John Henry Parker

We have now arrived at the period of those rich Norman churches which may still be considered as amongst the glories of our land.

It is very remarkable that so large a number of buildings of the rich character which generally distinguishes this style should all have been built in about half a century, from 1120 to 1170 or 1180; yet such is clearly the case. The early Norman style has been already described; the late or rich Norman is chiefly characterized by the abundance of ornament and the deep cutting, the absence of which is the chief characteristic of the earlier period.

Before we proceed to describe it, a few of the buildings known to have been erected at this time may be mentioned.

Peterborough Cathedral was begun from its foundations in 1117 by John de Seez, who formed the plan of the whole of it, which was rigidly carried out by his successors, and it was consecrated in 1143; the work is very good, but not very rich. The Norman tower at Bury St. Edmund’s was commenced in the same year, 1117, and finished in 1130; the porch is an addition about half a century later. The nave of Norwich was built between 1122 and 1145: the work is still very plain, being in continuation of the previous work. Castor Church, Northamptonshire, bears an inscription recording its dedication in 1124: the tower is good, rich Norman work; the ornaments are the hatched, the square billet, and the scollop, all of very simple character, shallow, and easily worked. Furness Abbey was founded in 1127, but very little of the original work remains. In Canterbury Cathedral, the work of Prior Ernulf, under St. Anselm, was completed in 1130, and part of Rochester, where Ernulf had become bishop, in the same year; so that we need not be surprised at finding more ornament in these two cathedrals than is quite consistent with the usual character of early Norman work, and the same ornaments repeated in both these churches. St. Martin’s priory at Dover was founded in 1131; the refectory is still standing, and is a good example of plain Norman work, neither very early nor very late.

The small Norman church of Newhaven in Sussex is unusually perfect, and gives a good general idea of a parish church of the twelfth century. At first sight it looks earlier than it is; the bold projection of the buttresses indicates a later period, early Norman buttresses are very flat, the greater the projection the later

[Image unavailable.]

Newhaven Church, Sussex, c. A.D. 1120.

The apse is usually an early feature; in this instance the projection of the buttresses and the ornamental string round it shew it to be later. The side-window is an insertion in the Early English style.

they are, as a general rule. The spire is an early one, though that is not likely to be Norman. The belfry-windows in the tower, and the corbel-table under the eaves of the roof, are early. The porch is evidently a later addition.

At Iffley the tower is later; the original choir was square, with a flat east end, and another square bay has been added eastward at a later period, more in the Early English style.

The Augustinian priory of Dunstable, in Bedfordshire, was also founded in 1131; the original parts of the west front and of the nave are remarkably fine and rich Norman work.

In the time of William Rufus the work begun by the Norman bishops was carried on so vigorously, that, before the close of this century, every one of the Saxon cathedrals was undergoing the same process of destruction, to be rebuilt on a larger scale and in a better manner. Some of the buildings which remain to us of the work of this reign are the crypt of Worcester; the crypt, the arches of the nave, and part of the transepts of Gloucester; the choir and transepts of Durham; the nave and transepts of Christchurch in Hampshire; the choir and transepts of Norwich.

The history of Canterbury Cathedral has been so carefully preserved by contemporary records, and these have been so thoroughly investigated by Professor Willis, and compared with the existing structure, that we may almost put a date upon every stone of this magnificent fabric; it is therefore our best and safest guide in the study of the architecture of that period in England. The work in the older part of the crypt agrees exactly with that at Lincoln, and the other early Norman works above mentioned. The crypt is, however, not part of Lanfranc’s work, for it is remarkable that his church was entirely pulled down and rebuilt by his successor, St. Anselm, between 1096 and 1110, under the direction of Priors Ernulf and Conrad. Even in the time of Gervase, writing in 1170, he says, “You must know, however, good reader, that I never saw the choir of Lanfranc, neither have I been able to meet with any description of it: Eadmer indeed describes the old church, which before the time of Lanfranc was constructed after the Roman manner; he also mentions, but does not describe, the work of Lanfranc, which succeeded this old church, and the choir of Conrad, constructed in the time of St. Anselm.” From this we may fairly conclude that the work of Lanfranc was of very inferior character. It is now said by some, that parts of the walls of the present crypt at the west end belong to this early period, “after the Roman manner;” Willis considered this to be of the time of Lanfranc.

