The Romano-British Settlement at Springhead. Excavation of the Bakery, Site A

THE ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD. EXCAVATION OF THE BAKERY, SITE A By W. S. PENN, B.SC. INTRODUCTION THE settlement is situated on the north boundary of Southfleet, Kent, where the raUway to Gravesend crosses the modern Watling Street (A.2), the area being referred to locaUy as Springhead (Kent 6-in. sheet X S.W., National Grid Reference No. 617,725, Fig. 1). The Ordnance Map referred to identifies the site with Roman Vagniacae, although the only evidence for this is a reference to Iter. I I in the Itinerary of Antonine to the fact that Vagniacae is IX Roman mUes from Durobrivae (Rochester) and that substantial remains have been found in the vicinity. The site was re-discovered by members of the Gravesend Historical Society and has been excavated by them since 1951. Prehminary work indicated the importance and extensive nature of the site so that it has now been scheduled as an Ancient Monument and is being excavated on behalf of the Ministry of Works by the Gravesend Historical Society. In this work we are most fortunate in having the valuable advice of Mr. Sheppard Frere, M.A., F.S.A., who has most kindly agreed to act as consultant to the Society. The present report is concerned with Site A which was excavated in 1951-2 when the bakery was discovered. The other sites wUl be reported in due course and, therefore, until much more data have been presented it has been thought wise to defer a general discussion on the topography of the site, its identity and other features. This report is thus mainly devoted to a factual record of the excavation of Site A, although a brief general account has aheady been given in Archceologia Cantiana (1). PREVIOUS WORK The site has been the subject of much digging since about 1775 but, although many finds have been reported, they can hardly be said to have been recorded. However, the earliest important references are in Hasted (2), although the most important document is Memoranda of Springhead, by A. J. Dunkin, 1848. Philip (3) describes the Northfleet VUla and the Victoria County History (4) contains a useful summary 53 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD of the findings. Many volumes of Archceologia Cantiana contain references to Springhead. The most important finds were an aUeged mUestone (c. 1750) (5), various tombs (1799 onwards) (6, 7), baths (1814) (8), flint footings (1844) (9), a tUed platform (1845) (10), part of a building (1864) (11), a tUed arch (1921) (12), the Roman road (1921) (12), and a smother 880 760 SCALE OF FEET // //// •; Vi // Vv ll II Nursery JI C3 '/ «£ WINGFIELD BANK « 'O/N E TREE & M 300 .ps SCALE OF FEET K> ONE TREE \ FIELD SOLE 8 FIELD Fig. 1 [Crown Copyright reserved kiln (1921) (13). The most interesting of these finds was the walled cemetery (6) consisting of a buttressed enclosure wall, 58 ft. by 55| ft. and containing a stone tomb with two lead coffins. These contained the remains of two young children, one with a gold chain set with pearls, two gold bracelets and a pair of leather sandals (aU in the British Museum). The main importance of aU these isolated finds is that they do indicate that a substantial settlement existed in the area. 54 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD SUMMARY OF WORK, 1951-2 The excavations on Site A (Fig. 2) have indicated four weU-defined superimposed phases of occupation. Phase I extended from about A.D. 43-90 and is represented by a pit and the chalk floor of a wooden buUding ; Phase I I extended from A.D. 90 to about A.D. 120 and is represented by a well-defined walled structure tentatively identified as a bakery ; Phase I I I extended from about A.D. 130 until the early %!>>&_& iilll fM EBBLE/CHALK ROAD PIT fS PESTROTtD LLLUU 1 StCTIOH OF BAKERT m V \$esSsa S 555 ism8 i <=_ wmrnmi m 9m n OC11S111t1 f1 c1 S1 1 P a 00 WW DOD •i> • 111 _ 1 0 DECOJSATED SAMWN-HApRIANIC FOR KEY - SEE KEY DEPOSITS. Fia. 7. Over the top is the extensive rubble layer of the derehct buUding level. There were many animal bones under this, and in the layer were tUe fragments, naUs, much pottery, oyster sheUs and the like. Some of this rubble had also found its way into the top soU. Figure 8. Fourth-Century Section This section is in complete contrast to aU of the others. Instead of the usual remains of early and late first-century occupation, the soil was completely stripped to base clay in Antonine times. This includes a partial destruction of the waU. In the base clay was a pit and a post hole (containing Antonine pottery) but there was nothing to indicate the purpose of either. There was a lining of flints, however, in the pit which sealed an Antonine mortarium sherd. 66 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD Over the Antonine occupation layer was a late occupation layer, containing third- and fourth-century coins and pottery. This was ' clearly the edge of a third-fourth century occupation area, but it may be noted that in complete contrast to the rest of the site, there were very few small finds in the Antonine or later strata. CONCLUSIONS Even from the slender evidence so far avahable, the pattern of occupation on Site A appears to be quite clear. There was a settlement at a very early period, possibly as early as A.D. 43, which is quite consistent with the proximity of the invasion port. There are hints of Belgic influence in pottery and fibulae but so far there is no indication of a pre-Roman occupation, but this possibility cannot be ruled out. The early settlement was of wood, but towards the end of the first century a flint building, the bakery, was constructed. This was possibly a result of the buUding wave inspired by Agricola and in this the settlement may foUow'the general pattern. However, there is an important difference in the next phase compared with other settlements, in that about A.D. 130 this first stone buUding was destroyed and replaced with a wooden structure. In most places it appears that wooden buUdings were replaced by stone at this time. The explanation can only be decided by future excavation, and the main Antonine settlement could no doubt have had its centre elsewhere. There is no doubt, however, from the large number of smaU finds, that there was an extensive community in the Antonine period which survived in some force until the first quarter of the third century—at least. There is only a hint of the final phase, the main centre of which is somewhere south of Site A. That it existed, however, is certain, and there is much fragmentary pottery to indicate a late third/fourthcentury occupation. It is interesting to note, however, that the Antonine Site A was deserted and not buUt over in the fourth century. Does this indicate a contraction in the hving area in the fourth century as at Verulamium ? Only future excavation can answer this question. Site A has only provided a hint of what is to be found at Springhead, and excavation wUl proceed to find answers to the many obvious problems, to provide a picture of a provincial Roman settlement and post-station and to identify the settlement with Vagniacae of Iter. II. THE FINDS The smaU finds on Site A were extremely numerous and interestmg, and detaUed accounts are given below. A few have been singled out for special attention, and they are aU preserved at the Museum of the Gravesend Historical Society, Milton Chantry, Gravesend and can be viewed by arrangement. 