The fabric of the church: The eastern limb

CHAPTER II - THE EASTERN LIMB

The Quire, Apse, and Presbytery

The quire (not the monastic quire which occupied also the crossing and one bay of the nave) was, as already stated, built with four bays terminated by an apse of seven bays, the western bay being the work of Prior Thomas. It comprised a main arcade, averaging 17 ft. 10 in. in height, a triforium of 13 ft. 8 in., and a clerestory of 16 ft. 6 in., or 48 ft. in all. (fn. 1)

Though Rahere's work is shown by the measured drawings to have been set out with considerable exactness, there are certain peculiarities about the quire which want explanation.

The two eastern bays diminish in width westward by about 6 in., whilst the two western bays of the quire are parallel. The cause of this has been thus explained. (fn. 2)

The western bay of the apse is wider, both on the north and south sides, than the other bays, measuring 9 ft. from centre to centre against a measurement of 7 ft. of the other five bays. An examination of the setting out of these other bays shows that if the two western bays had been of the same width the apse would have been an exact and symmetrically divided semicircle; but with the two widened bays the apse is more than a semicircle. Owing to the eastern sides of the bases and capitals of the western piers of these bays having been set out also to radiate from a common centre like the sides of the other piers, there is a natural coming inwards which causes a slight horse-shoe shape to the apse, and, the sides of the quire being on tangential lines, the two eastern bays naturally approach each other. This may, or may not, have been intentional, but it is possible that, when setting out the apse, it was not borne in mind that the western bay on each side, being opposite to the entrance to a chapel, would require to be wider than the rest, and it was not noticed until it was too late to alter it.

The arches of the main arcade of the quire and its apse are roundheaded, the narrower ones of the apse being stilted to range with the others (pl. XXIX b, p. 23). The heights of the piers average 10 ft. 6 in. from the floor to the top of the abacus. The arches are recessed with a second order. The soffits of the arches are flat, as at St. Albans, Malvern, Tewkesbury, and Norwich. The hood moulding has the plain round billet ornament finely worked, which, besides passing over the arch, is also carried immediately over the abacus of the capitals into the adjoining bay. The capitals of the pillars follow the recessing of the arch above and are ornamented with the scallop (pl. XXII, XXIII, p. 17). The bases have the small flat moulding characteristic of the period, measuring only 2½ in. wide and 2½ in. high.

On both the north and the south sides of the quire the abacus of the capital is carried as a string across the face of the compound piers from one half column to the other, but it is only carried some 18 in. on the face of the eastern piers of the crossing, probably indicating the point to which the canons' stalls extended eastward (pl. XXII). The mouldings of the base of these great piers project only 1¼ in., to allow of the stalls being placed against the face of the wall.

On the north face of the compound pier, on the south side of the quire, is the 2¼ in. set-back already alluded to (pls. XVII, p. 9, XXII). (fn. 3) It extends up to the string below the clerestory but does not appear on the south face of the pier in the aisle, though it reappears in the triforium above. (fn. 4)

Twelve inches above the arches of the arcade, and immediately below the triforium, is a bold string which, on the north side, stops 2 ft. 2 in. short of the western angle of the crossing pier, but on the south side it is carried right up to the angle (pl. XXII), another indication of the rebuilding of the north-east pier.

Some of the piers were renewed and some largely repaired in 1864, as will be seen by reference to the plan above, p. 3. (fn. 5) Coney's engraving made for Caley and Ellis's edition of Dugdale in 1818 (fn. 6) shows the arch of the north-west bay supported by a corbel with Perpendicular mouldings on the capital, instead of a half-column with twelfthcentury mouldings as now, showing that the rebuilding of that side of the pier of the crossing, in 1405, affected this bay. Presuming that Caley's drawing is correct, it would have been well if that feature could have been retained.

On the south side of the quire, Malcolm's (fn. 7) drawing made in 1803 shows that, after the suppression, the piers of the ground arcade suffered very materially, the half-column on the western side of the compound pier on the south side of the quire having been hacked back. Other prints show the piers covered with wainscot with hatpegs attached! (Pl. XXVI, p. 20.)

In the twelfth century the whole of the stone work in the quire was coloured, as was usual. The voussoirs were treated in bands of red, black, and yellow in succession, each being 5 in. to 6 in. wide. This arrangement on the face of the walls was also carried through the soffits of the arches of the arcade, as was found in the quire beneath the whitewash in the restoration of 1864, (fn. 8) and again, in 1892, in the recess in the east wall of the north transept. Fragments with the colour still adhering have been found of the thirteenth-century work of the nave, the roof of the Lady Chapel and portions of canopied work of the Decorated period, showing that the interior of the church was coloured throughout.

On the conversion of the apsidal end of the church to a square termination, about the year 1405, the five eastern bays of the apse were cut off (as explained above), (fn. 9) whilst the western bay on each side, with the adjoining three bays of the quire, went to form a square presbytery. The floor of this presbytery was raised about 2 ft. 3 in. above the twelfth-century floor level (as already mentioned) (fn. 10) and was probably approached at this period by five or six steps.

The north side of the presbytery was occupied in its western bay by the founder's tomb, (fn. 11) the mural arcading or panelling of which was carried across the eastern bay. (fn. 12) In this latter bay was a priest's door to give access to the high altar (pl. VI a, Vol. I, p. 78). It has been suggested that the monument, as well as forming the north wall of the sanctuary, was also itself a southern 'parclose' to a chapel built by Roger Walden for his own chantry. This is quite possible, although there are now no remains of any walls, or signs of any walls having existed, on the north side of the monument. Withers, however, records in his diary (October 1864) that the excavation then going on 'in the north aisle near the east end had disclosed another block of masonry and the base of a tomb carried round on the west side of Rahere's in the north aisle, probably originally a portion of the same'; and again (February 1867), that 'a low piece of wall at the back of the monument had been taken up and some exquisite fragments of transition Norman work discovered therein'. It is quite possible that this masonry was a portion of the walls of a chantry chapel on the north side of the tomb built by Roger Walden for his chantry, though his chantry was eventually founded at St. Paul's, as he was Bishop of London when he died. (fn. 13)

The straight wall on the south side of the sanctuary filled both bays. We have no record concerning it before the suppression, but Sir John Deane, the first rector, desired, in 1563, to be buried in the western bay, Sir Walter Mildmay's monument being placed later in the eastern bay. (fn. 14) Hayter Lewis records that the latter cut very awkwardly into one of the main piers and arches. (fn. 15) It is shown by Malcolm (fn. 16) surrounded by iron rails (pl. XXIX a, p. 23), with the pier to the west of it cut square. Withers, in his diary, (fn. 17) records (February 1865) that when the south wall was taken down 'numerous small fragments of beautifully carved stone work' were discovered, and in a cavity at the back, above the floor level, were the coffins of Sir Walter and Lady Mildmay, which, together with the monument, were at that time removed to the present position in the south aisle.

