Springhead Temples II and V

SPRINGHEAD : TEMPLES II AND V By W. S. PENN, B.SC. PART 1. GENERAL INTRODUCTION TEMPLES I, III and IV in the temple complex at Springhead have already been described (1 and 2). The present report continues the story with an account of Temples I I and V. The latter is described in part only as it was not possible to excavate that section of it under the railway embankment. The relationship of the present two temples to Temple I is indicated in Fig. 1. It is still not possible to make an overall assessment of the temple complex as work on it continues. Although no new temple has been found, an examination of the Watling Street road-side ditch, within the temenos, has revealed some interesting features. The most outstanding of these are ten child inhumation burials and six ovens, probably for baking bread. These features wUl be described in the next report. No attempt has been made to give individual dating evidence in the text. Detailed evidence, however, is provided in Tables 1 and 2. Two detailed sections are provided, one for Temple I I (Fig. 2) and one for Temple V (Fig. 3). The position of the former is West/East including the cult statue base and the position of the latter is the same at right angles to the north wall. SUMMARY Temple I I structurally is the most interesting of the two buildings and is unique. Although a " concentric square " type basically, the cella walls are replaced by a number of plinths. The temple is approached from the east by a flight of tiled steps (gradus) flanked by wings. An altar base was discovered in front of the gradus. The most important small find was a rare coin of Carausius. Temple V is a small room connected to a larger building. It is unimportant structurally except for the remarkable tile fall, but included a number of interesting finds of the greatest significance. These were several groups of coins, three face urns, six bracelets, some small beads and a large votive pot. The work carried out on these two temples has indicated, without doubt, that the temple complex at Springhead was a very important pagan rehgious centre and possibly the most important in Britain. The finds stUl continue to reflect, in no uncertain way, the close relationship 110 SPRINGHEAD : TEMPLES II AND V of the complex to similar features in Gaul. Other temples in Britain are usually but a pale reflection of their Continental counterparts. ACKNOWLEDGEMENTS We are once again very grateful to Mr. J. Bartholomew of Messrs. J. Bartholomew & Sons for allowing us the unrestricted use of One Tree Field. Our work has now been proceeding for eleven years which clearly indicates assistance and patience of the highest order. Various authorities were again most helpful to us. Special thanks are again due to Mr. R. A. G. Carson, M.A., for examining and reporting on the coins, particularly his comments and report on the coin of Carausius (3). Miss D. Charlesworth, M.A., was also kind enough to comment on the glass and Mr. R. P. Wright, M.A., F.S.A., reported on the graffiti. The author wishes to thank his friends and colleagues on the site for their unfaihng assistance. Mr. E. W. Tilley, as usual, examined all the coins on site and cleaned, classified and reported on all the small finds, except the pottery. Mr. W. Gee prepared his usual excellent drawings and Mr. P. A. Connolly the photographs, which illustrate extremely well the beautiful state of preservation of some parts of the buildings. The regular excavators to whom the author is most grateful for their continued support were Messrs. E. W. Tilley, W. Gee, G. Buries, F. Turner, P. A. Connolly, S. Harker, R. Ivell, M. Groombridge, and E. Ellet. Others who assisted from time to time were R. G. Bartlett and D. Bartlett. The work at Springhead is carried out under the auspices of the Gravesend Historical Society. All of us who excavate are grateful to the President, Chairman and Council of the Society for their help and encouragement and to the members of the Society for their continued support. PART II. TEMPLE II INTRODUCTION A plan of Temple II is given in Fig. 1 and a section through it in Fig. 2. It was not possible, and in any case it was undesirable, to continue the section through the gradus which would have meant its destruction. As may be seen, the temple was adjacent to Temple I from which it was only separated by a few feet. It was, in fact, connected to Temple I by means of a curtain wall which must have made the two buildings appear to be quite impressive. Temple I was originally built c. A.D. 90 but was reconstructed c. A.D. 200 when its wings, the curtain wall and the whole of Temple I I were built. It is probable, therefore, that the 111 SPRINGHEAD : TEMPLES II AND V two buUdings were in use simultaneously for the last 150 years of their fives. Unfortunately, the detailed data which was obtained from Temple I regarding the cultural use of the building could not be obtained here. In the first place the floor was only 9 to 12 in. below ground level so that everything had been removed by the plough to this level, including most of the tessellated floor. In the second place all the walls of the temple, except those of the wings, were robbed in Victorian times. STRUCTURAL FEATURES AND DATING As in the case of the other temples, almost the whole building was erected on a chahk raft about 14-16 in. thick. This raft was not thought necessary on the west side since a late first-century floor served a similar purpose. The walls around the ambulatory must have been about 21 in. thick (as for Temple I) and were of one build with the walls of the wings. This is different to Temple I where the wing walls made a breaking joint with the main temple. The walls were rendered externally with red painted plaster vividly indicated on the north side, since although the walls had been completely robbed, the plaster remained in situ. There was a rectangular gap in this plaster where the curtain wall joined the ambulatory wall of Temple II. Although the temple is a concentric square type the cella walls were replaced by tUed plinths, four on the front (east side) and three on each other side, the centre one of the west, larger than