67 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD THE SEEDS One of the most important finds at Springhead was the carbonized seeds, found in Key Deposit VI near Fig. 8. These were identified as • Atriplex patula or the Common Orache, by the British Museum (Natural History). Hundreds of seeds were found in a cluster, and it was clear that this number must have been deliberately coUected for use in some way. "FOURTH CEHTUKY SECTION" SOUTH NORTH SCAUi OF MET TT £D II © \ H» eO. 1 \ OQ 1i i i \ _i i ©C0IN-VIC70BINUS ©COIN- VALENS (F)MORTARIUM' HADRIAN L FOR K E Y - SEE KEY DEPOSITS. FIG. 8. The problem is clearly to estabhsh what this use was, since today the plant is a weed. ActuaUy there are hundreds of species of Atriplex, and a consideration of some of these affords several clues. The Red Orache (A. hortensis) (16) is a plant cultivated for its red fohage, which is due to the development of anthrocyanin pigment. Species of Atriplex (Salt Bush) (16) appear to form useful fodder plants in dry countries. These plants contain up to 30-40 per cent salt in their ash. In the genus Atriplex (17) there is the Mountain Spinach or Garden Orache (A. hortensis L.), which was introduced from Tartary about four hundred years ago, has since run wUd but is still occasionaUy grown in gardens. It seems not unreasonable from the above facts to assume that Atriplex p. was once cultivated by the Romans in this country, either for fodder or even human consumption, and that the plant has since run wUd. This is confirmed by the fact that seeds of Atriplex sp. were 68 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD ^ S3 iwawwairas FIG. 9 16 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD found at the Roman MUitary Station, Newstead (18). From this it would appear that the species Atriplex was of importance in ancient times, which may be a new and important contribution to agricultural history. INDUSTRIAL REMAINS There is definite evidence of much industrial activity at Springhead, particularly in working metals : iron, lead, bronze and brass. As far as bronze is concerned, Mr. HuU writes regarding one brooch (No. 1) : " I conclude this is an unfinished casting and indicates that brooches were made on the site." This is the main evidence for the working of bronze, but a few bronze droppings have also been found. As far as brass (orichalcum) is concerned, only one fragment has been found. However, this had clearly fallen from the spout of a crucible, since one side was very rough (in contact with the rough crucible) and the other curved and smooth (clearly due to a pouring action. The composition of the brass is as foUows : Copper .. 45-01% Zinc .. .. 23-84% Tin .. .. 19-85% Lead .. .. 11-30% This is of great interest and Ulustrates the difficulty of drawing a line between the composition of brass, bronze and pewter. It is, however, yellow like brass and was quite free of verdigris, i.e. it appears to be corrosion-resistant. In many places over Site A there was much iron slag and ore, indicating the working of iron. However, some of it was found over roads, which indicates its use for road metalling, although there was far from sufficient to make a continuous layer. It seems hardly likely, however, that high quahty ore would be used in road-making, and the finding of parts of a furnace base, and waUs with fused slag and flux adhering, definitely indicate iron smelting. The ore had an average iron oxide content of 72 -4 per cent and the slag 22-7 per cent. As far as lead is concerned, two rivets were found as weU as several indeterminate fragments. One of these had the foUowing composition : Iron Silver .. Tin .. Copper Bismuth Antimony Cadmium Arsenic present . 0-002% . 0-08% . 0-004% trace • 0-01% absent absent 70 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD o-o • S r * \ \ \ 7 \ \ \ 10 II 12 13 14 15 18 19 16 1 J 17 20 21 Fia. 10. 71 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD This analysis indicates quite a high state of purity and a very advanced extraction technique. In view of the considerable importance of the study of these leads, two of the finds on Sites B and C have been anticipated. Two lead fragments were found and a detaUed chemical study of these has been made. It is hoped that this wUl throw light on Roman metaUurgical techniques and the origin of the lead. The first object found (in a late second-century context) appears at first sight-to be an axe mould. It was rather thick, made of lead with the outer surface covered with a ground stone or sand, and the inside surface coated with iron oxide. The inside was shaped very much like an axe (a film of iron oxide was found aU over the inner surface) as if an axe had been inside. Now it is clear that an iron axe could not be cast in a lead mould, so that the problem exists of trying to estabhsh why the shape of an axe-like object was found inside a mass of lead. The following theories suggested themselves : (1) The iron had been tempered inside the lead sheath ; (2) The iron axe had been covered with a lead sheath to function as a blunt instrument; and (3) The iron had been covered with lead to act as a protective cover. Alternative (1) was untenable, since cast iron cannot be tempered. Alternatives (2) and (3), however, had certain merits, and possibly the investigation would have finished there had it not been that additional facts were discovered. A close examination of the object revealed certain flow lines which indicated with little doubt that the lead had been poured around the hon object. Then a detaUed chemical analysis was made (spectroscopically) with the results given below : Copper Tin .. SUver .. Zinc .. Iron Antimony Cadmium Bismuth Nickel .. . 0-047% . 0-66% . 0-0067% . 0-0016% . 0-00094% . 0-0048% . 0-012% . <0-0005% Not detected The outstanding feature of this lead, compared with the one given above, is the very high tin content. Could it have been that this tin had been added to give some special characteristic to the lead, such as to make it harder ? It is almost certain that the tin was deliberately added and that the lead was not a natural alloy. An extraordinary stroke of fortune decided the matter. A large 72 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD ? P Si i$ww FIG. 11. 73 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD stone was found (Site C) containing a hole in which was an iron rod, cemented in with lead (second century). The lead had been poured around the iron, into the hole in the stone and made a very secure joint indeed. Could the original lead have served this purpose since (a) it had been poured round an iron object and (b) it had clearly been in contact with stone on the outside and the particles of stone could only have been picked up when the lead was molten ? For this argument to be true it seems reasonable to assume that the lead would have to be the special aUoy with the high tin content. The lead was accordingly analysed, with the following result : Copper Tin SUver .. Zinc Iron Antimony Cadmium Bismuth Nickel .. . 0-031% . 0-63% . 0-0103% . 0-00077% . 0-0014% . 0-0092% . 0-022% . 