During the first fifteen or twenty years of the twelfth century, and of the reign of Henry I., there was no perceptible change of style; the numerous great works which had been begun during the preceding twenty years were carried on, and many of them were completed. During this period we have the dedications,—which shew that the work was sufficiently forward for the choir to be used,—of Ely, Rochester, Norwich, Canterbury, and some others. Several new works were commenced also, as Tewkesbury Abbey, St. Botolph’s, Colchester, St. Bartholomew’s, Smithfield, the nave of Durham, the choir of Peterborough, and Reading Abbey: but we do not find any difference between the early parts of these and those which immediately preceded them. There is no difference whatever between those built on the sites of the Saxon cathedrals, and those which were now first erected on entirely new sites.

We find in early Norman work that the chisel was very little used; most of the ornaments are such as could be readily worked with the axe, and whatever sculpture there is appears to have been executed afterwards, for it was a general practice to execute sculpture after the stones were placed, as is evident in the early work at Westminster: some of the capitals in the crypt of Canterbury are only half finished to this day, the work of carving having probably gone on until it was stopped by the great fire in 1174. If the sculpture is early it is very rude, and the work is shallow. But shallowness of carving depends partly on the nature of the material to be carved; from this cause buildings of a hard stone, such as granite, often appear much older than they really are. Baptismal fonts especially are frequently made of hard stone or marble, which admit of shallow sculpture; and rich Norman work cut shallow may be found as late as the time of Henry II.

Crypt, Canterbury, A.D. 1110.

Norman capital, with carving commenced and left unfinished.

Although the roofs of the aisles at Canterbury had been vaulted, the choir itself had a flat boarded ceiling, painted like that still remaining at Peterborough. The vault of the choir of the cathedral of Sens, from whence came William, the architect of the choir of Canterbury, is also an addition of later date. The same change was made in many other churches of that period. The builders of the early Norman period did not venture to erect a vault over so large a space; we do not find any early vault over a space above twenty feet wide, and few of so wide a span. Many of our Norman cathedrals still have timber roofs over the large spaces, and the aisles vaulted. In Normandy vaults were more frequently used than in England, even at this early period; and this was still more the case in subsequent times, for the fine open timber roofs for which some parts of England are distinguished are unknown in Normandy, where almost every village church is vaulted over.

Here it may be well to mention, that down to the early Norman period the eastern limb of a cruciform church, or the chancel of a plain oblong plan, was always short, rarely more than a single square, or at the utmost two squares, in length, and was frequently terminated by a round east end called an apse. Immediately after this period the custom of lengthening the eastern limb of the church became so general, that the original dimensions have been almost lost sight of. The history of nearly every one of our cathedrals gives the same result: first, the choir was lengthened by the addition of a presbytery, and afterwards still further by adding a lady-chapel, which did not come into fashion until quite the end of the twelfth century.

Gervase and William of Malmesbury have furnished us, as we have seen, with a clue by which to distinguish the work of the early Norman period from that of a later age, namely, wide-jointed masonry, and shallow sculpture executed chiefly with the axe instead of the chisel. The best and safest test is the wide-jointed masonry, where it is found; but in some cases the joints can hardly be said to be either wide or fine; they are of a moderate width, and not of marked character either way.

The arch is generally at first not recessed at all, afterwards only once recessed, and the edges are either square, or have a plain round molding cut upon them; the zigzag ornament is used, but not so abundantly as at a later period; the dripstone is frequently ornamented with what is called the hatched molding; the billet is also used, but sparingly, and perhaps not before 1100; it is found in the early parts of Peterborough, but not in the later parts. The head of the door is generally square with a round arch over it, and the intermediate space under the arch, called the tympanum, is either left plain, or ornamented with shallow sculpture of rude character, sometimes preserved from an earlier building.

Rich Doorways form one of the most important features of late Norman work. The examples given from Cuddesdon and Middleton Stoney are good ordinary specimens, such as may be found in scores of parish churches. They are generally round-headed, very deeply recessed, and frequently have shafts in the jambs. The tympanum is frequently filled with rich sculpture, which becomes deeper and better executed as the style advances. The moldings are numerous, but not of much variety in section,

consisting chiefly of round and quarter-round members, but all preserving a general square outline. These moldings, however, as well as the jambs and shafts, are frequently entirely overlaid with ornament, which, though of a peculiar and somewhat rude character, produces great richness of effect; and few features of churches are more generally admired than these rich Norman doorways, which are very abundant in many parts of the country, quite as much so as in Normandy itself. The examples in England are quite as fine and as numerous in proportion as in Normandy; and these doorways were so much admired for their rich character, that they have often been preserved when the church has been rebuilt, perhaps several times. The doorways of Iffley Church are among the richest that we have anywhere; not only the very fine one at the west end, but the north and south doors.