the others, also presumably serving as a cult statue base. All were substantially made and were carried down to the chahr. raft. This construction suggests that they supported a heavy load, presumably stone or brick pillars, rather than wooden ones. Unfortunately, no traces of these columns were found and could not have survived in situ, due to the proximity to the surface. The upper tiles were in fact loose, clearly scoured by the plough. The spaces between the ambulatory walls and phnths were filled with a light clayey earth imported from elsewhere on the site. It was capped with a weU-made floor consisting of ohalk blocks (lowest), concrete and red tesserae. Unfortunately the plough had removed most of the latter (thus explaining the discovery of the § in. square tesserae over a wide area in the field) but by a lucky chance enough survived on the south-west side to show that the tesserae were continuous from the ambulatory, through the spaces between the columns into the cella. Thus, there was no wall between the columns, indicating an open, cloister-hke arrangement. This is important and the significance is discussed later. The floor base was examined carefuUy all over but there were no 112 PLATE I «. % mm. A. The Gradus of Temple II with part of north wing. B. Detail of steps of gradus. [fare p. 112 PLATE II * * M I • * * - jr i J3*. * . .. if 1 m # » A. Column plinth set in chalk sub floor of Temple II. Plinth strip] ons of Temple II. PLATE ill A. Tile and Stone base of altar east of Temple II. • 1 B. Foundation Pot, Temple II in situ. C. Reverse and Obverse of Rare Coin of Carausius from " Hoard ", Temple II. PLATE IV A. General View of Tile Fall and West Wall of Temple V SPRINGHEAD : TEMPLES II AND V indications of a structural arrangement in the centre of the ceUa. Some type of impluvium might have been expected and its absence almost certainly indicates that the cella was not open to the sky. No trace of tesserae remained in the centre of the cella to indicate a mosaic. A curious 9 in. wide concrete strip existed all round the inside of the ambulatory wall. Its possible significance is discussed later. Although the main body of the temple had a clayey soil filling, there was no such feature in the wings. These had certainly been hollow in antiquity and were similar to the wings of Temple I. The offset inside the wings possibly indicates a wooden floor, but this is unlikely as is discussed later. In any case, there was no internal plaster rendering. The wings were filled with rubble, mainly consisting of red roofing tiles. Some of these were undoubtedly from the roof of the main temple and had been scraped there by the plough, but others may have been from the covering of the wings themselves. A " hoard " of coins and other small objects was found lodged in one part of this rubble in the north wing and is discussed under the heading of " Finds ". The tessellated floor of the temple was about 2 ft. higher than the latest external Roman floor level, which in the front of the building was a pebbled, thick, hard courtyard. This difference in height, clearly a deliberate podium, necessitated the gradus. Two of the steps (1 and 4 from the bottom) were made of pounded tiles and two from broken tUes (2 and 3). There appeared to be a narrow division down the centre of the steps (from top to bottom) perhaps suggesting a light fence. On either side of the bottom step was a quarter round plaster moulding (painted red) and each step was approximately 5 in. high. There were five steps, the odd number thus conforming with the dictum of Vitruvius. A tiled altar phnth was found a few feet in front of the gradus thus simulating many classical examples. A recessed stone altar base was found just to one side of it, a socket for a spigot being bored in the centre. No trace of the altar has been found. THE SIGNIFICANCE OF THE TEMPLE FEATURES THE CLASSICAL APPEARANCE When the gradus had been found, surmounted by a row of column bases, it was at first thought that the structure was that of a classical temple in the Grseco-Itahc style. This, in fact, is the clue to the meaning of some of the structural features of the temple. There can be little doubt that, from the front, it was meant to look like a classical style building. First there was the gradus itself, a common feature of classical temples, with such fine examples as the Maison Carree at Nimes, the 113 SPRINGHEAD : TEMPLES II AND V temple of Augustus and Livia at Vienne and the temple of Fortuna Virilis in Rome. A few Romano-Celtic temples in Britain had steps including the one at Harlow, Essex (4), one at Silchester (5) and the one at Lydney Park (6). Similar examples exist on the Continent (4). However, none of these had the row of columns at the top except for one doubtful example at Jordan Hill (7). This had been dug in 1842 when a gradus and four column foundations were reported, but the site was re-excavated by Lt.-Col. Drew in 1931 when he failed to confirm these findings. Second, there is the altar base in front of the steps. This is a constant feature of temples in Italy, whereas few examples have been reported in Britain or in north-west Gaul. An example occurs at SUchester (5). Third, there is the podium on which the temple was built. Although a raised cella is not uncommon in Britain, a true podium is, although again examples occur at Harlow and Lydney Park. The podium of Temple I I at Springhead must have looked very much like the classical examples quoted above with its high wall all round. The fourth similarity is the cheek walls (the wings of Fig. 1) on either side of the gradus. These had possessed no floors, were not plastered internaUy, and contained many broken roofing tiles. It is possible that the " wings " were covered with tUes to simulate cheek walls as they were in Temple I where the subject is discussed more fully (1). The final feature is the four columns themselves. The well known classical temples are usually hexastyle or more, but the temple of Fortuna Virilis is tetrastyle and Temple II, from the front, must have resembled it greatly. It wfll be noted that the intercolumniation is not constant, the central opening being greater than