pestooseen ts Spasatesteteetsseess esescsssrseiiseee: Siispesscnsssansinaesssie: mel casSunsssicere sees sro Spaces isaseaae oitbetesey hey , : - an ake ; au " : 1 y - ; ; = Rta y : : f ‘ : * : i a : ~ + ‘ ; ' a 7 : i t 2 7 1 om ~ ; 7 ) ee , : : 7 : Pi . at 7 7 . i a H ‘ 4 ’ 4 7 = an nt x th ‘ ‘ ’ ics : 1 "SOs . , : : [ 7 ; = F - A 7 4 1 am fj : ; 7 1 ‘ 7 ' : , ; : ; * UJ + ; : - ‘ - 7 ‘ + ; ; ; +, ; ee . i 7 : ie 4 ; ; ot ge \ ‘ " t oe * : et gt 8, | 7 if rm : : : 7 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE Volume 59 1964 PRINTED IN GREAT BRITAIN BY HEADLEY BROTHERS LTD I0Q KINGSWAY LONDON WC2 AND ASHFORD KENT THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE VOLUME 59 1964 GON TENTS THe NEoLITHIc CausEWAYED Camp at Rosin Hoop’s Batt, SHREWTON, by Nicholas Thomas - - - : - - = P 2 : : ExcAvATIONS AT AVEBURY, 1960, by Stuart Piggott - - - - - A BronzE AGE Rounp BaRROw ON EARL’s Farm Down, AMEsBurY, 0) Patricia M. Christie - 2 = 2 x a a 2 - Z a = Cross-DykEs ON THE EBBLE-NADDER RuipGeE, by P. J. Fowler - - - - A CATALOGUE OF PREHISTORIC PLANT REMAINS IN WILTSHIRE, by J. D. Grose and R. E. Sandell . - - - - = 2 ‘ é é ‘ EXCAVATION OF THREE RoMAN TomBs AND A PREHISTORIC Pir ON OVERTON Down, by I. F. Smith and D. D. A. Simpson - - - - - - - A Saxon CEMETERY AT WINTERBOURNE GUNNER, NEAR SALISBURY, by John Musty and J. E. D. Stratton - - - - - = ‘ : THE Saxon BOUNDARIES OF FRUSTFIELD, by C. C. Taylor - - - - A Pre-ConQuEst CHURCH AND BaPpTisTERY AT POTTERNE, by Norman Davey - SAXON AND MeEpIEvVAL FEatTurEs at Downton, SALIsBury, 6) Philip A. Rahtz THE SuBuRBs OF OLD Sarum, by John Musty and Philip A. Rahtz - - ‘BALLE’S PLAcE’, SALISBURY: A I4TH-CENTURY MeErcHANT’s House, by Helen Bonney - - . - - - - - - - - - - Two WiLTsHIRE Fonts IN SouTH AUSTRALIA, by H. de S. Shortt = - - - THE DECLINE oF A Recusant FAMILY: THE KNIpEs OF SEMLEY, by J. Anthony Williams - - : f : 2 z : : , i NoTEs: A Stone Axe Fragment from Hamshill Ditches, Barford St. Martin; Avebury: the Northern Inner Circle; Barrows near Clatford; Barrows in Savernake Forest; Siting of Barrows round Springs; Romano-British Interments at Parsonage Farm, Winsley; Benson’s Folly, a Forgotten South Wiltshire Field Monument; An Elusive White Horse on Rockley Down; Cropmarks near Crofton — - - - - - - - - ili PAGE I 28 30 46 58 68 80 110 181 EXCAVATION AND FIELDWORK IN WILTSHIRE, 1963 - - - - - - 185 OBITUARIES - - - - - : 2 2 : : s SC eTot REVIEWS - - - - - 2 : S Z 2 : a i Eh 1Q2 ACCESSIONS TO THE LIBRARY, 1963 - - - - . : : : =r HOA) ANNUAL GENERAL MEETING, 1963 - - - “ = 2 = ‘ ST eOS ANNUAL GENERAL MEETING, 1964 - - - : = = E = - 199 Orricers’ REPORTS READ AT THE ANNUAL GENERAL MEETING, 1964: Report of the Secretary - - - - - - - = 200 Report of the Curator - - - - - - -is - 2, 202 NATURAL HISTORY SECTION Roap CASUALTIES AMONG MAMMALS, REPTILES AND AMPHIBIANS IN WILTSHIRE, by A. A. Dunthorn and F. P. Errington - - - - - ok 207 THe WEATHER OF 1963, by R. A. U. Jennings - - - - - 2 OTT WILTsHIRE Birp NoTeEs FoR 1963 - - - - - - - - hil ato3h >) WILTsHIRE PLant Notes (24), 6y Donald Grose - - - - - =) 226 ENTOMOLOGICAL REPORT FOR 1963, by B. W. Weddell — - - - - SHG 2 REPORT OF THE HON. SECRETARY FOR 1962 - - - - - - Bhs 22 irs REPORT OF THE HON. SECRETARY FOR 1963 - - - - - - = 1) 226 REPORT OF THE Hon. MEETINGS SECRETARY FOR 1963 - - - - Sez or ACCOUNTS ACCOUNTS OF THE SOCIETY FOR 1962 - - - - - - - =7 | 289 ACCOUNTS OF THE SOCIETY FOR 1963 - - - - - - - = 2 242 NOTES FOR THE GUIDANCE OF CONTRIBUTORS TO THE MAGAZINE - - Si AOA INDEX - - - - - - - - - - - - - Sei Aue List oF ILLUSTRATIONS — - - - - - - - - - - =e 25 PLATES - - - - - - - - - - - at end of volume The Wiltshire Archaeological and Natural History Magazine Volume 59 1964 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY SOCIETY The Society was founded in 1853. Its activities include the promotion of archaeological and historical work within the County, and of the study of all branches of Natural History; the issue of a Magazine and other publications; excursions to places of archaeological and historical interest; collaboration with a Records Branch; and the maintenance of a Museum and Library. The subscription rate for membership of the Society is at present as follows: Annual Subscription (minimum), £1 12s. 6d.; Junior Subscription, 1os.; Life Membership, £25. Enquiries about membership should be made to the Secretary of the Society, 41 Long Street, Devizes. OFFICERS OF THE SOCIETY, 1963-64 PATRON: The Lady Colum Crichton-Stuart PRESIDENT: E. G. H. Kempson, Esq., M.a. TRUSTEES: E. QC. Barnes, Esq. The Lord Devlin Sir Michael Peto, Bt. Bonar Sykes, Esq. VICE-PRESIDENT: R: B, Pugh, Esq., M.A., F.S.A. CURATOR OF THE MUSEUM: F. K. Annable, Esq., B.A., F.S.A., A.M.A. HONORARY LIBRARIAN: R. E.. Sandell, Esq., M.A., F.S.A., F.L.S8. HONORARY MEETINGS SECRETARY: J. B. B. Roberts, Esq., m.a. SECRETARY AND TREASURER: The Rev. E. H. Steele, m.a. COMMITTEE: The Most Hon. the Marquess of Ailesbury, D.L. representing the Group Captain F. A. Willan, c.B.£., D.F.c. Wiltshire County Counct! M. G. Rathbone, Esq., A.L.A. (ex officio as County Archivist) M. Fitzgerald, Esq. J. W. G. Musty, Esq., F.s.a. C. Floyd, Esq., 0.B.E., B.A. H. Ross, Esq., B.A. J. S. Judd, Esq., T.p. H. de S. Shortt, Esq., M.A., F.S.A., F.R.N.S. Group Captain G. M. Knocker Miss I. F. Smith, B.A., PH.D., F.S.A. Miss J. de L. Mann, M.aA., F.R.HIST.S. Miss W. M. Stevenson, B.sc. ASSISTANT CURATOR: Mrs. G. P. Mitchell, B.A. HONORARY ARCHITECT: D. A. S. Webster, Esq., F.R.1.B.A. HONORARY SOLICITOR: H. G. Awdry, Esq. EDITORIAL BOARD: Miss I. F. Smith, Mr. F. K. Annable, and the Rev. E. H. Steele (Secretary) THE NEOLITHIG CAUSEWAYED CAMP AT ROBIN HOOD’S BALL, SHREWTON by NICHOLAS THOMAS INTRODUCTION DURING THE COURSE of a series of lectures given by the writer at Salisbury Museum in 1955-56, it was suggested that the digging of trenches across the earthwork at Robin Hood’s Ball would provide a useful exercise as well as estab- lishing its date and character. Through the enthusiasm of the class secretary, J. W. G. Musty, F.S.A.,' a digging party was provided for two week-ends in August, 1956. A preliminary ground survey was also arranged by Major Sweet of the School of Artillery, Larkhill, a member of the class. We would like to express our gratitude to Col. H. R. L. Hodges, O.C. School of Artillery, Larkhill, and his colleagues for permission to excavate, and for their help and general support. The majority of the finds has been deposited at the Salisbury, South Wilts. and Blackmore Museum. A small type-series of potsherds is at the Museum of the Wiltshire Archaeological Society, Devizes. Some of the animal bones were left at the Institute of Archaeology, University of London, after they had been identified by Miss M. M. Howard. GEOGRAPHICAL SETTING Robin Hood’s Ball lies 24 miles north-east of Shrewton and is crossed by the Shrewton-Figheldean parish boundary. On the r-in. O.S. map it is at SU (41) 102460, while a diagrammatic plan is shown on Sheet LIV N.W. of the 6-in. map. This ground, including the clump of trees from which the camp takes its name, falls within an artillery range which makes access to the site difficult. Built upon chalk at a height of 430-465 ft. above sea-level, the earthwork has clearly been designed to command country to the south. In this direction it overlooks, without truly dominating, a very broad sweep of rolling downland which includes Stone- henge and its cursus. Eastwards Woodhenge, where Windmill Hill pottery has been found in some quantity beneath the bank,? is only just out of sight behind a ridge which descends to the River Avon. As a strategic earthwork Robin Hood’s Ball is weak, because there is an almost equally broad area of ground to the north which is hidden behind the slight rise on whose southern slope the site is set.3 The nearest water supply today is the River Till which lies 2} miles to the south-west. The geographical setting of Robin Hood’s Ball is shown in Fic. 1, where it is seen to occur centrally, amid a scatter of causewayed camps extending along the edge of the chalk from Dorset to the Chilterns, with an outlier in Devon and two I 50 100 150 KMS. # CAUSEWAYED CAMPS AMES ee rome RES CHALK LAND UNSTIPPLED LONG BARROWS, EARTHEN BRISTOL LONG BARROWS, CHAMBERED SWINDON CAUSEWAYED CAMPS SALISBURY ) 2 = LONG BARROW \ 2 \ A TusHEAD ROBIN HOOD’S a BALL , Ge eo @ (As BS. e ——__» LESSER | ne o> = p = a AMESBURY p MORTUARY —~ ¢ ENCLOSURE® Wie Se chee Oe ee ie ee EO 7 Se iKOME RES et ae: Fic. 1 Locality map. Based upon the Ordnance Survey map, with the sanction of the Controller of H.M. Stationery Office. Crown Copyright reserved others on the Thames gravels. Within the area of Salisbury Plain, the camp is situated in a region where Neolithic sites abound. Between the Rivers Avon and Till this concentration is noteworthy: thirteen long barrows, a mortuary enclosure and two cursuses are known to exist between the camp and Lake Down, while Neolithic occupation of some sort beneath Woodhenge has already been noted. PREVIOUS HISTORY Sir Richard Colt Hoare made a characteristic record of the existence of this causewayed camp thus:4 ‘The first object of our attention, near a clump of trees called Robin Hood Ball, is one of those ancient circles, which I have before mentioned and described in the Heytesbury Station, p. 80. This, like the generality of them, is placed on an elevated and commanding situation, but has this peculiarity, of having one circle within the other, with an entrance towards the north. We have to regret the great injury these circles have sustained by the plough, as in their original state they must have been highly curious, and are the more remarkable, from representing a double circle.’ The site was described again by E. C. Curwen® in 1930, when a rough sketch- plan was included. Meanwhile, the late Alexander Keiller had unearthed potsherds of Windmill Hill ware from the earthwork close to the 19th-century rifle-butts which overlie its south-west quarter. In 1954 the true shape and character of the earthwork was established by 8. Piggott.7 The aerial photograph which he published is reproduced here (pL. I) by his kind permission and by courtesy of the Cambridge University Press, to whom thanks are due for the loan of the original half-tone plate. Among the long barrows nearest to Robin Hood’s Ball, one was excavated by Dr. Thurnam in the last century.’ The primary burial comprised a single inhuma- tion, with a pit to the south-east. A Long-Necked Beaker occurred in a secondary context. THE CAUSEWAYED CAMP Robin Hood’s Ball is an earthwork of irregular shape, having two banks and ditches which enclose 74 acres of grassy downland (Fic. 2). The inner bank and ditch surround an area of 24 acres which is approximately oval in plan. Its long axis runs north-east to south-west, giving it a maximum diameter of 450 ft. The inner enclosure is much more regular in shape than the other, and has been more nearly levelled. Its bank is hardly anywhere visible and even at the north-west Trench 1 failed to reveal any trace of its presence despite indications of a slight rise in the ground. The outer ditch and bank are separated from the inner by a space about 100 ft. wide. In plan this earthwork is very irregular and seems to have been designed in a series of straight sections. Around the south, east and north-east it follows the line of the inner earthwork fairly closely. On the north-west, however, it bulges first inwards and then out, while on the western side, immediately north of the rifle-butts, the aerial photograph and ground survey suggest an unusually broad 3 ROBIN HOOD'S BALL PIG Bib AN . ww » 5 “7 Z \ . Uy \\ Meyryy 0 SHREW TON CALE. FOR. SECTIONS : METRES SCALE FOR SECTIONS . FEET (0G. "0, IEEE TF ele) O SCALES FOR. PLAN Fic. 2 Plan of the site. Surveyed by R.C.H.M. (England), Salisbury. Crown Copyright reserved, (Drawn and lettered by N. Thomas) causeway, with a section of narrow ditch swinging outwards around it and the main rampart running behind, as if ignoring it. This enclosure has a maximum diameter of 750 ft. On the north and west sides the ditch is still fairly prominent, with a surface width of 30-40 ft. It is interrupted in this section by at least six causeways, one of which was verified by our Trench 4. These causeways vary in width from 12 to 25 ft. Along the north side the bank is also visible, having a maximum height of 18 in. It does not appear to be interrupted by gaps corresponding with those across the ditch. Where the Shrewton-Figheldean boundary crosses the site a small mound (diameter c. 35 ft.) overlies the bank of the outer enclosure. There is a disturbance at its centre, but no record of discoveries survives.9 The material for the mound has been quarried from the outer enclosure ditch, which is deeper and broader at this point. THE EXCAVATIONS, 1956 THE INNER EARTHWORK; TRENCHES I AND 3 (FIG. 3) Trench 1 (6 ft. wide, 404 ft. long) was designed to expose the cross-section of a typical length of the inner ditch and the site of its bank. Trench 3 was opened to examine the causeway at the northern end of this length of ditch. At chalk surface the ditch was found to be about 13 ft. wide, with a depth of 74 ft. The undisturbed chalk was firm, and stratified in thin layers which tilted down very slightly to the south. The walls of the ditch were almost straight and each ended at a ledge nearly 12 in. above the floor. At this point the ditch was 54 ft. wide. From here the ditch narrowed sharply to a final width of 3 ft. Its floor sloped southwards, following the natural tilt of the bedded chalk. The following layers occurred in the silted-up ditch: A. Modern turf and humus, a scatter of chalk crumbs and some flints at its base. Thicker over the site of the bank than across the ditch. Included one Romano- British potsherd. B. Chocolate-coloured humic soil crossing the ditch. Contained an irregular scatter of chalk and flints; at its base a scatter (buried by earthworms) of Neolithic, Romano-British and Medieval sherds. C. Dense spread of loose chalk chips in a matrix of chocolate-coloured soil, entering from the outer side and not extending across trench. One Romano- British and two Neolithic potsherds occurred in this material. D. Compact, fine red-brown loam, almost free of chalk and flints. Spreading across, and deepening, from the outer side but not sealing ditch. Burnt flint and animal bones included. E. Fine, red-brown loam, paler and less compact than D and including chalk chips. Derived from inner side of ditch. F. Compact, fine loam, greyer than D and charged with chalk chips larger in size than those from G below it. Included substantial spread of charcoal, but only one find of Neolithic pottery. Derived from outer side of ditch. G. Grey-black soil with some chalk chips. High concentration of Neolithic pottery; flint flakes and serrated blade with charcoal at lowest point of layer. VvAdIGaW 2 HSILNG-ONVWOU 2 DIHLITOAN @ qWosAVHD * ‘SGNI4 *"suUOTIIES :][eG S,POOP{ UIqoYy 6 ol TIVG SAGOOH NIGOU H. Very fine, compact chalky rainwash from inner side of ditch; interrupted by recent animal burrow (J). Some isolated potsherds. Large chalk rubble blocks, concreted. Small chalk chips in pale yellow-brown soil matrix. Larger, looser rubble in lens running through middle of layer. Derived from inner side of ditch. Some Neolithic potsherds, most below the loose rubbly lens. M. Primary ditch filling, no soil or chalk matrix; loose chalk blocks with air spaces, liable to run if touched. Largest blocks, fist-size, in channel between ledges at bottom of ditch. Heavy concentration of potsherds and flint tools, particularly in upper part. Secondary concentration of potsherds at level of inner ledge on floor of ditch. N. Site of bank. Slight increase in quantity of chalk chips below modern humus, and general increase in thickness of modern humus. Underlying natural chalk surface level from inner edge of layer B to point 23 ft. from inner end of trench, where chalk surface drops slightly. bres Trench 3 was a cutting 6 ft. square laid out to examine the causeway which apparently interrupted the northern end of the ditch sectioned in Trench 1. Undisturbed chalk was encountered everywhere below the modern turf and humus. The latter thickened towards the south side of the cutting, where the butt-end of the ditch began to appear. No post-holes or other features were found upon this causeway. OUTER EARTHWORK; TRENCHES 2 AND 4 (FIG. 3) Trench 2 (6 ft. wide, 66 ft. long) was designed to examine a representative length of the outer ditch. Trench 4 was dug to confirm the presence of a causeway of undisturbed chalk at the west end of this segment of ditch. Trench 2 was extended inwards so as to expose a section across the outer bank which here stands to a height of nearly 2 ft. The ditch at chalk level was about 15 {t. wide and 4 ft. deep. Its profile was a flattened U, the outer wall sloping upwards more sharply than the inner side. Today, a gently sloping berm nearly 7 ft. wide separates the inner edge of the ditch from the bank (taken to begin where layer L replaces layer C, Fic. 3). The following layers were noted in the filling of the ditch: A. Modern turf and humus, noticeably thicker than that covering the inner ditch. Fine lenses of chalk crumbs and flints along base. Contained Romano- British and one piece of Neolithic pottery. B. Loam, paler brown than modern humus above, with more general scatter of chalk. On outer side of ditch it merged gradually with more chalky soil inter- preted as weathered bedrock. Beyond Neolithic ditch at northern end of trench, bedrock rises to a peak, with 6-8 in. of weathered chalk beyond it. In the ditch a general scatter of Neolithic pottery, with some Romano-British sherds. Finely broken weathered chalk, with flints, in loose grey-brown soil matrix. Chalk content gradually decreased as it entered ditch to merge with B. Began again at south end of trench, at tail of Neolithic rampart. Pale buff chalky rainwashed soil; compact, with almost no rubble. At lowest point two isolated patches of dark brown soil (indicated in section, FIG. 3). Pale brown loam, paler than C, darker than D; more compact than C but less compact than D; derived from outer side of ditch. Lower end almost reached floor of ditch, and ran under end of D, F and H. Neolithic potsherd at lowest point. Very compact, hard chalky loam entering ditch from inner side. Upper surface almost concreted but became looser with increased depth. At end of deposit, where it joined E, much loose heavy chalk rubble. Fine chalky loam, resembles colour of E, but has texture closer to J. Grades evenly into both. Loose grey-brown soil speckled with much fine chalk: more earthy than G but not so loose. Primary silt, chalk rubble. On bank side of ditch chalk is finer, looser and less mixed with loam than that on other side of ditch. More compact as it approached bottom of ditch. Chalk on outer side more earthy: looser and more rubbly towards ditch-centre. Scatter of flint flakes and a Neolithic potsherd on floor of ditch, outer side. Neolithic sherds and flint flake in J on bank side also. The following layers were uncovered during excavation of the bank behind the outer ditch: K. L. Very diffuse small chalk rubble mixed with grey-brown loam, top of bank make-up. Thinned out to north and merges with L. Brown loamy soil, mixed weathered chalk and spread of K. At north end merged with weathered chalk of C. High concentration of Neolithic potsherds at base, with charcoal. Neolithic land surface not preserved beneath L. Brown soil, thinly speckled with small chalk. Absence of finds except at base, on top of Neolithic land surface. Bank make-up behind, and sealed at north end by M. Compact small chalk rubble with some grey-brown loam. Rubble more concentrated and compact than K. Much charcoal with potsherds to south of post-hole, at base of N. Interrupted by ?badger hole extending down from modern humus (Q and R). Died out beyond latter, merging into weathered natural chalk with concen- tration of Neolithic potsherds here. Neolithic land surface. Dark brown, almost black soil, free of flint and chalk, but with a concentration of the latter at its base; lying directly on undisturbed chalk. Much charcoal on top of, and in P, together with potsherds. Became less black south of post-hole and was distorted, then broken by entrance to ?badger hole. Post-hole. Found at base of old land surface (N); not noted whether it extended through N to bank material above, but probably did not. Diameter at chalk surface, 6 in.; 7 in. deep, with gently tapering profile. Fill similar to N, with no packing. 8 QO and R. Fill of ?badger hole. Clean brown loam, becoming progressively blacker with increased depth. Trench 4 was a cutting 6 ft. square to examine the causeway of uncut chalk at the west end of this section of the ditch. Bedrock was found beneath the turf, which established that the causeway was genuine. No artificial features were found upon. it. DISCUSSION? The inner ditch was deeper, and its sides originally steeper than those of the outer ditch, whose profile was altogether more gentle. Layer M in the former was so loose and free from the remains of turf and humus that it must represent the sudden and complete collapse of the upper part of the inner ditch wall, together with the forward section of the chalk bank behind it. Layers H and L represent the return of the bank at a much slower rate, where wind and rain were the chief agencies involved. The deposition of layer M could have occurred as a result of a few sharp frosts, or because of a structural weakness in the rampart, or as a deliberate slighting of the earthwork. Layers K and L would seem to have accumu- lated over a longer period of time. The outer edge of the ditch, represented by layer K, did not collapse, however, until this primary silting had formed. The secondary silts were mainly of earth, in contrast with the chalk layers below them, and except for E they all came from the outer side of the ditch. Layer G contained much occupation rubbish, while a considerable scatter of charcoal unaccompanied by artifacts occurred in layer F above. The noticeably low phosphate figures for these layers (p. 24) prevent them from being considered as an occupation horizon. Finally, a considerable spread of chalky soil entered the inner ditch, again from the outside, accompanied by a thin scatter of Neolithic potsherds. Romano-British activity was represented by a handful of small potsherds in the modern humus. The filling of the outer ditch presented the same picture—a heavy collapse of chalk from the upper walls of the ditch and the front of the bank followed by a thick, more gradually deposited layer of very compact chalk which must have been derived also from the bank. The secondary silts, like those in the inner ditch, were much more earthy and had entered an already half-filled ditch mainly from the outside. Of these, layer E had started to accumulate early, but had been inter- rupted by the collapse of the front of the bank (layers D, F and H). Here, too, no Romano-British pottery was dropped until the filling of the ditch had been completed. It was noticeable that the outer walls of both ditches sloped to the surface more gradually than the inner, and it seems possible that when the camp was constructed some surface scraping took place along the outer edge of each ditch to augment the spoil already dug out for the banks. This has been recorded at Wind- mill Hill and at Whitesheet (Wilts.) and is hinted at in some of the published sections of Maiden Castle (Dorset) ;1! it may account for our original calculation of an outer ditch 30-40 ft. wide when Robin Hood’s Ball was surveyed (p. 5). The excavators of Windmill Hill and Whitesheet were similarly misled. In Trench 2 a good cross-section of the bank was obtained. Layers K and L (FIG. 3) suggest that its front was of chalk, most of which now lies in the ditch as layers D, F and H. A less easily explained feature was the heavy dump of chalk (layer N) at the back of the bank, whose face seems to have been revetted with turf and humus (layer M). The front of this rearward chalk deposit coincided with the post-hole, though since this was not noticed until the undisturbed chalk was reached, too much cannot be made of it. Although layer N might represent an original bank re-faced at a later stage by layers M and K, there was no evidence for the enlarging or re-cutting of an early ditch to provide this extra material. And a bank excluding layers M and K would have involved an unusually broad berm between its front and the inner edge of the ditch. At present we must regard all these bank layers as of one period: that they seal a land surface of even thickness shows that even if they do represent more than one stage of construction, the whole earthwork cannot have taken long to build. The discovery of a Neolithic rampart interleaved with turf and soil is not new. The latest work at Windmill Hill'? shows that the outer bank incorporated the same feature. Here, however, the earthy layers slope from front downwards to rear, unlike layer M at Robin Hood’s Ball. The old land surface beneath the outer rampart at Robin Hood’s Ball included a stone- and chalk-free zone which establishes that it had not been disturbed deeply by agricultural activity within at least a generation of the construction of the camp. As at Windmill Hill, however, it was covered with occupation rubbish which also occurred all through it. This included potsherds, animal bones and an even scatter of charcoal. Pending further excavation it seems likely that the post-hole belongs to this occupation, and suggests that there had been considerable activity in the area before the causewayed camp was built. Dr. Cornwall has shown (p. 25) that this buried soil was of rendsina character, representing a cool, moist Atlantic climate like that of today. Comparison of the ditch silts with those at Windmill Hill fails to find close agreement. The frequent lenses and spreads of loamy material in the rubble of the primary silts of the latter, entering from both sides of the ditch, show that turf and topsoil had been incorporated in the front of the main rampart, and that in quarrying their ditch the Neolithic diggers had left humic material to overhang the edges. At Windmill Hill, Dr. Smith justifiably regarded the primary chalk and soil silts as the product of a collapsing rampart.%3 In contrast, the absence of soils from the lower filling in the ditches of Robin Hood’s Ball supports the suggestion (p. 9) that layers J and M represent the collapse of the upper edges of the ditches through frost and other natural causes, and at the same time a fairly rapid slumping of the front of ramparts which were made of unrevetted chalk rubble. The crumbling of the ditch walls could have caused sudden collapses of the banks if the latter had been set closely behind their ditches. The total absence of soil in the lowest part of the inner ditch is still curious. Observation of the silting process in the experimental earthwork ditch on Overton Down (Wilts.), however, is showing that earthworms will rapidly eat up humus trickling into a ditch, to the extent that their castings are pure white." IO Occupation refuse in the inner ditch was deposited in two stages. The first group of potsherds and animal bones occurred scattered near the top of layer M and may have been derived from a surface occupation layer disturbed as the ditch- edge crumbled. The second group of sherds occurred in and on the top of layer G and resembled more closely an actual occupation of the ditch, since it included much charcoal and decayed organic material. From the lower group the animal remains included two bones with joints still articulated. That it was of very short duration, however, has been shown by soil analysis (p. 24), and there was no sign of a trampled layer in the fill. In Trench 2 the finds suggested an occupation at the back of the outer rampart and also an intensive pre-earthwork phase. Contrast in the quantity and disposition of finds from Trenches 1 and 2 was striking. 4 lb. 3 oz. of potsherds came from Trench 1, 2 lb. 6% oz. from Trench 2. Of this material, everything from Trench 1 occurred in the ditch, whereas only 1? oz. occurred in the outer ditch in Trench 2. No sherds or other occupation refuse were found on the site of the inner bank: 2 lb. 5 oz. of pottery was collected from under and at the back of the outer rampart. ‘This trial excavation, if typical of the site, suggests that at Robin Hood’s Ball the human occupation contemporary with the camp was concen- trated between the inner ditch and the outer rampart, while a detailed study of the potsherds suggests that about 30 different vessels are represented by these finds.*® The absence of later Neolithic wares from the four trenches cut in 1956 was complete. Likewise no products of the Middle and Late Neolithic axe-factories were found, and when Romano-British farmers appeared on this part of the downs, both ditches had become completely filled. Roman ploughing played no part in the levelling of these Neolithic earthworks. The results of such a small-scale excavation cannot be expected to throw much new light on the purpose for which the camp was built nor do they explain how the ditches came to be filled up. That this process was rapid is suggested by the presence of two long-bones in articulation low down in the inner ditch. Scavenging animals would surely have disturbed these unless they had been buried very soon after being dropped. The chief agency in the levelling of Robin Hood’s Ball, like that of all the other known causewayed camps, must have been something more than just extreme antiquity, for long barrow ditches and cursus earthworks often survive more distinctly, although of similar age. The causewayed camp ditch at Maiden Castle was nearly levelled, and its bank almost flattened, by the time the Long Mound was built across it.16 Like most causewayed camps, Robin Hood’s Ball yielded a very great deal of broken pottery and domestic rubbish. The yield of potsherds from the ditches of an Iron Age hill-fort in the south of England, even one permanently occupied, would have been much less. Judging by the amount of pots which were broken and meat consumed, occupation in a Neolithic causewayed camp was intensive, and beyond the rate suggested by the remains from Iron Age hill-forts. It should not be forgotten, however, that much Iron Age rubbish, which might otherwise have been tipped into the hill-fort ditch at least in times of peace, was carefully preserved II instead for spreading on fields as manure. Moreover, hill-fort ditches may have been scoured out in times of war. As at Windmill Hill, there was strong evidence for lengthy occupation of the area before these earthworks at Robin Hood’s Ball were built. The presence of one post-hole hints at the existence of wooden structures prior to the building of the camp: and surely the curious bend in the line of the outer earthwork on its western side suggests the presence of a feature more substantial than the trunk of a tree. At least one grain impression in a potsherd (p. 19) is further evidence for the pre-existence of Neolithic farming in the region, and thus of sizeable areas of cleared ground available for those whose task was the construction of a cause- wayed camp. So, too, the unusually high proportion of sheep/goat bones from the site must imply that much cleared land was available for grazing. The presence of an undisturbed soil profile beneath the outer rampart, however, established that the site of Robin Hood’s Ball had not been dug over for cultivation immediately prior to the building of these earthworks. One more observation on the purpose of causewayed camps must be made. The writer believes that among the thirteen to seventeen sites of this type so far identified in southern England!7 more than one function impelled their construction. At the Prehistoric Society Conference of 1962, to which allusion has already been made, support was given to the thesis that these camps may have been used as fair grounds and occasional gathering places for large numbers of people. While this may be true of Windmill Hill, and may even fit in with the unusual density of occupation refuse at Robin Hood’s Ball, the writer feels that it does not suit the peculiarly isolated situation of Knap Hill (Wilts.) nor a camp like Hambledon (Dorset) which has outworks. It may be no coincidence that some of these places, situated so as to command country and make use of natural protection like steep hill-slopes, were selected in the Iron Age as sites of hill-forts, for example, Maiden Castle (Dorset) and Maiden Bower (Beds.). Although all causewayed camps were constructed using a common digging technique, this ought not to rule out the possibility that they served a variety of purposes which may have included defence of settled communities, even if the presence of causeways across the ditches raises a problem of its own. A little light was thrown upon the culture to which the builders of Robin Hood’s Ball belonged. The style of pottery agrees in general with the material from Maiden Castle (Dorset), from the pre-earthwork phase at Windmill Hill and from the pit at Waden Hill, south of Avebury (Wilts.).18 The mineral inclusions other than flint in the clay of the potsherds from Robin Hood’s Ball (50 out of 230 sherds from the excavation) show that a high proportion of the vessels had been made in the Jurassic zone of southern England. Moreover, three fragments of unusually well finished vessels included grits obtained from Dartmoor granite or even from as far afield as Brittany. Pots made from clay charged with weathered oolite or with crushed shell from Jurassic deposits have been found in the causewayed camps at Maiden Castle (Dorset), Whitesheet Hill, above Mere (Wilts.)!9 and at Windmill Hill. The presence at Robin Hood’s Ball of vessels made of clay perhaps from Dart- moor, a distinctive ware characterized also by thinness, a high surface burnish 12 and general excellence of workmanship, emphasizes also close contacts between Salisbury Plain, Dorset, and the south-west peninsula, for vessels of this ware have been found at Windmill Hill (Wilts.), Maiden Castle (Dorset) and Hembury (Devon), where radio-carbon dates of 3480-3000 B.c. have recently been obtained.?° The diversity of elements in the earliest Neolithic of southern Britain, already well known, has thus been stressed by the evidence from Robin Hood’s Ball. The relative date of Robin Hood’s Ball within the chronology of British cause- wayed camps was not established by these excavations, for it is now clear that pottery styles changed not at all during more than the first thousand years of Neolithic occupation in the south of England. The evidence for much pre-camp occupation in the vicinity would suggest that in this region, at least, Robin Hood’s Ball was not one of the first pieces of construction carried out by these farmers. The survey of Robin Hood’s Ball by Mr. D. Bonney for the Royal Commission on Historical Monuments (FIG. 2) established for the first time the presence of what is probably a small barrow mound on top of the outer bank at the north. His work also made it clear that at this point the outer ditch had been widened to provide material for the feature. Respect, even veneration for earthworks left by preceding groups, especially when of a ritual character, appears to have been a recurring feature among pre-Roman groups in Britain. The large barrow overlying the ditch of the causewayed camp on Whitesheet Hill, above Mere (Wilts.), the even larger Conquer Barrow overlapping the henge monument of Mount Pleasant, near Dorchester (Dorset) and a mound occupying a similar position at the henge monument of Arbor Low (Derbyshire) all suggest sympathy with the works of earlier peoples. The small bank and ditch dug around the Heelstone at Stonehenge by the builders of phase ii there illustrates the same tendency. And it is surely not just because of a local high point that a fine Bronze Age barrow cemetery extends eastwards from the middle earthwork of Windmill Hill (Wilts.) itself. THE EINDS POTTERY About 230 potsherds, 6 lb. 9? oz. in weight, were recovered during the excavation. Each fragment was studied minutely in the laboratory and particular attention was given to the nature of the ‘filler’ incorporated in the clay of the pieces. A Gowllands pocket magnifying glass (x10) and a Sterimag (20) microscope were used for these identifications, inclusions being removed from the sherds with a needle set in an Eclipse pin-vice. Note was taken of the colours of the sherds and the textures of their surfaces, for the information this might throw upon methods of firing and degrees of hand-finishing of the pottery. Finally, careful search was made for indications of the method of manu- facture of this group of early Neolithic wares. Three sherds were found to include grain or seed impressions: these were sent to Dr. Hans Helbaek, whose report is included below. FILLERS The word filler is used here as a convenient term to include all material incorporated in the clay with which the pots were made, from chopped straw or crushed shell to pulverized rock or sand. Those fillers which have been identified are set out below in a table. It shows the numbers of individual potsherds in which the various fillers occurred, 13 eG | a | th |] Vv] 9 iP GE} osz=pGr | 1 | o | er Shilcoe |r | Jake] Jad seo, |e ee eae ee 0-Gz | le | — ig eae aa rt ti - a}TJOO parayiVIM pur WIP o1| 1 | a pan I | 0-1 | br | — es ee eee | auole zi1eNn¢) es ae a ee sie ew Spec Sam ners (aes Fd eo ar C.z € (doowIVd? ) WvpIOISe]g ae a eles pe ae 1) I Fea eae (ees g-0 I Jul} pue suo0g 25 ae poe = a en aie I z fears 5 Fea (ec | Gas ¢ IU[OO paroy}V9 AA i ae ae ee I PS (A (pers CRE nS | ea ee 3-0 I JU, pues oyIseoIeyy ee — |1 v Pe secs a z.V G "+ JUIEE pue 4s proysiog oe | | tf | et, [9-3 z — aa I g-0 I uleig ee ee (ned Pema fee eee a= a ral pee tae el ee pee | eee ee Wt z "+ JuIyT pue spoo1/sseis paddoyg I ae ee ee 6.0 I = ae |e = | B20 i PD He EY Oe zo |—|F par G.G 9 II be eo a O-O1 | ZI aUOTL [[PYS I11]0Q ay eS pales | ree pam ae eae z Leet z JUIE pue [[PUs I11[0O bo|—|e I ee G.9g L € —. | TN 136 I L.9 8 "* QIQISIA SUIION Fie = kos ares 6-6 or | — oe ee a ee z|/—]| 1 | SG.z & zyzenb pue 4urpy 93 I oe € | b | o-gS | IF | zz 4o/}gr}/g |— eee I} 6 | —|24.99] 0g auore 1UIT,] Ned | O-|. 4 TF % | S$pseys) W as Pa 4 | @ | Ge) 5\ ‘alley % | Sproys JoAey tod spiays Jo [v0], feo BE rake] iad spiays Jo [210 J, AG L JOT UNV UILAO HOLI(] YANN | SS SS J T1avbe the layers in the ditches of the camp and in (and under) the outer bank where they were found, and the percentages of sherds incorporating the various fillers to the total number of sherds from each cutting. Flint ranging in grain size from pieces } in. in width to a microscopic grain, but with an average size from 1-4 mm., was the most popular filler in use among the potters of Robin Hood’s Ball. From the old land surface beneath the outer bank a notably high proportion of sherds included flint and weathered oolite mixed together, while from both earthworks quartz was occasionally found mixed with the flint. Weathered oolite and crushed Jurassic shells constituted the other important materials used (on their own) as fillers. Quartz used alone as a filler was conspicuous from the outer earthwork. Rarer materials included chalk, potsherd fragments, bone and marcasite, chopped grass and reeds. The weathered plagioclase found in sherds from the inner ditch, here called West Country ware, constitutes a class apart, whose significance is discussed below. In one instance the interior surface of a potsherd from the old land surface beneath the outer bank included a large amount of minute quartz grains, whereas its core and exterior surface showed only minute flint. This was the only example where specially prepared, quartz-gritted clay seems to have been applied to one surface of a pot to achieve a desired surface. Presumably it was applied as a slip, although there was no evidence of this in the cross-section of the sherd. METHODS OF MANUFACTURE No evidence was found to suggest how the vessels had been made. The rims are so simple that all could have been everted and thickened using finger pressure: no examples of rims added as a separate coil were noted. At least one lug had been applied as a separate plug pressed into a hollow prepared for it in the wall of a vessel (a fragment, not illustrated). Impressions on the surfaces of the carinated sherd (Fic. 4: 3; pL. II: 6) suggest that this part of the vessel had been shaped with fingers and thumbs. The vast majority of the sherds had received at least some smoothing of their surfaces; a few were almost untreated in this way. Three sherds only, those of ware from the West Country, could be described as burnished. COLOUR In colour the vessels ranged from black, through many shades of brown and red to bright orange. About 70 per cent. of the potsherds exhibited three areas of colour— exterior and interior surfaces and central core or body. These changes are shown in the table below, the colours being divided into groups of browns, reds, black and grey. Percentages of colours for areas of sherds are shown. Tase IT Exterior surface Core Interior surface Browns Ae ny. ss 48: 32-6% 32: 24°2% 36: 25-72% Reds... iA a i. 78: 53°1% 22: 16°6% 33: ag1% Grey .. a Ae - Te 027% 42) 322%, 2) Teg cs Black .. ae ae aa 20: 136% 74: 56:0% 72: 50°3% Totals of samples ee 147 132 143 This table emphasizes that the exteriors of the majority of vessels had been fired in an oxydizing atmosphere. About 50 (21-9 per cent.) of the sherds from the site were evenly fired throughout 15 to one of four groups of colours as follows, with percentages of the total of evenly-fired sherds: Reds Browns Grey Black 16: 32% D2 AAs Tee 1252209, Here again the majority of evenly-fired vessels had been submitted to an oxydizing heat. About 10 per cent. of the sherds had been fired so that half the thickness of the piece was the same colour as the exterior surface and the other half that of the interior. SHAPES The forms of the vessels (FIG. 4) represented by the potsherds are simple. Rims have the minimum of definition and only one vessel (FIG. 4: 3) has any suggestion of a carinated profile. ‘The two vessels of ware from the West Country (Fic. 4: 12, 17-18) have been drawn as simple bowls with almost vertical sides. The rim sherd No. 17 is so small, however, that we cannot be certain of the angle of its wall and it is just possible that the vessel from which it came may have been very much shallower and more plate-like. Also the incised sherd (No. 18) which is clearly part of the same pot, may as well have come from its base as from its wall. Of the three lugs in the assemblage (one fragmentary and not illustrated), two were oval (including Fic. 4: 7) and one (FIG. 4: 2) more elongated, with a concave upper surface. COMPARISONS As a group, the vessels from Robin Hood’s Ball bear fairly close resemblance to those from Maiden Castle (Dorset) ,?! while lacking their specialized lug forms. There is a resem- blance, too, to the pottery from beneath the outer bank at Windmill Hill?? and from the isolated pit on Waden Hill (Wilts.)?3 just to the south of Avebury. But from both these sites rim forms tend to be a little more elaborate. Attention has already been drawn to the presence, at all the sites mentioned, of Jurassic shell and other materials used as fillers, and also of vessels which must have been made in the West Country, perhaps somewhere on Dartmoor. The presence of chalk as a filler in two sherds from Robin Hood’s Ball has also been recorded from Maiden Castle (Dorset) .74 It was noted that a number of potsherds from Robin Hood’s Ball had a sooty deposit on the interior which may be the remains of their contents. Traces of soot were also noted in cracks and hollows on the exterior of some fragments. Details of the potsherds illustrated in fig. 4 are as follows: | (1) Rim and wall of bowl, diam. 142 in. Rim slightly everted and thickened on interior. Exterior dark red-brown; interior pale brown deepening towards base. Surfaces smoothed. Flint gritted ware with some potsherd grit. Trench 1, top of layer K. Depth 4 ft. 8 in. (2) Rim and shoulder, with lug, of bowl, diam. 5 in. Everted rim, elongated lug with concave upper surface and tilted slightly out of horizontal. Uneven orange-black exterior, orange interior, deep brown core. Smoothed, almost burnished surfaces. Much fine flint grit and well-pounded Jurassic shell. ‘Trench 1, layer H, depth 4} ft. (3) Upper wall of carinated bowl, rim missing. Diam. more than 63 in. Exterior mottled grey-brown; core and interior almost black. Smooth surfaces. Shell gritted with some flint (pL. II: 6). Shoulder formed by pinching out with fingers and thumbs. Trench 2, top of layer J, depth 4 ft. The shells used as filler in the clay of this sherd were identified by Vincent C. Smith, Bolton Museum, as follows: ‘1. Modern shells (i.e. not pre-Pleistocene). 2. Shells of fresh water species, probably Corbiculidae. Matrix probably a glacial or river mud with high organic content.’ 16 @ IZ Fic. 4 Robin Hood’s Ball: pottery. (Drawn by D. Grant King.) (4) (5) (6) (7) (18) 18 Rim of bowl, diam. c. 7 in. Rim sharply everted and rolled over. Surfaces pale red-brown, core brown-black. Surfaces smooth with slight sandy texture. Fine quartz sand and sparse flint grit. Trench 2, layer P, in old land surface, depth i ft.3. in. Rim of bowl, diam. 5§ in. Rim expands inwards slightly with flat top. Exterior grey-brown, interior uneven red-brown; core dirty yellow darkening towards exterior. Fairly high concentration of flint grit, some c. } in. wide. Surfaces fairly rough and gritty. Trench 2, layer P, in old land surface, depth 1 ft. 7 in. Rim of bowl, diam. ¢. 9 in. Simple rounded rim. Exterior, interior and core deep red-brown. Exterior smoothed, interior appears slightly decomposed. Slightly sandy ware with one flint grit visible. Trench 1, layer M, primary rubble silt, depth 73 ft. Rim of bowl, with lug, diam. c. 10 in. Rim everted and thickened, lug oval. Deep brown ware with smooth surfaces. Large flint grits (some c. } in. wide). Trench 1, layer M, primary rubble silt, depth 7} ft. Rim of bowl, diam. 9 in. Slightly everted rim; body expanding to a greater diameter (c. 123 in.). Exterior and core deep brown, interior paler brown. Exterior almost burnished with irregular vertical and oblique grooves; interior more sandy surface. Sparse flint grit. Trench 1, layer M, top of primary rubble silt, depth 54 ft. Rim of bowl, diam. ¢. 10 in. Rim slightly everted and rolled over. Exterior orange- brown, interior and core grey-black. Exterior smooth, interior more so. Much small and medium flint grit. Trench 2, layer P, on old land surface. Rim of bowl, diam. c. 10 in. Simple, flattened rim slightly inverted. Mottled orange- brown, surfaces slightly sandy. Some fine flint grit. Trench 2, layer P, on old land surface. Bowl with flat-topped, slightly thickened rim and vertical wall. Exterior pale brown, core and interior black. Surfaces somewhat rough. Much flint and shell grit. ‘Trench 1, layer M, primary rubble silt, depth 6# ft. Of the shell grit, F. Woodward, Birmingham City Museum reports: ‘Plentiful remains of bivalve shells. None identifiable, but probably derived from a marine deposit of Jurassic or Cretaceous Age.’ Rim of bowl, diam. 9-10 in. Plain rim, curved on inside, upright wall. Black bur- nished surfaces, grey core. West Country ware, resembling Nos. 17 and 18 in hardness and surface burnish, but more evenly dark in colour and includes quartz grits. Trench 1, layer L, depth 53 ft. Bowl with slightly everted and thickened rim. Exterior mottled brown-black, interior and core red-brown. No special treatment of surfaces. Much fine quartz and some large (c. $ in. wide) flint grit. Trench 1, base of layer G, depth 4? ft. Rim of bowl, diam. 6} in. Rim slightly inverted, with rounded top. Grey-brown ware, exterior a little darker and rim redder. Surface smoothed and rim somewhat burnished. Fairly plentiful small flint grit. Trench 1, layer M, top of primary rubble silt, depth 64 ft. Bowl with slightly everted and rolled rim, inside bevelled. Brown-black ware. Sparse small flint grit. Trench 2, layer P, on old land surface. Bowl with rounded rim slightly rolled and thickened. Red-brown ware. Some large flint grits. Trench 2, layer P, on old land surface at centre of bank. Bowl (or perhaps a platter) with pointed, inverted rim. Dark brown-black surfaces with pale brown core. Thin, unusually evenly and hard fired, and highly burnished. West Country ware, like No. 12; perhaps same vessel as No. 18 (see below). Faint scored lines on exterior and interior. Trench 1, layer M, top of primary rubble silt, depth 64 ft. (pL.-IT: 2). Wall or base sherd of same vessel as No. 17. Group of criss-crossed scratches on interior surface, perhaps the result of repeatedly cutting up food in the bowl (PL. Ika). These two sherds, together with two body fragments containing flint and quartz filler, and a third fragment containing oolite grains (pL. II: 3-4), were submitted to H. W. M. Hodges, University of London Institute of Archaeology, who reports as follows: ‘Hodges Nos. 71 and 72: both from Trench 1, layer C. These two sherds are of a ferruginous clay and contain coarse flint and quartz particles as a filler. The filler is of very variable size, but the majority of it appears to be a deliberate addition. The sections compare quite well with those of other Neolithic sites in the general area and it would seem that these sherds were made of local materials. ‘No. 73: Trench 1, layer K. This sherd is of a ferruginous clay containing a considerable proportion of oolitic material as a filler. The oolites have been weathered from their matrix, each oolite being a discrete body. It would thus seem that the material comes from an area in which oolitic limestone has been naturally weathered. The alternative possibility that a crushed oolitic limestone was added as a filler is untenable since the matrix has been weathered away. ‘No. 70: FIG. 4: 17-18. These sherds are remarkable for the very high burnish of the surfaces. Essentially the clay is ferruginous while the filler is of a very mixed mineral content. The most significant mineral present is in the form of fairly large but nevertheless con- siderably weathered fragments of a plagioclase felspar; other minerals present include quartzite and olivine. Here again, the various mineral fragments are each discrete bodies and one must assume, therefore, that the filling material was gathered at or near the source of weathering. ‘Comment: The last three sherds described must be looked upon as imports into the region and it is suggested that it is more probable that these were imported as pots than as the material from which to make pottery. The significance of these imports has been discussed by Cornwall and Hodges (‘“The thin sectioning of Neolithic pottery’’, Bulletin of the Institute of Archaeology, Volume 4, 1963). While no precise area from which these imports may have come can be pin-pointed, it is clear that they must have travelled considerable distances: the oolitic material presumably from some point along the Jurassic ridge and the sherds containing plagioclase quite possibly from the area of the Dartmoor granite massif. For the latter sherds there are, of course, even more distant possible sources as, for example, Brittany, but this method of examination will not allow a more precise determination of this point.’ GRAIN IMPRESSIONS Several sherds showed cavities left by plant parts, but only one could be identified. This occurred in a potsherd from Trench 1, layer G. Dr. H. Helbaek, National Museum, Copenhagen, very kindly examined the sherds and reported on the identifiable specimen: ‘It is the lower (embryo) end of the grain of Emmer, Triticum dicoccum. It displays the embryo and the end of the ventral furrow, and along one flank a slight ridge is visible which excludes the other possible identification, Bread wheat. Only about one-third of the grain is represented, the imprint being interrupted by a fracture of the sherd.’ FLINT AND STONE Flint flakes and implements were found scattered throughout the filling of both ditches and on the old land surface beneath the outer rampart. The disposition of the 59 specimens recovered from Trenches 1 and 2 is most easily seen in the following table: Inner ditch Outer ditch Outer bank Layers... te DK M E F H Jj di Specimens ae 2 he 4. 3 28 3 «6 8 Ig Only six pieces had been sufficiently shaped by retouch to merit description as implements. Two of these were typical Neolithic end-scrapers, two had wide concave scraping edges, one had a long cutting edge retouched with a narrow concave notch at one end; and one blade flake was serrated, its teeth showing a little gloss under the micro- scope. ‘These implements, all pale mottled grey in colour, and fresh, are illustrated in FIG. 5, and can be described as follows: (1) Narrow parallel-sided blade, length 13 in. Both edges very finely serrated and bearing traces of gloss. Trench 1, top, layer K. (2) Large broad flake, some cortex at non-bulbar end; one long edge battered by use, a short length on other side worked into a shallow hollow. Extensive wear on all edges. Trench 1, top, layer M. (3) Broad oval blade, some cortex at non-bulbar end; both edges battered by use and one retouched to give a small hollow at non-bulbar end. This edge bears traces of gloss and is heavily worn. Trench 1, top, layer K. (4) Broad flake with parallel, almost microlithic blade-flake scars on non-bulbar face and cortex along non-bulbar end. One long edge has two hollows, one retouched from non-bulbar face, the other bearing very fine serrations with traces of gloss. Trench 2, layer E, outer wall of ditch. (5) End scraper worked on heavy irregular flake with cortex on part of non-bulbar face. Steep flaking to form scraper, with additional retouch around most of remaining edge of flake. ‘Traces of gloss on scraping edge and some wear. Trench 1, top, layer M. Found close to No. 6. (6) Scraper at end of heavy oval flake with some cortex on non-bulbar face. Steep retouch to form working edge; no traces of gloss or wear. Found close to No. 5, Trench 1, top, layer M. One concentration of 27 unworked flint flakes was found on top of layer F in the outer ditch, and the two scrapers described above lay almost touching near the bottom of the inner ditch. Otherwise the flints were evenly scattered throughout the ditch-fillings and on the old land surface beneath the outer bank. The assemblage can be compared with material from all the causewayed camps so far excavated in Wessex. From such a restricted excavation as this, too much cannot be made of the absence of specialized implements like axes and arrowheads. One large, unworked flake of stone was found in Trench 3, base of topsoil, which has been identified by R. V. Melville of the Geological Survey: ‘This resembles very closely both Albian Upper Greensand from Blackdown and Melbury, Devon, and Cenomanian “‘Upper Greensand’’ from Warminster, Wilts., and neighbouring localities. At first I thought Blackdown more likely on general grounds, but Warminster provides just as good a match and is, of course, much nearer to the Neolithic site in question.’ UTILIZED BONE No manufactured implements were found during the excavation, but Miss M. M. Howard noticed that the teeth in a lower jaw of ox, from the inner ditch, layer H, were battered as if from use. This utilized jaw must be regarded as one of those ad hoc implements which can occur among the refuse of any primitive group of people. ANIMAL BONES The finds were very fragmentary, although the joints of several long bones were still in articulation when discovered. Miss M. M. Howard kindly identified the specimens: she was not able to distinguish between sheep and goat. 20 Fic. 5 Robin Hood’s Ball: flint implements. INNER DITCH, Trench 1: Layer B. PA obo ae Z Red deer: metatarsal. Red deer: proximal end metatarsal. Ox: distal end metapodial, rib. Ox: molars, frag. skull, rib; sheep or goat: humerus, frag. long bone. Ox: scapula, distal end large immature metatarsal, lower jaw (teeth battered as if from use as hammer); sheep or goat: frag. long bone. Ox: proximal end tibia. Ox; distal end tibia, sacral vertebra, frag. ulna and patella, astragalus, frag. sacrum, humerus; sheep or goat: frag. carpal or tarsal, frag. rib, frag. immature long bone, metatarsal; pig: frag. lower jaw and teeth. Ox: patella and frag. ribs, proximal end metatarsal, frag. skull, distal end humerus, half hoof, frag. jaw, molars, distal end tibia, frag. horn core and pelvis, articulated radius and ulna, astragalus, proximal end scapula, proximal end metacarpus, cervical vertebrae, frag. pelvis, phalange; sheep or goat: scapula, distal end humerus, humerus, vertebrae; pg: three immature molars; red deer: frag. antler with rose, immature molars. OUTER DITCH, Trench 2: Layer A. B. D. A = Ox: distal end metacarpal, tooth, frags. scapula, long bone, vertebrae. Ox: distal end humerus. Ox: lumbar vertebrae, frag. proximal end radius; red deer: frag. acetabulam, metatarsal. Ox: vertebrae. Ox: condyle of femur; pig: jaw. . Red deer: long bone; pig: jaw, immature molar. Ox: ulna, frag. jaw, phalange, distal end immature metacarpal; sheep or goat: tooth. Quantities of the remains of various species can be expressed in a table: Cattle Sheep/Goat Red deer Pig Trench 1 42 II 4 3 Trench 2 14 I 3 3 Totals EG 12 7 6 The priority of cattle over other animals, with sheep or goat second, above deer and pig, is true also of Windmill Hill (Wilts.) and Abingdon (Berks.). Yet the pre-camp occupation at Windmill Hill and at many other Neolithic sites produced more pig than sheep.25 The presence of much cleared grazing ground around Robin Hood’s Ball, which may be inferred from the high proportion of sheep or goat,?® has already been mentioned (p. 12). 22 CHARCOAL Charcoal occurred in very small fragments scattered throughout the filling of the inner ditch and beneath the outer bank associated with the occupation on the old land surface. A small amount was large enough for identification and this was kindly undertaken by Mrs. F. L. Balfour-Browne, British Museum (Natural History) : INNER DITCH, Trench 1: OUTER BANK, Trench 2: Layer F. Oak. Layer P. Hazel. Gy Oak. K. Ash. L. Oak, hazel. REPORT ON SOIL SAMPLES Thirty-one samples, in all, were received, but some of these only duplicated others and were not needed. ‘’'wo main problems were posed: PROBLEM I This concerned the nature of disturbances at the tail of the outer rampart in Trench 2, cutting through the stratification, possibly attributable to burrowing animals, e.g. badgers. All samples were examined for pH-value, humus-content and, in the case of those concerned with this problem, for phosphate-content also. Run IT: Outer Bank ee (0. 3) ee eee | eerie a + A | 8-1 | 7 10°0 | g°6 b | 8+4 2-7 7:6 c QO 8-1 3°4 8-6 d R fas 4°9 | 6-9 h | A 6-7 | 14-0 | 8-5 J | M | 7°2 3°4 | 75 | IE | 6-8 4°2 4°5 The question of possible animal burrows is not clearly answered by the experimental results. ‘The component most likely to show activity by carnivorous burrowers is phosphate and, though this was indeed higher by a single unit in IIc than in its immediate predecessor and successor, had there been food-bones and excreta in a burrow, a phosphate-level several times higher than normal would have been expected, and it is doubtful, therefore, whether the small increase observed is significant. The general level of phosphate seems to be round about 7 units, with some increase towards the present-day soil, probably owing to modern cultivation or grazing. Badgers, however, are known to ‘spring-clean’ their setts regularly, so the lack of evidence does not exclude badgers as the disturbing agency. Foxes and rabbits are not so particular. 23 PROBLEM 2 Was the outer ditch deliberately filled ? Compare with the filling of the inner ditch. Run I: Inner Ditch Spraple tes ole ng /e) Goes b | A apg} ia 20°0 = (e B 7:5 11-0 — d C 7:8 3°2 aan : D 0 | 2°5 75 f F Oot OG} 7:0 g G 8-1 3:7 7-6 a (g) G 8-1 6°55 a h H 8:1 0:3 4-0 k L 8-2 1-1 = j M 8-2 0:6 = Run IV: Outer Ditch seme | ae | |e) | elton) a5 A 6-1 ame = r B 6-7 0-5 = 5 D 7 » 3000, FEET Fic. 1 Location map. Based upon the Ordnance Survey map, with the sanction of the Controller of H.M. Stationery Office. Crown Copyright reserved SUMMARY A large mound with turf core was revetted with chalk and separated by a narrow berm from a deep, flat-bottomed ditch, slightly oval in plan. A central post-hole had originally contained a post, 5 in. in diameter with wedge-shaped base, packed round with chalk, while an adjacent circular pit 2 ft. deep contained 30 the cremation burial of a young adult, possibly male. Finds included fragments of Bronze Age pottery, a bone pin and a number of flints, including some worked implements. Later Roman activity in the area is attested by potsherds and a penannular bronze brooch from the ditch. EXCAVATION Before excavation the barrow appeared as a large, low mound some 100 ft. in diameter, steeper on the west side, and thickly covered with charlock, thistles and other weeds and pitted with burrows—known locally as a favourite haunt of badgers and pre-myxomatosis rabbits. Although there was no record of the mound having been previously explored, a shallow depression in the centre suggested robbing had taken place. There was faint indication of a ditch on the west only, with a slight rise beyond it caused by ploughing round the barrow. Excavation was carried out by the quadrant and octant method (see plan, Fic. 2) and the following features were revealed: DITCH A continuous ditch, 4 ft. deep from the present surface of the chalk, but originally over 5 ft. deep (allowing for denudation of the chalk bedrock since the Bronze Age), was 2 ft. wide with a flat bottom and had a diameter of 102 ft. from north to south and 107 ft. from east to west (measuring to ditch centre). The ditch was excavated to the floor in all sectors where possible: elsewhere the upper edges only were defined, owing to lack of time and the thick layer of plough-soil which had to be removed—giving an overall depth in some places of more than 6 ft. Where the outer ditch edge lay beyond the limits of the original cuttings, these were extended on the main axial sections only. The deep primary fill of loose chalk rubble contained flint nodules, struck flakes and cores (particularly in the south and south-west where a number were recovered from the bottom 12 in.); over this was a fine, compact rain-washed chalk silt with buff or yellow streaks, stained yellowish brown towards the top where it merged into the upper ditch filling of dark brown soil (brownearth) containing small pieces of chalk and numerous flints. The top 6 in. of this layer was a chalk-free soil interpreted as an old land surface, marking the level to which the ditch had silted up in ancient times. Contained in this upper ditch filling, notably on the east and north, were burnt flints, Roman and Romano-British sherds and a complete bronze penannular brooch (Fic. 6: 5). The final layer of modern plough-soil was between 12 and 18 in. thick—indeed the plough-soil over the entire field was thicker than is usual on chalk lands and was thought to be due to the former presence of clay-with-flints in the area, pockets of which were noted under the barrow mound. The ditch was found to be regular, its flat floor covered in smooth chalk mud, with steep, well-cut sides, disturbed in places by earthworms or burrows along the upper edges. In the west certain irregularities and indentations were noted in the otherwise smooth chalk mud of the ditch floor. 31 position of small pit Sr as ) , bush Sis hole” interspace ? ip 4 8 12 16 Fic. 2 Plan of the barrow on Earl’s Farm Down (Amesbury G.70). 32 BERM The gently-sloping space between the inner ditch edge and the beginning of the mound was about 15 ft. wide on the north, east and south sides, but slightly wider and more steeply sloping in the west, where the downhill run of the land would have speeded up the process of erosion. In all sectors it was covered with rubble, composed of chalky soil and quantities of flint—in the east-south-east octant the layer between plough-soil and natural chalk on the berm was a solid mass of flints, including struck flakes and some finished tools. A fragment of ferru- ginous sandstone, part of a whetstone (FIG. 6: 1), was found embedded in flints on the natural chalk in the east-north-east. On the east the berm was riddled with large burrows penetrating deep into the natural chalk and making the excavation and interpretation of certain features in this area extremely difficult. These features were: (1) A spread of burnt flints and charcoal in what appeared at first inspection to be two or more pits, the larger running under the chalk ring at the edge of the mound. The smaller circular pit, 1 ft. 7 in. diameter with irregular base and 1 ft. 10 in. at the deepest part, was filled with dark soil and some struck flakes, and proved to have been dug into the chalky silt of the larger, more irregular depression. The latter, which ran under the chalk ring, was filled with chalky silt and grey mould, with a few large flints; the charcoal at the top was superficial and the filling otherwise compact and homogeneous, of the kind so often met with in natural hollows of this type on chalk sites. (2) A roughly circular pit, cut by the east baulk, was filled with chalk rubble and flints, but it was impossible to ascertain whether it was of human origin or made by burrowing animals. Parallel grooves were noted in the chalk of the berm, particularly on the north and east, varying between 2 ft. 3 in. and 2 ft. 5 in. apart, extremely regular and running mostly east-west. While mainly visible on the berm, some lines were seen to run under the chalk revetment, but were not noticed under the turf stack. Erosion channels on a downhill slope may be the explanation, but similar parallel lines, crossed at right angles in many places, were also noted under the second barrow (G.71) and in the light of other evidence from that site the possibility of their being caused initially by pre-barrow ploughing cannot be ruled out,? while Neolithic and Beaker sherds indicate earlier activity in the area. Their absence under the turf stack may be due to chemical weathering (see Discussion below). CHALK ‘RING’ As the turf and topsoil was stripped off the mound, patches of clean compact chalk were uncovered in all octants and proved to be the remains of a revetment (pL. IIIa). It was badly damaged by animal burrows in all sectors, but in places could be seen to lie against the rising turf stack with a band of dark soil between its base and the natural chalk—showing that the old land surface with its turf 33 cover ran underneath it. In one of the better preserved sectors in the north-west, however, an interspace of 3 ft. 6 in., filled with light brown chalky soil, between the inner edge of the chalk revetment and the start of the turf stack was clearly seen (section E-F, ric. 3). On the east and south the sections were too disturbed to state with certainty whether such an interspace existed there also, but it seems unlikely since the interspace on the north-west is believed to be connected with the ‘ramp’ in the turf stack (see below). On the west the chalk was definitely packed against the turf stack itself as a retaining wall, and here too, where the land sloped downhill, a layer of flint nodules had been placed on the old land surface under the revetment as a deliberate build-up on this side of the mound. The upper and outer limits of the chalk could not be determined with certainty, owing to its dispersal by rabbits, but there was no sign of it on the top of the mound. It appeared to be more in the nature of a retaining wall round the possibly flat- topped mound, rather than a complete chalk envelope. TURF STACK A central mound, best preserved on the north and north-east, but badly rabbited elsewhere, had a wider east-west axis (E.-W. length 60 ft., N.-S. length 54 ft.) and was built of turves, some measuring over 18 in. long, carefully laid horizontally one on top of the other. While some of the sods could be seen to have chalk adhering and had been cut from chalk soil, most of the stack, particularly in the lower and central portions, was composed of dark and light brown clay with patches of yellow and red. The effect of these colourful clayey turves, with the thin black lines of compressed decayed humus standing out clearly between them, was most striking and quite unlike the ordinary soil core with faint outlines of sods usually met with in barrows on chalk-lands3 (pL. IIIb). The irregular form of the turf stack illustrated in the sections shows the extent to which disturbance had rendered the turf structure indistinguishable. In the best preserved portion on the north-east a diagonal break was noticed in the turf stack (see section E-F, ric. 3) and is believed to show the existence of some sort of ramp in this sector during an intermediate stage of building. At the interfaces of turves, particularly in the lower levels where the stack was compact hard clay, a film of redeposited iron had formed. This pan-like deposit was particularly striking on the old land surface, where it was possible to peel off layers of clay all over the barrow floor under the turf stack, exposing a bright red surface mottled with the black of decayed vegetation, which quickly faded and cracked as it dried out. In one place, due west of the central post-hole (Fic. 2) the disturbance of the turf stack reached the barrow floor and broke through the old land surface. Elsewhere the bright red floor was intact, except over the cremation (see below). The thickness of this old land surface varied, giving a deep profile over several pockets of dark brown or orange-brown clay (residual clay-with-flints) and a shallow one of the normal rendsina type over the chalk elsewhere. Under the mound the chalk was soft and pasty—section A-B (FIG. 3) shows the actual level to which 34 - Va LRONEAN <=> IRON “PAN POCKET OF ORANGE CLAY WEATHER! ‘HAL ¢ | MOUERN PLOUGH SOIL (e] ne meD CHa fry TURF & TOPSaL Ws CHALK RING EINE) FEY EE] CHALK CHIPS DISTURBED BROWN SOIL PASTY CHALK (BEOROCK) & BROWNEARTH OVER MOUND UNDER MOUND B+ RECENI BURROW COMPACT CHALK SILT TURF STACK i= O10 LAND SURFACE —) FUNTS es] cua RUBBLE LISTURBED | BUSH” HOLLOW 2 1 RON PAN LARGE FUNT FLINT S FEET OO 0 5 10 ai Ta arS TURF & TOPSOIL [==] TURF STACK fax] BROWN SOIL fam (01D LAND SURFACE CHALK RING (==) FLINTS (inner edge) Fic. 3 Sections of the barrow on Earl’s Farm Down (Amesbury G.70). SE LT the ground was excavated in an attempt to find better chalk, but in fact the rotted, pasty bedrock appeared to continue for a considerable depth. A small amount of pottery, two worked flints and part of a bone pin were found scattered at random through the turf stack. CENTRAL FEATURES (1) Post-hole (pt. IIIc). A low chalk mound was covered over with turves placed upside down, with impressions of decayed vegetation clearly visible on both chalk and the turves as they were removed. This chalk mound had a superficial but regular hollow on the north side (made by rabbits) and a regular hole, 5 in. in diameter, in the top. On sectioning the feature the hole was found to be filled with chalk rubble to a depth of 63 in.; a band of unburnt wood (unfortunately unidentifiable) adhered to the side and larger chalk lumps formed a plug at this point, below which the hole was empty (FIG. 4), with a total depth of 2 ft. 9 in. wood band aT Tlbilii iti > SOOT WASSRBEBESE i — chalk ie | co El ORaUR th cele iss) flinks om) old land} Surface A B < FesT rr | ° 74 Fic. 4 Section through central post-hole. and a wedge-shaped base. Surrounding this socket, the outlines of a small pit, filled with compact yellow-buff chalk, stained with earth in places, could just be distinguished against the natural chalk. From these factors it can be deduced that a substantial post, trimmed to a screwdriver-shaped end, had been set upright in a pit dug into the subsoil, packed round with chalk and further supported by the mound of chalk heaped round its base on the old land surface—this latter feature suggesting that the post may have stood to a considerable height. (2) Cremation pit. A break in the ‘iron pan’ on the floor of the barrow just south of the post-hole was caused by a circular pit cut into bedrock (diameter 18 in., depth 2 ft.) with nearly vertical sides and slightly rounded, uneven base. 32 It contained the only funerary remains from the barrow: a large quantity of incompletely cremated bones and a small amount of charcoal in the bottom 8 in. of the pit (see Appendix I). The cremation was covered over with clay sods, some with redeposited iron staining, which reached to the top of the pit, forming a slight dome at the level of the old land surface (Fic. 5). Since no chalk was found in the Diagrammatic section through cremation pit. fill of the pit, the excavated material had been removed elsewhere, and may have contributed to the chalk packing round the post-hole. On cleaning out the pit, which showed no discoloration from burning, tool marks were clearly seen, particularly on the north side (pL. II1Id). Long vertical marks measured 1 ft. 3 in. and shorter ones 6 to 4 in. These marks averaged 1 to } in. in width and could have been made by a bone or wood tool, possibly of rectangular section, used vertically to straighten the pit sides. The only identifiable fragment of charcoal was obtained from the pit and proved to be oak. A soil sample from the pit was examined for land mollusca by Dr. M. P. Kerney of the Department of Geology, Imperial College, and the results are given in Appendix Ila. PREHISTORIC FINDS FLINT An analysis of the flint industries from both barrows is being undertaken as a separate study and for this reason none of the material is included in the present report. STONE (FIG. 6: 1) Whetstone (fragment) of dark red ferruginous sandstone, from berm. FOSSIL Two beads made by enlarging the natural perforation in the fossil Porosphaera globularis4 were found, one in the ditch, the other in the turf stack. The latter had 36 the perforation enlarged at one end only, and does not appear to have been strung, while a small indentation on the side indicates that an attempt to perforate the fossil laterally was made, probably before the natural perforation (which could have been blocked by mud) was noticed. BONE (FIG. 6: 2) Part of a finely-made bone pin of D-section was found in the lower part of the turf stack. The flattened side is smooth and polished, the convex side bears parallel grooves on its upper portion, while numerous small cuts and marks occur round the needle-sharp point. Fic. 6 Finds from the barrow on Earl’s Farm Down (Amesbury G.70). 