An entirely new state of affairs was inaugurated in 1066 by the Norman Conquest of England. A new aristocracy, new laws, and a new language infused new life and energy into every department of the State, and an age of unwonted activity and brilliancy superseded the lethargic misrule of the Saxon period.

In nothing was this more manifestly evident than in architecture. Instead of a barbaric and debased style, a real lithic art was introduced and adopted at once, on a scale of magnificence but little known even in France at that time. Almost all our great cathedrals were either rebuilt, or at least remodelled, at that time, and great monastic institutions were founded all over the country, demanding churches and buildings on a scale undreamt-of before that time. The impulse thus given lasted for nearly five centuries, till the Saxon element in the population again came to the surface at the Reformation; but during that long period it continued without break or drawback, and forms a style complete and perfect in itself,—imported, it is true, in the first instance, but taking root in the soil, and with little aid from abroad growing into a thoroughly vigorous and acclimatised style. So completely is this the case, and so steady and uninterrupted was its progress, that it is impossible to separate its various stages one from another, but it is proposed to treat it as one style and in one chapter in the following pages. In a larger work it might be necessary to divide it into parts, but within our limits it will certainly be found more convenient, as it certainly is more logical, to treat it as a whole.

Plans of English Cathedral Churches.

The most remarkable and universal peculiarity in the arrangement of English churches, when compared with those on the Continent, is their extraordinary length in proportion to their breadth. In this respect they seem to stand alone when compared with any buildings existing in other parts of the world. The ancients affected a double square; in other words, their temples were generally twice as long as they were broad. In the Middle Ages, on the Continent, this proportion was generally doubled. Practically the internal width was multiplied by 4 for the length. This at least seems to have been the proportion generally aimed at, though of course it was often modified by circumstances. In England the larger churches generally reached the proportion of 6 times their width for their length. Most of our cathedrals have been so altered and modified by subsequent additions that it is difficult now to trace their original arrangements; but Norwich exists in plan almost exactly as originally erected (A.D. 1096-1135), as will be seen from the plan (Woodcut No. 802). The nave to the west of the intersection is more than 4 times its width (70 × 295). The rectangular part of the choir is more than a square, and with the apse and its aisle, exclusive of the chapels, makes altogether a length of 410 ft. internally, or nearly 6 squares. At Peterborough and Ely the proportion seems to have been as 5 to 1 to the centre of the apse; but if there was a circumscribing aisle or chapel, the longer proportion would obtain. At Canterbury and Winchester, and generally in the south-eastern cathedrals, as built more immediately under French influence, the original proportion was somewhat shorter; but so impressed were the English architects with the feeling that length was the true mode of giving effect, that eventually the two cathedrals last named surpassed it. Canterbury (Woodcut No. 803) attained an internal length of 518 ft. while the width of the nave is only 72, or as 7 to 1. At Winchester (Woodcut No. 806) these dimensions are 525 and 82, or something less than 7 to 1, owing to the greater width of the nave.

It is extremely difficult to assign a satisfactory reason for this peculiarity of English plans. It arises so suddenly, however, in the English churches of the Norman age that it must have pre-existed in those of the Saxons; though why they should have adopted it is by no means clear. If these churches had wooden roofs, which was almost certainly the case, their naves might easily have been wider, and it can hardly have arisen from any æsthetic motive. As we now judge them, these early naves were badly proportioned for hearing an address from the bishop or prior, and as ill adapted for a multitude to see what was passing at the altar; but for pictorial effect they surpass everything erected on the Continent, unless with greatly increased dimensions of height or width. Whether, therefore, it were hit upon by accident or by design, its beauty was immediately appreciated, and formed the governing principle in the design of all the English cathedrals. It was a discovery which has added more to the sublimity of effect which characterises most of our cathedrals than any other principle introduced during the Middle Ages.

All the cathedrals above enumerated, indeed most of those which were designed by Norman prelates during the first half-century after the Conquest, were erected on very nearly the same plan as that at Norwich. Durham (1095-1133) was the first to show any marked deviation from the type^[98] (Woodcut No. 804). The nave and choir became nearly proportioned to one another, and for the first time we see a distinct determination from the first that the building should be vaulted. All this involved an amount of design and contrivance which entirely emancipated us from the Continental type, and may be considered as laying the foundation of the English style.