Norman Windows are in general long and rather narrow round-headed openings, but sometimes of two lights divided by a shaft, included under one arch, more especially in belfries; in rich buildings they are frequently ornamented in the same manner as the doorways, with recessed arches, zig-zag and other moldings, as at Iffley, Oxfordshire, and sometimes with sculpture; other examples have shafts in the jambs carrying the

arch-moldings, and others are quite plain. At Castle Rising, Norfolk, is a very rich late example, with intersecting arcades on each side, ornamented chiefly with the lozenge molding. In Romsey Abbey, Hampshire, Waltham Abbey, Essex, Christ Church Cathedral, Oxford, and very many other examples, the clerestory window has a smaller blind arch on each side of it, making a triple opening within to a single window; and the shafts of this triple opening are made to carry small shafts to the upper arches. This is a common arrangement of Norman clerestory windows: at St. Stephen’s, Caen, there is only one subarch to each light instead of two, but this arises from the arrangement of the sexpartite vaulting.

The fine circular windows with wheel-like divisions belong to this period: Barfreston, in Kent, is a good example. St. James, Bristol, is a singular one, the effect of which is rich and good. There was frequently one in the centre of the west front, which was called the oculus, or eye of the building. These large round windows are much more common on the Continent than in England. In Italy there are many fine examples, as at Toscanella, Perugia, and Assisi. The French also appear to have always had a particular fondness for this kind of window, which in the later styles becomes the magnificent rose-window, so often the glory of the French churches.

Norman windows are far less common than the doorways, having frequently been destroyed to make room for those of later styles; probably for the purpose of introducing the painted glass of those periods, which did not suit well with the early windows. Small circular openings are also a common feature, as in the clerestory of Southwell Minster. The zigzag molding is frequently used in the arches of windows, as at St. John’s, Devizes, p. 44; occasionally, but not so frequently, this ornament is also carried down the jambs, as at Iffley. Windows of two lights divided by a mullion were not introduced until after the Norman period. The walls being generally very thick, the opening is small and narrow on the outside of the wall, and is very widely splayed to admit more light; so that while the glass is less than a foot wide, the opening of the splay on the inner side of the wall is three feet wide.

The Arches are generally round-headed: in early work they are plain and square-edged, with or without a recess at the angle; sometimes doubly recessed, and still square-edged, as in the early work at Westminster, p. 11, the White Tower, London, and the transept of Winchester, p. 27; sometimes molded, with plain round moldings. In the later period they are more richly molded than in the early part of the style: the chancel-arch especially is very much enriched; and the western side, facing the spectator when looking towards the altar, is generally much more ornamented than the eastern side. The chancel-arch at Iffley is one of the richest and best examples: where there is a central tower, as in that instance, both the tower-arches across the church are usually ornamented in the same manner; the side-arches, where there are transepts, are frequently much plainer, and often pointed. In the later part of the Norman style, without any other change, they are still quite plain and square-edged. In this manner the pointed arch occurs quite as early as 1150, or even earlier; at a later period they become much more common, and are gradually developed into the Early English style, which some call the “first pointed style;” but the pointed arch alone does not make a change of style.

The Small Arcades which are frequently used as decorations of the walls, and for sedilia, have scarcely any separate character; they are diminutives of the larger arches, except that the shafts are smaller and shorter in proportion: in rich work they are used both inside and outside of the walls, and frequently on the outside of the clerestory, as well as on the inside in front of the blind-story, now called the triforium. Intersecting arches occur in these arcades from a very early period; and Rickman observes, that whoever constructed them, constructed pointed arches; and he adds, “It appears as if the round and pointed arches were for nearly a century used indiscriminately, as was most consonant to the necessities of the work, or the builder’s ideas.” At Canterbury, an ornamental arcade of intersecting arches occurs both on the inside and outside of the wall in St. Anselm’s tower.

In the apse in the White Tower the arches are stilted to accommodate them to their position. The arches of the triforium are generally wide and low; sometimes they are divided by two sub-arches.

The form of the arch was at all periods dictated partly by convenience, and is not to be relied on as a guide to the date or style; but there was a prevailing fashion, and that form was usually followed at each period, unless there was some reason for changing it, which is generally obvious if we look for it. To

judge of the age of any building we must look at the general character of the work, and not seize upon some particular feature to ground any rule upon. The moldings are generally the safest guide, but even these sometimes require to be qualified by comparison with other parts.