0-0005% Not detected It is clear that this lead is also the special aUoy and it therefore seems reasonable to assume that the first lead object found was merely the cement to the iron foot of something else. Although confirmation of the use of a lead cement wUl obviously be required, this discovery is clearly of first-class importance to the history of metaUurgy. I t is proposed to study lead ingots from a variety of sources, and leads from various sites to estabhsh the trade routes of the different leads and, if possible, a new method of dating. This may be possible from variations in trace elements which may have changed as supphes worked out and new mines came into operation, etc. The field is wide but may weU repay study. There is much evidence that bone was worked on the site. Some bone pins had clearly been turned whUst others had been worked by hand, and a smaU coUection of partly worked bones was found. It is also obvious that the playing counters were turned on a lathe which was probably worked by a treadle. One of the counters found had the lathe marks clearly visible as weU as the point for securing the object on the revolving table. THE COINS Many coins have been found at Springhead and district over the past 150 years, most of which have been lost and not recorded, although many were only surface finds. These finds included a hoard (19) of 114 coins, the latest Emperor being Tetricus II. 74 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD ; - ^ s FIG. 12. 75 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD The earhest and latest coins found during the present excavations were of Claudius and Valentinian II respectively. DetaUs of the coins are given in Table II. TABLE II THE COINS (The coins were identified on site by Mr. E. W. Tilley and finally by R. A. G. Carson, Esq.) Key Deposit Position of Find and Description vni/vn X II IX X V II VI vr VI IV VII VI IV IV VI/III } Robber trench, Figure 6. As. Cohen 84 ; M. & S. p. 129, 66. Claudius I. In pit, 1st century. Figure 5. 2 AE Dupondius. Cohen I ; M. & S. p. 129, 67. Claudius I. Over pebble/chalk road. 2 AE (As). Cohen 6 ; M. & S. p. 132, 82. Antonia. In Bakery clay seal, near Figure 5. Denarius. Obv. illegible. Head laureate right. Vespasian. Rev. illegible. Seated figure left. In pit, 1st century. Figure 5. 2 AE (As). Cohen 85 ; M. & S. II, p. 106, 775. Titus. Outside walls N.W. corner of Bakery, Figure 4. 2 AE (As). Cohen 122 ; M. & S. I I , p. 196, 333. Domitian. Over pebble/chalk road. Dupondius. Obv. Head to right. lst/2nd century. Rev. Standing figure left. In field S.C. Above pebble/chalk road, Figure 6. Denarius. Cohen 272 ; M. & S. II, p. 268, 353. Trajan. Over Antonine flint floor, Figure 5. Denarius. Cohen 99 ; M. & S. II, p. 374, 297. Hadrian. At side of pebble/chalk road. Dupondius. Cohen 724 j M..& S. I I , p. 381, 360. Hadrian. Over pebble/chalk road. Dupondius. Cohen 724 ; M. & S. II, p. 381, 360. Hadrian. In chalk/flint road over 1st century pit, Figure 5. Denarius. B.M.C. I l l , p. 23, 139. Faustina I. Over Antonine flint floor near Figure 5. Dupondius B.M.C. I l l , p. 330, No. 1969. M. & S. I l l , p. 142, 932. Antoninus Pius. Just under plaster layer of Antonine building, near Figure 5. As. B.M.C. AP. 1983. M. & S. I l l , p. 142, 932. Marcus Aurelius. Derelict building layer, near Figure 7. 2 AE As. M. & S. I l l , p. 194, No. 1405a. Faustina II. At side of pebble/chalk road, probably after Antonine occupation. Denarius. Cohen 468. M. & S. I, p. 117, 196. Septimius Severus. 76 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD TABLE II—continued Coin No. 18 19 20 21 Key Deposit } IV III III III III III III III III III III VI/III VI/III III III III VI/III III III III Position of Find and Description Just under plaster layer of Antonine building, near Figure 5 Denarius. M. & S. IV, Part II, p. 71, 7. Severus Alexander. Near destroyed south-east wall of Bakery. Antoninianus. Cohen 199. M. & S. V, Part I I . D 362 315. Postumus. P ' Near destroyed south-east wall of Bakery. Antoninianus. Cohen 199. M. & S. V, Part II, p. 315. Postumus. 362, Near destroyed south-east wall of Bakery. 3 AE. Obv. illegible. Bust radiate right. Rev. Standing figure left. Postumus. Over destroyed south wall of Bakery, Figure 8. Antoninianus. Cohen 101. M. & S. V, Part II, p. 392, No. 61. Victorinus. Over destroyed south wall of Bakery, near Figure 8. Antoninianus. Cohen 87. M. & S. V, Part I, p. 214, 36. Claudius H. Gothicus. Over destroyed south wall of Bakery, near Figure 8. 3 AE. Antoninianus. Cohen 37. M. & S. V, Part II, p. 407, 70. Tetricus I. Over destroyed south wall of Bakery, near Figure 8. 3 AE. M. & S. V, Part II, p. 407, 56. Tetricus I. Over destroyed south wall of Bakery, near Figure 8. 3 AE. Obv. Ricus PF AVG. Bust radiate right. Late 3rd century. Over destroyed south wall of Bakery, near Figure 8. 3 AE. M. & S. V, Part II, p. 424, 270. Tetricus II. Over destroyed south wall of Bakery, near Figure 8. Barbarous imitation of radiate coin. Late 3rd/4th century. At side of pebble/chalk road. Cohen 4. Helena. At side of pebble/chalk road. 3 AE. Cohen 254. Constantine I. Over destroyed south-east corner of Bakery. 3 AE. Cohen 508. Constantine I. Over destroyed south-east corner of Bakery. 3 AE. Cohen 690. Constantine I. Over destroyed south-east corner of Bakery. 3 AE. Cohen 618. Constantine I. At side of pebble/chalk road. Cohen 114. Constantine II. Over destroyed south wall of Bakery, near Figure 8. 3 AE small. Cohen VII, p. 468, 188. Constantius I I. Over destroyed south wall of Bakery, Figure 8. 3 AE. Cohen 47. Valens. Over destroyed south wall of Bakery, Figure 8. Obv. DN VAL ENTINIANUS PF AVG. Valentinian II Bust diademed and draped right. Rev. VICTORIA AVG. GG. Victory advanoing left with wreath and palm. In ex. LVG. 77 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD fcKACVNAF) CORI-^OP) fATIt-IANI Ml IAVITVSFEC) 4- I1 ELV<1^ ^L>L . I I (MJcCtO M1M> 10 13 lgU'"y 7 ^ ^ 14 (CAN 8 )H ITHOB=10J 9 12 Fig. 13. 78 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD THE POTTERY A considerable amount of pottery was found at Springhead, there being upwards of 20,000 sherds. These include 800 sherds of plain and decorated Samian ware and 1,000 datable coarse ware rims. The majority of this pottery is confined to the Antonine period, although the full range of Roman occupation is included. First there have been traces of Belgic combed ware, foUowed by a sprinkling of Claudian and Flavian pottery, represented mainly by Drags. 29, 30 (decorated) and 15/17, 18, 24/25 and 27. The Antonine pottery is outstanding for its quantity represented mainly by Drags. 37 (decorated) and 31 and 33 (plain). One bowl of Drag. 37 of Hadrianic date was two-thirds complete, but not one complete pot has been found. There was also much coarse ware. Late second century, third and fourth centuries are not so well represented, but examples of all have been found. These include roughcast ware, Castor ware, Rhenish ware, New Forest ware, red colour-coated ware and imitations of Drag. 38. The amount of fourthcentury pottery is not inconsiderable, although most is very fragmentary. It has clearly been necessary to reduce the amount of pottery by a considerable degree for the purposes of the report. However, plain and decorated Samian, potters' marks, mortaria and coarse ware are aU reported on to a certain extent. TERRA SIGILLATA The plain Samian has been reduced to tabular form (Table III) .from which the incidence of various types at the different periods may be clearly seen. As would be expected, Drags. 