In addition to what was doing at Durham there prevailed an extraordinary activity in church-building in the North of England during the whole of the 12th century, owing to the erection of the great abbeys whose gigantic fossils still adorn every main valley in Yorkshire. As this part of the country was more remote from foreign influence than the South, the style developed itself there with a vigour and originality not found elsewhere; but its effect was appreciated, and when Lincoln was rebuilt, about the year 1200, the English style was perfected in all essential parts. This is even more remarkably shown, however, at Salisbury, commenced in 1220 and completed in 1258, with the exception of the spire, which does not appear to have formed part of the original design.

In this church we have a plan not only extremely beautiful, but perfectly original. There is scarcely a trace of French or foreign influence; everything is the result of the native elaboration during the previous century and a half. The internal dimensions, according to Britton, are 450 ft. by 78—a little under the English standard, but sufficiently long for effect. The apsidal arrangement, so universal in Norman cathedrals, has disappeared never to return, except in Westminster Abbey (1245-1269), and in some readjustments, as at Tewkesbury; and the square eastern termination may henceforth be considered as established in this country—the early symbol of that independence which eventually led to the Reformation.

Once the Salisbury plan came to be considered the true English type, the Norman cathedrals were gradually modified to assimilate their arrangements to it. The nave and transept of Winchester were already too extensive to admit of a second transept, but the choir was rebuilt on the new model; and when afterwards the nave was remodelled by William of Wykeham it became one of the most beautiful, as it continued to be the longest, of English cathedrals (556 feet, over all).

About the same time Ely had a choir and presbytery added to it in lieu of the old Norman choir, which raised it to the very first rank among English churches;^[99] and when, in 1322, by a fortunate accident the old Norman tower fell, the intersection was rebuilt in a manner that rendered it exceptionally pre-eminent among its rivals. There is perhaps no feature in the whole range of Gothic architecture either here or on the Continent more beautiful than the octagon of Ely (Woodcut No. 808), as rebuilt by Alan of Walsingham, the sacrist at the time the tower fell. He, and he alone of all northern architects, seems to have conceived the idea of abolishing what was in fact the bathos of the style—the narrow tall opening of the central tower, which, though possessing exaggerated height, gave neither space nor dignity internally to the central feature of the design. On the other hand, the necessity of stronger supports to carry the tower frequently contracted still more the one spot where, according to architectural propriety, an extended area was of vital importance to the due harmony of the design.

In the present instance the architect took for the base of his design the whole width of the nave and aisles, constructing in it an octagon, the sides of which are respectively 25 and 30 ft., and the diameter 65 ft. in one direction east and west, and 70 ft. transversely. By this arrangement a central area was obtained more than three times the extent of that originally existing, and, more than this, a propriety and poetry of design which are not to be found elsewhere. All this too was carried out with the exquisite details of the best age of English Gothic, and the effect in consequence is surpassingly beautiful. Unfortunately, either for want of funds, or of confidence in their ability to execute it, the vault, like that of York, is only in wood, though, from the immense strength of the supports, and their arrangement, it is evident that a stone vault was originally intended. The very careless—one might almost say ugly—way in which the lantern was finished externally, shows unmistakably that it was not intended to last long in its present form. Be that as it may, this octagon is in reality the only true Gothic dome in existence; and the wonder is, that being once suggested, any cathedral was ever afterwards erected without it. Its dimensions ought not to have alarmed those who had access to the domes of the Byzantines or Italians. Its beauty ought to have struck them as it does us. Perhaps the true explanation lies in the fact that it was invented late in the style. New cathedrals or great churches were very rarely commenced after the death of Edward the Third; and when they were, it was more often by intelligent masons, than by educated gentlemen, that they were designed.

After this, very little novelty was introduced into the design of English cathedrals. York, however, was almost entirely rebuilt in the form towards which the architects were tending during the whole of the Middle Ages, and it may consequently be considered as the type at which they were aiming, though hardly the one to which we can give the most unqualified praise. The nave was erected between the years 1291 and 1331, the choir between 1361 and 1405; the length internally is 486 ft.; the width of the choir, 100 ft.; of the nave, 106 ft.; both these last were, unfortunately, dimensions which the architects did not feel themselves equal to grappling with in stone, so that the roof, like the lantern at Ely, was constructed of wood, in imitation of a stone vault, and remains so to this day.