The work is frequently quite as massive, and in all other respects of as early character, with the pointed arch as with the round one; they occur in Malmesbury Abbey Church, apparently in the work of Roger, bishop of Salisbury, A.D. 1115-1139, without any other apparent difference of character from the rest of the work. The pointed arch, taken by itself, is therefore no proof of the change of style, nor even of late work.

St. Cross Church, near Winchester, founded by Henry de Blois in 1136, has pointed arches; and the triforium has intersecting arcades, with the intervals left open as windows. To these may be added, Fountains Abbey, Yorkshire, founded in 1132: pointed arches occur in the early part of the work, which is of pure Norman character, and appears to have been built before the fire in 1140;—and Kirkstall Abbey, built between 1152 and 1182; here the work is of later character, but still pure Norman. All these are previous to the period of transition, and have not transitional moldings.

The Piers in the earlier period are either square solid masses of masonry, or recessed at the angles, in the same manner as the arches, or they are plain round massive pillars, with frequently only an impost of very simple character, but often with capitals. The round pillars are sometimes ornamented with a kind of fluting, as in the crypt at Canterbury, sometimes with a rude and shallow zig-zag pattern, as at Waltham Abbey, Durham, and Lindisfarne.

In the later period the pillars are in general not so massive as in the early part of the style, and are frequently ornamented with small shafts; and these as well as the pillars are sometimes banded, as at St. Peter’s, Northampton.

The Capitals in early work are either plain cubical masses with the lower angles rounded off, forming a sort of rude cushion shape, as at Winchester, or they have a sort of rude volute, apparently in imitation of the Ionic, cut upon the angles; and in the centre of each face a plain square block in the form of the Tau cross is left projecting, as if to be afterwards carved: this remarkable feature is found in the chapel of the White Tower, London, in the early part of the crypt at Canterbury, at St. Nicholas, Caen, and other early work, but it has never been observed in late work.

The scolloped capital belongs to rather a later period than the plain cushion or the rude Ionic, and does not occur before the time of Henry I.; as at Stourbridge, Malmesbury, and Kirkstall. This form of capital was perhaps the most common of all in the first half of the twelfth century, and continued in use to the end of the Norman style.

The capitals were frequently carved at a period subsequent to their erection, as in the crypt at Canterbury (p. 37), where some of the capitals are finished, others half-finished, with two sides blank, and others not carved at all. In the early work at Westminster (p. 13), before mentioned, this is equally evident. At Castle Ashby, Northamptonshire, is the jamb of a Norman doorway with the pattern for the sculptor scratched upon it with the chisel, but never executed.

In later Norman work the capitals are frequently ornamented with foliage, animals, groups of figures, &c., in endless variety. The abacus throughout the style is the most characteristic member, and will frequently distinguish a Norman capital when other parts are doubtful. Its section is a square with the lower part chamfered off, either by a plain line or a slight curve; but as the style advanced it had other moldings added, and the whole are frequently so overlaid with ornament that it is difficult to distinguish the section (or profile) of its moldings.

The Bases are at first very simple, consisting merely of a quarter-round molding; then of two quarter-rounds, or two and a chamfer; or else of a round, or a chamfer and a quarter-round: as the style advanced they became more enriched, and the number of members more numerous: the earlier examples resemble the Tuscan, the later appear to be imitated from the Attic base. They always follow the form of the shaft or pillar, and stand upon a square pedestal or plinth; the angles of this square plinth being frequently filled up with some ornament, called foot-ornaments, or base ornaments: these increase in richness and boldness as the style advances, and their use was continued for some time in the subsequent style.

The Niches, or, Tabernacles, are small shallow recesses with round arches, frequently much enriched; they are chiefly placed over the doorways, and generally retain the figures which they were constructed to receive. These figures being executed in low relief upon the surface of the stone, were less liable to injury than the figures of the later styles, which are carved on separate stones and inserted. The most usual figure is that of Christ, distinguished by the cruciform nimbus. At Dorchester we have St. Peter with the key, under a semicircular arch, resting on cushion-capitals to twisted shafts, with molded bases. This example is from the font. The sculpture is at first very shallow, but becomes deeper as the style advances.

The Moldings have been already mentioned in describing the doorways, where they are most abundantly used; they are, however, freely employed on all other arches, whether the pier-arches, or over windows, wall arcades, &c., and frequently also as horizontal strings or tablets. One of the most usual and characteristic Norman strings exactly resembles the abacus of the capital, or the impost of the pier, with a hollow chamfer under it; another is merely chamfered off above and below, forming a semi-hexagonal projection.