At the east end of the quire the straight wall was carried to a height of 23 ft. 6 in. to the level of the springing of the arches of the triforium. Remains of this wall, some 2 ft. wide, still project both on the north and the south sides at the triforium level. The wall was surmounted by two large four-light windows, the shafts on the jambs of which, together with the springers of the window arches, and a small portion of a stone string that ran below the sills, were found, at the restoration of 1865, concealed behind the plaster. (fn. 18) The shafts, which have a plain moulded base and capital, measure 16 ft. 3 in. in height: the capitals range with the springing of the arches of the clerestory, the bases with that of the triforium. By means of the tracery of these windows, which was found in the year 1885 among the eighteenthcentury brickwork, and by means of the window jambs and springers, it has been possible to reconstruct on paper these two windows (pl. XXVa, p. 19). They probably remained until the eighteenth century, because, in the year 1704, 'the east window' was ordered by the vestry to be repaired in stone, (fn. 19) but later on, either in 1720 or 1791, they were taken out and replaced by two round-headed Georgian windows built in brick. The windows, we may assume, were filled with stained glass in the fifteenth century, from fragments that have been found from time to time.

The high altar, (fn. 20) in 1405, was placed on the west side of this straight wall, which was built with an ashlar face. Early in the nineteenth century this wall was found to be painted red with black stars, (fn. 21) probably the original decoration.

Eastward of this wall was another wall enclosing a space about 7 ft. wide, known in later times as 'purgatory' (pl. XXXIII, p. 27). This may have been a corruption of the word 'presbytery', or the place may have been so called because the chamber was, in the early nineteenth century, filled with human bones. (fn. 22) It was entered by a door in its north end, close to the priest's door, east of Rahere's tomb. The floor was lower than that of the ambulatory, for we read of steps leading down to it. (fn. 23) It was lighted by a small window above the door. (The arches, shown in some of the drawings (fn. 24) on the east face of this wall, were made of Roman cement and merely placed there for ornament by John Blyth, the architect, early in the nineteenth century.) This eastern wall, in 1405, was carried up parallel with the wall of the high altar to the height of the walls of the Lady Chapel. A floor was thrown across the space between these two walls at approximately the same level as the triforium floor, thus forming a covered passage-way (fn. 25) from one triforium to the other behind the high altar (p. 7). (fn. 26) This arrangement continued up to the time of the suppression. Afterwards the Lady Chapel, as will be seen, was converted into a private dwelling-house and then this passage gave access from it to an eastern bay of the north triforium, for use as 'a chapel chamber'; and later, when the Lady Chapel was used successively by printers and fringe-makers, the passage gave access to rooms in the south triforium.

The old engravings by Malcolm in 1803 and Wilkinson in 1822 (pl. XXIX, p. 23, and pl. XXVI, p. 20) give an idea of the appearance of the straight east wall early in the nineteenth century. They show a large altar-piece reaching in the centre to nearly a third of the height of the two round-headed windows. It consisted of a painting of a Tuscan temple with columns, arches, and obelisks surmounted by the royal arms, and is thus described in a parish document of the year 1821: (fn. 27) 'An altar piece 32 ft. high consisting of a very spacious piece of architecture painted on canvas; between the columns are painted the Commandments, the Creed and the Lord's Prayer. On the upper part are the arms of King Charles I with the initials C.R.' The altar itself is the most insignificant thing in the picture, being much lower than the pews. It is closely surrounded by railings the same height as the table and entirely without ornament of any kind. In the year 1828, when some £800 was spent in alteration and repairs to the church, this altar-piece was taken down and replaced, by Blyth, with an arcade of small arches in Norman style of the same height as the arches of the ground arcade rendered in Roman cement, with a plain wooden panelled screen behind the altar (pl. XXVII, p. 21). (fn. 28)

In the year 1863 Hayter Lewis and William Slater advised that the fifteenth-century square termination should be altered back to the apsidal one of Rahere's time. The grounds for this recommendation were that the square end contained no Perpendicular work of such importance as the Norman work which still remained hidden by the walls of 'purgatory'. Of moulded stone work of the former period there were only the shafts of the jambs of the east windows; whilst of the latter there remained two bays of the ground arcade, one on each side, with their four piers and two arches embedded in the straight east wall and the wall of 'purgatory', neither of which had any architectural feature. The architects' advice was carried out with the entire approval of the leading architects of the day. (fn. 29) The square presbytery was altered by removing the straight east wall and the Mildmay tomb on the south side; and, whilst retaining the Rahere monument on the north side, the continuation of the monument with the priest's door in the adjoining bay was removed. (fn. 30) This last is a matter of regret, for the removal was not necessary. The east wall and the walls of 'purgatory' were removed in March of 1865 to the height of the abacus of the capitals of the piers. The two centre piers were rebuilt and the sanctuary enclosed by means of wooden panels placed between the piers. The east ambulatory still remained part of the church, but on the first floor everything on the east side of the straight east wall was in possession of the fringe-maker, whose premises therefore extended some 17 ft. into the church; and, as all efforts to induce him to relinquish possession failed, his factory had to be supported by an iron girder upon iron columns placed within the sanctuary (pl. XXVIII, p. 22).

This disreputable state of things continued for twenty years, until, in 1885, a second effort at restoration was commenced, under the direction of Sir Aston Webb, when the fringe-maker's property was purchased. The question of the retention of the square end, or of a return to the apsidal end, had been settled twenty years before by the completion of the ground arcade: the apse, therefore, had to be completed, as it now exists, the patron, the Rev. F. P. Phillips, bearing the whole of the cost (Frontispiece, Vol. II).

The round-headed windows with the brick wall were removed, but the shafts of the fifteenth-century window jambs and the portion of the wall supporting them were retained. The five bays of the triforium were rebuilt in Norman style, to correspond with the existing triforium, with a subsidiary arcade of two arches and a central shaft. To make a difference from the old work the shafts were fluted and the tympana were hatched. The mouldings of the arches follow the old ones, those of the centre arch being—as already stated (fn. 31) —the original stones. Mouldings of the arches of the ground arcade below were found during the demolition of the windows, but, as the arches had been rebuilt twenty years earlier, these stones could not be re-used, so they have been set up in the north triforium. The clerestory bays of the apse were rebuilt to harmonize with those of the quire in early Perpendicular style, but, as already stated, a slender shaft (fn. 32) has been carried between each bay from the base of the triforium to the cornice above the clerestory to show that these two stories were built at one and the same time.

To incorporate the jamb shafts and springers of the fifteenthcentury windows into a harmonious whole with the new work, Sir Aston turned an arch across the apse springing from these jambs, thus preserving all the indications of the square east end (fn. 33) and forming a sanctuary arch with a straight wall over. This arrangement had also other advantages; the fifteenth-century clerestory, built when the eastern termination was square, ran in a straight line unlike the triforium below, which, being built when there was an apsidal termination, had begun to curve (as already stated) before the fifteenthcentury wall was reached. The retention of these jambs, with an arch above, enabled the junction of the straight clerestory with the curve of the apse to be masked and the new work to be carried out without any interference with the twelfth- or fifteenth-century work. On the other hand, it fixed the cord of the apse at the position of the fifteenth-century straight wall instead of at its original position one bay farther west, thus practically reducing the apse to one of five bays instead of seven bays, as in the twelfth century. The altar now stands on the cord of the new apse.