31 and 33 represent the bulk of the forms, but Drags. 18 and 18/31 are quite weU represented. Antonine pottery is by far the most prominent, but first-century types are not inconsistent with a reasonable occupation at that time. The decorated Samian is described in Appendix I and Ulustrated in Fig. 9. The potters' marks are summarized in Table IV, described in Appendix I I and illustrated in Fig. 13. COARSE WARE The mortaria are described in Appendix III and Ulustrated in Fig. 13. The main body of coarse ware has had to be reduced to manageable proportions by a system of classification. Fortunately about 500 datable rims have been reduced to 15 standard forms and these are Ulustrated in Fig. 10. The frequency of occurrence of these is given 79 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD TABLE in ANALYSIS OF POTTERY TYPES. PLAIN SAMIAN Drag. 15/17 „ 18 „ 18/31 „ 24/25 „ 27 » 31 „ 33 „ 36 „ 36/36 „ 36 „ 38 „ 42 „ 44 „ 45 Curie 11 „ 23 Ritt. 1 9 Walters 77 79 „ 81 Total No. Percentage of Total No. 9 ••a 1 5 5 2 4 1 1 18 7-0 1 1 2 3 1-2 1 o •» eg r-t 19 11 1 31 12-1 l l 4 1 7 2-7 1 CJ 43 rH O 2 2 2 6 2-3 l tr a? H i-H > 3 •s o 3 1 3 2 1 1 2 9 3-5 S o 1 8 81 49 3 1 7 1 7 157 61-1 1 j.-*~.-.-.-. 33C5JG __. 8 Fig. 15. 83 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD be an unfinished casting, and Mr. HuU concludes that it was made on the site. Since it is of Flavian date, it does indicate fairly early industrial activity. MISCELLANEOUS FINDS FuU descriptions and Ulustrations of other finds are given a's foUows : Find Appendix Figure 1. 2. 3. 4. 5. 6. 7. Bronze objects .. Iron objects Bone objects The Charcoals .. The Bones Graffiti Other objects .. . VI . VII . VIII . IX . X . XI . XII 14 and 15 16 and 17 18 — — 17 18 ACKNOWLEDGMENTS Our thanks are first of all due to the farmer, Mr. P. Bartholomew, who has aUowed us to excavate since 1951 on his valuable arable land at some considerable inconvenience to himself. Our thanks are also due to many authorities who commented on the finds, including M. R. Hull, Esq., M.A., F.S.A., for commenting on the fibulae ; R. A. G. Carson, Esq., for his comments on the coins ; J". W. BraUsford, Esq., F.S.A., for his comments on many of the small objects ,- R. P. Wright, Esq., M.A., F.S.A., for his comments on the graffiti ; Dr. F. C. Fraser for his comments on the animal bones ; F. C. Balfour Browne for comments on the-charcoals ; A. Warhurst, Esq., B.A., A.M.A., for comments on various small objects ; A. Melderis, Esq., for comments on the seeds ; and Mrs. J. E. Morey for comments on the stones. The author also wishes to offer special thanks to Mr. Sheppard Frere, M.A., F.S.A., for his criticism, advice and help in the preparation of the report and to Mr. E. TUley for his" help in preparing the report and dealing with many of the small finds. Thanks are also due to Mr. W. Gee for embellishing the report with his excellent drawings and to Mr. P. Connolly for assistance with the photography. Of the many excavators, particular help was given by Messrs. Tilley and Gee, Mr. and Mrs. Tidby, Miss A. Burvill, and Messrs. Connolly, Parsons, Talbot, Voisey and Walton. Finally, the author wishes to thank the British Museum, Maidstone Museum, the Gravesend Pubhc Library and British Insulated CaUenders Cables Ltd. (for the lead analyses) for much assistance, and the President, Chairman and CouncU of the Gravesend Historical Society for much help, advice and encouragement at all stages of the work. 84 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD i'.l •! m iMiiiiMiiiiiii(iililiif|ltl]ili||j||||TRi[i]i;iiPnif||yr' Fig. 16. ,85 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD (SS ~-ffi iiiii)iiiiiii_iiiuiiiipi)iTiimirimTtnTiiiTiHniNnimrTp!iTmnT jfafflmPircniy^ fA a ^ S B i l ^^ V» / ^ 0 M K to M AAAVNA I II Fig. 17. 86 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD I 1 rn • nrnn .• Iii / A ff : \ Fig. 18. REFERENCES (1) The Romano-British Settlement at Springhead, W. S. Penn, Arch. Cant., LXV, 171, 1952. (2) History of Kent, E. Hasted, Vols. 1, 2 and 3, 1797. (3) History of Gravesend, A. J. Phihp, Vol. 1, 1914. (4) V.C.H., Vol. 3 (Victoria County History of Kent). (5) Hasted, Vol. 2, 422-3, 1797. (6) Dunkin, 135. (7) V.C.H., Vol. 3, 91 (see Ref. 62). (8) Arch. Cant., XI, xhi. (9) Dunkin, 140. (10) Dunkin, 144. 87 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD (11) H. Ross, Arch. Journ., XXII, 63. (12) R. F. Jessup, Antiq. Journ., VIII, 338. (13) V.C.H., Vol. 3, 29 ; Antiq. Journ., VIII, 339. (14) Arch. Cant., LXVIII, 103. (15) Cicero and the Roman Republic, F. R. Cavell, Pitman, 1948, 70. (16) The Shorter British Flora, Prime and Deacock, Methuen, 1948, 222. (17) British Plants and Their Uses, Edlin, B. T. Batsford Ltd., 1951, 50. (18) Newstead, Curie, 1911, 360-1. (19) Arch. Cant., XVII, 209. APPENDIX I i DECORATED TERRA SIGILLATA 1. FIGURE 9.1. D.37, K.D. IV. Found over Antonine flint floor with plaster rubble. This was the largest fragment found on the site and is about twothirds complete. The large leaf scroll is its outstanding feature. (1) The absence of a demarcating motif indicates second century. (2) The cable motif of the tongue first appears at the turn of the first and second centuries and the asymmetrically placed tongue also indicates second century. Since the terminal swelling is not at an obtuse angle, however, this would indicate fairly early in the second century. (3) The use of this type of scroll is fairly widespread. The main leaf used is simUar to Wroxeter, XXV, 3 assigned to first half of second century. A somewhat similar scroll is figured in Silchester XXIII which may be CINNAMUS. Great similarity is also shown to a scroll of CINNAMUS (also with rings and fowl free in the field) dated Hadrian/ Antonine (O. and P. XXXI, 37), although the tendril bindings are not the same. (4) The tendril binding is the astragalus type (O. and P. XXXII, 34, 35, 36) which may be assigned to latter end of first century or beginning of second century. It is rare in the Antonine period. This could well be an early pot of CINNAMUS of Lezoux. Probably Hadrianic. 2. FIGURE 9.2. D.37, K.D. IV. Found over Antonine flint floor. A typical bowl of PATERNVS with animals and spindle-shaped ornaments free in the field. See bowl found at Wingham, Kent (Walters, C.R.P. XXXI, M.1463). Bowls found at Warrington, Wroxeter, Colchester, Corbridge, Silchester, Hengistbury and London. Probably pot of PATERNUS of Lezoux. Date : Hadrian/Antonine. 88 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD 3. FIGURE 9.3. D.37, K.D. IV. Found over Antonine flint floor. (1) The cupid contained in the festoon is dated Trajan/Antonine in 0. No. 440 (Lezoux). (2) The divided metope would normaUy indicate comparatively early in the second century, but medalhons, festoons and astragalus bordering and dividing motifs indicate later and possibly East Gaul. Date : c. A.D. 150. 4. FIGURE 9.4. D.37, K.D. VI. Found over Antonine floor. (1) This sherd has a narrow rim and a glaze dulled by wear. (2) The ovolo is divided from the decoration by a bead row, the tongue is placed asymmetricaUy and the terminal is quite simple. This would all indicate second century and possibly Antonine. (3) The divided metope with large medallions and festoons is typically Antonine. (4) The two bearded masks, facing left, are figured in 0. Nos. 1216 and 1218, both of Lezoux, the former dated Antonine and the latter Trajan/Antonine. (5)' The gladiator is very much hke 0. No. 1057 dated' Trajan/ Hadrian and of Lezoux. Thus, although there are many features of this sherd which may be Antonine, there are also several early features. The date may, therefore, be late Hadrianic rather than Antonine. Probably Lezoux and late Hadrianic. 