Norman ornaments are of endless variety; the most common is the chevron, or zig-zag, and this is used more and more abundantly as the work gets later; it is found at all periods, even in Roman work of the third century, and probably earlier, but in all early work it is used

sparingly, and the profusion with which it is used in late work is one of the most ready marks by which to distinguish that the work is late. The sunk star is a very favourite ornament throughout the style; it occurs on the abacus of the capitals in the chapel of the White Tower, London, and at Herringfleet, Suffolk, and it seems to have been the fore-runner of the tooth-ornament. The billet is used in the early part of Peterborough, but discontinued in the later work, and does not often occur in late work. It is sometimes square, more frequently rounded, as in this example. The beak-head the cat’s-head, the small medallions with figures, and the signs of the zodiac, all belong to the later Norman period. In the later Norman moldings a mixture of Byzantine character is seen on the ornaments, as at Durham.

Sculptured ornament made great progress during the twelfth century. We have seen by the testimony of Gervase that the chisel was not used in the “glorious choir of Conrad” at Canterbury, which was built between 1096 and 1130, and an examination of the old work proves the exactness of his statement; all the sculptured ornament on the old work is shallow, and such as could very well be executed with the axe, which is not a bad tool in the hands of a skilful workman, and is still commonly used in many parts of England and France. On comparing this early work at Canterbury with other early Norman buildings, it is plain that they all had their ornaments executed in the same manner: the chisel is only required for deep-cutting and especially undercutting, and that we do not find on any buildings of ascertained date before 1120. The chisel was used for carving in stone in Italy and the south of France at an earlier period, but not in Normandy nor the north of France much earlier than in England. After this usage was introduced, the workmen seem to have gloried in it, and revelled in it, and the profusion of rich Norman sculptured ornament in the latter half of the twelfth century is quite wonderful.

It has been observed, that in the sculpture of the period of the late Norman style there is frequently a certain mixture of the Byzantine Greek character, brought home from the east by the Crusaders, who had returned. This is also one of the characteristics of the period of Transition.

The Corbel-tables are at first very plain, consisting merely of square blocks at intervals, carrying the beam on flat stones which support the roof, or with small arcs between them, or merely rude triangles, like the Anglo-Saxon arches; and these are sometimes continued in late work, as at Iffley, but in general, in late work the corbels are carved, and the small arcs more or less enriched. The buttresses are usually flat and plain in early examples, but have moldings on the angles in late examples.

Corbel-tables under the eaves of the roof are very abundant in late Norman and Transitional work, and are often proof that the walls are Norman, when this is not otherwise evident, later windows having been inserted. They are frequently square blocks of stone only, as if intended to be carved subsequently, when convenient, and this has never been done; more usually they are heads, or grotesque masks, as at Romsey.

The earliest Norman Vaults are quite plain, and of the barrel form, as in the chapel of the White Tower, London. In the next stage they have flat transverse arches only; they are then groined, but still without ribs: these plain groined vaults without ribs, over aisles or other narrow spaces, are often contemporaneous with the barrel vaults, and generally belong to the latter half of the eleventh century, or the beginning of the twelfth, as at Sherborne Castle, built by Roger, bishop of Salisbury, A.D. 1115-1139; at a later period ribs are introduced, at first square, then plain half-rounds, then molded, as in Peterborough Cathedral, A.D. 1117-1143, and they gradually change their form until they almost imperceptibly assume the character of Early English work.

The Norman architects did not venture to throw a vault over a wide space until very near the end of the style, and various contrivances were necessary for vaulting over spaces of unequal width, such as stilted arches, and horse-shoe arches, before the difficulty was solved by the use of the pointed arch. The absence of vaults over a wide space is a proof that the Norman was not a continuation of Roman work, as is sometimes assumed, but that there was always an interval of at least a century in which there were no masons.

Early Norman Turrets are very rarely to be met with, but there are good examples at St. Alban’s; at a later period they are frequent as stair-turrets, but have generally lost the original roof or capping; sometimes, as at Iffley, and Christchurch, Hampshire, they die into the tower below the corbel-table; in other instances, as at Bishop’s Cleeve and Bredon, they are carried up above the parapet and terminate in pinnacles; they are sometimes round and sometimes square.

At St. Cross, Winchester, there is a remarkable example, something between a turret and a large square pinnacle, rising from the top of the side wall to the level of the front of the gables, and even above it.