Owing to the great width attempted for the nave, York has not the usual proportion of length affected by other English cathedrals, and loses in effect accordingly. Its great peculiarity is the simplicity and squareness of its plan, so unlike what is found anywhere abroad. The church is divided into two equal parts; one devoted to the laity, one to the clergy. There are no apsidal or other chapels. Three altars stood against the eastern wall, and it may be 3 or 4 in the transept. Beyond this nothing. There is none of that wealth of private chapels which distinguishes Continental cathedrals and churches, or even Canterbury, the most foreign of our English examples. The worship even at that early period was designed to be massive and congregational, not frittered away in private devotion or scattered services, and marks a departure from Continental practices well worthy the attention of those who desire to trace the gradual development of the feelings of a people as expressed in their architecture, and the architecture only.

The abbey church at Westminster is exceptional among English examples, and is certainly, in so far at least as the east end is concerned, an adaptation of a French design. The nave, however, is essentially English in plan and detail, and one of the most beautiful examples of its class to be found anywhere. So, too, are the wide-spreading transepts; but eastward of these the form is decidedly that of a French cathedral. Henry VII.’s Chapel now stands over the space formerly occupied by the Lady Chapel; but before it was pulled down the circlet of apsidal chapels^[100] was as completely and as essentially French as any to be found in the country where that feature was invented. In the choir, however, the architects betrayed their want of familiarity with the form of termination they had selected. The angle at which the three bays of the apse meet is far from pleasing, and there is a want of preparation for the transition, which tends to detract from the perfection of what would otherwise be a very beautiful design.^[101]

As the choir was sepulchral, to accommodate the shrine of the Confessor, the design was appropriate, and its introduction in this instance cannot be regretted; but on the whole, there is nothing in the church of Westminster to make us wish that this feature had become more common on this side of the Channel.

Notwithstanding the beauty of the result, it may still be considered as open to discussion whether the English architects were always correct in adhering to length in preference to height as the modulus of their designs. When, however, we reflect how immensely the difficulties of constructing a stone roof are increased by every addition to the width or height of the vault, we cannot but acknowledge their wisdom in stopping at that point where sufficient spaciousness was attained, without increasing constructive difficulties. Nowhere in English cathedrals are we offended by mechanical tours de force. Everywhere there is sufficient solidity for security, and a consequent feeling of repose most conducive to true architectural effect.

It may also be remarked that the strain of turning the head upwards detracts considerably from the pleasure of contemplating tall interiors, while the eye likes to dwell on long-drawn vistas which can be explored in a natural position. But, perhaps, the greatest advantage of moderate dimensions in section is that they do not dwarf either the worshippers or the furniture of the church. Everything in an English cathedral is in just proportion, which is certainly not the case in many Continental examples; and there is variety and a play of light and shade in the long aisles of our churches which is wholly wanting in French and German examples.

Another point on which a difference of opinion may fairly exist, is whether the square termination of our cathedrals is or is not more beautiful than the apsidal arrangements so universal abroad.

When, as at Salisbury, or Wells, or Exeter, there is a screen of open arches below the east window, it may safely be asserted that a polygonal termination would have been more pleasing; but when, as at York, or Gloucester, or Carlisle, the whole eastern wall is a screen of painted glass, divided by mullions and tracery of most exquisite design, judgment will probably go the other way. Such a window as that at York, 33 ft. in width by 80 ft. in height, is a marvellous creation, which few architectural developments in any part of the world can rival or even approach. On the whole, perhaps, the true answer to the question, is that, where a number of smaller chapels are wanted, the chevet form is the best and most artistic termination for a church; where these are not required, the square form is the most beautiful, because it is the most appropriate, and, like everything appropriate, capable of being made beautiful in the hands of a true artist.

Vaults.

Whatever opinion may be formed as to the proportions of English cathedrals, or the arrangement of their plans, there can be no dispute as to the superiority of their vaults over those of all their Continental rivals. The reasons for this are various, and not very recondite. The most obvious is the facility of construction which arose from the moderation just pointed out in the section of our churches.

The English always worked within their strength, instead of going to the very verge of it, like the French; and they thus obtained the power of subordinating constructive necessities to architectural beauty. Thus the English architects never attempted a vault of any magnitude till they were sufficiently skilled in construction to do it with facility. In a former chapter it has been pointed out how various and painful were the steps by which the French arrived at their system of vaulting—first by pointed tunnel-vaults and a system of domes, then by a combination of quadripartite and hexapartite intersecting vaults, of every conceivable form and variety, but always with a tendency to domical webs, and to the union of all pre-existing systems. This experimentalising, added to the great height of their roofs, and the slenderness of their clerestories, never left them sufficiently free to admit of their studying æsthetic effects in this part of the construction.