those on the flanks. This is vaguely in accordance with the precepts of Vitruvius who says that, in the case of the eustyle, where the distance between columns is normaUy 2\ times that of the column diameter, the central intercolumniation can be three times that of a column diameter. Thus, taking all the features mentioned into consideration, Temple I I from the front must have looked very much hke a classical temple. There can be httle doubt that this was the impression which was meant to be given. On the negative side, it should be recorded that no fragments of entablature were found. THE OPEN CELLA The problem of the nature of the superstructure of concentric square temples was raised in a previous report (1). There, doubt was thrown on the conventional picture (for this see References 4 and 8) of the superstructure which is normally represented as a high central 114 SPRINGHEAD : TEMPLES II AND V tower, with clerestory windows, and a pent roof over the portico, supported on dwarf columns on the portico walls. As an alternative to this a solid outer wall with a possible open cella was suggested. The nearest proof it is possible to have for this suggestion is provided by Temple II. It has already been shown that the ceUa was quite open, the roof being supported by columns. Surely in such a case the outer walls had to be sohd? Accepting this, it may be stated that the basic plans of Temples I and II are almost identical which suggests that it was intended that the superstructure should be more or less the same. Other examples of an open cella exist. The best example is at Lydney Park (6) where the theory of a basilica type building is expounded at length. Solid outer waUs and an internal colonnade were proposed for the concentric octagonal temple at Chew Stoke (9), where examples from Stuttgart and Avallon as well as the early Christian Lateran Baptistry in Rome (A.D. 432-40) were quoted. A simUar temple at Weycock, Berks (10) had outer walls surviving to a height of 8 ft., which certainly suggests that there could not have been dwarf columns on top. Other examples can be quoted, particularly from the Continent (11). What is proposed, therefore, for Temple II is an internal colonnade with solid outer walls. The columns may have extended shghtly above the pent-roof of the ambulatory to provide a clerestory. There was no hole in the roof. A similar arrangement probably existed for Temple I. I t is now possible to consider the interesting concrete surround on the inside of the ambulatory wall. The best explanation of this is that it was a stone seat placed there for the benefit of the devotees. It is possible to find a distant analogy for this in the Mysteries of Eleusis. There the Hall of Initiation survives and on all sides of the square bunding are stone seats, eight steps high. The initiates sat on these steps and watched mimed plays. Something similar could have happened at Springhead but in any case the devotees must have watched some rehgious ceremony. This feature in turn raises another interesting point. In Italy temples were originally merely storehouses for the cult statue and its treasure and sacrifices were performed outside. No doubt this also happened in Gaul and Britain, as evidenced by the altar in front of Temple II. Such northern altars are rare, however, and the ceremonies probably took place inside the buildings, particularly during the winter. This movement probably took place early in the development of concentric square temples and, in fact, the temples may well have been designed to hold a " congregation ". The design was, in fact, an inverted classical temple and would be ideal for protecting the devotees from the weather and allowing them to watch the ceremony. 115 SPRINGHEAD : TEMPLES II AND V THE FINDS Since the temple floor had been scoured by the plough and the walls robbed by the Victorians there were no finds on the temple floor which was a marked feature of Temples I and IV. The only finds in their original positions were the votive pot, buried outside the base of the west ambulatory wall (Table 5,7) and the stone altar base already mentioned. This lack of in situ finds was compensated by a discovery in the tile fall of the north wing of the temple. In one of the crevices, protected by the surrounding tiles, was a collection of finds, the contents of which are summarized in Table 12. At first sight it would appear that this remarkable coUection had been scoured off the temple floor by the plough and had been trapped among the tiles. This is, of course, possible but it is also possible that many of the objects were buried there deliberately, when the temple was partly ruined, sometime in the mid-fourth century. For this reason the collection may reasonably be called a " hoard ". The hoard can only be dated from its own contents. There were many standard Springhead sherds of fourth-century date and the latest coin was dated A.D. 337-40. The hoard may well date to A.D. 350-60 when the temple too was possibly destroyed. Certainly this was the date of the destruction of Temple I. Some of the finds are of interest. There were 23 coins in all, one being a rare coin of Carausius (Table 3,27) kindly examined and reported on by R. A. G. Carson (3). Briefly, the coin is outstanding because of its reverse MONETA AVGGG whereas the more common reverse is that of Pax. Only one other coin of the Springhead type has been previously found and the present coin has been presented to the British Museum by the Gravesend Historical Society. Some of the other finds may have had a ritual character. The two bronze plaques with holes may have been nafied to the temple wall and have carried graffiti, but these unfortunately were lacking. The tiny glass beads may have been associated with a child or mother cult as was the case with Temple IV. The broken flint blade certainly had a votive character and many examples have been found on the Continent (see Reference 11—Cite de Limes, near Dieppe ; Saint-Ouen-de-Thouberville ; Le Catelier de Criquebeuf-sur-Seine, etc.). Finally, the threelegged candle-holder may have been a means of lighting the buildings but more possibly for religious purposes (i.e. a