The pulpit formerly stood against the compound piers on the north side of the quire. The pre-suppression structure remained until 1828, when Allen says it was destroyed by the clumsiness of a workman in an attempt at its removal during repairs then going on. (fn. 34) It was a wooden pulpit with five sides and Gothic traceried panels of the late Decorated period, (fn. 35) painted red, (fn. 36) and a wood panelled back carrying a large sounding-board (pl. XXIXa, p. 23, and pl. XXX, p. 24). In place of this pulpit there were erected, in 1828, two polygonal massive wood panelled pulpits on pillars (pl. XXVIII). (fn. 37) The present stone pulpit is the work of Sir Aston Webb in 1893. Like an ancient ambo (fn. 38) it has two flights of stairs, one on the western side for ascending and one on the eastern side for descending. The centre projects, forming a three-sided bay supported by a moulded corbel. The front has six panels, four of which are pierced and cusped (pl. XVI, p. 8).

The Ambulatory of the Quire.

The ambulatory of the quire formed a spacious processional way which, as was said when describing the plan, encircled the apse and gave access to the chapels. (fn. 39) It contained in all fifteen bays divided by heavy transverse arches, and these, for easier reference, are here designated numerically, commencing with that on the east side of the south transept (see plan, p. 3).

As already mentioned, (fn. 40) the ambulatory was originally vaulted throughout with simple groined vaults, as at the chapel of St. John in the Tower, but with certain important differences. These differences are admirably described in the late Sir Gilbert Scott's Lectures (fn. 41) thus: 'At first sight the two may appear to be similarly treated, but on closer examination there will be found to be much difference between them. In the Tower chapel the transverse ribs are made to increase prodigiously in width towards the outer wall, so as to reduce the want of parallelism of the groined compartments, a very unsightly expedient, and the capitals of the columns are square, which makes the backs of the arches they support nearly double the width they present in front: while at St. Bartholomew's the ribs are of uniform width, and the capitals, instead of being square, have their sides radiating from the centre of the apse, so as to share with their arches the spreading of their outer sides' (Fig. 2, p. 42).

Only two bays of the vaulting at St. Bartholomew's—those at the east end of the north ambulatory—may be original work, the others having all been renewed at different times.

After the fire of 1830 we have a record (fn. 42) that, the south wall having to be shored, the damaged vaulting was seen to be rubble, and that 'lathing' and 'tiling' was carried out over the aisle, referring apparently to the first three bays on the south side. In November of the same year, Mr. Blyth, the Surveyor, was instructed by the vestry 'immediately to take down the two faulty groins adjoining the vestry-room' (the fourth and fifth bays) and 'reinstate them like the others'. (fn. 43) These latter vaults of laths, tiles, and plaster were taken down in 1891 and replaced in concrete.

The vaulting of the three western bays on the north side (the thirteenth, fourteenth, and fifteenth) probably gave way after the abutment of the chapels on that side was removed in the sixteenth century. In September 1866 the transverse arches of these bays were rebuilt, (fn. 44) but from lack of funds the vaulting was not renewed until 1893. In the interval the floor boards of the school above alone separated that noisy assembly from the church.

The transverse arches of the vaults have flat soffits and square arrises. The responds or pilasters on the walls which carry the arches have capitals which are variations of the scalloped capitals of the columns of the quire (pl. XXXI a, p. 25). The first and third pilasters in the south walk were renewed in 1865; the rest, including those in the north walk—with the exception of that of the arch opening into the transept—seem to be original.

In the first bay of the south walk of the ambulatory of the quire is an arched doorway which led to the sacristy. It probably dates from the fourteenth century. It was at some time after the suppression closed and hidden under plaster; it was only uncovered in 1865. The stone sill, which is much worn, is 1 ft. 8 in. above the present floor level; (fn. 45) but there is evidence of there having been either one or two steps up to it. The present door was made in 1867 (fn. 46) and remodelled in 1914. The transverse arch, and the responds which carry it, between the first and second bays are thicker and stronger than the others in the ambulatory and are of two orders (pl. XXXV, p. 29). The second order on the west side of the arch and on the east side of the arch on the opposite side of the bay westward are either concealed by the plaster vault erected after the fire of 1830, or were never built.

In the second bay is the Mildmay monument, removed here in 1865 from the south side of the sanctuary (pl. XCIX b, p. 451). (fn. 47) This and the other monuments are described in a later chapter.

The wall of the third bay is occupied entirely by one large arched opening. The arch is four-centred, is slightly pointed, and has a shallow moulding carried round and brought down to within about 5 ft. of the present floor level. It would seem to date from Prior Bolton's time. It opened into a space paved with large red tiles, between the sacristy and the south apsidal chapel. This opening was finally built up in 1914 and the Roycroft monument fixed to the filling.

In the fourth bay is a memorial with the half-length figure of Edward Cooke, moved here in 1864 from the south wall of 'purgatory' (pl. XXXI b, p. 25). (fn. 48) On the opposite side of this bay, between the columns of the arcade, the organ was placed in 1867–1868, and there remained until 1886. Since then, the inlaid brass memorial to the first rector, Sir John Deane, has been placed on the floor where the organ stood. (fn. 49)

We now enter the apsidal portion of the ambulatory.

In the fifth bay is the twelfth-century opening to what was originally the south chapel, believed to be that of St. Stephen (pl. XXXI b, p. 25). In 1914 a choir vestry was built on what remained of its walls. The opening at its entrance from the ambulatory is 6 ft. 9 in. wide—as already seen (fn. 50) —and 13 ft. 9 in. high to the soffit of the crown of its roundheaded arch.

The vaulting and original walls of the sixth and seventh bays have been taken down, excepting a small section, about 2 ft. in width, of the wall of the seventh bay, which remains in the south-west pier of the Lady Chapel. The demolition to the south of this pier must have been the work of Prior Bolton when he formed the present rectangular end, as already described. (fn. 51) The vaulting here has been replaced by a flat ceiling which was renewed in wood in 1886. In the south wall of the rectangular end is the doorway which was formed by Bolton as an entrance to the church from the new Prior's House that he built (pl. XXXVI b, p. 30). It bears the 'Bolt-in-tun' rebus of the prior (fn. 52) in the spandrels of the arch on the church side. The walls are of brick, a material in vogue in the time of Henry VIII. They are thickly covered with plaster and thus would have corresponded with the walls of the ambulatory, which, being built of rubble, were no doubt similarly treated.

Bolton's doorway is 1 ft. 11 in. above the present church floor level, and the present top step is 7 in. higher still.