A second reason was, that for 150 years after the Conquest, our architects were content with wooden roofs for their naves. One of the earliest vaults we possess is that at Durham, commenced by Prior Melsonby, 1233. Long before that time the French architects had been trying all those expedients detailed at pp. 113, 114, and had thus succeeded in vaulting their central aisles a century before we attempted it. In doing so, however, their eyes got accustomed to mechanical deformities which we never tolerated, and they were afterwards quite satisfied if the vault would stand, without caring much whether its form were beautiful or not.

A third cause of the perfection of English vaults arose from the constant use of ornamental wooden roofs throughout the Middle Ages. The typical example of this form now remaining to us is that of Westminster Hall. But St. Stephen’s Royal Chapel had one of the same class, and there is reason to believe that they were much more common than is usually supposed.^[102] All these were elaborately framed and richly carved and ornamented, often more beautiful than a stone vault, and quite as costly; and it seems impossible that a people who were familiar with this exquisite mode of roofing could be content with the lean twisted vaults of the Continental architects. The English alone succeeded in constructing ornamental wooden roofs, and, as a corollary, alone appreciated the value of a vault constructed on truly artistic principles and richly ornamented. Their eyes being accustomed to the depth and boldness of timber construction could never tolerate the thin weak lines of the French ogive, just sufficient for strength, but sadly deficient in expression and in play of light and shade.

Although it is, perhaps, safe to assert that there is not, and never was, a Saxon vaulted church in existence; and that, during the purely Norman period, though the side-aisles of great churches were generally vaulted, the central aisle was always ceiled with wood; yet, from a study of their plans, we are led to conclude that their architects always intended that they should, or at least might, be ornamented with stone roofs.

In the first place the area of their piers is enormous, and such as could never have been intended to support wooden roofs. Even making every allowance for the badness of the masonry, one-tenth of the sectional area would have sufficed, and not more was employed cotemporaneously in Germany when it was intended to use wooden roofs. There is also generally some variation in the design of the alternate piers, as if a hexapartite arrangement were contemplated. But the evidence is not conclusive, for the vaulting shafts are usually similar, and in all instances run from the ground through the clerestory, and terminate with the copings of the wall, so that, in their present form, they could only be meant to support the main timber of the roof. It may be that it was intended to cut them away down to the string-course of the clerestory, as was actually done at Norwich in 1446, when the nave was vaulted; but at present we must be satisfied with the evidence that the architects were content with such roofs as that of Peterborough (Woodcut No. 810), which is the oldest and finest we possess. It is very beautiful, but certainly not the class of roof these massive piers were designed to support.

Though we may hesitate with regard to the intention of the builders of Norwich, Ely, or Peterborough, there can be no doubt, from the alternate piers and pillars, that when Durham (Woodcut No. 804) was commenced it was intended that the nave should be covered by a great hexapartite vault. Before, however, the intention could be carried out, the art of vaulting had been so far perfected that that very clumsy expedient was abandoned; and, by the introduction of a bracket in the nave, and afterwards of a vaulting shaft in the choir, a vault of the usual quadrilateral form was successfully carried out between the years 1233 and 1284.

It is probably to St. Hugh of Lincoln that we owe the first perfect vault in England. Coming from Burgundy he must have been familiar with the great vaults which had been constructed in his country long before the year 1200, when he encouraged his new followers to undertake one not necessarily in the Burgundian style, but in that form with which they were conversant from their practice in erecting smaller side-vaults. He built and roofed the choir of Lincoln, immediately after which (1209-1235) the nave (Woodcut No. 811) was undertaken by Hugh of Wells, and its roof may be taken as a type of the first perfected form of English vaulting. It is very simple and beautiful; but it cannot be denied—and this is felt still more at Exeter—that the great inverted pyramidal blocks of the roof are too heavy for the light pier and pierced walls which support them. Another defect is, that the lines of the clerestory windows do not accord with the lines of the “severeys” of the vault. This defect was remedied at Lichfield, but nowhere else, until the invention of the four-centred arch and of fan-tracery. At Lichfield (Woodcut No. 812) the triangular form of the clerestory windows afforded a perfect solution of the difficulty, and gave a stability and propriety to the whole arrangement that never was surpassed, and never might have been relinquished had not their fatal fondness for painted glass forced the architects in this, as in other instances, to forego constructive propriety for indulgence in that fascinating mode of decoration.