1. Whetstone fragment; 2. Bone pin; 3. Beaker sherd; 4. Sherd of miniature vessel; 5. Penannular bronze brooch. Scale: 1, $4; remainder, 1:1. POTTERY? With the exception of the two sherds illustrated in Fic. 6, the fragments of pottery found scattered in and under the turf stack are all featureless. They can, however, be separated into groups on the basis of fabric. Western Neolithic. ‘Twelve somewhat weathered body sherds, none larger than 35 X35 mm., and a few crumbs, come from one or two relatively thin-walled vessels (6-8 mm.). The paste contains varying quantities of medium-sized shell fragments. The outer surfaces are pale brown over darker cores and interiors. The fabric, taken together with the absence of any decoration or indication of form, suggests that these sherds may have come from round-bottomed Western Neolithic vessels. 37 Beaker. There is a single small sherd bearing the characteristic hyphenated decoration (FIG. 6: 3). It comes from an undistinguished Beaker with a reddish- brown outer surface, flaky black core and black interior. The paste contains a little sand and probably grog (comminuted pottery). The decoration has been made by means of a stamp with poorly-formed and irregular teeth. A further five undecorated body sherds, none larger than 30 X20 mm., and a dozen crumbs appear to represent a second Beaker (though the possibility that they belonged to a Food Vessel of the Southern English type cannot be ruled out). The heaviest piece is 9 mm. thick and the outer surfaces are reddish over a black core and interior. The paste contains grog and occasional small pieces of flint. None of the sherds appears to be weathered. Collared Urns. Three urns appear to be represented by four sherds, all somewhat larger than the Neolithic and Beaker sherds and two of considerable size (75 and 85 mm. in maximum dimension). Urn fragments of these dimensions would not survive long if exposed to weather and trampling, and it seems probable that these pieces were freshly broken when incorporated in the barrow. (1) Sherd 10-11 mm. thick with slightly concave external profile, probably from the neck of a tripartite urn. The outer surface, well smoothed and slightly glossy, forms a very thin and even, reddish-orange skin over a flaky black core; the red of the inner surface is variable in thickness. The paste contains abundant grog particles, up to 3 mm. in diameter. This was the only sherd not contained within the turf stack, but its position in the rubbly soil over the berm may well be due to rabbits, etc. (2) The two largest sherds probably come from the body of a second urn. Although one is perfectly fresh and the outer surface of the other has been abraded and scratched—apparently by the claws of burrowing rabbits—they seem to be otherwise identical. On the outer surface of one is a small impression, measuring 8 X10 mm., such as could have been made by a textile pressed into the damp clay. Both inner and outer surfaces are irregular, with small pieces of grog protruding; the inner surfaces exhibit numerous fine shrinkage cracks. ‘The exteriors are a pale orange-buff in colour, the interiors brownish black, and the cores black and flaky. Thickness varies from 9-12 mm.; in addition to the abundant grog there are a few small pieces of flint, probably incidental inclusions. (3) A sherd 10 mm. thick, with irregular surfaces and containing grog only, has a reddish brown exterior and seems to come from a third urn. Miniature vessel. A single sherd (Fic. 6: 4), found on the old land surface, represents part of the rim and body of a small cup-like vessel with an estimated diameter of 42 mm. The original height is unlikely to have been much greater. The vessel appears to have been modelled, somewhat crudely, from a single ball of clay. The surfaces are red over a flaky black core and the paste contains sparse and fine particles of flint and perhaps grog as well. A small tool with a wedge-shaped end has been used to make two lines of deep vertical impressions below the outer edge of the rim and slightly oblique ones across its inner edge. This may have been an accessory vessel of a type often found with Bronze Age cremations. ‘The wedge- 38 shaped impressions and the narrow straight-sided form are both known from barrows in Wiltshire.® FINDS OF ROMAN PERIOD BRONZE BROOCH (FIG. 6: 5) A complete penannular bronze brooch, type D.57 was found at a depth of 2 ft. in the upper ditch filling near the east baulk. The hoop of the brooch is not much worn, the patina good. The pin, in contrast to most penannular brooch pins, is wound twice round the hoop instead of being looped over, and this fact, together with its generally coarser appearance and indifferent patina, suggests that this pin is not the original one but a replacement in antiquity. An almost exact parallel, apart from the pin, comes from Hod Hill® and this heightens the supposition of a late ist-century date for the Earl’s Farm Down brooch. E. FOWLER POTTERY The foregoing report describes only the excavation of Amesbury G.70. For the sake of completeness, however, the Romano-British pottery from Amesbury G.71, excavated at the same time, has been included, along with the G.7o finds, in the description below. AMESBURY G.70 A small group of some two dozen fragments of coarse pottery and six small Samian sherds was recovered from the topsoil and upper layers of the barrow mound and ditch. The fabric of the coarse ware is gritty in texture, and varies in colour from light grey to grey or reddish brown at core and surface. A few sherds are somewhat abraded at the edges. Only three sherds were rims, and these are described below: Mortarium, hard fabric, grey core, brick-red surface with traces of a cream slip on the rim underside. Rim type suggests 2nd century A.D. Small fragment in a hard blue-grey fabric of a late New Forest indented beaker. Normal cooking pot type rim in a dark grey gritty fabric. Undatable. The Samian pottery comprises four worn body fragments, and two rim frag- ments of forms Dr. 18/31 and ?33 (Layer 2a, chalk ‘ring’), probably of gnd- century date, although every sherd is too small to be accurately identified and dated. AMESBURY G.7I Considerably more Romano-British pottery was recovered during the excava- tion of Barrow G.71, mainly from Layers 2 and ga. The group as a whole comprises some 110 sherds of native wares, 12 Samian fragments, oyster shells and a few flanged tile fragments. The fabric of the coarse pottery is again predominantly grey to greyish-brown in colour at core and surface, gritty in texture and frequently abraded. Some 17 39 of the sherds were rims, and as would be expected, this larger group features a wider range of types than was found in Barrow G.70. They include cooking pots, shallow dishes, flanged bowls, mortaria, heavy storage jars, and colour-coated wares, bottles and indented beakers typical of the New Forest series. Of the rim types, none is earlier than the end century, and the presence of New Forest fabrics from both barrows implies an occupation extending beyond the grd, and into the 4th centuries A.D. There were 14 Samian sherds which are listed in detail: Layer 2. Fragment from footring of a large bowl. Probably f.18/31 or Lud.Sb. ?Central Gaulish ware c. mid 2nd century. This fragment bears a cross, perhaps the owner’s mark. Small chip from rim of platter. f.18 or 18/31, probably 2nd century. Part of rim of mortarium f.45. Second half of end century. Two other chips are too small for identification. Fragment from base of large and heavy vessel which cannot be accurately identified. Central Gaulish ware, probably 2nd century. Layer 3a above Ditch. Large fragment of footring of bowl or platter. Probably f.18/31 or Lud.Sb. Central Gaulish ware, c. mid 2nd century. A second fragment found with the above is too small for identification, but may be end century. Fragment of bowl f.18/31 or Lud.Sb. Probably Central Gaulish ware. Small rim fragment either f.18 or 18/31. Small chip from body of cup f.33, but difficult to be certain. The alter- native is that it comes from the upper surface of the base of a platter with a high kick. The last three sherds are all probably 2nd century. Topsoil and Layer 3a above Ditch. Two small unidentifiable fragments. The darker piece may be Lezoux ware, the lighter is perhaps East Gaulish. ?Layer. Rim fragment of mortarium f.45. 2nd century. In any further consideration of the coarse wares, the unfortunate absence of well-stratified pottery groups in the region makes it impossible to suggest accurate dating for individual rims, and at the same time, emphasizes the difficulty of attempt- ing to distinguish possible gaps in the occupation such as might be tentatively suggested by a close analysis of the material from both barrows. Little more can be said except to suggest that the entire series indicates occupation within the area from the end to the 4th centuries a.p. It is worth noting that Romano-British pottery, building foundations and surface finds have been previously recorded immediately north-west of barrows G.7o and 71 (W.A.M., XLV, 173, 8a). The present finds may well indicate occupational activity connected with this site, and the tile and oyster fragments emphasize the domestic nature of the occupation. F. K. ANNABLE DISCUSSION Any assessment of the structure and finds from this barrow must be made in the light of the considerable disturbance caused by burrowing animals, which 40 upset not only the stratigraphy of the turf stack but in places penetrated deep into the pre-barrow land surface. Enough evidence was obtained, however, to permit comparison with other sites and to place the barrow in a broad chronological setting. STRUCTURE Although technically a bell-barrow, the berm is rather narrower than is usual with monuments of this class, while the deep and narrow ditch has analogies with a neighbouring site—Amesbury G.61.9 The reason for such a deep, narrow ditch is puzzling—it would certainly present a barrier to anyone attempting to enter the precincts of the barrow. The impressions—if such they were—on the ditch floor were too amorphous to allow any definite inference to be drawn from them, but the possibility that they were caused by feet cannot be ruled out. The chalk ring is believed to have been a revetment to the turf stack and not the overall chalk envelope usually met with in Wessex. Such a revetment, with build-up on the downhill side, recalls the stone retaining walls and kerbs of composite barrows.!° The absence of any chalk or even chalky soil over the turf stack supports this interpretation. Even more significant is the suggestion that rain could percolate freely through such an unprotected turf stack (causing deposits of iron manganese to form on the interfaces of turves and on the barrow floor)! and result in erosion of the natural chalk under the mound, thus producing a level or slightly lower profile than that of the unweathered chalk outside. This is in marked contrast to the ‘proud’ land surfaces met with under many barrows and earth-works on chalk lands. The building of the mound with turves carefully cut and laid like bricks was the most striking feature of this site, and suggests the existence of local deposits of residual clay-with-flints, pockets of which were noted under the barrow itself. These deposits must have been extremely localized—there was no sign of clay-with- flints in the second barrow (G.71) in the same field—and it is interesting to note that the lower and innermost part of the turf stack contained predominantly clay turves, while the outer part was mainly composed of chalky turves. Presupposing that the first turves would have been cut from the immediate vicinity—namely the berm and ditch—this indicates that this area was richer in pockets of clay-with- flints than the surrounding land.'? Such turf structures are more common on acid subsoils, as at Swarkeston (Derby) where the secondary barrow over a cremation had nine layers of turves from 2 to 5 in. thick.t3 Turf is known as a roofing material from the late neolithic settlements on Orkney, while a turf structure with walls standing 3 ft. high was noted in the long barrow on Thickthorn Down (Dorset)."4 Since nothing is known of settlements of the Wessex Culture, it is pure conjecture to suppose that, while temporary shelters may have been of the tent type, using skins, or light wattle structures, turf may have been used for more permanent dwellings. The writer suggests, however, that barrows which show such deliberate ‘brick wall’ construction in turf as the present one may reflect a form of domestic building that was currently in use in areas, both in Wessex and outside, where the 41 turf was suitable. Sods cut from clay soil would make better building material than those cut from the more crumbly rendsina. If the diagonal break noted in the turf stack on the north-east is correctly interpreted as a ramp used at an intermediate stage of building, then the ramps at the turf barrow of Sheeplays 293,*° although structurally different, and that noted recently in a barrow on Arreton Down!’® are relevant. BURIAL AND RITUAL While unurned cremations are frequently met with in Wessex barrows, their occurrence as the primary burial in a pit is rarer.17 But the pottery finds show that cinerary urns were present, while the miniature vessel sherd from the barrow floor suggests that a funerary assemblage was intended. Since the cremation pit was sealed with turves, and no pottery was found in its immediate vicinity, such an assemblage must have been deposited elsewhere—possibly after ritual breakage (see pottery report above). If the disturbance in the centre which reached to the old land surface west of the central post-hole was a robber pit and not due to burrowing animals alone, it is possible that the funerary vessels, broken or unbroken, could have been disturbed and the sherds disseminated subsequently by rabbits. The deep cremation pit is unusual—deep pits containing cremations are known, but are mostly of the large cist type; a nearby bell-barrow had a primary unurned cremation accompanied by a knife-dagger in a pit of similar depth, though elongated to grave shape, with the sides showing vertical tool marks.t8 The central post-hole is, as far as the writer can discover, a unique feature under a Bronze Age round barrow, although multiple post- and stake-holes from pre-barrow structures are well known. It was apparent that it was withdrawn and the plug of chalk inserted in the upper part of the socket before the mound was built, since the socket and chalk packing were sealed with undisturbed turves. It must, then, have been connected with whatever pre-barrow rituals took place: while there was no conclusive evidence for trampling of the barrow floor, such as was found at Pond Cairn,'9 it seems likely that any rituals connected with the central post would have involved movement of people in the area (and such tramp- ling would undoubtedly have favoured the subsequent formation of the ‘iron pan’). Its purpose must remain a mystery—a totem or other kind of ritual pole seems probable, as it is hard to think of any functional structure that could be supported by a single tall post. It could, as has' been suggested, be seen as a temporary marker for the site of burial, though in the writer’s opinion such a substantial post with its elaborate packing indicates that it served some other function, probably ritual, as well.?° ENVIRONMENT The presence on chalk land of a barrow built mainly of clay turves which appeared sufficiently acid to permit the survival of pollen seemed a source of infor- mation on the vegetation at the time of building and samples were taken.?! Unfor- tunately, the soil was too alkaline, so no results were forthcoming, and _ local 42 botanists who viewed the macroscopic plant remains on the site were unable to identify any significant species, though grasses predominated. However, examination of the land mollusca and the bones of small mammals adds yet further to the evidence gradually accumulating from other sites, that a very dry, open environment, with short-turfed grassland, prevailed. DATING On the grounds of its bell form, the barrow can be placed broadly within the Wessex Culture. The pottery and other finds do not provide sufficient evidence for suggesting a more precise date within the period. APPENDIX I CREMATION This was a large cremation, though more than half was in unrecognizably small pieces and crumbs. A good part, however, was in fairly large fragments, as cremations go. All parts of the body were represented, and despite diligent search, no evidence of more than one individual was found. The skull was represented by no piece of more than 5 cm. in maximum dimension, but a few were significant. Two fragments, thinnish bones, of orbital margin included enough of the brow ridges to show that these were only slightly developed, but a piece of the anterior part of the frontal in the midline exhibited a fairly large frontal sinus. Only a single piece of the region of the external occipital protuberance was found; it was badly warped but nevertheless showed the superior nuchal line, which was fairly well marked. A number of biggish vault fragments showed anomalous sutures clearly indicating wormian bones, though owing to warping and fragmentation the exact position of these could not be ascertained. The sutures were here already obliterated on the inner table, so that one would guess at an age between twenty and thirty-five. The squamous suture was still open. As usual, only the bodies and roots of teeth were found, the enamel having perished in the fire. Individual teeth, in consequence, were not recognizable and all that can be said is that from the roots recovered, the individual was fully adult. A fragment of mandible, much shrunk, showed the alveolus of a fully-erupted M3, confirming the adult age. A few vertebral centra were reasonably complete and showed no signs of pathological change. Only one axis fragment, with the odontoid process, was found. Little or nothing can be said about the long bones and extremities. The longest shaft fragment measured only 10 cm. and was so warped, as were most of them, as to be almost unrecognizable. In sum: a young adult, possibly male, in good health as far as these remains can show. I. W. CORNWALL Institute of Archaeology, London APPENDIX II ENVIRONMENTAL (a) Land Mollusca The following land Mollusca were obtained by washing a sample of the cremation material, weighing approximately 0-5 kg.: Truncatellina cylindrica (Ferussac) 2 Vertigo pygmaea (Draparnaud) 7 43 Pupilla muscorum (Linné) 14 Vallonia costata (Miiller) 6 Vallonia excentrica (Sterki) 16 Helicella itala (Linné) 2 Punctum pygmaeum (Draparnaud) 3 Although it is difficult to be certain on the point, some of these shells appear to be partially calcined, suggesting that they had lain on the surface where the body was burnt and were subsequently gathered up with the ashes. If this is so, it should be borne in mind that the Mollusca are not necessarily contemporary: some may have lived at the time, but others may equally well have been present as dead shells in the soil on which the cremation took place. In spite of its very small size, the assemblage points clearly to dry and very open, short-turfed grassland. Of particular interest is the presence of Truncatellina cylindrica, a species unrecorded living in Wiltshire.*? 7. cylindrica is restricted to dry exposed places, such as scree slopes, sand hills and maritime grassland. M. P. KERNEY Department of Geology, Imperial College (b) Animal Remains?3 Owing to the disturbance of the site already mentioned, little stratigraphical value can be placed on this material. However, of the 266 bones and fragments of animals examined, mostly from the body of the mound, sheep and/or goat predominate and are thought to be prehistoric—occurring as they did throughout the turf stack. Sheep and goat are not distinguishable on the material available except for two horn cores and a cranium which are certainly of sheep, a small breed resembling the present Dartmoor sheep. Of the 141 fragments, 43 show signs of immaturity such as new unworn molars, unfused epiphyses or deciduous teeth, indicating that about 30 per cent. of the animals were slaughtered when young. Only three of the bones are complete enough for measure- ment and these are metacarpals of 11-5, 11:1 and 10-65 mm. overall. Cattle remains (32 fragments) occurring in the mound and in the chalk revetment were very fragmentary, but the breed appears to be small and of the size of Bos longifrons, common to the period. A few fragments of red deer and hare were present, while the only pig remains from the site occurred in the old land surface, under the ‘iron pan’ of the central area, and are either contemporary with or even earlier than the building of the mound. They consist of one upper tush of boar, one upper tush of (probably) sow, and one fragment of maxilla. Dog, occurring in the central robber pit, and fox were represented, as were modern species such as rabbit and the common fowl. There are no remarkable features about the collection, which is fairly typical of material from such a site, although a larger wild fauna might have been expected. (c) Small Vertebrates The small bones found in the barrow were recorded on six separate finds and in total they comprise the remains of at least three moles, two weasels, one short-tailed vole and a great many small (young) amphibia. The small vertebrate bones must be the remains from the pellets that are disgorged by raptorial birds. In other archaeological excavations, notably those in the barrows on Snail Down (Wiltshire) these tiny bones have been found in small aggregates just as they were in the original pellet. ‘The present finds are rather few in number—if a large sample of items of prey can be examined, then some indications about the environment can be obtained. However, it can be noted that the short-tailed vole (Microtus agrestis) is present and this is typically an animal of open grassland. The bank vole, an indicator of scrub 44 and woodland, is absent. ‘The presence of the mole and an abundance of amphibia (probably mostly toad) are quite consistent with an open environment. Weasels are often taken by the larger birds of prey. If the environment were open, then it is easy to understand how any stakes erected by man would make attractive nesting-posts for birds like buzzards and kites. Even without stakes, a turf pile would have the same attraction. If the bodies of the dead had ever been left exposed at the place, there is even more reason to see why birds frequented the site. P. A. JEWELL Wellcome Institute of Comparative Physiology, Zoological Society of London 1 The writer is indebted to the Ministry of Public Building and Works and to numerous colleagues and specialists; to Dr. Cornwall and his colleagues at the Institute of Archaeology, to Drs. Kerney and Jewell, to Mrs. E. Fowler and Kenneth Annable, to M. R. Hull for additional comments on the Samian pottery, to Neil Jenson for taking the photographs, and to the tireless group of volunteers whose magnificent work on the site enabled the work to be completed in the time. Thanks are especially due to Dr. Isobel Smith for all her help and advice in the preparation of this report. 2 This subject will be discussed in more detail in the report on Barrow G.71. 3 In view of this factor, combined with the environmental evidence from this and other sites, the terms turf and turf stack are used throughout, since they are considered both accurate and appro- priate. 4 The writer is indebted to Dr. Kenneth Oakley for this identification. 5 The writer is indebted to Dr. I. Smith for this report on the pottery. 6 Abercromby, Bronze Age Pottery, u (1912), nos. 237 and 242. 7 Fowler, E., P.P.S., xxvt (1960), 153, fig. 1. 8 Brailsford, Durden Collection from Hod Hill (1962), 13, fig. 11, E.17. For other examples of Ist or 2nd century A.D. see Fowler, loc. cit. 9 Ashbee, P., The Bronze Age Round Barrow in Britain (1960), 45. 10 Ashbee, loc. cit., 47 ff. 1 But the ‘iron pan’ occurred on the old land surface under the chalk dome round the central post-hole. 12 A barrow on Boscombe Down also had clay- with-flints soil; the finds included a bone-eyed pin and whetstones, and the mollusca indicate a similar though slightly damper environment (Newall, W.A.M., xLv (1931), 432-58). 13 Greenfield, E., 7. Derby. Arch. G Nat. Hist. Soc., LXxx (1960), Barrow 4, Swarkeston. 14-P_ PLS. (1936), Sr. 15 Fox, C., Life and Death in the Bronze Age (1959), 129 ff. 16 Alexander, J., P.P.S., XxvI (1960), 263-302. 17 Ashbee, loc. cit., 81 ff. (See also W.A.M., Lvul (1963), 370-82—a barrow on the Stonehenge Cursus excavated by the present writer had a primary unurned cremation.) 18 Ashbee, loc. cit., 82. 19 Fox, loc. cit., 105 ff. 20 A barrow on Overton Down, had a smaller post-hole at the head of an off-centre grave and is interpreted in this way (information from Dr. I. Smith). 21 By Dr. G. W. Dimbleby, to whom the writer is indebted. 22 Ellis, A. E., Census of the distribution of British non-marine Mollusca. Journal of Conchology, XXII, 171-244. 23 Thanks are due to Mrs. B. Westley for examining and reporting on these remains—a shortened version of her report is given here. 45 CROSS-DYKES ON THE EBBLE-NADDER RIDGE by P. J. FOWLER “COVERED WAYS’, ‘cattle-ways’ and ‘cross-dykes’ are familiar terms of field archaeology, particularly in the works of those earlier writers on whose observations in the Chalk country much of our understanding of earthworks is still based.* It is, however, 150 years since Hoare (1812) and 50 years since Sumner (1913) described and interpreted the linear earthworks to be discussed here; and more than a generation has passed since Clay (1927) re-examined them. His interpretation, despite Williams-Freeman (1932), came to be accepted, and furthermore, accepted as generally true of similar earthworks elsewhere. To a certain extent, this view was supported by the earlier work of the Curwens in Sussex (1918). Now, with little qualification, ‘cross-dyke’ is often synonymous with ‘cattle-way’.? It was partly to examine the basis of this interpretation, i.e. that bivallate cross-ridge dykes were cattle-ways, that this survey was made, bearing in mind the pleas for further work on the subject by both Curwen (1951, 107) and Grinsell (1958, 147). The only reasons for choosing the Ebble-Nadder ridge were its ease of access, and the continual ploughing of the earthworks on it. ‘The main purpose here is to publish a map of the ridge at a reasonable scale showing cross-dykes, settlements, some other relevant earthworks, and areas of ‘Celtic’ fields: firstly, to illustrate the relationships between the local topography and the dykes and between the dykes themselves; and, secondly, to give a rather fuller picture of the settlement pattern on and associated with the ridge than some recent maps have afforded. With the map, a description of the cross-dykes (Appendix I) and settle- ments (Appendix II) form the most important part of the article, preceded by a discussion. SUMMARY Univallate and bivallate cross-dykes are distinguished, both by form and position. It is argued that the bivallate cross-ridge dykes are primarily land boundaries, not tracks, though all may have had secondary purposes as protective or obstructive earthworks and, possibly, as tracks. Univallate dykes are in some cases parts of a track system; in others, notably on spurs, they diverted and perhaps controlled tracks leading to and from the ridge. DISCUSSION Between the valleys of the Rivers Ebble and Nadder in South Wiltshire is a well-defined ridge running south-south-west to north-north-east and then east from an abrupt rise at Whitesheet Hill to the west bank of the River Avon below 46 Salisbury, a length of some 14 miles. Throughout it consists of Upper, Middle and Lower Chalk, capped by Clay-with-Flints only for about 2} miles east from Compton Down. On the north, broken only by two spurs, is a scarp to the Greensand Plateau; on the south, gently rounded spurs slope towards the Ebble on either side of steep- sided combes, sinuous in plan and often deep in shadow. The ridge top descends gradually west-east, from about 775 ft. above O.D. to about 150 ft. above O.D., and is usually about 200-300 yds. wide. Along it, often just south of the highest ground, runs the former Salisbury-Shaftesbury turnpike road with, now, arable or recent arable on either side for most of the way. The scarps, however, are mostly unploughed. Below them, small woods or copses—locally often called Ivers— persist. Along a nine-mile stretch, from Whitesheet Hill to the northern spur towards Barford where the ridge bends east, are eighteen comparable linear earthworks. Their frequency is impressive in any case,3 but particularly so here since the ridge top is not otherwise remarkable for its earthworks. ‘There are only one, possibly two, long barrows and some forty round barrows along the whole ridge, while Chiselbury is the only hill-fort and the enclosure on Swallowcliffe Down the only certain settlement. Great and Little Woodbury lie 5$ miles east of the Barford spur at the east end of the ridge, while the two Fyfield Bavant Down settlements and a third one, first noted here, on Prescombe Down East are all on spurs to the south. The cross-dykes consist of a bank and ditch (univallate) or a ditch between two banks (bivallate).4 They run approximately at right angles across the ridge, or roughly parallel with the ridge across a spur. All the dykes which cross right over the ridge, except for three (D.45-7) at the west end on what is virtually a spur, are bivallate, and all the dykes on spurs except for D.21 are univallate and associated with terrace-ways. There are no multivallate dykes, though a bivallate and uni- vallate (D.19) run together for a short distance on Middle Down. The dykes are remarkably consistent in size: about 30 ft. overall width for univallates on the ridge and 40 ft. for those on spurs, while bivallates are about 40-50 ft. in overall width, with the largest of them (D.25) covering 70 ft. Banks are usually about 2 ft. high and ditches a similar depth, D.25 again being the excep- tion. The length of the dykes is largely but not wholly determined by the relief immediately adjacent and is not presumably of much significance in itself. The longest (D.57) is only 1,000 yds. long, the shortest (D.79/) 100 yds. long. Not one has a certain original entrance. All the dykes off the ridge except D.172 definitely have no original break; and it has to be argued that the turnpike road passes through an original entrance in every case if they are to be sought on the ridge top. If no entrances existed, as seems the case, then presumably the ridge was not a ‘way’ when the cross-dykes functioned. Although all but the pair of dykes D.79a and b have been ploughed to a greater or lesser extent, in most cases the ends are well-preserved. The ends are here given special consideration since they have not been touched on in V.C.H., and some at least of the uses to which the dykes have been put should, both in theory and in the light of experience in Sussex, be ascertainable by ground inspection at these points. The ends in fact vary greatly. They lie on or near the ridge top; some way 47 or immediately short of the shoulder above a scarp; on the shoulder; immediately below the shoulder; and well down the scarp, in a few cases almost at its foot. Not one, however, actually continues on to level ground beyond the foot; and not one dyke has both ends dropping well down the scarp. The form of the ends also varies, though in all cases the bank or banks stop before or level with the ditch. They either come to a well-marked end or simply fade out. The ditches often narrow before ending. They either run out, have a cupped end, or splay out, the last type occurring only on ‘spur-dykes’ (below, p. 50) above terrace-ways. In three cases only, and these close together (D.45-7), a terrace-way emerges from one end. Except on the spurs, all the dykes (D.22 possibly excepted) have one end related to a combe. In ten cases an end occurs above the head, and in four above the side, of a combe. The cross-ridge dykes do not, however, link combes, though D.18 and D.21 run across spurs from combe edge to combe edge; nor do dykes occur at every point where a combe from the south runs right up to the foot of the ridge (contra Clay, 1927, 61). Indeed, granted that the south ends of the six bivallate cross-ridge dykes are related to the end of a combe, and that the dykes therefore occur along lines where the ridge top is narrower than elsewhere, there nevertheless seems to be an artificial factor determining their particular position, determining why they cross the ridge top from the head of one combe rather than from another. This factor is distance. The following facts seem relevant. D.107 is roughly equidistant from D.25 and D.75, and D.20 is exactly equidistant from D.25 and D.107. The distance between D.75 and the west end of D.57 is just under 3 miles, less than 4 mile greater than the distance between D.25 and D.107, and between D.107 and D.75. Between D.75 and the west end of D.57, D.74 is about halfway between D.75 and D.73, and D.73 is about halfway between D.75 and the west end of D.57. In other words, the bivallate cross-dykes (with one exception, D.21) apparently divide up the ridge top in a significant manner, suggesting that the dykes themselves are primarily boundaries between units of land. If this premise is correct, then the distribution of the bivallate dykes can be interpreted thus. Between the escarpment at the west end of the ridge and the spur between Barford and Burcombe, the ridge top was divided into four major units, with a possible fifth on the Barford spur immediately north of the Clay-with-Flints, by D.25, D.107, D.75 and, plausibly, D.57. The major unit between D.25 and D.107 was subdivided into two equal parts by D.20; the adjacent major unit to the east, containing the only hill-fort on the ridge, remained in full; but the next major unit was split into three very nearly equal lengths by D.74 and D.73. Unfortunately, further interpretation would be increasingly hypothetical since there is little relevant evidence to develop the argument. There are, for example, unanswered questions about what the postulated land units represent, and whether the boundaries relate only to the ridge top or whether they continued, using either natural features or fences for example, down on to the Greensand Plateau to the north and into the combes to the south. The size of the units, even if they include the spurs between the lines of the bivallate dykes, suggests that they relate to small 48 ay a «4 AN nen rm GS Peet Smeal = a * Se oN =a Bank Celtic field area eceeccose Pitch Terrace way Probable + Fifield Bavant settlements miles —————— ans Fic. 1 Cross-dykes and settlements on the Ebble-Nadder Ridge. settlements, perhaps even single farms, rather than to territorial areas of the type recently postulated on the Chilterns.5 This suggestion is supported by the Swallow- cliffe Down settlement which, contained within an enclosure of some 3 acres, is presumably something akin to the Little Woodbury type of settlement, although no timber house was found in the excavations (Clay, 1925). The Swallowcliffe Down site is, however, the only certain settlement on the ridge-top. Chiselbury is similarly situated right at the edge of a land unit, but, although twice dug into, has produced no evidence of occupation (Hoare, 1812; Crawford and Keiller, 1928, 76). The only other known settlements are on spurs to the south, a mile and more from the ridge top. It is impossible to know how, if at all, they related to the cross-ridge dykes or the land units defined by them. It is nevertheless curious that all three settlements should be on successive spurs from that stretch of the ridge between D.107 and D.75 which is the longest unit between cross-ridge dykes and which ends to the east at Chiselbury. The evidence from settlements is, then, suggestive but inconclusive. It was hoped that more settlements would come to light during this survey, but despite the use of very good air cover® and a considerable amount of fieldwork, no certain examples have been found other than that on Prescombe Down East. Two possible sites were noted (at 93852341 and 95272469), but both are so uncertain that they have not been included on the map, although they would fill out the picture of settlements on the ridge-top, one to each land unit. On the other hand, it would be surprising if any major settlements have been missed, and it is practically certain that no other enclosed sites existed, if only because virtually all the areas where they might have done so were under crop or ploughed when the air photographs were taken and should therefore have produced crop or soil marks. ‘The cross-dykes, then, exist largely in a settlement lacuna, despite their purposive appearance. The survey has, however, added considerably to the acreage of ‘Celtic’ fields previously noted,7 though this has helped little in considering the alleged close relationship between such fields and cross-dykes, previously used to explain the purpose of the latter. Cross-dykes, it was argued (Clay, 1927, 64), were not only ‘cattle-ways’, but ways which led through arable fields; though Curwen (1951, 101) makes the point of Sussex dykes that most are not in ‘Celtic’ field areas. On the Ebble-Nadder ridge, the area on which Clay based his suggestion, it is now too late to determine the relationship. Although the map appears to show that, with minor exceptions, the distributions of cross-dykes and ‘Celtic’ fields are separate, such has been the extent of modern ploughing that it is impossible to be certain whether ‘Celtic’ fields were formerly widespread on the ridge-top. However, the fact remains that the only known association between fields and a bivallate cross- ridge dyke is just south of the Swallowcliffe Down settlement (Clay, 1927, 63). For the rest, ‘Celtic’ fields are (or were, for many areas were already ploughed in 1946 when the air cover used for Fic. 1 was taken, and other have been ploughed since) largely confined to the scarps and particularly south of the ridge. Two points are relevant to this apparent scarp distribution: firstly, faint traces of fields spread over the tops of some spurs, linking two or more superficially separate areas and suggesting what might have happened over the ridge top; secondly, it hardly seems likely 49 that the scarps would alone be cultivated when much more easily worked land lay immediately above. On the whole it must be concluded that, despite the increased area now shown on FIG. 1 to have been formerly cultivated in ‘Celtic’ fields, the distribution represents only the fragments of a once much more extensive pattern; but exactly how, if at all, the dykes fitted into this pattern is uncertain. It remains to consider the univallate dykes, and the slender dating evidence for all the dykes. It is impossible to generalize about the purpose of the univallates, except to say that some had different purposes from others. But the function of some is by no means clear, and as a whole they do not fit into a definite pattern like the bivallates. Three (D.45-7) at the west end of the ridge where it curves south and then south-east form a little group which is part of a track system or systems. ‘There is no doubt that these dykes have been used for foot traffic, and if any of the cross-dykes were ‘cattle-ways’, these are they. It must be stressed, however, that all three are univallate, and that they are the only dykes on the ridge with terrace-ways issuing from one end of their ditch. There is no such evidence from any of the bivallate dykes commonly stated to have been trackways. It is also interesting that the terrace-ways to or from D.45-7 run obliquely up and down the scarps, as indeed do all the paths and tracks on the sides of the ridge whether or not they are associated with cross-dykes. This is in direct contrast to the bivallate dykes whose ends, where they do continue down the scarps, drop straight down with no attempt to counteract the effects of the slope. Another point is that, whatever the date of D.45-7, the trackways of which they are a part imply traffic to and from the plateau below, a point which is perhaps significant in view of the sparsity of settlement evidence on the ridge top, and the probable prehistoric date of the dykes. The function of three other univallate dykes (D.172 and D.79a and 6) can reasonably be inferred. They lie across spurs, with their ditch on the uphill, southern side facing the ridge top, and apparently were so placed to obstruct and possibly control traffic to and from the ridge and the plateau to the north. D.79a and # are unbroken, spanning the spur but with a terrace-way passing either end. D.107 is now broken west of centre by a hollow-way, probably not through an original entrance, and a terrace-way passes its west end. Good reasons can be postulated why tracks should have descended both spurs and why some sort of control might have been exercised. Firstly, the spurs provide obvious places to descend an other- wise steep and unbroken scarp; secondly, a way past D.107 would lead to Castle Ditches, the largest and strongest hill-fort in the Nadder Valley, and that past D.79a and 4 to a possible ford over the Nadder, perhaps, if one is dealing with a prehistoric way, reflected in the later name Barford (D.B. Bereford). Despite their tactical siting, however, the three dykes would hardly have been large enough as earthworks to fulfil even this limited deflective or obstructive purpose, and it is conceivable that a palisade was incorporated in their banks.® The remaining univallate dykes are not readily comprehensible, though the recent suggestion (Bowen, 1961, 33) that some cross-dykes ‘are probably to be connected with the needs of enclosing stock’, or the similar idea that they may have divided arable from pasture, help explain the sort of context into which they might fit. 50 The primary evidence of date for the whole group of cross-dykes, or for any individual example, is very limited. That they are early is not in doubt, since many are cut by the 18th-century turnpike-road and the line of four dykes is followed exactly or in general by parish boundaries. Two (D.46 and D.107) are apparently referred to in Saxon charters, and D.20 was apparently earlier than a Saxon barrow (Clay, 1927, 64). There is no evidence concerning the Romano-British period, though the absence of settlements, and indeed the almost total lack of any finds from this period, might well suggest that the dykes do not fit here. Similarly, there is no evidence of Belgic activity. On the other hand, D.75 and its northern counter- part are either contemporary with or later than Chiselbury, which, whatever its oddities, is presumably of Early Iron Age construction and (contra Sumner, 1913, and Crawford and Keiller, 1928, 76) likely to be early rather than late in that period; and D.20, flanking the Swallowcliffe Down Early Iron Age settlement, was probably constructed ‘at a time when [it] was extant’ (Clay, 1927, 64), and was earlier than or contemporary with ‘Celtic’ fields on its south-west. D.172 bears the same relation- ship to a ‘Celtic’ field north of its west end; and D.57 1s almost certainly later than round barrows Burcombe 1-4. A date about the middle of the first millennium B.c. would take account of these facts, and a cultural context within Iron Age ‘A’ appears suitable. On the other hand, the evidence is extremely slight, and either the group as a whole or any individual dyke could be earlier or later. This survey has incidentally indicated some of the many problems associated with cross-dykes, and particularly those in Wiltshire where knowledge of them is limited. The suggestions that the bivallate cross-ridge dykes on the Ebble-Nadder ridge are primarily land boundaries, and that the univallate spur-dykes in some way controlled traffic, plus the statement that some univallate dykes are trackways, are made without prejudice as to the functions of similar earthworks elsewhere in the county. Each group of dykes, and each individual dyke, will have to be considered separately before generalizations about cross-dykes in Wessex can be made. This examination of one area supports the suggestion (Bowen, 1961, 32-3) that there are significant regional differences masked under the omnibus term ‘cross-dykes’, in that, ‘spur-dykes’ apart, there are many differences between the dykes considered here and those in Sussex (cf. Gurwen, 1918 and 1951). But these eighteen dykes are only a random sample, which includes, for example, no truly defensive examples of a type known elsewhere in Wessex to protect settlements and to lie outside hill-forts. However, whether or not the suggestions made here are acceptable, it should be clear how much is still to be learnt, in the first place from fieldwork. APPENDIX I The following is a description of the cross-dykes, subdivided into bivallates and univallates, in the order that they occur from west to east along the ridge. ‘The numbers are those given in V.C.H. Wilts. (1957), 249-60.9 A. BIVALLATE CROSS-DYKES D.25. The largest of all the cross-dykes on this ridge, it lies at right angles across the ridge on slight eastern slope east of Whitesheet Hill (947244). Length 470 yds., 51r overall width 70 ft. West bank 30 ft. wide and 4 ft. high, markedly flat-topped; ditch 25 ft. wide, 5 ft. deep and 4 ft. across bottom; east bank 15 ft. wide and 2 ft. high. To south, banks fade on shoulder of scarp, dropping steeply at head of combe, and ditch apparently runs out on slope immediately below them. To north, banks fade and ditch runs out well down scarp, above tracks climbing obliquely to east. Length south of road and both ends well-preserved, rest ploughed almost flat. Hoare (1812), 249, and (wrongly) Station VIII Map; Sumner (1913), no. 10, plan XXXVIII. D.20. At right angles across ridge, here kinking to north-east around head of deep combe on south (966252). Length 400 yds., overall width 50 ft. West bank 10 ft. wide and 1 ft. high, now almost completely ploughed out; ditch 20 ft. wide, 2 ft. deep and 10 ft. across bottom; east bank 20 ft. wide and 2 ft. high. To south-east, banks and ditch stop exactly on shoulder of scarp above c. 30° slope down to combe; to north-west, banks and ditch stop just over shoulder but above steepening slope. Ditch end cupped. South-east end broken by hollowed track (D.23) and north-west end continues just beyond line of a lynchet (D.24). Swallowcliffe Down Early Iron Age settlement lies 200 yds. north-east, and occupation débris, similar to that from the settlement, underlay the east bank of the cross-dyke. ‘Celtic’ fields had modified the structure and shape of the west bank. Ploughed almost flat north of road, but two ends, and ditch and east bank south of road preserved. Hoare (1812), 249, Stn. VIII Map; Sumner (1913), no. 8, plan XX XVII; Clay (1927), 62-4. D.21. At right angles across spur jutting south from ridge (968248), is the only bivallate dyke not on ridge proper, the only bivallate of the four dykes across spurs, and one of the only two cross-dykes on a spur running south from the main ridge. Length 200 yds., overall width 45 ft. North bank 20 ft. wide and 2 ft. high; ditch 15 ft. wide, 2 ft. deep and ro ft. across bottom on centre of spur, narrowing to 4 ft. on east; south bank 15 ft. wide and 1 ft. high. To west, banks stop, but ditch continues slightly beyond banks to go just over shoulder and stop on steep slope to combe. West end and length east of road preserved; length between ploughed. Hoare (1812), Stn. VIII Map; Sumner (1913), no. g and plan XXXVII. D.19. A bivallate and univallate dyke (972252) run together for 70 yds. west-north- west from the shoulder above a steep slope at the head of a combe. Overall width of bivallate 36 ft., both banks and ditch each being 12 ft. wide. North bank 14 ft. high, slightly higher than that on south. Ditch 2 ft. deep and 4 ft. wide across the bottom. Only 6 ft. lies between the south bank of the bivallate and the north edge of the ditch of the univallate to the south. Its overall width is 22 ft. of approximately equal parts of bank and ditch, the former being 1 ft. high and the latter 1 ft. deep and 5 ft. wide across its bottom. Ditch and north bank of bivallate run to shoulder above the scarp, but the south bank and the univallate, which has a cupped end, all stop some 10 yds. short of the edge. Immediately west of this short preserved stretch, the earthworks have been almost flattened by ploughing. On air photographs the dykes diverge, the bivallate apparently becoming univallate, swinging north-west and then south-west to end at a short stretch of surviving bank and ditch which runs out on the shoulder of the spur above a steep slope. The univallate curves round to the north, apparently to join it and form an enclosure. None of this can now be ground-checked, an unfortunate fact since these dykes were unique on this ridge. D.107. Row Ditch (987261), at right angles across ridge south-east of Buxbury. Length 530 yds, overall width c. 45 ft. West bank 15 ft. wide and 2 ft. high, ditch 15 ft. wide, 2 ft. deep and ro ft. across bottom; east bank 15 ft. wide and 2 ft. high. These dimensions become less at both ends. To south, runs over shoulder down comparatively gentle slope towards gradually rising combe floor, the earthwork continuing almost to bottom of scarp before running out. To north, ditch stops on steep slope above Sutton Ivers, the banks having faded out immediately before. ‘Two ends, and length south of road remain, rest ploughed almost flat. ‘Celtic’ fields immediately east of north end on steep slope. Hoare (1812), 249 and Stn. VIII Map; Sumner (1913), no. 6, plan XXXVI; Williams-Freeman (1932), fig. 1. 52 D.75. Two dykes project from Chiselbury hill-fort, one to north (019279) and one to south (018282) (pL. IV). The north one not noted in V.C.H. (1957) under Section D (though noted under Chiselbury, p. 266). “Two short parallel ditches’ south-east of Chiselbury appear under D.75. One is presumably that shown here; the other, 35 yds. west, is presumably a length of slight bank with narrow ditch on both sides (i.e. unlike any other cross-dykes on the ridge), associated at the roadside with a small rectangular earthwork, almost certainly the site of the toll-gate shown by Andrews and Dury, Map of Wiltshire (1773). The south cross-dyke ran from the ditch bounding the small D-shaped earthwork in front of the south-east entrance to Chiselbury to just over the shoulder of the scarp at the head of the combe. There the banks fade away. The ditch also apparently stops, but has in fact been blocked by fill from above. It reappears as a slight ditch only, ro ft. wide, and continues uninterrupted almost down to the foot of the steep scarp, giving a total length of c. 150 yds. Here it is undisturbed; above the scarp, however, it has been ploughed almost flat and much cut by tracks (and surely this is the explanation of the ‘horn’ or splay noted by Hoare and Crawford). Its dimensions were: west bank, width 10 ft., height 1 ft.; ditch, width 18 ft., depth 2 ft. and basal width 10 ft.; east bank, width 15 ft., height 1 ft. The north cross-dyke runs from the north ditch of Chiselbury—ploughing has con- fused the junction between the two—to just over the shoulder of the hill, where the ditch narrows and runs out immediately beyond the end of the banks. Length c. 80 yds.; west bank 12 ft. wide and 6 in. high; ditch 15 ft. wide, 1 ft. deep and 5 ft. across bottom; east bank 12 ft. wide and 6 in. high. There is no evidence that these two stretches of dyke are parts of a true cross-dyke overlain by the hill-fort. Three points vitiate the idea: one, the line of the two lengths means that a considerable bend or kink must be postulated to join them; such a junction within the hill-fort cannot be seen either on the ground or on air photographs; excavation has not found any continuation within Chiselbury. It would therefore seem that the dykes are contemporary with or later than the construction of the hill-fort, and the former seems the more likely. Together with the hill-fort, the dykes draw a line completely across the ridge, and since the dykes on the flanks are not defensive, and the length of hill-fort ditch joining them was presumably not intentionally a trackway, the ditches seem best interpreted as a boundary line. Hoare (1812), 249; Sumner (1913), no. 5, and plan IX; Clay (1927), 62, 65; Curwen (1918), 62; Crawford and Keiller (1928), 7A-7,0Pl, Vil. D.74. At right angles across ridge, if the assumed line linking two short stretches of bivallate dyke is correct (032277). Length 750 yds. To the south, a ditch with slight banks curves down part of the west side of a shallow north-south combe before running out. To the north, only a short length of ditch with slight banks survives, narrowing and running out well down the scarp above Compton Ivers. The whole of the centre portion has been ploughed flat, and only the faintest of traces are visible on air photographs. Colt Hoare described it as a ‘strong bank and ditch’ (1812, 249) (though omitting it from the Stn. VIII Map), while Sumner (1913, no. 4) found it only as a ‘ditch between low banks for a few yards on the north and south scarps’. D.73. Apparently ran at right angles across ridge, but has been obliterated even more than the last (046289). Length 600 yds.? Only two short lengths of ditch, both on the north scarp, and a possible, even shorter, length in dense scrub south-east of the ruins of Compton Hut, survive. Immediately above Burcombe Ivers, into which it runs down a steep slope, the ditch is 20 ft. wide, 4 ft. deep and to ft. across the bottom, with a slight east bank 15 ft. wide and 6 in. high. Immediately south and uphill, it is cut by hollow-ways and the parish boundary, and after a brief reappearance is ploughed flat right up to the copse around Compton Hut. Nor does it survive, even if it ever existed, south of the copse in former arable, where it might have run into the gentle incline down into a narrow combe. Its line is not clear on air photographs, nor could Sumner (1913, 53 no. 3) trace it; but he described it as bivallate, and Hoare (1812, 250) recorded that it ‘traversed’ the ridge. He added that the dyke was the boundary between Compton and Burcombe Downs; its general line is still followed by the parish boundary and has clearly also been used by tracks from the north. D.57. Approximately west to east, parallel to the ridge and across the neck of a spur jutting north, this is the only dyke so sited in the area (067293). ‘The evidence of it now is even slighter than the last two. It has been completely ploughed out, except for slight depressions at either end, on west in former arable, on east in woodland, though it is not certain that the latter is in fact part of the dyke. Length c. 1,000 yds. ? Its general line is clear from Hoare (1812, Stn. VIII Map) who showed without describing it. The ditch can still be seen starting very close to the south side of three barrows (Burcombe 2-4) lying, unusually, on a marked incline above the true break of slope. It ran east to pass close to the south side of another barrow (Bishopstone [S.]9) and then apparently turned slightly north towards the head of a narrow combe running up from the north-east. Its bivallate nature is recorded by Sumner (1913), no. 2. B. UNIVALLATE DYKES These are deliberately not called cross-ridge dykes since not all certainly cross the ridge. ‘Those on the ridge are described first; the three across the two north spurs, apparently very similar to the ‘spur-dykes’ of Curwen (1951, 105-7), are described together at the end. (i) On the ridge: D.45-7. All three lie near together at the west end of the ridge where it curves south and finally south-east as it drops away towards the Ebble valley. 45. Runs south-south-west to north-north-east near highest land on top of what has virtually become a spur from the ridge (940231). Now almost completely ploughed out, the ditch and bank on its east are both about 15 ft. wide. ‘To the north, the dyke emerges from former arable, and from its ditch runs a 10 ft. wide terrace which zig-zags north-east, then north-west, to run round head of combe on north and below a ‘Celtic’ field lynchet to south. There are also slight traces of such fields to east. The terrace either joins or, more probably, is cut by the east end of D.46, and a ditch then continues its line to the north. Total length ¢. 500 yds. 46. Loops across saddle at lowest point of ridge between rises to north and south (938233). Ploughed out except on east where ditch 18 ft. wide and 6 ft. across bottom, with traces of a possible bank to south, before running out on slight slope short of shoulder to combe and immediately after cutting (possibly joining) terrace from D.45. On west, immediately beyond the edge of recent arable, two terraceways join and become the ditch of D.45. Length too yds. One terrace drops down the scarp to the south-west, followed by the parish boundary which has already run along the dyke over the saddle; the other drops more gradually down the scarp to the north-west. From it branch three other tracks; the most northerly, taking off from near the foot of the scarp and climbing steeply as a 3-ft. terrace, becomes, just below the shoulder of the scarp, the ditch of D.47. 4.7. Loops west to east across ridge on south slope (938235). Bank on south 15 ft. wide and 2 ft. high, ditch with similar dimensions. To west, bank fades on shoulder of scarp, the ditch continuing to become a narrow terrace down the slope (see above). To east, bank and ditch ploughed out, but probably continued down east scarp, possibly connecting with former track running up combe from east. Total length c. 500 yds.? These three ‘cross-dykes’ are, therefore, all part of a track system or systems, the ‘cross-dyke’ element being simply those parts of the track where it passes over high ground. The earthworks show, however, that these high parts were dug out and not simply worn down, and suggest that they were used only for foot traffic, presumably mainly cattle and/or sheep. It is perhaps significant that one end of at least two, and possibly all three, ‘dykes’ apparently rests on the high ground as if the tracks were designed to lead there oe and not over it. The facts that a parish boundary follows one track, that part of the terrace on the west scarp has fallen in an old landslip, and that the terrace from D.45 is associated with ‘Celtic’ fields, suggest that the track system including the dykes is of some antiquity; and the whole arrangement, whatever its date, suggests that the people who used it were occupying not only the high downland. Sumner (1913), no. 11 (D.47) and plan XX XVIII. D.41. Runs north from south shoulder of ridge, this south end being only preserved portion (944241). Rest ploughed out, and unknown whether dyke crossed ridge or not. Air photographs suggest not, and that it may have curved back to the south-east. Length 170 yds. or more. ‘To south, where dyke fades out above steep slope at combe head, bank and ditch are both 15 ft. wide and 1 ft. high and deep respectively. Hoare (1812), Stn. VIII Map only. D.22. Runs up north escarpment of ridge and fades out on ridge top in recent arable after curving slightly south-east (963249). May have continued further south towards head of combe (as Hoare’s account suggests), but there is no trace of it doing so. Length 330 yds., overall width 30 ft., from equal parts of bank and ditch, each only 6 in. high and deep respectively. Bank on east, the whole much damaged. To north, between zig-zag of modern road, ditch drops over shoulder and stops immediately above narrow terrace running south-west to north-east aslant the contours. Hoare (1812), 249 and Stn. VIII Map. D.18. Runs west to east across south spur from Swallowcliffe Down (972250). Length 210 yds. To west, which is only part not ploughed, bank and ditch are both 12 ft. wide and 1 ft. high and deep respectively, the width of the ditch bottom being 3 ft. The ditch end, slightly beyond the bank end, is cupped, lying just over shoulder of ridge above steep slope to combe. To east, ditch fades out c. 20 yds. short of shoulder above combe. The earthwork continues the line of D.21 across the next spur to the west, and is c. 250 yds. south of D.19 on the same spur, its west end being similar in position and form to D.19’s similarly surviving west end. (ii) “Spur-dykes’ : This is not simply a term to indicate the position of the dyke, since some dykes already described lie on spurs. The term is used of univallates with the connotation, implied by Curwen’s description of ‘spur-dykes’, that the dykes as well as being on a spur are also associated at their end or ends with trackways. D.172. Runs across Buxbury, a north spur from ridge, nearly at its lowest and narrowest point, facing up the slope towards the ridge (984266). ‘The spur widens and rises to the north. Length 200 yds. Bank 22 ft. wide and 4 ft. high, ditch 18 ft. wide and 3 ft. deep, apparently deeper because cut into slope. The outer southern ditch edge is about level with the top of the bank. Ditch 4 ft. wide at bottom. The earthwork narrows and diminishes to the west, where, immediately above a 10 ft. wide terrace curving along the side of the spur from the ridge to the foot of the escarpment, the bank fades and the ditch widens before running out. Immediately north of the west end is a ‘Celtic’ field, with a positive lynchet on its west running north from the end of the dyke and a negative lynchet—the field’s south edge—at the north foot of the dyke’s bank. The junction between the lynchet bounding the field on the east and the spur-dyke’s bank has been slashed through by a sharp ditch marking the parish boundary between Sutton Mandeville and Swallowcliffe. In the centre the dyke is breached by a wide and deep hollow-way, and to the east it has been flattened by slightly developed ridge and furrow which has caused previous observers to make the dyke much shorter than it is. The east end runs out on the shoulder of the spur above a steep scarp to Sutton Ivers. The ‘Celtic’ fields on the west and north scarps of the spur suggest that formerly much of the spur was similarly cultivated: the irony is that here, on one of the few fairly extensive areas of pasture undisturbed by modern ploughing where evidence of early fields might well have survived, medieval or later ploughing, evidenced by both lynchets and rig, seems to have been widespread. Nevertheless, the purpose of the dyke could well a5) have been both demarcation and protection: the ‘Celtic’ fields behind it would explain the need to delimit and also to force valley-ward traffic down off the spur to descend from the ridge by a terrace-way on the scarp. Hoare (1812), 249; Sumner (1913), no. 7; Williams-Freeman (1932), 26. D.79a and b. Two roughly parallel dykes, 65 yds. apart, run west to east across the narrowing, north-sloping neck of a spur which, to the north, narrows further before flattening and widening out to form a plateau above the River Nadder. The dykes, both facing uphill, i.e. with their banks on the north, are remarkably well-preserved, only the southernmost (79a) being damaged on the west by tracks. 79a is 200 yds. long (066296). The bank is 22 ft. wide, and 2 ft. high, the ditch 18 ft. wide and 3 ft. deep. To the west the bank fades out just over the shoulder of the spur, but the ditch continues down the scarp to splay out and end immediately above a terrace- way swinging round the curve of the scarp to join, behind the northerly dyke, a similar terrace. from the east. In the centre of the dyke the ditch broadens and there is a kink in plan but no break. On the east, the dyke ends immediately over the shoulder and immediately above the terrace already referred to as it skirts a steep drop into the head of a combe. Again the ditch splays out before ending. 79b (066297), only 100 yds. long since the spur has markedly narrowed, has a bank 22 ft. wide and 2 ft. high and a ditch 18 ft. wide and 2 ft. deep. To the west it fades on the shoulder of the spur, here flattening out, immediately above a relatively recent hollow-way slicing through the end of the terrace-way related to the west end of 79a. It is unbroken in the centre on the spur, and to the east, it drops just over the shoulder to end, like 79a, immediately above the east terrace-way, which here has developed into a hollow-way. The purpose of the dykes would seem to have been to divert any traffic between ridge and valley from the centre to the sides of the spur. Hoare (1812), 250; Sumner (1913), no. 1, plan XXV. APPENDIX II Settlements shown on FIG. 1.!° I. SWALLOWCLIFFE DOWN (968254): surrounded by bank and ditch, enclosing roughly circular area of c. 3 acres, with entrance on north-west and circular or semi- circular earthwork to south. Ploughed flat except for traces on south-west and south-east. Excavated 1925-26. Hoare (1812), 249, who notes it was ‘encompassed by a slight earthen agger’; Clay (1925); V.C.H. (1957), 111. R.A.F. vertical air photograph CPE/UK 1811, QT. 2. PRESCOMBE DOWN EAST, EBBESBORNE WAKE (995252): surrounded by ‘Celtic’ fields of group F.81 is a polygonal area of ¢. 12 acres about 550 ft. above O.D. on the top of a spur dropping south to the Ebble valley. The ‘Celtic’ fields are laid off from the edge of this area, which is almost certainly a settlement, clearly respecting it. In part overlaid by broad rig running up from south, and now completely ploughed. Known only from R.A.F. vertical air photograph CPE/UK 1811, 2119. 3. FYFIELD BAVANT DOWN, WEST (001255): Clay (1924); V.C.H. (1957), 68. 4. FYFIELD BAVANT DOWN, EAST (009257): apparently unenclosed, though a small rectangular enclosure lies immediately south-east, and surrounded by ‘Celtic’ fields, it lies on top of a spur dropping south-west and south-east. Clay (1924); V.C.H. (1957), 68. R.A.F. vertical air photograph CPE/UK 1811, 2121. 56 1 The main literature on cross-dykes, and in particular on those dykes and related sites discussed here, is as follows. References to these works, mainly in the text, are given by the abbreviations as indicated: Bowen, 1961: H. C. Bowen, Ancient Fields (1961). Clay, 1924: R. C. C. Clay, An Early Iron Age Site on Fyfield Bavant Down, W.A.M., xm (1924), 457-96. Clay, 1925: R. C. C. Clay, An Inhabited Site of La Téne I Date on Swallowcliffe Down, W.A.M., xm (1925), 59-93, and xLim (1927), 0-7. Cieooes: R. C. C. Clay, Some Prehistoric Ways, Antiquity, 1 (1927), 54-65. Crawford and Keiller, 1928: O. G. S. Crawford and A. Keiller, Wessex from the Air (1928). Curwen, 1918: E. and E. C. Curwen, Covered Ways on the Sussex Downs, Sussex Arch. Colls., uix (1918), 35-75: ; Curwen, 1951: E. C. Curwen, Cross-Ridge Dykes in Sussex, in W. F. Grimes (ed.), Aspects of Archaeology (1951), 93-107. Grinsell, 1958: L. V. Grinsell, The Archaeology of Wessex (1958). Hoare, 1812: R. Colt Hoare, Ancient Wilts. South (1812), 248-50. Sumner, 1913: H. Sumner, The Ancient Earthworks of Cranborne Chase (1913). V.C.H., 1957: Victoria County History of Wiltshire, 1, Pt. I (1957). Williams-Freeman, 1932: J. P. Williams-Freeman, Cross-Dykes, Antiquity, Iv (1932), 24-34- 2 Cf. O. G. S. Crawford, Archaeology in the Field (1953), 117, 187; G. J. Copley, An Archaeology of South-East England (1958), 83-4; Grinsell (1958), 147; E. S. Wood, Field Guide to Archaeology in Britain (1963), 182. 3 Cf. the Grovely Ridge, on the north side of the Nadder, where there is not one cross-dyke. 4 All the dykes discussed here are listed under Section D, Ditches, in V.C.H. (1957), 249-60, and the numbers given to them there are used here. The category letter and number under which other sites are referred to here is also taken from V.C.H. (1957). All the monuments on Fic. 1 which are listed in V.C.H. are also identified in the same way. 5 J. F. Dyer, Dray’s Ditches, Bedfordshire, and Early Iron Age Territorial Boundaries in the Eastern Chilterns, Antig. Journ., xL1 (1961), 32-43; and The Chiltern Grim’s Ditch, Antiquity, xxxvu (1963), 46-9. The dykes themselves are also different as earthworks from those discussed here. Cf. other evidence from recent excavations of cross-dykes in J. S. Wacher, Interim Report on Excavations at Bowden’s Hill, Melcombe Horsey, Proc. Dorset Arch. @ Nat. Hist. Soc., UXXIx (1957), 115, and Litton Cheney Excavations, ibid., Lxxx (1958), 160-77. 6 The air photographs used were R.A.F. CPE/UK 1811, 1121-33, 2105-2124, 2240-6, 3115-25. 7 Cf. the distribution of ‘Celtic’ fields in ‘the area given on the O.S. Map of Roman Britain (1956) and esp. fig. 5; in V.C.H. (1957), Maps VII and VIII; and in Grinsell (1958), Maps IV and V. 8 Little is known about the use of timber in linear earthworks in Wessex, though one suspects it may have been quite common; cf. J. F. S. Stone, A Doubly-Stockaded Early Iron Age Ditch at Winterbourne Dauntsey, W.A.M., xtv1 (1934), 450-3. To test the suggestion that the spur-dykes might have incorporated a palisade, D.107 was sectioned in March 1964, after this paper had gone to press. No palisade, but a ditch much wider and deeper than anticipated was found. See W.A.M. forthcoming. 9 V.C.H. (1957), 260, usefully breaks down the 202 ditches listed in Section D into regional groups, otherwise obscured by the parish basis of the inventory. The Ebble-Nadder Group consists of these listed here, minus nos. 18, 19, 41, 45 and 79a and b which for some unknown reason are omitted. On the other hand, it includes five ditches —nos. 23, 24, 26, 100 and 1o0a—not shown here on Fic. 1. D.23 and 24 are not dykes, nor do they form a ‘kite’ enclosure with D.20, 21 and 22 (contra V.C.H. (1957), 250). The former is a lynchet perhaps of modern origin; the latter is a slight hollow-way beneath a lynchetted former hedge or fence line. The hollow-way narrows and deepens where it cuts through the south end of D.go. D.26 appears to be a double-lynchet track asso- ciated with ‘Celtic’ fields and later cut across and in part followed by the Berwick St. John-Alvediston parish boundary. D.100, east of Chiselbury and evidenced only by reference to a Dic in a Saxon Charter, was looked for and not found. D.100a is not a cross-dyke, and is indeed only ‘a suggestion on air photos’. Although not included in the V.C.H. grouping, D.17 falls within the area of FIG. 1, but is excluded from this survey since it is a system of slight banks and ditches, quite unlike the cross-dykes and certainly not a ‘kite’, probably later than round barrow Alvediston 5 and possibly later than D.25. 