In the eighteenth and first half of the nineteenth century this doorway was used as an entrance to the church. It was approached by a flight of steps descending from east to west on the outer side of the wall. At that time there was, as now, a window in the east wall of the rectangular end; but in 1849 the then rector, John Abbiss, who had built a school-house which obstructed Bolton's doorway, made a new entrance with a flight of seven steps by removing the window in the east wall. So it remained until 1864, when an eastern entrance was given up, and the window again inserted, as now. Bolton's doorway, after 1849, formed an entrance from the Infants' room of Rector Abbiss's school to the church. Owing to the lowering of the church floor in 1864 a flight of wooden steps was made for this doorway, but these were replaced by the present stone ones on the occasion of the royal visit in 1893, when the north transept was re-opened after restoration.

The small stone balcony outside the door was built for safety when Rector Abbiss's school-house was removed in 1914. The present doors were made in 1866. (fn. 53)

Until the latter year the south walk of the ambulatory was shut off by two sets of double doors erected in the year 1720, (fn. 54) one being placed under the vaulting arch east of the entrance to the south chapel, the other under the arch on the west side of the south transept (pl. XXXII b, p. 26). The east end of the north walk was enclosed in the same way (pl. XXXIII, p. 27). (fn. 55) These doors were probably erected to keep out the draughts from the various entrances; but also perhaps to exclude what Malcolm describes as 'the humid exhalations from a number of bones in a semicircular dungeon (purgatory) at the back of the altar', which, he remarks, 'made the air unpleasant'. (fn. 56)

In the east end of this south walk some tiles, found during the restoration of 1864, have been fixed to the small piece of the north wall for preservation. They are considered to be not earlier than the fourteenth century.

Here also stands the church chest. It is of oak and has three locks, but it is perfectly plain and there is nothing to indicate its date. It measures 5 ft. 3 in. long, 2 ft. wide, and 2 ft. 4 in. high.

We now enter the eastern part of the ambulatory. The vaulting of the three centre bays (the seventh, eighth, and ninth) must have been first taken down when the Lady Chapel was rebuilt about 1335 (fn. 57) because the increased width of the opening into the chapel necessitated the removal of the responds which carried the transverse arches between these vaults. The small triangular portion left of bays seven and nine was then revaulted by turning a new transverse arch, of similar design to that removed, across the ambulatory and by filling in the triangular space with a section of a simple barrel vault. The new transverse arch was kept a little back from the face of the extension of the chapel wall above, and a second or mural arch, in the style of the period, was formed at a higher level to carry the wall, the space between it and the arch below being filled in (pl. XXXIV, p. 28).

These mural arches, both on the north and on the south sides of the entrance to the chapel, and the section of the barrel vault on the north side still remain, but the vault on the south side was again removed when Bolton formed his square end to the aisle, as previously described. (fn. 58)

It may be that the portion of the ambulatory enclosed between the extensions of the chapel walls was also revaulted at this time, but in the style of the period, for there remain above the capitals of the columns of the apse, which carry the fourteenth-century arches, two pieces of decorated moulding brought down to small corbels (broken away from that on the north column) which may have been portions of stilted springers of such a vault (pl. XXXIV, p. 28), and provision to carry a vault on the east side of the space could have been made by means of an open arcade across the entrance to the chapel proper. There are, however, it must be admitted, no signs of vaulting above the mural arches of the chapel which span the ambulatory, nor is there any trace of such an arcade, and the existence of hood moulds to the mural arches is against the probability that a vault was ever carried out. It would seem most likely, therefore, that such a vault—if ever intended—was abandoned before the rebuilding of the chapel had proceeded far. In any case, if new vaulting was built, it must of necessity have been taken down again when the high altar was rebuilt about the year 1405, and a straight east-end wall substituted for the apse, because the two centre columns of the apse were then removed, which would have involved the destruction of the vaulting.

The extension of the Lady Chapel walls that span the ambulatory converge about 1 ft. 2 in. each, because the columns of the apse which carry them at the one end are only 19 ft. 8 in. apart, while the width of the chapel is 22 ft. The mural arches of these converging walls spring at their east ends from the capitals of decorated wall shafts, of which capitals that on the south side has much perished and that on the north is also damaged.

The arch on the south side is original as to its eastern half only and has a hood moulding (as mentioned above), which at its eastern end terminates in a carved head. The arch on the north side is original throughout, though patched in places in 1886. The hood moulding of this arch has perished, but the head at its east termination remains well preserved.

When Sir Richard Rich came into possession of the monastery in 1544 and converted the Lady Chapel into a dwelling-house he formed in it a first floor which he carried westward across the ambulatory, 13 ft. 6 in. above the present floor, or 11 ft. above the then existing floor, resting on the walls of the 'purgatory' chamber behind the altar and entirely covering the western mural arches of the Lady Chapel (pl. XXXII a, p. 26). He also incorporated the gallery above 'purgatory' into his house, but for some reason he did not absorb either 'purgatory' or the eastern portion of the ambulatory. (fn. 59) The whole encroachment was swept away in 1886.

Of the original walls of these three bays of the east ambulatory there remains that portion which forms part of the south-west pier of the Lady Chapel and a corresponding portion of the north-west pier. In this latter pier the twelfth-century work has been brought forward or 'made out' on its western face by the fourteenth-century builders to provide a better springing for their arch across the ambulatory, and on this 'making out' the Decorated shafts have been worked. On the south-west pier the fourteenth-century builders satisfied themselves with merely plastering wedge-shaped stones on to the Norman walling, some of which stones have now come away, revealing the Norman splayed face of the apse behind.

This is not only of interest as showing, as it does, that part of these piers is Norman masonry, and establishing with exactness the line of the original apsidal walls in these bays, but also because it shows by the twelfth-century tooling on the return faces that these Norman stones were originally quoins and that therefore there must have been an opening, or at least a recess, to the immediate south of this pier. Such an opening still exists in a corresponding position by the north-west pier of the chapel. In this latter case the inner part of the north reveal appears to be twelfth-century, and the inner arch and south reveal fourteenth-century work. The line of the latter is at a different angle to and in front of a line corresponding to the Norman face of the south-west pier.

The original purpose of these openings is somewhat obscure, as is also the reason for their having been placed out of the centre of their respective bays. The opening that remains on the north side measures in width 3 ft. 6 in., and in height from the present church floor to the crown of the arch 9 ft. 2 in.; whilst outside the church there is a covering brick arch which measures a further 1 ft. 8 in. to the crown, suggesting an outside stair: on the other hand, when the arch was opened there was found what may have been an original filling of chalk and stone 4 ft. 4 in. in height. It is therefore possible that these openings were originally Norman windows, and that there were turrets at the junction of the original Lady Chapel with the apse, which made it necessary to place the windows as far to the north and south respectively as possible: had these openings themselves led to such turrets we should assume that they would have been placed nearer to them. At Tewkesbury there are turrets in this position, but they were not apparently carried down below the Lady Chapel roof. At Peterborough such turrets are approached from the clerestory passage.

