Page images

interiorally with nine courses of tiles to a height of 18 in., thus forming a ledge for the tiles to rest upon that were laid across from the pila. Adhering to the ledge was a portion of the floor of the room, 6 in. thick, composed of a bed of rough concrete 6 in. thick, upon which was an additional layer of red cement 4 in. thick, giving, with the tiled floor on which it originally laid, a total thickness of 12 in. In the walls above were the smoke flues, four in both the north and south walls, and one in the west wall, the orifices of the flue tiles being 3 in. by 5 in. In the north wall of the hypocaust was an opening 10 in. wide and 1 ft. high, neatly constructed with tiles laid in courses (Plate F). This was for the admission of hot air from the furnace to the adjoining chamber.

18. Room, 19 ft. 3 in. by 16 ft. 3 in., the entrance in the south wall being 4 ft. 3 in. wide. The floor of the apartment was covered with white concrete composed of lime and small pebbles, with a halfround moulding of red cement round the edge. The walls were adorned with fresco painting. On referring to Plate F the method of heating this chamber from the hypocaust of No. 17 will be seen. On cutting out a portion of the concrete floor to a depth of 4 in. a layer of large tiles was disclosed reaching across the room in the form of the letter Y. A few of these were removed revealing a channel 9 in. wide, walled on each side with flints firmly set in mortar to a height of 8 in., the bottom of the channel being the natural soil. Upon the top of the little flint walls, tiles 8 in. square were laid overlapping inwardly 2 in.; over these came the large tiles, their undersides being blackened with smoke, then the layer of concrete. In the north wall of this chamber were two imbrex roof tiles serving as smoke flues, which communicated below the floor with the two arms of the Y channel. From this room we obtained a Roman spear-head 10 in. long, and a turning tool 13 in. long with a gouge-shaped tip, both of iron. The antiquity of the latter is doubtful.

19. Room, 9 ft. 3 in. by 9 ft. 6 in., with an entrance into No. 18 4 ft. 3 in. wide. The west wall was much broken away, and it is therefore almost impossible to say how it related to No. 22, but there was probably an entrance into it from that room. The floor of the chamber was of white concrete with a tile embedded here and there.

20, 21. Appeared to have formed the head of the inner corridor (No. 25). If No. 20 was an apartment the short wall seen in the Plan, opposite an external buttress in the south wall, must have continued to the south-west angle of No. 19, with perhaps a doorway in

[graphic][merged small][merged small]

it. In the north-east corner of No. 20 a small portion of red tesselated pavement remained.

We have now to describe a series of summer apartments with a corridor back and front. The walls of this block were so low that none of the entrances to the rooms were visible.

22. Room, 22 ft. 6 in. by 5 ft. 9 in., paved with white concrete. Portion of a coping-stone of sandstone found here.

23. Room, 22 ft. 6 in. by 10 ft. 8 in., paved with white concrete. Other fragments of coping-stone found.

24. Room, 22 ft. 6 in. by 10 ft. 4 in., paved with white concrete. 26. Room, 22 ft. 6 in. by 19 ft. 8 in., paved with red tesseræ. 28. Room or passage, 22 ft. 6 in. by 7 ft. 3 in.

29. Room, 22 ft. 6 in. by 21 ft. 11 in., paved with red tesseræ. The floor of No. 24 was 6 in. below the level of No. 23, and Nos. 26, 28, and 29, 1 foot lower than No. 24.


These six rooms had round the margin of the floors the same kind of moulded skirting of red cement that was present in some of the rooms in Section A, already described. Upon the floors of these six apartments we found numerous fragments of plaster decorated with fresco painting, the colours being as bright as when first laid All the pieces having patterns upon them were saved, and these Mrs. Payne faithfully copied in water-colours before the tints faded. Out of the fifty-five examples no two are alike. These will be described hereafter. Perhaps the most interesting feature of these rooms is that they were divided by plaster partitions, the bases of which remained set upon a foundation wall of flint level with the floor line. The plaster walls of each partition were 3 in. thick and 9 in. apart, the intervening space having been apparently filled in with timber, as the impress of it could be distinctly seen in the mortar upon which the first slab was laid. On the inner sides of the plaster walls there was a coating of mortar a quarter of an inch thick, which could be easily detached with a knife, shewing that as the timbers were placed in position mortar was run in down the sides in order to fill up the crevice between the plaster and wood. This careful method of construction is a striking contrast to the flimsy lath and plaster partitions of the present day, which keep out neither cold nor sound.

The corridors on either side of these rooms will now be dealt with. 25. A corridor on the south side, 68 ft. 2 in. by 9 ft. 9 in. A portion of its pavement of red tesseræ remained at the western end.

27. The corridor on the north side. This was evidently paved with tiles, as a large portion of the floor at the western end was in

its original state. Upon the tiles a quantity of bones of oxen were found. At the opposite end a mass of blackened earth was met with which yielded oyster shells, bronze tweezers, fragments of iron, bone pins, bronze bangles, a glass disc, a bronze handle terminating in the head of a lioness with eyes of some red glistening substance, a flat piece of bronze in the form of a bird's head, an iron hoe, and one of those curious objects that have been erroneously termed "hippo-sandals," of which more will be said later.

30. A room, 12 ft. 3 in. by 9 ft. 9 in. One half projects into corridor No. 25, the other half into the great corridor No. 39. The walls of this room had not been plastered, and there was no trace of a floor. It was excavated to the foundation of the walls and down to the natural soil in search of it. On the southern side of the chamber we met with quite a barrow load of red tesseræ piled against the wall. The cubes had certainly never been set in a pavement, and a workman may possibly have sat in this room and fashioned them.

31. A corridor, 43 ft. 2 in. by 8 ft. 5 in., paved with white concrete. There was a break in the east wall suggesting the site of a doorstep up into No. 29.


The area occupied by rooms Nos. 32, 33, and 35 underwent an entire alteration in Roman times. The whole was originally heated, but subsequently converted into cold rooms, the hypocausts under each having been filled in with earth, broken tile, flints and mortar rubbish to a depth of 3 ft. 8 in. Upon the top of this débris a new floor of white coarse concrete had been laid nearly level with the adjoining corridor (No. 31). The writer suspecting the presence of older work below decided to clear out the area to its base, which resulted in the exposure of the extensive hypocaust seen on Plate G. 32. The stokehole, 11 ft. 7 in. by 8 ft. 5 in. To the east of it is a hypocaust 12 ft. by 7 ft. 5 in., roughly built with low walls of tiles.

33. The great hypocaust. At the bottom of Plate G may be seen the passage, 2 ft. 1 in. wide, leading to it from the stokehole, the south end of which shews the springing tiles of the arch in position. The block of masonry on either side of the passage is simply the filling in with broken tiles set in clay of two recesses that were probably used as receptacles for charcoal for the supply of the furnace. To facilitate description the hypocaust may be divided into three compartments.

« PreviousContinue »