10 In reviewing the O.S. Map of Southern Britain in the Iron Age (W.A.M., Lv (1963), 461), I wrongly implied that the proximity of settlements and cross-dykes affected the ‘open’ nature of the former. Obviously the dykes do not enclose the settlements, except in a general way if the thesis about their purpose is correct. (Swallowcliffe Down is, nevertheless, an enclosed site for another reason.) For my ill-advised remark in the review, this survey is a sort of expiation. ay) A CATALOGUE OF PREHISTORIC PLANT REMAINS IN WILTSHIRE by J. D. GROSE and R. E. SANDELL INTRODUCTION THE PUBLICATION in 1956 of Professor Godwin’s great work The History of the British Flora, covered the wide field of the investigation of plant remains in Britain from the end of the Tertiary era until the beginning of historic times. The evidence of plant life there given for the early periods from the Inter-glacial, through the Late-glacial and through much of the present Post-glacial age does not concern us here, for our first record in Wiltshire is Mesolithic. Most of the evidence we have has been obtained from the excavation of archaeological sites in which Wiltshire is rich, and therefore contributions of plant records for the Bronze and Iron Ages are plentiful. Many of the earlier British records cited by Godwin are based on pollen analysis of material recovered from peat bogs and alluvial deposits, in which the outer structure of the pollen grain has been preserved, but sites of this nature in Wiltshire are few. We have, however, some recently-identified pollen and spore records from Neolithic excavations on the chalk where conditions are usually con- sidered to be unsuitable for the preservation of the grains. Such records as an indication of the former occurrence of the species in the localities must be treated with caution. Pollen grains can be air-borne for vast distances and it is only when they are found in quantity that the assumption of local origin can be safely made. In this catalogue these entries are classified as (A) abundant; (B) frequent; (C) scarce. Similarly the presence of charcoal is not conclusive evidence that the provenance of the wood was in the immediate neighbourhood. It is, however, quite improbable that wood for fuel should have been transported from a distance at periods when local supplies were abundant almost everywhere. The present list is extracted from that given in The History of the British Flora, with the addition of some later records, prominent among which are those from the Windmill Hill (Avebury), Overton and Wilsford sites kindly made available by Dr. G. W. Dimbleby of the Commonwealth Forestry Institute, Oxford. An important investigation not included in this list (owing to the mixture of pollen of unknown age with that from the modern flora) is that carried out on Overton Down in 1960. This has now been published in The Experimental Earthwork on Overton Down (British Association, 1963), and the interesting results obtained there by Dr. Dimbleby should be closely studied. Material identified from the Wilsford Shaft is prehistoric but cannot be allocated to a particular period. We are indebted to Professor Godwin for permission to use his list. The 58 comments are mostly ours, but are prompted by the History, and it is hoped are a true assessment of the views there expressed. LIST Ranunculus repens Creeping Buttercup Lidbury Camp Fruit Iron Age Cunnington, 1917 Papaver somniferum Opium Poppy Fifield Bavant Seed Iron Age Helbaek, 1953! This record must be considered doubtful. Helianthemum chamaecistus Rock Rose Wilsford Shaft Pollen (C) Ashbee Viola canina Dog Violet Fifield Bavant Seed Iron Age Helbaek, 1953 Melandrium album White Campion Fargo Plantation Seed Early Bronze Age Stone, 1938 The first British record of this weed and an indication of early cultivation. Tilia sp. Lime Downton Charcoal Late Neolithic Rahtz, 1962 Tilia cordata| Small-leaved Lime Windmill Hill c.c.? Pollen (B) Neolithic Smith Windmill Hill 1.b.3 Pollen (B) Neolithic Ashbee & Smith Linum usitatissimum Common Flax Windmill Hill Seed impression Neolithic Helbaek, 1953 Winterbourne Stoke Seed impression Middle Bronze Age Helbaek, 1953 The cultivation of flax was probably widespread on the Wiltshire Downs in the Bronze Age. Tlex aquifolium Holly All Cannings Charcoal Iron Age Cunnington, 1923 Swallowcliffe Down Charcoal Iron Age Clay, 1925 Boscombe Down West Charcoal Iron Age Richardson, 1951 Bury Wood Camp Charcoal Iron Age Grant King, 1962 Rhamnus cathartica Buckthorn Downton Charcoal Bury Wood Camp Charcoal Late Neolithic Iron Age Rahtz, 1962 Grant King, 1962 Aesculus sp. Horse Chestnut ? Avebury Charcoal Late Neolithic Gray, 1935 A very doubtful identification. Acer campestre Maple Downton Charcoal Late Neolithic Rahtz, 1962 Down Farm, Pewsey — Charcoal Early Bronze Age _Vatcher, 1960 Wylye Charcoal Bronze Age Passmore, 1940 All Cannings Charcoal Iron Age Cunnington, 1923 Bury Wood Camp Charcoal Iron Age Grant King, 1962 éy) The wood and charcoal of the Maple cannot normally be distinguished from that of the Sycamore (A. pseudo-platanus), but the latter species, although a native of the mountainous regions of central Europe, is believed to have been introduced into Britain only in comparatively recent times. Sarothamnus scoparius Broom Windmill Hill c.c. Charcoal Neolithic Smith Medicago lupulina Black Medick Lidbury Camp Seed Iron Age Cunnington, 1917 There is only one earlier British record. Vicia tetrasperma Smooth Tare Fifield Bavant Carb. seed Iron Age Helbaek, 1953 The smooth tare occurs sometimes as a weed of cultivated or disturbed ground, but is more frequently a plant of grassy sheltered habitats. There is one earlier record for Britain. Prunus spinosa Blackthorn Windmill Hill c.c. Charcoal Neolithic Smith Windmill Hill 1.b. Charcoal Neolithic Ashbee & Smith ? Downton Charcoal Late Neolithic Rahtz, 1962 West Overton Charcoal Early Bronze Age Smith & Simpson Beckhampton Charcoal Late Neolithic Young, 1950 Rubus idaeus Raspberry Fargo Plantation Fruit stone Early Bronze Age Stone, 1938 The Raspberry still grows at Fargo Plantation. Poterium sanguisorba Salad Burnet Wilsford Shaft Pollen (B) Ashbee Sorbus aucuparia’ Mountain Ash Swallowcliffe Down Charcoal Iron Age Clay, 1925 Some doubt must be attached to this record; the charcoals of the Rosaceae in which are included Sorbus, Pyrus, Malus and Crataegus cannot readily be distinguished from each other. The Mountain Ash is unlikely to grow on the chalk of Swallowcliffe Down and transport for even a few miles from the west is hardly credible when abundant supplies of other woods were at hand. Malus sylvestris Crab Apple Windmill Hill Seed impression Neolithic Helbaek, 1953 ‘The presence of apple pip impressions on no less than six separate shards at Windmill Hill warrants the assumption that the fruits were gathered for food in fair quantity.’ Godwin. Crataegus sp. Hawthorn Windmill Hill c.c. Charcoal Neolithic Smith Avebury Charcoal Late Neolithic Gray, 1934 Ratfyn Charcoal Late Neolithic Stone, 1935 West Overton Charcoal Early Bronze Age Smith & Simpson Swallowcliffe Down Charcoal Iron Age Clay, 1925 The uncertainty of charcoal identification in this genus is mentioned under Sorbus aucuparia above. There remains the further problem of specific naming. We have in Wilt- shire two hawthorns, C. monogyna, the plant of hedgerows which is common over the whole 60 county, and C. oxyacanthoides, which is more particularly a woodland species and is now known only in the north-west. Thus it is probable that the records should be referred to C. monogyna, the only species growing in the districts concerned. There is, however, the possibility that under the more densely wooded conditions of earlier times C. oxyacanthoides could have been present. (Unidentified pollen (B) of a member of the Umbelliferae has been found at Windmill Hill and at Wilsford.) Hedera helix Ivy Windmill Hill c.c. Pollen Neolithic Smith Windmill Hill |.b. Pollen Neolithic Ashbee & Smith Bury Wood Camp Charcoal Iron Age Grant King, 1962 Wilsford Shaft Pollen Ashbee Sambucus nigra Elder Windmill Hill c.c. Charcoal Neolithic Smith All Cannings Charcoal Iron Age Cunnington, 1923 Viburnum sp. Bury Wood Camp Charcoal Iron Age Grant King, 1962 This was probably V. lantana, the Wayfaring Tree, which is common in this district. The other possibility is V. opulus, the Guelder Rose, which prefers damper conditions and is less likely to have grown on the summit of a hill. Galium aparine Goosegrass Lidbury Camp Fruit Iron Age Cunnington, 1917 Winkelbury Fruit Iron Age Helbaek, 1953 Succisa pratensis Devil’s Bit Scabious Windmill Hill c.c. Pollen (C) Neolithic Smith Centaurea nigra Hardheads Wilsford Shaft Pollen (C) Ashbee Artemisia sp. Windmill Hill 1.b. Pollen (C) Neolithic Ashbee & Smith Wilsford Shaft Pollen (C) Ashbee Lapsana communis Nipplewort Fifield Bavant Fruit Iron Age Helbaek, 1953 Calluna vulgaris Ling Windmill Hill c.c. Pollen (C) Neolithic Smith Fraxinus excelsior Ash Windmill Hill c.c. Charcoal Neolithic Smith Ratfyn Charcoal Late Neolithic Stone, 1935 Down Farm, Pewsey Charcoal Early Bronze Age _-Vatcher, 1960 All Cannings Charcoal Iron Age Cunnington, 1923 Fifield Bavant Charcoal Iron Age Clay, 1924 Bury Wood Camp Charcoal Iron Age Grant King, 1962 The Ash thrives particularly on the shallow soils of the chalk downs and is often dominant in woodland. It is, however, somewhat intolerant of shade and under the less open conditions of the deeper soils is largely replaced by other species. Professor Godwin suggests that its growth may have been encouraged by forest clearance. 61 Ligustrum vulgare Privet Windmill Hill 1.b. Pollen (B) Neolithic Ashbee & Smith Veronica hederifolia Ivy-leaved Speedwell Great Bedwyn Seed ?Middle Bronze Age Meyrick, 1949 The first British record. Galeopsis sp. Hemp Nettle Fifield Bavant Fruit Iron Age Helbaek, 1953 This is probably G. tetrahit which is common in the district. Plantago spp. Windmill Hill c.c. Pollen (A) Neolithic Smith Windmill Hill 1.b. Pollen (B) Neolithic Ashbee & Smith Overton Down Pollen (A) Bronze Age Birmingham Wilsford Shaft Pollen (A) Ashbee (Unidentified pollen (B) of a member of the Chenopodiaceae has been found at Windmill Hill and at Wilsford.) Polygonum convolvulus Black Bindweed Fargo Plantation Fruit Early Bronze Age Helbaek, 1953 A weed of agriculture. Rumex sp. Windmill Hill 1.b. Pollen (C) Neolithic Ashbee & Smith Lidbury Camp Fruit Iron Age Cunnington, 1917 Wilsford Shaft Pollen (C) Ashbee Ulmus sp. Elm Windmill Hill 1.b. Pollen (C) Neolithic Ashbee & Smith Down Farm, Pewsey Charcoal Early Bronze Age Vatcher, 1960 Bury Wood Camp Charcoal Iron Age Grant King, 1962 Wilsford Shaft Pollen (C) Ashbee The charcoals of the different Elm species cannot be distinguished from each other, but it is probable that the Bury Wood Camp record should be referred to the Wych Elm (U. glabra) which attains its finest growth in that part of the county. Urtica urens Nettle Lidbury Camp Fruit Iron Age Cunnington, 1917 Urtica sp. Wilsford Shaft Pollen (C) Ashbee Betula sp. Downton Charcoal Neolithic Rahtz, 1962 Windmill Hill 1b. Pollen (C) Neolithic Ashbee & Smith Woodhenge Wood Late Neolithic Cunnington, 1929 Highfield Charcoal Iron Age Stevens, 1934 Records of Birch wood or charcoal cannot be assigned to either B. verrucosa or B. pubescens. Both species are now common in the lowlands of Wiltshire but scarce on the downs. From the paucity of remains discovered in excavated prehistoric sites, most of which are on the downs, it seems that the general distribution is unaltered. 62 of Salisbury Plain. This has been done in an attempt to resolve many of the ambiguities and inconsistencies which have arisen relating to the precise attribu- tion of some of the finds. Whilst the catalogue of objects illustrating the Windmill Hill and Secondary Neolithic Cultures is not exhaustive, every find in the collections from the Beaker period up to and including the Late Bronze Age, incorporating the famous Stourhead Collection, has been described and illustrated, with the exception of unlocated flint-work, and some miscellaneous stone and bone artefacts. There are over 50 pages of line drawings; many of the objects are now illustrated for the frst time since their discovery by Cunnington and Hoare in the initial years of the nineteenth century. The third section comprises a Concordance of References in which the sites recorded in the Cunnington Letters and Hoare’s Notes are listed together with the corresponding references in Ancient Wiltshire. This should be of immense value to scholars and students alike of British prehistory. The volume is completed by bibliographies, a site Index and a number of halftones illustrating decorative detail, greatly enlarged, on selected objects from the Stourhead Collection. The latter should also prove of great interest as they demonstrate in striking fashion the technical achievements of Bronze Age craftsmen. Published by Tue WittsHIRE ARCHAEOLOGICAL AND NATURAL History SOCIETY Price 255. net Publication end of July 1964 Crown 4t0, 144 pages including 51 pages of line illustrations, and 11 half-tone plates. Specimen page of illustrations NEOLITHIG AND BRONZE AGE COLLECTIONS Sr TC en Ee an rE Ce uh e vee We WO Eiko om 176 Scale: 171-73 (%); Rest (3) Specimen page NEOLITHIG AND BRONZE AGE COLLECTIONS 45 outline at the heel of the knife dagger; traces of vertical wood grain suggest the haft was of that material. The bone fragments are part of the hollow long bone of a swan (Cygnus cygnus) with one end smoothly finished. When excavated it was 7-8 inches long, and had a single perforation near the centre. The bone is at present in two pieces; there is a single, crudely made hole at the centre, and a possible indication of a further hole one-third from the narrow end. It has been conjectured that it is a bone flute. Longest dagger L. 7-8 in. (20 cm.). Dagger Anal. No. 7. D.M. 1042-6. (A.W., I, 199, pl. xxiv; Antiquity, XXXIV, 1960, 6-13.) 168-78. Grave group. Primary inhumation in bowl barrow (Bush Barrow), Wilsford G.5. NGR. 11644128. P., 53. 168. Lozenge-shaped plate of sheet gold with incised ornament. 169. Copper dagger. 170. Bronze dagger. 171. Thirty-three rivets or fragments of rivets of bronze. 172. Large bronze rivet. 173. Small hook-like bronze object. 174. Three cylindrical bone mounts and two end-pieces or ferrules of bone. 175. Polished mace-head with cylindrical perforation. 176. Belt-hook of hammered gold with incised ornament. 177. Small lozenge-shaped plate of sheet gold with incised ornament. 178. Flanged axe of copper or bronze. The large gold plate has a single perforation at each end of the long axis, and a single broad groove around the edge of the overlap. The small plate has a single groove around the edge of its overlap. Adhering to the blade of the smaller dagger are fragments of the wooden outer casing and the inner leather lining of the sheath. In the wood of the haft adhering to the upper part of the blade is a row of seven holes each 0:75 mm. in diameter. Also discovered, but not illustrated, were nine small wood fragments containing an inlay of minute gold pins, and several thousand loose specimens, each 1 mm. long. All these represent the remains of the decorated haft of this dagger; the pattern, suggested by the fragments, is that of running chevrons. Three of the wood fragments, presumably part of the pommel, have an additional inlay set in their ends. The hilt of the large dagger is represented by fragments of wood adhering to the heel of the blade. Traces of the wooden sheath are visible on the lower portion of the blade. The mace-head is of a rare type of limestone consisting mainly of the Devonian Stromatoporoid (Amphipora ramosa). Cunnington (Cunnington MSS., x, 12) says of this mace-head that it ‘had a wooden handle with a neat ornament of brass on the top, the latter being fastened to the top of the handle by a brass pin’. The hook and plate of the belt-hook were manufactured separately, the ends of the former being hammered over to grip the plate. There is a single perforation in each corner of the plate through the overlapping portion. GUIDE CATALOGUE of the NEOLITHIC and BRONZE AGE COLLECTIONS DEVIZES MUSEUM This book is the fourth in the well-known series of printed Catalogues published by the Wiltshire Archaeological and Natural History Society, and is intended to replace the prehistoric sections of the former publications (Devizes Museum Catalogues Pt. I (1896) and Pt. II (1911 and 1934), now out of print, and largely out of date. Professor Stuart Piggott who contributes a Foreword to the Catalogue rightly emphasizes that ‘the Devizes Collections have an international fame among archaeologists and prehistorians, especially among those concerned with those critical and formative centuries of ancient barbarian Europe, the earlier second millennium B.c.’. Not only is this material famous, but it is indispensable for an understanding of this period, and the essential purpose of the new edition is to present a corpus of antiquities dating from the Early Neolithic up to and including the Late Bronze Age, now preserved in Devizes Museum. The Catalogue is divided into three sections. The first consists of Introductory chapters providing summary accounts of the Neolithic and Bronze Ages in southern Britain, and a description of the Colt Hoare-Cunnington partnership. These are intended particularly for the interest of the general reader in the hope that the Catalogue may thereby attract a wider sale. The second section is the illustrated Catalogue, and it should be stressed that in the compilation of this fourth edition, every effort has been made to base the descriptions on a close study of the original sources, in particular the manu- script letters of William Cunnington to Sir Richard Colt Hoare, and Ancient Wiltshire, vols. I and II, the classic account of their joint researches on the barrows Alnus glutinosa Windmill Hill c.c. Windmill Hill 1.b. Downton Wilsford Shaft Alder Pollen (B) Pollen (B) Charcoal Pollen (C) and wood Carpinus betulus Hornbeam Downton Charcoal ?Woodhenge Charcoal Avebury Charcoal ?Down Farm, Pewsey Charcoal Beckhampton Charcoal Neolithic Neolithic Late Neolithic Late Neolithic Late Neolithic Late Neolithic Early Bronze Age Late Neolithic Smith Ashbee & Smith Rahtz, 1962 Ashbee Rahtz, 1962 Cunnington, 1929 Gray, 1934 Vatcher, 1960 Young, 1950 The Hornbeam is now known in Wiltshire only as a planted tree. Its charcoal closely resembles that of the hazel and some of the records are suspect. It should, however, be noted that Gray mentions both species for Avebury. Corylus avellana Avebury Windmill Hill c.c. Windmill Hill 1.b. Ratfyn West Overton Overton Down Fifield Bavant Highfield Bury Wood Camp Swallowcliffe Down Wilsford Shaft Quercus sp. Oak Downton Windmill Hill c.c. Windmill Hill Lb. Ratfyn Woodhenge Down Farm, Pewsey West Overton Overton Down All Cannings Boscombe Down West Fifield Bavant Highfield Swallowcliffe Down Hazel Wood and charcoal Pollen (A) and charcoal, in- cluding nut- shells Pollen (A) Charcoal Pollen (C) and charcoal, in- cluding nut- shell Pollen (C) Charcoal Charcoal Charcoal Wood and charcoal Pollen (B) Charcoal Pollen (C) and charcoal Pollen (B) Charcoal Wood and charcoal Charcoal Charcoal Pollen (C) Charcoal Charcoal Charcoal Charcoal Charcoal Late Neolithic Neolithic Neolithic Late Neolithic Early Bronze Age Bronze Age Iron Age Iron Age Iron Age Iron Age Mesolithic Neolithic Neolithic Late Neolithic Late Neolithic Early Bronze Age Early Bronze Age Bronze Age Iron Age Iron Age Iron Age Iron Age Iron Age Gray, 1934 Smith Ashbee & Smith Stone, 1935 Smith & Simpson Birmingham Clay, 1924 Stevens, 1934 Grant King, 1962 Clay, 1925 Ashbee Higgs, 1959 Smith Ashbee & Smith Stone, 1935 Cunnington, 1929 Vatcher, 1960 Smith & Simpson Birmingham Cunnington, 1923 Richardson, 1951 Clay, 1924 Stevens, 1934 Clay, 1925 63 Bury Wood Camp Charcoal Iron Age Grant King, 1962 Wilsford Shaft Pollen (B) and wood Ashbee Most or all of these records are referable to the Common Oak Q.robur. The Durmast Oak, Q. petraea is extremely rare in Wiltshire and is not likely to have grown on the cal- careous soils of the localities listed. Quercus lex Holm Oak All Cannings Charcoal Iron Age Cunnington, 1923 It has been suggested that this wood had been imported from Europe as some manu- factured object. Castanea sativa Sweet Chestnut ?Down Farm, Pewsey Charcoal Early Bronze Age _-Vatcher, 1960 The Sweet Chestnut is thought to have been introduced by the Romans. The Down Farm specimen may possibly have been ‘part of a modern fence post’. Fagus sylvatica Beech Boscombe Down West Charcoal Tron Age Richardson, 1951 Wilsford Shaft Pollen (C) Ashbee The meagre prehistoric evidence of Beech in Wiltshire strongly suggests that it was not, as in several counties, a common tree. Salix sp. Willow Highfield Charcoal Iron Age Stevens, 1934 Salix alba White Willow Swallowcliffe Downs — Charcoal Iron Age Clay, 1925 ‘Charcoal determinations (of Salix) suffer from the natural untrustworthiness of such material in species diagnosis.’ Godwin. S. alba is only doubtfully native in Wiltshire and is most unlikely to have grown on Swallowcliffe Downs. Salix or Populus sp. Willow or Poplar Down Farm, Pewsey Charcoal Early Bronze Age Vatcher, 1960 Bury Wood Camp Charcoal Iron Age Grant King, 1962 Phleum nodosum Cat’s ‘Tail Fifield Bavant Carb. fruit Iron Age Helbaek, 1953 The only prehistoric British record. Avena fatua Wild Oat Fifield Bavant Carb. fruit Iron Age Jessen & Helbaek, 1944 Then, as now, a weed of cornfields, probably having been introduced from the Continent. Arrhenatherum tuberosum Onion Couch Rockley Down Carb. tubers and stems Late Bronze Age — Allison and Godwin, 1949 About fifty tubers were found in association with barley grain; possibly the plant may have been collected for food. The Onion Couch, a form of False Oat (A. elatius) with bulbous stem bases is characteristically a plant of chalky cultivated fields. In the History of the British Flora, Professor Godwin illustrates the Rockley Down carbonized tubers, together with fresh ones. 64 Anisantha sterilis Barren Brome Fifield Bavant Carb. fruit Iron Age Clay, 1924 Winkelbury Carb. fruit Iron Age Helbaek, 1953 Bromus spp. Brome Lidbury Camp Fruit Iron Age Cunnington, 1917 Fifield Bavant Carb. fruit Iron Age Clay, 1924. Winkelbury Carb. fruit Iron Age Helbaek, 1953 Identification of the fruit at species level cannot be made, but it is probable that it belongs to B. secalinus or B. mollis. CEREALS Avena sativa Cultivated Oat Fifield Bavant Carb. fruit Iron Age Clay, 1924 Avena strigosa s.1. Black Oat Fifield Bavant Carb. fruit Iron Age Jessen & Helbaek, 1944 The cultivation of Oats in Britain appears only to have commenced in the Iron Age. Hordeum polystichum Six-rowed Barley ‘Naked Barley’, chaff not adhering to grain. Windmill Hill Fruit impression Neolithic Helbaek, 1953 . Winterbourne Stoke — Fruit impression Late Neolithic Helbaek, 1953 Collingbourne Ducis — Fruit impression Middle Bronze Age Helbaek, 1953 Tan Hill Fruit impression Late Bronze Age Jessen & Helbaek, 1944 Rockley Down Carb. fruit Late Bronze Age — Allison & Godwin, 1949 Fifield Bavant Carb. fruit Iron Age Helbaek, 1953 ‘Hulled Barley’, chaff adhering to grain ?Windmill Hill Fruit impression Neolithic Helbaek, 1953 Winterbourne Stoke — Fruit impression Late Neolithic Helbaek, 1953 Beckhampton Fruit impression Middle Bronze Age Helbaek, 1953 Durrington Fruit impression Middle Bronze Age Helbaek, 1953 Amesbury Fruit impression Middle Bronze Age Helbaek, 1953 Collingbourne Fruit impression Late Bronze Age — Helbaek, 1953 Fifield Bavant Carb. fruit Iron Age Helbaek, 1953 Winkelbury Carb. fruit Iron Age Helbaek, 1953 Hordeum hexastichum Erect six-rowed Barley Swallowcliffe Down Fruit Iron Age Jessen & Helbaek, 1944 Fifield Bavant Carb. fruit Iron Age Clay, 1924 Hordeum vulgare Bere Lidbury Fruit Iron Age Cunnington, 1917 Fifield Bavant Carb. fruit Iron Age Jessen & Helbaek, 1944 Barley constituted the major corn crop in Britain throughout the Bronze Age, but took second place to wheat in the Iron Age. Triticum monococcum Small Spelt Windmill Hill Spike impression Neolithic Helbaek, 1953 Triticum dicoccum Emmer Windmill Hill Spike and fruit impression Neolithic Helbaek, 1953 65 Upavon Spike impression Late Neolithic Helbaek, 1953 Beckhampton Spike impression Middle Bronze Age Helbaek, 1953 Fifield Bavant Carb. fruit Iron Age Clay, 1924 Triticum compactum Club Wheat Fifield Bavant Carb. fruit Iron Age Jessen & Helbaek, 1944 Material from Fifield Bavant may have included a form of Bread Wheat (T. vulgare). Triticum spelta Spelt Fifield Bavant Carb. spike and carb. fruit Iron Age Jessen & Helbaek, 1944. Casterley Camp Spike impression Iron Age Helbaek, 1953 Highfield Spike impression Iron Age Helbaek, 1953 Winkelbury Carb. fruit Iron Age Helbaek, 1953 The cultivation of wheat in Wiltshire appears to have been on a small scale during the Bronze Age, but to have increased considerably in the Iron Age. Small Spelt and Emmer were grown in the Neolithic period, the former apparently only rarely, the latter abundantly and continuing into historic time. It is thought that Club Wheat and Bread Wheat were cultivated to a very limited extent. Spelt was grown more commonly but was local, the pre-Roman records with two exceptions being confined to the counties of Somerset, Dorset and Wiltshire. Secale cereale Rye Fifield Bavant Carb. fruit Iron Age Helbaek, 1953 Winkelbury Carb. fruit Iron Age Helbaek, 1953 There is but one other Iron Age record for Britain and the cultivation of Rye as a crop must have been almost unknown. Taxus baccata Yew Boscombe Down Wood and leaf Bronze Age Newall, 1931 Pinus sylvestris Scots Pine Windmill Hill c.c. Pollen (C) Neolithic Smith Windmill Hill 1.b. Pollen (B) Neolithic Ashbee & Smith Overton Down Pollen (C) Bronze Age Birmingham Pteridium aquilinum Bracken Windmill Hill c.c. Spores (A) Neolithic Smith Windmill Hill 1.b. Spores (A) Neolithic Ashbee & Smith Overton Down Spores (C) Bronze Age Birmingham Wilsford Shaft Spores (A) Ashbee The abundance of Bracken spores in the Neolithic soils of Windmill Hill cannot be easily explained. Bracken is a pronounced calcifuge and if it did, in fact, formerly grow in great quantity on the chalk downs, soil and/or climate factors must have been more favourable to it than they are now. A former more extensive clay-with-flints area with a milder climate and some shelter in open woodland or scrub could have provided the necessary conditions for its growth. A second suggestion has been made that bracken was carried to the hills for some agricultural reason or as litter for animals. There remains the possibility that the spores were blown by the prevailing south-westerly winds from the greensand soils at the foot of the downs (where bracken still grows abundantly) and were trapped and held on the higher ground. The recent discovery that bracken spores are frequent in ‘modern’ soils on Overton Down and Overton Hill seems to support the air-borne theory. 66 Polypodium vulgare Common Polypody Windmill Hill c.c. Spores (C) Windmill Hill 1.b. Spores (B) Wilsford Shaft Spores (A) 1 Proceedings of the Prehistoric Society for 1952, published 1953. Neolithic Neolithic Smith Ashbee & Smith Ashbee 2 Causewayed Camp. 3 Long Barrow. REFERENCES Allison, J., and Godwin, H., New Phytologist, XLVI (1949), 253- Clay, R. C. C., W.A.M., xin (1924), 457. Clay, R. C. C., W.A.M., xiut (1925), 59. Cunnington, M. E., W.A.M., xu (1917), 12. Cunnington, M. E., All Cannings Cross (1923). Cunnington, M. E., Woodhenge (1929). Dimbleby, G. W., personal communication. Grant King, D., W.A.M., vm (1962), 185. Gray, H. St. G., Archaeologia, LXxxtv (1934), 99- Helbaek, H., Proceedings of the Prehistoric Society, n.s., XVII (1952), 194. Higgs, E., Proceedings of the Prehistoric Society, n.s., XXV (1959), 209. Jessen, K., and Helbaek, H., A. danske vidensk. Selsk., Wt (1944). Meyrick, O. (Allison, J., and Godwin, H.), New Phytologist, XLvim (1949), 253. Newall, R. S., W.A.M., xLv (1931), 432. Passmore, A. D., W.A.M., xLrx (1940), 117. Overton Down, the Experimental Earthwork (1963), ed. Jewell, P. A. Rahtz, P. A., W.A.M., tv (1962), 116. Richardson, K. M., W.A.M., Liv (1951), 123. Stevens, F., W.A.M., xLv1 (1934), 579. Stone, J. F. S., W.A.M., xLvm (1935), 55+ Stone, J. F. S., W.A.M., xivi (1938), 357- Vatcher, F. de M., W.A.M., Lvu (1960), 339. Young, W. E. V., W.A.M., Li (1950), 311. 67 EXCAVATION OF THREE ROMAN TOMBS AND A PREHISTORIC PIT ON OVERTON DOWN by I. F. SMITH and D. D. A. SIMPSON INTRODUCTION THE MONUMENTS to be described below lay on a tract of downland on Overton or Seven Barrow Hill, in the parish of West Overton, which was first brought under cultivation in 1962. When the examination was carried out on behalf of the Ministry of Public Building and Works in the summer of that year, the sites and the spaces between them had been left as an unploughed island in the midst of a standing crop. Now, after total excavation, they too are under plough. Before excavation the three sites were marked by small mounds, so low as to be almost imperceptible under their cover of rough downland vegetation. Narrow bands of more luxuriant growth round the circumferences pointed to the existence of ditches. The mounds appear in the list of (presumptively prehistoric) bowl barrows in V.C.H. Wilts., 1, Pt. 1, as West Overton 6, 6a and 7. As will be seen, they proved to be neither prehistoric in date nor barrows in the normal sense. ‘They were situated on the flat top of the hill, at about 575 ft. O.D., and were accurately aligned in a north-south direction, parallel with and about too ft. to the east of The Ridgeway, and almost at right angles to the Roman road from Cunetio to Verlucio, which passes about 150 ft. to the south of No. 6 (see ric. 1). The National Grid references are SU/11936832/34/37. The mounds had all been opened by Colt Hoare, who described them as follows: ‘In our way to this barrow [his No. 6; V.C.H. Wilts., West Overton 8] over a fine piece of verdant down, we traverse an elevated ridge of the Roman road leading eastward to Cunetio near Marlborough, and westward to the Stations of Verlucio and Aquae Solis. Some very flat barrows then occur, which, on opening, proved to have been most effectually robbed in former times, not even the fragment of a bone having been left behind; but from the shape and dimensions of the cists, it is evident that cremation had been performed in all of them.’! No. 6 (the southern mound) and No. 7 (the northern) were also dug into? by Thurnam: ‘Attention was next directed to the miniature mounds in this group, immediately to the north of the Roman road, and to the south of the barrow numbered 6, on the plan of Sir R. C. Hoare. It was extremely doubtful, from their form and trifling elevation, whether any of these were sepulchral. In the most southern nothing whatever was found. The second was not examined. In the third, of rather larger size, at a depth of less than two feet, were a few bits of decayed bronze, of doubtful purpose, and two or three fragments of black pottery, with a thin coin the size of half-a-crown, 68 which fell to pieces on removal. There were also some ashes and slight traces of burnt bones.’3 In 1962 the mounds were excavated by the quadrant method and the inter- vening spaces, 40 ft. between Nos. 6 and 6a and 44 ft. between Nos. 6a and 7, were investigated, with negative results, by means of north-south cuttings. The geological solid is Upper Chalk, the surface of which is broken by a series of long parallel natural channels filled with decayed chalk and yellowish clay. The topsoil is a rendsina with a maximum thickness of 1 ft. It will be convenient to describe the sites in order from north to south, starting with No. 7, the largest and the only one to produce conclusive evidence of date. A prehistoric pit, uncovered in the course of excavating No. 6a, is described in an appended note, together with the few prehistoric objects that had incidentally been incorporated in the mounds. WEST KENNETT a RIDGE WAY THE SANCTUARY Fic. i Location map. 69 Based upon the Ordnance Survey map, with the sanction of the Controller of H.M. Stationery Office. Crown Copyright reserved NO. 7 The mound reached a maximum height of 2 ft. and was surrounded by a ditch with an external diameter of 23 ft. (Fics. 2, 4; pL. Va). The ditch had vertical sides and a flat bottom, and varied in width from 1-6 to 2 ft., and in depth from 1 to 2 ft. Removal of 2 to 3 in. of fill from the top exposed 54 cylindrical pockets of soft brown soil which extended to the bottom of the ditch and were surrounded by a firm packing of chalk rubble, occasionally supplemented by flint nodules and pieces of sarsen. The soft soil was clearly replacement material, filling the spaces formerly occupied by the bases of timber uprights; the ditch had thus been the bedding trench for a circular setting of posts (pL. Vb). Charcoal recovered from the bottom of one cavity proved to be oak,4 and may indicate charring of the base of the post as a preservative measure. Despite the pains taken to bed the posts securely, little effort seems to have been made to provide timbers of uniform girth, for the diameters of the cavities ranged from 0-6 to 1-8 ft. The spacing had also been some- what irregular, but there was no definite interruption in the ring and the only anomaly was on the east side, where the intersection of two cavities suggests that one of the posts may have been replaced at some time after the original construction was finished. An unweathered sherd of Romano-British coarse pottery, measuring about 2 in. by 14 in., was found firmly embedded in the chalk packing in the north- west quadrant at a depth of 1°5 ft. During the preparation of the bedding trench, the loosened chalk had been heaped round its inner edge; part was then returned as packing for the posts and the remainder left as a low bank (see sections, Fic. 4). Traces of the old land surface were detectable only where covered by this chalk bank. On this buried surface lay part of a tile- or brick-shaped object which, though not readily identifiable, appears more likely to be attributable to the Roman period than to any earlier or later one.5 The remainder of the mound consisted of topsoil. Owing to the numerous previous disturbances in the central area (at least two of which were traceable in the sections), as well as to spreading and sorting by natural agencies, it was not possible to determine the original relationship between this material and the chalk bank. It is evident, however, that the mound had never been much larger than at the time of the excavation. At the centre was a circular pit 2 ft. in diameter, with vertical sides cut 2 ft. deep in the chalk. Its contents had been completely disturbed, and during one of the previous investigations, probably that made by Thurnam, a deep cut into the chalk natural had destroyed the western arc. Scattered through the fill of the pit, the soil of the mound, the top of the ditch and, to a smaller extent, the soft fill of the post-cavities, were numerous fragments of bronze, a few pieces of cremated bone, a small quantity of broken animal bone (ox and sheep or goat), and sherds of Romano-British pottery. The pieces of bronze ranged from pin-head size to a fused lump, 65 mm. in length; some of the larger pieces are distorted by heat and there were a number of solidified globules. The recognizable fragments appear to have come from bronze vessels and bronze- 70 Dass 2OVERTON G7: PLAN. LIMIT OF EXCAVATION “=~ Ori =f 5 10 0 1 4 ee FEET — = METRES Fic. 2 71 WEST OVERTON > G@a-PEAN: LIMIT OF — EXCAVATION Nia is 1 PREHISTORIC PIT Fic. 3 72 mounted caskets; they are described in detail on pp. 77-79 and the more impor- tant ones are illustrated in Fic. 6: 1-5. Of the eleven fragments of cremated bone recovered, ten are of human origin (see report, p. 81). It may therefore be inferred that a cremation deposit had been accompanied by the remains of metal objects gathered up from the site of the pyre. Whether or not all had originally been placed together in the central pit, later to be thrown out by robbers, remains uncertain; as Mr. Jessup points out on p. 79, material from the pyre was often accorded somewhat casual disposal in Roman barrows. In the circumstances it is not possible to determine whether any of the pottery was associated with the cremation. The sherds are small and weathered, and the character of the whole assemblage is suggestive of domestic refuse which had already been strewn over the area when the tomb was built. This interpretation is supported by the presence of a sherd in the chalk packing in the ditch, as well as by the fact that, as Mr. Annable reports (p. 79), the pottery from No. 7 is indistinguishable as a group from that obtained from Nos. 6a and 6 and from the surface of a nearby prehistoric barrow, West Overton 6b, which was excavated at the same time.® Unfortunately the sherd found in situ in No. 7 is not closely datable, but the entire collection from all four sites appears to be a homogeneous one, pointing to a limited period of occupation during the early years of the gnd century a.p. A shallow grave intersecting the outer edge of the ditch in the north-east quadrant contained the extended skeleton of a child, much disturbed by burrowing animals. The corpse had been placed on its back, with the head to the south-west. At first sight the situation of this grave suggests that it was related to the post-circle in the ditch, but the juxtaposition was probably fortuitous, for, though there was no direct dating evidence,7 the posture and orientation of the skeleton were similar to those of a series of secondary pagan Saxon interments discovered in West Overton 6b. There is also some evidence, in the form of an unburnt skull fragment, an iron knife of Saxon type and sherds of grass-tempered Saxon pottery from the mound and top of the ditch, that other secondary burials had been made in No. 7.° NO. 6a This site (FIGs. 3, 4; PL. VIa) exhibited, on a smaller scale, the same structural features as No. 7. The mound, nowhere more than 1 ft. in height, was surrounded by a ditch 15 ft. in external diameter. The ditch was similar in form, width and depth to that of No. 7, but negative casts left by upright posts were less perfectly preserved. However, even where no casts could be traced, the unweathered sides and edges indicated that the ring had again been continuous. Apart from slight traces of upcast chalk round the inner edge of the ditch, the mound was featureless. The circular central pit was 2 ft. in diameter and 1 ft. deep. There were no finds in the disturbed fill of the pit. A bronze rivet and a decorated bronze suspension attachment (Fic. 6: 7) came from the mound and an iron object (Fic. 6: 6) from the upper fill of the ditch. None shows signs of exposure to heat. Some forty pieces of cremated bone, all but one probably of human origin, 73 S33 LIN i: = ——— —— —— —=————— 1334 ——— —— 7 1 0 01 § Nagi ‘WHO 7 0: Wis. 25a) d3dINLSid ~~ Sostaae " , ent Born) 2 Sqn 7 Silber ccna mats ; 3 an 4 Ranva WWH) eetet stat gacstbaats }j anya INI HOLId F i : aaa gas HOLId @3103£0%4) Lid WiysLvW : ANNOH IWNIDIIO Wiaslvw ANNOW “TWNIDI30 Sd Sos: . oe 0.5". 3d3nl Q O88 oF a9 SBOP” 9OQ0 2 Sow. 7 od ri yl “SLNSHOWYS AWH) “awa: SIV = Sava inn? | Li SING IELS cia ee) NOLYWIAO ISaa 14 ~~ YIVH) q3WWVd @ a et svadl HDL 3) q3g3nsi ea WYIVHD AgWnVa a >Lsvdn HILIG Fo? “SNOILOdS #92 NOIWAAO LSA WEST TOV ERION Go, PLAN a = SF noveen : CUT UMIT OF 7 EXCAVATION : ov te DISTURBED “5 ° rool 3 “DITCH UPCAST 7 S0ye =i 5 —— J fer oe = verees came from the mound in the area of the pit. There were also scattered sherds of Romano-British pottery and fragments of the bones of domestic animals. At the base of the humus layer, a little to the north of the central pit, an iron buckle and part of a vessel of Saxon grass-tempered ware were found together; a number of sherds of similar fabric were also distributed elsewhere in the mound. Despite the absence of unburnt skeletal remains, it is probable that here, too, there had been secondary pagan Saxon interments. The prehistoric pit described on pp. 82-83 and visible in the lower left-hand corner of pL. VIa lay 1 ft. south-west of the outer edge of the ditch. NO. 6 The overall dimensions of this monument (FIG. 5; PL. VIb) were similar to those of No. 6a, except that the ditch was somewhat narrower and did not exceed 1 ft. in depth. The fill consisted entirely of undifferentiated brown soil, but the vertical and unweathered sides show that it had not been left open and it is a reason- able inference that here too a ring of posts had stood. Since no part of the ditch upcast had been returned as packing material, the internal bank of chalk rubble was correspondingly larger. The central pit was 1-4 ft. in diameter and 2 ft. deep; a roughly square area of the natural chalk round it had been dug away to a depth of about 6 in., presumably again by ‘Thurnam. There were no finds from the pit and those from the mound and ditch were less numerous than at the other two sites. ‘They consisted of five scraps of cremated human bone, a few Romano-British sherds and a small quantity of animal bone. An unburnt human atlas, an iron stud and sherds of grass-tempered pottery probably indicate the former presence of secondary pagan Saxon interments. DISCUSSION Although No. 7 was the only site of the three to produce indisputable evidence of date, Nos. 6a and 6 resembled it so precisely in structure and were so obviously related to it spatially, that there can be no real doubt that all were Roman funerary monuments. Assuming that they were constructed fairly soon after the early end-century pottery was spread over the area, they would fall well within the period when cremation is known to have been practised in Roman Britain (up to 6.°A.D 225) 29 The finer details of the original appearance of these tombs, and in particular of superstructure, must of course remain unknown, but the general effect may have been that of a drum-shaped monument. Allowing for the loss of about a foot of the original chalk surface through weathering since Roman times, the ditches may be assumed to have been 2 to 3 ft. deep when first dug out; the timbers they supported could then have risen above ground to a height of at least 6 ft. It seems clear that the rings of closely-set posts (and any elaboration of this basic structure above ground level) were intended to be the conspicuous, monumental, features of these tombs. Neither the chalk rings inside the ditches, which were perhaps more or less incidental to the plan, nor the heaped soil (perhaps originally piles of turves) over the central pits would have been visible while the posts still stood. Since the low mound that was left after the tmbers decayed was merely the durable residue of a monument that had never been intended to be a barrow in the sense of a large permanent memorial, we have avoided the use of the term ‘barrow’ throughout this report. We are much indebted to Mr. R. F. Jessup for the information that the West Overton tombs are the only examples of their kind known in the British Isles and that the type is also virtually unknown on the Continent. The one exception appears to be the cremation cemetery at Gaalse Heide, North Brabant, Netherlands,'° where pits containing cremation deposits were in some instances surrounded by narrow vertical-sided trenches, circular or rectangular in plan. Impressions left by decayed posts could only once be detected, but the excavator concluded from the shapes of the trenches that all must in fact have supported upright timbers. Whether or not small mounds, as at West Overton, had been contained within the post-settings remains unknown, since the sandy soil had been under cultivation for many centuries. The pottery accompanying the Gaalse Heide cremations suggests that the cemetery had been in use for about two centuries, beginning in the third quarter of the rst century A.D.; those of the cremations inside timber structures that are closely datable seem to belong to the ist and end centuries. The West Overton tombs are likely, therefore, to have been contemporary with some of the Dutch ones, and may perhaps be interpreted as manifestations of the same funerary tradition. In conclusion it may be remarked that there is some reason to suspect the existence of other Roman cremation tombs in North Wiltshire, whether of West Overton type or true barrows it is of course impossible to say. In view of the objects recovered, three very low mounds described by Merewether might reasonably be attributed to the Roman period, and in fact he himself so interpreted them." THE METAL OBJECTS FROM NOS. 7 AND 6a DESCRIPTION Nou Bronze fragments from disturbed fill of central pit This assemblage of eleven major fragments provides several features of interest, and one can only regret that the bronze vessels represented received such rough treatment in antiquity. There is first the flat rim of a cup or jar which could have had a diameter of about 8 cm. (FIG. 6: 2); a portion, length 2-4 cm., of a narrow tube or piece of beading; two pieces of much-distorted plating, the larger, 5 cm. in length, exhibiting a rivet- or pin-hole for attachment; two fragments of the side of a curved cup or jar (just possibly the vessel noted above) ; a length of 4. cm. of the flat rim ofa cup or jar, with slight internal moulding at the lip and a small exterior lug, no doubt part of an attachment for a ring or handle (Fic. 6: 1); part of such a ring, diameter 2 cm. (FIG. 6: 3); a much-corroded piece, perhaps a lug or escutcheon attachment for a handle; a fragment of a (circular) 7] attachment-plate with cable-and-line decoration deeply cut to receive enamel enrichment (FIG. 6: 4); and another small fragment which appears to have traces of much-burnt enamel still adhering to the surface (Fic. 6: 5). There are also eighteen scraps of fused bronze which include one or two pieces of plating. Bronze fragments from mound material There are about thirty-six scraps, the largest 1-5 cm. in length, much corroded and showing signs of fusing by quite intense heat. Among them it is possible to recognize four small rivets and four or five fragments of thin plating. All could have been part of the decoration of a bronze-mounted box, and there are minute fragments of charred wood on one scrap. There is also a cone- shaped mass of bronze, length 6-5 cm., maximum diameter 4°5 cm., weight 101-7 grammes. It does not appear to be an object; there is no trace of a socket or other means of attachment. It could have resulted from the melting of bronze objects, the molten metal having run fortuitously into a conically-shaped (?stake-) hole in the ground. The crystalline appearance of the rough surface might result from cooling in contact with a particular soil. Fic. 6 West Overton. Metal objects from No. 7 (1-5) and No. 6a (6-7). Scale: 4. Bronze fragments from fill of post-cavities in the ditch There is the head and part of the shank of a thin bronze rivet, diameter of head 0-6 cm. It is similar to other rivets from the site, and again was probably used to fix bronze mountings to a wooden casket or chest. The larger of the other two objects, length 3-5 cm. and width 1-3 cm., is a corroded, fused and distorted fragment of a bronze mounting with one hole for its rivet attachment still visible. The third object is indeterminate. No. 6a Bronze fragments from mound material A suspension attachment (Fic. 6: 7), damaged and much corroded, consists of two butts, each 1 cm. in width, the more complete 3-5 cm. in length, each provided with rivet holes and a plain solid ring fixed at the upper end. Moving within these two fixed rings is a heavy moulded ring, 2-4 cm. in diameter and 0-6 cm. in thickness, decorated with a cable pattern. The butts are too small and the slots too narrow to accommodate any sort of leather strapping. Indeed, a fragment of corroded metal in one butt suggests that the fitting was part of a bronze bowl or pan, but there can be no certainty. There is also a scrap from the shank of a thin bronze rivet, length 1 cm. Tron object from upper fill of ditch This is a handle or attachment (Fic. 6: 6) of rectangular iron strip, length 9 cm., with closely-coiled terminal and part of a rivet hole for horizontal attachment in the opposite, expanded, end. No parallel is known to me, but just possibly it could be the finial of a lamp-stand or part of a candelabrum. COMMENTS The nature and condition of the material makes anything more than a general statement impossible. Bronze objects, including jugs and bowls and wooden caskets mounted with bronze fittings, are often found in Roman barrows in Britain as well as on the Continent and were sometimes burned on funeral pyres. The rough-and-ready treatment of the remains which were placed in the barrows has also been noted, and it is not uncommon to find parts of the same objects scattered at different places in the mound and ditch. Here at West Overton the problem is also complicated by the previous disturbances. The traces of enamel work on the decorated fragment from No. 7 are of particular interest, and recall the richly-enamelled casket from the largest of the Bartlow Hills barrows.!2 R. F. JESSUP THE ROMANO-BRITISH POTTERY GENERAL REMARKS The three groups of Romano-British sherds recovered from Nos. 6, 6a and 7 are indistinguishable from one another both in type and fabric, and are also indistinguishable from the group obtained during the excavation of the nearby prehistoric barrow (West Overton 6b), which it is convenient to include in this discussion. Collectively, the pottery could well represent a single contemporary occupation within the area. The cooking pot rims are few in number and as a type difficult to date closely, but all could fall within the early years of the end century a.p. A ring-necked jug fragment from the mound of No. 7, the only other possibly datable coarse sherd, is an early type and likely to be of similar date. From Nos. 7 and 6b were recovered body fragments which, as their fabric shows, are indisputably products of the Romano-British kilns in Savernake Forest,%3 fe) in production during the first years of the 2nd century A.p. Body sherds with grooves and cordons, again reminiscent of Savernake forms, also appear amongst the pottery. No single rim type in any group is late and, on the basis of a general assessment, an early gnd-century date may fairly be inferred for the whole assemblage. The three Samian sherds are much worn and too small for accurate identification, although here also a gnd-century date is a reasonable assumption. It is interesting to note the great similarity in fabric and type between the pottery from the West Overton sites and the coarse wares recovered from Windmill Hill.14 SCHEDULE No. 7 From chalk packing in ditch 1-sherd: grey ware, undatable. From surface under chalk bank 1 fragment of brick- or tile-shaped object: the soft and coarse reddish fabric is not easily matched, but is probably of Romano-British manufacture. From central pit 2 body sherds: gritty grey/brown ware, undatable. From mound 1 sherd: Samian Drag. F. 18, or 18/31; late 1st century A.D.; 1 sherd: ring-necked jug; compare Jewry Wall, where dated A.p. 110-20 (Kenyon, Excavations at the Jewry Wall Site, Leicester, Soc. Ant. Lond. Reports of Research Committee, XV: fig. 28: 4)3 1 grooved sherd: Savernake Forest type; 1 body sherd: certainly Savernake manufacture; 1 body sherd: probably Savernake manufacture; 5 rim sherds: cooking pot types, grey wares; all could certainly be and century; 35 body sherds: grey, brown, and reddish brown wares, fabric gritty and much abraded. From soft fill of ditch 1 rim fragment: Samian Drag. F. ?22; late 1st century ?; 1 cooking pot rim: too small to be datable; 24. body sherds: light to dark brown, dark grey and reddish brown, gritty and abraded wares. From fill of secondary inhumation grave 1 small chip: Samian, undatable; 1 body sherd: light grey coarse ware, undatable. No. 6a From mound : 1 rim fragment: Samian Drag. F. 33, cup with concave sides; c. early and century A.D. 2 rim sherds: a bead rim in a grey/brown gritty fabric; a light buff/brown ware; undatable; 5 sherds: brown to grey abraded wares, not distinguishable from material from Barrow 6b (see below). From soft fill of ditch 3 rim fragments: cooking pot types; rough fabric, dark grey surface, brown core; one could be of gnd-century date. From top of prehistoric pit 1 rim fragment: brown/grey gritty ware; cooking pot type; again need not be later than early 2nd century. 80 No. 6 From mound 1 rim fragment: dark grey core and surface; 13 body sherds: grey to reddish brown ware. From soft fill of ditch 1 rim fragment: cooking pot type: light grey core and surface; rim shape suggests likely 2nd-century date; 7 body sherds: gritty greyish brown fabric, core and surface colour similar; 1 large sherd: fabric as above, containing grits. No. 6b From humus layer over and round the barrow 1 rim fragment: light grey ware with moulding at base of rim; probably a copy of Samian Drag. F. 24/25; post-Claudian, made up to Flavian times; 4 rim fragments: cooking pots; none need be later than early 2nd century; 2 rim fragments: cooking pots; light/dark grey core and surface; 1 sherd: definitely Savernake ware; 2 body sherds: light grey core and surface; one has cordon, the other grooves reminiscent of Savernake ware; 1 rim fragment: greyish brown rough fabric; 1 grooved pie-dish fragment; 120 body sherds: light/dark grey, light/dark brown, reddish brown and red wares, gritty fabric; one or two probably and two possibly of Savernake manufacture. F. K. ANNABLE THE SKELETAL MATERIAL No. 7 CREMATED BONE From central pit There are five pieces of calcined bone in all; probably two pieces are femur or tibia and two from ribs. From mound Only six fairly small pieces of calcined bone were present. Probably five of these are from long bones. The other piece may not be human. UNBURNT BONE From secondary grave, probably pagan Saxon From the developmental condition of the permanent teeth, the child was probably about 5 years old. The skull consisted of parts of a mandible, parts of the left temporal and sphenoid, and pieces of parietal, frontal and occipital bones. What remains of the orbits shows the presence of usura orbitae. A small area of the left parietal fragment displays changes probably indicative of slight inflammation. The dentition is as follows (mandi- bular teeth only): Right: deciduous teeth missing, all sockets present; first permanent molar possibly erupting; permanent incisors, canine and premolars present, unerupted; second permanent molar missing, unerupted. Left: sockets of deciduous incisors and canine only; permanent incisors present, unerupted; permanent canine missing, unerupted. The post-cranial skeleton is represented by fragmented and very incomplete vertebrae, ribs, right clavicle, right scapula, humeri, right ulna and radius, two metacarpals, the pelvis, both femora, and the right tibia and fibula. From mound The individual is only represented by a fragment from the central region of a frontal bone, certainly of an adult. Sex could not be determined with any certainty. 81 No. 6a CREMATED BONE From mound There are about forty small pieces of calcined bone, very probably human. Some could be distinguished as belonging to the femur or tibia; also a metacarpal or metatarsal, and small fragments of skull. A further fragment of a large long-bone is unlikely to be human. No. 6 CREMATED BONE From mound and soft fill of ditch There are five calcined fragments, possibly all derived from long bones. UNBURNT BONE From soft fill of ditch This is an adult atlas vertebra. The morphological features are neither sufficiently slender nor robust to suggest the sex of the individual. R. POWERS and D. R. BROTHWELL Sub-Department of Anthropology, British Museum (Natural History) THE PIT NEAR No. 6a AND OTHER PREHISTORIC FINDS THE PIT This pit (FIG. 3; pL. VIa) was the only prehistoric feature encountered during the excavation of the three Roman tombs. It had a diameter of 2-6 ft. and a depth of 1-6 ft. below the surface of the chalk, with vertical sides and a rounded bottom. The sharpness of the upper edges and the unweathered sides indicate that it cannot have been left open for long and must have been deliberately refilled shortly after excavation. An animal burrow (?rabbit) extended through the centre of the fill to the bottom. Elsewhere the fill consisted of an apparently fairly uniform grey material, which gave the impression of a water-sorted silt sparsely flecked with charcoal. Three samples of this filling from different levels were submitted to the Institute of Archaeology for analysis. On these Miss Andrée Rosenfeld reports as follows: ‘All three consist of a mixture of chalk, subsoil and soil. No significant distinctions can be made on the basis of any physical characteristics. Chemical analysis shows a slight increase in phosphates in the lowest sample, but no further distinction. It would seem probable, therefore, that the pit was simply refilled with nearby surface material, and that it may have contained some organic debris at the bottom.’ Apart from one Romano-British sherd from the top of the fill, all the artifacts found in the pit were prehistoric. They consist of some twenty sherds, all confined to the top few inches of the fill, and a number of unretouched flint flakes. There were also a few animal teeth, including those of ox, sheep or goat, and dog, small fragments of ox and sheep/goat bone, two of them burnt; two large pieces of fractured sarsen; three fragments of ferruginous sandstone; and a number of snail shells (see below, p. 84). The prehistoric sherds represent a minimum of nine vessels, all apparently late Peterborough ware (Fengate style), as follows: 82 West Overton. Prehistoric finds: sherds from pit near No. 6a (1-5); sherds (6-9) and flint scraper (10) from mounds. Scale: 4. I. (FIG. 7: 1). Two sherds from an overhanging rim; on the exterior oblique lines of indistinct impressions and an interrupted band of closely-spaced rectangular impressions on the interior. Brown ware with a flaky black core, containing abundant shell fragments. 2. (FIG. 7: 2). Fragments of overhanging rim with part of a trellis pattern in twisted cord on the outer surface and deep finger-nail impressions on the internal bevel. Flaky dark brown ware with black core, containing sparse particles of flint. 3. (FIG. 7: 3, 4). Two body sherds with impressed herring-bone pattern. Reddish surface, flaky black interior; no conspicuous grits. 4. Two further sherds with ornament similar to No. 3 and one plain sherd. Orange surface, black core; tempered with crushed pottery. 5. (FIG. 7:5). Two fragments of flat base, 20 mm. thick and slightly hollowed beneath. Pale orange exterior and black core; abundant particles of flint. Six small sherds, probably part of the same vessel; two exhibit traces of impressed finger-nail decoration. 6-9. Some thirty small fragments representing a further four vessels, two with finger- nail impressions. Pits of this character are commonly the only surviving traces of Neolithic settlements in southern England. It seems probable that they were primarily storage-pits, subsequently refilled with domestic refuse and other material when they became unserviceable. The sherds of overhanging rims and of a flat base from the West Overton pit indicate a date at the end of the Neolithic period. The complete absence from this pit of flint or bone artifacts of Secondary Neolithic type should be stressed. Such artifacts typically occur with Rinyo-Clacton ware, often in considerable numbers,!5 but are rarely, if ever, found in closed contexts, e.g. in pits, with Peterborough ware. 83 OTHER FINDS The few objects of prehistoric character that were found scattered over the three Roman sites are listed below. The sherds are all small and weathered. Western Neolithic ware. Sherd from simple rim of small diameter (Fic. 7: 6); hard brown ware containing some flint. ‘Two body sherds of similar fabric. Peterborough ware. Small fragment of overhanging rim with oblique incised grooves and part of a deep pit in the neck (Fic. 7: 7); hard brown ware with abundant flint. Five featureless sherds of ?Peterborough ware. Beaker. Small sherd with single line of fine twisted cord impression; reddish ware. Featureless red sherd of ?coarse Beaker. Deverel-Rimbury ware. Two sherds with low cordons bearing vertical finger-tip impressions (FIG. 7: 8, 9); hard black ware containing abundant finely-crushed white flint. Six featureless sherds of similar fabric. Worked flint. Scraper with bulbar end removed (Fic. 7: 10) and a crudely retouched or utilized flake. Worked stone. Vhree small fragments from a polished implement, all found scattered over site 6a. One fragment has been sectioned'¢ and proves to be ‘a very fine grain brownish sandstone’. Other stone. Numerous unworked pieces of ferruginous sandstone were encountered over the whole of the excavated areas. Since similar pieces occurred in the fill of the pre- historic pit, it is reasonable to suppose that the scattered ones were introduced in the same period. THE MOLLUSCAN FAUNA FROM THE PREHISTORIC PIT The following species were identified: From upper fill of pit Arianta arbustorum (L.): 5 (2 are juvenile) Cepaea nemoralis (L.): 2 (I is 00000 and probably brown, certainly not yellow; the other is poorly preserved) Hygromia striolata (C. Pfeiffer): 1 Hygromia hispida (L.) : 2 Helicella itala (L.): 3 Discus rotundatus (Miiller) : 2 From all levels in pit Pomatias elegans (Miller) : I Arianta arbustorum (L.) : 3 Cepaea sp. juvenile: I Cepaea nemoralis (L.): 3 (1 is unbanded, 00000; 2 are poorly preserved) A. J. CAIN Department of Koology and Comparative Anatomy, University Museum, Oxford ACKNOWLEDGEMENTS We should like to express our thanks to the owners of the sites, English Farms Limited, for permission to excavate. We are also much indebted to Mr. R. F. Jessup for advice on the interpretation of the Roman sites and for his report on the metal finds, and to Mr. F. K. Annable, Miss R. Powers and Dr. D. R. Brothwell, Miss Andrée Rosenfeld and Dr. A. J. Cain for their respective contributions. Miss Frances Lynch, Miss Susan Nicholson, Mr. Michael May and Mr. A. H. Stokes were of great assistance throughout the excavation and should in particular be credited with the successful accomplishment of the delicate task of exposing the post-rings in the ditches of Nos. 7 and 6a. 84 t Ancient Wilts., u (1821), 91. 2 The term is used advisedly, for the severe mutilation of the central areas of these two sites must be attributed to Thurnam. 3 Examination of Barrows on the Downs of North Wilts., W.A.M., vi (1860), 330-1. 4 Kindly identified by Dr. G. W. Dimbleby of the Commonwealth Forestry Institute, Oxford. 5 See comment by F. K. Annable on p. 80. 6 The excavation report on West Overton 6b will be published separately. 7 The few sherds recovered from the grave included prehistoric, Romano-British and Saxon wares. 8 The Saxon material recovered during the 1962 excavations at West Overton will be pub- lished in a separate report by Dr. V. I. Evison. 9 See R. F. Jessup, Roman Barrows in Britain, Collection Latomus, Lv (1962), 853-67. to P, J. R. Modderman and C. Isings, Het grafveld uit de Romeinse tijd op de Gaalse Heide, gem. Schayk (N.-Br.), Berichten van de rijksdienst voor het oudheidkundig bodemonder zoek, 10-11 (1960-61), 318-46. ™ Diary of the Examination of Barrows and other Earthworks in the Neighbourhood of Silbury Hill and Avebury, Proc. Archaeol. Inst., Salisbury, 1849, 82-112. The barrows are No. 10 (V.C.H. Wilts, Avebury 352), No. 19 (V.C.H. Wilts, Avebury ?1g9), and an unnumbered one (p. 84) on ‘Bye Down Hill’ (Monkton Down), apparently not identified in V.C.H. Wilts. It would appear to have been situated near the Roman sites on Monkton Down (Nos. 171-2 in W.A.M., xLv (1930), 209). The first two covered cremations in pits; the third was only superficially dug into, but seems to have yielded a quantity of Romano-British pottery, including an intact vessel. ™ See generally R. F. Jessup, Barrows and Walled Cemeteries in Roman Britain, Journ. Archaeol. Assoc., 3rd series, xxi (1958), 1 ff., and Roman barrows in Britain, Collection Latomus, Lyi (1962), 853 ff. 3 F. K. Annable, A Romano-British Pottery in Savernake Forest, W.A.M., Lv (1962), 143-55. ™ See report by F. K. Annable in Windmill Hill and Avebury: Excavations by Alexander Keiller, 1925-1939 (Clarendon Press, forthcoming). 15 See, for example, J. F. S. Stone and W. E. V. Young, Two Pits of Grooved Ware Date near Woodhenge, W.A.M., Lit (1948), 287-306. 16 Implement Petrology Survey of the South- West; No. 1219. A SAXON CEMETERY AT WINTERBOURNE GUNNER, NEAR SALISBURY by JOHN MUSTY and J. E. D. STRATTON INTRODUCTION THE sITE of the Winterbourne Gunner cemetery, which is approximately 4 miles north-east of Salisbury (FIG. 1), was discovered accidentally during the digging of a narrow pipe trench. The digger of this trench (Mr. Andrews) reported the finding of a spearhead, but before we could visit the site, four graves had been cut through by the trench, and the end of another exposed in section. Arrangements were then made for a proper archaeological examination on behalf of Salisbury Museum, and this was carried out during the summer of 1960. The greater part of the site lay in a garden and the owner insisted that disturbance of his ground should be kept to a minimum. It was not possible, therefore, to strip the area completely, but cuttings were laid out in such a manner as to give the maximum coverage of the ground, and it is unlikely that any graves were missed. Ten graves were excavated and these contained the skeletal remains of men, women and children. For a small cemetery the grave goods were particularly rich and included objects not previously recorded in the Salisbury area, and only two of the graves lacked grave goods. There is no previous record of Saxon finds from Winterbourne Gunner; there are, however, a number of other cemeteries in the greater Salisbury area (FIG. 2). These are (A) Petersfinger (excavated 1949-51: 65 graves, 70 skeletons), (B) West Harnham (excavated 1852: at least 64 graves, 67 skeletons), (C) Council House Grounds (St. Edmund’s), Salisbury (excavated 1771-74: about 30 skeletons),3 and (D) Broadchalke (excavated 1925: at least 25 skeletons),4 all of which lie in, or overlooking, river valleys. In addition, there are the two upland cemeteries at (E) Roche Court Down (excavated 1932: 13 graves, 16 skeletons),5 and the ‘Execution Cemetery’, also on Roche Court Down (excavated 1932: 16 skeletons, mostly decapitated).6 These cemeteries, with the exception of that at Broadchalke, are within a 5-mile radius of Winterbourne Gunner. THE SITE The site (Fics. 3 and 4) lies in the valley of the River Bourne, south-east of Winterbourne Gunner church, not far from the river and the church (SU/182352). This church is surrounded by house platforms, but otherwise, except for a recently- erected house, stands alone. These house platforms probably represent the original 86 GUNNER SALISBURY ¢ WINTER BOURWE Y% MILE 2 MILES Key to Fig2 Cemeteries Valle |_| Valley A Petersfinger FJ LowChalk Plain © 8 West Harnham — C Salisbury Council House Grounds High Chalk Pici D Broadchalke seuO ih E Roche Court Down ye Scarp Fig2 Fics. 1-3 The position of the Winterbourne Gunner Cemetery and its relationship with other Saxon Cemeteries in the Saiisbury area. nucleus of Winterbourne Gunner, the village having subsequently shrunk to its present form away from the church. To the north is Gomeldon Hill on the slopes of which lie the earthworks of the deserted medieval village of Gomeldon, now resettled along the top of the hill. Nearby the Roman road from Old Sarum to Silchester (the Port Way) fords the river. The newly-discovered cemetery therefore lies close to a Roman road and to deserted early medieval settlements. The graves have been dug in a slight natural chalk rise, possibly selected because it would provide slightly drier ground than its surroundings where the chalk falls away rapidly under clay. The ground, until recently worked as allotments, is now being converted into permanent caravan sites or building plots. The trench which revealed the cemetery was being dug as part of a small local scheme to lay on water to these properties. Most of the cemetery was situated on a plot where a bungalow had been built recently. Nothing was found during the erection of this bungalow, or during the 87 sinking of a septic tank in its garden, although a pipe trench running from the tank to the house had cut through the edge of a small, oval pit. THE GRAVES The graves (FIGS. 4, 9-10) were all cut in solid chalk to a depth of 1 ft. 6 in. to 2 ft. All the graves had west to east orientations and occupied a strip 36 ft. wide and 56 ft. long, five of the graves lying in one west to east line (Fic. 4). An account of the skeletal remains, prepared by Miss Powers of the British Museum (Natural History), is given in Appendix I. All the skeletons were extended and supine with the skulls to the west. Many bones were missing, a fact which may be attributed to wetness due to the low-lying position of the cemetery, although poor preservation of skeletal material appears to be a feature of many Saxon cemeteries. That wet conditions may have always existed is suggested by the presence in the graves of wood and leather which might not otherwise have been preserved. After burial the graves had been refilled with chalk rubble and this contrasted ~ in colour with the bedrock chalk, so that the graves were easily recognized when Fig.4. WINTERBOURNE GUNNER 1960 Water pipe trench Grass path Fic. 4 Plan of site. 88 the top soil had been removed (approximately 10 in.). No evidence for coffins was found, but three graves were partly lined with large flints. The only trace of clothing was in the form of textile remains preserved on a few of the objects from the graves. The presence of a coarse weave textile on the throwing axe from Grave VI suggests that the body in that grave might have been covered with a blanket or coat. One small empty oval pit had been cut by Grave VII. As Grave VII contained the intrusive sherds of comb-stamped and cord-zoned beakers, it is possible that this pit is connected with earlier prehistoric activity on the site. Although the form of the pit resembles that of a Beaker grave, there was no other evidence to suggest that a prehistoric barrow preceded the Saxon Cemetery. One of the ten graves (Grave X) was also empty of skeletal remains, but the presence of a string of Saxon beads in the grave showed that it had contained a burial. The smallness of the grave implies that this burial had been of a child whose remains had completely decom- posed. GRAVE INVENTORY All burials were extended. The following abbreviations have been used: D. = depth of grave in chalk; DS. = depth from ground surface; L. = length; H. height; Diam. = diameter. I GRAVE I (FIG. 9) Disturbed by pipe trench. D. 1 ft. 2 in.; DS. 2 ft. Adult male, poorly preserved and lower limbs broken by trench. Associated finds were (ric. 5, I): (a) At right shoulder—zron spearhead (L. 44 cm.); narrow leaf-shaped blade with split socket (Diam. 2 cm. internal) containing traces of wooden shaft. (b) On left half of pelvis—zron shield boss (Diam. 16-8 cm.); low-conical with point ending in a button (Diam. 2-3 cm.) and grip (L. between rivets 11-8 cm.) with traces of wood. Originally attached to the shield board by means of the four flat-headed rivets. The boss had been re-riveted at some time as there is another set of empty rivet holes. The grip has side flanges; there are traces of wood on the inside face of the grip and also on the undersurface of the boss. (c) In lumbar region above left half of pelvis—a group of objects consisting of: (1) Purse Mount. L. 11-5 cm. Remains of a textile around the central area. (2) Lron Knife. Overall L. 15-4 cm. with a tang 5:0 cm. L. Traces of wood on the tang. (3) Bronze Tweezers, decorated. L. 6:2 cm. with jaws 1-1 cm. wide. Decoration consists of stamped circles and also a series of grooves at the butt end. GRAVE II Badly disturbed by pipe trench. D. 6 in.; DS. 1 ft. 4 in. Adult female, poorly preserved. No grave goods. 