We now pass to the tenth bay, which is one of those which apparently retains the original vaulting, though now cemented over. The whole of the outer face of this bay is occupied by a round-headed and probably original Norman opening, the upper part of which was filled in as a mullioned window in 1864–1865 (pl. XXXIV). (fn. 60) The filling below the window is probably post-suppression. It is thought that when—as is described later (fn. 61) —the chapel of St. Bartholomew was rebuilt about the end of the fourteenth century, it was placed outside this bay and was entered through the arched opening, and that when the chapel was destroyed after the suppression the opening was roughly filled up, as we see the lower part to-day.

The next bay (the eleventh), which also seems to have its original vaulting, contains the entrance to the twelfth-century north external chapel (pl. XXXIV), corresponding to that on the south side (the fifth bay). This Norman arch was at first—as is shown later (fn. 62) —the entrance to St. Bartholomew's chapel, but now leads to a modern building used as a robing room, and to a turret stair to the old school-house, now used as a sacristy.

The twelfth-century chapel has been completely destroyed, probably when the Walden chapel was built and when the St. Bartholomew's chapel was 'newly founded' at the end of the fourteenth century. This twelfth-century arched entrance may later have been used as an entrance to St. Anne's chapel as suggested presently. (fn. 63) The opening has been partly filled and a doorway formed in the filling. The door is modern, but the frame is Tudor and was brought from the schoolroom above when the west approach to the school was removed in 1866. (fn. 64)

On the south side of this bay was the fifteenth-century priest's door to the sanctuary already referred to. (fn. 65)

With the next bay (the twelfth) we enter the north ambulatory of the quire (pl. XXXV). The vaulting of this and the next three bays is a restoration of 1891. (fn. 66) The original vaulting was removed in 1791, for there is a copy of an agreement among the parish papers of that year 'to take down the stone groin in the north aisle and to replace with good sound lath and plaster' and to relay the school floor above. (fn. 67) The object may have been to lessen the weight of the thrust against the north wall, then getting into a precarious condition for want of abutment. To the south of the twelfth bay is the back of Rahere's monument. The tiles which covered the open tracery were removed in 1867. The rubble wall, in part covered with plaster, shows where the monument, without authority, was opened in 1866. On the opposite side of the bay and about 8 ft. from the floor is a square deeply-splayed mullioned window, of the date of which we have no record. It was probably inserted after the demolition of the old parish church some time in the reign of Elizabeth, (fn. 68) and it is likely that the filling in below dates from the same time.

Below this window, outside the church, a wall was built in 1866, parallel to the church wall, forming a long narrow cupboard.

The last three bays (the thirteenth, fourteenth, and fifteenth) form an arcade on the north side consisting of three Decorated arches inserted within the Norman arches of Rahere's work. They opened into a long chapel built by Roger Walden about the year 1395, (fn. 69) called the All Saints or parish chapel—the latter because it was an extension of the parish chapel in the north transept. (fn. 70) The floors of the fourteenth and fifteenth bays were always, as now, 1 ft. 3 in. to 1 ft. 4 in. above the present church floor level. (fn. 71) The level of the thirteenth bay, however, is 3 ft. 8 in. above the church floor, but that it was not so originally is shown by the evident 'make up' character of the bases of the shafts which are upon it.

The arches are good examples of the late Decorated period; their outer order is continued down without capital or base, but the inner order, which consists of shafts, has both. At the junction of the two orders is a deep hollow.

In 1864 the work was found to be much decayed, so that only the thirteenth bay consists entirely of original work, but the bases and the lower part of the shafts and jambs of the other two bays are original and the remainder was at that time carefully restored. (fn. 72)

The capitals have an octagonal abacus with circular bell and necking; the bases have the upper members circular overhanging the octagonal plinth.

A drawing made in 1864 (fn. 73) shows the capitals, shafts, and bases of the thirteenth bay and an iron grill in the upper part of the arch (the holes of which still remain), but the other two openings are shown entirely filled with rubble and plaster. It may be that all the openings were originally furnished with grills.

The filling behind the arches and piers together with the windows is in each case the work done in 1865. Investigation shows that on the north side of these piers, though now covered, is faced work proving that they once stood free. The filling is 4 ft. thick and forms a much-needed buttress. The north wall of the church had long been a cause of anxiety. In 1812 the vestry directed a buttress to be placed against it. In 1820 the wall was ordered 'to be immediately repaired'. When Walden built his chapel the walls between the three western bays were taken away, his chapel giving the necessary abutment, but when it was pulled down the thrust of the aisle vault seems to have pushed the remains of the north wall outwards. In 1865 the straightening of this wall was an important part of the work of the Restoration Committee. Withers records in his diary on September 20th of that year that the western end had been made perpendicular by means of a screw-jack, and five days later that a portion had been pushed in three inches. (fn. 74) A buttress was then built behind each of the piers of the north wall and the space between was roofed and the ends enclosed by the insertion of the present mullioned windows, which, besides strengthening the wall, enabled the jambs of the arches opening to Walden's Chapel to be left exposed.

Between the fourteenth and fifteenth bays the pilasters, both against the external wall and on the compound pier, are of two orders, as in the corresponding bays one and two on the south side; and the transverse arch replaced in 1866 has been made to agree with the arch between those two bays (i.e. with two orders on the east side and only one on the west): consequently the present vaulting of bay fifteen (inserted in 1866) has had to follow that of bay one, and may or may not be as originally built.

The arch at the west end of the ambulatory was built up (fn. 75) when the north transept, into which it opened, was pulled down at the time of the suppression. It probably formed one item of Hugh ap Harry's Bill (recorded in the Augmentations in 1542) rendered to the king for 'repairs to the late priory £80'. (fn. 76) Later on, at the back of this west wall, a stair was built leading to the boys' school in the triforium, which was taken down again in 1866, in order to re-open the arch. (fn. 77) In taking down the wall Withers records the finding of a crowned head, the fragments of a canopy, and two Norman capitals, all of which are now in the museum in the cloisters, and the matrix of a brass now fixed in the cloister floor. The crowned head has been a corbel and is very similar to the two at Westminster Abbey from which the label springs on the doorway below the Abbot's Chamber built between the years 1360 and 1390. At the base of the wall was found a portion of the old tile pavement then still in position. (fn. 78)

The stone coffin now preserved in this last (the fifteenth) bay was taken from the western bay of the arcade under the north arch of the crossing in 1865: (fn. 79) it measures 6 ft. 10½ in. in length and 2 ft. 2½ in. in breadth at the head and 1 ft. 3 in. at the foot. The leaden coffin, now with it, was found in the burying-ground of the canons where the schools now stand.

The Triforium of the Quire.

