pointed arches of the internal vaulting, and therefore proving the intention of raising such a tower. From the frequent mention we find of central towers in the descriptions of the French churches, it is probable that this intention was coeval with the design of the church. Of all these described stone central towers or pyramids, however, I have not had the good fortune to meet with one, and a large proportion of them seems to have been destroyed.

Besides its Gothic architecture, Rheims has to boast an interesting relic of Roman times. It is unfortunately built up in the modern wall of the city, and not easily seen, except in a general view of such parts as project from the face of the later work. Three columns and an arch are sufficiently visible, and parts of three other columns; and enough remains to enable one to make out the plan, and to shew that there were three nearly equal arches, and eight columns disposed in four pairs; the larger intercolumns are of course occupied by the arches, the smaller have niches and medallions; the entablature is entirely gone, but it is possible to creep into a vault in the thickness of the city wall, where we see near at hand the soffite of one of the arches, with the ancient stucco. The design was not very simple, but the execution is good for its purpose. The flutes of the columns finish square under the capital. The astragal of the necking has been cut; the capitals are too much wasted to form any decided opinion from them, but on the whole it appears to be a monument of a good period.

There is also a vaulted chamber, probably a sepulchre, which was discovered in 1738. The vault is ornamented with octagonal compartments and roses, in stucco, and the walls with painting. I did not see it, and only know it from a little printed description, for which I am indebted to the kindness of Comte Gregoire.