89 iMwili RD) yu Tn emia bt Wigan Ai] Tel AA it Fic. 5 The objects from Graves I, IV and V. The tweezers (Ic3 and Vb) 3; the remainder 4. GRAVE III Badly disturbed by pipe trench. D. 1 ft. 8 in.; DS. 2 ft. 6 in. Adult female poorly preserved. No grave goods. GRAVE IV (FIG. 9) Undisturbed. D. 1 ft. 4 in.; DS. 2 ft. 2 in. Adult male, poorly preserved. Associated finds were (Fic. 5, IV): (a) At right shoulder—iron spearhead (L. 23:7 cm.) with traces of wooden shaft. Ogival-shaped blade. (b) Above left knee—zron shield boss (Diam. 18:9 cm.); low-conical, with exceptionally large stud (Diam. 5-7 cm.). Originally attached to shield board by means of four flat-headed rivets (Diam. rivet heads is 2-5 cm.) approximately 11 cm. from each other. The grip is flat and without side flanges; there are traces of wood on it and on the boss flange. (c) Above spine in midriff position—iron belt buckle and fittings with bronze stud. Traces of leather on the fittings. GRAVE V (FIG. 9) Disturbed by pipe trench. D. 6 in.; DS. 1 ft. 4 in. Lower limbs of adult male. Poorly preserved. Grave flint-lined on left side and at the foot. Associated finds. were (FIG. 5, V): (a) Beneath pelvis—iron knife. Overall L. 11-7 cm. with a tang of L. 3-5 cm. (b) Beneath digits of hand and on right femur—a pair of plain bronze tweezers. L. 7-5 cm. with jaws 9 mm. wide. A slight chamfer has been applied to the edges at the butt end. GRAVE VI (FIG. 10) Undisturbed. D. 1 ft. 1 in.; DS. 1 ft. 11 in. Adult male, poorly preserved. One large flint on either side of grave at skull. Associated finds were (ric. 6, VI): (a) At left shoulder—iron throwing axe (francisca) with traces of wooden shaft, and replaced textile on blade. (Overall L. 16-3 cm.; cutting edge 7-5 cm. wide.) (b) At left elbow—iron buckle loop and tongue (b1) and ? buckle plates (b2). (c) In right midriff position—an iron stud (b3) and a pair of decorated bronze tags or strap ends (c1), L. 6-5 cm.; a pair of undecorated bronze tweezers (c2) L. of tweezers 5 cm. with jaws 6-5 cm. wide, attached to a split ring (Diam. 1-4 cm.) formed from 2 mm. rod; and an iron buckle plate (c3). (d) Two joining sherds of Saxon pottery. GRAVE VII (FIG. 9) Undisturbed. D. 1 ft. 1 in.; DS. 1 ft. g in. Juvenile, poorly preserved, c. 8 years. The sex could not be determined in the anatomical examination, but the nature of the grave goods suggests female. Associated finds were (Fic. 6, VII): (a) At left side of skull—tinned square-headed brooch, iron pin missing. (b) At base of skull—sherd from cord-zoned beaker (not illustrated). gL CZ - (ee < A a | d UTD @y Fic. 6 The throwing axe, perforated spoon and other objects from Graves VI and VII. (Scale of a, b1, b2, b3, cz and d from VI, 4; remainder 1:1.) (c) Grouped about right elbow—+three clear glass beads (two blue, one lemon coloured). (d) Between femurs—a tinned bronze perforated spoon, L. 10 cm. (e) Two broken amber beads. (f) Outside right femur in fill—base of a comb-stamped beaker (not illustrated). GRAVE VIII (FIG. 10) Undisturbed. D. 1 ft. 6 in.; DS. 2 ft. 4 in. Adult female, poorly preserved. Associated finds were (FIG. 7): (a) At each collar-bone—a bronze, gold-plated applied saucer brooch (Diam. 6-1 cm.), decorated, with central blue glass bead. Identical design. (b) Grouped about right wrist—a bracelet of 42 amber, paste and glass beads. These comprised 17 amber beads (three cylindrical, seven quoit-shaped and seven cigar-shaped) and 25 pigmy paste beads (including eight dumb-bell shaped with spherical members and two with cylindrical members). (c) Lron chatelaine ring (Diam. 3:5 cm.) and fragments of latch lifter. (d) near (c). At inside of upper femur—zron object with wood facing, possibly a small knife or similar object in a sheath. (e) Indeterminate iron object. GRAVE IX (FIG. 9) Undisturbed. D. 1 ft. 3 in.; DS. 2 ft. 1 in. Adult female, poorly preserved. Grave lined with large flints on either side. Associated finds were (Fic. 8, UX): (a) At each collar bone—a bronze, gold-plated applied saucer brooch (Diam. 6-1 cm.) with central blue glass bead. Identical with pair in Grave VIII. (b) Bronze clip. (c) Tang of iron knife near chatelaine ring. (d) In midriff area—iron belt buckle. (e) A large green glass bead with serrated outer edge. (f) At left elbow—zron chatelaine ring (Diam. 4:5 cm.). (g) Iron pin (L. 8-2 cm.). GRAVE X (FIG. Q) Undisturbed. D. 11 in.; DS. 1 ft. 9 in. Completely decomposed. Probably young female (c. 4-5 years). Associated finds were (Fic. 8, X): (a) In supposed neck area—necklace of 11 glass and paste beads (five of opaque glass: one brown, one greenish-white, two green, one blue; five blue clear glass; and one amber). (b) In grave fill—an irregularly-shaped piece of lead. DESCRIPTION OF INDIVIDUAL OBJECTS SPEARS (Graves I and IV: ric. 5: Ia, [Va) The spear from Grave I is of leaf-shape and exceptionally long (circa 44 cm.) ; that from the other grave is only half this length. It is, however, possible that the 93 94 Fic. 7 Applied saucer brooches and other finds from Grave VIII. (Scale of cr, c2, c3, d and e, 4; remainder 1 m2) shorter spear reached its present length (and to some extent its ogival shape) through excessive sharpening. Both spears can be paralleled in form and dimensions with spears from the Petersfinger Cemetery. SHIELD Bosses (Graves I and IV: Fic. 5: Ib, IVb) The shield boss from Grave IV has an exceptionally large stud (5:7 cm.). Only one other boss with such a large stud has been recorded from Wiltshire and this was found at Wanborough in the north of the county.7 The Winterbourne boss is 9 cm. high and 18-9 cm. in diameter and is therefore slightly larger than the bosses from Petersfinger. The grip, however, is small and the distance between the attaching rivets is only 9 cm., insufficient for a large hand. Both the boss flange and the grip had remains of wood on them. That on the boss belonged to the shield board which can be shown (from the length of the rivets) to have been approximately 8 mm. thick, a thickness comparable to that observed elsewhere. The boss was attached to the board by means of four rivets. The wood on the grip has its grain running in two directions at right angles, and the two grains overlap around the rivet. This suggests that the grip was faced with wood, and that this wood formed a halved joint with that of the shield. The other boss is smaller in all its dimensions. The stud is only 2:3 cm. in diameter and the overall diameter 16-8 cm. The grip is, however, larger and the distance between the rivets more normal (11-8 cm.). It also differs in another respect from that found in IV in that it has side flanges. The boss has obviously been refitted to a shield board at some time as it has two sets of rivet holes, the grip also probably comes from another shield as its size is out of proportion to that of the boss. In fact the two grips would be more appropriately matched to the bosses if changed over. THE THROWING AXE (Grave VI: Fic. 6: VIa) The throwing axe lay on the left side of the skull with its cutting edge downwards. The axe is a Frankish type (a francisca) and rare in this country. It is the first example of this type to be found in Wiltshire, although an axe was found at Peters- finger (Grave XXI). The socket for the handle is of oval section, and runs approximately parallel to the edge of the blade. The socket still contains the wood of the handle and a large rivet which originally held the head to the handle. Textile remains are present on the blade and the axe may have been wrapped, therefore, in cloth before burial, or the textile may have come, as Miss Crowfoot suggests (Appendix III), from a cloak or blanket. The date for this axe according to B6hner’s classification® should be 450-525 A.D. THE KNIVES (Graves I and V: Fic. 5: Ic2 and Va) Two complete blades and fragmentary remains of others were found. The knives are 11-7 cm. and 15°4 cm. in length, with tangs approximately one-third of the overall length. 95 96 OO ee SP < pene, ae LPI Fic. 8 The objects from Graves TX and X. (Scale of c, d, f and g from IX and b from X, 4; remainder, 1:1.) ralsne0 2) The position of the knives in the graves is of some interest. That from Grave I, along with a pair of tweezers, was in a purse which could be identified from its mount. The purse probably hung from a belt. The other knife was under the right side of the pelvis of Skeleton V, and a similar positioning of a knife was observed by Knocker at Snell’s Corner.9 TWEEZERS (Graves I, V and VI: rics. 5, 6: Ic3, Vb and VIc2) Three pairs of tweezers were obtained, all from male graves. They are of bronze and are 5:0-7°5 cm. long respectively, with jaws 6-5-11 mm. in width. One of the pairs is decorated with stamped circles and grooves; the other two pairs are plain except for slight chamfering on the edges of one. The tweezers from Grave VI have a bronze ring through the eye. THE PERFORATED SPOON (Grave VII: Fic. 6: VIId) Perforated spoons, usually associated with a crystal ball, are a well-known feature of Kentish graves, although comparatively rare. An example also comes from the Isle of Wight (Chessell Down, Grave 15). The Kentish examples, and that from the Isle of Wight, were found in richly-furnished graves and, with their garnet settings, are more elegant examples. The Winterbourne Gunner spoon is best paralleled by three examples from Bifrons (Kent): two from the Conyngham Collection (both unassociated) and the third from Grave 6. This latter grave is one of the earliest in the cemetery. The Winterbourne Gunner spoon is of tinned bronze and has an overall length of 10 cm. It has a slightly oval bowl (3-5 cm. by 3:8 cm.) perforated with five holes 2-5 mm. in diameter. The handle is slightly off-set, and widens where it joins the bowl. It is decorated with a series of crescentic punch marks and three scored lines at the widening of the handle; the other end of the handle is turned over. STRAP END WITH ZOOMORPHIC DECORATION (Grave VI: Fic. 6: VIcr1) A pair of bronze tags approximately 6-5 cm. long came from Grave VI. These presumably fitted each side of a piece of leather as there are matching holes through either end of them, and the inner surfaces have a fibrous deposit which is probably the remains of leather. The outer face of one tag is plain, but the other is decorated with zoomorphic ornament and a chased border. The ornamentation consists of an engraved beast which runs the whole length of the strip; the back of this animal is decorated with crescentic punch marks similar to (but smaller than) those used to decorate the spoon. The only parallels (and not exact ones) for this object can be found amongst Roman material. Thus in the Second Richborough Report!® is a bronze tag of similar general form but with a nicked end and described as a ‘nail cleaner’. The tag, like that from Winterbourne, is decorated with an engraved beast which has crescentic punch marks on its back. An example from Rivenhall, Essex, has been published by Tonnochy and Hawkes'! who also describe parallels from Cirencester and from excavations by Pitt-Rivers at Rotherley and Wor Barrow. Other examples oO” qa GRAVE IY. GRAVE I. GRAVE Y. PWN NY | | vo ®& GRAVE X. GRAVE Ik. Ele GRAVE WU. Fic. 9 Winterbourne Gunner. Saxon cemetery. were found at Lydney” and Silchester. Boon'3s has described and illustrated the four Silchester examples. Engraved beasts do not appear on these, but there is decoration in the form of crescentic punch marks and dot and circle motifs. The crescentic punch mark is therefore common to all the examples. The significance of this type of decoration, and the zoomorphic ornament, has been discussed by Mrs. Hawkes,'4 but the strap end is clearly late Roman and not yet of her Jutish Style A, i.e. of 4th/5th-century workmanship and not 5th/6th. BRONZE CLIP (Grave IX: Fic. 8: [Xb) A small bronze clip was found in Grave IX in the region of the skull. Five bronze clips were found at Petersfinger, Grave LVII (No. 151, pl. IV, Petersfinger Report) in a similar position, and it was suggested by the excavators that these were the remains of a hair ornament or a comb. Parallels from Suffolk and Cambridge- shire graves are also quoted in the same report. PURSE MOUNT (Grave I: Fic. 5: Icr) Objects of this type have been described as both strike-a-lights and purse mounts. That from Winterbourne Gunner is clearly a purse mount. The presence of decayed leather and a buckle, and the position of the mount in the grave, along with an iron knife and a pair of tweezers, is consistent with its being part of a bag which contained these items. Replaced textile was also present and Miss Crowfoot has suggested (Appendix IIT) that this may have come from a tunic against which the purse could have lain. A purse mount of somewhat similar form, but without a buckle, was found at Petersfinger (Grave XX, No. 57). BUCKLES (Graves IV, VI and IX: rics. 5, 6 and 8) Three iron buckles were found. The buckle from Grave IV (Fic. 5) is of oval shape (2-0 cm. by 2-5 cm.). Attached to the buckle is a plate with a square bronze stud and this has not been paralleled. The buckles from the other graves are of figure-of-eight form, and that from VI (Fic. 6) is of special interest as it has been inlaid with bronze wires. Inlaid buckles have been studied by Miss V. Evison's and shown to be of early date. She considers the Winterbourne Gunner buckle to date to the last half of the 5th century. The form of both buckles can be paralleled at Petersfinger (Grave LXIIIb, No. 180). Two iron objects which appear to be buckle plates were also found in Grave VI (FIG. 6: b2 and c3). CHATELAINE RINGS (Graves VIII and IX: rics. 7, 8: ci and f) These are iron rings 3:5 cm. and 4:5 cm. in diameter. With the ring from Grave VIII were the fragments of another ring and an iron pin, possibly a latch lifter. There was also an iron pin (with a ring head) in Grave IX (ric. 8:g). 99 GRAVE YI GRAVE “VIII. Fic. 10 Winterbourne Gunner. Saxon cemetery. APPLIED SAUCER BROOCHES (Graves VIII and IX: rics. 7 and 8) Four identical applied saucer brooches were found as two pairs. The brooches — are of the ‘Kempston Cross’ type and Mrs. Margaret Saunders has reported on them as follows: 100 “They are a Bedfordshire-Cambridgeshire type: 5 pairs and several fragments were found at Kempston (Beds.), 7 pairs or single examples came from the various sites near Barrington (Cambs.), single examples came from Haslingfield and Linton Heath (both Cambs.), and one pair came from Frilford (Berks.)—previously the only Thames valley or southern site. “The brooches are remarkably uniform in design, and many of them (including these new ones from Winterbourne Gunner) may well have been struck from the same die. “The overall design is closely related to one of the Kentish jewelled disc brooch types (Leeds type Ia), that with 4 wedge-shaped settings forming a cross, usually with a profile head and a schematic leg between each of the settings. This design has been adapted for the applied brooches by the development of the head into a full-face mask inside the arms of the cross, while the legs only remain in the spaces between the arms. The border of zoomorphic elements is in fairly simple Salin Style I zoomorphic ornament. “The fact that the design on these brooches is an adaptation of the design on the richer Kentish type means that this group of applied brooches must date to the second half of the 6th century, because the Kentish disc brooches are now generally agreed to date from the second quarter of the 6th century at the earliest. “The blue bead studs are found on several types of the applied brooches, including ones with floriated cross and star patterns, which may go back into the 5th century. These blue studs, therefore, are not an imitation of the central garnets or coloured bosses on the Kentish types, but a fairly frequent feature of applied saucer brooches.’ The method of construction of the brooches is as follows: The catch-plate and pin mounting are fixed to the undersurface of a slightly dished bronze base- plate. On this base-plate is seated a thin disc of gold-plated bronze which is held in place by a bronze collar. The disc is stamped with the characteristic Kempston Cross design, and the centre of the disc is perforated to receive a blue glass bead. The overall diameter of each brooch is 6-1 cm. and the collar is 6 mm. deep. None of the brooches is complete, although drawn as such. The blue glass bead is missing from one of the brooches (VIII left), otherwise in good condition. ‘The bead is present in the other brooches but one lacks part of the disc (VIII right), and others the collar (IX left and right). In addition, the pin mechanism of IX (left) has been torn out of the base plate. This suggests that the brooches were probably worn out before they were put into the graves. THE SMALL-LONG BROOCH (Grave VII: ric. 6: VIIa) The brooch is of tinned bronze and is 4-8 cm. long. The square head with slightly incurved sides is decorated with four dot and circle punch marks. Between the square head and the triangular foot is a pyramidal bow and a series of four ridges which have been formed by three saw cuts. The foot is decorated with three dot and circle punch marks. ‘The catch-plate is of bronze, and the rest of the fastening mechanism of iron. Approximate parallels for this brooch can be found at Bifrons, Kent, and the brooch is clearly an early type. THE BEADS The beads consist of two strings (one of forty-two beads, the other of eleven beads) from Graves VIII and X, and loose beads from Graves VII and IX. A variety of forms occur, and these are basically barrel-, quoit-, “cigar’- and segmented-dumb-bell-shaped. The materials used are amber and glass (both clear and opaque). (a) The string of beads from Grave VIII (Fic. 7) Approximately half of the beads in this string are of amber, including the principal bead. This latter bead (br) is a large cylinder approximately 1-0 cm. thick and nearly 3-0 cm. in diameter. The other amber beads are either small IOL cylinders (b2) or are quoit-shaped (b3) or cigar-shaped (b4) or irregular variants of these. The glass beads, whose fabric and colour resemble a skinned onion, are mainly segmented-dumb-bell-shaped (b5), and eight of these (plus a half bead) have been formed with spherical members, two with cylindrical members. The remainder of the beads are either fragments of segmented beads including a black bead (b6) or small quoit-shaped beads as b7. (b) The necklace from Grave X (FIG. 8) The necklace is made up of quoit-shaped beads formed from coloured glass, clear and opaque. ‘There is one brown opaque bead (ar), one greenish-white opaque bead (a2), two green opaque beads (a3), one blue opaque and five blue clear glass beads (a4). The bead a4 is not characteristic of all the blue glass beads as it has a larger central perforation than the rest. ‘There is also one amber bead (a5). (c) The green clear glass bead from Grave IX (ric. 8: e) This bead which is 2:1 cm. in diameter and 1-4 cm. thick, with a central perforation of diameter 5 mm., is of the so-called melon type. It has nine pronounced lobes. The bead is to be compared with the paste bead from Snell’s Corner,?® Grave $28, although this bead is not a close parallel. (d) Loose beads from Grave VII (ric. 6: VII) The beads from this grave are of clear glass and amber. There are three glass beads; one a prussian blue cylindrical bead 2-0 cm. long and 5 mm. in diameter (cr); the other two are quoit-shaped and in colour ultramarine blue and lemon of diameters 1-2 and 1-75 cm. respectively (c2 and c3). The blue beads have striations in the glass and the lemon coloured bead has occluded air bubbles. The amber beads have cylindrical form and are approximately 1-0 cm. thick and 1-2-1-5 cm. in diameter (only one of these is illustrated, e). Neither of the beads is complete, and they must have already been broken when they were placed in the grave. THE POTTERY In addition to the intrusive sherds of the two beakers from Grave VII (see W.A.M., 58, 414) and five small sherds of Roman pottery and one sherd of 12th-century ware from the top of grave fills, there were two joining sherds of Saxon pottery in the fill of Grave VI. The Saxon sherds (Fic. 6: VId) are in a brown sandy fabric with a dark brown polished surface and are decorated with a series of stamped cross-hatched squares, possibly arranged in a triangular setting. Stamped pottery of this type has hitherto been dated as not earlier than the 6th century. However, its association with 5th- century material in Grave VI shows that it was being made in fact before the end of the 5th century. 102 CHRONOLOGY In discussing the results of the excavation of the Petersfinger cemetery, Leeds!7 maintained the accepted view that an English invasion force (described by him as Juto-Frankish) first penetrated to the Salisbury area by a.pD. 552 after the initial landings from Southampton Water around 495. Because of the gap of 57 years between the initial landings and the battle of Old Sarum (552), Leeds suggests that the progress northwards was slow, and that the invasion was more in the nature of a military adventure than of a large-scale immigration. This accepted view for the dating of the settlement of Wiltshire by the English depends upon a rigid adherence to the Anglo-Saxon Chronicle, and a consideration of it as a step by step record of the progress of the invaders. The finds from the present excavation are such, however, as to suggest that one should re-examine the available evidence and approach the problem of the date of the English settlement of Wessex with an open mind, and not too rigid an acceptance of one set of interpretations of the Anglo-Saxon Chronicle. At Winterbourne Gunner at least two graves appear to be earlier than 552 (Graves VI and VII), and neither need be later than circa 520. Three of the objects from Grave VI (Fic. 6) are demonstrably early—the francisca, the buckle with metal inlay and the bronze tags with zoomorphic decoration. All the material from Grave VII (ric. 6) could be of early date and most certainly the perforated spoon and the small-long brooch, parallels for which occur in the earliest graves at Bifrons, Kent. Later material in the form of Kempston Cross brooches (Fics. 7 and 8) was found in two of the graves (VIII and IX) and these graves may there- fore be later than 552. A provisional view would be, therefore, that some of the graves at Winterbourne Gunner should be associated with primary settlement begun not later than circa 520, resulting from a penetration of a group from Southampton Water, and that the later burials belong to a consolidation period after circa 550, when contact had been made with the Saxons to the north. No attempt has been made by us to re-examine in detail the finds from the other South Wiltshire cemeteries for evidence of this earlier settlement date, but we have noticed one feature of interest in connection with the cemetery at Peters- finger which should form the basis of any re-examination of the chronology of that cemetery. The graves have two orientations: south-north and west-east. These orientations are not randomly distributed and comprise two distinct groups, of which the west-east group appears to be the earlier, as in two instances south-north graves cut west-east graves. Early (Frankish) material occurs in the west-east group, but not exclusively, as there are also saucer brooches, applied brooches and disc brooches of Saxon origin present in some graves. It is possibly with this west-east group of graves that the new finds at Winterbourne Gunner should be compared, in view of the west-east orientation of the Winterbourne Gunner graves and the similar mixture there of Frankish and Saxon material. A west-east orienta- tion was also found at the Roche Court Down Cemetery and at Harnham Hill, but the ‘Execution Cemetery’ on Roche Court Down had a south-north orientation and the graves at Broadchalke had a random one. An additional feature of the two phases represented by grave orientations at 103 Petersfinger is the large number of children present in the second (south-north) phase. These comprise 43 per cent. of all the interments (nearly 60 per cent. if adolescents are included) as compared with 11 per cent. in the west-east graves. The reasons for this difference are not immediately obvious. It remains to be considered how a pre-552 date can be reconciled with the entries in the Anglo-Saxon Chronicle. It must be assumed that the chronicler can have recorded only a small number of events occurring during the settlement of Wessex, and some of the statements made are not necessarily explicit. Thus the statement that ‘In this_year (519) Cerdic and Cynric took the Kingdom (of the West Saxons) ; and the same year they fought against the Britons at a place called Cerdicesford. (And from that day on the princes of the West Saxons have reigned)’ has been taken to imply that Cerdic and Cynric sustained a defeat, were unable to effect a river crossing and could not succeed in penetrating to the Salisbury area until many years later (i.e. 552, when it is recorded that the battle was fought at Old Sarum). However, it does not necessarily follow from the Chronicle entry that they were defeated, nor in fact that this particular group was the only one attempting to penetrate to the Salisbury Plain area. There can have been others; and still more pushing inland perhaps into Hampshire too. For why should Cerdic and Cynric ‘take the kingdom’ or ‘begin to reign’ if their people remained confined to this single group ? If all the various groups of ‘West Saxons’ throughout all these parts acknowledged them as kings at that time, the passage would be much easier to understand. It is unrealistic to consider that even a defeat at this river crossing (Charford, south of Downton) could have hampered progress by the invaders to the extent of nearly 33 years between the two battles. At that time the New Forest spread far beyond its present confines, right into the Salisbury basin in fact, and there is no reason why the invaders should not have worked along the fringes of the forest and secured river crossings (over the Avon or the Bourne) at other points. Or, as a first phase, penetration could still be achieved to the vicinity of Salisbury by establishing settlement on the east (New Forest) side of the rivers. This is in fact the position taken by the Winterbourne Gunner and Petersfinger cemeteries which are both at strategic points on roads leading to Old Sarum, near the points where these roads ford the River Bourne (at Winterbourne Gunner and Milford respectively). In these circumstances Old Sarum may have been outflanked, if indeed the Chronicle entry of 552 is to be read as meaning that the invaders fought a battle to obtain possession of it in that year, and had made no attempt to secure it previously. In fact, however, the archaeological evidence today points to Saxon settlement so near to Old Sarum before 552 that it can hardly have remained a hostile British stronghold down to that year, and the fact of a battle there need not imply that it did so. Professor Hawkes has even suggested to us that in 552 Old Sarum could have been long in Cynric’s hands already and the battle one of Saxon defence against a British attempt to win it back. Whatever interpretation be preferred, the Chronicle need in no way be at variance with the early date, well before A.D. 552, which archaeology now requires for the penetration of groups of English into the Salisbury area. 104. APPENDIX I THE HUMAN REMAINS All skeletons in the series were incomplete, and their surfaces much eroded. No signs of disease or abnormality were seen, other than dental disease. I. Male, aged 25 to 30 Skull nearly complete. Post-cranial skeleton incomplete and no long bones measurable. Cranial sutures obliterated, although the lambdoid can still be traced externally, but the dental attrition is moderate with M3 only slightly worn, so age estimated as above. There is slight torus mandibularis, and the chin region of the mandible is noticeably deep. Cranial measurements: Facial measurements: (L) 197 (G’'H) 81 (B) 138 (ZM.ZM) 97 (B’) 97 (G2) 41 (BY) 109 (J) 122 (St) 133 (N.B.) 33 (S2) 140 (S3) 116 Mandibular measurements: (S’1) 116 (R.B.) 35 (S’2) 124 (GnDt.) 38 (S’3) 92 (M?H.) 34 Il. Sex uncertain An aged individual, too fragmentary to measure. Part of the left ramus of a mandible was present, indicating that the left molars, at least, had been lost for some years before death. Ill. Male. ?Aged about 16 Left side and back of skull missing, skeleton incomplete. Cranial sutures: coronal and sphenoidal open, sagittal shows slight endocranial closure about the middle of its length. Cranial measurements: Facial measurements: (B’) 98 (G2) 38 (S1) 146 (3/1) 122 Mandibular measurements: (ZZ) 43 (RB) 32 (G’2) 75 IV. Male. Aged c. 20-25 Skull fairly complete. Skeleton imperfect. Cranial sutures: open, with sagittal and lambdoid beginning to fuse internally. Cranial measurements: Facial measurements: (L) 183 (ZM.ZM) — 95 (B) 149 (G2) 43 (BR) 88 (J) 125 (B’) 122 (N.B.) 24 (Sr) 134 (S2) 135 Mandibular measurements: (Sit) < 115 (ZZ) 47 (S’2) 116 (RB) 33 (GnDt) Pay (M?H.) 25 One femur is complete. 105 V. Male No skull present. None of the long bones are complete. A section of tibia shaft is somewhat platycnemic. VI. Male Middle aged, probably aged between 40 and 50. Fragmentary skull and skeleton. Calculus deposit on teeth, with alveolar recession. The remaining vertebrae show no sign of osteo-arthritic changes. There is a little arachnoid granulation on the inner table of the skull; this is fairly common in middle-aged skulls. Cranial measurements: (L) (180) measurement affected by post-mortem deformation. (B’) 92 Cranial sutures: coronal, sagittal, lambdoid, all completely closed internally, but still traceable externally. ‘The lambdoid is the most advanced. VII. Juvenile. Aged about 8 years, sex uncertain Fragmentary skull and skeleton, mandible almost complete. The inner end of the clavicle is stained green from contact with some metallic object. One upper outer incisor of an adult was found among the teeth. VIII. Female. Aged 35 to 45 Skull with imperfect base, mandible, incomplete skeleton. Tibia complete. Cranial sutures: coronal, sagittal and lambdoid all obliterated internally, slight traces still visible externally. Cranial measurements: Facial measurements: (ia) 178 (Zm.Zm) go (B) 199 (G2) 41 (BY) 98 (J) (117) (B") 110 (N.B.) 25°5 (Sr) 119 (S2) 110 Mandibular measurements: (S’1) 107 (ZZ) 46 (S’2) 108 (RB) 33 (BiB) 108 (GnDt) 30 (CYL) 21°5 (CH) 15 IX. Female Very fragmentary skull, female, with no teeth or jaws present. Middle aged? Only fragments of leg bones. Cranial sutures: the coronal and the short length of sagittal remaining are completely obliterated, but there are no signs of old age apparent otherwise. Cranial measurements: (B) 94 (BY) — (124) (Sr) 112 (S’r) 102°5 As far as I can judge, there is no evidence that any of the individuals were ‘foreign’ to the group, as suggested by the brooches found with two of the females. However, only one female skull was sufficiently complete for reliable measurements to be taken, so the evidence on this point is inconclusive. ROSEMARY POWERS British Museum (Natural History) 106 APPENDIX II DENTAL REPORT Six skulls with dentition ranging from complete to incomplete sets of teeth. A number of loose teeth, one skull, No. II, with only the left fragment of the mandible (ramus), and absence of teeth. Skeleton [X representing only cranial bones. Grave I complete dentition (pronounced attrition). II incomplete. IV incomplete. VI incomplete. VII mixed dentition (7$ or 8} years old) incomplete set of teeth. VIII incomplete. MEASUREMENTS MANDIBULAR TEETH: mesiodistal buccolingual N. R. N. 108 N. R. N. lie M; 4 10°5 4 10°2 4 g°2 4 9°4 M, 4 10°6 4 LO"? 4 10°00 4 10°O M;, 5 10°8 5 10°2 5 10°0 5 10°2 PM, 4 7:0 4 6-9 4 8-0 4 Lo PM; 2 70) 4 70 2 7°2 4 7-0 C 5 6-9 3 se: 4 7°3 2 7°5 I, 3 6-3 4 6-2 I 6-0 2 6-0 I, 3 5°3 2 5:2 I 5:0 I 5:0 MAXILLARY TEETH: M3 3 g°2 3 9°3 2 II-2 2 I1-O M? 4 9:8 3 10°2 3 Lid 2 11-7 M? 4 10°7 3 10°9 3 Lid 3 TI4: PM? 3 6-6 G: 7:0 3 8-8 2 g'l PM? 2 6-6 3 6-7 2 O°1 2 9:0 C 4 8-1 3 8-1 3 8-0 2 8-0 Z 3 6-9 2 6-8 2 6-0 I 5°5. 1S 3 85 3 7°83 2 6-9 2 6-2: N = Number. R = right. L = left. All measurements are in millimetres. MANDIBULAR DECIDUOUS SECOND MOLAR: I 9°5 I 9°5 I ous I 8-5 MORPHOLOGICAL CHARACTERISTICS Shovel-shaped Incisors Slight shovel-shaping on the lingual surface of the central and lateral incisors was: observed on the teeth of skulls IV and VII (SIV-Rt. It and I?, and SVII-Lt. I*). Lingual Tubercle Lingual tubercle was observed on the central and lateral incisor of Skull I (lingual surface of the incisors), Skull IV on the canines only, and on the lateral incisor of Skull VIII (maxillary incisors). 107 Carabelli’s Cusp Carabelli’s cusp was observed on the maxillary first molar of Skull III, which was characterized with a moderate expression of cusp on the mesio-lingual surface of the molar. Maxillary Molar Cup Patterns M: M:; M3 Skull I 4 4 — 3 Skull III 4 3+ — Skull IV 4 3 3 Mandibular Molar Groove Pattern M; M. M; Skull II Y5 +5 +4 Skull TV Y5 +4 +4 Attrition Pronounced or marked attrition was observed on the teeth of Skulls I, III, [V and VIII, and also on Skull VI where no measurements of the teeth present were taken. Alveolar Abscess Alveolar abscess was observed on the left maxillary second molar. The diseased bone involved the area surrounding or around the roots of M:. No evidence of caries present, but the occlusal surface of the tooth has been very much worn down, hence causing pressure or trauma on the underlying bone which may have resulted in a severe periodontal condition, thus affecting the alveolar bone. ‘Periodontal’ or Marginal Alveolar Bone Relation to the Teeth Loss of the marginal alveolar bone was observed in most of the skulls, which can be associated with the presence of calculus deposits around the necks of teeth. Skull III has a normal relation cf the marginal alveolar bone to the cervical portion of the teeth present. VIRGINIA CARBONELL British Museum (Natural History) APPENDIX III THE TEXTILE REMAINS Purse Mount, Grave I. Fic. 5: Ici Fragments of replaced textile on one side of the mount, the clearest being 4 x5 mm, Both systems Z-spun, twill weave, probably 2 <2, count 6 on 4 mm. by 7-8 on 5 mm., 1.e. ¢. 15 X15 per cm., a suitable gown or tunic weave. On the other surface, traces of Z-spun threads and an area of leather fibres. Throwing Axe, Grave VI. Fic. 6: VIa. Small area of replaced textile, roughly 2-7 x2 cm. at widest. Both systems Z-spun, weave coarse regular 2 x2 twill, count 9 x9 threads per cm. This would almost certainly have been a woollen textile, probably a cloak or blanket. which lay across the axe. ELISABETH CROWFOOT 108 ACKNOWLEDGEMENTS Our thanks are due to Mr. Cooke and Mr. Adams for permission to excavate on their land; to the Inspectorate of Ancient Monuments, Ministry of Public Building and Works, for providing certain special facilities including the cleaning, drawing and X-ray examination of the finds; to Mrs. E. Fry-Stone for drawing the finds; to the British Museum (Natural History) for the examination of the skeletal material which was undertaken by Miss Powers and Dr. V. Carbonell; to Miss E. Crowfoot for the examination of the textile remains, and to Mr. B. F. White and Mr. K. Grinstead for help with photography. We also thank Mr. Gerald Dunning, Miss Vera Evison and Professor and Mrs. C. F. C. Hawkes for their advice and encouragement during the preparation of this report, and the Members of the Salisbury Museum’s Research Committee Group for their assistance throughout. 1 Leeds, E. T., and Shortt, H. de S., An Anglo- Saxon Cemetery at Petersfinger, near Salisbury, Wilts., Salisbury Museum, 1953. 2 Arch., XxV, 259-78, 475-9- 3 W.A.M., XLVI, 155-6. 4 W.A.M., XLII, 94-101; XLVI, 153. 5 W.A.M., xLv1, 568-99; XLVI, 156-7. 6 W.A.M., XLv, 570. 7 W.A.M., xLIxX, 542-3, 503. 8 Bohner, K., Die frankischen Altertiimer des Trierer Landes, t (1958), 166-7. 9 Knocker, G. M., Proc. Hants Field Club, XIX (1958), 121. to Bushe-Fox, J. P., Richborough, u, 46 and Plate XIX, No. 34. 11 Tonnochy, A. B., and Hawkes, C. F. C., Antiq. Fourn., x1 (1931), 123. 12 Wheeler, R. E. M., and Wheeler, T. V., Research Report, 1x, Soc. of Antiquaries, 1932, fig. 19, No. 82. 13 Boon, G., Med. Arch., ut (1959), 80. 1% Hawkes, Sonia, Arch., xcvut (1961), 29. 15 Antiq. Journ., XXXV (1955), 20. 16 Knocker, G. M., Proc. Hants Field Club, xix (1958), 139. 17 Leeds, E. T., and Shortt, H. de S., op. cit. NOTE The Society is much indebted to the Council for British Archaeology for a grant towards the cost of publishing this paper. 109 THE SAXON BOUNDARIES OF FRUSTFIELD by C. C. TAYLOR THANKS TO THE WORK of Dr. Grundy, Dr. ‘T. R. Thomson and others, many of the Saxon Land Charters of Wiltshire have been put in their correct geographical position, even if in detail there are doubts about specific boundaries. 'wo charters which have not been assigned to any place are those which refer to grants of three hides of land in Frustfield. One dated a.p. 943 (Birch 782, Kemble 395) was pub- lished by Grundy, but with no geographical identification. The other, allegedly dated A.D. 968, was published by Dugdale,? Hoare,3 and reprinted by Matcham,¢ and is clearly a late copy, though referring to the same land. Both are included with charters of land in South Newton. The difficulty of these charters was ‘not a question of the position of a hundred, or even a parish, but of a small area of three hides so that the only possible clue to their locality would be an exact knowledge of Frustfield. Even were that known it would not necessarily be possible to trace the site of an area the bounds of which are not likely to be represented to any great extent by modern boundaries.’5 It is the purpose of this paper not only to give the general location of these three hides, but also to suggest the actual bounds of the land in question. The first necessity is to establish the whereabouts of Frustfield. The name occurs in many documents, usually as ‘the Hundred of Frustfield’. ‘This hundred, the smallest in Wiltshire, appears to have contained only the modern parishes of Whiteparish and Landford.* This general area can be narrowed still further. In Domesday Book? for the area covered by the hundred there are eight entries. One is for Landford (No. 528), two are for Cowsfield, in the eastern part of White- parish (Nos. 423 and 446) and the rest (282, 380, 471, 507 and possible 529) refer to Frustfield, spelt in various ways. Of these five entries, one (282), the land of Humphrey de Insula, is probably Whelpley Farm (N.G.R. SU/232240), a mile north-west of Whiteparish Village.’ Another (471), the land of Aldred, is certainly the eastern half of the present village, later known as Alderstone.9 Of the other three Frustfield references, one is probably the western half of the present village and another probably Moor Farm (N.G.R. SU/226221), 13 miles south-west of White- parish Village. This site is called More Hamlet and Manor in many later documents, '° and earthworks of the former hamlet can still be seen due east of the farm. From these entries in Domesday Book it seems that Frustfield was the name given to the western half of the present parish of Whiteparish and possibly the name of the village itself before the church, built about 1190, gave rise to a new name."! Having established the position of Frustfield, it is possible with the aid of one of the charters (B.782) to narrow down the area of the three hides still more. The introduction to the charter states: “These are the bounds on the west side of Fyrste rro Felda which belong to the three hides.’ Thus it is probable that these three hides lay towards the extreme western part of the parish of Whiteparish. The actual bounds given in both of the charters are so vague that even knowing the specific area it is almost impossible to trace them. However, by following the subsequent history of the three hides this difficulty can be partly overcome. The earliest charter (B.782) is a grant by King Eadmund to his thegn Wulfgar in a.p. 943. The later one is a grant of the same land by King Edgar to Wilton Abbey in A.p. 968. How the land came back into the hands of the Crown is not known, but the subsequent history of the land is easy to trace as it remained part of the Wilton Abbey estates until the Dissolution. After the original grant to Wilton there is no mention of it until the 13th century. It is not mentioned specifically in Domesday Book, though it is almost certainly included under the entries for South Newton’ where it is stated that the Abbey had the right of 80 cartloads of wood from Melchet Forest. The original grant of the three hides was included with the lands of South Newton and the whole of Whiteparish lay within Melchet Forest in the 13th century."3 By the 14th century at least, the land was leased by the Abbey at the annual rent of £5'4 and right up until the Dissolution the rent remained at this figure.ts The land is referred to either as Frustfield'® or more usually as Abbotstone'7 (the Abbess’ Farm).18 The present name ‘Abbotstone House’, given to a house in Whiteparish Village, is a 19th-century alteration from the older name of ‘Street Farm’ and has no connection with the original Abbotstone. By the late 16th century the land of Abbotstone was held by the Compton family who sub-let it to Richard Light, and the land is then called the Manor of Titchborne.'9 The later descent of this land is not necessary for our purposes. The important feature is that Titchborne Farm (N.G.R. SU/220219), 2 miles south-west of Whiteparish Village, close to the western parish boundary, together with its land, has some relationship to Abbotstone and the three hides. Indeed it is probable that Titchborne Farm itself is on the site of the original Abbotstone. From the Tithe Map of Whiteparish, 1842,2° it is possible to plot the bounds of the land belonging to Titchborne Farm, and this has been done (Fic. 1). If the above argument is correct, these bounds ought to have some relationship to the bounds of the original charters. However, there are four pieces of land that must be added to the Tithe Map boundaries of Titchborne Farm, which seem to have originally formed part of Abbotstone or Titchborne Manor. First, Privett Wood (now Cheyney’s Wood, etc.) in Downton Parish (N.G.R. SU/210237), 14 miles north-west of Titchborne Farm. This is specifically stated as belonging to Abbot- stone in a 14th-century Perambulation of Melchet Forest.2! ‘The second is Newhouse Park. This appears to have been originally part of Titchborne Manor sold in 1619 to Sir Edward Gorges of Longford, who probably built New House.?? Third, the land in the north-west of Whiteparish Parish between the A.36 road and the western parish boundary. This land was probably part of Abbotstone, and was sold in the early 17th century by the Stockman family who then owned Titchborne Manor,?3 and who appear to have enclosed this previously open downland and built a farm called Dry Farm there (at N.G.R. SU/217241). Fourth, 54 acres of As Ga A36 A The Hien -- Holl ‘500ft N 4 (7300 ft ~ \ be ‘ -. © pry see ; : FIRM So 40Q ft PRIVET E he. WOOD a ; : \ CLAPGATE zy COPS t i y ie ‘ea y ' S53 beh ; . i i a oe / ! 4 ea 1 BLACK i \ \ i DOWN ly ; , \ > Ve eae a NA AS Wheel Seat ee jee oeN RS TheWedge aaa ‘a = 2-2-5 : > | Hy AN aw ; Aes {Bend tn my a ~ The Hedve ? : t x ag ne A36 m% Chalk crs. cha “on as 5 . . s 59% - Pit ‘ \ pee tee Mecic 10 rg au are oer Foote \ ‘ _ MOOR >>. 4 FARM \ \Titchborne ‘e : \ Ne ne : \ Red Plough a rie8e R \\ Lands Ea NEWHOUSE seis ae “ PARK OF aes Boundury of Titchborne Farm 1842 ‘ rs ae oeeesccsees . | / ie comat Ps . ( 200 fe --.-, 76 2 Sat Parish Boundary Ne hae i ji as \ 1 MILE Tree i} Fic. 1 Bounds of the three hides at Frustfield. land in Downton Parish (since 1895 in Redlynch Parish) called Black Down.74 This land was part of Titchborne Farm certainly as early as 1689.75 If these four blocks of land are added to the land of Titchborne Farm it can be shown that the block of land thus formed is probably the land granted in the two charters. Using Grundy’s translation of B.732, the bounds may be as follows: 1. That is first at the Lea of the Swing Gate. Clapgate Copse lies 1 mile north-north-east of Titchborne Farm. The name Clapgate mean a Swing Gate.?6 The Lea of the Swing Gate probably lay north and west and south-west of the present Clapgate Copse, on land which was certainly still open until the early 17th century when Dry Farm was built and the land enclosed.?7 If the other bounds are correct, this point must be on the south-west side of the ‘Lea’. 2. From the Lea of the Swing Gate along the Hedge or Game Enclosure to Heath Combe. These two landmarks are unrecognizable, but they probably lay along the south side of Black Down. 3. To the Cut Wheel to the (Hedged?) Way. These are the first two bounds of the later charter. What the ‘Cut Wheel’ was, is again unknown, but it could be the place where two tracks cross near Langford Lane Wood (N.G.R. SU/207229) at the south-west corner of Black Down. The Way is presumably the north-south track running west of Black Down and Privett Wood. 4. From the (Hedged?) Way to the High Hill to the Rough Thorntrees. From the Way the boundary must have run up to the highest point on Standlynch Down, 504 ft. above O.D. (N.G.R. SU/207243), which must be the ‘High Hill’. This is confirmed without any doubt by the second charter which at this point gives ‘the ancient citadel’. On this hill, almost completely destroyed by later tracks and modern ploughing, is a Romano-British settlement of 20 acres, consisting of at least two enclosures and surrounded by ‘Celtic’ fields. It can still be seen on the ground, but it is clearest on air photographs.? The thorn trees, whose successors still grow there, must have lain just north-east of this settlement. 5. From the Thorntrees to the Bramble Thorns to the north end of the Dyke. The Bramble Thorns naturally cannot be traced. The Dyke may be either the dry valley draining south-east from the High Hill, or a ditch long since vanished on the same line. 6. Then along the Dyke to its south end. This should be near the modern A.36 and A.27 road junction. 113 7. From that end to the Way. This is probably the track running east-west near Clapgate Copse. Though now only a track it clearly continues the line west of the modern A.27 road running along the edge of the former forested area to the south. 8. From the Way to the Old Dyke. The Old Dyke is completely unknown. There is no ditch or any similar feature in the area of its apparent position. g. From the Dyke to the bend in Wulfstans Hedge. _ This bend, in the second charter the next boundary point after the Thorntrees, is probably one of the sharp bends in the boundary of ‘Titchborne Farm south-east of Clapgate Copse, which lies at the head of a dry valley (see ro below). 10. From the bend on the Head of the Coombe to the Chalkpit. In the second charter between the bend in the hedge and the Chalkpit is the ‘Foxes Hill’. The whereabouts of this is not known, but it may be in or near two fields called ‘Hilly Gilberts’ on the Tithe Map (N.G.R. SU/225226). The ‘Chalkpit’ is also unknown, but it must have lain near the modern chalkpit north of Moor Farm (N.G.R. SU/225224), for south of this the chalk is overlain by the sands of the Reading Beds. 11. from the Chalkpit to the Damp Meadow. The ‘Damp Meadow’ is almost certainly the valley of a small stream flowing south-west of ‘Titchborne Farm. In the second charter the boundary points given after the “Chalkpit’ are ‘the old garden’ and ‘the Military Way’ and ‘by the Way to the Dry Brook’. The position of ‘the old garden’ is unknown. ‘The Military Way’ is probably the lane running south-east from Titchborne Farm to Landford, which is in fact part of a continuous track from Landford along the western boundary of White- parish to Pepper Box Hill. 12. From the Damp Meadow to the Red Plough Lands. Redhills Farm (N.G.R. SU/239218)29 could possibly be the boundary, but as this part of the parish is on London Clay any cultivated land would be red in sharp contrast to the sand and chalk soils to the north. It seems more likely that the ‘Red Plough Lands’ lay west of Redhills Farm, and so the boundary would agree roughly with that of the second charter and the boundary of Titchborne Farm. 13. From the Tree to the Dry Brook. This brook is the only boundary point in the charters which appears in any other Wiltshire Charters. It is in the Surveys of that group of charters labelled Downton (B.27, 391, 863, 690 and K.698). It was identified by the Rev. Boulay Hill 114 as the stream which forms the south-western boundary of Whiteparish.3° This appears to be correct, and certainly in its northern part at least, it is a ‘dry’ brook for most of the year. 15. From the Dry Brook again quickly to the Lea of the Swing Gate. This completes the circuit of the first charter. In the second charter after the ‘Dry Brook’ it gives ‘then to its source and then to the Haesfloran(?) then to the Cut Wheel’. At the end of the first charter a note is added: ‘And the Lea on the Open Heathland is for the common use of everyone.’ These bounds appear to fit in fairly well with the reconstructed boundaries of Abbotstone, but the fact remains that they are very vague and the actual correlation may well be wrong. However, it does seem to be fairly certain that the three hides of land in Frustfield lay in the western part of Whiteparish, around Titchborne Farm, somewhere between the ‘High Hill’ to the north and the ‘Red Plough Lands’ in the south, the only two points which can be fixed with reasonable accuracy. G. B. Grundy, Arch. Journ., LXxvi (1919), 16 G. Matcham, op. cit., 68. 277-8. 17 Eg. Wilts. ILP.M., 1327-1377, 191. 2 W. Dugdale, Monasticon Anglicanum (1819 ed.), 18 Wilts. Place Name Survey, 388. Il, 324. > 19 G. Matcham, of. cit., 50 and 51. a0 Wilts. C.R.O. 21 V.C.H. Wilts., 1v, 428. 22 G. Matcham, op. cit., 51-2. 23 G. Matcham, of. cit., 46 and 51. 24 Tithe Map of Downton. Wilts. C.R.O. 25 Nelson Estate Papers. Wilts. C.R.O., 464/55. 6 Wilts. Place Name Survey, 451. 3 R. C. Hoare, Registrum Wiltunense (1827), 15. 4G. Matcham, Hundred of Frustfield (1844), 67. 5 G. B. Grundy, op. cit., 277. 6 Wilts. Place Name Survey, 386-91. 7 V.C.H. Wilts., 1, 113-68. 8 V.C.H. Wilts., 1, 146, n. 91. 9 The present Alderstone Farm (N.G.R. SU/245245), } mile north of Whiteparish Village, » is a late transferred name. The original Manor 27 To the south of Clapgate Copse is woodland, House of Alderstone lay near the Church. to the east lay the open fields of Whiteparish. G. Matcham, of. cit., 26. Thus the Lea must have been to the north, west 10 E.g. I.P.M. 1428 P.R.O. G/139/50/7, etc. and south-west. 28 R.A.F. Vertical Air Photograph CPE/UK 1811, no. 2150. 29 The farm is called Redhills on Andrews’ and 1 Wilts. Place Name Survey, 387. 12 V.C.H. Wilts., u, 129, no. 107. is oes Wilts., 1V, 427-31. Dury’s Map of Wiltshire, 1773. 4 Wilts. ILP.M., 1327-1377, 128. 30 W.A.M., xxxv1 (1909), 50-6. The other 15 Valor Eccl., u, 108. bounds identified in this area are more doubtful. 115 A PRE-CONQUEST CHURCH AND BAPTISTERY AT POTTERNE by NORMAN DAVEY IT IS RECORDED that in the latter part of the 8th century Potterne and its appurtenances were given to Sherborne, a see constituted in 705.' In gog a see was formed at Ramsbury in Wiltshire, and just before the Conquest, in 1058, Herman, then bishop of Ramsbury, united the two sees, Sherborne and Ramsbury, to create the bishopric of Salisbury. In 1075 he fixed the site of the new see at Old Sarum. Herman, one of Edward the Confessor’s chaplains, had been appointed to the bishopric of Ramsbury in 1045, and he leased the property at Potterne to various tenants. At the time of the Conquest, according to the Domesday Inquest, the tenants of the manor of Potterne were two Englishmen, one a nephew of Bishop Herman, and the other a man named Alward, who had bought his portion of land from Herman to hold for life only, with reversion to the bishop.? Herman was succeeded as Bishop of Salisbury by Osmund, a nephew of William the Conqueror. To him is ascribed the original ‘Use of Sarum’, which, later on in the early part of the 13th century, was organized and codified in a Consuetudinary and an Ordinal by, or at the instance of, Richard le Poer, dean and afterwards bishop, who transferred his seat from Old Sarum to New Sarum (Salisbury). The ‘Use of Sarum’ remained for several centuries preceding the Reformation the most outstanding of five ‘Uses’, the others being Bangor, Hereford, Lincoln and York. In about the year r1og1 bishop Osmund granted the church at Potterne, and the neighbouring one at West Lavington, to his chapter at Sarum. When the first church at Potterne was built has yet to be established, but it is likely that it was during the period of the ecclesiastical revival in the roth century, which resulted from the more settled conditions in Wessex. ‘That there was already a church at Potterne at the time of the Conquest, may be inferred from a statement in the Domesday Inquest which reads ‘what the priest of the manor holds is worth 40 shillings’. The site of the church remained unknown until 1962 when the writer found it in his grounds at the Porch House, at a spot approximately 100 yds. to the south of the 13th-century church of St. Mary the Virgin. It stood on Plot 205 shown in the tithe map of Potterne dated 1839 (Fic. 1). This particular plot was seen to differ from the others which made up the Porch House property, for there was no access to it. This led the writer to consider the possibility that it might be an abandoned plot of one-time consecrated ground, and for that reason, and unlike the neighbouring plots, no subsequent development had taken place upon it. 116 The conjecture appeared to be confirmed, for a trial trench revealed a slot cut in the greensand for positioning a foundation pad; this proved subsequently to be for the east wall of the baptistery at the south-east corner of the church (pL. VIIa). Subsequent excavation revealed that the whole church lay within the confines of Plot 205. St Mary’s Ch 195 X i Site of Timber. , 205 Church Dr Fic. 1 Site of the pre-Conquest church at Potterne and its relation to the adjoining tithe plots. During the erection of the church the site was cleared of topsoil down to the greensand, and in this latter longitudinal slots for positioning timber foundation pads and holes for vertical posts were cut. There was ample evidence from the pottery sherds in the primary layer which had accumulated on the floor of the building after its abandonment to show that the church survived into the 12th century, when it was replaced by a church built in stone, on or near the site of the present church. 117 It may be that the wooden church suffered neglect between the years 1139 and 1148, and fell into disrepair, for in the former year King Stephen, having quarrelled with Roger, bishop of Salisbury, seized Devizes Castle and other domains belonging to the bishop. In 1141 the property passed into the hands of Matilda, daughter of Henry I, and on 1oth June 1148, after Pope Eugenius III had acknow- ledged Potterne to be part of the episcopal estates, she formally restored the manor to the bishops of Salisbury in whom it remained vested until well into the 19th century.3 It is possible that during the nine-year period of its confiscation the little timber church fell into disrepair and that on its restoration to the bishops it may have been decided to replace it by a more substantial stone building on a more suitable site. A few stones of this Norman church, either loose or built into the walls of the present church, are the only traces that remain visible. Replacement of timber churches by stone ones seems to have been a common practice at this time. Excavation showed that the timber building had been completely removed. Following a decree made by Bishop Theodore (668-go0), it was forbidden to use the timbers of a church for any other purpose, and he explicitly forbade that timbers should be used for firewood.4 Whether this decree remained operative for very long is not known, but there was certainly no indication of timbers having been destroyed by burning on the site. All that remained was the matrix of slots and holes as shown in FIG. 3. The original building was very small and consisted of the three essential elements (FIG. 2): a nave approximately 15 ft. square internally, a presbytery or chancel, 8 ft. square, and a baptistery just over 14 ft. square. At a later stage the nave was extended westwards to give it a length of just over 21 ft.; chapels, C — CHANCEL. N- NAVE. B- BAPTISTERY S — SIDE CHAPEL. P~PORCH. Fic. 2 Plan of the pre-Conquest church at Potterne and stages of development. 118 approximately the same size as the chancel, were added on the south and north sides, and a small porch entrance arranged at the west end. These additions were laid out with a change in orientation from 82° E. to 87° E. of magnetic north. The discovery of a baptistery attached to the early church is particularly important and significant. Some of the early and larger Christian churches had separate baptisteries built in close association. Examples survive on the continent of Europe, but there are few if any remnants of them left in Britain. The only such baptistery known to have been built in England was apparently the work of Archbishop Cuthbert in the mid-8th century. It stood at the east end of the Cathedral at Canterbury, almost adjoining it.s The separate baptisteries were intended to serve the needs of a large area and often to be administered by the bishop of the diocese. During the roth century in Britain, however, when the parochial system took shape, bishops tended to delegate their powers and authority to the parish priest, and this meant that for purposes of baptism, fonts were installed in many village churches. An additional incentive to erect fonts resulted from a Canon enacted under King Edgar (crowned 959, died 975), which read, ‘we enjoin that every priest grant baptism as soon as it is demanded, and everywhere, in his shrift- district command that every child be baptised within xxxvi days’.® Potterne may represent an early and transitional phase in that the font was installed, not in a completely separate baptistery, but in one which adjoined the church, and not in the body of the church itself as in so many other instances. As the bishops owned the manor of Potterne and this church, it may be that they officiated on occasion at ceremonies in this baptistery, following firstly, no doubt, the ancient procedure outlined by Aelfric (955-c. 1020), monk at Winchester, Cernal and later Sherborne, and later that laid down by Osmund, who as bishop and owner of the manor may himself have done so. Aelfric in the 1oth century instructed that ‘with the holy oil ye shall mark heathen children on the breast, and betwixt the shoulders, in the middle, with the sign of the cross, before ye baptize it in the font water; and when it comes from the water, ye shall make the sign of the cross on the head with the holy chrism. In the holy font, before ye baptize them, ye shall pour chrism on the figure of Christ’s cross, and no one may be sprinkled with the font water after the chrism is poured in.’7 A recess cut in the greensand in the centre of the baptistery to accommodate a tub font (pL. VIIb) has dimensions which match those of the 1oth-century font preserved at the west end of the present church, and it is beyond reasonable doubt that this font came originally from the baptistery now discovered. Local tradition states that this font came from an earlier church. When the present church was being restored in 1872, the Saxon font was found below the floor level, under the present 16th-century font. It is possible that on removal from the timber church to the Norman church the font had been deliberately set down in the floor to facilitate the act of immersion, in much the same manner that it had been installed in the earlier church. The font was approached from the entrance to the baptistery down two steps, and access to the chancel from the font was up three steps, shown in FIG. 3. The stone flags which formed the steps were removed long ago, but the 119 PRE-GCONQUES). CHURCH. PO R.ERNE a &) = Present fimit of excavation 82°E mag) FOUNDATION MATRIX CUT IN GREENSAND 87°F (mag) Fic. 3 impressions remain, so that their size and shape are known. It is hoped to replace them in due time, so that the site can be preserved. The font which is one of the earliest in the country has already been described.® It is a tub font, for immersion, and is clearly based on a timber prototype. It is in the form of a tapered tube, 2 ft. 3 in. internal diameter at the top, with a circular stone 74 in. thick fitted in to form the bottom, and caulked with lead. A rim which bears the inscription in Latin, + SICVT CERVVS DESIDERAT AD FONTES AQVARVM ITA DESIDERAT ANIMA MEA AD TE DS. AMEN 4,9 is 120 rebated for a cover. It is impossible to date the inscription closely, but from its style D. M. Wilson of the British Museum considers that a roth- to 12th-century margin embraces its proper date. The inscription could, of course, have been added some years after the making of the font itself. In the base of the font is a draining hole, and this would have been located centrally above a soakaway revealed in the baptistery floor. There were also two recesses cut in the greensand close to the site of the font, and these could quite well have accommodated vessels containing oil or water for ceremonial use.?° The method of construction of the timber church was to cut shallow trenches in the greensand, and into these to lay timber pads varying in length from 2 ft. to over 5 ft., to form a level base on which to position the sills which carried the timber walling (Fic. 4; pL. VIIc, d). There were no signs of wattle or daub having been used in the construction of the walls, and it must be assumed that they consisted of vertical staves located on the sills, most probably in the manner adopted in the contemporary church at Greenstead, near Ongar (c. 1013), where the walls were formed of half trunks of oak set upright with the split face inwards, and where originally the upper end of each trunk was chamfered and roughly tenoned into a plate, and the lower end into a sill at ground level, and fixed with two pegs in each tenon. In the first instance the nave at Potterne seems to have been of the typical Anglo-Saxon ‘four-poster’ type, for there were four pads in the floor of the original nave, which may have formed either a direct bearing for four vertical posts carrying the roof, or alternatively may have provided a bearing for longitudinal timbers which in turn carried the vertical posts, in the manner illustrated by Braun.1! The type of construction was varied at the south-east corner of the chancel and along the south face of the baptistery. In the former position five very substantial vertical posts were set 3 ft. down into a trench cut in greensand. This may have been done to strengthen and stabilize the superstructure. Along the baptistery wall there were three substantial vertical posts, one at the south-east corner and two forming the portal entrance. There were less substantial vertical posts along the rest of the wall. Spaces between the upright posts could have been utilized as window openings enabling light to enter the chancel and on to the altar, also to give direct vision from the chancel into the baptistery embracing the font and the entrance beyond; and lastly to provide daylight to the baptistery and the font. The floor of the chancel, side chapels, and western extension to the nave were all about 10 in. lower than the floor level of the original nave, and were each entered down a step. The floor of the chancel, and that of the north chapel, were lowered by excavating the greensand, and clearly this was deliberately done. The lower floor of the south chapel and the western extension of the nave resulted more naturally from the falling away of the ground in the south-west direction. In the chancel floor and also in the floor of the south chapel recesses about 15 In. square were cut in the greensand, where pedestal altars may have stood. At the time of writing the north chapel has not been fully explored, and therefore it is not known yet whether there was a similar recess there as well. The primary layer, which varied in depth up to 6 in., contained many types 121 PRE-GONOUEST sCHURGCH. POT TERNE ee | Present limit of excavation 5. = @ — ! ARRANGEMENT OF | | FOUNDATION. PADS a] Je | AND PosTs. Reel Fic. 4 of pottery from the Roman period onwards up to the latter part of the 12th century. The Roman fragments, which included Samian and New Forest wares, were found in the area of the baptistery, and presumably had come from an uphill site to the south-east. The bulk of the sherds were, however, of 12th-century pottery types, including cooking pots with sagging bases, cooking pots with straight, nearly vertical sides, flat-bottomed vessels and vessels with inturned basal angle, scratched ware and ware with finger-tipping on rims. Glazed sherds were significantly absent, however, from the primary layer. This layer was sealed by a loamy soil varying 122 up to 12 in. in depth depending on the position on the site. This second layer was practically devoid of sherds, and presumably it represents a period when the site was lying comparatively undisturbed. Above this layer, however, the topsoil up to 18 in. thick, and in places disturbed by cultivation and planting, contained a large quantity of pottery sherds of varying date up to recent times. Green-glazed wares of the 13th and 14th centuries were present.?? CONCLUSION The evidence suggests that the timber church at Potterne lasted from the roth until well into the 12th century. The tub font, considered to be of roth-century date, and preserved in the present church of St. Mary the Virgin at Potterne, could have come from the baptistery of the timber church, for the shape and dimensions of the recess cut in the greensand to accommodate a tub font match those of the Saxon font; the inscription cut round the rim of the font is of roth- to r2th-century date, and is therefore contemporary with the timber church; the terminal date for the timber church, suggested by the pottery sherds in the primary layer covering the floor of the abandoned church, appears to be around the middle of the 12th century; the church may in fact have fallen into disrepair during its nine years of confiscation from the bishop by the Crown (1139-1148), and have been superseded by the new stone church soon after the restoration of the property to the bishops of Salisbury in 1148. ACKNOWLEDGEMENTS The writer wishes to acknowledge with gratitude the help given in many ways by friends, but particularly Messrs. R. E. Sandell, F. K. Annable, P. J. Fowler, P. Rahtz and E. H. Ruffett. ' V.C.H. Dorset, 1, 63, quoting Cotton MS. Faustina Au, folio 23. 2 V.C.H. Wilts., vu, 209. 3 Church Commissioners’ Document 136793 at Wren Hall, Salisbury. 4 J. Godfrey, The Church in Anglo-Saxon England (1962). 5 J. Godfrey, op. cit. 6 Benjamin Thorpe, Ancient Laws and Institutes, Il, 247. 7 Aelfric’s Epistle, entitled Quando Dividis Chrisma, reproduced in Benjamin Thorpe, of. cit., 390-3. 8 W.A.M., Lut (1950), 462-3. 9 Psalm 42, verse 1. to The name Potterne may have derived from this baptistery. In 1165 the spelling was ‘Poterna’, and it has been assumed to derive from O.E. pott, meaning ‘pot’, ‘hole’, ‘pit’, or ‘hollow’, and O.E. gern, meaning ‘house’, ‘store’, or ‘container’ (The Latin urna means a ‘waterpot’, ‘waterjar’, or ‘urn’), and the word Potterne has been variously interpreted as ‘the house where pots were made’, or ‘the pot house’. The writer suggests that the word ‘Poterna’ could equally well have meant ‘the urn, or water pot in a pit, or hollow’, and have referred explicitly to the baptistery, which must clearly have been a very important feature of the Bishop’s manor. 1 H. Braun, An Introduction to English Medieval Architecture, 118. 1% JT am indebted to Messrs. P. Rahtz and P. J. Fowler for their observations on the pottery. r22 SAXON AND MEDIEVAL FEATURES AT DOWNTON, SALISBURY by PHILIP A. RAHTZ PREHISTORIC AND ROMAN settlements at Downton, excavated in 1953-57,! have been described in previous issues of this magazine. This final report describes Saxon and Medieval features and finds from Castle Meadow and the Old Court site (see area and site plans in W.A.M., Lv (1962), 116-17; figs. 1 and 2). THE SAXON GRAVEL PIT IN CASTLE MEADOW The pit (section, FIG. 1) was visible only as a large depression about 100 ft. in diameter and 1-2 ft. deep. A trial-hole in the centre of this in 1956 showed that disturbed soil continued to a depth of at least 6 ft., so the hole was extended into a trench, parts of which were excavated down to the undisturbed natural, as shown in Fic. 1. The actual pit is between 60 and 70 ft. in diameter, and roughly circular, with a maximum depth of 13 ft. from the present surface, or 14-15 ft. from the original surface. It is mostly filled with a loose soil containing Roman and medieval sherds (the latter down to 4 ft. only; layer 2 in section), and it seems probable that this is derived from ploughing. The surrounding area has been slightly eroded in this process, accounting for the larger depression visible today. The pit is dated by a layer of charcoal and Saxon sherds near the base (layer 3), separated from the base only by layers 4, 5 and 6, which may be interpreted as primary and quick silting and weathering of the sides of the pit. It is evident that the accumulation layer 3 does not long post-date the digging of the pit, and was apparently tipped in from the north-east side, i.e. from the direction of the Moot. The dating layer was also seen, as a less distinct black layer, on the south-west side of the pit, and is separated from the plough-soil in the centre by a clayey layer, which yielded only Roman sherds, presumably derived also from ploughing the other parts of the field. Worked and unworked flints, of Mesolithic and later types, were also frequent in layer 2.3 The gravel from this pit is most likely to have been used in the Saxon period as a make-up. This would hardly have been necessary in Castle Meadow, where the sub-soil is mostly hard and relatively dry. It would, however, have been essential for any building on the flood-plain below the terrace. Such a building would not be far from the gravel-pit, and may perhaps be looked for in the meadow immediately north of the pit and below the Moot. Here are substantial medieval foundations which were seen by General Pitt-Rivers in the last century.4 The site is known locally as Old Court (and also as King John’s Palace and the Bishop’s Palace).5 124 It is possible that these buildings had a Saxon precursor. Indeed, only a few hundred feet away three late Saxon battle-axes were found during the construction of the generating station.® A trial hole was dug in the field known as Old Court in August 1957; the stratification is shown in Fic. 2. The hole was fortunately sited, as a medieval wall-trench crossed the extreme south side, from west to east at a bearing of 105° east of magnetic north. The lowest gravel layer encountered (layer 6) may have been derived from the pit, but only further excavation could prove this. The pit was apparently used as a rubbish dump after removal of the gravel, and since occupational debris would hardly have been carried up from the Old Court site, it is suggested that layer 3 is in fact derived from Saxon occupation on that part of the terrace where the Moot now lies. The sherds are described on pp. 127-28, and shown in Fic. 3, and pt. VIII; they may be of middle or late Saxon date. THE LATE SAXON AND MEDIEVAL OCCUPATION IN CASTLE MEADOW The Saxon gravel pit and its possible association with the ‘Old Court’ site have been described above. It remains to consider the late Saxon and medieval occupations of Castle Meadow, and their relationship to the Moot. The only features which may be dated to shortly before the Norman conquest are the ditches in the road junction area (plan of junction cuttings, W.A.M., Lv (1963), fig. 13). They cut features associated with Roman roads (ibzd., fig. 14, sections S16, $18 and S1g). They are D48, D4g and D57. D48 appears to end in Trench 4 (zbid., fig. 13). D49 does not appear to go very far to the west (see W.A.M., Lv, fig. 2, Castle Meadow), but may extend further east. D57 is apparently localized to the junction area. They may be associated with some dumping of clay in the Roman ditches (though there is also clay in D48). They are thought to be contemporary only because they are all secondary to the Roman roads: but they may in fact be of widely differing dates. The only dating evidence is from Ditch 49; in the base of this, in Trench 5, were several sherds of a cooking pot which may be of pre-Conquest date, probably of the 11th century (p. 128, FIG. 3: 7). Ditch 336, in the Beaker area, yielded sherds of an early medieval cooking pot in its base (p. 129). A few dozen medieval sherds (p. 128) were found in the upper levels of Castle Meadow; some of them, together with a coin, and two ditches, were found by the Cambridge University excavation in the Mesolithic area.7 Any finds such as these from Castle Meadow may reasonably be thought to be associated with the major earthwork known as “The Moot’ (see area plan W.A.M., Lviu, fig. 1). This is not the place in which to discuss this massive structure in detail, but it seems worth pointing out that the one thing it is unlikely to be is what its name implies—and what it is locally considered to be—a Saxon Moot. ‘The amphi- theatre-like terracing which has given rise to this view, is wholly the result of land- scape gardening.* There seems to be no reason why the earthwork should not be a ring-work of Norman date, presumably the castle of Downton built in 1138 by Henry de Blois, Bishop of Winchester, and plundered in 1147.9 125 ‘Arayjod uoxeg ‘u0jUMO § ‘oq 6 ‘Old NOLNMO *LYnOdD GIO ———— 100.4 INO aE rae eal “uscwVd OILvVAVOXINA “5-9-2 337 URE ORCAS: ONILOO.4 WAAN TIES ADD TVenLyn, FOVISNS ONIAIT LEYS éIld jonos6 MOP2IW B/1S0DE “IN INVW A sy ———— SFHIN "UOMOSE ~J00/f 4/0? PIIPENS HW INTHL TIM WABI SGHINSYL T7YM FWAROIW WOHs 4SvIEN wnt jO_~ =~ p3ysddofias) ~ W9es6 halo f= TAIT YOOTY WABI “| I ‘OL 250 S EV “S{Ul/J SNEBIYSO PUP AYTD GDIHDITA TWOISWHD NMOYE SI/LS- let d qi 4 AV YD) NOX VY S (SOYZHS NOxvS) TWwODYvHD PUE TOS ABIAVID NMOXG Mev: APAVID FYOW >" SE NOINMOTW j SNLIYLIG HIVHD(IWYNLYN) Q3LVAVOX3NN | nopvayt a21svg | | ANIDS 39HV7T PUB TRAVYD FJWOS 'WOS AONWS NMOYE JING © --~AvI2D AONYS JONvYO FINE qos: ANN i t LSIM-HLNOS LSVI-HLION | SAXON POTTERY FROM THE GRAVEL PIT About fifty sherds (Fic. 3, pL. VIII) were recovered from a limited area of layer 3 of the gravel pit in Castle Meadow;!