candle on an altar). PART III—TEMPLE V INTRODUCTION The first feature of this temple to record is the interesting way in which it was discovered. A section was taken out from the south wall 116 F I G U R E I. P O S I T I O N OF TEMPLES ROOF FALL ^ / ^ ^ S\ /^ /^\ ^s TEMPLE I UNEXCAVATFD RAILWAY E M B A N K M E N T -CONCRETE BENCH VOTIVE is roOPclD m STATUE TEMPLE V ©. R O B B E R TRENCH CURTAIN WAU WING TEMPLE I S GRADUS QUARTER. ROUND P E B B L E D C O U R T Y A R D MOUI P ING TILED ALTAR fASE FEET PIG. 1. Position of Temples. SPRINGHEAD : TEMPLES II AND V of Temple II, to determine the extra-mural stratification. Within a few feet a tegula was found standing at an angle to the horizontal. This discovery was followed by another, an imbrex still in position over the flange of the tegula. Further discoveries then followed rapidly and in all eight tegulae and imbrices were found in a row and they still continued into the railway embankment. I t was clear that this was a complete hne of tiles from the roof of a building which were still cemented together. It was surmised that the wall was nearby and the supposition proved to be true. The building was Temple V. Unfortunately, the building could only be partially examined as most of it is covered by the railway embankment. Pig. 1 shows the relationship of the position of the building to Temple II, to which it is parallel and close. STRUCTURAL FEATURES AND DATING As will be seen from Fig. 1 the main part of the building excavated consisted of a small room 15 ft. 8 in. by 10 ft. 9 in. The walls were 16 in. wide. The walls were plastered internally and were buff coloured with a few thin red lines in unusual shapes (Fig. 4,12). However, a few more regular black lines were found, two of these terminating in a Fleur de Lys motif (Fig. 4,13). The in situ plaster finished a few inches above a clay fill on to which more plaster had fallen. This suggested that the temple had had a wooden floor and the significance of this is suggested later. The clay fill, clearly a part of the temple building operations, contained several coins all of which indicated a late third-century or early fourth-century date for the building. This is the first time that stratified coins, of the third century, and dating a building, have been found at Springhead. It is clearly of great importance since it suggests that there was no thought of abandoning the site at this time. The devotees still had sufficient faith in their settlement, their temples and their gods to believe it worthwhile to erect yet another temple. At this point it may be noted that the temple complex at Springhead gradually developed over many years. Temple I was first (A.D. 90) ; followed by the pedestal in front of it (A.D. 90-110) ; followed by Temples I I I and IV (c. A.D. 140-60) ; followed by Temple II (late second or early third century) ; and finally Temple V (late third, early fourth century). There may, of course, yet be other temples. I t might be thought that the small room was an annexe to the temple proper. An examination of the plan in Fig. 1 will show that this was not so. The only wall of the remainder of the building discovered forms a breaking joint with the small room wall and there is no 117 SPRINGHEAD : TEMPLES II AND V '•%';&» V wrnrrrnn i Mm$ nmmMWm 0 FIG. i. The Pottery and Plaster. (All J) 118 SPRINGHEAD : TEMPLES II AND V north-south wall. This is a most unusual arrangement and difficult to explain. Only complete excavation can give the answer but one possible explanation is that the small room blocked a previous entrance in the main building. All the walls were well made of flints. There were three courses of bonding tiles in the north-east corner of the small room and the same in the end of the solitary wall of the main building. THE FINDS AND THEIR SIGNIFICANCE The first indication that the finds were to be at all unusual came at the start of the clearing of the plaster rubble from just inside the west wall of the small room. On top of the plaster was found a small " hoard " of eight coins (Table 4, Coin Nos. 1-8) all of virtually the same date ending at A.D. 375. The significance of this was not realized at first, but only when the finds under the plaster rubble had been made. It should be stressed that all the finds about to be described were discovered along the west wall of the small room starting at the north end. Since the whole length was only 10ft. 9 in. the concentration of finds will be obvious. There can be httle doubt that votive offerings were only placed on or by this wall. The first find was a small group of four coins of third-century date. Obviously no significance could be attached to this. About 9 in. away, however, another group of four coins, a little scattered, was found. A little further away was another group of four coins, then another, then one of three and finally another of three. This made 22 coins in all (Table 4, Coin Nos. 9-30) discovered in a length of wall about 4 ft. The discovery of these coins in groups suggested that they may have been attached to the temple wall in small bags. These had eventually fallen to the ground, the bags had decomposed and left the coins only. Such votive coins appeared at first to represent the usual Roman donarium given for favours expected or received. An analogy from Lydney Park, however, indicated that the coins may have served a different purpose. In the first place when Charles Bathurst undertook an excavation in 1805 he found in the north-west of the cella " . . . close adjoining, and chiefly in a space only a few inches between the edge of the pavement walls and floor of the three compartments, and along their respective walls, were found not less than 531 coins ". It will be noted that the Springhead finds were on the north-west wall, a fact, perhaps, more than a coincidence and indicating some special significance for this. The west wall of Temple I was provided with a suggestus for a cult statue and the west wall of Temple III was higher and more strongly made than the others. It seems from the finds at Lydney and 119 SPRINGHEAD ; TEMPLES I I AND V ^ c f e o ^ ^ w®. G ;&*(?