which, however, cannot be identified with certainty. The puddled and rammed chalk here was exceedingly hard and more difficult to remove than any of the rock-chalk near by. In it was found a fragment of red Samian pottery (No. 194). The gangway, or passage, between this trench and the inner trench was about 2ft. wide, and was fairly even in all but the S.E. part, where, for a length of 6ft., the chalk was rugged, and left at a level equal to, or even higher, than the arenafloor.* The smooth part of the gangway was 0.7ft. lower than the nearest part of the arena-floor. Over it several fragments of Romano-British pottery (No. 185) were found, and at a depth of 4.2ft. below the surface an iron cleat (No. 181) was uncovered (Fig. 3). Its use as portion of the iron furniture of sandals, or shoe leather, is proved by the discovery of specimens with a quantity of hob-nails at the feet of skeletons at Rotherley and Bokerly Dyke. A specimen was also found on the old surface line under the rampart of the Wansdyke in N. Wilts, and others have been found at Silchester, &c. The inner trench which was previously traced for some distance in Cutting XII., 1909, was very clearly defined in this cutting, extending from the large smooth-sided circular pit containing Post-hole No. ix. to the extreme E.S.E. corner of the digging. It varied in width at top from 1.7ft. to 2:2ft.; at bottom from 0.7ft. to 1.5ft. It was about 1.2ft. deep below the gangway and from 1.8ft. to 2:4ft. below the arena-floor. This part of the inner trench differed from other parts (Cuttings XII. and XX.) in having no ledge, or step, on its outer side, but the filling consisted of chalk rubble along the outer half and mixed mould and fine chalk along the inner side. In two places post-holes (Nos. xli. and xlii.) were distinctly traced in the trench, the bottom of the holes. *The reason for this cannot be determined unless the cutting is extended further southwards. "Excavations in Cranborne Chase " II., 190; III., 102, 106, 129, 270, &c. Report, 1909, pp. 9-10. Proc. Dor. Field Club, XXX., 221–2. slightly penetrating the rock-chalk bed.* Very few objects were found in this trench; they consisted of a few scraps of iron, including some nails, a piece of New Forest ware, small fragments of Romano-British pottery, and part of the base of a red Samian bowl (No. 189), bearing the maker's mark OF. SEVER I. Severus was a potter of La Graufesenque about A.D. 50-60. An entirely new structural feature was met with here, which has given rise to some discussion. I refer to the rough, shallow, flat-bottomed gutter connecting one oval and three oblong pits, the centres of which were 5.5ft. apart. This row of holes, which was on the curve, was on the arena side of the inner trench, to which it did not run quite parallel. The nearer margins of the two trenches on the east were separated to the extent of 1·75ft., but they met on the west at a distance of 16 7ft. E. of the long axis line; and the westerly oblong pit was recessed into the inner wall of the inner trench. All the pits were about 19ft. deep below the arena-floor; the oval pit was 21ft. by 18ft. at the top and the oblong pits averaged 2.25ft. by 1.25ft. Nothing was found in any of these holes, and no traces of the former existence of posts, as would be expected, had posts ever been permanently fixed in this position. The holes may continue in a S.E. direction, but the point could not be determined this season. It is uncertain, therefore, whether this line of holes was constructed at the same time as the inner trench, or whether it existed previously; in support of the latter view it is seen that the inner trench unites with the northern side of the most westerly of the oblong pits.† * In one or two places in the trench traces of a buff-coloured substance were observed, which Mr. Jukes-Browne thinks very possibly may be burnt chalk. Over the inner trench in the E.S.E. corner of the digging traces of charcoal were noticed. † It has been suggested that these pits may have contained the bases of posts or columns for the support of a velarium in this part of the amphitheatre. IV.-CUTTING XII. EXTENSION. In the first place, this cutting was pegged out on the W. side of the long axis to correspond with Cutting II. Extension, to the W. end of which it joined, and the excavation was begun for the purpose of tracing the inner trench and examining the border of the arena-floor. (See Plate I.) FIG. 4.