° most came from the deepest part of the trench in an area of 6 ft. « 4 ft.; there are doubtless several hundred more still in the pit and lying too deep to be disturbed by building operations. The sherds represent at least four vessels, and possibly as many as eight. The fabric of all is hand-made, with an uneven surface similar to Iron Age pottery. Their colour is black, but some have brown or reddish surfaces inside or out. Some are heavily ‘grass tempered’! to such an extent that the surfaces are covered with tiny grooves where the grass or similar material has burnt out; others are seen to be ‘grass-tempered’ in section, but the surfaces are smoother. One pot (No. 6), though containing some grass, is much more sandy. A few sherds have what looks like burnt matter adhering to their inner surfaces. Only one pot (No. 1) is represented by enough sherds to show its form; of the others, No. 6 was probably a large vessel, to judge by its thickness. The sherds are described in detail below; they are among the first Saxon sherds to be found in Wiltshire,'? other than from cemeteries. Dr. Brian Hope-Taylor, F.S.A., who has excavated large quantities of Saxon pottery from Old Windsor, has examined these sherds; in his view they are not really datable, but might tentatively be attributed to the 7th-8th centuries 3%3 the sherds are neither sufficiently numerous nor individually characteristic to permit any closer dating. Minor variations in fabric and form, locally and regionally, make a large sample essential for dating purposes. At Old Windsor™ nearly every phase produced a range of fabrics similar to these from Downton, and sherds could only be dated there by statistical analysis of large quantities. Mr. Hope-Taylor says that a few of the sherds from Downton are of a fabric which occurred frequently in late contexts at Old Windsor, but that there are no specifically late forms among the few rims. DETAILED DESCRIPTION (FIG. 3: I-3) 1. Complete section of bowl, two joining sherds; there are three other rim sherds which do not join, and eleven further sherds probably of this vessel. The fabric is dark grey with buff-brown exterior surface towards the base; it is “grass-tempered’, with a very little fine grit, making the section appear irregularly laminated. The interior is heavily incised with grooves, mostly horizontal; some of these could be the result of burning out of grass in the surface, but most look as if they were the result of wiping the inside of the bowl with a bundle of grass or a coarse brush before firing; the outer surface has a few ‘burning-out’ grooves, but most of the surface has been carefully smoothed, probably with the fingers, and it is quite polished on some ‘high-spot’ areas, either by burnishing or by use. The sherds are heavy in texture, and rather friable; a low firing temperature may have been responsible for their slightly ‘soapy’ surface. The drawing is based on that by Mr. G. C. Dunning,’ but the rim is irregular, and the bowl may have a rather wider mouth with more sloping sides. 2. Sherd with part of a small horizontally perforated lug, of similar fabric to No. 1 and possibly from the same pot; the sherd is the top or lower half of the lug (shown as lower in the figure) and appears to have been made by pulling out part of the clay of the sey into a ‘nose’, which was then perforated with a small tool or stick hardly more than ¥ in. in diameter; the groove representing the inner edge of this hole extends laterally beyond the sides of the lug. 127 3. Rim sherd, probably of a jar or a bowl with everted rim; the fabric is black and soapy, and heavily ‘grass-tempered’ with no visible grit; both surfaces are ‘crazed’ with a network of fine ‘burning-out’ grooves; there are four other sherds of this vessel, and several others which might belong, but whose surfaces are rather smoother. 4. (Not illustrated.) Three joining sherds and five others from near the base of a large heavy vessel of probably more than 12 in. diameter at the base; the joining sherds are ¢. 5 in. in maximum dimension, and }-? in. thick. The fabric is dark grey to black, with a reddish-brown exterior surface, and a brighter reddish-brown in places inside; it is ‘grass-tempered’, slightly soapy ware, with no ‘grooves’, but with much more grit, giving it a more.sandy feel. Another sherd of similar fabric but only + in. thick may be from another part of this pot. 5. (Not illustrated.) Two sherds of dark grey ‘grass-tempered’, soapy, slightly sandy fabric, } in. thick, black with dark grey-brown surfaces, with some black (apparently burnt) substance caked on the inner surface and a few ‘burning-out’ grooves. This pot is probably not more than 8 in. in maximum diameter. 6. (Not illustrated.) Four fitting sherds (making sherd 4x3} in.) and one other, of a large vessel c. 11 in. diameter at this point; the upper part of the joined sherd is } in. thick, the lower part } in.; the fabric is ‘grass-tempered’, slightly soapy, with some fine grit, and a few pieces of larger grit up to 3; in. diameter, which looks like flint; the colour is dark grey with reddish-brown exterior surface and buff to dark grey inside. The surfaces show only a few small ‘burning-out’ grooves, but there are irregular circumferential grooves on the inside which may result from smoothing with the potter’s hand on a turntable. LATE SAXON OR EARLY MEDIEVAL POTTERY FROM CASTLE MEADOW These sherds (FIG. 3: 7) were found in Ditch 49 (see p. 125). They comprise a rim sherd, fitting shoulder sherds, and thinner body sherds, some of which join. The rim sherd does not join the shoulder sherds, and the drawing of the upper part of the pot is based on the comparative thickness and angle of the neck and rim sherds; it must not be taken as necessarily accurate. The shape of the rest of the pot is conjectural, as is the position suggested for the thinner body sherds; the reconstruction is based on a sketch by Mr. G. C. Dunning. There is a slight ridge on the inside of the rim, which is not clear on the drawing. The fabric is dark grey, coarse, hand-made, gritty, with slightly soapy surfaces; the exterior is uneven, with some irregular thumb-pressed ‘dimples’; the thinner body sherds are more reddish. Mr. G. C. Dunning suggests that this pot belongs to the long-lived Saxon tradition in the south'® and that the fabric is approaching that of scratch-ware, so that an 11th-century dating seems likely; he compares the form with that of pots from Oxford.17 MEDIEVAL POTTERY FROM CASTLE MEADOW A scatter of a few dozen medieval sherds was found in Castle Meadow. None of those found in medieval features (see p. 125) is worth illustrating, and _ this pottery is described here only as a record of the fabrics represented. A. Scratch-marked pottery: coarse, sandy, grey, sometimes reddish surface, or reddish-brown with blackened surface; scratch-marks range from coarse obvious 128 scratches to slight brushing only visible in cross-lighting. This type of pottery is well known in the Salisbury district; recent excavations at Laverstock'® suggest that its dating range may extend from the 11th-13th centuries, though the later examples are rather finer than those from Downton which should be r1th-1ath century. B. As A, but without scratch marking; also common in 12th-century levels at Old Sarum. C. Green glazed ware: medium fine off-white, pale grey-buff with varying colour and uniformity of glaze; some of these may be from the Laverstock kilns; all are probably of 13th- to 14th-century date. Sherds from features excavated are as follows: From the upper 3 ft. of the Saxon gravel pit came sherds of all three fabrics. In Ditch 45% there was a single sherd of fabric A and a fragment of 13th- to 14th-century glazed roof crest. In Ditch 336 (p. 125) were three sherds of fabric B, one 3 in. across. t For the Ancient Monuments Inspectorate of the Ministry of Public Building and Works. 2 W.A.M., Lv (1962), 116-41; ibid. (1963), 303-41. 3 Also medieval scratch-marked sherds at 39 in. 4 Arch. Journ., xxxm (1875), 290-309. The Earthworks of the Wiltshire Avon, by G. T. Clark, who dug with the General. 5 See The Moot and its Traditions, by E. P. Squarey (1906), 7, who described visits by King John in 1206, 1207, 1209 and 1215, and a special court held as late as 1578. 6 In the collections of Salisbury Museum; information from Mr. H. de S. Shortt; see W.A.M., XLV (1931), 489, and xLvr (1932), 171. 7 P.P.S., xxv (1959), 212 and fig. 2; the coin is there stated to be of 13th-century date. 8 Squarey, op. cit., 24. 9 Ibid., 29, quoting Passages in the History of Downton, A.D. 1138-1380, chiefly from the Public Records, by the Rev. I. Kestall Flayer, M.A., F.S.A., in W.A.M., xxtx (1897), 102-12, who in turn quotes the Annales Monastici Winton. 10 See W.A.M., Lvmt (1962), 117, fig. 2, for position, 11 This term has been adopted by Mr. Hurst (Med. Arch., 11 (1959), 21) and is distinct from “grass-marked’. 1 There is a single sherd of uncertain date in the Avebury Museum. 13 See Med. Arch., ut (1959), 13-31, for general discussion of Middle Saxon pottery; here Mr. Hurst suggests a Middle Saxon date for the Down- ton sherds, probably before the 9th century (p. 25), but Mr. Dunning, in the same paper, would date them to the Late Saxon period; no. 1, above, is shown by him in fig. 9, no. 6, from which our no. 1 has been redrawn. 14 Med. Arch., 1 (1958), 183-5. 15 Med. Arch., ut (1959), fig. 9, no. 6. 16 His Group A (Insular pottery) ; see Med. Arch., UI (1959), 31-49. 17 Clarendon site, Oxoniensia, xxi (1958), 45, two pots each in figs. 11-12. 18 Information from J. W. G. Musty. 19 See W.A.M., Lv (1962), 116-41, fig. 3, ibid. (1963), 323. 129 THE SUBURBS OF OLD SARUM by JOHN MUSTY and PHILIP A. RAHTZ SUMMARY DURING THE last few years there has been a change in the use of the land around Old Sarum, and a small part of Old Sarum Farm has been sold for building. New ploughing, pipe trenches and an excavation (in 1958) of the land sold for building have provided fresh information on the Old Sarum suburbs. New lght has been thrown on the Roman road system which converged on the East Gate. It was replaced by four medieval roads and these were located in the 1958 excavation, together with a building of 13th- to 14th-century date and many cesspits. Chance finds resulting from ploughing and pipe-trench digging included a series of mass graves and pits of medieval date. PREVIOUS HISTORY Old Sarum has been, in succession, an Early Iron Age hill-fort, a meeting place of Roman roads and a motte and bailey castle. In its decline it gave its name to a rotten borough. The hill-fort was transformed into a motte and bailey castle by throwing up a mound in the centre of the hill-fort area and placing on this mound the keep and other principal buildings of the castle. The remainder of the hill-fort interior then became the bailey of the castle; it was an unusual one in that it also contained the cathedral. Braun! in his scheme for the evolution of medieval Old Sarum also sites the Saxon town and part of the medieval city in the bailey. However, it has yet to be proven if any part of the city other than the cathedral lay inside the earthworks, nor is there any direct evidence for the existence of the Saxon town within the area, but there is no doubt at all that large areas of the medieval town — existed outside the earthworks—which latter would have exercised an unnatural constraint to its growth. These areas can be described as the East, West and South Suburbs. There is no evidence for a north suburb and its absence is virtually proven.? The East Suburb was first noted in 1540 by Leland, who also recorded the former existence of a suburb outside the West Gate of the castle.3 The only building of any age now standing in the East Suburb is the Old Castle Inn. The site of the West Suburb is a ploughed field, and fields also cover the South Suburb. Burgage tenures+ in the East and South Suburbs conferred the right on certain individuals to return two members to Parliament, but this surviving relic of the town of Old Sarum disappeared with the passing of the Reform Bill. The first archaeological excavation in the suburbs was carried out in 1855. Akerman, in a search for Saxon cemeteries following his excavation of the cemetery 130 at Harnham, carried out an excavation near the Old Castle Inn and discovered several skeletons, including one buried with a paten and chalices (p. 143, pL. IX). Further skeletons were discovered during work in connection with the building of a by-pass road in 1931,° and again during excavations in 1933 by Stone and Charlton.7 The latter also discovered a building. In the last two years there has been a change in the use of the land around Old Sarum following the sale of Old Sarum Farm. Old pasture has been converted to plough, and part of the farm (that between the old Amesbury road and the 1931 by-pass road) has been sold for building. In addition, several pipe trenches have been cut through the area. The new ploughing and the pipe trenches have all produced archaeological finds, and the land sold for building was the subject of an excavation in 1958 (during four weeks in November and December) under the auspices of the Inspectorate of Ancient Monuments, Ministry of Public Building and Works.’ The results of this excavation will be described, as will the finds resulting from the ploughing and pipe-trench digging,9 with the exception of those from the first stages in the construction of the Castle Hill reservoir which have already been reported.'° The latter finds have, however, been shown on the plan of the suburbs for the sake of completeness. Leland, in 1540, said that ‘there hath bene houses in tyme of mind inhabited in the est suburbe of Old-Saresbyri; but now ther is not one house neither within Old-Saresbyri, nor without it, inhabited’. He also mentions the parish church of St. John and the existence in his day of a chapel still standing on the site, and of ‘another over the est gate whereof yet some tokens remayne’. Benson and Hatcher? in their plan of Old Sarum show St. John’s Hospital in the position now recorded on the 25-in. O.S. map, and the Church of the Holy Cross between the East Barbican and the modern Salisbury-Amesbury road; they describe (p. 62) the ‘Chapel of Holy Cross’ as without the (east) gate, and quote references of 1237 (when Henry III maintained its chaplain) and 1280 (the chaplain’s salary). They also mention (p. 62) the Chapel of St. Mary ‘in the tower’, and the Chapel of St. Nicholas; the maintenance of its lamp in 1246, and the ordering of 10,000 shingles for its roof; at St. Nicholas they record that there were five chaplains, but only one was retained in 1273. They also describe further churches of St. Peter and St. Lawrence, which enjoyed privileges in 1281 (St. Lawrence was possibly the present church of Stratford-sub-Castle). For St. Peter, they quote an Jnqu. post mort.t2 as follows (p. 59): ‘the mill of Sarum with a mill 10 m. nothing belongs to the castle that is outside the castle—with regard to the advowsons, John of Sturminter (1255) holds patronage of Church of St. Peter worth } m.’; and record an Inquisition of Edward I in which an unnamed church was in the King’s advowson and vacant in the third year of Edward I. Jackson, in 1867,%3 also listed the chapels of Old Sarum: in 720, a chapel of St. James (in a charter of Ina); a chapel of the Virgin Mary, long maintained in some part of the fortress, apparently of older foundationthan St. Osmund’s Cathedral; Our Lady’s Chapel, im the cathedral, which was still standing and maintained at the time of Leland’s visit; the Church of the Holy Rood, called in the reign of Edward II ‘the Chapel of Holy Cross’ ;'4 the Parish Church of St. Peter, of which 131 the incumbents are named in the Wilts. Inst. from 1298-1412 (quoting Benson and Hatcher); a reference in 1381 to the ‘Free Chapel in the castle of Old Sarum’; and the Church of St. John in the East Suburb (named in the Val. Ecc. as ‘the Hospital or Free Chapel of St. John, near the castle of Old Sarum’). The Church of St. Peter is further referred to in 1327:15 “Thomas de Tolre, parson of St. Peters of Old Sarum, claimed in 1327 that his predecessors had tithes, etc. of the men dwelling in the castle of Old Sarum until a certain chaplain appointed to celebrate in the Chapel of the Holy Cross in the said castle, occupied the said: tithes. Sixteen men of Old Sarum say that the castle of Old Sarum is notoriously situated outside che bounds of St. Peter, and that after the foundation of New Sarum, priests celebrating for the time being in the said Chapel of the Holy Cross received these tithes.’ In the V.C.H.76 there is a reference to the Hospital of St. John the Baptist and St. Anthony, near Salisbury Castle: this was a leper hospital as early as 1195, received a grant of royal protection in 1231; is referred to in 1255, 1260, 1348, 1361 and 1387; was in existence in 1535 when tiles had been removed; and owned half an acre of pasture and fifteen acres of arable (Val. Ecc.). Montgomerie,'7 in 1947, referred to ‘the great (east) gate, which carried the Church of the Holy Cross’. The 25-in. O.S. map!® shows the ‘site of? St. John’s Hospital in field 2372, some 250 ft. north-east of Old Sarum Farm, and the ‘site of? Holy Cross Church just south-east of the south-west corner of the barn of Old Sarum Farm (see Fic. 1); the reasons for their placing are now unknown. These are the only buildings referred to. It seems clear that the ‘Free Chapel in the castle’, the Chapels of St. Mary, the Virgin Mary and that of Our Lady, are alternative names for a chapel within the precincts of the castle; St. Lawrence was the church of Stratford-sub-Castle: the positions of the pre-conquest Church of St. James and the Chapel of St. Nicholas are unknown; the Chapel of the Holy Cross on the Holy Rood was ‘over’ or outside the East Gate;'9 it was apparently closely associated with the castle, and at one time received the tithes of the ‘men in the castle’. It would also seem that the Churches of St. Peter and the Chapel or Hospital of St. John (the Baptist) (and St. Anthony) were outside the castle, and probably in the East Suburb. The 1958 excavation has shown that the Chapel of Holy Cross does not lie where the site is marked on the 25-in. O.S. map; it must be looked for in the earthworks of the East Gate, or between there and the present main road, unless it was destroyed by the deep cutting made for the main road in 1931. In 1933 a building was found by Stone and Charlton which they tentatively identified as part of St. John’s Hospital,?° though it was not at the place shown on the 25-in. map; there were foundations of puddled chalk on flint, well-tooled Chilmark stone, greensand, shale, Cornish slate and ridge crests, Purbeck and Portland stone, and plaster floors. There were also many skeletons, from a cemetery which had been partly destroyed when skeletons had been found in 1834, 1881, 1931 and 1932; the easterly limits of this cemetery were located in 1960. Twenty skeletons were found in 1931 on the north-west side of the site excavated by Stone 132 and Charlton (in 1933) which is very close to the north-east corner of the present excavation; although no skeletons were found in 1958, the north-east part of the area was only slightly explored; it seems likely, however, that the cemetery did not extend in a westerly direction as far as the road junction. The site of St. Peter’s church remains unknown. A number of roads were found in the 1958 excavation and there is no previous record of these. The tree-lined scarp on the south-east side of the site was thought to mark the line of the Roman road from Dorchester to Old Sarum, and possibly even to represent part of its agger; but the 1958 excavation has proved this bank to be of medieval or later date, and there is no reason to think that any of the roads found are Roman. The scarp continues in a south-west direction down the slope, a Se rane ///] a ite /// Sk RAN \v\\ \\Ad/ Why, iy i, tt f/f 2 Cathedral >= ay ROMP ST JOHNS HOSPITAL +e Pl SS, aaa ° Z od PIT 1957 CHALK PITS N.GESS.PITS. | 4, \ > Zs SLIME KILNS “ay SS H /3 4 1957 ‘ ie Re - SINGLE ERNE ce = . “CESS PITS = \ ‘ ROY Ez WO » “SEEN AFTER Ny OLLI sy” PLOUGHING 1961 / / /, Wy fea tt HTT iy HH OF OLD SARUM Feet SITE OF PARLIAMENT * TREE Fic. I 133 and here it may be a continuation of one of the roads found in 1958; the 1680 plan, of the Burgage Tenures at Old Sarum, shows the scarp further down the slope as ‘the beginning of Portway as it was formerly, ye western road to London before Harnham Bridge was built’. THE SUBURBS IN PRE-MEDIEVAL TIMES Despite the large amount of excavation work which has been carried out in the Old Sarum suburbs, scarcely any evidence has been obtained for occupation prior to the medieval period.?! Evidence of prehistoric and Roman occupation was obtained inside Old Sarum?? and on Bishopdown just east of the suburbs. The latter settlement area is extensive and in a sense might be considered as the prehistoric suburb of Old Sarum. However, five Roman roads are aligned on Old Sarum and four of these must meet outside the East Gate, namely the roads to Cunetio, Silchester, Winchester and Dorchester. It has been pointed out?3 that visible remains of Roman roads are absent in the immediate vicinity of Old Sarum, but in 1961 the side ditches of the Silchester road were exposed in a pipe trench within 400 yds. of the East Gate. During the 1958 excavations a number of roads were exposed outside the East Gate, but none of these could be shown to be Roman and they are probably medieval (see p. 136). THE ROMAN ROAD TO SILCHESTER In 1961 a pipe trench was laid by the Esso Petroleum Company4 from Fawley to Avonmouth. This trench passed through Clarendon and Laverstock, traversed the northerly slopes of Bishopdown and then crossed the fields at the westerly end of Old Sarum aerodrome. The trench provided an opportunity to test the line of the Portway (the Roman road to Silchester). Near Old Sarum the Portway is covered by the modern metalled road which passes through the built-up area of Old Sarum aerodrome and this stretch is aligned on the East Gate of Old Sarum, but at a point near the Old Isolation Hospital the modern road bends northwards to join the Salisbury-Amesbury road at “The Beehive’ (the former turnpike house). It has always been assumed that the > Roman road continued on its straight alignment from the Isolation Hospital, but this assumption has been based on the idea that Old Sarum was an important Roman centre. An alternative hypothesis might have been that the Roman road also veered northwards to skirt Old Sarum and join the road to the Mendips, the course of which road in the vicinity of Old Sarum has not so far been discovered.?5 However, the side ditches of the Roman road were exposed in the Esso trench and indicated a line which is a prolongation of that in the built-up area of the aerodrome, thus confirming the accepted line for this road (as shown on O.S. maps). The side ditches were all that remained of the road, and we must assume that either the agger has been completely ploughed away or, alternatively, the ditches were the main feature (i.e. marking the line of the road and defining its eye limits) and the agger was never very pronounced. The side ditches were 100 ft. apart, centre to centre (although it should be pointed out that the pipe trench did not section the road at right angles) and this compares well with the 93 ft. observed by Wright?¢ in his excavation of the Portway in Newton Tony parish at a point some seven miles along the road to the north-east. The profiles of the ditch sections also compare well. These were of V-shape, 3 ft. wide and 2 ft. deep (cf. Wright’s ditches which were 3 ft. wide and 1 ft. 3 in. deep). The ditch on the south side had an earthy chalk rubble fill, whilst that on the north had a humus fill which was slightly chalky at the bottom. Beyond the side ditches were broad U-shaped ditches 9g ft. to 10 ft. wide and approximately 4 ft. deep. The inner lip of the ditch on the south side of the road was 5 ft. from its corresponding V-ditch; that on the north side was 33 ft. from the \V-ditch on that side. THE ROMAN ROAD TO CUNETIO After leaving the aerodrome area, the Esso pipeline crossed the Amesbury road and headed for Woodford. An opportunity was therefore presented to locate the line of the road to Cunetio, as it should be under or near the present Amesbury road. This road is well defined at the Cunetio end as far as Easton,?7 but its con- tinuation from there to Old Sarum has not been proved conclusively. The evidence produced by the trench was not definite on this point. Approxi- mately 400 ft. north-west of the Amesbury road (and } mile north of the East Gate of Old Sarum) large flints were exposed in the trench section and were present over a length of some 160 ft. Associated with this flinting was a pair of what might have been the side ditches of a road. These were V-shaped, 132 ft. apart, 1 ft. 6 in. wide and of a similar depth. These features could not be examined in any detail as the pipe-laying proceeded too rapidly. Other pits and hollows were seen further west, but no associated finds. Further patches of large flints were also encountered, so it is not possible to be firm on the conclusion that the first area of flints formed part of a road. Certain identification of this line can only follow further field work and especially observation from the air. It can be said, however, that the line might be found away from the present metalled road in this area. Just as the Silchester road has been, in its present metalled state, ‘pulled’ into the Toll House so, likewise, this could have happened to the Cunetio road. In other words the focus of the roads in the immediate area of Old Sarum has, with turnpiking, changed from Old Sarum to this Toll House to the north. THE ROMAN ROAD TO DORCHESTER Recent work (1962) by J. E. D. Stratton, following his observation of a parch- mark in a field west of the accepted line of the Roman road to Dorchester (Fic. 1), points to the possibility that the Dorchester road did not follow the route of the present track, but was approximately 70 yds. west of it. The present track may be of medieval date; if so, this is a further example of the replacement of the Old Sarum Roman roads by medieval roads not on the same line. The dating of the parch-mark 139 is, at the time of writing, under investigation by selective excavation, and the results will be published separately.’ At the present time the identification of the parch- mark as a Roman road rests solely on the fact that it runs in a straight line from Old Sarum to the river (at the point where there is a small island), and can be traced on the other side of the river, from which it continues in the direction of Devizes Road. THE MEDIEVAL SUBURBS: THE 1958 EXCAVATION GENERAL DESCRIPTION (FIG. 2) The site is on Upper Chalk, with ‘high level gravel’ and ‘valley gravel’ quite close to south-east, through south-west to north-west.29 The ‘natural’ encountered in the excavation was hard chalk, weathered and worn in some areas by human activity. Many cuttings were made to give as complete a picture as possible; this is an unsatisfactory method for dealing with a medieval site, but the most important area of a building discovered at the south-west end will be preserved. This was of 13th- to 14th-century date, and secondary to Road C which lay below its walls; there were foundations of stone, slots for timbers, post-holes, numerous cesspits, and other features. This building appeared to be in the angle between a major Road B, from the Wilton direction, and Road D, a minor road from the direction of ?Old Sarum Farm. Further to the north Roads B and D merge, and are joined by Road A; this was a hollow way for foot and animal traffic, and is the only road whose extension can be traced today; further to the south-east, on the south-east side of the old main road, its line is continued as the present hollow track up to Bishopdown. The junction of the four roads (Road X) was found further north, and was shown to continue below the barn of Old Sarum Farm; here it begins to turn to the west, a curve continuing outside the barn up to the point where the road is cut away by the modern roads. There can be little doubt that the combined medieval Road X led up to the East Gateway of the castle, on the north side of the barbican. This apparently round-about way to the castle was that of least gradient. It is probable that travellers, foot and horseback, from south and west entered the castle by the steeper more direct route through the south side of the barbican. East of Road X, the ground was hardly explored; a single trench eastwards from the barn was negative, but a few features were found further north on the edge of the road, indicative of some occupation here. West of Road X, there were numerous features and cesspits; there was probably a building here, destroyed by the modern road cutting. Two more cesspits were found (Pits 56 and 57) close to the road, in a pipe-laying trench. DETAILED DESCRIPTION OF FEATURES (FIG. 2) The site is a long narrow triangle, bounded on the west by a deep cutting made for the modern Salisbury-Amesbury road and on the north by the slope down to the old Salisbury-Amesbury road; on the south-east side there is a steep scarp 136 S piste l= | section —= OLD SARUM EAST SUBURB 1958 Scale A Se a aaa | ie | a K | | | : TIATT AOOTA 43/ DNILSIXNF VA | I} i! | PU i ] i | eee a ee eee oe a Uy | = a V-K NOMLOFTS 7VM/G)LIONO 7 oes se Se, ee we = 4 vA vA x YS So. ss JIOINAOD PGITTNOW = OL NOPD aS, 11 which provided an alternative method of spanning a hall to that of the early medieval aisled hall, which had two arcades like the nave of a church.3 The hammer- posts in this roof, as in the few other examples, are in fact truncated aisle-posts, carried by the hammer-beams which project from the walls at the normal tie-beam level, and are braced beneath by heavy curved timbers springing from wall-posts. Above, the hammer-beams and posts are braced together by heavy raking struts, set almost parallel to and just below the common rafters. The hammer-posts perform the normal functions of aisle-posts, and carry both the main collar of the open truss and the longitudinal plates or square-set purlins. ‘Together with the collar-purlin these are the only purlins in the roof. They are heavy timbers, laid, as if they were aisle-plates in an aisled roof, horizontally upon the heads of the posts, not parallel to the slope of the roof as in later types. Moreover, they are embellished, also as if they were aisle-plates, by moulded cornices. These are small separate timbers set upon the inner top angle of each plate, and slotted at the end of each bay into the collars of the main trusses. Their section and simple double roll-moulding are illustrated in the drawing. ‘The hammer-posts were braced both to the collars and the plates by long arch-braces, each pair meeting centrally to form an ‘arch’. Only the pair in the west open truss survive, but the central compartment of each bay was once surrounded by four such suspended ‘arches’. The positions of the longitudinal ‘arches’ has been deduced from existing mortice-slots and peg-holes. Below the long slots which housed these braces in the hammer-posts, are shallow depressions in the flanks of the hammer-beams, where the feet of the braces originally rested. This device ensured that the longitudinal braces sprang from the same level as those in the open trusses. In the open trusses the ‘arch’ was combined with the braces below the hammer-beams to form a clear-cut cusped profile. This was obviously a deliberate effect and was contrived by extending the lower braces beyond the hammer-beams so that they help to support the braces which form the ‘arch’. Moreover, the effect was emphasized by infilling the spandrels behind the upper braces with vertical boarding. An oak fillet was attached to the upper side of the braces, under side of the collar, and flanks of the hammer-posts, and thin oak boards, carefully cut and fitted, were nailed to this fillet on both sides, using short square-section nails with flat square heads. Thus the spandrels were made_ flush with the main timbers of the truss. Many of the oak boards survive i setu and share the thick smoke blackening of the remainder of the roof. They are clearly visible in pL. XIa, but are not shown on the drawing. The upper part of the roof, above collar level, is of orthodox crown-post construction. There are 19 pairs of trussed rafters and a collar-purlin. This is carried, in the open trusses, by crown-posts of the usual Salisbury type, with four heavy curved braces springing from the base of the post. The end trusses have remains or traces of crown-posts with three braces. There was no evidence for any structural smoke louvre. The west end truss was probably a spere truss, providing a screen between the open hall and a screens passage which would have crossed the house from south to north. This cannot be proved definitely, since below collar level only the south half of the truss exists. However, the truss retains the aisled construction usual 158 in spere trusses. The south end of the collar is supported, not by a hammer-post but by a complete ‘aisle-post’, reaching from ground level. It measures 13 in. across as against the 10 in. of the hammer-posts. Like them, it is broadened at the head where it carries the south square-set purlin, but this thickening commences lower than in the open truss; just above the arch-brace to the collar. This too survives in situ, together with a long mortice for the longitudinal brace to the purlin. The existing brace rested at the foot, not on an extended lower brace as in the open truss, but on a (destroyed) horizontal between the two main posts for which a 6-in. slot exists, in the south post. In the position of the hammer-beam a heavy horizontal timber, in effect a truncated tie-beam, extends from the position of the outside wall to the south post. There can be no certainty about the arrangement of the truss below this level as all that exists is the damaged lower half of the south post. Three peg-holes, and a long mortice in its north face, at a height between 4 and 6% ft., may indicate the position of the arched head of a former opening at ground level. The walls which carried this very massive open roof had been removed com- pletely, probably during the 18th century when the house was refronted on the north side and extended on the south. Only stone walls, of ashlar or rubble, would have been equal to the task; their thickness could be calculated from the roof timbers as about 2 ft. This is the distance between the wall-posts, from which spring the lower braces of the open truss, and the outer termination, raked in line with the roof slope, of the hammer-beams. These were laid across the top of the walls. The rectangular cut in their lower edge housed the wall-plate, laid along the top of the wall. It is probable that this timber carried ashlar-pieces to the common rafters, but no trace survived of this feature, commonly used to mask the gap between the top of a stone wall and the sloping timbers of the roof. It appears, for example, in the ‘Stranger’s Hall’ at Winchester, where also the hammer-beams and wall- plates are similarly interlocked.5 ‘The two roofs are generally similar in their structural system, and their obvious derivation from aisled construction, but at “Balle’s Place’, a fully-developed crown-post and trussed rafter upper roof replaces the single posts and scissor-braced rafters of the ‘Stranger’s Hall’. In decoration, it lacks the carved heads and stops of the Winchester roof, and relies on run-out chamfering, the moulded cornices, simple outlines, and careful relationships between the different curving members. These characteristics indicate that, on typological grounds, it can be regarded as a later development of the Winchester roof. If the early 14th-century dating of the ‘Stranger’s Hall’ can be accepted, then this derivative in Salisbury may be as much as fifty years later than its prototype, and therefore already somewhat conservative in character. The documentary evidence indicates a building date of ¢. 1370-85. Considering, however, that it was built by a merchant of Salisbury, not by a Cathedral, nor by a manorial lord, like the hall of Tiptofts Manor of ¢. 1340 in Essex which also has a hammer-beam roof,° this conservatism is hardly surprising. That the roof covered an interior of considerable grandeur cannot be doubted. Unlike the main house, the street range of Balle’s Place (Nos. 25 and 29 Winchester Street) was a two-storied and completely timber-framed building, 159 containing three equal bays, each 11 ft. long, and a shorter bay at the west end only 7 ft. long. It is clear from the documents that the entry from the street to the courtyard originally occupied this narrow bay. At some date, and before 1716, when the range had already been made into two houses, this entry was moved to a central position between them. The doorway leading to it can be seen in this position in pL. Xa. The main three bays of the range, and probably a fourth referred to in the documents but destroyed when Nos. 31-33 were built c. 1500, were occupied by shops. If each shop occupied a whole ground-floor bay, they would each have measured 11 ft. by 16 ft. The range was jettied along the street front, so that the first-floor bays measured 11 ft. by 174 ft. In pL. Xa the projecting end of a first-floor beam, together with a short supporting brace and the sawn-off end of the ground-floor wall-plate are visible, embedded in the half-demolished wall beside the doorway to No. 27. Above can be seen the remains of a first-floor partition. These were very simply framed, with long arch-braces springing from the wall-posts to the tie-beam. Similar braces had existed in the side walls, a pair meeting in the centre of each bay, but only their mortices in the underside of the rear wall-plate at eaves level survived. The roof (pL. XIb) was of the simplest crown-post type, with slightly cambered tie-beams, trussed rafters, a collar-purlin, and crown-posts each with two longitudinal chamfered braces springing at 2 ft. above the tie-beams. Each end truss had one brace only. pL. XIb illustrates the crown-post of the truss first from the east end. Both braces had deep cross cuts near their upper ends, and the chamfers were run-out just below them. These must therefore have been an original feature, but their purpose remains a mystery. The other crown-posts lacked this feature. It would be difficult to date such a simple building merely on the basis of its construction. Crown-post roofs in Salisbury are usually datable to the 14th century or the early 15th. In general scantling this one was similar to the upper part of the roof of No. 27, although the crown-posts of the two roofs were so different. On documentary grounds the close relationship between the two buildings is certain, and it seems safe to regard them as of similar date. DOCUMENTARY HISTORY A, DESCENT OF THE PROPERTY The calendar of charters and memoranda which precedes the third surviving volume of the Salisbury ‘Domesday Book’ includes the following entry for the tenth year of Richard II (1386): “The will of John Balle in which he bequeathed a corner tenement with appurtenances in Wynmanstrete and Brownstrete’.7 (These were the medieval names of Winchester Street and Rollestone Street.) Other references to John Balle occur intermittently between 1374 and 1386, concerning transactions over properties in the City.® They are sufficiently numerous to indicate a man of considerable wealth. In 1383 he bought two tenements in Wynmanstrete,? and these may have comprised the eastern part of the Balle’s Place site (see below), but there is no record of his acquisition of the main corner tenement. That John Balle was a wool merchant or dyer can be inferred from an 160 episode recorded in the Close Rolls concerning a dispute with Richard Bernewelle, also of New Sarum, over a cargo of twenty tuns of woad ‘of the growth of Normandy’ shipped in the ‘Seint Johan’ of Caen.'° John Balle does not seem to have been a native of Salisbury, where the surname is not recorded before his lifetime. Earlier in the 14th century there were Balles, who may have been other members of the family, living in North Wiltshire.! Nor did John hold any public office in the City, and he is not among the senior citizens who, as witnesses, signed the numerous deeds enrolled in the Domesday Book. Perhaps he died relatively young, before the wealth which must have been his brought him into public life. His wife Agnes survived him for nearly thirty years, until 1414 when, since their sons Nicholas and Thomas had also died, no heir to the family property survived. Accordingly, ‘Balle’s Place’, where John Balle had lived—as the document specifically states—was sold under the terms of his will, his family and executor having all died, by his executor’s executor to Walter Shirley and two others.!2 Walter Shirley owned ‘Balle’s Place’ from 1414 until his death in 1424, and it is from his deeds and will that we learn most about the lay-out of the site. In the same year, 1414, he bought the next tenement adjoining ‘Balle’s Place’ on the north and this became a permanent and valuable adjunct to the original property.%3 He was a wealthy and important man, Mayor and M.P. for New Sarum many times,'4 and one of the first Salisbury merchants to have his will enrolled in the Prerogative Court of Canterbury instead of the Court of the Sub-Dean of Sarum. That part of the will which concerned his numerous tenements in Salisbury was, nevertheless, transcribed into the Salisbury Domesday Book, in 1429, five years after his death.'5 It is clear from his will that Walter Shirley did not himself inhabit ‘Balle’s Place’, but sub-let it in several parts. He bequeathed the capital messuage, then occupied by John Hobbes, to his wife Johanna for the term of her life, with reversion to John and Ann Estbury for the term of their lives. The other parts he bequeathed to his tenants, on various conditions, but usually for life. On the expiry of these terms each part or sub-tenement was to be sold by his executors and the money thus received to be given to the College of St. Edmund, Salisbury. In the course of time, these instructions seem to have been carried out. Bishop Beauchamp’s Rental for 1455 records that the tenement ‘formerly John Estbury’s’, together with some shops in Brownstrete, once Walter Shirley’s, were held by John Whittokesmede, and certain shops in Wynmanstrete also once Walter Shirley’s were held by the Mayor and Commonalty.'® Walter Shirley’s property had therefore been sold by his executors by 1455. John Whittokesmede, purchaser of the capital messuage and the north tene- ment, held the office of Bishop’s Bailiff and so was the most important secular official in the City.17 He held no other property in Salisbury, according to the Rental, and it is likely therefore that he lived in ‘Balle’s Place’ himself, and did not sub-let it. As a man of law, John Whittokesmede of Beanacre, near Melksham, held various offices in Wiltshire c. 1433-81. He represented several boroughs in Parliament, including Salisbury in 1448, became one of the County justices, and 161 in 1450 was made a Knight of the Shire.'? His ownership of ‘Balle’s Place’ had ended by 1463. From at least 1463 until the late 19th century the capital messuage was the property of the City. In 1463, Simon Poy acting in his office of Mayor, leased ‘Balle’s Place’ for 98 years to John Aport, a famous citizen of Salisbury who needs no introduction here.'9 He may have lived in the house, but more probably sub-let it. The City Chamberlain’s Account Roll for 1469-70 records the receipts of 66s. 8d. rent ‘for the tenement formerly John Estbury’s, given by Walter Shirley, sometime an elder citizen of this city, now leased to John A Port’.2° This statement of a gift cannot be reconciled with the terms of Walter Shirley’s will, nor with the apparent descent of the property. In 1473-74 John Aport was expelled from his tenure for non-payment of rent,?! and in 1476 we find Nicholas Edmonds, Mayor, leasing “Balle’s Place’ to John Ingler, Gentleman, and Elena his wife, for 70 years.?? He was Mayor of Salisbury in 147823 and died in 1487.24 By 1498-99 Elena had probably died also, as the house was occupied by Thomas Holbeme, City Chamberlain ¢. 1498-1508.75 B. THE SITE The same documents which tell us of the descent of ‘Balle’s Place’ from the late 14th century to the early 16th century also provide much information on the lay-out of the site, and the function of the various buildings. ‘The gradual consolida- tion of the property under private and public ownership can be envisaged. It seems safe to assume that ‘the corner tenement and capital messuage in Wynemanstrete called Ballisplace’ for which the City Chamberlain collected rent from 1508-9 to 1526-2726 was one and the same as ‘the corner messuage and garden lyinge in Winchester Streate and Rowlston Streate, containing in lengthe’ nine poles from north to south and six poles from west to east, which John Lyminge held in 1618, and which his father, Zachary Lyminge, leased from the Mayor and Commonalty in 1591.27 A block of property with frontages having approximately these measure- ments, 148 ft. 6 in. by 99 ft., is immediately apparent on any large-scale plan of the City, and these dimensions have been taken as a basis for the following account of the site. The accompanying sketch plan (Fic. 2) is based upon Botham’s detailed and magnificent plan of Salisbury, drawn in 1854, and now kept in the City Engineer’s Office. In 1618, a small tenement ‘adjoyning to the corner tenement’ on the south-east and measuring only 12 ft. across was separately leased to William Kent, but it was probably originally part of the corner tenement, and has therefore been included on the sketch. Its separation provides the first example of a gradual process; by 1716 all the buildings along the Winchester Street frontage were let separately by the City?® instead of being held by a single lessee, who then sub-let on his own account. At its greatest extent, therefore, the Balle’s Place tenement may have measured 148 ft. 6 in. by 111 ft., as shown in Fic. 2. The boundaries of the sub-tenements cannot be drawn precisely, but are based on detailed measure- ments given in 1716. When John Balle died in 1386, it is recorded (in 1414) that his tenement contained, as well as his dwelling house, ‘some shops, a cottage, a gateway, a court- 162 Fic. 2 ‘Balle’s Place’, Salisbury. Sketch plan of the site. yard, a garden, a dovecot, and other appurtenances’. This description immediately indicates an extensive complex of buildings and an establishment of high status. Only a few of the most important houses in Salisbury had dovecots. John Balle’s property comprised the parts marked A, B, and C on the sketch plan. A was the main house, with its garden on the north and courtyard on the south. B and C were shops ranged along ‘Wynmanstrete’. Those in B may have been the two tenements bought by John Balle in 1383. The details of this lay-out have been concluded from an analysis of the medieval documents and later surveys, together with the information provided by the buildings still standing in August 1962. These are blocked in on the plan. The hall of the main house was aligned, as we know, roughly east-west, with a probable spere truss at the west end, with, presumably, a screens passage behind it and butteries beyond. The documents confirm this, generally, by mentioning that the kitchen abutted on to ‘Brownstrete’. We can therefore envisage it not as a detached kitchen but as completing the west end of the hall range, and approached in the orthodox medieval manner by a passage between the butteries. If the service rooms were at the west end of the house, then we can probably assume that the solar end was beyond the hall to the east, and so overlooked the garden with its dovecot, on the north side of the house. The outline of this suggested plan is drawn in on the sketch. 163 Several documents refer to the buildings on the south side of the courtyard, along the street, as shops. ‘The one on the corner was probably a fairly substantial building as it is given the description ‘with solars built above’, that is with more than one upper room. The ground belonging to it, perhaps totally occupied by the building, extended as far north along ‘Brownstrete’ as the kitchen of the capital messuage, and as far east as the entry (‘hostes introitus’) leading to the courtyard. Measurements of the same tenement taken in 1716 confirm that this entry was at the west end of the street range which survived until 1962, and we can presume that it occupied the narrow bay in that position. Significantly, this narrow bay was opposite the presumed position of the screens doorway of the house, across the courtyard. Beyond and east of the entry, four shops are mentioned, and the remains of three of these stood in 1962. All these buildings had the right of ingress and egress via the entry. When the corner shop was sub-let, as in 1423 by Walter Shirley to Richard and Alice Couper,?9 these four shops were regarded as part of the same sub-tenement (C), and were therefore sub-let by the tenant of the corner shop. The eastern part of the Wynmanstrete frontage was occupied by two more shops, and a ‘cottage’ between them and the four described above. The word ‘cottage’ seems to imply merely a dwelling house, as opposed to a shop with living accommodation. It was not necessarily very small. These three buildings are tene- ment B on the sketch plan. Walter Shirley, in 1424, bequeathed the ‘cottage’ to his cousin Isabella, and the two shops to John Park.3° It is from the transactions and will of Walter Shirley that we also learn most about the northern part of the site. In 1414, the year in which he acquired ‘Balle’s Place’ and its appurtenances, Shirley bought a tenement (D) which lay further north in Brownstrete. It contained either two or four ‘cottages’. The documents provide conflicting descriptions of it, but it is certain that it adjoined the capital messuage on the north and that it had no street frontage.3! It lay within the gateway of another, separately-owned tenement (E) and possessed a right of access through the gateway. The sketch suggests a possible plan. Walter Shirley made use of the gateway to provide a rear entrance to ‘Balle’s Place’. Evidently the entry from ‘Wynmanstrete’ was adequate only for people on foot, or packhorses. At his death, he bequeathed the north tenement (D) separately from the capital messuage, but _ with the proviso that there should always be free ingress and egress through the gateway for carts going to the main house. Another probable consideration was that any warehouses belonging to the capital messuage must have been on the north side of the house, as the courtyard was surrounded with other houses and shops, and merchandise brought into the courtyard by the south entry would have to have been carried through the screens passage. Between the capital messuage and the north tenements D and E was another tenement (F'), with a frontage on ‘Brownstrete’. It must have abutted immediately against the north wall of the main house, otherwise a rear entry for Balle’s Place could have been inserted between the two. This tenement (F) was in separate ownership from at least 1376 to after 1455; from 1415-55 in that of the Fadur family. There is no description of it, but one document mentions a shop there, another a ‘cottage’ .3? 164 Most of these properties can be identified in Bishop Beauchamp’s Rental for 1455.33 John Whittokesmede, the Bishop’s Bailiff, held the capital messuage of ‘Balle’s Place’ (A), described simply as ‘formerly John Estbury’s’, and the north tenement (D), described as ‘cottages in ‘‘Brownstrete”, formerly Walter Shirley’s’. The corner tenement C is not identifiable at this date as we do not know who the occupier then was. Walter Fadur held the tenement F’, and the Mayor and Com- monalty held tenement B, in two parts: ‘A shope in Wynmanstrete formerly John Park’s’, to whom Walter Shirley bequeathed it, and another shop there, ‘once Walter Shirley’s’, and probably the ‘cottage’ he bequeathed to Isabella. Subse- quently all the other tenements became the property of the City, presumably by purchase. After 1455, the first surviving rent roll of the City Chamberlain is dated 1469-70. ‘Balle’s Place’, which did not appear in the roll for 1453-54, 1s now included. It had been leased to John Aport, together with tenement D and probably tenement B which does not appear separately, and therefore had presumably been re-amal- gamated with the capital messuage. In 1477 the same tenements were leased to John and Elena Ingler, with the addition of tenement C which the Mayor and Commonalty had leased from John Wareyn in the same year. Subsequently they acquired total rights over it. In 1484-85, the Inglers paid 60s. rent, and the Chamber- lain also received, from other tenants, ros. ‘for 3 tenements called Faders next to the gate of Ballisplace’, and 10s. for ‘three tenements next to the gate of Ballesplace’. I suggest that these are, respectively, my tenements F and E. They do not appear separately in the 1508-9 rent roll, when £4 was received for ‘the corner tenement and capital messuage in Wynemanstrete called Ballisplace’.34 The Mayor and Commonalty, by systematic purchase, had consolidated this valuable property. The history of this tenement is interesting both for itself and with regard to certain more general questions relating to the plan of medieval Salisbury. Founded as a ‘new town’ in 1225, the City was laid out on a gridiron plan of intersecting streets with rectangular blocks of tenements between, known since the 17th century as ‘chequers’, and called each by a distinguishing name, often that of an inn.35 Certain pre-existing factors, such as the existence of the old road from Clarendon to Wilton, which passed along the present line of Milford Street and New Canal, created diversity in an otherwise regular lay-out. Along the streets ran narrow channels of water diverted from the Avon, which were also directed across a few of the chequers, and constant references in medieval deeds, both to the watercourses and to angle tenements, make it possible to envisage the division of several of the chequers into their component tenements. Within the City, property was held by burgage tenure. In 1225, the Bishop granted that all his free tenants in New Sarum should pay quit rent for their tene- ments: ‘Each holder of a plot measuring seven perches in length and three in breadth was to pay twelve pence annually, sixpence at Easter and sixpence at Michaelmas, and holders of more or less than this in proportion.’3° One would therefore hope to find, ‘fossilized’ within the modern plan of the City, some burgage plots of approximately the regulation size, 1154 ft. by 494 ft. Those plots which were 165 from the beginning either greater or smaller are indistinguishable from tenements which have been subdivided or amalgamated at a later date. Some impression of the diversity which existed in the mid 15th century may be gleaned from the 1455 Rental, when the amounts paid in quit rent varied very greatly. One street where standard burgage plots can be distinguished, although now subdivided, is the modern Queen Street, the west side of Cross Keys Chequer. Starting at the north-west angle, there appear to have been five such plots in a row, of which the northernmost measured 1093 ft. by 50 ft. and had a long north frontage along Winchester Street as well as the shorter west frontage; it therefore gained the greatest possible advantage from its corner site.37 The south-west angle of the chequer presents a contrast to this simple arrangement; it appears to have been divided between the tenements facing west and the equally important tenements facing south with their frontages on Milford Street. ‘The two rows were interlocked so that the tenements were progressively shorter, the nearer they were to the corner.38 A similar contrast existed between the two southern angles of Three Cups Chequer, where the south-west angle was occupied by ‘Balle’s Place’ and the tenements associated with it. We have seen how these were gradually amalgamated into a single angle tenement of unusually large dimensions, but at the earliest date that we can envisage the site, it must have resembled the south-west angle of Cross Keys Chequer in that the tenements differed in size and were closely interlocked. The remaining south frontage of the chequer appears to have been divided into more or less equal tenements each 45 ft. to 50 ft. across, and extending at the rear to a straight boundary running east to west across the chequer. ‘The south-east corner tenement, which had a long east frontage on St. Edmund’s Church Street, was called ‘Pette’s Corner’, and later the ‘Three Cups Inn’.39 The occupier paid 1s. 3d. quit rent in 1455, whereas 4s. 5d. was paid altogether for ‘Balle’s Place’ and its associated tenements, including 2s. 14d. for the capital messuage. One cannot deduce from this how the rents were assessed, but they certainly reflect the difference in size of the two southern angle properties of the chequer. The Rental contains references to a considerable number of tenements whose rents were 2s. or over, and one can probably envisage these as having been, like Balle’s Place, large court-_ yard houses, with valuable appurtenances. It is clear that the west frontage of Balle’s Place was less important than the south with its crowded row of shops along ‘Wynmanstrete’. There were shops between the kitchen and the back gate, but on the north side the site was bounded by a piece of open ground used as a rack close.4° ‘There were many such open sites in the City. They were not, in any sense ‘vacant lots’, but valuable plots of ground, used for semi-industrial purposes by the fullers, dyers or tanners, or as gardens. A garden, which sometimes, like John Balle’s, had a dovecot standing in it, usually appears in the list of appurtenances of an important tenement.4! ‘Balle’s Place’ was therefore typical in several respects of the larger houses of medieval Salisbury. Other examples could be mentioned, to show that John Balle’s house was not, in its own day, a remarkable or isolated phenomenon, but merely a reflection of the solid prosperity of its builder. 166 t We would like to thank Mr. C. H. Harris, M.B.E., and Mr. J. A. Harris, together with the Salisbury City Council, for the help which, as joint owners of the site, they gave to us during our investigation. We would also like to thank the City Engineer of Salisbury, Mr. H. Rackham, Mr. F. Crocker his Planning Officer, and Mr. S. E. Rigold of the Inspectorate of Ancient Monu- ments (M.o.P.B.&W.) for conducting the nego- tiations which led eventually to the preservation of this medieval roof. 2 W. A. Pantin, Some Medieval English Town Houses: a Study in Adaptation, in Culture and Environment: Essays in Honour of Sir Cyril Fox (1963). 3 J. T. Smith, Medieval Aisled Halls and their Derivatives, Arch. Journ., cxtt (1955), 76. 4M. W. Barley, The English Farmhouse and Cottage (1961), 32. 5 Illustrated in J. T. Smith, Medieval Roofs: A Classification, Arch. Journ., cxv (1958), 111, fig. 9. 6 J. T. Smith, op. cit. Note 3 above. Fig. 5. 7 Salisbury Corporation MSS. Domesday Book, 1413-33. Calendar folio 44. 8 Ibid., folios 34, 37, 38, 42. 9 Ibid., folio 42. t© Calendar of Close Rolls: 1384, p. 485; 1385, p. 29; 1386, pp. 51, 63, 135. u E. A. Fry (ed.), Abstracts of Wiltshire Inquisitiones post mortem, 1242-1326 (Index Library XX XVII), passim. 1 Salisbury Corporation MSS. Domesday Book, 1413-33. Transcripts, folio 21. 13 Ibid. Transcripts, folio 22. Calendar, folio 67. % V.C.H. Wilts., v, 773 VI, 105. 1s Somerset House. P.C.C. 3 Luffenham (1424) and Salisbury Corporation MSS. Domesday Book, 1413-33. Transcripts, folio 243-4. 1 KE. R. Nevill, Salisbury in 1455, W.A.M., XXXVI, 66-91. 17 R. Benson and H. Hatcher, Old and New Sarum or Salisbury (1843), 698. 18 V.C.H. Wilts., v, 78. 19 Salisbury Corporation MSS. Z/240, folio 5. 20 Salisbury Corporation MSS. O/103/4. 2t Salisbury Corporation MSS. O/103/5. 22 Salisbury Corporation MSS. Z/240, folios 10 and It. 23 R. Benson and H. Hatcher, of. cit., 696. 24 Public Record Office. P.C.C. 4 Milles. (1487.) 25 Salisbury Corporation MSS. O/103/7-8. 26 Salisbury Corporation MSS. O/103/8-11. 27 Salisbury Corporation MSS. O/117/1, folio 56 and 1618 Survey. aie Corporation MSS. O/117/4, folios 21-28. 29 Salisbury Corporation MSS. Domesday Book, 1413-33. Transcripts folio 171. 3° See note 15, above. 3« See note 13, and note 15, above. 32 Salisbury Corporation MSS. Domesday Book, 1413-33. Calendar, folio 37. Calendar, folio 67, and Transcripts, folios 21-22. 33 See note 16, above. 34 Salisbury Corporation MSS. O/103/4, 6, 8. 35 V.C.H. Wilts., vi, 114. Plan. 36 Fanny Street, The Relations of the Bishop and Citizens of Salisbury (New Sarum) between 1225 and 1612, W.A.M., xxx1x (1917), 190. 37 Salisbury Corporation MSS. O/117/1, folio 64. 38 Wiltshire Record Office. Deeds 164/1/9, 10, 12, 13. Dean and Chapter of Salisbury. Deeds. Box 3 (Deeds of the Harding Family). 39 Salisbury Corporation MSS. O/103/1-11 and O/117/1, folio 59 and 23 (N.S.). 4° Salisbury Corporation MSS. Domesday Book, 1413-33. Transcripts, folio 22. 41 E.g. Salisbury Corporation MSS. Tailors’ Guild Deeds, Nos. 16-89, passim, referring to ‘Sowthwel Place’, which occupied the north-east angle of Marsh Chequer. In 1455, the rent paid for this tenement was 2s. 543d. TWO WILTSHIRE FONTS IN SOUTH AUSTRALIA by H. pe 8S. SHORTT A CERTAIN AMOUNT of mystery surrounds the fact that in 1877 the Venerable C. W. Morse, Archdeacon of Adelaide, Rector of Yankalilla and Vicar of Glenburn (now called Delamere) from 1869 to 1901, took two fonts with him on a return journey to Australia. One of the fonts came from Salisbury Cathedral (pL. XIIa), and this he put in Christ Church, Yankalilla (1857). The other came from St. Peter’s, Britford (pL. XIIb), and this he put in St. James’s Church, Glenburn (since re-named Delamere) (1871). While the Cathedral font was well known, from two copper engravings, one by James Biddlecombe (pt. XIIc) and the other (1754) by I.S. Miiller after a drawing by Biddlecombe, to be of English Renaissance style and late 17th- or early 18th-century date, the Britford font had apparently never been recorded and was said to be of the Saxon period. Through the kindness of the present Rector of Yankalilla, the Reverend R. O. Nichols, Mr. Guy Wells of Wirrega, South Australia, and especially Mr. J. E. Webb of the School House, Britford, photographs of the two fonts have been obtained, together with records of their being set up, so that most of the mystery can now be explained. Writing in his Brief Historical Survey of St. Jame’s (sic) Church Delamere, a dupli- cated pamphlet written for the nonagenarian celebrations of the church in 1961, the late Rector of Yankalilla, the Reverend C. F. Sexton, says of the font on page 4, ‘This is priceless and unmatched in Australia—a touch with the Saxon period of our Church in England, possibly 900 years old. Archdeacon Morse in 1877 obtained this disused marble font from England, the gift of the Vicar and Church-wardens of Buxton (sic), a parish near Salisbury, Wilts.’ The inscription on a brass plaque over the font is more accurate. It reads: “To the Greater Glory of God / This Font / Was given in 1877 by the Vicar of St. Peter’s Church, Britford / in the Diocese of Salisbury / to the Venerable C. W. Morse, Archdeacon of Adelaide, Vicar of this Parish / Who brought it out and erected it here. / It had been in use in that Church for over 200 years.’ The photograph (pt. XIIb) shows that the Britford font differs little in date from the Cathedral font, though, as might be expected in a small country parish, it is of somewhat coarser workmanship and design, and the rectangular panels round the upper part of the stem are quite unharmonious. The Britford font is octagonal, the other hexagonal and the Cathedral font has a more polished finish and the embellishment of acanthus leaves on the lower part of the stem. Although the bowls of both fonts are circular, a rebate round the lip is polygonal in both cases, showing that wooden covers were once intended. Unfor- 168 tunately, neither now exists, though Biddlecombe’s print gives a good impression of the Cathedral cover, with a charming finial, possibly a rebus.! The three steps on which the Cathedral font once stood are now also absent. It is not easy to clear up the mystery at the English end of the transaction. The Vicar of Britford was Canon A. P. Morres, well known in Wiltshire as an antiquary and an ornithologist. How did he get a faculty to move his font from Britford, and how did Archdeacon Morse persuade the Dean and Chapter to dispose of their font from the Cathedral? Probably, in the case of Britford, by the provision of funds for a new one (pL. XIIIa). The Cathedral font was apparently removed about 1850 when the present one in 13th-century style was put in as a memorial to Dean Lear (pL. XIIIb), and the Dean and Chapter may have been only too glad to be rid of an embarrassing redundancy. It must be remembered that in the mid rgth century classical styles were entirely out of fashion—everything had to be Gothic. Gothic ornament was regarded by Pugin’s disciples as Christian, while classical ornament was pagan. In 1879 the Rector of West Knoyle, Canon E. Inman, removed a small but no less charming classical font (pL. XIIIc) from his church, no doubt by faculty, and replaced it with a Victorian Gothic font (pL. XIIId) as a memorial to his two infant children, also commemorated by a stained glass window and a brass plaque on the south side of the chancel. Today, taste goes the other way and the present Rector, the Reverend N. E. E. Johnson is obtaining a faculty to remove the Victorian font and replace the classical one, which was in the churchyard. The Victorian fonts in the Cathedral and Britford Church have so far survived.’ 1 The design of the finial appears to be an upright barrel or tun, on which stands a bushy- topped tree. There are bench-end finials of an earlier date in Britford Church with a similar rebus, except that the tun is on its side. The tree is described as a service tree and the family com- memorated by this shocking pun is Cervington. A search through Jones’ Fasti failed to discover any such name. The nearest, though improbable, candidate was Francis Horton (Haw-tun) who was a residentiary canon and in 1683 was locum tenens for the Dean. Of course, the font and cover may have been presented by a layman. 2 I am grateful to Mr. E. W. Fimmel for photographing the two West Knoyle fonts, to Major H. F. W. L. Vatcher, M.B.E., for photo- graphing the Victorian fonts at Salisbury and Britford, and to the Royal Commission on Historical Monuments (England) for the photo- graph of Biddlecombe’s engraving, which is in the Salisbury Museum. THE DECLINE OF A RECUSANT FAMILY: THE KNIPES OF SEMLEY by J. ANTHONY WILLIAMS READERS OF THE CHAPTER ‘Roman Catholicism’ by the late Brigadier T. B. Trappes-Lomax in the third volume of the Victoria County History of Wiltshire (1956) will have been left in no doubt as to the supremely important role of the Arundells of Wardour in sustaining a ‘pocket’ of Catholic recusancy throughout the 17th and 18th centuries—so much so that by 1767 a group of parishes in the Wardour area contained between 50 and 60 per cent. of all the Roman Catholics in the county.! One of these parishes was Semley, near Shaftesbury, but within the borders of Wiltshire, among the inhabitants of which a persistent element of popish recusancy is reflected in Anglican returns of papists for over a century. In an undated list, compiled—from internal evidence—between the Restoration and 1672,? five adult ‘Romanists’ are named in the parish of Semley; the Compton Census of 1676 reported twice as many3 (giving no names, however), but the House of Lords’ return of thirty years later¢ omits this parish and it is not until 1767 that a further census of papists was compiled, by which time the Catholic population of Semley, including children, had risen to thirty-nine.s In 1780 (the year of the Gordon Riots, when a party of rioters may have passed nearby on their way to attack the Wardour chapel—which in fact they failed to locate)® the number stood at forty-three, the fourth highest Catholic population in any Wiltshire parish7 and we find the Anglican incumbent explaining to the Bishop of Salisbury: ‘If the number of Papists should seem large for this Parish, which is not a populous one, your Lordship will easily account for it from its vicinity to Wardour Castle.’® Among those reported in 1780, as in the earlier lists of Catholic names, we find the surname of Knipe, or Knype, a prominent Wiltshire recusant family which can be traced at Semley for some two hundred years, but which is mentioned in neither of the two printed accounts of Wiltshire Catholicism.9 It is the purpose of this article to trace in some detail the fortunes of this family and in so doing we shall note how they had declined from the days of William who left his children well- endowed, gave generous bequests to charity and possessed a ‘Silver Boule with Knype’s Armes upon it’, to those of Mary, ‘married to a poor labourer’, and Jane who was unable to sign her name in the register when she married an equally illiterate widower in 1826.!° One might be tempted to suppose that this sorry state of affairs was brought about by the imposition of the various fines and forfeitures to which Catholics were statutorily subject, but in the absence of family papers we have no direct evidence of this, and the ‘public record’ evidence points the other way: the absence of any Knipe entries in the post-Restoration recusant rolls for 170 Wiltshire ;!! the mere £48 10s. paid into the Exchequer by the Receiver of Recusants’ Forfeitures for this county between 1680 and 1685; the general leniency in exacting recusancy fines in Charles IT’s reign*3 and, after 1692, the double land-tax to which Catholics were theoretically liable ;4 the fact that in the years immediately following the ‘fifteen’, when the oaths were most likely to be pressed upon Catholics, no Catholic property-owner in the county came into the unenviable category of ‘popish recusant convict’.t5 The decline of the Knipe family may, then, have been due to other factors than religious persecution and it may be, perhaps, that whereas the earlier members of the family derived their prosperity from their practice as land- agents and lawyers, the later attempted to wring a living from the stubborn soil of the family farm,'® met with little success and at last left the farm (which bears their name to this day) and the district, having already abandoned the religion of their forbears. A fleeting reference to this family occurs in volume IX of the Catholic Record Society’s publications,'7 where the late Edward Doran Webb—editor of a privately printed history of the Arundells of Wardour!*—is quoted as saying that ‘the family was an old Catholic one, settled at Semley, their house with its little chapel still standing’. The house, known as Knipe’s Farm and printed as such on the current Ordnance Survey maps, is still in existence today and until alterations were carried out a few years ago the room on the first floor once used as a chapel could be readily detected as such, for, alone of all the rooms in the house, it possessed a tall, arched window (or ‘door’—see below) of ecclesiastical character, surmounted by a stout stone cross—originally a crucifix, the figure having been chipped away. This room—hardly a ‘little chapel’ (it measures 15 ft. by 18 ft.)—was not accessible from within the house, but only by ladder from outside and neither the window, which was the only mode of entry, nor the cross could be seen from the road in front of the house, though they must have been clearly visible from the fields at the back and the existence of the chapel must have been an open secret in the neighbour- hood. No doubt the priests from Wardour said Mass there from time to time, as may the Father John (or Joseph) Sebastian from Marnhull, Dorset, who ministered to the Catholics of the Shaftesbury area in the 1750’s, and who died on 27th July 1757.19 From a variety of sources—some of them not recognized as ‘recusant history sources’ at all—it is possible to reconstruct something of the history of this family over a period of exactly two hundred years (1638-1837). In 1638 William Knipe drew up the conveyance of the Manor of Christchurch, Hants, which the first Lord Arundell intended to bequeath to the Church,?° and five or six years later he appears again in connection with the Arundell property—he would seem to have acted as steward or land-agent for the family—the occasion being a dispute over the Manor of Hanley in Dorset.?! He also appeared in cases involving the Dorset properties of two other notable Wiltshire Catholics: the second Lord Baltimore (Lord Arundell’s son-in-law) and Edward Codrington, of the Sutton Mandeville family.2 William Knipe does not appear to have taken up arms during the Civil War, but had two-thirds of his Wiltshire estates sequestered for recusancy?3 and was similarly deprived of at first two-thirds and then (from October 1648) one-third iyi (y aaystBay) SPL1 Areniqay pi& uo adiuy aovin & jo [eting 9y1 Sutps0901 yey) st 9a1ZIpag sty} OUT pay ueeq you sey YY m AgTwag Jo siaysiday ysteg ayy ut Aue adiuyy AjUO ayy, :ELON LEgi “q PEgi ‘p &EgI ‘q ofgi ‘q gzgi “q *9B1095) ‘1oura|q ‘uyof ‘uuy Aleyl Logi 92g! ‘poosoo J, oogl “q ‘pjoury Logi -q youIePY ‘Wi 6981095) paempyg ‘wi ‘suet {—________________| 1Sg1 10 1¥gI ut Aafwiag oog! Logi ye you ynq qyeieg “‘ayAeg SIIION, ‘cgi ut aatle uyof ‘wi ‘uuy sopiey ‘wi ‘ouel soure( ‘Ww ‘10URIT A — AIvyy “Ur SUeTTTT MA oLL1 19 pose gl pose gl pase zl pase ‘Ale “1G pesyuod Loli Jo uINIIy nedaq ‘60g1 ‘p ‘PEQI “p ‘1ZQI ‘p ‘6641 ‘p LELi‘p jo a8priqnory, ueso001q, Ul svoulyg ‘ur ‘AIvy ‘— yjoqezipy ‘wi Sparempy “Wet ‘que SUTIG ‘Ul QUTIZYIeD) “ a8e jo srvad Lo 0g jo spremdn, se poqiiosap 6g pese 6L pose ‘6gL1 “p ‘G¢L1 ‘p ‘ecLr-p 6EL1 “p PELip elisp ¥PSL1-p vEL1 ogf1 ‘p | 6bLi ‘p Sidi urate sey ‘Jouea]q ‘gytaspuryy uyof tw Yespug ‘PELT ‘sued Ul: oulotpNe) ‘931005, ‘sewoyy, ‘uyof ‘stueg ut: Asepy ‘— ouef-w ‘prempy “Wey AA J9UINO J, Gil1-p €121 ‘p UOSMETA, — “UG uyo{ ‘ui ‘uuy sewoy ], ‘_ auuayieg “WI ‘9d1025) zLgi ut parp oym Aapuag jo adiuy “ura, ay Aqeqoid £C€gi ‘uuy s Avid 0} paytUIpe “WITT | (uos yg) Auoywuy purparowjsa AA “YULqIey jo uos{I\, — JO Jarysnep ‘ouef cw “arent 00g! v2419 "p “souRT ‘[faUILD Jo sdtuy WITT ‘SLTIM ‘ADTTIWAS JO SAdINN AHL of his Dorset property at Marnhull.?4 He was, moreover, sufficiently ardent a Catholic to send his fifteen-year-old son, George, overseas in 1653 to be educated at Douai College.25 After the Restoration William was repeatedly presented at Quarter Sessions as a popish recusant,?