wide. The arch is mostly formed of converging blocks of Kentish rag, generally about 1 in. apart, though somewhat closer at the crown. The span at the springing is an inch or two wider than the span of the jambs. The imposts are formed of two Roman tiles, the upper one overhanging the lower, and the lower overhanging the jamb. The doorway is lined throughout with plaster, on which at its first openingout were seen what looked like rough mathematical figures. The jambs internally are of Roman tiles, with occasional pieces of Kentish rag. Externally they are almost entirely of Roman tiles, though under the west impost, 3 ft. 10 in. above the sill, there has been inserted a fragment of freestone about 2 in. high, brought from elsewhere. On it are parts of an inscription, which has been supposed to date from the ninth or tenth century. The letters HONORE.. STÆ.. ET OMNIV SCORV are still decipherable; and the whole may perhaps be read as, "To the honour of Saint (Mary ?) and all Saints." This may have been the dedication-stone of a Church, or it may not impossibly have been the dedication-stone of an Altar, as an order was issued in the ninth century by a Saxon Archbishop that a stone should be placed at the corner of each Altar specifying the name of the Saint or Saints to whom it was dedicated. A parallel to this has been found in the discovery of a stone from the Saxon Church of Deerhurst, the fragmentary inscription of which has been conjecturally read as, "In honore Sanctæ Trinitatis hoc altare dedicatum est."

This round-headed doorway has been hitherto supposed to be of the same date as the wall, but closer investigation has clearly proved that it is a later insertion, probably made in the Saxon period.

While in the surrounding wall there are (as I have before stated) only four Roman tiles to the foot, there are in the jambs of the doorway six tiles to the foot; and at the time of the insertion nearly 1 ft. of the surrounding wall was broken away, as will be noticed by any observer outside the Church.

The early brick wall extends eastward for 6 ft. 9 in. beyond the round-headed doorway till we reach a break in it, which was clearly the termination of the original Chancel. For the last 2 ft. the work is somewhat irregular, and from this circumstance (and from some evidence that has been discovered at this spot on the outside) a conjecture has been hazarded that here we have the beginning of a Roman apse. Eastwards of this break the walling is of different workmanship, shewing with the mortar-joints six tiles to a foot; and after 3 ft. 5 in. we come to a Sedile, which was discovered a short time ago blocked up with medieval brickwork (see Illustration). It had apparently a pointed arch of which about 5 ins. have been cut away. The springing line is about 2 ft. 9 in. above the seat; the radii are about 3 ft. 9 in., their centres being on the springing line. This would fix its measurements as follows-span 5 ft., depth about 1 ft. 3 in., height from seat to springing line 2 ft. 9 in., and from seat to apex about 6 ft. 4 in. A difficulty has arisen as to the date of the Sedile from the fact that the top of it has been cut away by the insertion of a lancet window, appearing at first sight to belong to the Early English period, so that the Sedile would seem as if it must be of an earlier date than the window. But Mr. Livett, though believing it not impossible that the Sedile and lancet window were built at the same time, and the sill of the window

altered afterwards, thinks it more probable that the Sedile and the brickwork in which it is placed were built late in the twelfth century, and the lancet window inserted subsequently, perhaps in the fourteenth century. The position of the Sedile would seem to point out that the Altar stood, in Early English times, immediately east of the step whereon the present Altar-rails are placed.

66

Little or nothing fresh has been discovered on the N. side of the Chancel. The so-called Queen Bertha's tomb," which is now surmounted by a pseudoNorman arch, is probably the tomb of the Restorer of the Church at the end of the twelfth century, and is coeval with the later brick wall. Below ground, in the North-West angle of the Chancel, were found two or three projecting Roman tiles, apparently the beginning of a cross wall which was destroyed when the present Chancel Arch was erected. Some slight signs of the cross wall have also been detected above the stalls in this angle.

It now only remains to mention the discoveries that have been made outside the S. Wall of the Chancel. Near the square-headed doorway described above there have been found underground the remains of two walls, running at right angles to the Chancel, and forming two sides of an Adjunct or side-chapel, the southern side of which has been destroyed in the process of digging graves. These walls are 4 ft. 9 in. apart, and are each of them 26 ins. wide, built entirely of Roman tiles. The Western Wall runs 8 ins. beneath the Eastern angle-wall of the Nave. Between the walls there is still existing part of a flooring of opus signinum. There can be no doubt that this adjunct is of the same workmanship, and the same