-Relics found at Maumbury Rings, 1910. (From Drawings by Mr. Ernest Sprankling.) The floor (average depth 5.5ft. below the surface) was found to be well preserved and fairly even up to a point 20ft. along the S. margin from its junction with Cutting II. Extension; but transversely the arena-floor in this position was found to fall from W. to E. to the extent of 0.6ft. in a length of 20ft. The inner trench running obliquely across the western end of the cutting had the usual ledge, or step, along its outer margin, which supported rammed chalk, leaving a width of 1ft. for the trench proper. 66 On or near the level of the arena-floor a third brass coin of Tetricus I., A.D. 267-273 (No. 171), and a British imitation coin, IV. Century (No. 172), were found; also a few fragments of pottery of the Roman period, including a piece of red Samian (No. 173). At a depth of 4.8ft., a globular bead (No. 175), formed from a fossil hydrozoon, Phosphæra globularis, was discovered (Fig. 4). On the floor and just over it ten iron nails were collected. The usual "shingle was also observed; it contained a large proportion of small water-worn fragments of quartz. At 20ft. to the W.N.W. of the E. end of the cutting the Roman floor was no longer formed of solid chalk; consequently we decided to extend the digging in a W.S.W. direction, the whole cutting in a straight line attaining a maximum length of 73ft., bringing the work practically up to the margin of Cutting X., 1908 (through the prehistoric shaft No. I., the terrace, and up the great embankment to its crest). In this large area a number of interesting features became involved (see Plate III.), and we saw nothing more of a solid chalk arena, as the N. margin of an immense shaft (No. IV.) extended for a length of 49ft. to the W.S.W. end of the cutting, where, however, it did not terminate. The solid arena-wall also proved of interest, as it was more irregular here than elsewhere, and gradually diminished in size towards the W.S.W. In this excavation, too, the inner trench cut into solid chalk gradually disappeared as we proceeded, but its line was most clearly defined in the rammed chalk deposit over the mouth of the great shaft. The solid chalk gangway, or intermural space, between the arena-wall and the inner trench, also diminished in size as our digging progressed, on account of the position of the shaft. Bearing in mind the enormous amount of labour that would be entailed in clearing out the shaft and the limited time at our disposal, it was decided not to attempt its complete reexcavation, but we followed its N. margin down to a depth of 9'6ft. below the surface. The sides exposed were far from steep, shelving away at an angle of only 27° (Plate III). A large number of flint flakes, some very large, of Neolithic type, were found below the Roman level, but no antler tools were discovered. We were rather reluctant to leave the mouth of this great shaft, but its re-excavation was not the primary purpose of our researches at Maumbury. In tracing the inner trench westwards it was found that its inner edge and the margin of the shaft united, and still further west the trench became more feebly represented by rockchalk (Plate III.). Ultimately the deepest part of the trench no longer had a solid bottom; and as far as solid chalk was concerned, it ceased to exist; but the Romans, finding this soft place made the best of the difficulty, as we shall show later on. The trench in this part contained seven iron nails and two pieces of Romano-British pottery. Returning to the N.E. end of the shaft, an interesting structural feature was noticed close to its mouth and its sloping side, viz., a smooth-sided hole, semi-circular in outline, and having an even and flat bottom (Plate III.). Its N. margin was within 0.25ft. of the outer edge of the inner trench, the cutting of the latter by the Romans not having disturbed it, It contained a very fine deposit of marly chalk and about sixty well-struck flint flakes. It was quite evident (1) that the hole was once circular (diam. 2.2ft.), with sides almost vertical, and (2) that it had been sliced through obliquely at the time the shaft was sunk. No doubt it belongs, like the shaft, to the Neolithic period, but in point of years (or months) it was certainly anterior to the shaft itself. Nothing earlier in date than this circular flat-bottomed hole has been revealed at Maumbury. |