® but he appears to have escaped conviction and its attendant penalties, since his name is not to be found in the Wiltshire portions of the post-Restoration recusant rolls.27 Perhaps his legal knowledge assisted him in eluding the status of “popish recusant convict’, or possibly the donation of £5 6s. 8d. which he subscribed towards the Voluntary Presentation to Charles II, raised by the city of Salisbury in 1661,?* was proving a sound investment. Two years later we find William Knipe named as executor of the will of Dorothy Arundell of Chidcock, Dorset,?9 while another will—that of the eleventh Lord Stourton, who died in 1672—refers to ‘William Knipe, gent.’ as ‘my Steward’.3° William himself died in the same year,3! having made a will two years earlier3? which reflects family pride, Catholicism and a sense of obligation towards the poor. To the latter in a number of parishes in Dorset, Hampshire and Wiltshire (including Semley and Sutton Mandeville in the last-named) he left bequests; to his sons George and Thomas he left lands at Lydlynch, Dorset, in trust for the payment of his debts,33 and added ‘I give to my said sonn George my Silver Boule with Knype’s Armes34 upon it, my silver salt and a dozen of silver spoons’. To his wife he left ‘my great silver Cupp and Cover’ and to their daughter ‘my little silver Cupp’. The will continues: ‘I give to my said sonn Thomas Knype my silver Tankard and eight hundred pounds in money, to be raised out of my Estate next after my Debts.’ A most interesting legacy is one of a hundred pounds ‘to my friend Mr. Dowaie’— clearly a veiled reference to Douai College—while, after various bequests to relatives (‘my son Mawson’, ‘my daughter Tourner’ and ‘son Tourner’) and godchildren, he left legacies which shed a little light upon the relationship of two of the Wiltshire papists, Elizabeth Drew and Joan Barter, who are listed in the recusant rolls: ‘And to Mrs. Elizabeth Drewe, widow, my silver Candle-Cupp with a cover, and to her three children twenty shillings apiece. And to her two sisters Joane and Dorothy Barter,35 five pounds apiece.’ The residue was left equally between William’s two sons, George and Thomas. Both George and Thomas Knipe were presented at Quarter Sessions as popish recusants—George repeatedly but Thomas only twice3’—but, like their father, they appear to have escaped conviction.37 ‘Thomas may perhaps have spent some time outside the county for, as his father’s will indicates, the family also owned property in Dorset; this, indeed, was registered in 1717 by Thomas Knipe in his name and those of Ann (Tourner) his sister and Mary, daughter of George Knipe.3® The latter who had been admitted to Gray’s Inn in 1660 and who practised as a solicitor in Shaftesbury39 was sufficiently prominent a Catholic to be included in the House of Lords’ list of Wiltshire papists drawn up in 1680 in consequence of the ‘Popish Plot’ panic and selected for removal from the county (he was to be sent to Coventry).4° In the following year he appears as sole executor of the will (proved 18th June 1681) of Elizabeth Arundell of Corfe Mullen, Dorset,4! while among a number of legal papers bearing his name in the Wiltshire County Record Offices? is a mortgage which is particularly interesting in that it carries the signature not 173 only of George Knipe, but also of Nathaniel Pigott,43 the most eminent Catholic conveyancer of the late 17th and early 18th centuries, the epitaph on whose monu- ment at ‘T'wickenham, written by his friend, Alexander Pope, testifies to his ability in saving his co-religionists from the rigour of the penal laws: ‘Many he assisted in the law; more he preserved from it.’44 There is evidence that George Knipe acted in a legal capacity on behalf of the Catholic clergy; the minutes of the 1684 assembly of the secular Chapter4s contain a recommendation that its treasurer, the Rev. William Byfleet, who lived nearby,+4° should confer with him, and six years later—as we know from a document dated 23rd May 1690 in the Chapter archives47—he and Mr. Byfleet were entrusted with the disposal of four hundred pounds in lands or money bequeathed by John Foyle, a prominent Wiltshire recusant,4® ‘in order to maintain one or more priests of the secular clergy for the spiritual assistance of the poor Catholics persons residing in the counties of Wilts. Dorset and Somerset’. Meanwhile, in the reign of James II a Mr. Knipe, probably George, was included in a Wiltshire list of “Catholiques that are fitt to be made. . . Justices of the Peace’,49 and on 1oth September 16go—Catholics having been earlier ordered not to travel more than five miles from their usual places of abodes°’—George Knipe, described as ‘of Dorset’, was granted a licence to travel to Shrewsbury on business ‘as often as his occasion shall require within the space of six weeks’.5! George and his wife, Catherine, died in 1713 and 1715 respectively5? and in both cases administration was granted to their son William.s3 They are known to have had at least two other children, Edward and Mary,54 and they, together with William, are mentioned in the will of their uncle, Thomas Knipe (brother of George, see above) who died in 1720.55 His will is of considerable genealogical value in that it mentions, besides William, Edward and Mary, Thomas’s other nephews and nieces, George, ‘Thomas, John, Eleanor, Catherine and Anne.5® With the possible exception of Anne, these appear to have been brothers and sisters of William, Edward and Mary: this is shown by the will of the younger George Knipe who, with his brother Thomas, had registered property at Semley in 171757 and who died there in 1734.55 This will mentions George’s brothers Johns9 and Edward —but not Thomas, who had predeceased him,®° or William—and his sisters Mary, Catherine, Eleanor and Bridget. The first two sisters were stated in the will to be then (1734) in Paris, though Catherine later returned to Semley and died there in 1739.° Of the remaining brothers and sisters, John, Edward, Eleanor and Bridget, the last-named married a John Mandeville,®? while John Knipe was living at Semley in 1736 when he acted as one of the ‘overseers’ of the will of John Hussey of Marnhull, Dorset.®3 This will is of especial interest because it mentions the Bell-Tree House in Bath, ostensibly a lodging-house for fashionable visitors to that city, then in ‘the heyday of Beau Nash’s influence’,®4 but in actual fact a Benedictine headquarters, housing priests and containing a chapel.®s John Knipe died in 1754 and John and Bridget Mande- ville died in the following year.®°® John Knipe left the bulk of his estate to his sister Eleanor (whose will, to be cited later, conveys some idea of its nature) and, after her death, to his nephew William Knipe and his heirs, subject to £100 apiece 174 for William’s brothers and sisters and £20 to his (John’s) servant, Sarah Baker.®7 Before looking at Eleanor’s will, we may note that her other brother, Edward, occurs more than once in the first Wardour register (preserved at Wardour Presbytery). On 16th December 1744 he was a godparent at a baptism in Wardour chapel and on 27th May 1747 the christening of his own daughter Catherine was recorded; this entry shows that her mother’s name was Jane and from the list of deaths at the beginning of the first Wardour register we learn that Edward died on 29th January 1749, aged sixty-four, and his wife on 11th October 1760 (her age is not recorded). They had another daughter, Jane, who is described as such in the earliest of the Semley registers which records her burial on 15th September 1737, and it may be that the two nephews and a niece (William, Edward and Mary) mentioned in Eleanor’s will were also the children of Edward and Jane. That they were brothers and sister is shown by a ‘Survey of the Manor of Semley’ compiled in 1801,°8 which states that on 19th October 1754 a copyhold tenancy was granted to William Knipe for his own life and for those of his brother Edward and sister Mary. In 1767 the persons reported as papists at Semley included Eleanor Knipe, William her nephew and her two nieces Catherine and Mary, the former unmarried and the latter ‘married to a poor labourer’.°9 It is noteworthy that Eleanor’s other nephew, Edward, was not listed as a papist, and in view of evidence to be adduced later we can safely infer that he had conformed to the Established Church. The humble status of Mary’s husband is the first indication of some decline in the family fortunes—a decline to which a number of later parish register entries bear witness. William, however, 1s described in the 1767 report as a ‘gentleman’ and Eleanor’s will suggests that something at least remained of the family’s earlier possessions. She left five pounds to each of her nieces Catherine, Anne and Mary, wife of Phineas Wyatt, and added that apart from these bequests, ‘I give devise and bequeath all my Mony Plate Jewells Household Goods Cattle Stock Credits and Effects whatso- ever and wheresoever to my said nephew William Knype and appoint him my sole and whole Executor.’ At the foot of the document is added, ‘I likewise bequeath to my Nephew Edward Knype One Shilling and no more.’ A codicil was added to the will on 14th July 1770, when Eleanor was on her death-bed and too infirm to sign her name, and administration was granted to her nephew William on 16th Septem- ber 1771.7° In 1779 and for twenty years afterwards William (a customary tenant and freeholder) attended regularly as a member of the Homage of the Semley Manor- court.7! He also occurs in the Anglican return of 1780 which includes Mary Wyatt but not Catherine Knipe.7? William died in 1799 aged seventy-two and Mary Wyatt in 1809 aged eighty-one. 73 It appears from the Semley land-tax assessments74 that William was occupying Knipe’s Farm until his death and that his brother Edward, to whom administration of his will was granted in November 1799,75 lived there until 1807 or 1808. From 1808 to his death in 1821,76 however, Edward is listed as the proprietor but not as the occupier, the latter being his son-in-law Charles Baker, whose relationship we know from Edward’s will (to be cited later) and who continued in occupation 175 after Edward’s death, the farm having then come into the ownership of George Parham.77 That Edward had long abandoned his family’s religion is suggested by several pieces of converging evidence: his exclusion from the Anglican diocesan returns of papists in 1767 and 1780,78 his service as a land-tax collector and assessor in 1800 and 1801 respectively,79 and the presence of his name in the Wiltshire Poll Books for 1818 and 1819,8° when the granting of the franchise to Catholics lay still a decade in the future. Conformity to the Established Church, moreover, could well be the explanation of his aunt’s cutting him off with a shilling. When he died in 1821 Edward left his freehold property, Sevior’s Farm, to his widow Elizabeth, after whose death it was to be sold and the proceeds divided among his son William and his three sons-in-law (James Norris, James Barret and Charles Baker, husbands of his daughters Eleanor, Ann and Jane respectively).*! That he had enjoyed a com- fortable standard of living is suggested by the mention in his will of ‘Plate Linen China Books Pictures’ and ‘Monies Securities for Money Stock and Implements in Husbandry’, while in addition to Sevior’s Farm his will refers to another ‘Freehold Messuage or Dwellinghouse called Corner’ and to ‘all other my real Estate what- soever and wheresoever’. A William Knipe, probably Edward’s son, was presented before the Manor Court of Semley in 1832 and again in 1834 for encroachments on the common, *? and he appears earlier to have ceased to consider himself a Catholic—though he may well be the William Knipe who had been confirmed in Wardour chapel by Bishop Walmesley, the Catholic Vicar-Apostolic of the Western District, on 21st October 1787°83—for he and his wife Mary had their children christened in Semley parish church in 1807 and 1810.*°4 These were Jane and George respectively, the latter being described as a labourer when he married Harriet ‘Toogood in 1827,85 while Jane and her husband, a widower named Edward Arnold, could neither of them sign their names in the register when they were married in the previous year.*® George and Harriet had four children between 1828 and 1837, all of them baptized in the parish church,*7 and the latter date is the last on which any trace of this family has been discovered at Semley.*® ‘The information given in the preceding pages does, however, enable a genealogical table to be constructed with some degree of assurance and this is appended, while the early post-Restoration list of “Romanists’ printed below has a relevance wider than that of the family and the parish with which these pages have been concerned. APPENDIX MS. ‘List of the Romanists in the Diocese of Sarum’ (Wilts. portion), compiled between the Restoration and 1672.89 Wilcot : Mrs. Barbara Skilling. Mr. Edward Skilling. Dorothy Pontine. Willm. Pontine and his wife. Michael Pontine. Sutton Mandeville : Mr. Edw. Cuddrington and his wife. Mr. Tho. Anne and his wife. Mrs. Cuddrington, widd. Mrs. Trim. widd. Jo. Leastead. Susan Leastead, widd. Ursely Button. Salisbury, St. Thomas: Salisbury, St. Edmund’s : Kingston Deverill: Sr. Jo. Penruddock and Mr. Tettersill. Wm. Seam and Coronel Touchet. Jo. Skammell and his wife. Andrew and Lawrence his sons and Alice his daughter. The wife of Andrew Leversuch. Fonthill Gifford : Charles Woolmer. Edw. Basely and his wife. Rob. White and his wife. Hindon: Walter Kantelo and his wife. Phill. Kellaway and his wife. James Jopson and his wife. Semley : Mr. Wm. Knip and his wife and George his sonne. Widd. Trim and her daughter. t V.C.H. Wilts., 11, 92. 2 Diocesan Archives, Salisbury: Returns of Papists, Box 1 (‘List of the Romanists in the Diocese of Sarum’). See Appendix for further details. 3 Lambeth Palace MS. 639, f. 254, verso. A draft of the Salisbury diocesan portion of this census, and a notebook of Bishop Seth Ward, in which the incumbents’ original returns were rearranged, are among the diocesan archives. 4 House of Lords Record Office: 1705/6 Return of Papists (Salisbury section); diocesan archives, Salisbury: 1706 return (in ‘Returns of Papists’, Box 1). See also my article ‘Wiltshire Catholicism in the Early 18th Century’, in Recusant History, vu, no. 1 (Bognor Regis, 1963). 5 Diocesan Archives, Salisbury: Returns of Papists, Box 1. ® See The Diary of Fanny Burney (Everyman edition), 57: letter to her father from Salisbury, 1ith June 1780, and E. T. Long, ‘Wardour Revisited’, in The Tablet, 29th October 1955, 430. 7 Diocesan Archives, Salisbury: Returns of Papists, Box 2; V.C.H. Wilts., m1, 96 (Table ‘A’). 8 Diocesan Archives, Salisbury: Returns of Papists, Box 2. In 1767 the prevalence of Catholi- cism in two other parishes was attributed to the Wardour influence, viz. Fonthill Gifford (‘A much greater number than I could wish . . . But we are situated too near Lord Arundell’) and Tisbury (‘Ld. Arundell is Ld. of the Manor and living near us, and consequently many of the lower class of people depending chiefly on him for their subsistence and following his Lordship for loaves and fishes . . .’). Both letters are in Returns of Papists, Box 1, in Salisbury Diocesan Archives. See also my article ‘Some Eighteenth-Century Conversions’ (Essex Recusant, 11, no. 3, Brentwood, 1961) for further examples of the cwjus regio, ejus religio principle under Catholic lords of the manor. 9 G. Oliver, Collections Illustrating the History of the Catholic Religion in Cornwall, Devon, Dorset, Somerset, Wiltshire and Gloucester (1857), and T. B. Trappes-Lomax, ‘Roman Catholicism’, in V.C.H. Wilts., ut. 10 Not too much should be made of this; see W. P. Baker, Parish Registers and Illiteracy in E. Yorkshire (E. Yorks. Local History Society, York, 1961). The author’s remark (zbid., 7) that ‘there is evidence that the ability to read was more general than the ability to write’ should be noted and may be compared with The Children of the New Forest (end of Chap. 3): ‘The chaplain at Arnwood had taken a fancy to him’ (Jacob Armitage) ‘and taught him to read—writing he had not acquired.’ ™ Public Record Office: E.377/68-82 (Wilts. portions). ™ Public Record Office, Pells Receipt Books, series E.401/1965-1975; see also my article ‘English Catholicism under Charles II: the Legal Position’, in Recusant History, vu, no. 3 (Appendix). 177 "3 See my article ‘Some Sidelights on Recusancy Finance under Charles IP’ (Dublin Review, Autumn 1959, 245-54), and also the Recusant History article cited in the previous footnote. 14 Much work remains to be done on this subject; see, however, W. R. Ward, The English Land-Tax in the 18th Century (1953), 33; J. Anthony Williams, ‘An Unexamined Aspect of the Penal Laws: the Problem of the Double Land-Tax’ (Dublin Review, Spring 1959, 32-7), and Miss M. Rowlands, “The Iron Age of Double Taxes’, in Staffordshire Catholic History, No. 3 (Stafford, 1963), 30-46. ™s Of three lists, all compiled about 1720, among the Treasury papers (Public Record Office, series T.1/227, No. 6) one, after recording that the annual value of papists’ estates in Wiltshire was £5,232 13s. 6d., leaves blank a final column headed ‘Amounts of two-thirds of Estates of Popish Recusants Convict’; another leaves blank two columns headed ‘Amounts of Annual Rents of the Estates of Papists Convict’? and ‘Two Thirds thereof’, while the third list, in which convicted recusants in various counties are named, contains no Wiltshire names and, again, no figures in the ‘two-thirds’ column for this county. These docu- ments were drawn up as a result of an Act of 1714 (1 Geo. I, st. 2, cap. 55) ‘to oblige Papists to Register their Names and Real Estates’. which recalls that Catholic recusants were liable to forfeit two-thirds of their landed property—under 29 Eliz. I, cap. 6 and 3 Jac. I, cap. 4—and requires all who had not taken the oaths by the end of Trinity term 1716 to register their estates at Quarter Sessions ‘to the end that their estates may be certainly known and discovered for the purpose aforesaid’; hence the ‘Enrolments of Papists’ Estates, 1717-1788’ among the Wiltshire and other county archives, and in the Public Record Office. Abstracts of the latter are printed in E. E. Estcourt and J. O. Payne, English Catholic Nonjurors of 1715 (1886). ™6 Long known locally as a poor and difficult farm, with frequent changes of occupant (infor- mation from the present owner, Mr. 8. A. Godeseth). t7 Published in igt1. The reference occurs on pp. 127-8. ™8 Notes by the 12th Lord Arundell of Wardour on the Family History (1916). 19 For evidence of his death at Marnhull, probably ‘from a fever caught attending a poor family in Shaftesbury’, see Catholic Record Society, LVI (1964), 171. His death is also recorded in the first Wardour register (at Wardour Presbytery) and in the Tisbury parish register. I am grateful to Father J. Paine, S.J., and the Rev. F. H. Phillips for access to these registers. 20 Catholic Record Society, X (1911), 127-30. This transaction is also mentioned by J. Waylen, ‘The Wiltshire Compounders’, in the W.A.M., xx (1887), 323-4, and in an unpublished docu- ment of 1666, showing that the terms of the will had not been carried out at that date (Westminster Cathedral Archives: Series A, XX XII, no. 116). The only William Knipe in the registers of the 178 Inns of Court for this period* is one who was admitted to Gray’s Inn in 1635 and who was described as ‘son and heir of Anthony Knipe of Fairbanck, Westmoreland, gent.’. The latter was doubtless the sixth son of a William Knipe who married ‘Jane, daughter of — Wilson of Fairbanck, co. Westmoreland’ (Chetham Society, Uxxxv, Man- chester, 1872, 170). William’s father, another William, who died about 1600, was one of the Knipes of Cartmel, Lancs. (ibid.), a branch of a family of whom some at least were recusants (Catholic Record Society, vi (1909), 238, note 2), though this surname does not occur in The Lanca- shire Elizabethan Recusants, by J. 5. Leatherbarrow (Chetham Society, Manchester, 1947). 21 Waylen, op. cit., 322-3. The author describes William Knipe as ‘professional agent for the Arundells’. 22 Notes and Queries for Somerset and Dorset, 1x (Sherborne, 1905), 124. 23 Public Record Office: State Papers Domestic (Committee for Compounding), S.P.23/142, no. 578. See also Calendars of the Committee for Com- pounding, V, 3186 and introduction, xxxii-iv. 24 C. H. Mayo (ed.), The Minute Books of the Dorset Standing Committee, 29rd Sept. 1646 to 8th May 1650 (Exeter, 1902), 25, 50, 238, 444. The Marn- hull property, ‘Mr. Knype’s tenement’, described and valued in a survey dated 16th April 1657, was sold in the following January (Dorset County Record Office, Dorchester: MSS. 5853, 5854). A Mr. Knipe was assessed for hearth-tax on a house in Marnhull in 1664 (C. A. F. Meekings, Dorset Hearth Tax Assessments, 1662-1664, Dor- chester, 1951, 58), but not in 1672 (Dorset County Museum, Dorchester: Hearth Tax, Dorset, 347). There are no Knipes among the Marnhull payers of poor-rate, 1699-1740 (Dorset Record Office: Overseers of the Poor Accounts for these dates, ref. P.32/Ov. 1). 25 Catholic Record Society, x1 (1911), 526. 26 County Record Office, Trowbridge: Great Rolls, Hilary term, 1661-1672 (presentments by the constables and Hundred-jury of Chalke for each of these years). 27 Public Record Office, E.377/68-82 (1663-_ 1691). 28 Salisbury Borough Records: Document Z.228. 29 J. J. Howard and H. S. Hughes, Genealogical Collections Illustrating the History of the Roman Catholic Families of England, 1 (1887), 198. 30 C.B.J., Lord Mowbray, Segrave and Stourton, The History of the Noble House of Stourton (1899), I, 500. * The registers of the Inns of Court referred to above are: J. Foster, The Register of Admissions to Gray’s Inn, 1521-1889 (1889), in which the Knipe entry occurs on page 209; Records of the Honourable Society of Lincoln’s Inn: Admissions, 1420-1799 (1896) ; F. A. Inderwick, A Calendar of the Inner Temple Records, II: 1603-1660 (1898), and H. A. C. Sturgess, Register of Admissions to the Honourable Society of the Middle Temple, I: 1501-1781 (1949). For a tentative pedigree of the Knipes of Semley, see p. 172. 31 1672 is the year in which his will (see next footnote) was proved, and also the last year in which he was presented at Quarter Sessions (County Record Office, Trowbridge: Great Roll, Hilary term, 1672: Hundred of Chalke). 32 Somerset House, Principal Probate Registry: Wills (Prerogative Court of Canterbury) 77 Eure. 33 No Knipe entries are to be found in The Registers of Lydlinch, co. Dorset, 1559-1812 (ed. C. H. Mayo and F. G. Henley, 1899). 34 The name of Knipe, or Knype, does not occur in the Heralds’ Visitations of Wiltshire between 1553 and 1667, as summarized by F. A. Carrington in the W.A.M., u (1855), nor in the Harleian Society volume published in 1954 (Wiltshire Visitation Pedigrees, 1623, ed. G. D. Squibb), but ‘Knipes Arms’—those of the Lanca- shire branch from whom the Wiltshire family were probably descended (see note 20)—are described in Burke’s General Armory (1878 edition), 573, while the corresponding Knipe crest—a wolf’s head pierced through the throat by an arrow—is reproduced in Fairbairn’s Crests of the Families of Great Britain and Ireland (ed. J. Maclaren: 2 vols., London, no date) as crest no. 12 in plate 40. 35 Public Record Office, E.377/68, 73, 75-81 (Elizabeth Drew of Baverstock, Wilts.) and E.377/82 (Joan Barter, two entries). 36 George most years from 1661 to 1690: Thomas in 1662 and 1681 only (County Record Office, Trowbridge, Great Rolls, Hilary term, Hundred of Chalke). 37 They do not occur in the Wiltshire portions of the recusant rolls, 1663-1690 (Public Record Office, E.377/68-82). 38 E. E. Estcourt and J. O. Payne, English Catholic Nonjurors of 1715 (1886), 43. 39 Cf. J. Foster, The Register of Admissions to Gray’s Inn, 1521-1889 (1889), 289, and W. Good- child, William Horder, yeoman, 1683: a Story of Cranbourne Chase (8 pp., Devizes, 1930), 5-7. 4° House of Lords Record Office: Main Papers, 321, c66. His is the eighth name on the list. In the V.C.H. Wilts., m1, 94, n. go, the list is rearranged in alphabetical order and his name is printed as ‘George Kulpe’. In the original MS. it is clearly Knipe. This Papists (Removal and Disarming) Bill never became law; see Historical Manuscripts Commission, Eleventh Report, Appendix, Pt. 2 (1887), 222-5. 41 J. J. Howard and H. S. Hughes, Genealogical Collections Illustrating the History of the Roman Catholic Families of England, i (1887), 199. 4 E.g. Beaton MSS. 409/9; Simpson MSS. 130/4; Penruddocke MSS. 332/2, 27, 46. 43 Beaton MSS. 409/9. It also names Elizabeth Drew (see above, p. 173) and another convicted recusant, Francis Anne of Grovely who occurs in recusant rolls nos. 68, 75 and 82 (Public Record Office series E.377), and in the House of Lords list of prominent papists, 1680 (see note 40, above). 44 Cited by J. Gillow, A Literary and Biographical History or Bibliographical Dictionary, of the English Catholics, v (1902), 311. See also C. Butler, Historical Memoirs of English, Irish and Scottish Catholics (1822 edition), rv, 459, and J. Kirk, Biographies of English Catholics, 1700-1800 (1909), 184. 45 Westminster Cathedral Archives, Series A, XXXIV, 219. 46 ©. . with Mr. Gildon near Shaftesbury in Dorsetsh.’ (Westminster Cathedral Archives, Series A, XXXIV, 211). See also my article, ‘Who was William Byfleet?’ in The Downside Review, Winter 1960-61, 48-9. 47 Archives of the Old Brotherhood of the English Secular Clergy: O.B. IV, no. 48 (no. 49 is another document on the same subject, signed by William Stourton and William and George Knipe). I am most grateful to Monsignor D. Shanahan of Hornchurch, Essex, for this informa- tion. 48 He occurs in recusant rolls nos. 68 and 73 (Public Record Office series E.377) and in the 1680 list of prominent papists (House of Lords MS. 321, c66). 49 Sir G. Duckett, ‘Proposed Repeal of the Test and Penal Statutes by King James II’, in W.A.M., xvii (1879), 368; also Duckett, Penal Laws and Test Act (1882-83), 1, 220. 5° Public Record Office: Privy Council Register, P.C. 2/73, 458-9; R. Steele, Tudor and Stuart Proclamations (1910), nos. 4038-9. st Public Record Office: Privy Council Register, P:@. 2/74, 6. 5? Semley Parish Registers: Register I (Burials 1709-1799); Baptisms and Marriages to 1753) entries dated 21st January 1713 and 24th May 1715. I am indebted to the Rev. J. H. Westcott for access to the Semley registers. 53 Somerset House, Principal Probate Registry: P.C.C. Admon. Act Books, 1714 (April) and 1715 (June). 54 For Edward, see Notes and Queries for Somerset and Dorset, xv (1917), 222; for Mary, E. E. Estcourt and J. O. Payne, English Catholic Nonjurors of 1715 (1886), 43. 55 His burial on 6th July 1720 is recorded in the Register mentioned in n. 52 above. 56 Somerset House, Principal Probate Registry: Wills (P.C.C.) 256 Shaller. See also J. O. Payne, Records of the English Catholics of 1715 (1889), 73. 57 County Record Office, Trowbridge: Enrol- ments of Papists’ Estates, 1717-1788; see also Estcourt and Payne, op. cit., 286, 287 (George Knipe also registered property at Donhead St. Andrew). 58 His burial, on 17th September 1734, is recorded in Semley Register I. 59 Somerset House, Principal Probate Registry: Wills (P.C.C.) 33 Ducie; J. O. Payne, Records of the English Catholics of 1715 (1889), 73. In 1723 George and John Knipe of Semley registered property at Kingstone near Yeovil, Somerset (Somerset County Record Office, Taunton: Enrolments of Papists’ Estates, 1717-88). This property is mentioned in the will of their uncle, Thomas Knipe, 1720 (see n. 56, above). 60 Semley Register I: burial-entry dated 18th May 1730. 61 Jbid., burial-entry dated 2oth(?) March 1745. 179 62 J. O. Payne, Records of the English Catholics of 1715 (1889), 73. Her husband was not, one hopes, the individual of that name who turned informer and who solicited of the Commissioners for Super- stitious Lands a reward for revealing the contents of a Catholic will (cf. Payne, op. cit., 139). 63 J. O. Payne, Records of the English Catholics of 1715 (1889), 11-12. 64 R. A. L. Smith, Bath (1948 ed.), 62. 6s For further details of the Bell-Tree house and chapel and of Bath Catholicism generally, see my Bath and Rome: the Living Link (Bath, 1963), the index of which contains many Wiltshire references. 66 Semley Register I. John Knipe was buried in August 1754 and John and Bridget Mandeville in April and September 1755 respectively. 67 Somerset House, Principal Probate Registry: Wills (P.C.C.) 253 Pinfold. 68 County Record Office, ‘Trowbridge: ref. 413/37. 69 Diocesan Archives, Salisbury: Returns of Papists, Box 1. 7° County Record Office, Trowbridge: Wills, Archdeaconry of Sarum. 71 Devizes Museum, Library of the Wiltshire Archaeological and Natural History Society: Bennet-Stanford Papers (‘Notes from Semley Court Books’). 7 Diocesan Archives, Salisbury: Returns of Papists, Box 2. Catherine had married, ten years earlier, Silas Troubridge of the neighbouring parish of Donhead St. Mary (Semley Registers, III: Marriages 1754-1812). 73 Semley Registers, II: Baptisms and Burials 1779-1812. Mary Wyatt’s husband was presumably not a Catholic: he does not occur in the Anglican returns of 1767 and 1780 and his son was baptized in the parish church (Semley Register, I: Baptism of William, son of ‘Finis’—i.e. Phineas—and Mary Wyatt on 7th June 1767). 74 County Record Office, Trowbridge: Land Tax Assessments, Hundred of Chalke, 1780-1831 (the Semley sheet for 1794 is missing). 75 Somerset House, Principal Probate Registry: P.C.C. Admon. Act Book, 1799 (November). 76 Semley Register, VII: Burials 1813-1889 (4th October 1821). 77 County Record Office, Trowbridge: Land- Tax Assessments, Hundred of Chalke, 1780-1831. From 1822 the proprietor is shown as George Parham, and from 1828 to 1831 as the ‘trustees to G. Parham’ (or ‘to the late G. Parham’). By 1828 the tenant was no longer Charles Baker, but William Brown, and in 1831, Robert Brown. 180 The latter is listed in the 1841 census (Public Record Office, H.O. 107/1170—Semley booklet) as of ‘Amberley’s’ (i.e. Amber Lease Farm; cf. Gover, Mawer and Stenton, Place-Names of Wilts., 209; Andrews’ and Dury’s Map of Wilts., section 4; W.A.M., xxvr (1892), 358), which he may perhaps have been farming in conjunction with Knipe’s Farm, for two years later the marriage took place in Semley parish church of John Rogers, a labourer aged 23, and Jane Brown aged 24, both of Knipe’s Farm (Register V, 11th June 1843), the latter being the daughter of James Brown, labourer. The census of 1851 shows ‘Thomas Brown aged 37 at Knipe’s Farm, farming 54 acres and employing four labourers (Public Record Office, H.O. 107/1849, Book 266/1). Subsequent occupants of this property can be traced through the Post Office Directories for Wiltshire—beginning with that for 1855, which shows Thomas Brown still there— and Kelly's Directories. 78 Diocesan Archives, Salisbury: Returns of Papists, Boxes 1 and 2, respectively. 79 County Record Office, Trowbridge: Land- Tax Assessments, Hundred of Chalke, 1780-1831. 80 See respectively, J. Hodding, jun., The Poll for the Election of Two Knights for the County of Wilts. : 1818 (Salisbury, no date), 6; and G. Butt, The Poll of the Freeholders of the County of Wilts., 1819 (Salis- bury, no date), 6. The presence of his name in the Poll Books would not in itself be enough to show that he had conformed; several known Catholics did in fact vote in 18th-century Wiltshire elections (see my Recusant History article cited in note 4, esp. p. 15). 8: The land-tax assessments (n. 79, above) show that Elizabeth rented Sevior’s Farm first to James White and then to George Targett. 82 Devizes Museum (Library); Bennet-Stanford Papers: ‘Notes from Semley Court Books’. 83 Wardour Presbytery: First Register. 84 Semley Registers, II: Baptisms and Burials, 1799-1812. 85 Semley Registers, IV: Marriages 1813-1837 (23rd August 1827). 86 Jbid. (24th January 1826). 87 Semley Registers, VI: Baptisms 1813-1854; - Mary Ann (baptized 7th September 1828), John (25th December 1830), Eleanor (19th May 1833), George (27th August 1837). Eleanor died in 1834 (Register, VII: burial dated 2nd April). 88 The 1841 and 1851 censuses (cited in n. 77, above) contain no one of the name of Knipe in this parish. 89 1672 was the year of William Knipe’s death (see p. 173). NOTES A STONE AXE FRAGMENT FROM HAMSHILL DITCHES, BARFORD ST. MARTIN The end of a broken stone axe was found at the above site in April 1963. It lay only 5 in. deep in the face of a small exposure beneath the fence dividing Barford Down from Grovely Wood (SU/06203321). The spot is immediately north of feature H in enclosure G of Crawford’s sketch plan of the site.1 The fragment was kindly examined by the South Western Implement Petrology Committee (serial no. 1244) and identified as ‘an albitised dolerite or greenstone’. It is otherwise of minor interest for two reasons: firstly, the barest hint of Neolithic activity recorded at Hamshill Ditches? might provide a possible context for the axe or, conversely, be strengthened by it; and secondly, in view of the visibly Iron Age/Romano-British nature of the site, the fragment might be another example of a ‘thunderbolt’ having been deliberately preserved into a later period for magico-ceremonial purposes.3 P. J. FOWLER AVEBURY: THE NORTHERN INNER CIRCLE In his master plan of the Avebury monument, Stukeley shows (Abury, Tab. I) two circles centred on The Cove, an outer one of some thirty stones and an inner one of twelve. Four stones of the outer circle are still to be seen; eight or nine of those visible in Stukeley’s time have been destroyed; some of those shown as missing by Stukeley may in fact lie buried in the ground. There are now no surface traces of the internal circle. In Stukeley’s plan five stones are shown prostrate and one, on the north-eastern arc, standing. The latter survived until shortly after 1825; a water-colour by J. Browne, dated to that year, depicts this stone (a reproduction of the water-colour was published by H. St. George Gray in Archaeologia, LXxxtv, pl. xxxiii). A. C. Smith claimed (British and Roman Antiquities of North Wiltshire, 141 and pl. v) to have found a buried stone, his ‘18’, that had probably stood on the north-western arc. In February 1964, the owners of the monument, the National Trust, caused a trench about 3 ft. deep and 150 ft. long to be dug across part of the interior of the Northern Inner Circle for the purpose of moving a pipe that supplies a watering-trough for cattle. At a point 60 ft. south-south-east of the south-eastern corner of the southern stone of The Cove, the trench cut through part of a burning-pit some 15 ft. in length. The pit was filled with the charred straw and flakes of sarsen that invariably attest the destruction of a stone by the fire-and-water method. The destroyed stone was undoubtedly the one that Stukeley shows lying south-south-east of The Cove in a gap between buildings to the rear of the 18th-century inn. It now appears highly probable that the existence of Stukeley’s internal circle would be substantiated by excavation. It is also probable that within this small circle there had been another feature, a linear setting of small sarsens similar to the one revealed by Keiller’s excavations in the Southern Inner Circle. At a point 30 ft. east of the southern stone of The Cove the surfaces of three buried stones appear above ground. They were trenched round by St. George Gray, who found that they lie end-to-end in a shallow pit. The method of burial, exactly that employed for the concealment of the stones of the setting in the Southern Inner Circle, suggests that here, too, the sarsens had stood in a row and close together. I. F. SMITH BARROWS NEAR CLATFORD In 1955 two barrow rings were observed as cropmarks west of bowl barrow Grinsell No. 3 (Preshute parish), in a field just north of the London-Bath road near Clatford (W.A.M., Lvt (1955), 193). In August 1962 these were again conspicuous, and the more 181 easterly showed apparent traces of an inner ring as previously noted in the other. They were also seen independently by Dr. Isobel Smith and Mr. W. E. V. Young. In 1962, for the first time, another ring was distinct about 50 yds. south-east from Preshute 3, rather smaller than the other two. Dr. Smith reported marks as of another barrow just north-west of the most westerly ring, and this later showed up more clearly as a very small circle on the extreme edge of the field. It is remarkable that four barrow rings should be seen from the ground in this one field, and air photography might reveal others in a good season. Photographs taken by Dr. Smith have been deposited in Devizes Museum Library. O. MEYRICK BARROWS IN SAVERNAKE FOREST From the reservoir in the Deer Park south of Tottenham House a beech avenue runs south-east along the crest of the ridge. About 130 yds. along the line of the avenue is a mound some 4 ft. high and 36 ft. in diameter, which has all the appearance of a bowl barrow. ‘Two trees grow from it, together with a decayed stump. ‘The avenue was planted in the first half of the 18th century (WV.A.M., Lu (1948), 180), which gives an earlier dating for the mound. About 100 yds. south-south-west of this and 20 yds. inside the deer fence is another possible barrow, also planted on, 3 ft. high and 20 ft. across (NGR. SU/248631). O. MEYRICK SITING OF BARROWS ROUND SPRINGS In December 1960 there was an exceptional rising of springs, and the Rockley Bourne, for the first time since 1915, rose at the north end of Rockley Park beside the Marlborough- Wootton Bassett road (NGR. SU/164729). Here is a cluster of barrows dotted about on either side of the stream bed. With the higher water level generally prevailing in prehistoric times it can be surmised that the bourne may have risen from this point and that the barrows were sited round the springhead instead of on higher ground more commonly chosen. A further hint of a regular flow is provided by willow charcoal and water vole bones from a Late Bronze Age site half a mile to the north, indicating a constant water supply close at hand. A chance visit later the same winter revealed that the Lambourn in Berkshire had its source right in the midst of the well-known Seven Barrows group (in fact numbering over twenty), where they are most closely gathered at a sharp bend in the road from Lambourn. This is about 2 miles above its normal rising point, and from the contours of the land it does not appear ever to have flowed from a still higher spring. It can hardly be fortuitous that these two barrow concentrations were thus placed around springs. O. MEYRICK ROMANO-BRITISH INTERMENTS AT PARSONAGE FARM, WINSLEY In September 1962, an inhumation burial within a small stone coffin was discovered i ft. below ground surface, during ploughing, in a field approximately a quarter of a mile south-east of Parsonage Farm in the parish of Winsley (NGR. ST'/79936201). The coffin, a square-ended type, was hewn from a single block of Bath stone. Its internal measurements were as follows: length 3 ft., width (at the head), 1 ft. 14 in., 7 in. (at the foot), depth of walls 84 in. The thickness of the walls varied between 3? in. and 7 in. (at the ends). Orientation was north-west to south-east with head to the north- west as clearly shown by the internal narrowing of the sides to admit the lower part of the body. The grave cover and the contents of the coffin, with the exception of a few fragmentary rib bones, obviously of a small child, had unfortunately been dispersed, and could not be examined. 182 Whilst clearing away disturbed soil from the coffin a further interment was revealed jammed up against the head on its eastern side. This consisted of fragments of a single cooking pot, in a greyish brown fabric, of late grd- to 4th-century a.p. date, containing burnt bones. The vessel, which had originally been placed upright in the ground, was only fragmentary, and most of the upper portion was missing, doubtless removed by plough action owing to the shallowness of the burial. All the sherds were extremely worn and much abraded at the edges, suggesting that the breaks were of considerable antiquity; possibly the vessel was incomplete when buried. The position of the cremation, however, right up against the head of the coffin, does suggest that the two interments were almost cer tainly contemporaneous. Information has been recorded on the O.S. 6-in. maps, and photographs are now deposited in the Society’s library. The cooking pot and its contents (Accession No. 22/62), have been placed in the Museum through the kindness of Mr. H. S. Boles of Parsonage Farm. F. K. ANNABLE BENSON'S FOLLY, A FORGOTTEN SOUTH WILTSHIRE FIELD MONUMENT On Idmiston Down, slightly west of the spot marked ‘Tower Hill’ on O.S. maps, is a small hillock which can be identified as the site of Benson’s Folly, under which name it is described on Andrews’ and Dury’s map of 1773. William Benson replaced Sir Christopher Wren as Surveyor-General of Works, and was also Sheriff of Wiltshire in 1710. He came into possession of the Manor of Newton Tony in 1709 when he purchased it from the Fiennes family. According to Britton (Description of the County of Wilts (1814), 385), he built the present Wilbury House and also improved the estate by introducing extensive plantations and enclosures. Doubtless the improvements on the estate included the building of the Folly on the spot now identified as its site almost exactly 2 miles from Wilbury House. The hillock is overgrown with scrub, but a recent examination of its top revealed blocks of stone. These are not of local origin (i.e. Chilmark), but probably came from the Bath area, possibly at the same time as stone was obtained for the house. Although the Folly exists now only as foundations, we have a thumbnail sketch of it on Andrews’ and Dury’s map. This sketch shows it as a triple tower of similar size to Wilbury House, but doubtless its scale is much exaggerated. When this map was made the owner of Wilbury was a Mr. Greville. By 1841 the spot was marked as ‘Tower Hill on J. and C. Walker’s map of Wiltshire and it is not clear from the map whether the Folly was still standing—it possibly was because the spot is marked by a black square. In fact no information can be discovered as to when the Folly collapsed or was destroyed. We are on more certain grounds as to the date of its erection for, if one assumes that William Benson built the Folly at the same time as Wilbury House, then the date for its erection could not have been long after 1709, and it must certainly have been erected before 1754, as William Benson died in that year. Benson’s Folly might therefore rank as one of the earliest, if not the earliest, of this type of structure to be erected in Wiltshire. In this con- nection it is rather a coincidence that Celia Fiennes should have been travelling England at the beginning of the 18th century and observing a large number of newly-built houses and the associated novelty of landscape gardening. Did she, one wonders, observe the novelty taking form on her own family’s former estate at Wilbury ? JOHN MUSTY AN ELUSIVE WHITE HORSE ON ROCKLEY DOWN (PL. XIV) In 1948 the figure of a white horse was exposed by the plough on Rockley Down, beside the beech belt running west from the Marlborough-Hackpen road beyond Rockley Park, and was the subject of a note by the late Mr. H. C. Brentnall (W.A.M., Lu (1948), 396). It came as a complete surprise to all except old shepherds who had seen the form 183 of a horse showing through in dry seasons, before these downs were turned over to corn. It was subsequently photographed from an R.A.F. plane based at South Cerney and Dr. T. R. Thomson secured a print. This was on matt surface and so not suitable for reproduction, but is now in Devizes Museum library. However, an admirable air photograph had also been taken by Dr. J. K. St. Joseph, Curator in Aerial Photography, Cambridge University, and is now reproduced here. It shows a fair specimen of the breed, suggesting an early 19th-century date, when the road from Marlborough to Swindon past Four Mile Clump was in common use, from which it would show up prominently. Though Mr. Brentnall notes a proposal to preserve it, the horse is now again lost under corn for a great part of the year, and it is likely that its outline will crumble and the figure be obscured by soil slipping down on to it under constant ploughing. A permanent record of it in its prime is, therefore, desirable. Its remote position invites speculation as to the cutters. Perhaps it was the whim of the occupant of Rockley Manor and executed under his direction. The files of a local paper may hold the secret, but the Marlborough Times, which would have been the most likely source, did not appear till 1859. Probably the horse was not kept in trim for long, and soon was but a memory to be lost altogether in the passing of a generation. We are indebted to the Committee for Aerial Photography, University of Cambridge, for permission to reproduce the photograph. O. MEYRICK CROPMARKS NEAR CROFTON In July 1962 Mr. C. N. Tilley drew attention to a cropmark showing intermittently in corn for 2 miles and following roughly the 600-ft. contour line south of Savernake Forest from Durley to Langfield Copse, dropping down near Crofton Pumping Station and then winding up the hillside south of Wilton Brail. It may be a disused droveway; according to one of the oldest local inhabitants it was the line taken in his father’s time by shepherds going from Cirencester to Weyhill Fair with their flocks. It would be better for the purpose than the high road, affording pasture as they moved along. O. MEYRICK « Crawford and Keiller, Wessex from the Air 2 V.C.H. Wilts., 1 (1957), 36. (1928), 97-9, fig. 16, Pl. XIb. 3 Cf. Stone, W.A.M., Lrv (1951), 192. 184 EXCAVATION AND FIELDWORK IN WILTSHIRE 1963 ALL CANNINGS: RYBURY CAMP (SU/083640) Neolithic/Iron Age An examination of Rybury Camp in the course of field survey for the Royal Com- mission on Historical Monuments (England) was carried out by Mr. D. J. Bonney. In his admirable description of this site (Antiquity, Iv (1930), 38-40), Curwen suggested that it was of two phases, a Neolithic causewayed camp, defined by two lines of rampart, and a univallate Iron Age hill-fort overlying it and obliterating much of its outer rampart. He further suggested that a cross-dyke on Clifford’s Hill, 220 yds. to the south, was also Neolithic and associated with the causewayed camp. That there is an Iron Age hill-fort at Rybury has not been questioned, but subsequent writers appear either to have ignored or rejected Curwen’s view of a Neolithic origin for the earthworks. The present excavation did indicate, however, that Curwen was almost certainly right. There are, undoubtedly, two phases, and the interrupted nature of much of the ditches and the denuded banks behind them suggest strongly that the earlier is Neolithic. A trial trench, 15 ft. by 3 ft., was cut, with the ready permission of the farmer, Mr. H. W. Daw, into the outer ditch of the earlier earthwork, near where it emerges from under the Iron Age defences. It had silted level with its outer lip and appeared now as a narrow shelf set into the slope. It was flat-bottomed and 7 ft. deep at the outer face, but its width was not recovered since the inner face lay, thickly masked by soil creep, far back under the slope down from the bank. Waste flint flakes were found in abundance throughout the filling, over 600 all told, but no worked implements. A few teeth and scraps of bone were also found. Such evidence, though limited, lends further support to Curwen’s interpretation of Rybury, and it would appear reasonable to accept this until positive proof to the contrary may be offered. WOODFORD: WOODFORD LONG BARROW (SU/102377) Neolithic During September and October, the Woodford long barrow was excavated by Major and Mrs. H. F. W. L. Vatcher on behalf of the Ministry of Public Building and Works. The investigation revealed a sequence of five main phases of activity at the site, each separated by a considerable gap in time. In the primary phase six large pits were dug, probably in the course of open-cast flint mining. After partly silting up the pits were filled in and the area levelled, material being obtained from a quarry on the east side of the barrow. Two separate successive timber buildings (Phases 2 and 3), were later constructed, one rectangular, 36 ft. by 12 ft., the other irregularly trapezoidal, 16 ft. by 16 ft. at its widest point. Phase 4 was represented by the construction of the barrow which consisted of a rectangular flint cairn covering a few weathered human bones, afterwards capped by a chalk mound, the mound material being supplied from flanking ditches. The final phase, dated to the Late Bronze Age, was marked by the re-cutting of the barrow ditches, an inhumation on the edge of a newly-dug ditch, and two pits outside the latter containing unaccompanied cremations. FYFIELD AND OVERTON DOWNS: NEAR MARLBOROUGH Bronze Age-Medieval The fifth successive season of work in the area, directed by P. J. Fowler, completed the excavation of the medieval settlement at Wroughton Copse on Fyfield Down, continued 185 the investigation of ancient field boundaries begun in 1961, and tested other features with a view to further excavation. Many flint flakes and sarsen chips were found in excavating around the base of a much-polished and recumbent sarsen stone (SU/12887150), shown to have formerly stood upright. Perhaps of more significance were an iron wedge (for splitting sarsens ?) and a short cross silver halfpenny of c. 1200 A.p. Further south on Overton Down, much Iron Age pottery was obtained from the area of a probable settlement (B.3), and hut platforms (13127009), apparently of a small, unenclosed settlement integrated with ‘Celtic’ fields, were noted about } mile to the south towards enclosure C.2 (Second Report, fig. 2). A section through a ‘Celtic’ field lynchet of group D.1, lying between these two settlements, produced evidence of pre-lynchet marking-out features and suggested an origin of the field in the Early Iron Age or earlier, and an abandonment during the Romano-British period and after. In contrast, six cuttings through the boundaries of, and actually within, the distinctive oblong fields laid off from a straight central bank on Totterdown (within group D.2, clearly visible in First Report, pl. Ib), indicated an origin, use and abandonment within the Romano-British period, possibly limited in fact to the first two centuries A.D. Three sections through the hollowed trackway (F.4) bounding the whole of the ‘Celtic’ field group D.2 on the north-west showed, as anticipated (First Report, 107), that it followed the line of a silted-up ditch. The ditch itself, though clearly prehistoric, was not conclusively dated. A cutting on the north of Totterdown to pick up the ditch where it had been ploughed over within a ‘Celtic’ field showed that its bank here had originally been sarsen revetted, and produced beneath the tumble from the revetment shallow V-shaped grooves in the protected chalk surface of the negative lynchet which can reasonably be interpreted as plough-marks. Final work on the medieval settlement of Raddun (First Report, 113) attempted, with mixed success, to solve questions outstanding from the previous four years’ excavations. The main results were a stratified pottery sequence from south of Building IT; a pre- medieval ditch underlying the south end of House I (excavation of which, incidentally, produced an almost complete pot sunk into apparently undisturbed clay-with-flints beneath the house floor level); and, most satisfactorily, since it was the observation of this relationship on an air photograph which initiated the excavation, stratigraphical proof that the bank of Enclosure B really was built on top of a ‘Celtic’ field lynchet (First Report, 109). WYLYE: BILBURY RINGS (SU/010363) Iron Age A fifth season of excavation directed by the Rev. E. H. Steele took place at Bilbury Rings, Wylye, by courtesy of Messrs. A. and R. Barrett and Miss D. Barrett, from. 24th August to 8th September. The first object of the investigation was to make a further probe into the area of the Romano-British settlement site where traces of hut floors had been found in 1962 (W.A.M., 58 (1963), 468). It speedily became clear that any hope of similar finds was unfounded, for the settlement area lies just off the northern fringe of the clay-with-flints capping which overlies the chalk along the top of the Grovely Ridge. As a result, there is no more than a few inches of ploughsoil above the solid chalk, instead of a thick layer of clay dividing the two. These few inches have been so disturbed by ploughing that no trace of any structure could remain, while the underlying chalk is so riddled with solution hollows that post-holes cannot be detected. It must be considered a very fortunate circum- stance that the nearly destroyed remains of a floor were found previously. The second area of investigation lay at the southern rampart of the hill-fort, where Hoare believed that there were signs of an original entrance (Ancient Wilts., 1, 108). He also recorded a small barrow nearby. It is still possible to see a faint swelling in the ground at the site of the barrow, but the levelling has been too destructive for any traces of 186 structure to be observed in a trench section. However, further excavation revealed that the outer ditch of the ramparts had been interrupted at this point, and that the barrow, if such it really is, lay central to the causeway so created, and apparently a little outside the line of defences. A most inclement period of weather seriously hindered the work. A further investiga- tion will, it is hoped, show whether the causeway also interrupts the main ditch of the ramparts, attesting again the accuracy of Hoare’s observation. In this case the question will arise as to whether the barrow may not in fact be a part of a gateway defence work. If there is no way across the main ditch, the inference will be that the later outer ditch was deliberately interrupted in order to respect the barrow. MILDENHALL: BLACK FIELD (Cunetio) (SU/216695) Romano-British Walled Township During excavations carried out in 1960 by Messrs. F. K. Annable and A. J. Clark at the west entrance to the town, a well was located sealed beneath the massive footings of the west wall which runs through the small paddock flanking Cock-a-Troop Lane. The well was cleared during a week’s digging in the summer. It was an unlined, circular shaft 4 ft. in diameter dug through the natural chalk. The sides were remarkably straight, and the well diameter remained constant throughout its depth of 20 ft. below the base of the town wall. At the bottom the sides converged to form a rough hemisphere beneath a thin primary filling of grey clayey silt. The well filling was remarkably homogeneous, and apart from a handful of animal bones the finds consisted entirely of Samian and native coarse wares. So far it has been possible to make up two almost complete vessels whose sherds were scattered through a depth of 2-3 ft. within the filling. Savernake type pottery and plain and decorated Samian occurred together at all levels, and the entire Samian series has been dated to between A.D. 50-60. The obvious homogeneous nature of the fill, and the absence of other occupa- tional material implies a very rapid closing up once the decision to abandon the well had been taken. It would seem therefore that this was carried out within the second half of the 1st century A.D., and thus we have indisputable evidence of a 1st-century occupation, the earliest yet known, at this extremely interesting site. WESTBURY: WELLHEAD (S1'/873502) Romano-British Further investigations carried out by Lt.-Col. W. D. Shaw in the garden of his house, since his earlier discovery of a lime-kiln of 4th-century date (W.A.M., 58 (1962), 245), have produced evidence of a considerable settlement extending over at least an acre, with iron smelting, farming, and possible pottery manufacture amongst its chief occu- pations. Structural remains in situ were scanty, but the discovery of dry limestone footings and level floors some 2-3 ft. below ground surface suggests the presence of house or hut sites. Other dwellings nearby are indicated by finds of Pennant sandstone roof-tiles, floor and box-tile fragments, and quantities of wattle and daub. Samian and native coarse wares occur in great quantity; amongst the Samian types were bowl fragments of Forms 31 and 37, stamped bases, and some sherds repaired with rivets. Much of the coarse pottery was of New Forest type, and included flanged bowls, flagons, storage jars with heavy rims, indented beakers and colour-coated wares. ‘The prevalence of kiln wasters, and oval-shaped clay ‘saggers’ does suggest, however, the likelihood of pottery kilns being found in the locality. Proton magnetometer surveys carried out by Dr. Aitken, Oxford University, and Dr. Gill, Bristol University, during 1962-63 failed to locate the kilns, which may lie in a wooded area on the north side of the settlement. Many small finds were made which fully illustrate the life of the settlement. These included querns and whetstones, glass fragments, iron, copper and bronze objects of a domestic nature, and quantities of iron slag as proof of ironworking at the site. Bone pins 187 and beads, spindle whorls, antler tines worked into tools, and many animal bones were recovered; and to judge from the latter, sheep, including a horned variety, was the domi- nant animal in the local economy. By contrast, only nine coins of late grd-4th centuries a.p. have so far been found, but they confirm the essentially late character of the occupation within the Roman period. Perhaps the most striking discovery at the site consists of almost 100 sherds which have been identified by Mr. P. J. Fowler as grass-tempered ware of middle or late Saxon date. The site has evidently been much disturbed for they were closely associated with Romano- British pottery within the hut-sites almost at floor level. A limited excavation in the area is planned for 1964 to seek further evidence of Saxon occupation, and the identification of pottery kiln sites in the Roman period. CRICKLADE: PARSONAGE FARM (SU/097935) Pre-Conquest Excavations were undertaken by the Ministry of Public Building and Works on Parsonage Farm, Cricklade, in September 1963; the work was carried out under the direction of Dr. C. A. Ralegh Radford. The site covered the south-west corner of the enclosure. Sections across the earthwork revealed the existence of two periods. The earlier consisted of a clay bank with internal strengthening of timber. ‘To this was added a stone revetment. A small ditch, apparently of the earlier period, was found in front of a berm rather over 20 ft. wide. Disturbance on the outer edge of the field about 150 ft. in front of the bank, suggested the former existence of an outer ditch, similar to that recorded in the lower part of the site. The original make-up of the bank included fragments of early medieval pottery, which are not in themselves closely datable; there is proof that this bank is post-Roman and no evidence to contradict an ascription to the age of King Alfred. No new evidence for the date of the wall was found. It was destroyed during the 12th century and could well be of late pre-Conquest date. Both the timber reinforcing of the bank and its relation to the ditch suggest the influence of Carolingian fortifications. A trench within the bank brought to light the line of an original roadway. RESCUE .AND RESEARCH WORK IN THE SALISBURY AREA INVESTIGATIONS UNDERTAKEN this year by the Research Committee of the Salisbury Museum under the direction of Messrs. J. W. G. Musty and D. J. Algar are summarized below. GOMELDON: (SU/182356) Deserted Medieval Village The deserted medieval village of Gomeldon lies on the slopes of a hill and is over- looked by the present village, which is a modern resettlement situated on the top of this. hill. During 1963 the first of a series of excavations was undertaken to explore the village remains. Work was concentrated on a visible building site (Building 1) and a flat area adjacent to it; another structure (Building 2) was found in this latter area. Both sites were totally excavated. Building 1 was found to be 42 ft. long and 13 ft. wide with ground-walls of unmortared flint trimmed on the outer face. The east long wall bowed outwards, but the west long wall curved inwards. Only one entrance could be identified, this was in the west long wall and was denoted by two post-holes and a gravel ramp. Any evidence which may have existed for an ‘opposed entrance’ in the east wall had been destroyed by the trench dug for an electricity cable. At the north end, there was a drain running across the width of the building, and it seems likely therefore that this part of the building, at least, was occupied by animals—a number of horseshoes of late 13th-century date were also found on the floor near this drain. Indeed, there was no evidence to suggest that the building 188 was a house; no hearth was found and little pottery was recovered from inside the walls. Thus the building may have been a large byre or barn associated with a farmhouse not yet excavated (possibly Building 3 which is a prominent earthwork adjacent to it). All the finds from the building which include, in addition to the horseshoes, cooking-pot rims with well-moulded profiles, a strap-end buckle and a hunting arrow, point to a late 13th- or 14th-century date for it. Building 2 was found to be of ‘long-house’ type with opposed entrances and a hearth. As in Building 1 the ground-walls were of unmortared flint, but in addition, there was also evidence of timber trussing; the combination of the two types of construction in this building makes it one of unusual interest. The building was 24 ft. long and 14 ft. wide with stone walling zn situ in the southern half of the house. The main roof truss had consisted of timbers of cross-section 10 in. by 7 in. and these had been set into the chalk bedrock to a depth of 1 ft. 7 in. at an angle of approxi- mately 60°. This truss would have divided the house approximately in half. There was also possibly another truss of more slender proportions near the south end of the building, and a large post-hole at the mid-point of the south end wall. A series of posts on the line of the main truss had clearly served to support a partition which divided the living quarters (south end) from the byre (north end). These posts had been burnt to the ground and never renewed, an event which probably led to the replacement of the roof and the intro- duction of the flint ground-walls as an additional feature. Pottery found scattered across the floor is of 12th-century date, so that this building is a very early example of the use of stone in the construction of a medieval peasant house. Outside the west long wall, at the south-west corner, was a subsidiary building, wholly of timber and represented by an approximately square setting of post-holes asso- ciated with a rectangular shallow pit with burnt clay in its fill. This building is interpreted as an attached bake-house, possibly entered from the main house via an interior door. Above the flint tumble from the collapsed walls (which occupied approximately half the house interior and conveniently sealed the floor levels) was a thick spread of late 13th-century pottery. This level represents the conversion of the area of the house into a yard, possibly for Building 1, and a dumping area for rubbish (which contained a silver penny of Edward III) from another house. Exploration of the down-hill edge of the house platform revealed a quarried area of r2th-century date which was only superficially examined. It was also discovered that a (?)prehistoric ditch crossed the platform beneath the floor of Building 2. No dating evidence for this ditch was forthcoming, but the post-holes of one of the entrances to the building, and one of the main truss-holes, were cut in it. WILTON, KINGSBURY SQUARE: Possible site of Saxon Burh Kingsbury Square, Wilton, is thought to have been the centre of the Saxon Town and when a site in the Square (that of the demolished Talbot and Wyvern Hall) became temporarily available for examination the opportunity was taken to carry out a trial excavation in an attempt to obtain a stratified series of pottery and evidence for the Saxon Burh. However, the earliest pottery recovered was of 12th-century date, although the cuttings were taken down to below the water table. In the upper levels, the remains of a medieval house were uncovered, including a massive wall of greensand blocks and an area of herringbone tiling, the latter possibly belonging to the hearth. This house was probably of 13th- or 14th-century date. FIELD WORK IN THE SALISBURY AREA TISBURY: WARDOUR (ST/924273) Medieval Immediately before ploughing in 1962, earthwork remains indicating the former extent of the medieval settlement at Wardour were recognized on the north-west slope 189 13 down from the grounds of New Wardour Castle towards the south bank of the river Nadder. The part of the site now ploughed was planned by the Royal Commission on Historical Monuments (England); much of the rest remains in pasture. The settlement was bounded on the west by a ditch, with closes running back to it from building platforms beside a former hollow-way, now a hedge-line. On the highest ground was a 33-ft. deep, stone-lined well, topped by a large flat stone with a much worn bucket-hole and slots for the, presumably, wooden winding gear. Subsequently, much 12th- to 13th-century and 18th- to 1gth-century pottery was collected from the site. The recognition of the site might well be relevant to the disputed location of Saxon Wardour. WEST TISBURY: WICK FARM (ST'/94152861) Medieval While looking for the site of the possible Saxon burh at Tisbury, the earthworks of a small almost deserted settlement associated with a 14-acre rectangular enclosure were noted at Wick Farm, } mile south-west of ‘Tisbury church. Wyk is first recorded in 1225, and trial excavations carried out by P. J. Fowler, mainly to date the enclosure, indicated occupation in the 12th century as well. For a more detailed note with plan, see Antiquity, XXXVII (1963), 290-3. ACKNOWLEDGEMENTS Thanks are accorded to the following for their contributions to this summary of excavations: D. J. Bonney (Rybury Camp); P. J. Fowler (Fyfield and Overton Downs, Wardour and Wick Farm); J. W. G. Musty (Rescue and Research Work in the Salisbury Area); Dr. C. A. Ralegh Radford (Cricklade); Lt.-Col. W. D. Shaw (Westbury) ; Rev. E. H. Steele (Bilbury Rings); Mrs. F. de M. Vatcher (Woodford Long Barrow). 1 The two main earlier reports are in W.A.M., respectively. References B.3, F.4, etc., are to the 58 (1962), 98-115, and (1963), 342-50, here classification of sites by categories in First Report, referred to as the First Report and Second Report 99-109. 190 OBITUARIES The Rev. Stanley Harper died at Sunderland in February 1963, aged 73. From Durham University he was made deacon in 1915. Ordained priest at Durham in 1917, he served two years with the R.A.M.C. After the war he officiated at Trowbridge, Warminster and St. Thomas’ Church, Salisbury, before being appointed vicar of Worton and Marston in 1928. Here he was to remain till his retirement in 1961, when the parishes were amalgamated with Poulshot, to make way for a younger man. He leaves a widow and two sons. Obit.: Wiltshire Gazette, 21st February 1963. Arthur Julius Hosier died at Wexcombe on 3rd April 1963, aged 85. Son of a farmer, he started at 21 with a 150 acre farm at Chaddington, near Wootton Bassett. Next taking over for six years an agricultural engineer’s business at Pewsey, he gained much useful experience for later life. With his brother Joshua he then bought farms at Ramsbury, Froxfield and elsewhere before acquiring 1,700 acres at Wexcombe in 1920. Here they invented the open-air milk bail which was to revolutionize dairy farming. The partnership was dissolved in 1929 and Hosier bought the neighbouring holding of Brunton and subsequently added other farms in Wiltshire and adjoining counties. The open-air system was extended to pigs, and other inventions such as the car and tractor hay sweep brought him many awards. His advice was sought by scientists and politicians; he received the O.B.E. in 1949, and Cambridge University honoured him by the bestowal of the degree of LL.D. in 1951. With his son Frank, who died in August 1962, he published the Hosier Farming System in 1950. Twice married, he leaves a son and three daughters. Obit.: Wiltshire Gazette, 11th April 1963. Thomas William Blackburn Middleton died on 3rd February 1963. Educated at Christ’s Hospital and Trinity College, Cambridge, he joined the staff of Marlborough College in 1936. A brilliant teacher of mathematics, he was also proficient in coaching cricket and rugger. After the outbreak of war he served with the 2nd Battalion Wiltshire Regiment in the Middle East, Italy and Germany. Returning to Marlborough he took over command of the C.C.F. Since 1953 he was bursar as well as housemaster and the combined duties undermined his health, but he remained always genial and considerate. He married in 1940 Miss Ann Vigors of East Kennett, who survives him with a son and daughter. Obit.: Marlborough Times, 9th February 1963. Edward Robert Pole died on 19th November 1962, aged 80. Born at Great Bedwyn, he, like his brother, Sir Felix Pole, made his career with the railways, and for many years acted as his secretary. Living most of his life at Bedwyn, he was very active in local affairs and no one knew more about the village and its history. He formed a large collection of music boxes and mechanical toys, and this was put on exhibition in London a few years ago. In 1960 he left Bedwyn to live at Bramshaw, Hants. Obit.: Marlborough Times, 24th November 1962. George Smith died at Brighton on 17th March 1963, aged 91. Son of George William Smith, a noted geologist, he was sub-librarian of University College, London, before taking charge of the Linen Hall Library, Belfast. Returning to London, one of his first tasks was to arrange Westminster Cathedral Library for Cardinal Vaughan. He came to live at Great Bedwyn in 1927, taking a keen interest in the village, acting as churchwarden, and as representative on the Diocesan Conference. He was a Fellow of the Library Association and himself owned a large and valuable collection of books. He left Bedwyn in 1959 on the death of his wife. Obit.: Wiltshire Gazette, 21st March 1963. IQI REVIEWS Prehistoric England, by J. G. D. Clark. Pp. 200, 50 pls., 36 line illustrations. Batsford, 1963. 5s. This book, first published over twenty years ago, has long provided a reliable, and readable, introduction to English prehistory and a new edition, in paperback, is most welcome: At little more than the cost of twenty cigarettes it is value indeed. It is the author’s declared intention to describe ‘within the narrow limits imposed by deficiencies in our present knowledge’ and, it may be added, two hundred pages ‘the manner in which our forefathers lived before the dawn of history—no mean task. Professor Clark is, however, remarkably successful in this task, and his book scores over comparable works by its far more intelligible layout. Separate chapters are devoted to specific aspects of life in prehistoric times—the Food-Quest, Dwellings, Mining and Trade, etc.—in each case treated chronologically, in preference to the chronological treatment of the whole subject which is more often employed. The plates, though fewer than in earlier editions, are well chosen and reproduced. This book possesses the signal merit of perspective, not so very common in the literature of prehistory, and it deserves to be read widely, not merely by those seeking an introduction to the subject. D. J. BONNEY The Prehistoric Ridge Way: A Journey, by Patrick Crampton. Pp. vi and 78. Abbey Press, 1962. 6s. This is a useful guide-book to the Ridge Way, taking in the antiquities bordering on the track and some not so close, particularly on the Berkshire Downs. The writer has been at pains to keep abreast of recent researches and fieldwork, and a bibliography gives the references for these. It was unlucky for him that his journey should have been made before the latest excavation of Wayland Smith’s Cave. He and a companion followed the track from Avebury to Streatley, camping out en route, with their equipment carried on a donkey, whose occasionally capricious ways make entertaining reading. Probably a present-day Modestine is the best solution to the transport problem on the Ridge Way. Time was, not so long ago, when it was ideal galloping ground before army vehicles and tractors scored deep ruts over long stretches of it, whilst in the last century, as we are told, Welsh coal carters brought their waggons this way, as a better surface than the road through the Vale, selling to villagers below. This little book should certainly be in the rucksack of any future Ridge Way travellers. O. MEYRICK The Experimental Earthwork on Overton Down Wiltshire, 1960, edited by P. A. Jewell. Pp. viii and 100, 8 pls., 38 text figures and 15 tables. British Association for the Advancement of Science, 1963. 