5. FIGURE 9.5. D.37. K.D. VI. Found over chalk/flint road. (1) Bead row divides ovolo from decoration and indicates second century. (2) Two birds are in festoon which could be Antonine. (3) One bird is flying left and one right. Both are figured by Oswald No. 2324, Lezoux, Trajan/Antonine (former) and 0. No. 2316 (Lezoux) Trajan/Antonine. Probably Lezoux and early Antonine. 6. FIGURE 9.6. D.37. K.D. VI. Found over chaUc/flint road. This pot has a compound stalk scroll, with a large divided'leaf and ivy leaf. There is probably a medalhon in the concavity of the scroll. SimUar to 0. and P. XXXI, No. 37, by CINNAMUS of Lezoux, and Antonine. 7. FIGURE 9.7. D.37. K.D. VI. Found over Antonine flint &oor. (1) The usual ovolo, shaped hke a hanging ear-ring, without tongue, and bead row below would indicate Central and East Gaul, and fairly late Antonine. 89 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD (2) The animal free in the field (Antonine), shows no sign of degeneration and is not figured by Oswald. It appears to be a vicious dog. Date : Antonine (c. A.D. 160). 8. FIGURE 9.8. D.30. K.D. VI. Found over pebble/chalk road. (1) Bead rows separate ovolo from decoration and also panels of medalhons. A second century feature. (2) The tongue of the ovolo has a knobbed terminal bent at an obtuse angle which is probably Antonine. Very simUar to an ovolo of CINNAMUS (Newstead II, XL, IV): (3) MedaUion and festoons typicaUy Antonine. (4) Cupid as 0. No. 431 (Lezoux, Antonine) and 0. No. 431A (Rheinzabern, Antonine). Probably CINNAMUS of Lezoux and Antonine. 9. FIGURE 9.9. D.37. K.D. VI. Found over pebble/chalk road. (1) There is no bead row separating ovolo from decoration which could mean second century and East Gaul. (2) The corded tongue indicates second century and DIVIXTUS of Lezoux. (3) Bead row dividing individual motifs indicates second century. (4) The motif is unusual and consists of a medaUion containing a bird, the medaUion being at the centre of four leaves emanating from the corners of the metope. No analogy can be found. (5) The fine glaze indicates no degeneration and therefore not late in the second century. (6) The bird is similar to 0. Nos. 2296 (Lezoux and Hadrian/ Antonine) and 2296A (Lezoux and Antonine). Possibly DIVIXTUS of Lezoux and Antonine. 10. FIGURE 9.10. D.37. K.D. VI. Near section, Figure 4. (1) The bead row separating the ovolo from the decoration and dividing metopes indicates second century. (2) The head is simUar to 0. No. 1269 which is given as Rheinzabern, Antonine and La Madeleine, Hadrian. (3) There is rather an unusual feature separating the panels, with a crown over a head. (4) The style of metope decoration would indicate Antonine. Date : Antonine. 11. FIGURE 9.11. D.29. K.D. VII. Found in section, Figure 5. (1) MoriUiform tendrU binding is early, possibly extending to early Flavian period. (2) The scroll is also early. SimUar scrolls and bindings are to be 90 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD found in Camulodunum XXV Nos. 1 and 7a, both dated early Claudian. Date : Early ; early Flavian being terminal date. 12. FIGURE 9.12. D.37. K.D. VII. Found near section, Figure 6. (1) The finish is poor, but the wavy line dividing the ovolo from the decoration taken with the bead row separating metopes would indicate turn of first and second centuries. (2) The divided metope taken with absence of medalhons indicates Trajanic. SimUar bowl Wroxeter I, XIV, dated c. A.D. 110. Probably Lezoux and Trajanic. 13. FIGURE 9.13. D.37. K.D. VII. Found under Antonine flint floor. Figure 7. (1) Bead row separating ovolo from decoration would indicate second century. (2) The type of divided metope would indicate Trajanic period and is typical of South Gaul. (3) The bear is similar to 0. Nos. 1609 (Antonine) and 0. 1610 (Hadrianic) both of Lezoux. (4) A simUar type of decoration was found at Wroxeter (I., p. 37, 13) in a well dated deposit of A.D. 90-120 and again Lezoux. Probably Lezoux and Hadrianic. 14. FIGURE 9.14. D.29, K.D. IX. Found near section, Figure 7. (1) The rim is hardly everted which would indicate Tiberian example. The decoration is divided into two zones by the central moulding which is not rouletted. (2) The lower frieze has a godroon which is probably pre-Flavian. (3) The upper frieze contains discontinuous festoons containing animals, almost separated into metopes by tassels. This should, indicate Flavian but there is no sign of pinnate leaves, arrow heads or other Flavian motifs. Nearest analogy is bowl by PRIMUS who worked A.D. 40-51 and A.D. 76-83. A bowl by PRIMUS (SUchester, Vila) is dated A.D. 40-51. There is a lower frieze on another bowl at SUchester (IXA) simUar to the present upper frieze, dated before A.D. 45. (4) The bird is probably the Nile goose (Knorr, RottweU 1912, p. 7, PI. II, No. 10) La Graufesenque. Probably PRIMUS of La Graufesenque and Tiberius/Claudius. 15. FIGURE 9.15. D. 29, K.D. IX. Found in section, Figure 5. This is a lower frieze consisting of a continuous winding scroll with an unusual filhng in the concavity. This consists of an upright palmette type leaf from which curves away two other simUar leaves. This is a 91 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD most unusual feature in Claudian pottery (very few examples are figured in the pre-Flavian pottery of Camulodunum) and is more consistent with an early Flavian period. Date : Early Flavian. 16. FIGURE 9.16. D.29, K.D. X. Found in pit, Figure 5. (1) The upper frieze is divided from the rim by a bead row and from the lower frieze by a central moulding (without rouletting) and widely spaced beads on each side of it. An early feature. (2) The frieze is divided into metopes (with rosettes at the corners) containing animals and arrowheads alternately which is a pronounced Flavian feature (see bowl by PASSENUS London, Walters, C.R.P., PI. XXII (M.352).) (3) For Hare see O.2077 La Graufesenque. Claudius/Vespasian. Nero/Vespasian. Date : Probably Vespasianic. APPENDIX II NOTES ON POTTERS' MARKS 1. CRACUNA This potter has also been found at Treves and Xanten. 2. CORIUS A potter of La Graufesenque. 3. ATBDIANUS A Lezoux potter. His activity may be dated to between A.D. 140 and A.D. 190, since eighteen of his pots have been found at Pudding Pan Rock. One stamp is recorded from Colchester (Walters, M.1683) and his moulds have been found at Lezoux (Dech. i, 157) and Moulins. 4. AVITUS This is a well-known Rheinzabern potter of Hadrian/Antonine date. See Curie 232, Forrer, 215, 238 and Ludowici, I, 10, III, 8. Found at London (GuUdhall), Rheinzabern and Ruckingen. 5. ELVILLUS Probably a potter of Lezoux. Found in Britain at Richborough (I, p. 68, No. 26), Cirencester, Colchester, Corbridge, Leicester, Litlington near Royston, Westbury (Devizes Mus.), York and London (B.M.). Seven stamps of this potter were found at Wroxeter, dated c. A.D. 160. 6. Micoio A Rheinzabern potter. Found at Corbridge in Britain. 