The triforium arcade corresponds in spacing and in general design with the main arcade. It measures an average of 13 ft. 8 in. in height from the under side of the string below the arcade to the under side of the present fifteenth-century string above the arcade. When first built it was some seven inches higher, as shown by the Norman strings still existing in the west bay. The pilasters on the jambs of the arches have capitals, ornamented with the scallop on which is a very narrow impost only 2½ in. wide, from which spring, not only the triforium arch, but also the arch of the lesser arcade (pl. XXXVII, p. 31). In this, Rahere's triforium differs from any known to the writer in England. It would appear, as has already been said, (fn. 80) that Rahere intended to have an open arcade, as he made no provision for the present lesser arches. At Malmesbury the lesser arcade consists of four arches, as here, but the end arches rest on a half-shaft built for the purpose. At Gloucester, Ely, Peterborough, Romsey, and Shrewsbury there are lesser arcades of three arches, but in each instance there is special provision for carrying the terminal arch. At St. Remis at Rheims the round-headed arches of the triforium of the transepts are still open, (fn. 81) as we believe Rahere's to have been, but in the nave they have been filled, and the arches rest, as here and also at Basle Cathedral, on a narrow impost only. Anyhow, it is clear that the filling did not take place until after Rahere's time, because the capitals of the circular shafts are of a design of a later date, (fn. 82) those on the south side dating from about 1150, those on the north from about 1160 (pls. XXXVI a, p. 30, XXXVII, p. 31).

On the north side of the church the westernmost shaft of the west bay is longer than the rest, necessitated by the settlement that took place in the north-east pier of the crossing. If this shaft is original—and it has every appearance of being so—it shows that the settlement took place before the triforium arches were filled, and it may be assumed that Prior Thomas filled the arches for strength after the weakness in the north-east pier of his crossing showed itself. This was the case at Furness, where several arches near the tower were entirely filled for strength when the centre tower was heightened to carry more bells.

The arch of the fifth bay from the west, that is the one adjoining the present cord of the apse, is walled up entirely on both sides of the church (pl. XX, p. 12). (fn. 83) This was evidently done in the early fifteenth century when the east end was reconstructed. The shafts and capitals appear to have been renewed at the same time.

On the north side the other arches of the lesser arcade have all been walled up at one time or another since the suppression. In Coney's drawing in Dugdale, in 1818, (fn. 84) all the arches are so shown and, with the exception of the second from the west, they are shown as plastered over in a manner completely to hide the shafts and their caps. Storer and Greig's (fn. 85) and Coney and Skelton's views in 1804 show them in a similar condition; but Billings' drawing of 1838 (fn. 86) shows all the shafts free, though the openings were still filled, which was the work of John Blyth, the architect, in 1837.

The east triforium, as originally built, no doubt followed the line of the ambulatory arcade below. That it must have been remodelled when the Lady Chapel was rebuilt is evident, as the arches then formed arose to a greater height than the Norman floor level. How far this remodelling was carried is not known, but the remodelling of the apse, as already described, must have necessitated the entire demolition of this part of the triforium. (fn. 87)

On the south side, where the capitals of the shafts are not so varied, Malcolm, in 1803, shows the two western bays built up, but with the shafts and arcading still visible; the two eastern bays he describes as 'totally defaced and their outline barely discernible'. (fn. 88) The central bay is filled with Prior Bolton's window, built about the year 1517 (pl. XXXVIII, p. 32). It has two tiers of five lights, the upper tier having trefoil cusps at the head of the lights, of which there are three in the front and one on each side of each tier. They were always glazed, as now, as is shown by the original rebates which remain, but some time after the suppression the window was bricked up, for in 1836 the churchwardens paid for 'removing the brick filling into lights and glazing with lead lights and iron-work.' Above is an embattled cornice. Below are three quatrefoil panels in front and a trefoil panel on each side. The centre panel contains Bolton's rebus in low-relief—a crossbow 'bolt' piercing a wine 'tun'. Each of the other panels contains a plain shield.

There are several instances of similar internal windows in churches: thus at Peterborough and also at Gloucester there was an 'abbot's gallery chapel'. (fn. 89) In that at Gloucester is a window looking into the Lady Chapel from the east triforium in which the original mensa of the chapel altar remains. At Fountains the abbot's window opened into the chapel of the nine altars from the south end. At Christ's College, Cambridge, there is a window (restored by Bodley in wood) in a similar position to Bolton's and probably built by him. It opens from a chamber known as Lady Margaret's prayer room. At Westminster the abbot's window, with a balcony, opens into the nave at the west end of the south side. Probably this was used for observation, as the one at Malmesbury in a similar position, or Mr. Francis Bond suggests that it may well have been erected for the use of the semi-choir in the Palm Sunday procession. (fn. 90) There are other instances, as at Chichester, Oxford, St. Albans, St. George's, Windsor, and at Worcester, where the prior's oriel opens into the north aisle of the quire at its western end. For all of these various uses have been assigned. (fn. 91)

Prior Bolton's window, in the third bay from the west, no doubt opened from the new prior's house that he built, and enabled him to witness mass at the high altar.

In the fifth bay from the west, in the fourteenth century, was the domus inclusa occupied by John Mirfield, the physician, as already related. (fn. 92)

In the early part of the sixteenth century Prior Bolton greatly altered the south triforium internally when it was remodelled by him in connexion with the rebuilding of the prior's house. (fn. 93)

Bolton's house, as will be seen later, ran south from the east end of the south ambulatory of the quire, with which, on the ground floor, it communicated by the doorway known as Prior Bolton's door. On the triforium level, between the west wall of the house and the east wall of the sacristy, Bolton built a gallery parallel to and against the south wall of the triforium, very probably of wood like the one at Leicester Hospital at Warwick. (fn. 94) This gallery extended over the south chapel (fn. 95) and over the space between that and the sacristy (a space which was probably enclosed at this time); but it did not extend over the sacristy itself, as that was a lofty building. The floor of the gallery was probably at the same level as that of the triforium, as from Lord Holland's Rental of 1616 we learn that there was access from one to the other, and no mention is made of steps or stairs. We consider that Bolton used the eastern part of the south triforium as his private chapel, and it is possible that it was he who converted the two western bays into the 'chapel chambers', as they are subsequently described in the same Rental.

Bolton's private chapel would have extended from the bay in which his window is placed eastward as far as the line of the fifteenthcentury square east end of the church; it seems to have had an east window overlooking a flat or low roof above the square east end of the south ambulatory, because, in 1833, it was thus described by Allen: (fn. 96)

'It was one of the apartments erected by Bolton, and still exists in nearly a perfect state; it is now divided into two apartments; the walls are wainscoted with small panels, each contains a curious scroll-formed ornament, the roof also is of timber and panelled into square compartments; at the points of intersection are flowers; at the east end is a large window with wooden mullions; it is bounded by a low pointed arch, on one of the spandrels of which is the device of Bolton. In the window is a shield with many quarterings; the arms of Rich, a chevron between three [crosses botony], is the only one perfect; the same arms appear on the front of a house in Cloth Fair.' (fn. 97)

Although all this has now disappeared, we still have the name 'Bolton' incised in sixteenth-century characters on the wall where the straight east wall of the quire projects into the triforium.