25s. The Nature Reserve on Fyfield and Overton Downs, near Marlborough, is a precious and untouchable island of the old downland scenery safe from the tide of modern agri- culture. Here, among the luxuriant turf, the weathered thorns and the immemorial sarsen stones, the eye is startled by a neat enclosure containing a ditch and a gleaming white chalk bank, over 70 ft. long and visible for many a mile, from which project a series of calibrated steel poles. This is a scientific experiment in the construction and denudation of earthworks and the behaviour of small finds, planned to last 100 years, the first of a series initiated by the Research Committee on Archaeological Field Experiments of the 192 British Association for the Advancement of Science. The committee was formed after discussion at the meeting of the British Association in 1958, and is under the chairmanship of Dr. G. W. Dimbleby, with Dr. P. A. Jewell, the editor of this report, as Secretary. The actual earthwork was constructed by a team of well-organized volunteers in 1960. Similar experiments are planned for different soil types (a second earthwork has already been built on sandy heathland in Dorset), chalk being given priority because of the frequency with which it is encountered in excavation. This report is intended as a general introduction to the whole project. The tremendous expansion of the archaeological horizon by the analytical, environ- mental and conservational sciences in recent years is generally acknowledged. Experiment has received less attention, but promises to be equally fruitful: certainly a study of earth- work behaviour is overdue, for, apart from a pioneer experiment by (naturally enough) Pitt Rivers, and some observations by Curwen and others, archaeologists have had no quantitative data with which to assess the processes and speed of silting and denudation, and have had to rely mainly on intuition. Unfortunately such an experiment must take its own time, and we must fill in the 100-year wait with this first report and the interim reports which will appear in the Proceedings of the Prehistoric Society. The present volume is a feast of valuable material. The first chapter gives a historical introduction to the study of earthworks, followed by descriptions of the different types of weathering—hbiological, chemical, physical and differential—that are at work on them, and an assessment of the value of pollen analysis. Further chapters describe the inception of the project; the painstaking work of design, survey and construction; buried materials; work study and primitive tools; environment; publicity, volunteers, camp organization and cost; theoretical aspects of the morphology of banks and ditches; and there are several appendices. It was decided that the earthwork should be linear. Using a template of novel design, the ditch was cut to a carefully predetermined flat-bottomed profile; the bank was built up with great care in several layers, using calibrated vertical steel poles accurately set in concrete and stretched wires and strings as markers. The interfaces of layers not readily differentiated were marked with scatters of different types of roadstone, which will act as markers of soil movement and mixing. Polythene tubes set up vertically in the bank will, by their distortion, indicate the flow of the bank material. Specimens of textiles, leather, wood, cooked and uncooked animal bone, human bone both fresh and cremated, flints and simulated potsherds were buried in different environmental situations: under the turf stack forming the lowest layer of the bank, within the chalk body of the bank and just below the adjacent ground surface. The fresh human bone was of known blood group, so that a study could be made of the survival of this important characteristic. A further valuable secondary experiment was the digging of part of the ditch with primitive tools—antler picks, ox and horse scapulae as shovels and baskets for carrying. These, with the exception of the scapulae, compared surprisingly well in efficiency with modern tools. The bank is divided up so that a test section can be cut beside each of the calibrated poles in succession at intervals of 2, 4, 8, 16, 32 and 100 years after the construction. The first section has already been cut, and after the cutting of the second in 1964, an interim report will be published. There seems little to criticize but minor points. On page 27, horizontal is misspelt. On page 28 the authors have some difficulty in explaining why rubble transported in buckets compacts to a smaller volume when tipped into a heap: this is surely because mutual filling of the interstices is reduced against the side of the bucket, an effect which grows in importance with the size of the pieces. On page 34 the polythene tubes are mentioned for the first time with no explanation of their purpose, whereas they should have been introduced in the section on Planning, on page 23. While the bank was being piled around them, these tubes might have been kept straight more easily if steel rods had been temporarily inserted in them. 193 It is a pity that this beautiful work is marred by the loss of the main set of photo- graphic plates, their place being taken by a secondary set, some of which are not in focus: perhaps in subsequent work a double set of high-quality pictures could be taken. Trouble seems also to have been experienced with postal delivery of samples, some of which were late, and others damaged. Personal collection seems well justified in such an important matter. Finally, there is no index, but perhaps this is compensated for by a very full list of contents and references. Such is the exhaustiveness of this report, that it even contains an appendix listing the volunteers’ menus for a week—relevant perhaps to the work-study tests that were made, and quite a tempting guide to others faced with this difficult type of catering operation; no doubt it will also be of historical interest to the readers of the final report in 100 years. On close inspection of the list, one finds (unless it is a misprint), that tea is served with breakfast on every day but Sunday. Is this one more experiment—in human endurance ? The volume is a well-laid-out and workmanlike production, in keeping with its content, although a crisper text typeface would have been more in keeping with the severe sanserif of the main titles. It is rather vulnerably bound in paper, which seems hardly good enough for the price. This brief notice does little justice to the exactness and exhaustiveness of the measure- ments made on all aspects of the construction and environment of this experiment, nor to the careful thought that has gone into its design. I can commend it as a fascinating lesson in scientific method from which one must gain in archaeological awareness. ANTHONY CLARK The Scientist and Archaeology, edited by Edward Pyddoke. Pp. xiii and 208, 32 line drawings and 25 pls. Phoenix House, 1963. 30s. The publishers of this volume have a distinguished list of archaeological books to their credit, many of them standard textbooks or original studies. Against that back- ground, this book is somewhat disappointing. It is too diffuse, and the chapters are too short, for it to be a standard text; and too much of it has been published elsewhere, although admittedly not in book form, for it to be considered an original work. It is additionally open to criticism because, at the end of his Introduction, the editor expresses an expectation that ‘each paper will be recognized as an authoritative account of the principles of its subject which will stand for some time to come’. Such a view is perhaps possible of one or two contributions, but of the majority it is patently not true. Further, such a claim immediately invites unfavourable comparison of the book with the much more original, voluminous and expensive Sczence in Archaeology (ed. Brothwell and Higgs; Thames and Hudson, 1963), in the shadow of which it appears. Had the claim been that this was an honest attempt at popularization in the best sense, presenting to a lay audience (including non-scientific archaeologists) some of the scientific techniques now used to archaeological ends, then a judgment of the book must have been more favourable. The book contains nine chapters, each written by a separate specialist who is directly involved in, and in most cases has pioneered the archaeological use of, the techniques he describes. Some of these are already standard procedure in the correct circumstances— resistivity surveying, petrological analysis, pollen analysis and CG 14 dating—but the thin- sectioning of pottery, the fluorine, uranium and nitrogen dating of bone, and physical methods of chemical analysis, are as yet less well known. The plates are of variable quality, those of pollen grains and metal structure being, as always, captivating to the lay eye; but those of apparatus, and of people fiddling with same, are, as always to that same eye, singularly unilluminating. Despite the diverse authorship and, happily, the resulting stylistic differences, there is a certain dull uniformity about the presentation, degenerating at times into simple bad writing, e.g. (at random) ‘Consequent on the isotopic fractionation effect and other factors . . .’ Although interesting as far as it goes, the subject-matter 194 is by no means comprehensive: various new surveying methods, the increasing archaeo- logical use of computers, and the pathological approach towards bone examination, with all its economic and social implications, are but three developments coming straight to mind which are not covered. Anybody practising archaeology seriously today will realize that the study of man in antiquity is passing through a revolutionary phase. Antiquarianism on the one hand, and the hearty if depressing dedication of the ill-informed week-end digger in his solitary and misdirected trench on the other, are as irrelevant to the present developments in archaeological thought and practice as is a soup spoon to a fish dish. The subject ‘archaeology’, even as it was understood recently, is no longer self-sufficient. Probably it never has been, for its origins in and early links with some Natural and indeed Social Sciences have often been appraised. But now, suddenly, numerous research workers with a scientific rather than humanistic background, using a battery of both known and specially-developed techniques, have turned to tackle problems deriving from archaeology and the material on which it is based. Some of the problems themselves are new, in that previously the right questions had not been conceived, let alone asked. The possibilities of extending our knowledge seem limitless. What irony that, at this moment, the sites, the fossilized landscapes which most of all can give that knowledge a context, are being butchered and obliterated. It is at least a consolation that the archaeology of what was on and in the ground was the subject of earlier books from Phoenix House, books which now daily increase in worth as the things they speak of disappear. It will be interesting to note how well The Scientist and Archaeology lasts. P. J. FOWLER The Families of Allnutt and Allnatt, by Arthur H. Noble. Pp. 164, 17 pls. Privately printed, 1962. This elaborate and painstaking account of families which belonged mainly to Buckinghamshire and Oxfordshire only concerns Wiltshire because two brothers moved to New Swindon from the neighbourhood of Wallingford in the mid-1gth century. Its place in the Society’s library will, however, be amply justified if it moves any member to imitate its thorough method of recording each line of a family, and its concern with members of the clan as people rather than as mere entries in a pedigree. Its format and illustrations will probably only excite the envy of any who try to emulate its method. K. H. ROGERS Staying with the Aunts, by Ida Gandy. Pp. 144, drawings by Lynton Lamb. Collins, TG0G 21s: Despite the attractions of living with all mod. con., there remains for most of us a yearning for the days which were spacious and leisurely. Even those so-called ‘lower classes’ who uncomplainingly accepted conditions which now fill us with horror, seem to have enjoyed part of an existence which has vanished while amenities have been improving. Mrs. Gandy has recalled vividly the atmosphere in which her aunts passed their days, and has been able to admit us to a share in their outlook, with its tenacious clinging to the way of life they knew and loved. How keenly we sympathize with their desperate fight against the encroaching bricks and tarmacadam beyond the toll-bridge! It is almost ungrateful to wish that more had been said of holidays with the aunts, and less of their pathetic decline. Much of this book is concerned with Eling Manor, on the banks of Southampton Water, but Wiltshire readers will thank the author for the description of mid-Victorian days at Baverstock Rectory. E. H. STEELE 195 ACCESSIONS TO THE LIBRARY, 1963 BOOKS PRESENTED Hemingby’s Register, edited by Helena Chew. Records Branch. Vol. 18 Area Effects in Cepaea, by A. J. Cain and J. D. Currey. Phil. Trans. Royal Society Two Wiltshire Villages, by Miss D. Ibberson Royal Wilts, by Lt.-Col. P. W. Pitt A New Survey of England, by N. Salmon. 1729 Burke’s Landed Gentry. 1937 Melksham Spa: Guide and Directory. 1963 A Lost Roman Road, by B. Berry. Allen and Unwin, 1963 The Caliph of Fonthill, by H. A. N. Brockman. Werner Laurie, 1956 The Megalith Builders of Western Europe, by Glyn Daniel. Hutchin- son, 1963 Family Prayers on the Pentateuch, by W. B. Whitmarsh. Nisbet and Co., 1848 Clerical Reminiscences, by ‘Senex’. Seely, Jackson and Halliday, 1880 Anglo-Saxon England and the Norman Conquest, by H. R. Loyn. Longmans, 1962 Staying with the Aunts, by Ida Gandy. Collins, 1963 A collection of Inclosure Acts, Sale Particulars and other documents Bath and Rome, the Living Link, by J. A. Williams. 1963 Transcript of Parish Registers of Easton Royal In addition to the above gifts of books we must record with very great pleasure the most generous gift of books and pamphlets from the library of the late Miss Braidwood, generously presented by her brother, Mr. P. Braidwood, and amounting to 58 items Notes on the Episcopal Polity, by 'T. W. Marshall. 1844 The Ancestry of Abel Lunt, by W. G. Davis. 1963 Mere Parish Magazine, 1892-1911 Catalogue of the Earl of Radnor’s Pictures, Part 2 BOOKS BOUGHT DONOR Authors Author Mrs. Shadbolt R. Sandell Dr. T. R. Thomson Miss 'T. E. Vernon Publishers R. Sandell Publishers K. Rogers K. Rogers Publishers Publishers K. Rogers Author A. M. Colliard K. Rogers K. Rogers K. Rogers The Church in Anglo-Saxon England, by the Reverend C. J. Godfrey. 1962 The Story of the Wiltshire Regiment, by Col. N. C. E. Kenrick. Gale and Polden, 1963 Excavations at Camerton, Somerset, by W. J. Wedlake. 1958 British Copper Coins: Parts 1 and 2. Seaby, 1961 Notes on 18th-Century Tokens, by A. W. Waters. Seaby, 1954 English Silver Coinage from 1649. Seaby, 1957 Collins’ Field Guide to Archaeology, by E. S. Wood. Collins, 1963 W. L. Bowles’ Complete Works. James Nichol, 1855 Miss Scrope’s Answer. R. Baldwin, 1749 Chippenham New Hall: Minutes, 1834-1855 196 The Excavation of Staple Howe, by T. C. M. Brewster. Dennis and Sons, 1963 Wiltshire, by N. Pevsner. 1963 The Experimental Earthwork at Overton Down. British Association, 1963 Industrial Archaeology, by K. Hudson. John Baker, 1963 Fletcher’s Directory of Devizes. 1963 Wills and their Whereabouts, by A. J. Camp. Society of Genealogists, 1963 The Clay Tobacco Pipe in Great Britain, by L. S. Harley. Essex Field Club, 1963 Fasti Ecclesiae Anglicanae, 1300-1541; Salisbury Diocese. Institute of Historical Research, 1962 Marlborough, by Kenneth Mason. 1963 PAMPHLETS GIVEN The Three Winterbournes The Westward Expansion of Wessex, by W. G. Hoskins, 1960 Records of the Seasons; copy belonging to T. H. Baker and con- taining his notes Sale Catalogue of Avebury Manor and letter from Miss Kemm Various papers and estate maps relating to Avebury Map of Avebury, dated 1733: farm belonging to Richard Holford Index to Wills from the Diocesan Registrar Inclosure Act for Steeple Ashton Collection of antiquarian notes concerning the White family Notes on the Gacelyn family Rubbings of brasses at Blunsdon St. Leonard Water-colour of Stonehenge Westwood Church: Historical Notes Trowbridge St. James: Mission Room Alderton: account of village in 1845 Wiltshire Catholicism in Early 18th Century, by J. A. Williams English Catholicism under Charles I, by J. A. Williams DONOR J: S:. Judd Dr. T. R. Thomson Miss A. M. Baker Miss Kemm Miss Kemm Miss Kemm C.R.O. C.R.O. R. C. Hatchwell Mrs. Waley Miss Mepstead R. de C. Nankivel H. Ross K. Rogers H. Ross Author Author Special mention must be made of three volumes presented by Mr. R. S. Newall, F.S.A. These comprise two notebooks made by Sir Richard Colt Hoare and called Castles and Camps, and Roman Roads, respectively, and are complementary to those we already have in the Library. The third contains similar archaeological notes compiled by H. P. Wyndham, who made possible some of William Cunnington’s early excavations. 197 ANNUAL GENERAL MEETING, 1963 THE ANNUAL GENERAL MEETING for 1963 was held at St. Margaret’s Hall, Bradford- on-Avon, on Saturday, 25th May, at 11.15 a.m., the President, Mr. E. G. H. Kempson, taking the Chair. A short address of welcome was given by Mr. H. G. Baker, Vice- Chairman of Bradford-on-Avon Urban District Council. The Hon. Secretary and Treasurer (the Rev. E. H. Steele) being absent through illness, his report upon the Accounts and affairs of the Society was presented by the President. ‘The Report contained a warning about the increasing difficulty of balancing the budget, in face of continually rising costs, and made an appeal to Members for con- tinual efforts to obtain more support for the Society. The Meeting also heard reports prepared by the Hon. Librarian, the Hon. Editor, the Hon. Meetings Secretary, the Hon. Secretary of the Records Branch, the Hon. Secretary of the Natural History Section, and the Curator of the Society’s Museum, Mr. F. K. Annable. The President having been unanimously re-elected, the following Officers were appointed upon the nomination of the Committee: Hon. Secretary and Treasurer, the Rev. E. H. Steele, M.A.; Hon. Librarian, Mr. R. E. Sandell, M.A., F.S.A., F.L.S.; Hon. Meetings Secretary, Mr. J. B. B. Roberts, M.A.; Editorial Board, Mr. Sandell (Hon. Secretary), Mr. Steele, and Mr. F. K. Annable, B.A., F.S.A. To fill vacancies on the Committee, Miss Isobel Smith, Ph.D., F.S.A., and Mr. H. Ross, B.A. After the conclusion of business, an address on the Preservation of Ancient Buildings in Bradford-on-Avon was given by the County Planning Officer for Wiltshire, Mr. Kenneth Cooper. Luncheon was taken at the Swan Hotel, after which the company visited South Wraxall Manor, by kind permission of the Hon. Mrs. Sara Morrison, where the speaker was Mr. Hugh Braun, F.S.A., F.R.I.B.A. After tea, Mr. Ross conducted a visit to the Saxon Church at Bradford-on-Avon, and members of the Bradford-on-Avon Preservation Society led tours of the town. ANNUAL GENERAL MEETING, 1964 THE ANNUAL GENERAL MEETING Of the Society for 1964 was held on 23rd May, 1964, when there was a large attendance of Members. After a morning visit to Avebury Manor, by kind permission of Sir Francis and Lady Knowles, and lunch at the Georgian Restaurant, Marlborough, members and their guests gathered at the Adderley Room, Marlborough College, by the kind invitation of the Master; the Chair was taken by the President, Mr. E. G. H. Kempson. In presenting the Accounts for 1963, the Treasurer was glad to report a surplus of £899, but was compelled to point out that this favourable figure was due to the inclusion of three half-yearly instalments of the Wiltshire County Council grant, and a number of sums earmarked for the forthcoming Guide Catalogue. The true picture was of a close balance between ordinary income and expenditure which, while not unsatisfactory in itself, demanded much future work in improving the Society’s income if necessary expenditure was not to produce an adverse balance in a later year. The encouraging response to a recent appeal for increased subscriptions, together with plans for improving the yield of investments, and schemes for widening membership gave hope that in the long term it would be possible to face with confidence the outlay involved in renewing the heating system at the Society’s premises and in undertaking other capital works. The Accounts having been adopted, reports were made upon other aspects of the Society’s work by the Secretary, the Hon. Librarian, the Hon. Meetings Secretary, the Editorial Board, the Hon. Secretary of the Natural History Section, and the Curator of the Museum. The following Officers were elected to serve for the ensuing year: President, The Most Noble the Marquess of Ailesbury, D.L.; Hon. Librarian, R. E. Sandell, M.A., F.S.A., F.L.S.; Hon. Meetings Secretary: K. H. Rogers. To serve on the Committee: Miss T, E: Vernon, P. J. Fowler, M.A., Dr. T. R. Thomson, F.S.A., F.R.Hist.S. The formal business having been concluded, the President gave his address on “The Vicar’s Library at Marlborough’, which was illustrated with an exhibition of books and documents from the Library. He was warmly thanked, on the proposition of the President-Elect, Lord Ailesbury. Tea was taken in the College Dining Hall, at the invitation of the Master, after which an illustrated lecture on the buildings of Marlborough College was given by Mr. Guy Barton. The interest and enjoyment of the audience was manifest, and was expressed in words by Mr. Hugh Shortt. The company was then allowed to visit the rooms and buildings described, their features being pointed out by members of the College Staff. ask) OFFICERS’ REPORTS READ AT THE ANNUAL GENERAL MEETING, 1964 REPORT OF THE SECRETARY THE SOCIETY’S Statistical record for the year ended 31st December 1963, is that 61 new members were enrolled. Of these, 58 were Ordinary members, 2 were Juniors, and there was 1 Institution. During the same period 52 names were removed from the lists. Thirty of these were resignations, in some cases after a failure to send subscriptions had been pointed out; there were 13 deaths, and g members were removed from the books having paid no subscription for at least two years, and having not answered any letters. Thus the net increase in membership was g only, a number which is quite insufficient to maintain the Society as a going concern. As was pointed out in the Secretary’s Report last year, it is desirable that there should be a continual net increase of substantial propor- tions if the Society’s activities and services are to be maintained in face of rising costs. At the end of 1963 the total membership was 671, made up of 575 Ordinary members, 38 Life members, 5 Juniors, and 53 Institutions. So far this year the trend has been more favourable for, in spite of normal wastage, and the removal of a further batch of defaulters, the total membership on 7th May was 688, and we should like to see it well over the 700 mark before the end of 1964. Death has made sad havoc among old friends. We have recently lost the Patron of the Society, Lady Colum Crichton-Stuart; Mr. James Oram, a most loyal supporter, who was Mayor of Devizes and President of the Society in our Centenary Year; and only in the last week we have heard of the passing of Miss Seth-Smith, for long a familiar figure in the Society’s Library. During the latter part of the year under review, the Committee appointed a sub- Committee to examine ways and means by which the Society’s general position could be strengthened. A very comprehensive report, which included the conversion of the Hon. Secretary and Treasurer into a paid part-time official, was made known to the Society at a Special General Meeting in January. First steps in the task of implementing the Report have already been taken, and include the Appeal to members which was referred to in the Financial Report, and the mounting of an exhibition at the forthcoming Bath and West Show which will be mentioned by the Curator. The enrolment of a number of new members can be traced to the Appeal, and there are others, suggested by members, to whom literature has been sent, but from whom we have not yet heard. For the rest of the year, other steps will be taken on the lines of the sub-Committee Report, so that your part-time Secretary, who is already finding plenty to occupy him, does not anticipate that the Society will have cause to accuse him of spending idle moments. A not inconsiderable amount of time is spent in making enquiries into proposals to develop sites which contain either Scheduled Ancient Monuments, or Listed Buildings. Some of these cases are brought to notice by members, and we wish there were more. Most, however, come in the form of references by the Ministry of Housing and Local Government to the C.B.A., which passes on the information to local Societies. Many of these cases are not such as warrant any action; only too often the building concerned is already so dilapidated that no good case can be made for its preservation, but some are in another category, and we have rapidly fattening files of correspondence concerning them. During the past year we have dealt with a housing estate at Cricklade, which was impinging on the old Town Bank of Saxon times; a fine 18th-century house at Corsham; 200 the Lanhill Long Barrow; and, more recently, the Market Place at Devizes. This short list only mentions the more spectacular cases; there are many more, and all require correspondence and enquiry. Some of them involve one or more personal visits as well. We feel that this is a vital and important piece of work, and we would ask yet again that members should let us know of cases where ancient sites or houses are under threat. Time is the critical factor; it is not enough that we should know when the damage has actually started. The remaining work of the Secretary’s office, or at least that part which is not performed by his alter ego, the Treasurer, has consisted of the slow and sure building up of a sound administrative system, on the lines so admirably laid down by the previous Secretary, Mr. Hankin, with the aim that the business of the Society should be efficiently conducted, and the Librarian and Curator freed to do their important and absorbing work without having to worry about purely business matters as they so often have had to do in the past. 201 REPORT OF THE CURATOR THE HARVEST As our text for 1963 we may be said to have taken that much-used quotation, ‘Ask, and it shall be given unto you.’ Now, as the year is seen in retrospect, we can perhaps claim that our biblical adherence has brought forth fruit in good measure. Members will recall that our appeal made in 1962 to the British Academy resulted in a generous gift of £150 towards the cost of printing the projected Guide Catalogue. Early this year this sum was further augmented by a munificent grant in aid of £250 from the Council for British Archaeology. Coupled with the Society Committee’s own provision, this combined sum will go a long way towards meeting the heavy costs of production. As this report is being written, galley proofs of the Catalogue lie on the Curator’s desk. We can only hope that the final production will amply justify the generosity shown to us. Our efforts towards an appeal for financial help to the Carnegie United Kingdom Trust for the redisplay of our widely-known Iron Age collections were also continued. Early in January, Dr. D. B. Harden and Mr. N. C. Cook inspected the museum, and gave a most excellent report on the progress achieved since their last visit in 1956. A layout design was then drawn up at the museum, and Messrs. Hurst & Co., museum specialist fitters, of Cheltenham, have since submitted estimates of the entire cost of the proposed display, amounting to some £1,750. All this has now resulted in a grant-in-aid of £750 being formally offered to the Society by the Joint Committee of the Carnegie Trust and the Museums Association, towards the construction of the new room. Generous as this offer is, particularly when seen in conjunction with their magnificent gift of £1,500 in 1956 towards our Bronze Age Room, we have somehow to bridge a monetary gap of £1,000 if we are to accept the designs as submitted by Messrs. Hurst & Co. Harassed as always by soaring costs, this is a well-nigh impossible task, and we may have to dras- tically cut our pattern according to the measure of our cloth. Nevertheless, we must continue in the earnest hope of arriving at some satisfactory scheme which will enable us finally to display these notable collections in a worthy setting. THE MUSEUM AND COLLECTIONS One long overdue task, begun this year by our Assistant Curator, Mrs. Mitchell, has been the refurbishing of the Recent History Rooms. Because of an antiquated heating system, we are bedevilled by dust which penetrates into every room, resulting inevitably in dingy cases. All the case interiors in the larger of our two ‘Recent’ rooms have been redecorated, re-arranged, and relabelled, with the addition of some new acquisitions. — The smaller room is due to be completed early in 1964. We are indebted to Mr. Michael May, a Cambridge graduate in anthropology who has given valuable assistance with further arrangement and listing of our reserve Neolithic material. Much still remains to be done, particularly in putting our large assemblages of flint artefacts in order, but Mr. May’s contribution is an excellent start. During the year the Recent Accessions case in the entrance hall of the museum has twice been arranged with new exhibits. Work on exhibition cases in the new Anglo-Saxon and Medieval room has been retarded due to the unfortunate indisposition of our carpenter/electrician, Mr. Cole. Happily he is now much fitter, and we sincerely hope that this room will be finished by the autumn of 1964. Display units have, nevertheless, been constructed on three sides of the room, including one of two large wall cases intended to house distribution maps of the period. The maps, constructed in g-mm. plywood to a scale of 1 inch to a mile, are in preparation, and in fact are almost complete. 202 LABORATORY AND TECHNICAL The Assistant Curator has spent a good deal of time in the treatment and conservation of material in preparation for future reorganization. Almost the whole of our Anglo-Saxon ironwork, with the exception of specimens too fragile for treatment, has been cleaned. Some two dozen iron keys of medieval and later date, and a number of iron tools from the Roman collection have likewise been treated. A few iron objects of Roman date, found by Lt.-Col. W. D. Shaw during excavations in his garden at Wellhead, Westbury, were cleaned at his request. Work has continued with repair and restoration of Iron Age and Roman pottery. Amongst other individual items restored in the laboratory this year are a large storage vessel of medieval date from Mr. P. J. Fowler’s excavations at Fyfield Down, and a Neolithic beaker from barrow excavations on Overton Hill, directed by Dr. Isobel Smith. A plaster facsimile of a Windmill Hill type bowl was also made at the request of the Northampton Museum. We are nearing completion with cleaning and consolidating the wall plaster fragments from the town house excavation at Cunetio in 1962, and further pottery finds from the early well near the west gateway of the town have been washed and examined. The excavation of this well was successfully carried out during the summer. PUBLICATIONS The text and drawings of the new Guide Catalogue of the Neolithic and Bronze Age Collections in Devizes Museum were finally delivered to Messrs. Headley Brothers Ltd., the printers, in December. The close checking of the entire text, and the correlating of references took far longer than was anticipated; it was also necessary in the final stages to check some of our manuscript material with copies preserved at the Society of Antiquaries library in Burlington House, and originally used by Sir Richard Colt Hoare when writing his Ancient Wiltshire. However, as mentioned above, Messrs. Headleys have responded most promptly with galley proofs, and at this rate we may confidently expect publication of the Catalogue in 1964. NATIONAL NATURE WEEK During the week of 17th to 25th May, the Council for Nature organized a National Nature Week to promote an interest in, and a greater understanding of, the countryside, its wild life, and its conservation. Members of the Natural History Section of the Society and museum staff collaborated to arrange an exhibition illustrative of the Wiltshire countryside and its wild life, and in particular, some of the valuable work of conservation being carried out by members. Display stands were constructed in the museum by Mr. Cole, and individual displays were set up by members working with the Curators. We are additionally grateful to the Wiltshire County Librarian who kindly provided an excellent book-stand of natural history literature. For individual displays we have to thank Mr. D. Tucker of the Bristol Avon Water Board, Mrs. Ruth Barnes, Miss W. M. Stevenson and pupils of Devizes Grammar School, Mr. R. S. Barron, Mr. B. Weddell, Mr. J. D. Grose, Mr. Inigo Jones, and by no means least, Miss Beatrice Gillam, whose skilful assistance in all stages of preparation was greatly valued. The exhibition was opened by the Mayor of Devizes, Councillor R. Kemp, and continued throughout the week, including evenings, with considerable success, a number of Wiltshire schools making a special journey to see the displays. LECTURES In collaboration with Bristol University, the Society again held its winter University Extension Lectures in the museum lecture hall. The lecturers this year were, Professor R. J. C. Atkinson (Cardiff University), Mr. D. R. Wilson (Ashmolean Museum, Oxford), 203 and Mrs. M. Aylwin Cotton (President, Berkshire Archaeological Society). The series was again very well attended, and the evident popularity of the lectures encourages us to continue them in future years. We owe much to Mr. Harry Ross for his efforts in their regular organization. The Curator gave a course of ten lectures on ‘Prehistoric Britain’ to a W.E.A. group in Corsham. Numerous lectures and guided tours were also provided for local groups, societies and training colleges, both in the museum, and at archaeological sites within and outside the county. Amongst the parties catered for were English and foreign students from Urchfont Manor; students of Shenstone Training College, near Kidderminster, also spent a study week in Wiltshire during the summer, and the Curator provided lectures on prehistory throughout one day of their tour. VISITORS It has been our pleasure again to receive students and specialists who visited the museum to examine our collections, among them Dr. Glyn Daniel and students from Cambridge University. A party of students from Berlin University led by Professor Kirchner also inspected our collections last September during a tour of the county. It is also gratifying to note that the Bath Academy of Art is making increasing use of the museum, and of the services we can provide for their students. All this is as it should be, for we are most anxious that our collections, both library and museum, should be fully exploited by educational institutions everywhere. Twenty-eight school parties visited the museum during the year, and in almost every case were given guided talks by the Curators. The total number of visiting pupils was 888. It is worth noting that a number of schools from outside Wiltshire are now beginning to make annual visits to the museum. We are pleased to welcome them, and hope very much that they will continue to do so in increasing numbers. The number of visitors to the museum, excluding specialist visits and school groups, is given below, with the figures for last year: 1962 1963 January-March 326 January-March 372 April-June 806 April-June 964. July-September r,120 July-September T1Ot October-December 347 October-December 261 Total 2,599 Total 2,778 LOANS Groups of museum specimens have again been lent out during the year to teachers and lecturers. Amongst other material loans provided were fossil specimens used during a week-end course on the Geology of Wiltshire, held at Urchfont Manor. A selection of prehistoric pottery, and some of our Bronze Age grave groups were lent to Reading Museum for use during a Museums Diploma Course held in September. A bronze brooch of early Iron Age date is now on loan to the Birmingham City Museum. A further group of prehistoric, Roman and medieval pottery and metalwork, originally found in the Pewsey region, was displayed at the Pewsey Local History Exhibition held during September. A pitch pipe of early rgth-century date was lent to the Council for the Care of Churches for their Church Music Exhibition held in London from 15th October to 1st November. SLIDE COLLECTIONS We are grateful again to a number of individuals and museums who have added to our growing slide collections. These now total well over 1,000 in number, and thus 204, constitute a valuable record of the county’s monuments, and its natural history. Let us remind teachers and lecturers that slides are available on short-term loan for educational purposes. FIELDWORK The excavation of a Roman well located beneath the west wall of the walled township of Cunetio (Mildenhall), was completed this summer. There were no spectacular finds, but pottery, including decorated and plain Samian fragments was recovered throughout the interior filling. All the Samian ware has been closely dated to A.p. 50-60, and it is clear that the well constitutes incontrovertible evidence for the earliest so far known occupation at this extremely interesting site. Our most grateful thanks go to Messrs. Norris Thompson, John Mead and Harry Ross for invaluable assistance during the dig. ACKNOWLEDGEMENTS We are particularly indebted this year to Mr. Michael May for his extremely valuable work as a volunteer in assisting with the reorganization of our reserve Neolithic collections. Mr. Charles Webster, of Bryanston School, has worked with us during vaca- tions, and has been immensely helpful in carrying out useful day-to-day tasks. It falls to us again to thank Mrs. Cole for another year’s service to the museum, and Mr. Cole for his continuing skilled assistance in this never-ending task of reorganization. Long may they both continue. ACCESSIONS The Society Committee extends its thanks to all who have made gifts or loans during the past year. These are listed below: GEOLOGY Fossil ammonite covered with calcite crystal formation. Possibly from Corsham. Mrs. C. Phipps. 8/63 NATURAL HISTORY Skin of Leech’s Fork-tailed Petrel (Oceanodroma leucorrhoa leucorrhoa). Found on 22nd December on Keevil Airfield. Col. C. Floyd. 4/63 PREHISTORIC Group of sherds of Iron Age (A2) date. Found on the surface in the garden of a house in Westbury, NGR. 88025118. L. F. T. Sawyer, Esq. 3/63 Barb-and-tang flint arrowhead. Surface find near Wilcot, NGR. 14456065. L. G. Burry, Esq. 11/63 Butt end of polished flint axe of Neolithic date. Found in the garden of 29 Cards Lawns, Devizes. Mrs. Goodman. 21/63 Shale pendant, shale bead, flint and bone material, sherds of Windmill Hill, Peterborough, and Beaker-type wares. From excavations directed by Dr. Isobel Smith on behalf of the Ministry of Works at a group of barrows on Overton Hill. English Farms Ltd. 22/63 Bronze fragments, bronze brooch, sherds of Neolithic and Bronze Age pottery, flint and bone artefacts. Found during excavations directed by Mrs. F. de Mallet Vatcher on behalf of the Ministry of Works at a group of barrows on Lamb Down, Codford St. Mary. J. M. Stratton, Esq. 23/63 Group of sherds of early to late Iron Age date. Found during surface digging at Erlestoke Detention Centre. H.M. Inspector of Prisons. 25/63 205 ROMAN Sherds of coarse pottery. Surface find in the garden of a house at Westbury. NGR. 88025118. L. F. T. Sawyer, Esq. 3/63 Sherds of coarse pottery. Surface finds in a ploughed field at Northwood Farm, Colerne. NGR. 87812734. E. A. Shore, Esq. 6/63 Samian fragments and iron nails. Surface finds in a ploughed field at Northwood Farm, Colerne. NGR. 87812734. E. A. Shore, Esq. 9/63 Coarse pottery and quern fragments. Found while erecting a fence near a kite-shaped enclosure on the western side of Yarnbury Castle. Col. B. G. Ivary. 10/63 Bronze ‘one-piece’ brooch. ist century A.D. Avebury parish. Miss E. A. Kemm. 14/63 Box flue-tile fragment. From the garden of 29 Cards Lawns, Devizes. Mrs. Goodman. 21/63 Bronze and iron fragments, originally part of a vessel. From excavations directed by Dr. Isobel Smith on behalf of the Ministry of Works, at a group of barrows on Overton Hill. English Farms Ltd. 22/63 Samian and coarse pottery, glass fragments, bronze coin of Valentinian I (A.D. 364-7) (Arles). From excavations directed by Mrs. F. de Mallet Vatcher on behalf of the Ministry of Works, at a group of barrows on Lamb Down, Codford St. Mary. J. M. Stratton, Esq. 23/63 Ae. coin of Constantinian Period. Obv. Helmeted bust of Rome 1. VRBS ROMA. Rev. Victory on prow of galley. Found in Nursteed Road, Devizes. R. G. Hatter, Esq. 24/63 Samian and coarse pottery. Recovered during surface digging at Erlestoke Detention Centre. H.M. Inspector of Prisons. 25/63 MEDIEVAL Pottery fragments. Found during surface digging at Erlestoke Detention Centre. H.M. Inspector of Prisons. 25/63 RECENT Silver bodkin, probably 17th century. Discovered built into a wall of Upham House. L. R. Chivers, Esq. 1/63 Mahogany map, cut into pieces as a jigsaw, of E. Bowen’s reduced (1762) map of Wiltshire of 1755. Miss E. S. D. Merriman. 2/63 China ash-tray, bearing a view of Stonehenge, and inscribed, ‘Stonehenge, a present from Salisbury’. Late 19th century. Mrs. B. Yeatman Biggs. 5/63 Leather travelling trunk, lined inside with pages from a stage-coach schedule, and a page of the Wiltshire Times, dated 1896. Mrs. E. E. Morton. 7/63 * Miniature hand cannon, or starting gun. 19th century. W. Hibberd, Esq. 13/63 A group of Victorian fans, an iron clucket bell and a box of assorted Valentines. All 19th century. A linen(?) sampler embroidered ‘Martha Davis finishedh er work 27th August 1752’. Miss E. A. Kemm. 15-19/63 Toothed wheel, used to facilitate drawing parallel lines on parchment. Used at the office of Wakeman and Bleeck, Solicitors, Warminster. 19th century. K. Rogers, Esq. 20/63 Stylographic pen, dated 1878, in its original black leather box inscribed Waterlow & Sons, London. Miss O. Gillman. 26/63 Model of the West Kennet chambered long barrow. Made by Professor R. J. C. Atkinson, Cardiff University. Granada T.V. Studios. 12/63 206 NATURAL HISTORY SECTION ROAD CASUALTIES AMONG MAMMALS, REPTILES AND AMPHIBIANS IN WILTSHIRE by A. A. DUNTHORN and F. P. ERRINGTON THIS IS A REPORT on the mammals, reptiles, and amphibians found dead, presumed killed by traffic, along certain roads in Wiltshire. It falls into two parts: one, from 21st December 1956 to 31st March 1960, when the sampling route was the road from Salisbury to Porton via Old Castle Inn, R.A.F. Station, Old Sarum, and West Gomeldon; and two, from 1st May 1960 to 30th April 1961, when the sampling route was the road from Salisbury to Porton via St. Thomas’ Bridge, Winterbourne Earls, Winterbourne Gunner, and East Gomeldon. The distance in each case was approximately 6 miles. The first route lay at about the 250-ft. contour line for the first 5 miles. It was, for the most part, bounded by wire fencing or clipped thorn hedges. Near the commence- ment it passed for about ? mile by the Salisbury Isolation Hospital and through the R.A.F. Station, Old Sarum. Here the roadside characteristics included hangars, messes, and married quarters with gardens. Trees fifteen to twenty feet high were present. The last mile dropped to the 200-ft. contour, and more or less followed the river Bourne. Here the hedgerows were high, mixed, and contained trees. Roadside ditches were present at West Gomeldon, and there was an area of woodland at Target’s Corner, Porton. The second route lay at about the 200-ft. contour, and more or less kept with the river Bourne for the first 5 miles, before climbing East Gomeldon Hill to the 250-ft. contour line, and dropping down to Porton to complete the 6 miles. Most of this route lay through villages, with old houses and buildings close to the roadside, and new dwellings set well back from the road. Some fields and woodland also occurred. These journeys were made morning and evening, on average nine days a fortnight. 19560-1960 During this period the journeys were made by bicycle. No animals were found dead during the last days of 1956. Table I shows the species and numbers found dead for each of the years 1957, 1958, and 1959, and for the first quarter of 1g60. It does not necessarily reflect the relative abundance of each species along this route. For many, the numbers are too few for any useful deductions to be made, but for others some generalizations may be offered. The numbers of Brown Hare, Common Shrew, and Hedgehog found dead each year were fairly constant, whilst the numbers of Brown Rat showed an annual increase. The first Rabbit was killed during September 1958, the only one that year, but fifteen were killed during 1959. The two Grey Squirrels were found dead during the autumnal migration period, when many others were seen in the district. The four species most frequently found dead were Brown Rat, Hedgehog, Brown Hare, and Common Shrew, in that order. This list contrasts with that of Daviest who found, from counts made in Hampshire between March 1952 and March 1954, that the four most frequent species were Rabbit, Brown Rat, Hedgehog, and Grey Squirrel. 207 TABLe [ Vertebrates found Dead, 1957-60 1960 Species 1957 1958 1959 (first Totals Quarter) Stoat .. 44 ae 3 Be — I — — I Weasel = — —= 2 Farm Cat I — — I 2 Brown Hare .. 9 Il 16 2 38 Rabbit ~ Be oe “A — I 15 3 1g Brown Rat .. oi os 28 58 78 16 180 Short-tailed Field Vole 9 5 2 I 17 House Mouse I I I — 3 Long-tailed Field Mouse 6 = I = 7 Water Vole — — I = I Grey Squirrel : ae - I I = = 2 Common Shrew a ae ss 10 9 9 — 28 Hedgehog... ae 7 ie 43 38 42 2 125 Moleo: . an me a — 2 7 = 9 Pipistrelle Bat —— — I = I Grass Snake .. 3 — I — 4 Slow Worm a oe hat I I I — 3 Common Frog ae Ee ke II 23 6 — 40 Common Toad ae = 13 84 19 7 123 Unidentified small mammals ne 3 2 4. — 9 Birds (all species). . be 6 415 434. 430 40 5319 He forecast that the advent of myxomatosis would change this order. The prominence of the Brown Rat and Hedgehog in both these regions, and in that covered by Table II is, in some respects, a measure of their successfulness, whilst the substitution of the Grey Squirrel for the Brown Hare as the next most frequent species reflects the nature of the regions under observation. That of the authors consisted mostly of large open fields, whilst that of Davies lay in the New Forest between Southampton and Bournemouth. Tas_e IT Vertebrates found Dead, 1960-61 1st May 1960 to pews: goth April 1961 Weasel .. ne ae ae I Brown Hare ; ee a I Rabbit .. BA x ve I Brown Rat ae 18 Short-tailed Field Vole. ra I Hedgehog : ae a8 II Common Frog... oF nih I Common Toad. J _ I Unidentified email amine eh I Birds (all species) = a g2 208 The seasonal variations in the numbers of the Brown Hare, Brown Rat, and Hedgehog found dead during three calendar years are shown in Fic. 1. About one-third of the Hares were killed during April, but deaths occurred every month except January. ‘The Brown Rat was killed every month, with a marked increase of casualties towards the end of the year. The histogram of the Hedgehog casualties agrees with the known habits of this animal, i.e. it is active during the summer months, when most deaths occurred, and it is a partial hibernator in the winter months, when few or no deaths occurred. This seasonal distribution agrees with that plotted by Davies for Hampshire, and with that of Brockie? for the North Island, New Zealand. The reaction of this animal to danger by rolling into a ball and waiting, makes it particularly vulnerable to traffic. There were relatively more Common Frogs and many more Common Toads found dead during 1958 than in the other two complete years. This could have been the result of a previous good breeding season. Their seasonal distribution is shown (Fic. 1). The 10 BROWN HARE BROWN RAT HEDGEHOG (ae) c-¢ w (a) Q ES COMMON FROG 9 20 a uJ (a0) 3 => ae COMMON TOAD HISTOGRAMS SHOWING DISTRIBUTION OF FIVE VERTEBRATES FOUND DEAD ON ROADS OVER THREE YEARS 1957-1959. Fic. 1 209 September peak shown for the Toad comes from a concentration of seventeen killed during September 1958, and the phenomenon is not explained. The main peaks for Toad and Frog mortality occur in spring and autumn respectively, when these creatures are actively seeking water. The Toad migrates to water in the spring to breed, dispersing after this function. Those killed during other months are indicative of activity during rainy periods. The Toad’s habit of moving slowly and resting for long periods makes it as vulnerable to traffic as the Hedgehog, and it appears to have been a common road casualty in the days of the horse and cart.3 The frog migrates to water at the onset of the colder weather when it hibernates in the bottom mud, or in holes, or under submerged logs. It emerges during the spring, when the temperature is suitable, to breed and to disperse. The small concentration in April might be due to breeding adults leaving the water. Most of the dead Frogs were found more than a mile from the nearest known surface water. The total number of birds of all species killed during each complete year was remark- ably constant. These have been the subject of part of a separate paper.‘ 1960-1961 Journeys during this period were made by motor-scooter, and the route lay through villages of the Bourne Valley. Table II shows the species and numbers found. These numbers are well below those for any one year recorded from the roads which lie between open fields. ‘The Brown Rat still has pride of place with a total of eighteen, 50 per cent. of all casualties, and the Hedgehog is a good second with eleven. The records of the bird casualties for this period were collected as part of the British Trust for Ornithology Road Death Enquiry (1960-61) and will be embodied in that report when it is published. ACKNOWLEDGEMENT The authors are pleased to acknowledge the criticism and helpful suggestions of Miss B. Gillam. 1 J. L. Davies, A Hedgehog Road Mortality 3 W. H. Hudson, The Book of a Naturalist. Index, Proc. Zool. Soc. Lond., cxxvitt (1957), 606-8. London. 2 R. Brockie, Road Mortality of the Hedgehog 4 A. A. Dunthorn and F. P. Errington. In the (Evxinaceus europaeus L.) in New Zealand, Proc. Zool. press. Soc. Lond., cxxxtv (1960), 505-8. 210 THE WEATHER OF 1963 by R. A. U. JENNINGS 1962 WAS a bad year: the ‘bad marks’ of 1963 in the summary below are 60 per cent. in excess of its predecessor’s. Except for us arithmomaniacs, who revel in eccentric meteorological statistics there is no redeeming feature apart from the late arrival of the autumn frosts. The incidence of these was capricious, but in several parts of the county dahlias, gladioli and comfry were still in bloom in mid-November; it is unusual, too, to have skating at both ends of a calendar year. We shall always remember the long and intense cold of the earlier months. 1881’s snow was not beaten, but by every other test last winter was the coldest for a century, though the fuel shortage and consequent electric cuts made 1947 more trying. The mean temperature of January at Marlborough was 23°7° F.: the previous coldest month was February 1895 with 27°3° F. As the spring flowers took heart they found strange bedfellows: in early April in the west of the county there could be seen in simultaneous bloom, and within 100 yds., snowdrop, aconite, crocus, celandine, dog-tooth violet, lent lily, daffodil and (with a little proleptic dishonesty) tulip. The summer months were all subnormal and August was really bad for the fourth year running. Television has brought weather maps and their interpretation to a far wider circle of Englishmen, and those who never used to read such newspapers as print these maps can now regularly see them on the box. The competent men who explain them deserve the highest praise: they do not always get it, because we are so ready to boast our wit when they make their rare mistakes. Yet our local forecasts are difficult because Wiltshire’s geographical position makes us a sort of meteorological 18th-century Poland with rapidly shifting alliances. Sometimes we are ‘S.E.’ in the forecasts: at others ‘S.W.’ or ‘South Midlands’. Except in an anticyclonic set-up, when the least skilled of us can do as well as the professionals, we have to strike a balance between the three forecasting areas. If we choose the least promising of them we insure ourselves against disappointment. There has been a steady decline in mean annual temperatures since 1959. In this summary R means rainfall; T, temperature; S, sunshine. -+- means excess; — means deficiency; 0 means nearly normal. R. ae S. January The coldest in the last 100 years. February —-—- —— O Nearly as bad. March ++ fe) — Wet after the first few days. April + oO — Normal but dull. May fe) a ) Undistinguished. June +--+ — Cold and wet, but with one hot week. July — — — Dry and dull; briefly hot at the end. August a _ —— Detestable. September ) — — Normal. October —— ) — Mild, dull and dry. November ++ ae — Mild and wet. December — — — Mild: very cold: mild. 1963 — —— —— _ Very depressing. 211 WILTSHIRE BIRD NOTES FOR 1963 RECORDERS: Ruth G. Barnes, M.B.o.u., Geoffrey L. Boyle, Dr. E. A. R. Ennion, M.A., M.B.o.u., C. J. Bridgman, M.B.o.U. THE MAIN INTEREST in the 1963 notes is the record of losses in many species from the exceptionally severe winter. In spite of this and the very late spring, the Chiffchaff’s arrival was early. Of rarities, the Rose-coloured Starlings, seen in March and July, have not been reported in the county since 1869, but the possibility of their being escaped birds cannot be entirely eliminated. A Knot seen in February, and presumably the same bird as was found dead in March, is only our sixth record. A Grey Phalarope was killed by a car. Unusually large numbers of Goosander were recorded on both the Rivers Avon. Whooper Swans, rare in Wiltshire, were seen as well as the less uncommon Bewick’s Swan. It was a year of irruption of Crossbills and a number were recorded. The Collared Turtle Dove, first noted as breeding in the county in 1962, seems to have become well established already. The number of birds ringed increases each year and returns are quoted from British Birds and from those notified by members. ‘The homing capacities of small migrants from overseas are well illustrated by two cases: three out of four Sedge Warblers ringed at Oakhill in 1962 were re-trapped there in 1963; and two Reed Warblers from the same nest at Coate in 1961 were found as a mated pair there this year. Contributors : R. Abbott .. es R.A. R. Dunthorn R.D. D. A. W. Alexander D.A.W.A. E. A. R. Ennion E.A.R.E. Sir Christopher Andrews . . C.B.A. F. P. Errington PPE: Miss M. F. Archer M.F.A. R. C. Faulkner R.G.F. C. S. Bailey C.S.B. Mrs. Forbes E.V.F. Mrs. Barnes R.G.B. Miss Forbes K.G.F. Mrs. Botting E.B. J. Ford a fal J. Bourne J.B. G. H. Forster G.HLF. G. L. Boyle G.L.B. Mrs. Forster G.M.F. K. C. Briand K.C.B. D. W. Free D:W.E-s J. S. Bridgman J.S.B. D. E. Fry D.E.F. N. M. D. Brown N.M.D.R. Mrs. Gandy 1.C.G. Miss A. Burnet A.B. Miss B. Gillam B.G. E. J. M. Buxton E.J.M.B. A. J. Gush .. as A.J.G. W. A. Cadman W.A.C. G. H. Hemmings .. G.H.H. D. Campbell i D.C. A. J. Horner A.J.H. Miss Joan Campbell JG; F. J. Hulbert F.J.H: A. A. Carpenter A.A.C. Sir Julian Huxley .. J.B: H. J. Clase .. an FL... C: Mrs. Johnston-Shearer G.J.S. Dr. R. C. C. Clay .. R.C.C.C. E. H. Jelly .. E.H.J. E. Cohen ae E.C. Dr. E. L. Jones... E.L.J. Major W. M. Congreve .. W.M.C. Miss C. V. Kendall a ae C.V.K. Mrs. Crichton-Maitland .. P.C.M. Wing Commander G. Knocker .. G.M.K. C. A. Cutforth C.A.C. T. Lanter. ay: ant T.L: Dr. W. Drummond W.D. Mrs. Lawson V.G.L. A. A. Dunthorn A-A-D: Julian Lawson J.R.L. 22) R. F. Lee R.F.L H., J. F. Smith FL PS: R. J. Lewis R.J.L R. J. Spencer : Reis: Miss M. K. Luckham M.K.L Miss J. M. Stainton J.M.S. J. E. Major J-E.M Miss E. Stewart .. E.S. Douglas Mann er D.M Miss M. J. Stokes .. M.J.S. Marlborough College Natural Mrs. Stopford-Beale JS: B. History Society .. M.C. B. M. Stratton B.M.S. J. G. Mavrogordato J-G.M. A. W. Sudbury A.W.S. Mrs. Morgan VM. P. W. Tanner P.W.T. Mrs. Morris P.W.M. Cc. N. Tilley C.N.T. R. 8. Newall R.S.N. M. E. Tyte M.E.T. Mrs. Newton-Dunn DEN=D. R. Upton R.U. J. G. C. Oliver J:€:C.O: C. R. Verner C.R.V. E. G. Parsons E.G.P. G. L. Webber G.L.W. R. Penney .. ie RP: G. Weyman G.W. Brigadier J. R. I. Platt JsResR: Ralph Whitlock R.W. Hon. R. O. Pleydell-Bouverie REOFEER: Miss D. Willams .. D.W. R. H. Poulding Ree: Rev. H. R. Williamson H.R.W. Countess of Radnor TERS R. W. Wood : R.W.W. Cyril Rice .. C.R. Lady Young R.M.LY. Peter Roberts PER& J.C. Rolls .. J.G.R. Abbreviations used in text: Wing Commander D.SeafieldGrant D.S.G. British Birds Journal B.B. Brigadier E. Searight = . E.E.G.L.S. Gravel pit GP. Mrs. Seccombe Hett wa OASWak Sewage farm S.F. Arnold Smith A.S. Reservoir Res. 4. RED-THROATED DIVER. One found dead on snow at Stokke, 2oth Jan. (C.N.T.). One in winter plumage on River Avon near Dauntsey, 17th-24th Feb. (R.C.F., J.C.R.). 5. GREAT CRESTED GREBE. I'wo pairs, one of the birds on a nest, at Stourhead, 18th Apr. (E.C.). Again low water prevented breeding at Coate Water (G.L.W.). Two pairs bred and probably reared two and one young respectively at Longleat (R.H.P.). Pair head shaking on Bowood Lake, 12th Apr. (G.L.B.). A pair, two young on the back of one of the adults, on Station Ponds, Westbury, 18th June (T.L., R.A., P.W.T.). A pair with four young, of which two were reared, at Corsham Lake, June-July (J.C.R.). A pair seen at Braydon Pond with one young bird, 1st Aug. (R.G.B.). Q. LITTLE GREBE. A few found dead in cold weather (M.C.). Decrease noted near Castle Combe (R.F.L.). Nine on Wroughton Res., 17th Nov. (G.L.W.). 28. CORMORANT. Several, some perching in trees, at Fonthill Lake, 8th Mar. (B.M.S.). 29. SHAG. One flew, at c. 200 ft., over Swindon, 12th Nov. (G.L.W.). 30. HERON. Six found dead in Marlborough area (M.C.). One dying at Seagry, 25th Jan. (R.G.B.). Two found dead at Longford ‘skin and bone only’ (I.R.). The numbers of occupied nests in heronries were as follows: Great Bradford Wood, 16 (R.J.S.) ; Bowood, 13 (G.L.B.); Leigh, 1 (E.J.M.B.); The Warren, Savernake, 6 (C.N.T.) ; Boynton, 5 (J.R.1.P.); Hurdcott, nil, Trafalgar, 1 (A.J.H., D.E.F.); Britford, 1 (I.R.). 38. BITTERN. One found in snow near Coombe Bissett, 29th Mar., had been dead some time (D.N.D.). One brought into a garden at Seagry, 7th Feb., presumably by fox (R.G.B.). One seen in Newcourt meadows, near Salisbury Avon, 22nd and 24th Dec. (R.O.P.B.). One seen at Corsham Lake, 26th Dec. and later (J.C.R.). 45. MALLARD. Maximum numbers: Axford, 500 on 20th Nov. (M.C.); Braydon Pond, ¢. 350 on 15th Dec., the highest number noted here by the observer in 25 years (R.G.B.); Corsham Lake, ¢. 215 on roth Oct. (J.C.R.); ¢. 100 on 5th Dec. (R.C.F.); Fonthill Lake, 100 plus on 18th Dec. (N.M.D.B.); Longleat, c. 100 on 25th Dec. (R.H.P.); Ramsbury Manor, c. 600 on 13th Oct. (V.C.L., J.R.L.); Shearwater, ¢. 165 on 7th Jan. and ¢. 100 on 19th Dec. (N.M.D.B.); Tockenham Res., c. 200 on 25th Nov. (R.W.W.). One duck with 17 ducklings (mixed broods?) at Coate on 12th May (G.L.W.). 213 46. TEAL. A few found dead in cold weather, Marlborough area (M.C.). Maximum numbers: Braydon Pond, c. 30 on 8th Nov. (D.A.W.A.); Coate Water, c. 200 on 24th Nov. (G.L.W.; Corsham Lake, 18 on 28th Nov., Lacock G.P., 25 on 25th Dec., and River Avon near Dauntsey, 17 on 24th Feb. (J.C.R.). A pair present at Coate Water all summer, but no proof of breeding (G.L.W.). First seen Shalbourne Mill, 19th July (E.A.R.E.). 47. GARGANEY. A pair at Lacock G.P., 26th Mar.-5th Apr. (J.C.R., G.W.H.). A male on the 21st and a pair 22nd-26th Apr. on Coate Water (E.L.J., H.R.W., G.L.W.). A first winter male at Coate Water, 6th Aug. (G.L.W.). 49. GADWALL. At least one amongst tufted duck at Chilton Foliat, 9th Feb. (B.G.). From one to four at Chilton Foliat in February (V.C.L., J.R.L.). An adult male ringed 24.4.61 at Abberton, shot 25.1.62 at Sutton Veney, Wiltshire, 135 miles west-south-west, cf. B.B., ‘LV1, 491. 50. WIGEON. Maximum numbers: Braydon Pond, 11 on 23rd Nov. (G.L.W.); Britford meadows, 100 on 1st Mar., and Clarendon Lake, 150 on 5th Dec. (D.E.F., A.J.H.); Corsham Lake, 6 on 12th Dec., and Lacock G.P., 6 on 2gth Dec. (J.C.R.); River Avon near Dauntsey, 34 on 6th Jan. (R.C.F., J.C.R.); River Kennet near Chilton Foliat, 12 on oth Feb. (J.R.L., V.C.L.). A pair seen on Coate Water, 12th Apr. (G.L.W.). 52. PINTAIL. One drake and g ducks seen at Wilton Water on 16th Nov. (M.C.). ‘Two drakes, with a possibility of ducks among 400 other birds, on Braydon Pond, 15th Dec. (R.G.B.). Seven at Corsham Lake, 15th Dec. (R.C.F., G.W.H.). A female and first winter male at Coate Water, 28th Dec. (G.L.W.). Two males amongst mallard on Ramsbury Manor Lake, 28th Dec. (V.C.L., J.R.L.). 53. SHOVELLER. At Stratford Tony, 1 male, 17th Jan. (C.H.A.). At Maiden Bradley, 1 male, 21st Jan. (J.C.C.O.). At Longford Castle, several 14th and 15th Jan. (R.W.). Two ducks, 27th Jan. (M.K.L.). At Chilton Foliat, 3 on 16th, 23rd and 24th Feb. (J.R.L.). At Corsham Lake, 2 drakes and 2 ducks on 4th Apr. (G.W.H.), one duck 1st Oct. (J.C.R.). 56. TUFTED DucK. Maximum numbers: Chilton Foliat, 145 on 16th Feb. and Wilton Water, 140 on 24th Feb. (V.C.L., J.R.L.); Braydon Pond, 20 on 15th Dec. (R.G.B.); Corsham Lake, 9 on 12th Dec. and River Avon near Dauntsey, 14 on ist Jan. (J.C.R., R.C.F.); Clarendon Lake, 50 on 28th Mar. (D.E.F., A.J.H.); Fonthill, c. 50 on 27th Feb. (B.M.S.), and 50 plus on 18th Dec. (N.M.D.B.); Longleat, c. 71 on 25th Dec. (R.H.P.); Longford, 200 on 16th Feb. (M.K.L.); Maiden Bradley, up to 12 throughout year (J.C.C.O.); Ramsbury Manor Lake, 42 on 17th Dec. (M.K.L.); Shearwater, 25 on 7th Jan., 20 plus on 13th Dec. (N.M.D.B.) and Stourhead, 7 each male and female on 18th Apr. (E.C.). A pair with 6 ducklings at Wilton Water on 28th July (J.C.R., R.C.F.), and a brood of 6 ducklings on the canal at Oakhill during summer (D.A.W.A.). MANDARIN DUCK. One on River Kennet, Marlborough, on 27th Jan. (D.W.F.). A male at Shearwater, 7th Jan. (N.M.D.B.). 57. POCHARD. Maximum numbers: Braydon Pond, ¢. 140 on 5th and 8th Nov. (D.A.W.A.) ; Chilton Foliat, 30 on 24th Feb. and Wilton Water, 40 on 16th Feb. (V.C.L., J.R.L.); Corsham Lake, 1o on 7th Feb. (J.C.R.); River Avon near Dauntsey, 18 on ist Jan. (J.C.R., R.C.F.) ; Longleat, 12 on 20th Mar. (R.H.P.); Longford, 110 on 16th Feb. (M.K.L.); Shearwater, 15 plus on 19th Dec. (N.M.D.B.); Fonthill, ¢c. 50 on 27th Feb. (B.M.S.), 70 plus on 18th Dec. (N.M.D.B.); Tockenham Res., 4 on 25th Nov. (R.W.W.). 60. GOLDEN-EYE. A female on Ramsbury Lake and nearby stream on 1oth and 12th Jan. and 17th Mar., 3 females on 3rd Feb. and 2 on 24th Feb. at Chilton Foliat (V.C.L., J-R.L.). A female on Wilton Water, 16th, 23rd and 24th Feb. (V.C.L., J.R.L., M.C.). A male and female on Braydon Pond, 3rd Nov. (R.G.B.). A drake at Longford, 30th Dec. (R.O.P.B.). A female or immature bird on Corsham Lake, 5th-19th Dec. (G.W.H., Jj:G-R.). 64. COMMON scoTOR. A female was with Goosanders on River Avon near Dauntsey, 6th Jan. (R.C.F., J.C.R.). One in a very weak condition, alongside the West Lavington- Erlestoke road, was caught, revived and freed, 18th Jan. (J.C.). 214 69. RED-BREASTED MERGANSER. One, in company of other ducks, at Tockenham Res., 25th Nov. (R.W.W.). 70. GOOSANDER. Twenty-three on River Avon at Longford, 13th Jan. (I.R., R-@:P-5:); 13 there 14th-15th Jan., one, oth Feb., and’ 3-5, 16th Feb. (R.W., D.E.F., A.J.H., M.K.L.). Twenty-six on River Avon (Bristol) near Dauntsey, 6th Jan. and 3 on 23rd Feb. (J.C.R., R.C.F.). A female on stream near Ramsbury, 12th Jan. (V.C.L.). A dead female at Longford, 13th Apr. (R.O.P.B.). 71. SMEW. One and two females seen on River Avon near Dauntsey, 6th Jan. and 23rd-24th Feb. respectively (R.C.F., J.C.R.). One male and two females on lake at Longleat, roth Mar. (R.H.P.). 73. SHELDUCK. One bird on Fonthill Lake, 26th Feb. to 7th Mar. Did not join other species (B.M.S.). Two or three at Wilton Water in Feb. (M.C., V.C.L., J.R.L.), and one in same area 15th Dec. (M.C.). At Corsham Lake, two on etst Nov., one on 2ondeDcc:, (J-C.R.). GREY GEESE. Thirty-seven heading south-east over Bromham, 8th Jan. (J.B.). Collingbourne, c. 100 overflying, 4th Jan. (A.W.S.). Seven landed in field near Odstock Hospital, 26th Jan. (M.K.L.). Twelve flying east-north-east near Melksham, 13th Feb. (R.J.S.). Sixty flying south-west at Marlborough during cold weather (M.C.). Twelve flew south-east over Cole Park, 4th Mar. (E.J.M.B.). Seen flying over Amesbury, 5th Jan. (D.W.) and Idmiston, 8th Jan. (C.M.F.). 75. GREY LAG GOOsE. Five on 17th, 20th, 23rd and 24th Feb., 4 on 3rd Mar., and 1 on roth, 17th, 23rd, 31st Mar. and goth Apr. on River Avon near Dauntsey (J.C.R., ReGon:): 76. WHITE-FRONTED GOOSE. Four in water meadows at Stratford Tony on 17th Jan. (C.H.A.). Flocks flying over Salisbury and Boscombe Down on 24th, 25th and 26th Dec. (G.H.F.). Sixty flying west over Boscombe Down airfield, 24th Dec. (A.J.H.). Fifty flew south over Pitton on 3rd Jan. (R.W.). Six flew over Swindon, 4th Mar. (G.L.W.). 82. CANADA GOOSE. Twenty-five at Lower Woodford, 25th Feb. (R.S.N.). One at Coate Water, 16th and 17th Mar. and probably the same bird at Walcot, 27th Mar. (G.L.W.). 84. MUTE SWAN. Badly affected by cold weather. One ringed at Coate, 27.3.62, found dying on Wanborough Down, 19.1.63. One ringed at Blunsdon found dead in Swindon, 9th Feb. (G.L.W.). 85. WHOOPER SWAN. Three on River Wylye at Upton, roth Feb. (Brigadier J. J. Kingstone per J.R.I.P.). Three adults and 2 immatures on River Wylye between Boynton and Sherrington, 17th Feb. (E.S., K.G.F.). A first winter bird at Britford, 1st-17th Mar., stillgpresent, 21st Apr. (A:J.H., D:E.F., G.H.F.). 86. BEWICK swAN. Three at Longford, 21st Jan. (I.R., H.J.C.). Six, feeding on water meadows at Britford, 1st Mar. (D.E.F., A.J.H., G.H.F.). Eight adults seen at Steeple Langford, 16th Dec. (N.M.D.B.). QI. BUZZARD. I'wo pairs reared young in the Maiden Bradley area. Up to 7 seen soaring together (J.C.C.O.). Pair bred near Semley with possibly a second pair breeding 2 miles away (J.E.M.). Pair seen at Fonthill, Bishops, 30th Mar. (F.J.H.). A pair seen, on several occasions in spring at Biddestone (R.F.L.). Nineteen records of single birds received. 93. SPARROW HAWK. Two nests on Salisbury Plain, only one successful, the other abandoned at time when the eggs should have hatched (J.G.M.). Nests, one with 2, another with 3 eggs, in West Woods and Belling Copse respectively in early May. A second nest in West Woods, but no history (M.C.). Pair of young at Chicksgrove, 28th July (G.H.F.). Many sightings throughout the year at Maiden Bradley (J.C.C.O.). Single birds seen at Bratton, 6th Jan. (E.E.G.L.S.); Trowbridge, 27th Jan. (A.W.S.); North Bradley, 27th Mar. (G.L.B.); Pitton, Apr. and Sept., Longford, Sept. (R.W.); Longleat, Apr., May, Sept., Oct. and Dec. (N.M.D.B.); Cole Park, Jan., Feb., Aug. and Sept. (E.J.M.B.); Everleigh, 17th Dec. (A.W.S.). East Knoyle (B.M.S., J.E.M.); Semley 215 (J.E.M.); Devizes, 26th May, Corsham, 29th Dec. (B.G.); Swindon, Rodbourne S.F. and Coate Water (G.L.W.). g5. KITE. One flying south-westerly over Pitton, 23rd Mar. (R.W..). g8. HONEY BUZZARD. A bird soaring near ‘Tilshead on 16th May was almost certainly of this species (J.G.M.). Q9. MARSH HARRIER. One, at Wylye Down, 23rd Nov. (E.G.P.). RINGTAIL HARRIER. One, in flight, between Shrewton and Chitterne, rath Oct. (J.G.M.). Possible Hen Harrier at Barton, Marlborough, 21st Oct. and also Widdington, end Dec. (C.A.C.). 104. HOBBY. First seen, one, 27th Apr. (G.H.F.); 29th Apr. (E.B.); 13th May (M.C.). J.G.M. reported 4 nests in 3 of which 2 young were reared. An unsuccessful nest found (G.W.). Two nests found, one definitely robbed (G.L.W.). An adult female seen, 22nd Aug. (E.A.R.E.). Last seen, 1, 20th Sept. (N.M.D.B.); 1, 22nd Sept. and 1, Oct. \(Dis:Gs): 105. PEREGRINE. One seen at Pitton, 16th Sept. (R.W.). One immature at Delling Copse, 12th Oct. (M.C.). An adult female flushed a covey of partridges near Marlborough, roth Oct. (J.G.M.). 107. MERLIN. A male at Granham Hill, 23rd Feb. (M.C.). A male at North Bradley, 28th Feb. (G.S.). One, Britford, roth Mar. (G.M.F., G.H.F.). One at Tilshead chattered defiance at a Saker Falcon on exercise, 11th Oct. (J.G.M.). A male at Colerne, 5th Nov. (G.L.B.). 110. KESTREL. Fewer in Warminster area (N.M.D.B.). Commoner than usual in Castle Combe-Yatton Keynell areas. Seen almost daily at Chippenham (R.F.L.). Several nests found, some with more addled eggs than usual, i.e. 3 out of 5 (M.C.). A pair bred in Coate area (G.L.W.). One nest found with 3 young (J.G.M.). ‘Two pairs reared young on downland near Alvediston, using an old crow’s nest (J.E.M.). Single birds seen in many areas in northern half of county (B.G.). In hard weather, one attacked birds feeding on a window table at Tockenham, 2ist Jan. (R.G.B.). An adult female and a young male together on Lyneham airfield, 18th Nov. (R.W.W.). 115. RED-LEGGED PARTRIDGE. Noted on Hack Pen Hill (J.C.R.); Morgan’s Hill (B.G.); and Yatton Keynell (R.F.L.). Two coveys of 15 and 20 birds near Ramsbury in mid January (M.C.). A covey near Berwick St. John, rath Aug. (J.E.M.). 116. PARTRIDGE. Birds observed feeding close to houses during the very cold spell (G_L:W..,. 1.C.G.,. R.F.L.). 117. QUAIL. First noted at Overton Down and Morgan’s Hill, gth June, also a single bird recorded at Liddington Hill and 3 at Tan Hill, 15th June, and one at Knap Hill, 13th and 14th July (M.C.). One calling in grass at Urchfont, 5th July (R.J.S.). One calling in barley near Aldbourne, 22nd July (G.L.W.). One, Stapleford area, 12th July (M.J.S., A.B.). ¢. 4 seen at Etchilhampton Hill, during harvesting, 27th Sept. (D.M.). BOBWHITE QuAIL. This American species was recently introduced on some estates and one was caught at Idmiston, 4th May. It can be distinguished by its white head with black eyestripe and black streak over the head (G.H.F.). 120. WATER RAIL. Sixteen sight records, widely distributed, of which 4 were in January, 2 in February, 1 in March, 2 in September, 1 in October, 1 in November and 5 in December (J.B., A.W.S., N.M.D.B., M.C., K.G.F., G.L.B., B.G., G.L.W., J.G-R., R.F.L., C.A.C., M.J.S.). In addition: Many seen in lower Kennet during the cold spell (M.C.). 125. CORNCRAKE. Heard in field between Baydon and Greenhill, 4th July, the first time for several years (Mr. Charles Hale per I.C.G.). One flushed from tall clover at Imber, 11th Oct., the first one seen since before 1939 (J.G.M.). 126. MOORHEN. Missing from usual haunts at West Ashton. High mortality (J.F.). Several found dead, Marlborough (M.C.). Twenty found dead along the Bybrook, Castle Combe to Ford (R.F.L.). Badly affected at Staverton (G.L.B.). Missing at Longleat (R.H.P., G.W., G.L.W.). Feeding in gardens, farmyards, etc., during cold weather 216 (J.C.R., W.M.C., I.C.G., R.F.L., B.G.). Not seen at Luckham, where usually numerous, until late in year (E.E.G.L.S.). 127. cooT. Breeding population at Corsham Lake down by 50 per cent. (J.C.R.). Dying bird at Yatton Keynell (R.F.L.). Poor winter record at Coate Water and Braydon Pond (G.W.). Great reduction in Kennet Valley (I.C.G.). 133. LAPWING. About 2,000 in Barbury castle area, 3rd Nov. (G.L.w.). One ringed as chick at Old Sarum on 23.5.59, shot at Meco, Madrid, Spain, on 27.1.63 (D.E.F., A.J.H.). Marked weather movements noted during December (R.H.P., R.G.B., R.F.L., WEB: RG. G.L.W.). 134. RINGED PLOVER. Two at Lacock G.P., 4th Aug. (J.C.R.). Two, 11th Aug., and one, Ist Sept., at Rodbourne S.F. (G.W.). 135. LITTLE RINGED PLOVER. One at Rodbourne S.F., 2nd and 8th Sept. (H.R.W., G.L.W.). 139. GREY PLOVER. Four, in winter plumage, near Biddestone, 2nd Mar. (R.C.F.). 140. GOLDEN PLOVER. Early in the year the largest flocks seen were: 30 near Maiden Bradley, 15th Feb. (J.C.C.O.); c. 500 below Barbury Castle, 16th Mar. (G.L.W.); 1,000 near Chisledon, 28th Mar. (M.C.); c. 100 overflying Beckhampton, towards east, 16th Apr. (E.L.J.). In autumn: 7 in a field near Barton Stacey at the early date of 23rd June (E.A.R.E.). Several hundred, Old Sarum airfield, Sept. (A.J.H.); probably in excess of 2,000 at Stourton in autumn (J.C.C.O.); ¢. 400 at Chisledon, 2nd Nov. (M.C.); 200 below Barbury Castle, 15th Sept., 500 between Hodson and Barbury, 16th Nov. (G.L.W.); 400 at Ford, 5th Dec. (D.E F., A. J.H.) ; 220 flying south-south-west at Pickwick, 24th Dec. (A.A.G)). 145. SNIPE. One feeding within 5 yds. of a house during the cold spell (G.L.B.). First arrived at Shalbourne Mill, 31st July (E.A.R.E.). Numbers very low in the Swindon area; following frost largest number at Coate Water, 26 on 17th Nov. (G.L.W.); 30 at Lacock G.P., 24th Nov. (J.C.R.). None heard ‘drumming’ in Kennet Valley (1.C.G.). 