Norman Central Towers are very low and massive, seldom rising more than a square above the roof, sometimes not so much, the ridge of the original roof, as shewn by the weather-table on the face of the tower, being only just below the parapet. These towers were intended to be, and without doubt originally were, covered by low wooden pyramidal roofs, resembling in appearance those which we find in some parts of Normandy of the same period, there executed in stone, on account of the abundance of the material, the facility with which it is worked, and the skill of the workmen.

When the towers are not placed over the centre of the church, but at the west end, it is remarkable that the later Norman towers are more massive and not so lofty as the early ones, as at Lincoln, Jarrow, &c., already described. They are comparatively low and heavy, sometimes diminishing by stages, and having buttresses of little projection on the lower parts. The belfry, or upper storey, has frequently been added in late Norman times upon the earlier towers. The belfry windows are generally double, and divided by a shaft. Towers of the pre-Norman period are generally remarkably tall, as at Deerhurst, one of the best dated examples.

The Round Towers which are so abundant in Norfolk and Suffolk are frequently of the Norman period; some may be earlier, and others are certainly later; they are often so entirely devoid of all ornament or character, that it is impossible to say to what age they belong. The towers themselves are commonly, but not always, built of flint, sometimes of rough stone rubble, and are built round to suit the material, and to save the expense of the cut stone quoins for the corners which are necessary for square towers, and which often may not have been easy to procure in districts where building-stone has all to be imported. The same cause accounts for the frequent and long-continued use in the same districts of flat bricks or tiles for turning the arches over the doors and windows, which are either of Roman manufacture, or an imitation of the same form. Some good authorities think that the Roman form of flat bricks or tiles was long imitated in England.

The Buttresses of this style were at first merely flat projections wholly devoid of ornament, and these are sometimes continued in late work; but in general, in late work there is a recess at the angle, in which a small shaft is inserted; the strings are sometimes continued round the buttresses and sometimes stop short at them, but in the latter case the buttresses have generally been added to strengthen the wall after it was erected, and are not part of the original work. In late Norman buildings the buttresses are sometimes square, and consequently have a much greater projection than the early flat buttresses. These square buttresses also have the moldings or shafts at the angles that the flat buttresses have not; an early Norman buttress never goes higher than the ground-floor, even when it is against a tower; at an angle, a flat buttress is placed on each side, nearly close to it.

Norman Porches have in general very little projection, sometimes only a few inches, but the thickness of the wall allows the doorways to be deeply recessed; they are sometimes terminated by a gable, or pediment, as at St. Margaret-at-Cliffe, Kent, where the projection is so slight that it may be called either a doorway with a pediment over it, or a shallow porch. More frequently the projection ends in a plain set-off, in which case the appearance is that of a doorway set in a broad flat buttress. There are, however, a few porches which have as great a projection as those of the succeeding styles, and the sides of these are usually ornamented with arcades: the outer archway is of the same character as other doorways. At Sherborne and at Southwell Minster there are good examples of these porches.

But the square east end is the usual characteristic of the Anglo-Norman style; the apse is comparatively a rare feature in England. In the diocese of Laon in the north of France, the cathedral and a large number of the churches have square east ends, under the influence of an English bishop, who was a leading man there in the early part of the twelfth century. The small parish church of Cassington, Oxon, has a Norman chancel with a Norman vault also. At Iffley the original chancel was like that of Cassington—one square bay; another bay eastward of this is of the Early English style; both bays are vaulted. At Cassington, the whole of the walls of the church are Norman, and the lower part of the tower, but the belfry-storey and the spire are of the Decorated style. The thick abacus shews this corbel to be of quite early Norman character.

The Fronts, particularly the west fronts of Norman churches, are frequently of very fine composition, having generally deeply-recessed doorways, windows, and arcades, all covered with a profusion of ornament in the later period, as at Iffley, and at Nun-Monkton, p. 73.

The Apse has been already mentioned as a characteristic of the Norman style. In England it is more frequently used in early than in late work, and is found at the east ends of the chancel and its aisles, and on the east side of the transepts; being, in fact, the places for altars, which were afterwards continued in the same situations, but either merely under windows in a flat wall, or under arched recesses which frequently remain in the transept wall, and are sometimes erroneously described as doorways.

The custom which has been mentioned of lengthening the churches eastwards, which commenced in the latter half of the twelfth century, was carried on vigorously in the thirteenth.

At Romsey there is an apse at the end of each of the aisles, not in the large central part.