After the suppression, the south triforium appears to have remained without further alteration for a long period, for in Lord Holland's Rental of 1616 we find it described as still forming part of what had been the prior's house, and in the same condition as we believe Bolton left it. In the second half of the seventeenth century, however, the two western bays were converted into a gallery of a Nonconformists' Meeting House, described later, (fn. 98) and so remained until destroyed by fire in 1830.

The eastern part of the triforium continued in the occupation of the tenant of the prior's house, probably until the middle of the eighteenth century, but we then find, by the parish rate-books, that this part was being used as a school, one Alice Russell being rated for the premises from 1748 to 1755. As there was a church school in the north triforium we assume that Alice Russell's was a Nonconformist school; at any rate her successor, Edward Cook, carried it on as a Presbyterian school, for he was rated for the school from 1756 to 1761. He was rated at the sum of £10, from which we gather that the school occupied the whole of this eastern portion of the triforium, or Bolton's gallery.

Edward Cook was succeeded by John Hitchcock until 1771, when his lease apparently came to an end, for the vestry in that year opened negotiations with the patron, Mr. W. Edwardes (a descendant of the Rich family), through the then rector, the Rev. O. P. Edwardes (also a descendant of Rich), (fn. 99) with the view of obtaining a lease of the buildings over the south ambulatory; but, as they limited the rector to a rent of £3 a year, nothing came of it. The patron afterwards concluded a lease with the trustees of a 'Protestant Dissenting Charity School supported by voluntary contributions' (as a notification on the door used to term it). (fn. 100) There were about 100 children, boys and girls, in this school, and the master lived on the premises. (fn. 101) The rector then came to an agreement with the trustees to close Bolton's window and the other openings into the church in consideration of £12 to be paid for opening skylights in their place. This the patron consented to and it was carried out. These schools, like the meeting-house, were continued until they were partly burnt out in the fire of 1830, the effect of which fire can still be seen in the badly flaked jambs of the western arches of the triforium, and in the transept below.

In the year after the fire, Mr. William Monney, who had then acquired the freehold of the Lady Chapel (fn. 102) and of all the south triforium, except the two western bays, (fn. 103) requested the rector and churchwardens to apply to the Bishop of London for permission to rebuild the rooms over the south and east aisles, (fn. 104) but the vestry resolved (fn. 105) ' that they could not recommend any measure which would tend to perpetuate the existence of a building on the church which they conceive ought never to have existed thereon' and they placed a tiled lean-to roof over the south triforium floor. The fire apparently either did not extend to the east end of the church or the vestry subsequently reconsidered their decision, for later in the century there existed over the east end of the aisle a brick-built chamber of recent construction, which was in the occupation of the tenant of the Lady Chapel, as was the space in the roof over the south aisle. This chamber may have been erected after the skylights were put into the school premises, either as an extension of those premises or of those of the Lady Chapel. It was removed when the apse was restored in 1886.

The lean-to roof erected by the vestry over the triforium after the fire of 1830 did not reach to the crowns of the arches. (fn. 106) In 1891 this roof was removed, the outer wall raised and a new roof built to the original height.

The north triforium, after the suppression, though secularized, did not fare quite so badly as that on the south side. In the year 1616, the two easternmost bays formed part of Sir Percival Hart's dwelling, which was in the Lady Chapel. The bay farthest east—the arcading of which had been built up in the fifteenth century—was utilized as a servants' room; the next bay, described as 'a chappell chamber opening into the church within a reasonable distance of the pulpit' apparently formed the private pew of the Hart family. Not many years after this—but at what precise date we cannot say—these two bays of the triforium formed part of the school-house, in connexion with the church schools of the parish, which occupied the remainder of the triforium gallery. The first reference we have to the school-house is a marginal note in the Rental of 1616. It is in a different handwriting from the rest and refers to certain entries with a cross; it runs, 'all these that is cros is into the skole hous and ames houses built by my Lady See'. As Lady Saye and Sele did not build her almshouses until 1631 the entry cannot have been made before that date, but there is no reason why the school-house should not have existed before. It is probable that, when Sir Percival Hart's lease or a subsequent one expired, Lord Holland granted a lease of these two eastern bays of the triforium to the parish to form a school-house. In fact, we learn from the vestry books (fn. 107) that in the year 1666 the churchwardens were paying the Earl of Holland at the rate of £10 a year for the school-house, which we assume included the two upper floors as well.

The school itself occupied the three western bays of the triforium. There is no record that any rent was paid to the Earl of Holland for these, nor of the manner in which the parish came into possession of them, neither are they referred to in the Rental; it is therefore probable that they were used for the monastic school before the suppression in 1539. That there was a school at St. Bartholomew's before the suppression there can be little—if any—doubt. John Stow, when referring to Fitzstephen's statement that in the reigns of King Stephen and of Henry II there were in London three principal churches which had famous schools (which Stow considered were St. Paul's, St. Peter's, Westminster, and St. Saviour's at Bermondsey in Southwark), says, 'Other priories, as of St. John by Smithfield, St. Bartholomew in Smithfield, St. Mary Overie in Southwark and that of the Holy Trinity by Aldgate were of later foundation . . . all which houses had their schools, though not so famous as those first-named'. (fn. 108) Had not these western bays of the triforium been in occupation for some such purpose as schools Rich would, no doubt, have incorporated them in the dwelling-house he set up in the Lady Chapel, and when required for schools his descendants would have exacted a rent for them, as was done for the eastern bays when those were wanted for a schoolmaster's dwelling-house.

Towards the latter end of the seventeenth century the churchwardens used to let the school-house on a repairing agreement for £10 to the schoolmaster, who sometimes failed to repair. This occurred in 1678, when the vestry 'ordered the churchwardens to take course by law against Mr. Henry Drake, clerk, for not repairing the school-house which he holdeth of the parish being much ruined and decayed'. (fn. 109) In 1681 it was let to Mr. Richard Sammon, schoolmaster, on a yearly tenancy, (fn. 110) the churchwardens doing the repairs, (fn. 111) which they continued to do for all subsequent tenants. In the year 1705 both the school and the school-house were let on lease to Mr. Charles Smith, (fn. 112) curate of the parish, for £14 a year. This may imply that the school was closed for lack of funds, but in the same year as a new seven years' lease was granted to Mr. Smith (1717), (fn. 113) we learn from Seymour's Survey that 'a school for boys was set up here', so we assume that the curate resuscitated the school and ran it himself.

In the year 1727 the rector and churchwardens came into the possession of John Whiting's legacy for the parish schools, (fn. 114) a small farm in the parishes of Navestock and Weald in Essex, bequeathed by him to the schools on the death of his widow. 'Then', Seymour says, 'a girls' school was also set up.' Mr. Smith dying the same year, the school and school-house were let to the treasurer and trustees of the Charity School on lease at £12 a year, the parish doing the repairs. The schools thus resuscitated have continued to flourish until the present time. In 1727 there were thirty-five boys and sixteen girls in the school; the numbers subsequently increased to about one hundred and fifty boys, girls, and infants; in 1918, owing to the air raids, the number fell to about a hundred, but when peace was restored the numbers increased to two hundred and twenty-four. In the year 1849 the rector, John Abbiss, secured a lease of land adjoining the south chapel, whereon, in 1853, he built a school for the girls and infants, and a school-house above for the mistress. He was enabled to do this by means of a bequest of £200 made by Miss Hardwick, and from other sources.