7. OSBIMANUS An unusual stamp probably of Lezoux and Trajan/Hadrian. SimUar stamps occur at Wroxeter (I, p. 56, No. 82) and Oswald gives 92 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD simUar stamps found at London, Cirencester, Corbridge and Stanwix (Carhsle). (Stamps on Terra Sigillata.) 8. CANTUS A potter of La Graufesenque. Also found at Augst. 9. FRONTINUS A potter of La Graufesenque. NormaUy dated Flavian but Ritterling thinks it may be after A.D. 80. (Hofheim, p. 250.) Another dated find (A.D. 80-110) was at Neuss in the Bregenz-ceUar find. Probably its date is therefore soon after A.D. 80 and the present find position confirms this.- Also found at Newstead (p. 236), Wroxeter, London (London in Roman Times, No. 92), Richborough (I, p. 69, No. 29), Camelon and RottweU. APPENDIX III MORTARIA 1. FIGURE 13.10. From first century Pit, Figure 5, K.D. X. Rim fragment of hard, pale, buff clay. Grit particles on rim and inside surface. Graffito on rim MAXIMA. (See Appendix XI, No. 6.) Coll. Roman Britain type 7 (Richborough, Claudian). Bushe- Fox Type 34. Date : Claudian. 2. FIGURE 13.11. From destroyed south waU, Figure 8. K.D. VII. Rim fragment of hard cream coloured clay. No grit particles. Variant of Wroxeter Type 66. Date : Hadrianic. 3. FIGURE 13.12. From near section, Figure 7. K.D. VII. Rim and spout fragment of white pipe-clay. Sparsely gritted. Inside surface smoked by fire. Wroxeter Type 54. Richborough IV, Type 508. Date : c. A.D. 80-120. 4. FIGURE 13.13. From near section, Figure 7. K.D. VII. Rim and side fragments of hard, pale, buff clay. Studded on rim and inside surface with white stone particles. Wroxeter Type 38. Date : c. A.D. 80-110. 5. FIGURE 13.14. From near section, Figure 7. K.D. VI. Rim and spout fragment of hard, pale, buff clay, early bead and roll type. Studded on inside surface with white stone particles. Herring-bone stamp on rim of double row separated by line, inner diagonals shorter. Wroxeter Type 58 (late first century-early second century). Richborough IV, Type 607 (first century). May, SUchester Type 128. Date : Late first/early second century. 93 10 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD 6. FIGURE 13.15. From near section, Figure 4. K.D. VI. Rim fragment of hard grey clay, buff interior (haematite wash). Two grooves on outside edge of the rim. Wroxeter Type 74 (end of first century to well into second century). Date : Early Antonine. APPENDIX IV COARSE POTTERY FIGURE 10.1. Table V Pie-dish, fumed grey ware, but also other fabrics, with black glaze (cf. Arch. Cant., LXHI, 102, No. 29 from Canterbury). Early first century. FIGURE 10.2. Table V Hard, fine grey ware from rim of carinated beaker. Mid first century. FIGURE 10.3. Table V Bead rim with internal projection. Grey clay with rough surface (cf Arch. Cant., LXVIII, 176, Nos. 9 and 10 from Joyden's Wood). Flavian. FIGURE 10.4. Table V Bowl, hard black matt fabric with in-curving side and internaUy projecting rim (cf. Arch. Cant, LXVIII, 176, No. 9). Flavian. FIGURE 10.5. Table V Large storage jar, coarse brown gritted paste, hght reddish-orange on surface. Zone of stabbed decoration above girth groove. Second half first century. FIGURE 10.6. Table V Lid, hard fumed grey ware, with rough black surface (cf. May, SUchester, Plate LXXIX, No. 7). Late first century. FIGURE 10.7. Table V Jar, with high shouldered pear-shaped outline of coarse buff clay (cf. p. 182, Nos. 50 and 53 from the Earthworks at Charlton). Late first century. FIGURE 10.8. Table V Olla, hard fumed grey ware, bearing relationship to Figure 10.3. Fhst century. FIGURE 10.9. Table V Bowl with down-turned rim, with diagonal line decoration. Grey ware with traces of black glaze (cf. Arch. Cant., LX, 77, No. 6 from Canterbury). Late first/early second centuries. FIGURE 10.10. Table V Pie-dish with angular rim. Hard fumed grey clay with scored 94 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD lattice pattern round the outside. Very common, sometimes black glaze (cf. Richborough I, No. 46). Antonine. FIGURE 10.11. Table V Pie-dish, straight-sided, grey ware (cf. Arch. Cant., LXVI, 28, Nos. 113 and 114 from Lullingstone). Antonine. FIGURE 10.12. Table V Olla with outbent rim. Black clay with black glaze (cf. Arch. Cant., LXVIII, 178, Nos. 22 and 23 from Joyden's Wood). Very common between A.D. 150-200. FIGURE 10.13. Table V Pie-dish, fumed hard grey clay with trelhs pattern (cf. Arch. Cant., LXVIII, 116, No. 73 from Canterbury), A.D. 160-190. FIGURE 10.14. Table V Obhque-sided bowl, bead and flange rim. Black burnished ware with tooled interlacing arcs (cf. Lydney Park. Fig. 27, No. 43). Fourth century. FIGURE 10.15. Table V Dish, hard black clay, with ghth groove near rim (cf. Richborough III, No. 345). Fourth century. FIGURE 10.16 Narrow-necked jar or vase. Two cordons on neck and decorated with row of incised ghth lines. Speckled grey clay with dull orange slip (cf. Richborough I, No. 5). Late Celtic/Claudian. FIGURE 10.17 ShaUow plate, hard grey clay (cf. Richborough III, Nos. 209, 210). Claudian. FIGURE 10.18 Bottle, hard grey clay. Claudian. FIGURE 10.19 Two-ribbed handle, reddish/orange clay, cream shp (cf. Richborough II, No. 143). Claudian. FIGURE 10.20 Butt beaker. Smoothed buff ware (cf. Arch. Cant., LXVIII, 120, No. 85 from Canterbury). Early first century. FIGURE 10.21 Three-ribbed handle of buff clay (cf. Richborough I, No. 35). Mid first century. FIGURE 11.1 Jar with recurved rim. Coarse grey fabric with orange/buff shp (cf. Arch. Cant., LXIII, 101, No. 24 from Canterbury). Mid first century. FIGURE 11.2 Wide-mouthed olla with recurved, thickened hp and cordon on shoulder. Hard grey clay (cf. Richborough IV, No. 384). Pre-Flavian. 95 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD FIGURE 11.3 Jug with five-ringed neck, moulded hp and handle missing. Hard sandy pinkish clay (cf. Richborough IV, No. 374). A.D. 70-90. FIGURE 11.4 Incense cup, with three rows of impressed decoration. Hard, rough buff clay (cf. Richborough IV, No. 421 for general type). Dated A.D. 75-100. FIGURE 11.5 Jar, reddish fabric with dark brown surface (cf. Arch. Cant., LXVIII, 122, No. 104 from Canterbury). Late first century. FIGURE 11.6 Pie-dish, intorned rim reeded on top. Smooth light brown surface. • Fhst century. FIGURE 11.7 Sherd with decoration of concentric incised semi-circles above incised vertical lines. Hard grey-black clay (cf. Richborough III, No. 283, A.D. 80-120). FIGURE 12.1 Carinated beaker, with shghtly raised disc beneath. Burnished grey ware (cf. Arch. Cant., LXVIII, 86, Nos. 2 and 3 from Upchurch Marshes). Fhst quarter, second century. FIGURE 12.2 Olla. Crude figure on side with crossed arms. Probably used as handle. Antonine. FIGURE 12.3 Poppy-head beaker, thin buff ware with black glaze (cf. Arch. Cant,, LIX, 73, No. 3). Antonine. FIGURE 12.4 Roughcast beaker, bulbous type with unusual everted rim (cf. Verulamium No. 9, Fig. 27). Late second century. FIGURE 12.5 Hunt cup, with running deer (cf. May, SUchester, Plate XLIV, No. 1, and Arch. Cant., LXVIII, 181, No. 33 from Joyden's Wood). Late second century. FIGURE 12.6 " Beaker of hard grey clay, smaU rim with two grooves on neck. Late second century. FIGURE 12.7 Bowl with oblique sides and flange. Hard grey clay and dull orange coating. Two burnished regions on side (cf. Richborough I, No. 121). Mid fourth century. FIGURE 12.8 Imitation Drag. 38. Brick red fabric with bright red slip (cf. CoUingwood 32). Fourth century. 