>.;p v@7wa n i W i i t i i l l D n i i i n i i i i i m i M m i i i i i i U i i i i r i l i i i i i l i i i m i l l l i l i i i i i i i i i i in iniiuD 5 FIG. 5. The Bronze Ware. (All 120 SPRINGHEAD : TEMPLES II AND V Springhead that the coins must have been offered for a similar purpose. It may well be that Lydney, unwittingly, has provided an answer to the problem of the purpose of the coins. When the temple was reexcavated by R. E. M. and T. V. Wheeler, the tessellated floor in the cella had gone. Fortunately, Bathurst had pubhshed a reasonable facsimile of it. There is some controversy over the exact wording but the general meaning is thought to be : " To the god Nodens, Titus Flavius Senilus, officer in charge of the supply-depot of the fleet, laid this pavement out of money offerings . . . " The conclusion seems obvious. At Lydney, the faithful had been making offerings for the laying of a tessellated floor and perhaps those discovered in the cella there were for some further embellishments. It seems reasonable to believe that the coins at Springhead were for a similar purpose. It will be recalled that the temple had had a wooden floor and the offerings may well have been placed there with the ultimate object of replacing the wooden with a tessellated floor. Two further points may be made regarding these coins. In the first place, although the temple was eventually deserted, the coins were not removed. Secondly, the eight coins on top of the rubble may have been placed there by some late devotee, knowing that there was a pagan temple below the rubble. Both features could have been due to a lingering regard for the old gods. Further coin finds along the wall were expected, but none were forthcoming. Instead, a bronze bracelet was discovered. Within the space of 2 ft. another five came to hght. Clearly these were votive offerings which had been attached to the wall. At Lydney, 270 bracelets were found within the temple area, although not in the temple itself. As mentioned previously (2), these bracelets were almost certainly associated with children and/or a fertility cult. This association with children received some confirmation from the next find. It consisted of a number of tiny beads from the same area as the bracelets. Similar beads (Fig. 6,6) were found in Temple II in the " hoard ". It seems reasonable to assume that the beads served a similar votive purpose. The next find of interest was a face urn (Fig. 4,3 and Table 5,3) whose crude features were easily distinguishable. The decoration on the throat of the pot was very reminiscent of Celtic art which suggested that the face was that of a Celtic god. This is confirmed by the discovery of face urns at Richborough (12) some with inscriptions of Celtic gods. The Temple V face was quite dissimilar to the Richborough examples and could not help with identifying the god. Nevertheless, it is probably the nearest approach that will be made to an image of the god or gods worshipped at Springhead. After this first 121 12 SPRINGHEAD : TEMPLES II AND V SSTtaM. - « « w « ^ •S«*«WM«™w 'Wfr^^'nTTnmiii,niroiiii,uliil|)|[tTn A IO t> II FIG. 6. Miscellaneous Objects. (All I) 122 SPRINGHEAD : TEMPLES I I AND V discovery a fragment of another face urn was found (Fig. 4,4) and then a free face. This had either become detached from a pot or had been a votive offering in its own right. It may be noted that in Britain and France, no other face urns have been found in situ in temples. The Springhead finds, therefore, confirm their votive character. The last find in the small room was a large buff pot (Fig. 4,6 and Table 5,4). This was found broken in the south-west corner and the fragments comprised about three-quarters of the original pot. It may have been placed in the corner to receive votive offerings. Its size, 24 in. high and 24 in. diameter, was ample for this purpose. TABLE 1 DATING THE STRATA, TEMPLE II Stratum B. B/C. G. Datable Object Coin Coarse ware Coarse ware Coarse ware Note: This stratum Decorated Samian Coarse ware Coarse ware Note: This stratum Coin Potters Mark Coarse ware Coarse ware Note: This stratum Coarse ware Coarse ware Coarse ware Coarse ware Note: This stratum Coin Coin Note: This stratum Date Reference A.D. 86 Table 3, 1 c. A.D. 75 Springhead Standard Form S.l (Arch. Cant., LXXII, 1958) c. A.D. 60 S.25 o. A.D. 60 S.30 may be dated late Ist/early 2nd century. Vespasianic Table 6, 2 c. A.D. 60 S.25 A.D. 50-80 S.27 may be dated late lst/early 2nd century. A.D. 119 2nd half 2nd century Antonine Late 2nd century may be dated late 2nd/early 3rd century, Late 2nd/early 3rd Table 5, 7 century Antonine A.D. 80-120 Late 2nd century may be dated late 2nd/early 3rd century A.D. 383-8 Table 3, 10 A.D. 351-3 Table 3,9 may be dated late 4th century. Table 3, 2 Table 5, 1 S.6 S.10 (2 sherds) S.6 S.9 S.10 (4 sherds) TABLE 2 DATING THE STRATA, TEMPLE V Stratum Datable Object Date Reference B. Coarse ware 1st century Table 5, 5 Coarse ware A.D. 75-90 S.40 (Arch. Cant., LXXIII, 1959) Note: This stratum may be dated late lst/early 2nd century. D. 11 coins General and latest Table 4, 29-4, 39 date A.D. 275 Note: This stratum may be dated late 3rd century. D/H. 18 coins Latest date 330/335 Table 4, 9-4, 26 Note: This stratum may be dated to the 2nd half of the 4th century. E. Coarse ware Late 2nd century S.10 (2 sherds) Note; This stratum may be dated late 2nd century. 123 SPRINGHEAD : TEMPLES II AND V Stratum Datable Object Date Reference F. 2 coins Latest date 330/324 Table 4, 27 and 4, 28 Coarse ware Late 4th century S.7 (2 sherds) Note: This stratum may be dated to the late 4th century. I. 8 coins General and latest Table 4, 1-4, 8 date 367/375 Note: This stratum may be dated to the late 4th century. Coin No. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Position Stratum B Stratum E. In robber trench, Over Steps. In tile fall. North wing. Over steps In tile fall. North wing Over steps. Stratum G. Stratum G. Over steps. TABLE 3 THE COINS, TEMPLE II (Reported on by R. A. G. Carson, Esq., M.A.) 12. 13. 14. 15. 16. 17. 18, Description Domitian. (As) RIC. II, p. 196, No. 335 (A.D. 86.) Hadrian. 2 AE. (As) RIC. II, p. 412. No. 577 (b). (A.D. 119.) Antoninus Pius (Denarius). RIC. I l l , p. 26. No. 9. (A.D. 138-61.) Caracalla. Denarius. RIC. IV, i. No. 141. (A.D. 201-06.) Valerian senior. 