In 1865 the approach to the boys' school, which had always been at the west end of the triforium by a flight of steps, was removed in order to facilitate the restoration of the west end of the north aisle of the church, when access was given by the turret stair of the schoolhouse. The eastern bay of the triforium, which had formed the chapel chamber for Sir Percival Hart, was subsequently taken out of the school-house and converted into a class-room for the boys. The original vaulting in the church below the school-house remains, as already stated, (fn. 115) but that under the school had been destroyed and was not made good at the restoration in 1866, so that pencils and other things were frequently sent down by the boys into the church between the floor boards.

In the year 1889 new schools were erected on the site of the burialground of the canons south of the Lady Chapel (pl. XLVIII c, p. 68); the foundation-stone was laid by the Duchess of Albany, July 5th, 1888, and the boys were transferred there in May 1889. It was whilst the school was in occupation of the triforium that the arches were bricked up, as described above. (fn. 116) The room was lighted by a skylight and by windows inserted in the north wall, which was of brick as now, but these were closed and the arches opened out as soon as the school left for its new premises. The class-room was then thrown into the triforium, the easternmost bay only remaining as part of the schoolhouse, which was occupied until 1911 by Mr. John Hope, the parish clerk and verger of the church.

The turret stair of the house stands on the eastern part of the site of St. Bartholomew's chapel. Excavations have disclosed some stone foundations, which probably belonged thereto, some 6 ft. 6 in. below the present ground level.

The Clerestory of the Quire.

The present clerestory was built, as already stated, (fn. 117) during the remodelling of the east end, about the year 1405. It measures in height 16 ft. 6 in. from below the string beneath the sills of the windows to the springing of the roof. It cannot be more than 12 in. higher than the Norman clerestory, as is shown by the position of the string on the east face of the quire arch of the crossing; a string which apparently supported a flat ceiling. On each side of the quire there are four windows, each of which consists of two lights within a single arched opening. On the jambs is a slender shaft carried as a moulding over the arch of the window; it has a small Perpendicular base but no capital at the springing. Above the arch is a plain hood mould springing from corbel heads. The filling consists of two lights with tracery by Hayter Lewis and William Slater inserted in 1865; the fifteenth-century tracery had disappeared when Carter wrote in 1809. (fn. 118)

The string under the sill is 7 in. below the Norman string, a portion of which remains on the wall adjoining the quire arch of the crossing. Higher up on this portion of the wall are two other Norman strings (more visible on the south than on the north side). These two strings were probably continuations of the abacus of the capitals of the shafts on the jambs of the Norman clerestory windows. (fn. 119) If so, it shows that the Norman clerestory consisted of triple arches, as has been already suggested, and as was usual at that period. The upper string then would indicate the springing of the arch of the central opening, which would have been above the apex of the lesser arches on each side; and the lower string would indicate the springing of those lesser arches. In the rebuilding of 1405 this portion of the wall was left, we may assume, owing to its proximity to the tower.

The passage in the clerestory wall has a shouldered lintel, as at Exeter and Llandaff, dating from about 1405. In this passage, on the south side of the church only, is a stone string which consists largely of fragments of Norman arch mouldings, ornamented differently from any other in the church (pl. LXVI (1), p. 128). The same ornament not infrequently occurs in churches elsewhere, but generally in the second half of the twelfth century, as over the east cloister door at Brinkbourne Priory, and over the late twelfth-century door at New Romney. There is not enough visible to suggest that this ornament occurred in all the clerestory windows; were it so it would be strong evidence in favour of Rahere not having built a clerestory; but we may assume that Thomas used the ornament when building his west bay. The united height of the ground arcade and triforium, in comparison with the width of the quire, is, however, conclusive evidence that Rahere intended that there should be a clerestory, whether he himself built it or not.

At the west end of the clerestory passage, both on the north and south sides of the church, a stone stair ascends, leading formerly to a roof turret or the tower. The stair is now blocked, but it appears again at the east end of the clerestory of the nave.

The Roof.

There are no records regarding the original roof of the church, nor of that which succeeded it when the east end was remodelled and the clerestory was rebuilt in the fifteenth century. That intersecting vaults were never contemplated nor carried out, either in the twelfth or fifteenth century, is evident from there being no sign of any provision having been made for vaulting. We must conclude, therefore, that the quire was always covered by an open or ceiled timber roof, and it is likely, as already mentioned, (fn. 120) that in the twelfth century it was ceiled by a flat wooden ceiling supported against the wall over the east arch of the crossing by the stone string-course that still exists.

In the fifteenth century the roof was probably open and of considerably steeper pitch than now, so as to allow height for the heads of the perpendicular east windows. At some subsequent period it is possible that the rafters, having decayed at the base, were shortened, thus lowering the pitch to what it is at present; or the roof may have been lowered to facilitate the use of lead as a covering. We know by wills (fn. 121) that lead was used in the fifteenth century and that the roof was well covered with lead at the suppression, because Bishop Grindal, in 1563, wanted the lead for St. Paul's.

In the year 1809 Carter described the roof as 'the wreck perhaps of some richer Tudor open-worked timber roof pared down to a common pediment covering, and cross timbers supported by cherubim heads'. (These cherubim heads suggest Sir Christopher Wren's time: they were simply wooden ornaments, not carved on the timber of the roof.) The roof has been renewed three times since the suppression, but from the vestry books it appears that it has required considerable repair about every forty years. The first record of the repairs occurs in the Churchwardens' Accounts of the years 1574 to 1578, when, in addition to the lead, timber, stone, iron and other materials, £6 13s. 4d. was paid for 'the new making of the roof of the church' and of the south side aisle. In 1737 a carpenter's bill of £150 was paid when the roof was releaded and the 'new and old timber and boards belonging to the new ruff' were whitened and coloured. The dust evidently caused these repairs to be thirsty work, because the last item on the account reads:

'Spent at the Half Moon Tavern and the Flying Horse &c. concerning these repairs £9/10/6d.'

In 1885 Sir Aston Webb reported that re-roofing the church was absolutely necessary if it was to be preserved, and that the roof was without architectural character. On his recommendation a new roof was made of similar pitch to the old one, but the ancient tie beams were retained. (fn. 122) The space between the king posts and the struts was filled with open work in place of the plain boarding of the old roof; (fn. 123) and the braces below the tie beams were enriched with cuspings; but the intention to place bosses where the purlins intersect the rafters was not carried out.

The roof of the crossing was similar in all respects to that of the quire until 1886, when a new one of oak was provided and the present flat panelled ceiling was added.