96 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD FIGURE 12.9 Beaker with rim recessed for cover. Hard coarse buff clay with orange slip (cf. Richborough IV, No. 469). Fourth century. APPENDIX V FIBULAE (Reported on by M. R. HULL, Esq., M.A., F.S.A.) 1. FIGURE 14.1. K.D. VI. Above pebble/chalk road. Figure 6. A small and well-preserved brooch after the general type of a derivative of Colchester A/B but it is not clear whether it was intended to be sprung or hinged. The head is possibly blundered, or possibly unfinished. There is a faint suggestion either that the crest (not a hook) has been pierced or that piercing has been attempted, but not carried out. If intended to be sprung the wings should be quite flat ; they are, however, small and cylindrical, as if for a hinged brooch. But there is no provision for a hinge. It may be concluded that this is an unfinished casting and that brooches were made on the site. It was intended to be hke B.M. 56.6.27.60 and several like it at St. Albans. Date : Flavian. 2. FIGURE 14.2. K.D. IX. Figure 4. A curious little brooch with hinged pin and belonging to the Hod- HU1 type, but not at all typical. The head looks stouter than it is, the bow has a deep, round-bottomed furrow (in which are two oblique cuts which may have held niello) and on each side of it the marginal ridge is minutely beaded, with shght swelling each side near head. Date : Mid-first century or a httle later. 3. FIGURE 14.3. K.D. IX. Near section, Figure 7. A very slender brooch of Colchester type. Spring of 6 turns held by hook ; wings very thin and plain. If the pin reaUy belongs the catchplate must have been rather long. There are several examples at Colchester, e.g. 1042.31 and Jos. 10009 E, compare Swarling No. 4. Date : Mid-first century or earher. 4. FIGURE 14.4. K.D. IV. Near section, Figure 7. Flat brooch in form of a stag, broken. The body is recessed for enamel which may have been blue on the body and red on the thigh. Compare a very similar example at Wroxeter (which, however, is not looking backward). Shrewsbury Museum, B.135. Date : Antonine. 97 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD 5. FIGURE 14.5. K.D. VI. Over chalk/flint road, Figure 5. Trumpet-headed brooch of a common type. Spring of 6 turns in smaU box behind head. On head three recesses fiUed with remams of whitish enamel. Central disc enameUed white in centre, red outside. (Many paraUels could be quoted.) Date : Antonine. 6. FIGURE 14.6. K.D. VI. Over chalk/flint road. Penannular brooch of very thin, round section, the ends flattened and roUed back, but much smaUer than usual. The pin is slightly ornamented by a smaU roll on the loop at the top. A well-known early type, see Cam. Report, Fig. 59. Date : First half first century. 7. FIGURE 14.7. K.D. II. Unstratified. A one-piece brooch with spring of 8 turns held in a hook. The wings have two sharp grooves and the bow is facetted or chamfered. The type is Cam. I l l , p. 310. Date : c. A.D. 50-65. 8. FIGURE 14.8. K.D. X. In pit, Figure 5. A somewhat simUar brooch with spring of 8 turns, much corroded. The catchplate pierced by 2 holes. Date : c. 60-70. 9. FIGURE 14.9. K.D. IV. Near section, Figure 7. Large brooch of Hod-HUl type but it belongs to a peculiar and interesting group which is growing in numbers each year. The characteristic is that of smaU iron rods put transversely through the bow with a smaU knob or baU on each end. (These are, of course, usuaUy missing.) Examples in which the bars go through the foot instead of the bow are not so numerous. There is one in London in Roman Times, Fig. 26, 14, and a huge one from Colchester ihustrated in " Camulodunum," PI. XCVIII, 159, and there is a very fine one from the Weisenau cemeteries near Mainz in the Maimer Zeitschrift, VIII, 49. The date is the same as for Hod-HUl type generaUy—about midfirst century. APPENDIX VI BRONZE OBJECTS (Reported on site by Mr. E. TILLEY and also by J. W. BRAILSEORD, Esq., F.S.A.) (1) COCHLEAR OR EGG-SPOON. Figure 14.10. K.D. VI. Found in section, Figure 4. Martial (Ep. XVI, 121) refers to its use for picking periwinkles or snaUs out of theh shells with the aid of the pomted end and for eating eggs with the bowl. 98 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD Wheeler, London in Roman Times, PI. XLV, No. 4. Curie, Newstead, PL LXXIII, Fig. 6. (2) LIGULA. Figure 14.11. K.D. VII. Found in section, Figure 4. (3) PAIR OF TWEEZERS. Figure 14.12. K.D. IV. Near section, Figure 7. (4) PENDANT. Figure 14.13. K.D. VI. Near section, Figure 5. There are several variations of this type of pendant from different parts of the country, although theh use and origin are not obvious. The present specimen (Antonine) may have been copied from a horse's barnacles, and consists of two serrated limbs fitting over the nose of the horse, which when puUed together drive the toothed edges into the flesh. This contrivance is stUl used in Spain in place of the bridle and bit. Alternatively for use as an ornament, the saw-edge being omitted to avoid fraying the clothing. See Fig. 52. British Museum, Guide to Roman Britain. (5) SMALL BRONZE RING. K.D. IV. Near section, Figure 5. 0-7 in. diameter with no decoration. (6) NAIL. K.D. IV. Over Antonine flint floor. Has a large hollow domed head. (7) NAO,. K.D. VI. Over Antonine flint floor. A smah naU 0 -9 in. long with a plain chcular knob. (8) BUTTON. K.D. VI. Over Antonine flint floor. A chcular disc of sheet bronze 1 -2 in. diameter pierced with a hole (diameter 0-125 in.) slightly off centre. Similar buttons were found at LuUingstone. (9) BRONZE FRAGMENTS. K.D. IV. Near section, Figure 6. Mr. J . N. BraUsford is of the opinion that these are several fragments of a mirror, sUvered one side but undecorated the other. (10) CRESCENT-SHAPED OBJECT. Figure 15.1. K.D. VI. Found near section, Figure 7. This object is hoUow and is possibly a scabbard fitting. (11) PINS. Figure 15.2. K.D.s II, IV and VI. Ten pins or fragments of common types were found, probably of Antonine date but found in Antonine and later strata. (12) AXE-SHAPED OBJECT. Figure 15.3. K.D. VI. Near section, Figure 5. This object is flat and Mr. J. N. BraUsford is of the opinion that it is a votive miniature axe-head. Alternatively it could be a spatula. (13) LION'S HEAD. Figure 15.4. K.D. IV. Over Antonine flint floor. This is a hoUow bronze casting depicting a hon's mask on a hoUow cylindrical mount. When found it had an hon fixing rivet on the inside 99 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD centre and there were traces of wood. It is probably a boss-cap or finial. Cf. Arch. Cant., 1950, LXIII, PI. VEc. (14) HANDLE OF LIGULA OR PROBE. Figure 15.5. K.D. III. Near section, Figure 8. (15) NEEDLE. Figure 15.6. K.D. IV. Over Antonine flint floor. Common type with the eye formed by splitting the shank and re-joining. (16) LEVER KEY. Figure 15.7. K.D. VI. Over Antonine flint floor. (17) BRACELET. Figure 15.8. K.D. VI. Near section, Figure 5. This is an expanding type bracelet made up of two single lengths of bronze wire, coUed into spirals some distance from the ends, and then turned about each other and shding on themselves. Its large size suggests that it is an armlet. (18) L-SHAPED OBJECT. Figure 15.9. K.D. VIII. Near section, Figure 4. Possibly a votive offering. SimUar to model axes found at Woodeaton, Oxon. (Oxoniensia, XIV, 39, Fig. 8, Nos. 2 and 3) but blade not very pronounced. APPENDIX VII IRON