Beautiful as these simple early roofs were felt to be, the great mass of the “severeys,” or inverted pyramids, formed a very obvious defect. It was, however, easily remedied when once perceived. The earliest example of its successful removal is probably in the roof of the choir at Gloucester (1337-1377) (Woodcut No. 813). In this instance the roof is almost a tunnel-vault with the window spaces cutting into it, so as to leave nearly one-third of the space unbroken; and, as the whole is covered with rich and appropriate tracery, the effect is highly pleasing. The same principle was afterwards carried to its utmost perfection in the roof of St. George’s Chapel at Windsor. In that case a flat band was introduced as a separate constructive compartment in the centre, supported by the severeys, and as the roof is ornamented with ribbings of the most exquisite design, it forms perhaps the most beautiful vault ever designed by a Gothic architect.

The great invention of the English architects in vaulting is the form usually known as fan-tracery. It is so beautiful in itself, and so exclusively English, that it may, perhaps, be worth while to retrace the steps by which it was arrived at. This may lead to a little repetition, but the stone vault is so essentially the governing modulus of the style that its principles cannot be made too clear.

The original form of the intersecting vault is that of two halves of a hollow-sided square pyramid placed opposite one another in an inverted position.^[104] One half of such a vault is shown at A and A A (Woodcut No. 814, fig. 1). The English seem early to have tired of the endless repetition of these forms, and, after trying every mode of concealing their sameness by covering them with tracery, they hit on the happy expedient of cutting off their angles, as shown at B and B B. This left a flat square space in the centre, which would have been awkward in the central vault, though in a side-aisle it was easily got over, and its flatness concealed by ornament. Arrived at this stage it was easy to see that by again dividing each face into two, as at C, fig. 1, the principal original lines were restored, and the central space could be subdivided by constructive lines to any extent required. By this process the square pyramid had become a polygonal cone of 24 sides, which was practically so near a circle that it was impossible to resist the suggestion of making it one, which was accordingly done, as shown at D and D D, fig. 1.

So far all was easy, but the fact of the flat central space resting on the four cones was still felt to be a defect, as indeed is apparent in such a vault as that of the cloisters at Gloucester (Woodcut No. 815), where a segment is used nearly equal to an equilateral spherical triangle. In this case they did not dare to employ a constructive decoration, but covered the space with circles so as to confuse and deceive the eye. At Windsor (Woodcut No. 816) the defect was obviated by using a low four-centred arch invented for the purpose, so that the outer tangent of the concoid was nearly flat, and the principal transverse rib was carried to the centre without being broken—as the others might have been had that mode of decoration been deemed expedient. This may be considered the perfection of this kind of vaulting, and is perhaps the most beautiful method ever invented. At Westminster (as shown in Woodcut No. 817) the difficulty was got over by reversing the curve by the introduction of pendants. This was a clever expedient, and produced a startling effect, but is so evidently a tour de force that the result is never quite satisfactory; though on a small scale perfectly admissible.

These devices all answered perfectly so long as the space to be roofed was square, or nearly so; but when this mode of vaulting came to be applied to the bays of the central nave, which were twice as long in one direction as in the other, the difficulties seemed insuperable. By cutting off the angle as in the former instance (as at B, fig. 2, Woodcut No. 814), you may get either a small diamond-shaped space in the centre or a square, but in both cases the pyramid becomes very awkward; and by carrying on the system as before, you never arrive at a circle, but at an elliptical section as shown at D, fig. 2 (Woodcut No. 814).

The builders of King’s College Chapel strove to obviate the difficulty by continuing the conoid to the centre, and then cutting off what was redundant at the sides, as in E, fig. 2, or, as shown in the view of the interior (Woodcut No. 846) further on.

The richness of the ornaments, and the loftiness and elegance of the whole, lead us to overlook these defects at Cambridge, but nothing can be less constructive or less pleasing that the abruptness of the intersections so obtained. In the central aisle of Henry VII.’s Chapel it was avoided by a bold series of pendants, supported by internal flying buttresses, producing a surprising degree of complexity, and such an exhibition of mechanical dexterity as never fails to astonish, and generally to please; though it must be confessed that it is at best a mere piece of ingenuity very unworthy of English art. By far the most satisfactory of these roofs is that at Windsor, where a broad flat band is introduced in the centre of the roof, throughout the whole length of the chapel. This is ornamented by panelling of the most exquisite design, and relieved by pendants of slight projection, the whole being in such good taste as to make it one of the richest and probably the most beautiful vault ever constructed. It has not the loftiness of that at Cambridge, being only 52 ft. high, instead of 78, nor is it of the same extent, and consequently it does not so immediately strike observers, but on examination it is far more satisfactory.