3 AE. RIC. V, i, p . 45. No. 86. (A.D. 254.) Tetricus senior. 3 AE. RIC. V, 2 p . 408, No. 80. (A.D. 268-73.) Constantine I I . 3 AE Q. LRBC. I. 63. (A.D. 330-35.) Constans. 3 AE. LRBC. I. 102. (A.D. 337-41.) Magnentius. 2 AE. LRBC. I I . 7. (A.D. 351-3.) Magnus Maximus. 3 AE Q. RIC. IX, p. 30, No. 87a ; LRBC. I I . 156. (A.D. 383-8.) ' Honorius? Obv. Illegible. Bust diademed and draped r. Rev. (VICTORIA) AVGGG Victory advancing 1. Cf. RIC. IX. Arelate, p. 70. No. 30. G. THE " HOARD ". (In tile fall in north wing of Temple II) Faustina I. RIC. I l l , p. 161. No. 1099. Gallienus. RIC. V, i, p. 146. No. 177. Gallienus. 3 AE. RIC. V, i, p. 145. No. 163. (A.D. 253-68.) (A.D. 253-68.) Uncertain. Probably Postumus or Victorinus. Obv. Illegible Bust, radiate, r. Rev. Illegible Sol, radiate, striding l.j r. hand raised, 1. hand holding whip. Fragment of an Antoninianus. Victorinus. 3 AE. RIC. V, 2, p. 393. No. 71. (A.D. 265-7.) Victorinus. Antoninianus. RIC. V, 2, p . 392. No. 59. (A.D. 265-7.) Claudius I I . Gothicus. 3 AE. RIC. V, i, p. 223. No, 149a. (A.D. 268-70.) 124 SPRINGHEAD : TEMPLES II AND V Coin No. Position Description 19. Claudius II. Gothicus. 3 AE. RIC. V, i, p. 219. No. 109 F. (A.D. 268-70.) 20. Claudius II. Gothicus. 3 AE. RIC. V, i, p. 212. No. 14. (A.D. 268-70.) 21. Tetricus I. 3 AE. RIC. V 2, p. 409. No. 101 c. (A.D. 268-73.) 22. Tetricus I. 3 AE. RIC. V 2, p. 409. No. 101 c. (A.D. 268-73.) 23. Tetricus I. 3 AE. RIC. V 2, p. 408. No. 90. (A.D. 268-73.) 24. Tetricus Senior. 3 AE. RIC. V 2, p. 408. No. 90. (A.D. 268-73.) 25. Tetricus II. 3 AE. RIC. V 2, p. 424. No. 272. (A.D. 268-73.) 26. Allectus. 3 AE. RIC. V 2, p. 561. No. 33 c. (A.D. 293-6.) 27. Carausius. 3 AE. Obv. CARAUSIUS ET FRATRES SUI. Jugate, radiate and cuirassed busts of Carausius, Diooletian Maximianus, 1. AH three with r. hand raised. Rev. MONETA AVGGG. Moneta standing 1, holding scales and cornucopiae. Mint mark SIP . C 28. Carausius. 3 AE. RIC. V 2, p. 493. No. 335. (A.D. 287-93.) 29. Not previously recorded but similar to RIC. V 2, p. 650. No. 1. (A.D. 287-93.) 30. Maximianus Herculeus. 2 AE. (Follis.) C.H. AOS. (c. A.D. 300.) 31. Constantine I. 3 AE. K. 39. (A.D. 307-13.) 32. Constantine I. 3 AE. CHAOS. 84. (c. A.D. 311.) 33. Constantinopolis type (Imitation of). Obv. . . . TINOP (blundered). Helmeted bust, 1. Rev. She-wolf 1. suckling Romulus and Remus ; two stars above. Cf. LRBC. I. 51. (c. A.D. 330-35.) 34. Constans. 3 AE. LRBC. I. 126. (A.D. 337-40.) 35. Constantius II. 3 AE. LRBC. I. 631. (A.D. 341-46.) (23 coins in hoard) TABLE 4 THE COINS, TEMPLE V (Reported on by R. A. G. Carson, Esq., M.A.) Stratum I (Just in top of rubble inside building) Coin No. Description 1. Valentinian 1. 3 AE. LRBC. II, 969. (A.D. 364-67.) 2. Valentinian 1. 3 AE. LRBC. II, 1003. (A.D. 364-67.) 125 SPRINGHEAD : TEMPLES II AND V Coin No. Description 3. Valentinian I. 3 AE. LRBC. II, 525. (A.D. 367-75.) 4. Valentinian 1. 3 AE. LRBC. II, 527. (A.D. 367-75.) 5. Valens. 3 AE. LRBC. LI, 352. (A.D. 367-75.) 6. Valens. Obv. DN VALENS PF AVG. Bust draped and diademed, rt. Rev. SECURITAS REIPVLICAE Victory to 1., holding wreath and palm. mm. OF/T "LV= 3 AE 7. Gratian. 3 AE. LRBC. II, 349. (A.D. 367-75.) 8. Gratian. 3 AE. LRBC. II, 1401. (A.D. 367-75.) Stratum DjH (Just under plaster rubble inside building) 9. Gallienus. 3 AE. RIC. V.i. No. 280K. (A.D. 259-68.) 10. Victorinus. 3 AE. RIC. V.2. No. 67. (A.D. 268-70.) 11. Victorinus. 3 AE. RIC V.2. No. 161C. (A.D. 268-70.) 12. Victorinus. 3 AE. RIC. V.2. No. 78. (A.D. 268-70.) 13. Victorinus. 3 AE. RIC. V.2. No. 69. (A.D. 268-70.) 14. Victorinus. 3 AE. RIC. V.2. No. 78. (A.D. 268-70.) 15. Claudius II. Gothicus. 3 AE. RIC. V.i. No. 67. (A.D. 268-70.) 16. Claudius II. Gothicus (Imitation). 3 AE. (c. A.D. 270.) Obv. Legend partly off flan, the rest meaningless. Bust, radiate (and draped?) r. Rev. . . . TVS AVG Virtus, helmeted, standing 1., holding spear and resting on a shield. Cf. RIC. V.i. No. 109. 17. Claudius II. Gothicus. 3 AE. RIC. V.i. No. 18F. (A.D. 268-70.) 18. Claudius Gothicus. 3 AE. RIC. V.i. No. 191F. (A.D. 268-70.) 19 and 20. Tetricus I. 3 AE. RIC. V.2. No. 101. (A.D. 270-74.) 21. Tetricus I. Obv. IMPTETRICUS P F AUG. Bust, radiate and draped rt. Rev. LAETITIA AVGG (or AVGN) Laetitia standing 1., with wreath and anchor (3 AE). Cf. RIC. V.2. No. 90. (A.D. 270-74.) 22. Tetricus II. 3 AE. RIC. V.2. No. 259. (A.D. 270-74.) 23. Carausius. 3 AE. RIC. V.2. No. 179. (A.D. 287-93.) 24. Carausius. 3 AE. RIC. V.2. No. 131. (A.D. 287-93.) 25. Carausius. 3 AE. RIC. V.2. No. 895. (A.D. 287-93.) 26. Constantine I. 3 AE. LRBC. I. 840. (A.D. 330-35.) 126 SPRINGHEAD : TEMPLES H AND V Coin No. Description 27. Constantine I (Urbs Roma). 3 AE. LRBC. I . 51. (A.D. 330-35.) 28. Constantine I. (Constantinopolis). 3 AE. (A.D. 330-35.) Obv. CONSTANTINOPOLIS. Bust of Constantinopolis, helmeted, 1., with sceptre over shoulder. - Rev. Victory standing 1., with spear and shield, placing r. foot on prow of vessel, m.m. lost. Cf. LRBC. I . 52. 29. As Coin No. 24 except on rev. before Victory a palm branch. 30. Constantinian. 3 AE. (A.D. 330-35.) Obv. Bust similar to Coin No. 24 but details obscure. Inscription illegible. Rev. Two standards between two soldiers, m.m. TRS. Inscription illegible. Stratum E (under rubble outside building) 31. Claudius I I . Gothicus. 3 AE. RIC. V.i. No. 74. (A.D. 268-70.) 32. Constantine I. 3 AE. K. 212 (iv.) (A.D. 320-324.) Stratum D (Levelled clay inside building) 33. Hadrian. Sestertius. RIC. I I , p. 438. No. 761. (A.D. 134-38.) 34. Commodus. Sestertius. RIC. I l l , p. 416. No. 440. (A.D. 183-4.) 35. Victorinus. 3 AE. RIC. V.2. No. 161C. (A.D. 268-70.) 36. Tetricus I . 3 AE. RIC. V.2. No. 142. (A.D. 270-74.) 37. Tetricus I I . 3 AE. RIC. V.2. No. 271. (A.D. 270-74.) 38. Tetricus I. (Imitation). 2 AE. (c. A.D. 275.) Obv. IMP TETRICUS P F AVG. Bust, radiate, r. Rev. . . . AVG. Prototype cf. Aequitas Avg. Cf. RIC. V.2. No. 62. 39. Tetricus I. (Imitation). 2 AE. (c. A.D. 275.) Obv. IMP C TET Bust, radiate, r. Rev. Inscription illegible. Standing figure, 1. TABLE 5 THE SPECIAL POTTERY 1. Potters Mark. DOCILIS in retrograde (Figure 4.1). Found in filling under T. I I floor (Stratum E). Drag 33. Rheinzabern, 2nd half of 2nd century. 