OBJECTS (Reported on site by Mr. E. Tilley) (1) HOOK KEY OR LATCH LIFTER. Figure 16.1. K.D. X. Near section, Figure 4. This key is of Early Iron Age type and may be pre-Roman or early Roman. For its use, see Ward, The Roman Era in Britain. Cf. Richborough IV, Fig. 346. Verulamium XI, p. 220. (2) KEY. Figure 16.2. K.D. VI. Over Antonine flint floor. (3) KEY OR HASP FROM LOCK. Figure 16.3. K.D. I I . Over pebble/ chalk road. This is either a padlock type key or the hasp from a lock (see Wheeler, London in Roman Times, PI. XXX, A, No. 5). (4) KEY. Figure 16.4. K.D. VI. Over Antonine flint floor. (5) KNIFE. Figure 16.5. K.D. VE. Over Antonine flint floor. (6) KNIFE. Figure 17.1. K.D. VI. Over Antonine flint floor. (7) KNIFE. K.D. VI. Over cha„/flint road. (8) SOCKETED ARROW HEAD. Figure 17.2. K.D. II. Over chalk/ flint road. 100 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD (9) AWL OR BIT. Figure 17.3. K.D. III. Near section, Figure 8. This has a pointed butt, of quadrangular section. (10) SKEWER. Figure 17.4. K.D. VI. Over pebble/chalk road. (11) LATCH LIFTER. Figure 17.5. K.D. VI. Over pebble/chaUc road. APPENDIX VIII BONE OBJECTS (Reported by Mr. E. Tilley) (1) COMB. Figure 18.1. K.D. III. Near section, Figure 8. The two rows of teeth are cut from a single piece of bone and held between rectangular bone strips by hon rivets at least four in number. These strips are bevelled on each edge and the surviving end, and decorated with groups of four cuts between the rivets, and vertical cuts along the bevelled edges. Normal late Roman type. Cf. Rich. IV, Nos. 216 and 266 (c. A.D. 400). (2) GAMING COUNTER. K.D. IV. Over Antonine flint floor. An interesting specimen showing the lathe marks. (3) GAMING COUNTER. K.D. VI. Over pebble/chalk road. (4) PINS. Figure 18.2. Many pins and fragments of pins were found over the Antonine flint floor, the pebble/chaUc road and the chalk/flint road. In Key Deposits II, III, IV and V. Most are of Antonine date. The simplest are mere skewers of bone shaped by hand and with Ul-formed heads but some have been turned on the lathe. The heads may be grouped as follows : " plain pointed," " plain flat," " bi-conical," " circular knob " and " decorated with incised lines." (5) NEEDLES. Figure 18.2. Found as for pins. Most were shaped by hand but a few were turned on the lathe. The eyes are made by drUling two small holes close together, although some were burnt through. (6) BONE CHTPPINGS. K.D. VI. Near section, Figure 5. A smaU heap of bone chippings was found, which appears to be waste from the manufacture of bone objects. 101 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD APPENDIX IX THE CHARCOALS Many fragments of charcoal were obtained from the site and a selection was identified by F. L. Balfour-Browne of the Department of Botany, British Museum of Natural History. The report is as foUows: Specimen Type of No. Tree K.D. Position of Find Approximate Date 1 2 3 4 5 6 7 8 9 10 Hazel Hazel Oak Oak Lime Oak Oak Oak Oak Oak VI III VI VI IV IV VIII VI X III Section, Figure 5 Section, Figure 8 Over chalk/flint road Over pebble/chalk floor Section, Figure 4 Over Antonine flint floor Section, Figure 4 Over chalk/flint road Filhng of robber trench Section, Figure 8 Antonine 4th century Early 3rd century Antonine Late 2nd century Antonine Late 1st century Early 3rd century Early 2nd century 4th century 102 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD APPENDIX X THE BONES Many hundreds of bones were found on the site, mainly the remains of food (one bone had been sawn in half). Specimens were identified by Dr. F. C. Eraser of the Department of Zoology, British Museum of Natural History. The foUowing is his report : Specimen No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14, 15, 17 18 19 20 21 Animal Ox Ox Ox Ox Ox Ox Horse Horse Horse Ox or Horse Sheep Sheep Sheep or Goat 16 Sheep or Goat Sheep or Goat Sheep or Goat Pig Dog Domestic Fowl Fragment Incomplete lower jaw Metacarpal Metacarpal Molar Molar Incisor 1st thoracic vertebra Terminal phalange Incisor Rib fragment Fragment of cranium Fragment of cranium Upper jaw fragments with teeth Incomplete lower jaws Hyoid fragments Proximal phalanges Fragment of lower jaw Canine Tibio-tarsus K.D. I ll IV VI VI VI VI III IV VI IV III IV VI VI VI VI VI XII VII 103 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD APPENDIX XI GRAFFITI The graffiti have been reported on by R. P. Wright, Esq., M.A., F.S.A. (1) FIGURE 17.6. K.D. VI. Near section, Figure 7. TA cut on underside of base of poppy-head beaker. (2) FIGURE 17.7. K.D. VI. Near section, Figure 5. VA % cut on sherd of Drag. 33. Part of the upright of the thhd letter survives. (3) FIGURE 17.8. K.D. IV. Over Antonine flint floor. VA (R) ? cut on a sherd of Drag. 33. The thhd letter seems to be " R " but could be otherwise interpreted. (4) FIGURE 17.9. K.D. VI. Over Antonine flint floor. PA cut on the underside of a two-thirds complete Drag. 33 pot. (5) FIGURE 17.10. K.D. VI. Over pebble/chalk road. DMR cut on shoulder of rim and shoulder sherd of a grey clay oUa. (6) FIGURE 17.11. K.D. X. In pit, Figure 5. MAXIMA cut on the large hooked rim of a Bushe-Fox Type 34 mortarium. The letters cut after firing are partly broad, partly thin. Part of M precedes. I and M are ligatured. The mortarium may well have had no more than MAXIMA, a feminine name cut on it. APPENDIX XII OTHER OBJECTS (1) STONES. Reported on by Mrs. J. E. Morey of Geological Survey and Museum. (a) K.D. IV. Over Antonine flint floor. Part of bangle. The specimen is made from a soft brown and speckled rock with shaley partings. It is composed of limonitized quartz, limonite, minute spherules of siliceous material, small flakes of sericite and micaceous material, and a quantity of finely crystalline calcite. The rock is probably a shale (origin unknown). (b) K. D. VII. Under Antonine floor. Fragment of carved circular moulding. A medium-coarse, grey-white, shelly and oolitic limestone with a well-crystalhzed ground mass of calcite. It is undoubtedly a Jurassic oolite, but it is a difficult stone to match precisely. (2) HONES Two fragments of hones both Key Deposit VI and found over Antonine flint floor. The composition of the stone was as follows : 104 ROMANO-BRITISH SETTLEMENT AT SPRINGHEAD A fine-grained, compact grey-brown sUtstone composed of grains (up to 0-1 mm. diameter) of quartz, hmonite, micaceous materials, feldspar, cherty sUica, accessory apatite and zircon. (3) QUERNS Segment of quern found Key Deposit I I I near Section 5. Neidermendig lava. (4) PART OF SHALE BRACELET. Figure 18.3. K.D. VI. Found over Antonine flint floor. Chcular in section with outside edges flattened and a ridge around the inner surface. The specimens are made from a soft brown and speckled rock with shaley partings. They are composed of hmonitized quartz, limonite, minute spherules of sUiceous material, and a quantity of finely crystalline calcite. The rock is probably shale. (5) PLASTER Much plaster was found in Key Deposit IV over the Antonine building. Some had wattle impressions directly on the back whUe some was attached to daub. The plaster had simple line designs on it in Vandyke brown, signal red, orange and green on a cream background. There was also the suggestion of a leaf. Much of it, however, was plain cream and it is probable that the waUs were plain, with a line design around the borders. 105