The truth of the matter seems to be that, after all their experience, the architects had got back to precisely the point from which they started, namely, the necessity of a square space for the erection of a satisfactory intersecting vault. The Romans saw this, and never swerved from it. The side-aisles of all cathedrals and all cloisters adhered to it throughout; and, when it was departed from in the wider central aisles, it always led to an awkwardness that was hardly ever successfully conquered. In some instances, as in the retro-choir at Peterborough (1438-1528), two windows are boldly but awkwardly included in one bay (Woodcut No. 818), and the compartments are so nearly square that the difficulty is not very apparent, but it is sufficient to injure considerably the effect of what would otherwise be a very beautiful roof.

In Henry VII.’s Chapel the difficulty was palliated, not conquered, by thrusting forward the great pendants of the roof and treating them as essential parts of the construction, and as if they were supported by pillars from the floor instead of by brackets from the wall. By this means the roof was divided into rectangles more nearly approaching squares than was otherwise attainable; but it is most false in principle, and, in spite of all its beauty of detail, cannot be considered successful.

Strange as it may appear from its date, the most satisfactory roof of this class is that erected by Cardinal Wolsey in the beginning of the 16th century over the choir of Oxford Cathedral. In this instance the pendants are thrust so far forward and made so important that the central part of the roof is practically quadripartite. The remaining difficulty was obviated by abandoning the circular horizontal outline of true fan-tracery, and adopting a polygonal form instead. As the whole is done in a constructive manner and with appropriate detail, this roof—except in size—is one of the best and most remarkable ever executed.

The true solution of the difficulty, in so far as the vault was concerned, would have been to include two bays of the side-aisles in one of the centre; but this would have necessitated a rearrangement of both plan and exterior to an extent the architects were not then prepared to tolerate, and it never was attempted, except perhaps in the instance of the retro-choir at Peterborough (Woodcut No. 818). Had it been done in King’s College Chapel at Cambridge (Woodcut No. 846), it would have been in every respect an immense improvement. At present the length of King’s College Chapel is too great for its other dimensions. Had there been six bays instead of twelve, its apparent length would have been considerably diminished, and the variety introduced by this change would have relieved its monotony without detracting from any of the excellent points of design it now possesses.

The English architects never attempted such vaults as those of Toulouse and Alby, 63 and 58 ft. respectively, still less such as that of Gerona in Spain, which is 72 ft. clear width. With our present mechanical knowledge, we could probably construct wider vaults still. Even the Mediæval architects in England might have done more in this direction than they actually accomplished, had they tried. On the whole, however, it seems that they exercised a wise discretion in limiting themselves to moderate dimensions. More poetry of design and greater apparent size is attainable by the introduction of pillars on the floor, and with far less mechanical effort. Unless everything is increased in even a greater ratio, the dwarfing effect of a great vault never fails to make itself painfully apparent. We may regret that they did not vary their vaults by such an expedient as the lantern at Ely, but hardly that they confined them to the dimensions they generally adopted.

Pier Arches.

Although the principles adopted by the English architects did not materially differ from those of their Continental confrères with regard to the arrangement of pier arches and the proportions of triforia and clerestories, still their practice was generally so sound and the results so satisfactory, that this seems the best place to point out what the Mediæval architects aimed at in the arrangement of their wall surfaces.