2. Decorated Samian Ware (Fig. 4.2). Found under pebble floor Temple H area (Stratum B/C). The multifld terminal (Drag 37) is suggestive of Flavian date but the four teeth suggest late in the period. The wavy line demarcation suggests a Flavian date. The style of the festoons, leaves and tassel is very reminiscent of Mommo of La Graufesenque (see Oswald and Pryce, Plate XV.3) so that the sherd may be attributable to the latest date of this potter, viz. Vespasian. 3. The Face Pots (Figs. 4.3, 4.4 and 4.6). All three pots were found under the rubble from the walls of Temple V (Stratum D/H). Fig. 4.3 represents a stylized face formed in situ on the pot. In the case of Fig. 4.4 it seems that the face was made separately and then pressed on to the unflred pot, the two being fired together. In the case of Fig. 4.5, this is the face only. It could have become detached from a pot or, alternatively, it may have been a crude representation of a god whioh had never been attached to a pot. 127 SPRINGHEAD : TEMPLES II AND V 6. Some similar face pots were found at Richborough (Third Report, Plate XLIII) and were attributed to the 3rd century. In the case of Richborough the faces were taken to represent those of gods. Two of the faces were identified as Sol and an inscription may be Ogmius the Gaulish Hercules. None of the faces, however, resembles the Springhead examples which, nevertheless, may be taken as certain to represent Celtic gods. Large Storage Pot (Fig. 4.6). This was found under the rubble of Temple V (Stratum D/H) and was probably to receive votive offerings. Two-thirds of it were recovered and its girth was 24 in., height 24 in. and diameter inside the neck 9£ in. This pot would normally be dated to the 1st century (Springhead Report, Arch. Cant., LXXII, Fig. 8.2, A.D. 50-80) and Springhead, Arch. Cant., LXXI, Fig. 10.5—2nd half of 1st century) so that the present pot must presumably be assigned the same date. Since it was certainly in use in the late 3rd century, it must have survived a long time and was probably used in its broken form. Large Storage Pot (Fig. 4.7). Found under pebble floor outside Temple V (Stratum B). Similar to 4 and probably of 1st century date. Incense Cup Fragments (Figs. 4.8, 4.9 and 4.10). Found in rubble, south wing Temple II. These are similar to the many other fragments of pale buff incense cups found at Springhead. They may be dated between A.D. 175 and 210. (See Springhead, Arch. Cant., LXXIV, Fig. 8.) The Foundation Pot (Fig. 4.11). Found at base of wall outside Temple II (Stratum F), see also Fig. 1. This is a very difficult pot to date and no exact analogy has been found. The vertical bead rim is not very pronounced but if it had been it would have had a resemblance to the 1st century pot from Lullingstone (Arch. Cant., LXV, p. 122, Fig. 4.15). There is a much greater resemblance to a pot previously found at Springhead (Arch. Cant., LXXI, Fig. 12.6) dated to the end of the 2nd century. It is somewhat similar to the pot from Wroxeter (Third Report, Plate XXVIII, Fig. 80) dated A.D. 80-130 and it also resembles numerous bulbous beakers of third century date. The date is, therefore, too vague to date the temple but it seems that it must be at least towards the end of the 2nd century. Object No. 1. TABLE 6 THE BRONZE OBJECTS (Reported on by E. W. Tilley, Esq.) Figure 4. 7. Position No. II. Rubble layer filling — of north wing. II. Over steps, unstrati- — fled. II. II. As object No. 1. As Object No. I II. As Object No. 2. — II. Unstratified, in front — of north wing (plough level). II. Plough level, west of 5.7 temple. II. As Object No. 6. — 9. II. As Object No. 6. Description Small bronze hollow-domed stud for fixing to leather. The dome is lead filled. Bronze needle. Common type with the eye formed by splitting the shank and joining up again. Expanding ring of square, bronze wire. Average diameter f in. Ring of heavy/thick bronze wire. Diameter i$in. Bronze needle of common type. Pair of bronze tweezers. Thin bronze bracelet, incomplete. The hooked end fastened into a loop. Bronze bracelet with transverse cuts, giving the appearance of beading. The fastening is the hook-and-eye type, which is usual in the Roman period. Part of a thin wire ring with the existing terminal folded back and flattened in the form of a snake's head. 128 SPRINGHEAD : TEMPLES II AND V Object Figure No. Position No. 10. II. As Object No. 1. 5.8 Description 11. II. As Object No. 1. 5.10 12. II. As Object No. 1. 13. II. As Object No. 1. 14. II. As Object No. 1. 15. II. As Object No. 1. 16. V. Under and in plaster rubble inside building. (Stratum D/H.) 5.9 5.6 5.5 5.2 6.3 5.4 5.1 Fragment of thin bronze strip. Possibly from a bracelet : decorated with a single central line of punched dot-and-circle design. Lion's head. A heavy hollow, bronze casting, depicting a lion's head in high relief, with an incised fringe representing the mane. It was anciently attached to wood (possibly a stave) by means of a long iron tang. Diameter If in. Circular plate of bronze, diameter 1J in., slightly convex, with a central hole. Possibly an ink-well top. Small, thin, bronze-wire finger ring of a common type. In rubble layer. A ligula, length 5-1 in. cf. B.M. Guide to the Antiquities of Roman Britain, 1951, Fig. 5, No. 4. SmaU uninscribed bronze plate, with hole. Bracelets. Six bracelets with different designs were found as follows : (a) Bronze striated bracelet of flattened section and rounded terminals. (6) Bronze striated bracelet. The overlapped joint is ohamfered to give an equal thickness all round. (c) Twisted bronze bracelet of the hook and eye type. Incised transverse lines at the terminals. (d) A fine heavy bronze bracelet of the hook and eye type. Elaborately decorated with transversed lines, concentric circles, and oblong panels. (e) Brass bracelet of the hook and eye type. Highly decorated with circular punch marks and incised lines, cf. Lydney Park, Fig. 17 D and F. ( / ) Bronze bracelet ornamented with punched dot-and-circle design. Cf. Lydney Park, Fig. 17, E. Object No. 2. TABLE 7 METAL WARE OTHER THAN BRONZE (Reported on by E. W. Tilley, Esq.) Figure Description Iron door-hinge of a smaller type than those from Temple I. cf. Richborough IV, pi. LIX, No. 318. Three-legged iron candle holder of common Roman type, cf. Lydney Park Report, Fig. 23, No. 191. T-shaped iron slide-key with looped handle, of. Wheeler, London in Roman Times, PI. XX A, No. 2. Position No. II. Rubble layer filling — of north wing. II. As for Object No. 1. 6.2 II. As for Object No. 1. 6.1 129 SPRINGHEAD : TEMPLES II AND V Object No. Position II. Over Steps. Figure No. V. Below plaster rubble inside building (Stratum D/H). V. As Object No. 5. Description Part of an iron netting needle. Similar examples were found at Wroxeter (Wroxeter, I I (1913), Fig. 5, No. 9, p. 13) and at the Saalburg. (Jacobi, Saalburg, p. 454, Fig. 71, No. 12.) A bronze example was found at Richborough (Richborough I, Fig. 22, PI. XIV, p. 46.) Part of split iron staple on an iron ring. Thin open ring of silver. Possibly an ear-ring. Object No. Position TABLE 8 BONE OBJECTS (Reported on by E. W. Tilley, Esq.) Figure No. 1. II. Rubble layer filling 6.3 Bone pin with ornamental head. of north wing. 2. II. Plough level, west 6.5 Bone pin with ornamental head. of temple. 3. V. Over building floor 6.4 Bone gaming counter. See Table 10.3. level, outside building (Stratum F). TABLE 9 THE GLASS (Reported on by Miss D. Charlesworth, M.A.) Object No. Position Figure No. 1. II. Rubble layer filling 6.7 of north wing. 2. II. As Object No. 1. 6.6 3. II. Under plough level, — east of north wing. 4. II. As Object No. 1. — II. Filling under cellar floor (Stratum E). Description Fragment of good colourless glass inscribed with the letter C. Probably from a cylindrical bowl. Six very small, oblate, blue-green glass beads. Oblate bead, of opaque green glass. Fragments of a bowl in colourless glass, with a ground rim, constriction below and bulbous sides, decorated with cut lines and circles. This style of decoration is quite common on round based bowls and globular bodied flasks of 3rdcentury date. Diameter of rim c. 0 • 10 m. Associated with rubble in rooffall amongst which were several late 3rd-century coins. Oblate bead of opaque greenish glass. 130 SPRINGHEAD : TEMPLES II AND V Object No. 5. Position V. Under plaster rubble inside building (Stratum D/H). Figure No. Description Several small beads and fragments of beads were found. All were of pale green glass : Cylindrical, 0 • 8 in. long. Cf. Verulamium, Fig. 47, No. 67C. Double or segmental. Cf. Verulamium, Fig. 47, No. 67p Triple or segmental. Minute and annular. Cf. Verulamium, Fig. 47, No. 67K. TABLE 10 GRAFFITI (Reported on by R. P. Wright, Esq., M.A., F.S.A.) Object No. Position No. 1. II. Rubble layer filling 6.11 of north wing. 2. II. Plough level, east of 6.9 north wing. 3. V. Over building floor 6.4 level, outside building and (Stratum F). 6.10 Description Sherd of poppy-head beaker with graffiti. Graffito reads " P I I I I " and gives the tare p(ondo) IIII " four pounds ". Samian sherd, Drag 33. Graffito TA. (See Arch. Cant., LXXI (1967), p. 86, Fig. 17.6 for similar graffito. Bone gaming counter. The bone counter reads " MR ". I grant that there is a vertical stroke through the second apex of the " M ", but I do not count this as the extra stroke to form an " A " tied to the " M " as the stroke carries on in a straight line upwards to the edge of the counter. So I read " MR " and, with regret, not the more attractive " MAR ". Object No. 1. 2. 3. 4. TABLE 11 MISCELLANEOUS OBJECTS Position No. Description I I . Rubble layer filling 6.8 Small double-edged flint blade (broken), of north wing. I I . In space between — Short leather thong tied into a knot, temples I and I I . Under pebble path (Stratum B/C). V. Stratum H. 4.12 V. Stratum H. 4.13 Plaster with line designs. Plaster with black fleur de lys on background. light TABLE 12 THE " HOARD" (From tile filling, north wing, Temple II) Object No. Description Reference 1. 23 coins. Latest date 337-40, including rare Table 3,12,3,34 coin Carausius 2. Small bronze stud Table 6,1 3. Small bronze ring Table 6,3 4. Small bronze ring Table 6,4 131 SPRINGHEAD : TEMPLES II AND Object No. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Description Small bronze ring Part of bronze bracelet Bronze plate with hole Small uninscribed bronze plaque with hole Bronze ligula Bronze lion's head Iron door hinge Three-legged candle holder (iron) T-shaped iron slide-key Bone pin Glass fragment with letter C Fragment of glass bowl Six tiny glass beads Broken flint blade Sherd of poppy head beaker with graffito Many sherds of Standard Springhead forms CONCLUSIONS Reference Table 6,13 Table 6,10 Table 6,12 Table 6,15 Table 6,14 Table 6,11 Table 7,1 Table 7,2 Table 7,3 Table 8,1 Table 9,1 Table 9,4 Table 9,2 Table 11,1 Table 10,1 I t has already been indicated that the present total of five temples indicates that the complex at Springhead was perhaps the most important in Britain. It is certainly the only one which reflects the finds of votive offerings found on the Continent. These votive offerings are of great interest and diversity. They include figurines, axes (flint and lead), beads, bracelets, bronze statuettes (represented by a lead base and bronze pin), pottery, coins, faceurns, rings, bronze plaques, bronze castings of parts of the human body, possibly bone pins, and bronze letters. Springhead is thus a strong reflection of the cults which were so common on the Continent. REFERENCES (1) W. S. PENN, Temple I, Arch. Cant. LXXIII, 1959. (2) W. S. PENN, Temples I I I & V, Arch. Cant. LXXIV, 1960. (3) R. A. G. CARSON, The Numismatic Chronicle, p. 10, XIX, 1959. (4) R. E. M. WHEELER, The Antiquaries Journal VIII, p. 300,1928. (5) Archaeologia, 1 ii, 745. (6) R. E. M. & T. V. WHEELER, Society of Antiquaries, Lydney Park. (7) C. D. DREW, Dorset Nat. History & Archaeological Soc. LIV, 1932, page 25. (8) Archaeologia, 1 xi, p. 207 & 215. (9) P. RAHTZ, Proc. ofSom. Arch. & Nat. History Society. XCVI, 1951, p. 112. (10) Archaeological,!. VI, p. 114, 1849. (11) LEON DE VESLEY, Les Fana ou Petit Temples Gallo-Romains de la Region Normande, 1909. (12) J. P. BUSHE-FOX, Richborough III. 132