In the Norman cathedrals the general scheme seems to have been to divide the height into three equal parts, and to allot one to the pier arch, another to the triforium or great gallery, and the third to the clerestory. In all the examples we now have, the upper is the smallest division; but I cannot help fancying that some arrangement of the timbers of the roof gave the additional height required. It is generally supposed that the roof at Peterborough (Woodcut No. 810) was originally flat. This, however, is by no means clear, nor that it started so low; but, be that as it may, the woodcut (No. 820) will explain the usual arrangement, as well as the changes afterwards introduced. At Winchester the two lower divisions are practically equal, the upper somewhat less, and the alternate arrangement of the piers hints at a hexapartite vault, if such should ever come to be executed. When William of Wykeham undertook to remodel the style of the nave, he first threw the two lower compartments into one, as shown on the left-hand side of the cut. He then divided the whole height, as nearly as the masonry would allow him, into two equal parts, allotting one to the pier arches, and apportioning the upper as nearly as he could by giving two-thirds to the clerestory and one-third to the triforium. With pointed arches this was the most pleasing and satisfactory arrangement adopted during the Middle Ages; but when something very like it was attempted in the nave of Gloucester with round arches, the effect was most unpleasing. Before the architects, however, settled down to this proportion, a variety of experiments were tried. One of the most successful was the nave of Lichfield Cathedral (Woodcut No. 812). Here the whole height is divided equally: one half is given to the pier arches, and the other divided equally between the clerestory and triforium. If the latter had been glazed externally, as was the case at Westminster Abbey and elsewhere, and made to look like part of the church, the whole might be considered as satisfactory. As it is, the area of the clerestory is so much less than that of the triforium, that the proportion is not quite agreeable, though the solidity and repose which this arrangement gives to the roof is above all praise.

All these objections were obviated in the three bays of the choir at Ely, which were rebuilt by Walsingham at the same time as the octagon. Here the triforium and clerestory are equal; but the upper window is so spread out, and so much is made of it, that it looks equal to the compartment below. The pier arch below is also subdued to less than half the whole height, so as to give value to the upper division. These proportions are derived from the very beautiful Early English presbytery beyond; but they are here used with such exquisite taste and such singular beauty of detail that there is perhaps no single portion of any Gothic building in the world which can vie with this part of the choir of Ely for poetry of design or beauty of detail.

The perfection of proportion, as of many other things, was reached in Westminster Abbey (1245-1269). Here the whole height is divided into two equal parts, and the upper subdivided into three, of which one is allotted to the triforium, and two to the clerestory. It is true this involves the necessity of springing the vault from a point half way down the clerestory windows, and thus the lines of the severeys do not accord quite with those of the lights; but at best it is a choice of difficulties, and the happy medium seems to have been reached here more successfully than elsewhere. The proportion of the width of a bay to its height is here also most pleasing; it is as 1 to 5½.^[105] Sometimes, as at Exeter, it sinks as low as 1 in 3, but the whole effect of the building is very much destroyed by the change.

Shortly after this, as in the choir at Lichfield (1250-1325) or at Exeter (1308-1369), the mania for the display of painted glass upset all these arrangements—generally at the expense of the triforium. This feature was never entirely omitted, nor was it ever glazed internally, as was frequently the case on the Continent; but it was reduced to the most insignificant proportions—sometimes not pierced—and, with the wider spacing just alluded to, deprived the English side screen of much of that vigour and beauty which characterised its earlier examples.

Window Tracery.

The date of the introduction of the pointed arch in England—for it may be considered as established that it was introduced—is a question which has been much discussed, but is by no means settled. The general impression is that it was at the rebuilding of the cathedral of Canterbury after the fire of 1174 that the style was first fairly tried. The architect who superintended that work for the first five years was William of Sens; and the details and all the arrangements are so essentially French, and so different from anything else of the same age in England, that his influence on the style of the building can hardly be doubted. Of course it is not meant to assert that no earlier specimens exist; indeed, we can scarcely suppose that they did not, when we recollect that the pointed arch was used currently in France for more than a century before this time, and that the pointed style was inaugurated at St. Denis at least thirty years before. Still this is probably the first instance of the style being carried out in anything like completeness, not only in the pier arches and openings, but in the vaults also, which is far more characteristic.

Even after this date the struggle was long, and the innovation most unwillingly received by the English, so that even down to the year 1200 the round arch was currently employed, in conjunction with the pointed, to which it at last gave way, and was then for three centuries banished entirely from English architecture.

Be this as it may, in their treatment of tracery, which followed immediately on the introduction of the pointed arch, the English architects showed considerable originality in design, though inspired by the same sobriety which characterises all their works. They not only invented the lancet form of window, but what may be called the lancet style of fenestration. Nowhere on the Continent are such combinations to be found as the Five Sisters at York (Woodcut No. 824), or the east end of Ely (Woodcut No. 825), or such a group as that which terminates the east end of Hereford (Woodcut No. 826). Tracery it can hardly be called, but it is as essentially one design as any of the great east windows that afterwards came into fashion; and until painted glass became all-important, such an arrangement was constructively better than a screen of mullions, and as used in this country is capable of very beautiful combinations.