147. JACK SNIPE. Several records in Kennet Valley, during cold spell, one found dead (M.C.). One, Yatton Keynell, 8th Jan. (R.F.L.). One, Lacock G.P., 3rd Mar. and goth Dec. (J.C.R.). One at Coate Water in March and May and ones and twos from 28th Sept. to the end of year (G.L.W.), 2-5 at same place, 2nd Nov. (M.C.). One at Sherrington, 26th Jan. (M.J.S.). 148. woopcock. Possible breeding at Clarendon, June (R.W.). Birds seen roding in two different areas at Longleat, 26th Apr. and 1st May (N.M.D.B.) and at Savernake, and June (M.C.). Many in copses on downs and some in water meadows during the cold spell (M.C.). One feeding with pheasants near Aldbourne (Mr. G. Wentworth per I.C.G.). 150. CURLEW. Heard near Urchfont, 31st Mar. (J.M.S.) and a nest with four eggs there on 9th May (R.J.S.) and 11th May (A.W.S.). A pair active in marshy meadows near All Cannings, 15th Apr. and a pair flying together over meadows between Kennet and Avon canal and Seend Cleeve, 25th Apr. (R.J.5.). Considered to be breeding in Lechlade-Hannington-Inglesham district (G.L.W.). One bubbling, Trowle Common, 21st Mar. (J.F.). Up to six birds in parties at Yatton Keynell, 16th June to 22nd July (R.F.L.). Noted in various parts of the county (J.B., J.F., M.E.T., R.G.F., J.C.R., P.R., G.H.F., R.W., M.C.). One ringed as nestling 30.5.61 Warcop: 54° 32’ N., 2° 24’ W. (Westmorland), found dead or dying c. 6.4.62, Oare, Marlborough (Wiltshire), 215 miles south., cf. B.B., Lv1, 497. 151. WHIMBREL. One seen at Idmiston, 23rd Apr. (G.H.F.), and one at Salisbury, 23rd May (D.C.). 156. GREEN SANDPIPER. Again noted throughout the county in February, March, April, June, July, August and December (J.C.R., R.C.F., D.W.F., A.J.H., R.F.L., R.G.B., .G.W., P.R., G.L.W., C.R., E.A.R.E.). 157. WOOD SANDPIPER. One to three, at times in company with green and common sandpipers, at Rodbourne S.F., 9th to 17th Aug. (P.R., G.L.W., G.W.). 217 159. COMMON SANDPIPER. Seven on Longford Castle lawn, 15th Apr. (R.O.P.B.). Three at Coate Water, 11th Apr., then seen every month (maximum 5) up to 15th Sept. (G.L.W.). Four and 7 on River Avon near Bradford-on-Avon, 18th Apr. (R.J.S.). Between 20 and 30 at Rodbourne S.F., 28th Aug. and 2 at Castle Combe, 5th May (R.F.L.). Varying numbers of 1-4, April and May, and 1-6, July to Sept. at Lacock G.P. (J.C.R.). Ones and twos seen in various parts of the county (A.A.C., N.M.D.B., J.C.R., GW... 1-C.€.0. MES): 161. REDSHANK. ‘Two pairs bred in Littlecote Water meadows (1.C.G.). Pair near Etchilhampton Water, goth Apr. (R.J.S.). Small party in watercress beds at Longbridge Deverill, 2nd Mar. (E.J.H.). First seen 7th Mar., but fewer than in previous years, Marlborough (M.C.). Seen at Coate Water, 16th Mar., but no breeding (G.L.W.). One, possibly two, calling on Marlborough Downs, 22nd June (R.J.L.). 165. GREENSHANK. Single bird at Coate Water, 28th Apr. (G.W.), 2nd May (G.L.W.). Two at Rodbourne S.F., 17th Aug. (G.L.W.). Ones and twos at Lacock G.P., 20th Aug. to 13th Oct. (J.C.R.). One at Idmiston, 12th Sept. (G.H.F.). 169. KNOT. One at Coombe Bissett watercress beds in Feb. (C.H.A.), and presumably the same bird found dead in observer’s garage, roth Mar. (W.D.). 171. LITTLE STINT. One at Rodbourne S.F., and and grd Sept. (H.R.W.). 178. DUNLIN. One feeding on mudbank in River Avon between Seagry and Dauntsey, 24th Feb. (R.C.F., J.C.R.). Three on gth May and one on 16th July and 28th Dec. at Coate Water (G.L.W.). One at Lacock G.P., 21st July (J.C.R.). Ones and twos seen at Rodbourne S.F. between roth Aug. and 5th Sept. (P.R., H.R.W., G.L.W.). 184. RUFF. One to four (1 ruff, 3 reeves) seen at Rodbourne S.F. on several days between roth Aug. and 5th Sept. (P.R., R.C.F., G.L.W., G.W., H.R.W.). 187. GREY PHALAROPE. One killed by car at Winterbourne Dauntsey in the middle of November. It had been seen in the area on several days prior to its death (R.W., D.E.F., A.J.H.). 189. STONE CURLEW. One pair reported breeding. Last seen 3rd Nov. (G.L.W.). A successful hatching after a farmer had removed and replaced the eggs during cultivation of the field (E.H.J.). Pair disturbed by the cultivation of what otherwise would have been a suitable nesting area (J.E.M.). Two pairs on ploughland 13th June, but nesting success unknown (D.E.F., A.J.H.). Several nesting pairs observed, but less numerous than formerly (E.G.P.). Nesting 11th May (R.W.). Seen and heard in other areas (R.M.Y., G;HF.,-M.J-S., A.B. G:L.B., .G.L.W., R-J.S., -G.W..,.B.G., E.L.J.). Party-of-7-with Lapwings, 12th Sept. (N.M.D.B.). 198. GREAT BLACK-BACKED GULL. ‘I'wo seen at Lacock G.P., 15th Sept. (J.C.R.). One at Rodbourne S.F., 26th Sept. (G.L.W.). At Longford, one on 14th and 15th Jan. and some on 3rd Oct. (R.W.). 199. LESSER BLACK-BACKED GULL. Some winter records: 30 on 6th Jan., 17 on 13th Jan. and a few on other days up until 24th Feb., 2 on 10th Nov., and 4 on 1st Dec. at Lacock G.P., 120 near Bradford-on-Avon, 25th Mar. (J.C.R.). 200. HERRING GULL. Highest numbers were: between 15 and 80 seen at Lacock G.P. between 6th Jan. and 24th Feb. Twelve at Corsham Lake, toth Feb. and 1oth Dec. (J.C.R.). ne BLACK TERN. Birds in full summer plumage recorded: One at Corsham Lake, 30th May (J.C.R.); 2 at Braydon Pond, 30th May (R.G.B.); and 3 at Coate Water, 31st May. Also one juvenile there, 3oth Aug. (G.L.W.). 217 or 218. COMMON or ARCTIC TERN. One at Corsham Lake, 11th Apr. and 2oth Aug. (J-C.R.). oe sTOCK DOVE. Thought to have suffered from the cold weather (M.C.). 234. WOOD PIGEON. All observers agree that this species suffered very heavy casualties. Estimates of 50 to go per cent. casualties in flocks given (J.F., R.G.B., B.M.S., R.F.L., G.L.W., J.C.). But: Still very common, flocks of several thousand over Dellings Copse in autumn (M.C.). 218 235. TURTLE DOVE. First noted 25th Apr., Bromham (J.B.), 27th Apr., Winterbourne (A.J.H.); 1st May, Shaw (C.A.C.). Last seen 24th Aug., Aldbourne (I.C.G.); 21st Sept., Corsham Park (J.C.R.); 24th Sept., Barton Bottom (M.C.). 236, COLLARED TURTLE DOVE. Heard calling at Devizes, 1gth, 23rd and 26th May. Not heard after the latter date (B.G.). Three arrived in garden at Amesbury, Ist Jan., later increased to 5, but no young birds seen (D.W.). At least one pair known to have nested and 12 to 17 flying about Marlborough in autumn (M.C., I.C.G., G.L.W.). One at Pitton, 27th Oct. (R.W.). 237. cuckoo. First heard 3rd Apr., Fovant (R.C.C.C.); and Neston, 9th Apr., Corsham Park (J.C.R.); 14th Apr., Southwick (G.L.B.). Last seen 12th Aug., Berwick St. John (J.E.M.); 7th Sept., Urchfont (J.M.S.); 26th Sept., Wardour Castle (C.A.C.). 241. BARN OWL. ‘I'wo died in Little Somerford and one at Tockenham as the result of the cold (R.G.B.). Drastic reduction in numbers noted at Marlborough (M.C.), and Tilshead (J.G.M.). Two, seen hunting in broad daylight at Longford, 13th Jan. (I.R.). Numbers have decreased during the last three years in the Chippenham, Castle Combe and Corsham areas (R.C.F., R.F.L., J.C.R.). One at Brinkworth in June and July (P.R.). None seen since hard weather, Codford (E.V.F.). 246. LITTLE OWL. One at Chippenham, Jan., Feb. and Nov. (J.S.B.). One killed and fed on a starling at a bird table at Maiden Bradley and this became its regular practice during the cold weather (J.C.C.O.). Two fed on fat at Corsham during Jan. and Feb. (J.C.R.). One feeding on a dead moorhen during the cold weather, Yatton Keynell (R.F.L.). A juvenile picking at the body of another recently killed on a road at Braydon, 8th July (P.R.). Two pairs bred at Yatton Keynell. A dead starling was found at the roost of one pair. Six young were being fed on top of a roadside wall, 5th June (R.F.L.). A nest with young in roadside oak near Chippenham (R.C.F.). 247. TAWNY OWL. Two or more pairs resident at East Knoyle (B.M.S.); breeding noted there (J.E.M.). Three eggs in a nest at Bromham, 14th Apr.; 3 nests locally (J.B.). Two nests, 4 and 3 young, others suspected in Marlborough area (M.C.). Seen and heard at Corsham Park and Lacock G.P. (J.C.R.). Two pairs in the Bybrook Valley and a pair nested at Yatton Keynell (R.F.L.). A pair heard in Chippenham in November (J.S.B.). 248. LONG-EARED OWL. Only one record: a single bird seen several times near Cherhill in May. Reason to believe one of a pair nesting in the area, but this was not confirmed (G.L.W.). 249. SHORT-EARED OWL. Seen flying over the same down near Wishford in groups of 3-6 in November, over a number of years (E.G.P.). Two at Hackpen, 26th Oct. (R.U.). Five on 28th Nov. and 10 on ist Dec. seen at Dean Bottom, and many observers saw up to 5 there subsequently. One seen at Fyfield Valley, 7th Dec. (M.C.). One near Shrewton, 4th Nov. (C.A.C.); 3 near Everleigh, 28th Dec. (A.W.S.). 252. NIGHTJAR. Usual territories occupied at Maiden Bradley (J.C.C.O.). At least 6 singing on Somerford Common, 11th June (P.R.) and several there, 1st Aug. (G.L.W.). Two pairs displaying at Fosbury, 1st July (G.L.W., C.N.T.). One in Savernake Forest, and June (M.C.), 3, 21st June (G.W.); 1, West Woods, rgth June (M.C.). Single birds crouching on the road, Grittenham, 19th May; Brinkworth, 31st May (P.R.); and Somerford Keynes, 6th Aug. (G.M.K..). 255. SwirT. First seen 23rd Apr., Harnham (E.B.) ; 24th Apr., c. 12, Gastard (C.S.H.) ; 25th Apr., 3 at Porton range (R.F.L.), 2 at Winterbourne (D.E.F., A.J.H.), Chippenham (R.C.F.), and 5 at Swindon (G.L.W.). Last seen 28th Aug. Chippenham (R.F.L.); goth Aug., Coate Water (G.L.W.), Lacock (J.C.R.); 15th Sept., Salisbury (F.P.E., R.P.). 258. KINGFISHER. Seen in various parts of the county (R.M.Y., P.C.M., R.F.L., B.M.S., D.W.F., O.V.K., M.K.L., J.C.C.0., R.C.F., J.C.R.). Seemingly fewer, Marl- borough (M.C.). A juvenile, Littlecote, 5th June (I.C.G.). An adult male and female and 5 juveniles ringed at Oakhill Water meadows (D.A.W.A.). None seen at Cole Park (where as many as 17 ringed in a season) until 21st Dec. (E.J.M.B.). 219 261. HOOPOE. One, in a house yard and allowing an approach to within a few feet, at Bradford-on-Avon in the beginning of May (Mrs. Butler per C.S.). One on a road at Gutch Common, 11th July (A.J.G.). 263. GREATER SPOTTED WOODPECKER. The general opinion is that this species survived the cold weather very well in most districts in the county. This is probably due to its willingness to visit bird tables and other feeding places in gardens. 264. LESSER SPOTTED WOODPECKER. Single birds at Porton, 24th Feb. (R.D.); Stratford Tony (R.M.Y.); Chippenham (J.S.B.); Corsham Park, 9th Mar. (G.W.H.); Marlborough, 3rd, 17th and 19th May, Mildenhall, rgth May (M.C.); Wroughton Res., 17th Nov. (G.L.W.), and Pitton, 2nd Dec. (R.W.). Singles and twos seen at Corsham Park in April and May (J.C.R.). 265. WRYNECK. One in an oak tree at Coate, 11th Sept. (G.L.W., H.R.W.). 271. WOODLARK. One resident pair, Maiden Bradley (J.C.C.O.). A bird singing from tall elm near Pewsey, 3rd May (B.G.). Since Grovely Middle Hills was ploughed up Woodlarks appear to have left locality (B.M.S.). Probably nesting on Porton Ranges in June (R.W.). 272. SKYLARK. Very badly hit by winter. Instead of usual ¢. 30 nests, fields between Coate and Swindon held one pair (G.L.W.). Near Chippenham few nesting sites but a steady flow passed through in autumn (R.C.F.). 274. SWALLOW. First seen 2nd Apr., Fovant (R.C.C.C.); 7th Apr., East Grimstead (W.M.C.), 9th Apr., Wishford (E.G.P.), Swindon (G.L.W.), and Corsham (J.C.R.). Numbers in a roost in willows at Coate built up from c. 200 on 18th July to c. 5,000 on 1oth and 14th Aug. There was a sharp drop in numbers to c. 200 on 27th Aug. probably due to spraying of vegetation with herbicide (G.L.W.). On 13th Aug. c. 800 were roosting in reeds at Lacock G.P. (J.C.R.). Last seen 23rd Oct., Chippenham (R.W.W.); 27th Oct., Biddestone (R.C.F.); 29th Oct., Bratton (E.E.G.L.S.). Less success in nesting this year noted by R.C.F. and many eggs failed to hatch on farms in Brinkworth area (P.R.). A swallow ringed at Coate 3.8.61 recovered inside a house at Coate 25.4.63 (G.L.W.). 276. HOUSE MARTIN. First seen gth Apr., Corsham Lake (J.C.R.); 11th Apr., Bromham (J.B.) and Longdean (R.F.L.); 12th Apr., Bowood (G.L.B.), and Coate (G.L.W.). A colony of 15 occupied nests unusually situated 6 ft. from ground inside a low farm building in daily use near Castle Combe (R.C.F., R.F.L.). An unusual type of nest built up from a metal bar 7 in. below eaves of a house in Devizes did not touch wall at any point, but was built up to the eaves (B.G.). Last seen 29th Oct., Corsham (A.A.C.) and Trowbridge (J.F.); 2nd Nov., Bratton (E.E.G.L.S.); 3rd Nov., one hawking insects over Trowbridge (J.F.). 277. SAND MARTIN. First seen 6th Apr., Salisbury (F.P.E.); 12th Apr., Coate (G.L.W.) ; 14th Apr., Ashton Keynes (G.W.). Breeding site in Chippenham destroyed by widening of River Avon (R.F.L.). About 5 pairs in Dinton Greensand Pit, 24th May (B.M.S.) and 10 pairs with young in nests at Sahara Sandpits, Calne, in July (J.C.R., R.C.F.)- Numbers rose to c. 400 on roth Aug. at Coate roost with Swallows (G.L.W.). 278. GOLDEN ORIOLE. An unconfirmed account of a cock bird seen from train between Dinton and Tisbury, 5th July, by an observer who knows Orioles in France. Described as a very striking yellow and black (Miss Smith per H.J.F.S.). 280. GARRION CROW. Forty-eight at a roost at Lacock G.P., 17th Mar. (J.C.R.). Not affected by hard winter, Bratton (E.E.G.L.S.). 282. ROOK. Observers in Chippenham, Bratton, Biddestone, Tilshead and Seagry noted that numbers appeared unaffected by hard winter (R.C.F., R.F.L., E.E.GL.S., j.G.M., R.G.B:). 288. GREAT TIT. Considerable decrease in Swindon area due to hard winter (G.L.W.). Fewer than usual in autumn, Yatton Keynell (R.F.L.). Losses of this bird and Blue Tit at Bratton more than made good by December (E.E.G.L.S.). 289. BLUE TIT. Conflicting opinions as to whether this species is decreased in numbers following hard winter. 220 292. MARSH TIT. Only one pair at Coate where normally 3 or more (G.L.W.). Other records indicate normal numbers. 293. WILLOW TIT. A pair singing, Grovely Wood, 22nd Apr. (B.M.S.). A pair in Hodson Woods, 5th May (G.L.W.). One with a party of Long-tailed Tits in garden at Pitton, 3rd Oct. (R.W.). One in Wroughton Copse, 19th Oct. (E.L.J.). Two with Marsh Tits at Coate, 2nd Nov. (G.L.W.), and 2 ringed near Aldbourne, 8th Dec. (K.C.B.). 294. LONG-TAILED TIT. It would seem that this bird survived the winter in some localities but not in others. Seen at Urchfont, 31st Mar. (J.M.S.). Parties of up to 20 near Castle Combe, July and Aug. (R.F.L.); several large groups in Grovely Wood in autumn (E.G.P.); 10 at Winterbourne, 25th Oct. (D.E.F., A.J.H.), and 10 plus with Coal Tits in conifers at Longleat, 19th Dec. (N.M.D.B.). Fair numbers in woods near Pitton, 2nd Dec. (R.W.). Small numbers seen by J.H., C.R., B.G., E.V.F. et al. None seen in Swindon area (G.L.W.), Hilperton (G.L.B.), Cole Park (E.j.M.B.), and Seagry (R.G.B.). 296. NUTHATCH. A nest in wood near Fonthill, 12th May, was deserted after laying had begun, when a Greater Spotted Woodpecker made a hole through the side of the nest chamber (J.E.M.). Seven observers noted that numbers appeared unaffected by hard winter. 298. TREE CREEPER. Only 2 singing at Coate indicating a winter mortality of c. 75 per cent. (G.L.W.). Last seen in hard weather at Seagry, 23rd Jan., then none until 25th Nov. (R.G.B.). Sitting on eggs, Wincombe Park, 2nd May (J.E.M.). Seen throughout year in Corsham Park (J.C.R.). ‘two at Clarendon Lakes, 16th Apr. (R.W.). Six records of single birds (M.K.L., K.C.B., 1.G. e¢ al.). Seven noted in woods at Castle Combe, 27th Feb., and survival good in Bybrook Valley (R.F.L.). 299. WREN. Suffered severely in winter. One, 29th Mar., Potterne, and 1 in December, Devizes, the only birds seen (B.G.). First seen since January, 30th Nov., in garden at Chippenham (C.R.). First seen since cold at Pitton, 31st Oct. (R.W.). None seen at Norton Bavant until 24th Nov. (K.G.F.). Decreased by 50 per cent., East Knoyle (B.M.S.). None seen at Ramsbury (V.C.L.). Only 1 seen in year, Winterbourne, 26th Apr. (D.E.F., A.J.H.). First song, 12th Apr. and this subsequently proved to be the only male at Coate. None in Hodson or other areas where normally numerous. Winter mortality in Swindon about 95 per cent. and no knowledge of any breeding in area (G.L.W.). A very rare bird in north-west Wilts. None seen by observer, a pair reported in Brinkworth, but no breeding (P.R.). In sheltered Bybrook Valley, although numbers below usual 24 were singing between Castle Combe and Slaughterford, 15th Dec. (R.F.L.). In one garden in Chippen- ham there were 2 pairs in August (J.S.B.). 300. DIPPER. A survey of birds on the Bybrook showed numbers apparently unaffected by hard winter. Breeding started at normal date and 5 pairs were found. One pair produced no young from 3 nests built. Two pairs produced 5 and 3 young respectively from first brood, but neither pair bred again. Observation of other 2 pairs impossible, but failure of one of them suspected. Disturbance was not greater than usual in the area and observers were unable to account for breeding failures (R.C.F., R.F.L.). Dippers appear to have left the Ebble between Stratford Tony and Bishopstone since the rebuilding of two bridges (R.M.LY.). — MISTLE THRUSH. Many feeding in suburban gardens in Swindon, Jan. and Feb. all in poor condition. Only 1 pair at Coate, where normally 3 pairs, and even this pair did not breed (G.L.W.). Numbers much diminished, Codford (E.V.F.). In autumn many parties were seen flying west in September, up to 30 in some groups at East Knoyle (B.M.S.). A flock of ¢. 25 at Cole Park, where yew berries were plentiful, 25th Sept., and continued more numerous than usual (E.J.M.B.). 302. FIELDFARE. During the hard weather a flock of ¢. 70 near Winterbourne, 4th Jan. (E.L.J.). The majority appeared to have left the Swindon area at onset of cold and most that stayed died (G.L.W.). Some came constantly to feed on rotting apples at East Knoyle, Jan. and Feb. (B.M.S.). Small numbers re-appeared in March and the last 221 15 seen in spring were 6th Apr., c. 20 at Hodson (G.L.W.), and 13 at Winterbourne (D.E.F., A.J.H.); 12th Apr., Etchilhampton (R.J.S.); 13th Apr., Pitton (R.W.). First autumn records: 13th Oct., Rood Ashton (J.F.); 14th Oct., Corsham (J.C.R.); 20th Oct., Ford (R.F.L.). Some observers note that numbers were lower than usual. 303. SONG THRUSH. Badly hit by cold weather, Jan. and Feb. Only 3 pairs at Coate instead of usual 12 or so. Both local pairs in Swindon died and not replaced (G.L.W.). Several found dead at snow thaw, Seagry, and very scarce for rest of year (R.G.B.). Only single bird came to garden at Hilperton in autumn and no breeding pairs (G.L.B.). Numbers much diminished, Codford (E.V.F.) and Bratton (E.E.G.L.S.). None heard singing at Longleat, goth Mar., but 2 seen in flight (R.H.P.). A solitary bird noted in Coppershell area from spring throughout year (C.S.H.). Some autumn movement noted: 20 plus feeding and drinking at a pond at Brinkworth for a few minutes, rgth Oct. (P.R.); 8 with Redwings at Barbury, 20th Oct., and night passage over Swindon, 16th-17th Nov. (G.L.W.). An adult ringed 3.5.60 at Mellum: 53° 44’ N., 8° 10’ W. North Sea, Germany, recovered dead or dying 16.4.62 at Calne: 51° 26’ N., "2° 00! W. (Wiltshire), ch BB, LVI (1963), 536. 304. REDWING. Many found dead in Swindon area, Jan. and Feb. (G.L.W.). Several found dead as snowdrifts melted in garden, Seagry, in March (R.G.B.). Last seen in spring, 12th Mar., Braydon (D.A.W.A.); 17th Mar., Corsham (J.C.R.); 18th Mar., Potterne Wick (B.G.). First noted in autumn, 2oth Sept., Pitton (R.W.); roth Oct., East Knoyle (B.M.S.); 11th Oct., Easton Royal (C.A.C.). Heard passing over Trowbridge at night, 15th and 22nd Oct. (J.F.). On 18th Oct., a night of low cloud and south wind, calls of single birds and small parties heard at intervals for an hour around 2100 hrs. and again at dawn the next morning (R.W.W.) ; a moderate passage also noted at Swindon the same night (G.L.W.). A roost of c. 200 located at Slaughterford, 12th Nov. (R.F.L.). 307. RING OUZEL. Six females and 1 male seen on slopes of downs at Barbury, 31st Mar., feeding among hawthorn scrub, when disturbed the birds moved out on plough feeding in loose contact. Very restless and active (G.L.W.). One reliably reported in a garden at Shalbourne, 26th July (E.A.R.E.). 308, BLACKBIRD. Some losses in Swindon area, but less than other thrushes, probably less than 50 per cent. (G.L.W.). A first winter male ringed at Brinkworth 4.10.62, recovered dead from exposure 19.1.63, near Launceston, Cornwall. A typical hard weather move- ment (P.R.). A juvenile male ringed 20.9.59 Amsterdam: 52° 21’ N., 4° 55’ E. Netherlands, recovered dead or dying 1.1.60 at Little Durnford 51° 07’ N., 1° 50’ W. (Wiltshire), et B.8 LVL 537. 311. WHEATEAR. First seen 11th Mar., Porton (F.P.E.); 26th Mar., Stanton. St. Bernard (B.G.); 28th Mar., near Marlborough (M.C.). Breeding noted at Middle Down, Alvediston (J.E.M.), a family party at Yarnbury Castle, 23rd June (B.M.S.) and on Tan Hill, 27th July (B.G.). Last seen 27th-29th Sept., Lyneham airfield (R.W.W.); 28th Sept., Wroughton airfield and 5th Oct., Ham Hill (G.L.W.). 317. STONECHAT. None wintering in Swindon area this year, possibly reduced by hard weather (G.L.W.). Two seen at Rockley, 3o0th-31st Mar. (M.C.). A male seen feeding near Hackpen, 18th Dec. (D.A.W.A.). 318. WHINCHAT. First seen 18th Apr., Coate (R.H.W.) ; 28th Apr., Wolfhall (C.N.T.), 4th May, Seend (G.L.B.). Frequent records on Marlborough Downs in May (M.C.). Singing at Everleigh, goth June (B.G.). Last seen 13th Sept., Yatton Keynell (R.C.F.); 14th-15th Sept., Rodbourne S.F., 15th at Barbury (G.L.W.). 320. REDSTART. First seen roth Apr., Upton Lovel (J.R.I.P.); 11th Apr., Ford (R.F.L.); 15th Apr., Pewsey (B.G.). One at Sutton Veney, 20th Apr. (J.R.I.P.). Seven singing at Castle Combe, 26th Apr., and at least 3 pairs bred there (R.F.L., R.G.F.). Four pairs in Coate area, 4th May (G.L.W.). Noted as usual in Savernake (K.C.B.). A regular nester at Maiden Bradley (J.C.C.O.). Heard singing in May in Stockton Wood (B.M.S.); Fonthill and Wincombe (J.E.M.). Nests in two distinct habitats at Berwick St. John, mature deciduous woodland and downland scrub (J.E.M.). Singing in 5 different 222 places in Longleat, 5th May (N.M.D.B.). Last seen 29th Aug., Shaw (C.A.C.); 8th Sept., one dead on road at Porton (A.A.D.) ; 2 passing through a garden at Chippenham, 7th Nov. (J.S.B.), a late date. 321. BLACK REDSTART. A male bird watched at ro ft. through window as it perched on a rose bush in Monkton Park, Chippenham, 26th Apr.; sooty black throat and breast, rufous rump and tail clearly seen (M.F.A.). 322. NIGHTINGALE. First heard 15th Apr., Sandridge Vale (R.J.S.); 22nd Apr., Grovely (B.M.S.); 23rd April, Coate (G.L.W.). A nest with 3 well-feathered young, a small clutch, at Coate, 6th June (G.L.W.). 325. ROBIN. Many died in Jan. and Feb. and several found dead in Swindon area. Numbers very low there until autumn when they began to build up, possibly migrants from continent (G.L.W.). More than usual in garden at Cole Park where they might have come in hard weather and stayed on to breed (E.J.M.B.). 327. GRASSHOPPER WARBLER. First heard roth Apr., Great Bradford Wood (R.J.S.); 22and Apr., Maiden Bradley (J.C.C.O.); 26th Apr., Winterbourne (D.E.F., A_J.H.). Reported in 11 other areas, including a record of at least 20 in fir plantations near Fosbury (G.L.W.). Last heard by River Avon, Dauntsey, 28th June (R.C.F.); 18th July, Savernake (M.C.). 33. REED WARBLER. First noted 12th Apr., Corsham Lake, where later 8-10 pairs bred (J.C.R.); 18th Apr., Coate Water (G.W.). About 15 pairs in reed beds at Coate, 26th May. Two ringed as nestlings in the same nest at Coate in 1961 were re-trapped as a breeding pair in 1963 (G.L.W.). Last seen 15th Sept., Corsham Lake (J.C.R.). 334. MARSH WARBLER. One singing at Coate, 20th and 23rd June, not seen or heard again (G.L.W.). One singing in Porton area, 15th June, when sound recordings were made. The bird flitting about and searching for insects in bushes and small trees. No sign of it until goth July when heard and seen working through same area (D.S.G.). 337. SEDGE WARBLER. First noted 12th Apr., Coate (G.L.W.); 13th Apr., Oakhill (D.A.W.A.); 21st Apr., Corsham (J.C.R.). On 12th May c. 20 nesting pairs at Coate (G.L.W.). Only 4 birds were ringed at Oakhill in 1962 and of these 3 were re-trapped at the same place in 1963 (D.A.W.A.). One ofa brood ringed at Maiden Bradley recovered at Chew on autumn migration of same year, 28 miles west-north-west (J.C.C.O.). Last seen 16th Sept., Coate (G.L.W.); 20th Sept., Corsham (J.C.R.). 343. BLACKGAP. A female observed at a bird table at Corsley for 3 or 4 days in second week of January feeding with tits and other birds on bread, fat and currants. It was very weak at first but seemed to recover. Not seen after very severe frost, 13th Jan. (E.H.J.). A male, possibly also overwintering, at Granham Hill, 24th Mar. (M.C.). Spring arrival noted 12th Apr., Castle Combe (R.F.L.); 13th Apr., Cole Park (E.J.M.B.), and Oakhill (D.A.W.A.). Last seen 5th Sept., Coate (A.W.S.); 14th Sept., Neston (J.C.R.). 346. GARDEN WARBLER. First noted 27th Apr., Longleat (N.M.D.B.) and Winter- bourne (D.E.F., A.J.H.); 30th Apr., Salisbury Racecourse (G.H.F.); 5th May, Coate (G.L.W.). Last seen 7th Sept., Bromham (J.B.), and Winterbourne (D.E.F., A.J.H.). 347. WHITETHROAT. First noted 18th Apr., Coate (G.L.W.); 20th Apr., Corsham (J.C.R.); 23rd Apr., Bromham (J.B.). Last noted 15th Sept., Goate (G.L.W.) ; 23rd Sept., Corsham (J.C.R.); 28th Sept., Yatton Keynell (R.F.L.). 348. LESSER WHITETHROAT. First noted 25th Apr., Idmiston (C.M.F.); 26th Apr., Cole Park (E.J.M.B.); 29th Apr., Seagry (R.G.B.). Observed feeding young at Yatton Keynell, 7th May (R.F.L.). Last seen 5th Sept., Coate (A.W.S.); 15th Sept., Seagry (RGB). a WILLOW WARBLER. First heard 16th Mar., Codford (K.G.F.); 17th Mar., Bromham (J.B.); 31st Mar., Corsham (J.C.R.). Last noted 8th Sept., Coate (G.L.W.); 15th Sept., Corsham (J.C.R.); 18th Sept., a passage movement, Shalbourne (E.A.R.E.). 356. CHIFFCHAFF. First noted 12th Mar., Idmiston (G.H.F., A.A.D.); 13th Mar., Bratton (E.E.G.L.S.); 14th Mar., Shalbourne (E.A.R.E.). Last noted 2nd Oct., Seagry 223 (R.G.B.); 3rd Oct., Corsham (J.C.R.); 6th Oct., Weavern (R.C.F.), and Rodbourne S.F.. (G.L.W.). 357. WOOD WARBLER. First noted 28th Apr., Longleat and 30th Apr., West Knoyle; singing in 5 localities at Longleat, 7th May (N.M.D.B.). Singing 13th May, Everleigh Ashes and 2nd June, Savernake (M.C.). Heard among beeches on Fonthill Terraces in June (B.M.S.). A pair were feeding young in the nest, Wincombe Park, 7th July (J.E.M.). A pair near Codford, 21st July (K.G.F.). 364. GOLDCREST. ‘wo seen feeding among ivy at Semley on a February day of temporary thaw (J.E.M.). One seen at Weavern, gth Mar. (R.F.L.). Seen on only 3 occasions in April and May in whole year (N.M.D.B.). Reported in Holt in May (C.S.B.). Very few records after cold spell, Marlborough area (M.C.). Seen in garden at Stratford Tony in late May (J.H.). No other records, but nil returns from B.G., R.H.P., E.J.M.B., B.M.S., G.L.W., R.G.B. 366. SPOTTED FLYCATCHER. First seen 8th May, Marlborough (M.C.); gth May, Corsham (J.C.R.); 12th May, Porton (A.A.D.). Fewer than usual this year in Swindon area (G.L.W.), and at Corsley (E.H.J.). Last seen 17th Sept., Corsham (J.C.R.); 22nd Sept., Coate (G.L.W.); 25th Sept., Seagry (R.G.B.). 368. PIED FLYCATCHER. A female watched for 10 minutes in a garden at Seagry, 5th Aug., as it worked its way from one perch to another and then away. Very distinct wing-bar (R.G.B.). Possibly 2-3 birds and at least one clearly identified by well-defined wing-bar at Urchfont, 7th Sept. (J.M.S.). 371. HEDGE SPARROW. Decrease estimated about 60 per cent. in Coate area, 12th May (G.L.W.). None present on 50 acres of Somerford Common where 4 pairs bred in 1962 (P.R.). Both these observers note some recovery in autumn when birds were noted as abundant in Chippenham (R.C.F.), and as having come through the winter well at Cole Park (E.J.M.B.). 373. MEADOW PIPIT. Noted on building sites and greens in Chippenham in parties up to 6 until October (R.C.F., R.F.L.). At Lacock G.P., ¢. 45 on 5th Apr. and c. 60 on 20th Oct. (J.C.R.). A small passage seen at Ham Hill, 5th Oct., and a heavy passage over Barbury scarp, 19th-2o0th Oct. (G.L.W.). 376. TREE PIPIT. First seen r1th Apr., 3 with Meadow Pipits, Coate (G.L.W.); 12th Apr., Maiden Bradley, where it is a very common summer resident. One pair occupied a territory in a 50-ft. high larch plantation with only limited breaks in tree canopy (J.C.C.O.). Last noted 15th Sept., 1 at Barbury (G.L.W.). 380. PIED/WHITE WAGTAIL. A roost of 20-30 birds on ledges of shops in Chippenham dispersed in hard weather at end of January (R.F.L., R.C.F.). Roost at Coate, c. 40, 18th July; at Rodbourne c. 50 and an adult male of White race (G.L.W.). Sixty at Lacock G.P., 4th Oct. (J.C.R.). Nine observed flying c. 100 ft. up below stratus in drizzle, almost exactly south, at Chippenham, 6th Oct. (R.W.W.). An adult female ringed 1.9.62 at- Idmiston 51° 08’ N., 1° 44’ W. (Wiltshire), found dead or dying 4.12.62, at Creon 44° 47’ N., 0° 22’ W. (Gironde), France, cf. B.B., Lv1, 518. 381. GREY WAGTAIL. On Bybrook only one nest near Castle Combe (R.F.L., R.C.F.). Two at Clatford, 23rd June, and 2 at Coate, 24th Oct., the only records from M.C. Six young at Rodbourne S.F., roth Aug. (P.R.). A few single birds noted in late summer and autumn. Last seen at Semley about half way through cold spell (J.E.M.). No records Warminster area, usual breeding sites deserted (N.M.D.B.). None seen at Cole Park (E.J.M.B.). 382. YELLOW WAGTAIL. First seen 11th Apr., Coate (G.L.W.); 12th Apr., Lacock G.P. (J.C.R.); 17th Apr., Cole Park (E.J.M.B.). Twelve at Coate, 29th Apr. (G.W.), but most of the pairs there lost their nests. A roost of c. 30 at Coate, 18th July (G.L.W.). Two to three pairs bred at Lacock G.P. (J.C.R.); at least 3 pairs in water meadows at Stratford and a nest in a tussock of grass in meadows at Britford in June (D.E.F., A.J.H.). A nest with young at Covingham, 29th Aug. (G.W.). Last seen 20th Sept., Yatton Keynell 224 (R.F.L.); 6th Oct., Rodbourne S.F. (G.L.W.). One ringed at Coate 11.9.62 found dead 8.4.63 on beach near Sentona, Santander, Spain (G.L.W.). 383. WAXWING. One at Longford, 28th-29th Jan. (I.R.). 384. GREAT GREY SHRIKE. One seen at Barton Down, 27th Apr. (R.U.). 389. STARLING. During January and February many reported dead in Swindon area (G.L.W.). In autumn and winter Starlings at Chippenham were heard to mimic the following calls: Redshank (alarm), Greater Spotted Woodpecker, Dunnock, Swallow, Little Owl, Domestic Hen (‘cluck’), Swift (scream), Yellow Hammer (song); at Castle Combe, Golden Oriole song and at Yatton Keynell the calls of Fieldfare (R.C.F.). A roost of c. 4,000 at Corsham Lake in July moved to Hawthorn where numbers in late autumn were estimated at c. 50,000. This roost was abandoned in December and flight lines were traced to 1 mile south of Badminton (R.F.L., R.C.F., J.C.R.). A roost in wood at Pitton estimated (method not stated) at half to one million birds. A very large flock of Rooks and Jackdaws previously roosting there were driven out (G.H.F.). 390. ROSE-COLOURED STARLING. On 16th Mar. one was seen to fly up on to some hurdles with a large flock of starlings feeding among sheep on downs between Mere and Kingston Deverill. An adult bird, it was watched for 5 minutes through fieldglasses at 25 yds. When the Starlings departed it followed behind but not with them (C.R.V.). An adult bird, first found by P.R. at Brinkworth, 22nd July, remained in the area until 28th and was seen by many observers. When first seen was feeding among new-mown hay and continued to feed mainly in this manner with starlings (P.R.). The tail was not abraded, which indicated long release from captivity or a genuine wild bird (G.L.W.). Both records accepted by the Rare Birds Committee of B.B. with caveat that they could be escaped birds. ‘The last Wiltshire record was in 1869. 391. HAWFINCH. In February one was picked up dying in Woodland at Hartham Park by Mr. Large, gamekeeper (G.W.H.). One seen in water meadows, Marlborough, 22and May, and 1 at Clatford, 2nd Nov. (M.C.). 392. GREENFINCH. Numerous in gardens in Swindon, Jan. and Feb. (G.L.W.). Parties noted of c. 40 feeding on spent hops, Savernake, 24th Mar. (R.F.L.); c. 50 near Corsham, 30th Mar. (J.C.R.) and ¢. 60 in Kale at Hodson, gth Nov. (G.L.W.). 393. GOLDFINCH. More numerous than usual in summer in Marlborough area (M.C.). 394. SISKIN. A least 2 with Redpolls at Coate feeding in elders and elms, 6th Apr. (G.L.W.). A flock of c. 30 at Maiden Bradley, 29th Apr., and first of winter seen there, goth Nov., ¢. 30 birds—increased to 40-50 birds by 3rd Dec. Several smaller parties about until New Year (J.C.C.O.). 395. LINNET. Scarcity noted in Swindon after winter (G.L.W.), but flocks seen later, as follows: c. 80 above Flinty Knap, 15th Apr. (B.G.); 40 Lacock G.P., 21st July (J.C.R.); at least 130, Lyneham airfield, 29th Sept. (R.W.W.); c. 350 on stubble at Lopcombe Corner, 26th Oct., and c. 200 in weedy kale at Laverstock, 16th Nov. (D.E.F., A.J.H.). An adult male ringed at Ford, 14.4.62, found dead at Bilbao, Spain, 20.4.63 (D.E.F., 397. REDPOLL. A flock of c. 30 singing and calling over suitable breeding habitat at Longleat, 27th Apr. One bird seen there 9th Oct., and 30 plus, 19th Dec. All records within 100 yds. radius and it is possible they may have remained through breeding season (N.M.D.B.). A flock of c. 8 in tops of alder and elm, Coate, 6th Apr. (G.L.W.). Seven feeding with tits on birches in garden at Seagry, 28th Dec., and 5 feeding on lawn under birches, 31st Dec. Their colouring, a warm brown with fawn, not white, wing-bars was clearly not that of the northern ‘mealy’ race (R.G.B.). 401. BULLFINCH. Apparently unaffected by cold spell and general increase noted (M.C., G.L.B., R.F.L.). Maintained numbers noted (E.H.J., J.E.M., E.E.G.L.S., G.L.W.). 404. CROssBILL. The following records all at Longleat: 11th Jan., 14 birds, 5th Sept., 20 plus, 6th Sept., 17 plus, 25th Sept., 13 plus, 3rd Oct., 8, these parties in 4 different places. On 7th Oct., 23 plus and 11 plus, a half mile apart, and on 19th Dec., 20 plus. At Shearwater, 27th Apr., 1 bird and 7th Sept., 5 plus (N.M.D.B.). At Ramsbury numbers 225 rose from 4 on 9th Feb. to 11, 16th Feb., and in Nov. and Dec. 7 birds (M.C.). At Maiden Bradley, 8th Feb., 20 birds, and rath Feb., 30. No breeding season records, but numbers varied from 12, 27th July, to 40, 3rd Aug., and c. 8 through Aug., but fewer to end of year as food supplies dwindled (J.C.C.O.). Four at Clouds, 14th Aug., and parties up to 10 around East Knoyle, ist-10th Sept. (B.M.S.). A pair feeding on pine cones at Chippenham, 16th Nov. (R.C.F.). Birds seen on Bedwyn Common throughout Nov. and Dec. (C.N.T.). A party of 6 seen, Porton, 20th Dec. (D.S.G.). 407. CHAFFINCH. Five males dead near Yatton Keynell, 15th Jan. (R.F.L.). Suffered badly in winter in Swindon area and was extremely scarce in spring and summer. There has, however, been an increase in late autumn and winter flocks appear only slightly less than normal (G.L.W.). Suffered considerable losses, Bratton (E.E.G.L.S.). An unusual type of nest in a laurel hedge near Corsham, 29th Apr., in which the usual moss cup had an external covering of grass forming a domed top with a side entrance (J.C.R.). A female ringed 11.10.55 Vosselaar: ie 19" N. 4° 53. E. (Antwerp), Belgium, recovered dead or dying 4.3.62, Devizes: 51° ON. * 59" W. (Wiltshire), cf. B.B., Lv1, 540. A female ringed 7.10. Gr, Castricum: 52° 33" N. 4° 38" E. eons Holland), Netherlands, recovered dead or dying 7.4.62, Grittleton: 51° 32’ N., 2° 12’ W. (Wiltshire), cf. B.B., LVI, 540. 408. BRAMBLING. Three feeding 2 yds. from house in a garden in Warminster, 13th-14th Jan. (N.M.D.B.). Seen in many gardens during bad weather even in the centre of Swindon, Feb. (G.L.W.). Quite common in cold spell in Savernake and stayed until and Apr. (M.C.). ¢. 20 feeding on beechmast, Bowood, when 1 male was displaying to a female, 2nd Apr. (G.L.B.). Last seen 7th Apr., Clarendon (J.C.C.O.); 12th Apr., Bowood (G.L.B.). First seen in autumn, 8th Dec., Sherston (R.C.F., J.C.R.); roth Dec., Savernake (M.C.). 409. YELLOW HAMMER. Flocks in Barbury appear normal in size and birds seem to have wintered well (G.L.W.). Normal numbers in Corsham area (J.C.R.). 410. CORN BUNTING. Six birds calling near Biddestone, 3rd Mar., remained there throughout year. Also noted near Ford and Burton (R.F.L.). ‘A single bird seen frequently near Castle Combe (R.C.F., J.C.L.). These records indicate an extension of area in north-west Wilts. (Ed.). Some seen since 1955 in fields around road junction south of Chapel Plaister (C.B.). 415. CIRL BUNTING. Two near Trowbridge, 12th Jan. (A.S.). Seen in a garden at Stratford Tony, 24th-27th May, in an area where there are also Yellowhammers (J.H.). 421, REED BUNTING. Breeding pairs at Coate slightly fewer than normal. A striking reduction in numbers in water meadows at Oakhill where normally abundant (D.A.W.A.). Suffered in Marlborough area (M.C.). Up to 3 birds visited bird table at Corsham in mid March and in early April, but did not do so in cold of Jan. and Feb. (G.W.H.). 425. TREE SPARROW. Several birds in mixed flock at Horton, 27th Jan., and at Bremhill, 13th Feb. (B.G.). c. 60 near Biddestone, 11th Mar., and 5 pairs bred at Yatton Keynell, where one pair raised 3 broods in a nest box (R.F.L.). Three pairs raised 5 broods in all in observer’s garden nestboxes, Seagry, and a flock of c. 50 seen on Seagry G.P., end Aug. (R.G.B.). The species appears to be common in Marlborough area where c. 35 at Rockley, 30th Nov. and ¢. 100 on Barton Down, 13th Dec. (M.C.). Near Barbury ¢. 50 were seen, 2nd Nov., and ¢. 200, 16th Nov. (G.L.W.). On gist Dec., ¢. 45 were seen near Ford and ¢. 30 near Castle Combe (R.F.L.). Not common at Maiden Bradley where 2 colonies of up to 6 birds are known (J.C.C.O.). The following species, though not mentioned in these notes, were recorded in 1963: Pheasant, Common Gull, Black-headed Gull, Jackdaw, Magpie, Jay, Coal Tit, Dunnock, House Sparrows. Correction to Bird Notes for 1962: c. 40 Teal, gth Dec., were on Lacock G.P. and not Corsham Lake; 78 Snipe by River Avon, Dauntsey, observed by J.C.R., not G.L.W. 226 STATUS OF CORN BUNTING IN NORTH WILTS It would appear from the report in W.A.M., Lvr (1956), by Cyril Rice, that the Corn Bunting is a non-breeding bird in North Wilts. During the last six years I have visited the northern escarpment of the Marlborough Downs frequently, and far from being absent, the Corn Bunting is a reasonably common breeding bird. From Foxhill in the east to Hackpen in the west, singing males can be heard during most of the year. At least six males in most years are present in fields below Barbury Castle and the same number at Liddington and Foxhill. At Hackpen up to twenty singing males have been counted in fields at foot of escarpment. In the valley between Preshute Down, Rockley Down and Barbury, Ogbourne Downs, numerous males sing on both slopes and valley floor. I have one isolated record of a male singing from telegraph wires at Inglesham in June. Crops in the area were cereals, but height above sea level c. 250 ft. As the original report was not a complete cover of the county, perhaps the foregoing notes will assist in extending this cover. G. L. WEBBER 227 WILTSHIRE PLANT NOTES (24) compiled by DONALD GROSE Records are for 1963 unless otherwise indicated. The numerals refer to the botanical districts as delimited in the Flora of Wiltshire. Helleborus viridis L. Green Hellebore. 2. Prickmoor Wood, D. Rice. 9. Upper Chicksgrove, G. H. Forster (261). Eranthis hyemalis (L.) Salisb. Winter Aconite. 7. River-bank, Pewsey, Miss Potter. Delphinium consolida L. Forking Larkspur. 7. The Butts, Salisbury, Dr. A. E. Walliams (261; WI). Aconitum anglicum Stapf. Monkshood. 5. Winterbourne Dauntsey, 1961, Mrs. Creighton. Papaver somniferum L. Opium Poppy. 9. Railway bank, Tisbury, B. M. Stratton. P. lateritium C. Koch. Armenian Poppy. 5. Porton Ranges, Dr. A. E. Williams, det. J. E. Lousley (261; G). Corydalis claviculata (L.) DC. Climbing Fumitory. 9. Garden weed, Milton, B. M. Stratton (G). Cardamine pratensis L. Cuckoo-flower. Form with double flowers. 1. Hilperton, H. ay. 4. Near Wootton Rivers, Major Cowan. 9. Grovely Wood, Mrs. Morgan (261). Hesperis matronalis L. Dame’s Violet. 4. Near Great Bedwyn, Major Cowan. 7. Marden, E. V. Cleverly. Sisymbrium altissimum L. Tall Rocket. 1. Westbury, R. Bennett. 7. The Butts, Salisbury, C. S. Gummer. Camelina sativa (L.) Crantz. Gold of Pleasure. 9. Garden weed, Harnham, M. A. Chaplin. Cardaria draba (L.) Desv. Hoary Cress. 3. Frequent. 7. Conock, E. V. Cleverly. 8. Warminster, Mrs. Berry. Thlaspi arvense L. Penny Cress. 1. Dilton Vale, R. Bennett. Iberis amara L. Candytuft. 5. Two localities on Porton Ranges, Dr. A. E. Williams (261). Rapistrum orientale (L.) Crantz. 7. The Butts, Salisbury, C. S. Gummer (261). The Everleigh Ashes plant recorded (257) as R. rugosum is also referable to R. orientale. Reseda luteola L. Dyer’s Rocket. 1. Westbury, R. Bennett. 9. Frequent. Viola odorata L. Sweet Violet. var. imberbis (Leight.) Henslow. 2. Monkton Park. Saponaria officinalis L. Soapwort. 3. Frequent. 6. Little Bedwyn. Cerastium semidecandrum L. Little Mouse-ear Chickweed. 8. Bare ground on clay-with-flints north of Great Ridge Wood (G). C. vulgatum L. Common Mouse-ear Chickweed. var. pentandrum Syme. 10. Witherington Down (G). Sagina apetala Ard. Annual Pearlwort. 1. Rail track, Westbury. 6. Track, Collingbourne Woods. 8. Bare ground north of Great Ridge Wood. Geranium nodosum L. Knotty Crane’s-bill. 2. Hedge-bank near Slaughterford, Miss Frowde. G. lucidum L. Shining Crane’s-bill. 10. Whitsbury (Hants), Mrs. Morgan (261). Oxalis corniculata L. Procumbent Yellow Sorrel. 7. Garden weed, Conock, FE. V. Cleverly. g. Garden weed, East Knoyle, B. M. Stratton. Impatiens glandulifera Royle. Himalayan Balsam. 5. Langley Wood, Major Cowan. Genista tinctoria L. Dyer’s Greenweed. 8. Corton Down. Ononis spinosa L. Spinous Rest Harrow. 7. Heale Hill, Dr. Hope-Simpson and R. E. Sandell. 228 Melilotus alba. Desr. White Melilot. 1. Dilton Vale and Westbury, R. Bennett. Trifolium pratense L. Red Clover. Proliferous form. 4. Manton, Major Cowan (G)!. T. arvense L. Hare’s-foot Trefoil. 1. Waste ground, Westbury, abundant, R. Benneit!. Vicia tetrasperma (L.) Schreb. Smooth Tare. 1. Waste ground, Westbury, R. Bennett!. 6. Frequent. V. dasycarpa Tenore. 7. The Butts, Salisbury, Dr. A. E. Williams (261; G). Lathyrus nissolia L. Grass Vetchling. 1. Westbury, R. Bennett. Prunus cerasifera Ehrh. Cherry Plum. 8. Wood on Norton Down, 7. Barkham. Avenue near Heytesbury Station and hedge near Stockton Rectory, B. M. Stratton. P. x insititia L. Bullace. 6. Shalbourne, Miss Frowde!. Geum rivale x urbanum. 8. North-east corner of Stockton Wood, B. M. Stratton. Rosa rubiginosa L. Sweet Briar. 10. Witherington Down. R. rugosa Thunb. 6. The rose recorded for Botley Copse as R. sherardi (262) is this alien species. Saxifraga tridactylites L. Three-fingered Saxifrage. 8. Artillery range near Redhorn Hill, E. V. Cleverly. Sedum album L. White Stonecrop. 8. Derelict army site, Codford St. Mary, B. M. Stratton. Peplis portula L. Water Purslane. 1. Derelict rail track, Westbury, R. Bennett!. Epilobium roseum Schreb. Pale Willow-herb. 2. Monkton Park (G). E. adenocaulon Hausskn. Glandular Willow-herb. 1. Westbury. Flinty Knapp, R. £. Sandell!. 8. Corton Down. Oenothera erythrosepala Borbas. Evening Primrose. 7. The Butts, Salisbury, C. S. Gummer (261). 8. Down one mile west of Imber, Major Cowan. io. Near Peter’s Finger, A. A. Dunthorn (261). Silaum silaus (L.) Schinz & Thell. Sulphurwort. 1. Westbury, R. Bennett!. Heracleum sphondylium L. Hogweed. var. angustifolium Huds. 2. Monkton Park. Erigeron canadensis L. Canadian Fleabane. 1. Waste ground, Westbury, R. Bennett (G)!. E. acris L. Blue Fleabane. 1. Waste ground, Westbury, R. Bennett!. Filago germanica L. Cudweed. 1. Waste ground, Westbury, R. Bennett. Anaphalis margaritacea (L.) Benth. Pearly Everlasting. 1. Waste ground, Westbury, R. Bennet (G)!. Ambrosia psilostachya DC. Perennial Ragweed. 7. The Butts, Salisbury, Mrs. Morgan (261). Artemisia absinthiuum L. Wormwood. 7. The Butts, Salisbury, C. S$. Gummer (261; G). Senecio viscosus L. Stinking Groundsel. 1. Waste ground, Westbury, R. Bennett (G)!. 7. The Butts, Salisbury, Miss Buckle (261). S. erucifolius L. Hoary Ragwort. 6. Frequent. S. integrifolius (L.) Clairv. Field Fleawort. 10. Frequent. Arctium lappa L. Great Burdock. 2. Frequent. Cirsium eriophorum (L.) Scop. Woolly-headed Thistle. 6. South border of Rag Copse, kK. Grinstead!. Centaurea nigra L. Lesser Knapweed. subsp. nigra. 6. Burridge Heath, Miss Frowde!. C. diluta Ait. Lesser Starthistle. 7. The Butts, Salisbury, Mrs. Morgan; Dr. A. E. Williams (261; G). Meracium maculatum Sm. Spotted Hawkweed. 1. Waste ground, Westbury, R. Bennett!. 2. Slaughterford, Miss Frowde. H. perpropinquum (Zahn) Druce. 7. Redhorn Hill. Leontodon leyssert (Wallr.) Beck. Hairy-headed Hawkbit. 1. Waste ground, Westbury (G). Taraxacum laevigatum (Willd.) DC. Lesser Dandelion. 10. Witherington Down, R. E. Sandell!. Campanuia latifolia L. Giant Bellflower. 8. Roadside near Newmead Farm, B. M. Stratton (G)!. Lysimachia vulgaris L. Common Loosestrife. 5. Near Hamptworth Lodge, Major Cowan. Buddleja davidi Franch. 1. Waste ground, Westbury, R. Bennett!. Gentianella anglica (Pugsl.) E. F. Warburg. 8. Cotley Hill. 229 Cynoglossum officinale L. Houndstongue. 4. Frequent. Nicandra physaloides Gaertn. Apple of Peru. 7. The Butts, Salisbury, Mrs. Morgan (261). Verbascum nigrum L. Black Mullein. 7. Foot of Rox Hill. hackxia elatine (L.) Dum. Fluellen. 8. Wylye, Mrs. Morgan (261). Veronica filiformis Sm. 2. Monkton Park. 9. Roadside between Dinton and Fovant, B. M. Stratton. Euphrasia nemorosa (Pers.) Wallr. Common Eyebright. 9. White Sheet Hill, Mere. Mentha x smithiana R. Graham. Red Mint. 7. The Butts, Salisbury, Mrs. Morgan (261). Acinos arvensis (Lam.) Dandy. Basil Thyme. 1. Rail track, Westbury, R. Bennett!. Calamintha ascendens Jord. Common Calamint. 2. Walled bank, Chalfield Drive, Col. C. Floyd. Chenopodium rubrum L. Red Goosefoot. 6. Frequent. Atriplex patula L. Spreading Orache. var. bracteata Westerl. 2. Monkton Park. Polygonum cuspidatum Sieb. & Zucc. Japanese Knotweed. 2. Monkton Park. 7. The Butts, Salisbury, Dr. A. FE. Walliams (261). Viscum album L. Mistletoe. 5. Between Whiteparish and Redlynch, on maple, Major Cowan. Betula pubescens Ehrh. Downy Birch. g. East Knoyle, B. M. Stratton (G). Carpinus betulus L. Hornbeam. 4. Near Puthall Gate (G). 8. Arn Hill Down, 7. Barkham. Salix viminalis x caprea. 9. East Knoyle, B. M. Stratton. Populus canescens Sm. Grey Poplar. 5. Frequent. P. tremula L. Aspen. 5. Frequent. P. x gileadensis Rouleau. Ontario Poplar. 2. Monkton Park. 9. Milton, B. M. Stratton. Sprranthes spiralis (L.) Chevall. Lady’s Tresses. 1. Valley near Littleton Down, L. F. Mead. Cephalanthera damasonium (Mill.) Druce. Large White Helleborine. 1. Frequent. Epipactis phyllanthes G. E. Smith. 9. Between West Harnham and Bemerton, C. S.. Gummer (261). Orchis morio L. Green-winged Orchis. 9. Hawking Down, L. F. Stearn. O. praetermissa Druce. Common Marsh Orchis. 8. Gow Down, Longbridge Deverill, Mrs. White (G). O. fuchsu Druce. Spotted Orchis. 8. Stony Hill, Sherrington. Gow Down, Longbridge Deverill, Mrs. White. O. fuchsii x praetermissa. 8. Cow Down, Longbridge Deverill, B. M. Stratton. Ophrys apifera Huds. Bee Orchid. g. Frequent. Gymnadenia conopsea (L.) R. Br. Fragrant Orchid. Form with a very short button-like spur. 8. Cow Down, Longbridge Deverill, Mrs. White. Platanthera bifolia (L.) L. C. Rich. Lesser Butterfly Orchid. 8. Cow Down, Longbridge Deverill, Mrs. White. Galanthus nivalis L. Snowdrop. 8. The record for Stockton (260) was an error. Allium vineale L. Crow Garlic. var. vineale. 4. Roadside near Walker’s Hill, EF. V. Cleverly. Ornithogalum umbellatum L. Star of Bethlehem. 2. Roadside near Charlcutt, H. Kay. ~ Colchicum autumnale L. Meadow Saffron. 8. Longdean Bottom, Gen. C. G. Lipscomb}. Schoenoplectus lacustris (L.) Palla. Bulrush. 1. Pond, Frogmore Road, Westbury, R. Bennett. Panicum miliaceum L. Millet. 7. The Butts, Salisbury, C. S$. Gummer (261). Echinochloa crus-galli (L.) Beauv. Cockspur. 7. The Butts, Salisbury, Mrs. Morgan (261; G). Setarta italica (L.) Beauv. Italian Millet. 7. The Butts, Salisbury, C. S. Gummer (261). Phleum nodosum L. Cat’s-tail. 9. White Sheet Hill, Mere. Agrostis gigantea Roth. Black Bent Grass. 6. Shalbourne. A. stolonifera L., var. palustris (Huds.) Farw. Marsh Bent Grass. 2. Somerford Common, R. E. Sandell!. Aira caryophyllea L. Silvery Hair Grass. 1. Waste ground, Westbury. Deschampsia caespitosa (L.) Beauv. ‘Tufted Hair Grass. var. parviflora (Thuill.) Coss. & Germ. 6. Collingbourne Wood. Helictotrichon pubescens (Huds.) Pilger. Downy Oat-grass. 9. White Sheet Hill, Mere. 230 Poa angustifolia L. Narrow-leaved Meadow-grass. 7. Railway bank, Salisbury, C. S. Gummer. Festuca arundinacea Schreb. Tall Fescue. 2. Frequent. F. pratensis Huds. Meadow Fescue. 6. Frequent. 9. White Sheet Hill, Mere. F, pratensis x Lolium multiflorum. 7. The Butts, Salisbury, C. S. Gummer (261; G). Ophioglossum vulgatum L. Adder’s Tongue. 2. Near Braydon Pond, 1962, R. E. Sandell. 4. Morgan’s Hill, W. D. Floyd. Asplenium adiantum-nigrum L. Black Spleenwort. 1. Westbury, R. Bennett!. Athyrium filix-femina (L.) Roth. Lady Fern. var. convexum Newm. 6. Near Burridge Heath, Miss Frowde (G)!. 257 Wiltshire Plant Notes—(20), 1960. WI The Herbarium of Dr. A. E. Williams, re i” Salisbury. eee woe Elant sNotes . (22) 2002. . ! Seen by the writer in the locality named. 261 Salisbury and District Natural History The use of the term ‘Frequent’ has a statistical Society Bulletins, 1963. basis and is explained fully in The Flora of Wiltshire, G_ The writer’s Herbarium. p. 76. 237 ENTOMOLOGICAL REPORT FOR 1963 by B. W. WEDDELL Looking back over 1963 one is hard put to find any bright spots at all. When the long freeze-up ended, we suffered a cold and wet spell lasting nearly to the end of May. Then, over Whitsuntide, we enjoyed a very nice fortnight, but alas, the weather reverted to ‘normal’ by mid-June, thereafter continuing, with rare breaks, thoroughly unreliable. Truly a season to write off! I saw my one and only Red Admiral on 30th September, and never a Painted Lady at all, though a Correspondent records one. ‘There were practically no migrants, at least, until late autumn when some counties as far north as the Borders were visited by Silver Striped Hawks. Apparently they missed Wiltshire. In such a year one wonders how and when many insects get the chance of a mating flight, but obviously they do manage to propagate the species, and this in spite of additional hazards from tidying hedgerows and, from often irresponsible use of toxic weedkillers. On the rare occasions when conditions for observation were good, it was seen that many species were still abundant. Such an occasion was experienced by those of us who foregathered round a mercury vapour lamp on 2oth July. Our hosts were Mr. and Mrs. David (Mrs. David is our old friend Christine Richards) at Berwick St. James, to whom we record our grateful thanks. It turned out to be the perfect night of the year, and no fewer than 76 species of Macros were logged. If Micros had been counted, this figure would have been doubled easily. The big fritillaries seem to have been well up to strength in the woodlands and on the downs of mid-Wiltshire, but the Blues have had a very hard time through unkind weather plus intensive agriculture. Some colonies of Chalkhill Blues have been wiped out by overgrazing. ‘The second brood of the Adonis Blue proved quite blank, and one wonders if the pupae from the first brood were drowned out, or may be lying over till next May. One hopes and prays that the latter is the truth, otherwise their future looks grim. It is hoped that the work of the Wiltshire Trust for Nature Conservancy will, in time, have its effect on the present trend, and prevent the extinction of many of our treasures of flora and fauna which have flourished in our countryside for so many centuries. Contributors M.C. Marlborough College N.H.S. C.F. Lt.-Col. Charles Floyd, 0.3.£., Holt. B.W. Mr. B. W. Weddell, Trowbridge. R.A. J. Capt. R. A. Jackson, c.B.E., R.N.(Retd.), F.R.E.s., Codford. C. Gi. Major-Gen. C. G. Lipscomb, c.B., D.s.o., Crockerton. C.M.R.P. Mr. C. M. R. Pitman, Salisbury. pF .C. Salisbury Field Club. R.W. Mr. Ralph Whitlock, Pitton. G.W.H. Commander G. W. Harper, R.N.(Retd.), visiting Tilshead. C.deW. Dr. Charles de Worms, M.A., F.R.E.S., visiting Salisbury Area. R.C.D. Mr. R. C. Dyson, F.R.£.s., visitng Heytesbury. M.J.L. Mr. M. J. Leech, visiting Codford Area. J.N. Mr. J. Newton, B.sc., F.R.E.S., visiting Tilshead Area. M.E.T. Mr. M. E. ‘Tyte, Hawthorn. 232 PHENOLOGICAL REPORT Average date Large White 25°4 Marbled White 25°6 Meadow Brown 15°6 Cinnabar 18°5 Garden Carpet 27-4 Brimstone Moth 14°5 Orange-tip Clouded Yellow Small Pearl-bordered Fritillary Large Tortoiseshell Painted Lady Duke of Burgundy Green Hairstreak Silver-studded Blue Adonis Blue Holly Blue Lime Hawk Poplar Kitten Sallow Kitten Puss Lobster Large Chocolate-tip Peach Blossom Muslin Ermine Wood Tiger Scarlet Tiger Green Arches Common Marbled Coronet Hedge Rustic Dusky Sallow Rustic Small Clouded Brindle Double-lobed Brindle Slender Brindle Flounced Chestnut Chamomile Shark Euchloe cardamines Coltas croceus Argynnis selene Nymphalis polychloros Vanessa cardui Hamearis lucina Callophrys rubi Plebejus argus Lysandra bellargus Celastrina argtolus Mimas tiliae Cerura hermelina Cerura furcula Cerura vinula Stauropus fagi Clostera curtula Thyatira batis Cycnia mendica Parasemia plantaginis Panaxia dominula Anaplectoides prasina Hadena conspersa Tholera cespitis Eremobia ochroleuca Apamea unanimis Apamea ophiogramma Apamea scolopacina Anchoscelis helvola Cucullia chamomillae me ae NN 1963 emergence Difference 7°4 +18 4 = 2°6 —7 4°5 = a) +13 eer eS C.M.R.P. 2.6, M.E.T. 15.6. M.C. 30.5 on Fyfield Down. Only record. S.F.C. 5.8. Partial second brood. Very unusual. C.M.R.P. 11.5. A very rare visitor to Wilts. of recent years. M.E.T. 28.7. Notable as our one and only record this year. Mi. [-Le 7:6, Gi MLR.P.3.6, C.M.R.P. 25.5, M.C. 25.5. bas 5-7. M.J.L. 8.6, R.C.D. 5.6. In good numbers. DEC. 2:6, M.C. 8.6. R.A.J. 26.5. An early date. C. 26.5. W. 20.7. Melanic. 28.5. 176: R.A.J. 31.5. Females, not usually attracted to light. RA. J. 8.6. C.M.R.P. 8.6. Early date. This species always does very well in the Avon valley south of Salis- bury. Why does it refuse to patronize the Wylye valley? M.C. 11.6 and 7.7. S.F.C. 10.6. Dark form. M.C. 8.7, B.W. 14.8. B.W. 9.8 and 15.8, J.N. 10.8. Up to 1960 this species appeared rarely and singly. Now it seems to be well established and wide- aa ee) 233 Small Purple Bars Old Lady Brindle-barred Yellow Scarce Tissue Royal Mantle Large Argent and Sable White-pinion Spotted Tawny-barred Angle Purple Thorn Small Engrailed Scarce Forester Wood or Orange Swift Beautiful Twist 234 Phytometra viridaria Morma maura Acasis viretata Calocalpe cervinalis Euphyia cuculata Eulype hastata Bapta bimaculata Semiothisa liturata, var. nigrofulvata Selenia tetralunaria Ectropis crepuscularia Procris globulariae Hepialus sylvina Eulia formosana SS ey ne er bo 82 . 20. C. 2.6. RP. 11.5, MiG s1235; ze OvwoF saz = ne) .M.R.P. 12:4. A.J. and M.J.L. 8.6. Right on cot a 3 oO M.C. 14.6. B.W. 22.7. A handsome tortrix only recently added to the British list. Remarkable that it should have spread so far west already. REPORT OF THE HON. SECRETARY FOR 1962 (PRESENTED AT THE ANNUAL GENERAL MEETING, 1963) In the past year Section membership has risen from 158 to 175, and it is particularly pleasing to note an increase in the number of young people who have joined the Section. The number of claims by the parent Society was reduced from 129 to 110, the losses being mainly due to departure from the county. Mr. J. D. Grose is chairman of a sub-committee formed to create a live Herbarium of typical Wiltshire plants in Monkton Park, Chippenham. The Hon. Treasurer, Mr. G. W. Collett, resigned after sixteen years in office, and has been replaced by Mr. A. Smith. Mr. E. C. Barnes was re-elected Chairman for a further year. Mr. B. W. Weddell has successfully launched the Macrolepidoptera List for Wiltshire, by Baron de Worms. This publication was made possible by Mr. Weddell and a group of supporters with no other financial assistance—an act of faith which has yet to be fully compensated. ‘The volume has gone to most countries in Europe, to America, and to Australia. The Hon. Meetings Secretaries, Mrs. S. Lee and Miss B. Gillam, besides producing and carrying through a programme of 33 field meetings and six indoor meetings, arranged a special programme for National Nature Week. An exhibition lasting for one week and jointly planned by Miss B. Gillam, Mrs. E. C. Barnes, Miss Stevenson, Miss G. Hellier, Messrs. R. S. Barron, Inigo Jones, B. W. Weddell, and D. D. Tucker (Bristol Avon Water Board), and the staff of the Museum, was also held at Devizes. A display of natural history literature was also kindly provided by the Wiltshire County Librarian. Other exhibitions were organized by members in Chippenham and Swindon. 239 REPORT OF THE HON. SECRETARY FOR 1963 (PRESENTED AT THE ANNUAL GENERAL MEETING, 1964) After seven years as Chairman of the Natural History Section, Mr. E. C. Barnes has now resigned, and Dr. E. A. R. Ennion has been elected to succeed him. Membership of the Section remains unchanged at 175, and the number of claims by the parent Society is now 134 as compared with 110 last year. The total membership is therefore 309, an overall increase of 24 against 1962. Mrs. 8. M. Lee, Hon. Meetings Secretary, arranged and carried through 39 meetings in 1963, and to date 18 meetings of the 1964 programme. The highlight of the autumn was the award to Mrs. E. C. Barnes of one of two 1963 Bernard ‘Tucker medals for her services to the British Trust for Ornithology, as Regional representative for Wiltshire, and for organizing various Surveys and the Swan censuses in 1955 and 1962. Mr. Geoffrey Boyle adds the following note: “The announcement of the award of one of the two 1963 Bernard ‘Tucker medals to Mrs. Ruth Barnes—the second time only that a woman has received this honour—will, without a doubt, be received with the greatest pleasure by all members of our Society. ‘For fifteen years Mrs. Barnes has been the Regional representative for Wiltshire for the British Trust for Ornithology, and has done outstanding work in organizing Trust enquiries. In addition, she has been Bird Recorder of the Natural History Section since its inception in 1946, bearing the brunt of this exacting and responsible task. ‘Mrs. Barnes does a tremendous amount of unobtrusive work on behalf of the Society, and we are more than fortunate to have someone of her great experience to help us. ‘I know everyone will wish to join me in sending Mrs. Barnes our sincerest con- gratulations and good wishes.’ 236 REPORT OF THE HON. MEETINGS SECRETARY FOR 1963 In spite of the cancellation of all the field meetings arranged for January and February, owing to the great ‘freeze-up’, 24 field meetings were held in 1963, with an average attendance, as in 1962 of 14. There was, however, a distinct drop in attendance in the second half of the year. On a number of occasions meetings were held jointly with the Melksham W.E.A., and respectively, the Bath, Salisbury and District, and Southampton Natural History Societies. We were pleased to make contact with the Southampton group, but regrettably only a few members undertook the long drive to Hatchett Pond in July to attend a most interesting meeting led by their President, Brigadier Venning. The first meeting of the year was not held until 3rd March at Cerney Wick gravel pits, but as the pits were still frozen over, Mr. C. M. Swaine, who led the party, could find little of interest except three Goosanders. Two other bird meetings were held outside the county: at Portland Bill, and at Chew and Blagdon Lakes. During May a dawn meeting was held in Great Ridge Wood for the purpose of recording bird song. After the nightingale, the first birds heard were skylarks, partridge and cuckoo. Nine roe deer were also seen. A later meeting in the Stert valley, Devizes, for evening bird song, was poorly attended owing to heavy rain. Birds were also studied at Etchilhampton, Corsham Park, and at Lacock gravel pits, where, despite a dull, cold July evening, Mr. Julian Rolls was able to show us a green sandpiper. Three botanical meetings were held during the summer. In June, a visit to Flinty Knapp, led by Miss W. Stevenson, produced a total of 47 plants, the most noteworthy being the Broad-leaved and White Helleborines, and the Fly Orchid. On this occasion inclement weather drove the party from the exposed hillside into the woods. Mr. J. D. Grose led an enjoyable expedition to Collingbourne Woods in August, when 43 species were recorded; the Great Burnett Saxifrage, which is not found elsewhere in Wiltshire was particularly interesting. The September excursion at Imber was spent studying garden survivals, and later, plant hunting on the Downs. Members met in Savernake Forest for the annual October fungus foray, when amongst the large number of species noted were the two most poisonous fungi, Fly Agaric and Death Cap. A freshwater life expedition, led by Mr. J. Field, aroused great interest, and specimens collected in Erlestoke lake were later examined under microscopes in the laboratory at Dauntsey’s School. Mr. M. J. Penistan, Forestry Commission Divisional Officer, acted as guide on two forestry walks. The first of these in Savernake Forest, enabled us to see besides the work of the Commission, some of the ancient oaks of the Forest. Among them, the Duke’s Vaunt is reputed to have been in its prime in 1540. A later meeting was held in the Springfield Estate, Cranborne Chase, when, after an enjoyable walk, a picnic lunch was taken in the garden of Mr. Rolf Gardiner, who afterwards talked about the Chase. For the geologists, two most instructive meetings were arranged by Mr. R. S. Barron, one near Devizes, the other near Marlborough, the object being in each instance to demonstrate how a geological map is constructed from relatively few exposures. Lepidopterists found conditions exactly right when a mercury vapour lamp was set 237 16 up in the garden of Mr. and Mrs. David at Berwick St. James, and resulted in a record total of 76 species identified by Mr. B. W. Weddell. A combined meeting in the Alton Barnes area was led by Miss B. Gillam, when Rampion, Saw-wort and Tuberous Thistle were amongst many interesting plants seen. During the walk, Mr. F. K. Annable, who was also with the section, enlarged upon some of the archaeological sites in the vicinity. Other meetings of general interest were held at Maiden Bradley, Porton Ranges, Bokerly Ditch, and on Scratchbury Hill near War- minster. Over Whitsuntide an expedition to Snowdonia was arranged, lasting for one week, eight members attending. Unfortunately Mr. Inigo Jones was unable to be with us as leader, but Mr. Evan Roberts, Warden of Snowdonia National Park, took charge on two expeditions—to Great Orme’s Head where plants peculiar to the district were seen, and to the Devil’s Kitchen in search of the rare Snowdon Lily which was eventually found and photographed. A further outing took us to Newborough Warren Nature Reserve in Anglesey. The Annual General Meeting of the Section took place on 11th May at Dauntsey’s School, by kind permission of the Headmaster. Exhibitions by the school on Geology and Freshwater Life were laid out for our interest after the meeting. Tea was followed by a tour of the garden. By arrangement with the Wiltshire Trust for Nature Conservation, Mr. Hurrell’s outstanding coloured films on ‘Pine Martens’ and ‘Atalanta the Seal’, were shown at the Art Centre, Swindon. Members were fortunate in being invited to attend lectures separately arranged by the Bath and Dauntsey’s School Natural History Societies. The talks were extremely varied, and covered such subjects as Snowdonia, Entomology, the Baltic Islands, and Birdwatching from a British Weather Ship. The British Trust for Ornithology organized a most instructive lecture on the subject of “Toxic Chemicals and their Effect on Bird Life’, held at Marlborough College, and attended by many members. The speaker, Mr. W. D. Campbell, formerly chairman of the B.T.O. Toxic Chemicals Committee, explained the dangers of some insecticide and seed dressings, and we hope that all who attended will actively refrain from using such preparations, and at the same time emphasize to their friends the danger to wild life consequent upon their use. National Nature Week was held from 18th to 25th May, when the main exhibition, arranged in the Museum, Devizes, was officially opened by the Mayor, Councillor Robert Kemp, on Saturday, 18th May. Miss Gillam and Miss W. Stevenson put a great deal of hard work into individual displays, with excellent results, and Miss Gillam’s own stand illustrating the care of garden birds, proved of great interest to the public. The Salisbury and District and Swindon Natural History Societies also held successful exhibi- tions in their areas, and further displays at Chippenham were augmented with slides and photographs provided by Messrs. ‘T. G. Collett and N. U. Grudgings. Further lectures by Dr. M. Burton on ‘Everyday Events in the Lives of Animals’, arranged by the Rural Studies Association at Southbroom School, and Mr. Ernest Neal on ‘Badgers’, held at Marlborough College, were well attended. During the week the Section conducted Nature Walks in various parts of the county, to which the public were invited. ‘wo of these to Fyfield Down Nature Reserve, under the leadership of Mr. Inigo Jones, Warden, attracted a number of schoolchildren. Other rambles arranged by the Salisbury and Swindon Natural History Societies, and the Richard Jefferies Society, failed, however, to attract participants other than members. We greatly regret the resignation of our Hon. Indoor Meetings Secretary, Miss Beatrice Gillam, who is now reading for a science degree at Bristol University. Our best wishes are extended to her for success in this new venture. Finally, we must express gratitude to all leaders and speakers, and to all landowners who have generously allowed section members to walk over their land. 238 ° ™ _ =~ — eoOonunvunuome CDONH DW ODO wd oom no WO Bom oem a tt) LiGh? roe Sine ES ae z aLP "+ geak oy} IOF 9UIODUT Jao oinjipusdxy jo ssaoxy Stov gI ae ‘+ puny suonvorqng | 6 ce ssunng apog ASO FTE yernjen Aimqoay | ‘+ spreojsog pur sapr[g Jo sae a sud} BIQUOD pue AIPUNG WON ais ee Es oh A oof‘1 =——__—_—_ G &g me ee oe Yoreasay [eoIsopoaeyoIy 0 0 o%%9 ae = nd 69/z961 6 Vz 3 s ie i sasuodxy Arpung 0 0 o%%9 oe ie a zg/1g61 €1 &g - es oe "+ saseyoing umnosny[ 1 40 oi “* PIOs sprvaisog pur sepig Jo. 180) O.- 65 a 7 ‘+ AQueUNODY Oe Cg on rir oe oe sosieyy yueg 6 Vv as we "+ suioqr yseo AI9g—J07eIND :plounory Ajunor) “si IAA a pOUIZESE IN, pur syoog jo 2[eg ei ‘+ sga,q AleIqry suon ) L OI co) 8 9 L -euog pure suorsstuipy unosnyy | b L of y ve sduieig. ERASED NCI) 9017 Sase}sog oz& a "+ spuopraig poxey, | o1 ir ZV oF as et: auoydaya J, :PdIIAODII XBT, BUIODUT | II ZI 6Y ae Ne ae .t ae Arerqry ae Q II ov suondriosqng 9 Visi1 ‘+ co/yy drysasquioyy ov] VY It 9 aa et a "* SoqeY 101B A, 6 S19 "+ O/W aArasaY Supping |>+ 9 a 53 me ms pat SUTT[OACLT, :ysoro}Uy yurg o1 1 MII - = ‘+ Suisqraapy pue Arou0neyg SSUIARG “SITIAA Pue yostouI0g | 1 G o0G — a ms "+ sgstutoid 03 satedoayy L¥g es a he spuopiigq | 1 GS F1z ai < A ai 1YsrT pue jon suorssruupy—diysioquis yy] Chg Ep aerccny A) oe os "+ -gouewinsuy JoyjO pur sty suoneuog sioquopy | G 6 &19'1 3 a "+ gouevInsuy "YN pur sorsiepes Goo‘! a es ‘+ suondiosqng | 6 G zZo&‘1 7 ae a oe QUIZeSRYY JO ISO F ps F pos F AWOONT aYNLIGNAdX 2961 “WAANAOAC 4816 QHGNA UVAA AHL WOA LNNOOOV AMNLIGNATXA ANV ANOONI 6961 GNV 27961 WOI ALAINOS FHL JO SLNQOODOV 239 Lig‘g1F |o Wr Lig‘gt joe premio} potsses souryeg Aq LNOODOV SLNANLSAANI Oo ViaLt 0 o 06 premioy posses souryeg Aq | 3 6 gzP Pe eSignal, dIHSUdaNAWN FATT 6 Gig oO “or 41 o g 1S§ II I1 €g2 “Pes HAYVASAY ONIGTING PIeMIOF potssed souryegq Aq o Vi Lig‘ol Fs o1 & g2SF 3S 69% : pseM10} SoU ‘+ SUOISSTUIPY soureg OF, 101910] spuog 20uajoq, 486 spuspraiq a s ae yser) sourleg Oy, MNVd SONIAVS FALSNUL SLIIM GNV LASYdWOS 240 “sqUuDUNOIIT “OD 2 NAMO GIAVd ‘S961 ‘yop ynSe saziacy ‘ssaquioy) yung ‘Jolfaq pue WspsyMouy sno Jo Jsaq ay} 0} Surprosoe 4991109 oq 0} Woy} AyNIIO pue ‘sn UIAIS UONVULIOJUT pu sYooq sy} WoT, syUNODIe sAoqe ay) poiedaid savy 9M o1 Gi 1067 4SOO 1e spreoisOog pue sapl[g jo YyI015' Sie te _ ToinseotlT, Jo spuey ul ysp’) g1 6Le a lee) tae. SEI 1 G ggg "+ qunoosoy yeudey "+ JuNODOYW [e1gUI4) ‘pry yueg spAoyT Jsjunosp yuog 0 6 Lgoit*t ———— O1 & g2G z G 6&9 9/W dryssoqurayy 2yrT (4) ‘O/V dArasay Surpying (v) :yueg SSUIARG “S]IAA PUe JOSIIUIOG /pisogaqy uo ysvr) IQ'SI'1E 1e sy :AJQTIIOG 24} 0} po}euOp UsYM onTeA JOYTEUL JO YSOd ye, squaupsoauy o1 V1 VOES ae Ig s1'1& 1 se qoang Suoq 1F/oP Oo. “41 LiI9e‘o1 :Gragoly Ployarsy pos F SLASSY | o1 G1 106baF o1 G1 600'¢s alt Tao Jea& IY} OJ IUIOIUT I9ZAO ati ETERS) JO Ssooxo ssaT 1961 ‘laquiacaqq ISI 1¥ sv aouLleEg qunos9p yojign’) "+ s4opipaayy Capuny II Cr 19h‘& 26351 i Oo oO sey os spies? 47, SALLITIGVI'T 2961 YWAMNAOAC 41818 LY SV LOAHS FONV IVA 241 BS | 1ShSF 5 Lx 19hG¥ an gS |b 2 668 : of oa ainqyipusdxy Cz sroquiayy VT : suor}diiosqng ONO OOO WT IOs Seno 0 eat eres -+ " qsoraquy 0 or’ SLE : O OI ZgI IAIISIY GS DUNGY ARM AINGEET Surping 0} palsoysuety, g Sr 2S fo - Pe \so19}U] 6 9 Vz “Ks "+ sotapung 0 OI zgI a ne + Qar89y o GI of ta sosreyO AARON Suprmg sof yUNo.0y 6 1 GP n : SUT[[IACL qUaLIND WO Jojsurty, o Lr gf 7 - suondriosqng CeUeOGe we - ansoeiep 0 o 0G ne YOICISIY, TROISOT apms) pung suonvoyqng | -oveyoary oy uoNeooTTy ‘siqiasay junossp ansasay | ; Bt ecdet ro) GAG, ee eaiente 19 & ggor Saas See Ont = suonvorqng A131I90¢ | ee Is Eker esd brs lopcteeets BL ss SISA org & 2 sopyg pur Oe oe 3 "+ onsorejen sprvoysog JO avg UO IYOIg o Sr éSr'r i oUIZUSseI 0 o OLF SYIOAA JO ArSTUTPY [eoIsopoowyory SATYs}LM, / suoyvoyqng agua |S gt 98 ae ee ae acon | O( 47 741 es ae "+ satapung 1 ADLQUT | Cane oy i “+ sogseyoumng o1 gS oe se ‘+ UOISSIUIPY | 1 AIDAQUT J wnasnyy Lf a “991 se giv =9=——______ L si 9 ag sosuadxy o Fil "* worjd0g ATOISTFT TeINIeN | ULOOY yeq pure Aroye10qe’T OMOzn ST = ae uo snjding | ? -6- 6S1 sjuatmaAoidury pue Avydstq € o1 61 is : sque Sunasnyy -UIAOT) UO I[GLIZAOIIY XT, | @ 61 €S9'% = ——_____ 0 o oS6%1 ——— Om ee juowdmby jo uonermoidaq : oof1F Fg/Eg61 € QI ol : “s SO1CY IIL AA o0So7F €9/zg61 :yUReID or L 99 a es "+ goueinsuy edroatere) peed) DITYSITAA, 6 o LGz Ae "+ yeozy pure 1ysrT Q. Q1e9 ‘+ sorapung Oe SOV pe tekst eet sireday vo: SY Iv - a, ep, Jo Way G o1 of a ‘+ guoydoajay, OI 0 gol oe: soreyg snuog jo avg | & V1 SOI at Sunulig pue oser 6 61 9Sg "* 4sarojuy pue spusplalq -rey ‘ArouoTIeIG §‘aseISOg 1 £ ¥o0%1 suoneuog pure suondriosqns g GSI 6z0c%% °° “ “+ soraepeg x !4grasey yossuayy Ag / Sat DY) JUaULYSIGDIST OT, + Ton Da hes: PP PSs Ii ie inet Beater = ANOON] FUNLIGNAdX | £961 “YadWaOAC 1818 CAGNA UVIA AHL YOA LNNOOOV FANLIGNAdXA GNV ANOONI O- Fi Lis oLy o 1 Lig‘o1 pis F or 4 9057 g o1 bes F g O1 Vzo'1 | o Fi LigorFs | | o V1 Lig‘o1 xi ibe ve 6961 premio} potareo souryeg Ag | ‘Arenuef $1 ye se souRleg OT, | ps F LNNODOV SLNANLSAANI ov £ 0057 or £ 90S a o F & oe e "+ 4sazaquy 0 o Gz ie bs uondriosqng or §& gzS ay “ €961 paleMJO}J patssies soueleg Ag ‘Arenuef 4sI 1e se souRleg OF, pis ¥ “ps F dIHSaddqWNaWN FATT @ Ol Vos 17 g o1 boost = —————— g Gr 2S e ae 2 18310107) O OI ZgI es - "+ QATOSIY Suipimg «oy JUNODY JUSTIN WON poraysuesy, 0 0 OoG€ a; anso[eyet) apm z G 6&9 be oe 6961 paeMio} potsses soueyeg Aq | ‘Arenuef{ jsI ye se s0ueleg OT, ‘p "S wi ‘Pp “g i INNOODDV DTAUdSAa MNVd SONIAVS FALSNUL SLTIIM GNV LaSaaNos 243 OI OI Li GI gI FI VI °S “SJUDJUNDIIP pasajsDYy") CoE beF ¢ 103 Gi = Sil of 60z o6L‘ o1 £ g OI Lig‘9l ou) Oo FI S9 o G I OI TOES F “p.s ‘ANAVd 8% TIVONVY ‘Poor “oy ys “uDysyjayy “Spooy vgs 1 “Eg61 S1aquia.9q 3sI1& 1e se AjaIN0g AIOISIFY TeIMIeNY pue [eorsopoavyoIY ITYSIIM 24} JO sureye Jo uortsod on1j ay smoys JooYg soUL[eg YONs uOTVUTIOJUT INO Jo 1S9q 94} 0} Surps1oooe puv sIoYINOA pue syood sy} WoI J99Yg soURTeg aAOqe oY} poredaid savy IAA pUuoy ui ysvy) qunosoy eiuder junosoy [e19ue+) tal i sazmag “pry yuvg spdoyT | #80) 1D aE pup sapis fo y20]5 s4ojgaq, dapuns | 99S “+ qunosoy drysisquriayyy ofr'T Vza‘1 : "+ JunosIY dArasaY Lyupg ssurvg *sq7uA4 puv yasaauos jisodaq) uo ysvry 00S yueg ssutaeg 90}sni J, ‘SHIM, PUR JesIOUTOg Aq PPPH LiS‘g1 ° "+ (08S%aaF ‘bob bs ye onjea JoyxIeW) "PYT yueg spAoyy Aq PlPH 1 4a1905' ayy 0} pajpuop uayn ANJD A JAYAVJAT 40 JSO') JV Spuausjsaauy L os "* uornernoidag ssaT ol 180) qo quaudinby dapuny €961 ‘Arenuef 4s1 ye se ooNg¢ SuoT 17/ov /Gaagolg pjoyaady Ee SLASSY Ol 2B Li GoSteF 9 618 juNODOY [e19ua4) :sazmag “pry yuvg spdoyT €1 L6 "+ saoqypasy Capuns QI g06‘Ez voz 668 aimjipuodxy 49AO SUIODUT JO SS2OxXe PPV o1 S1 600°€z% 6961 ‘Arenue[ jst 1e se oourleg /qunos9p p011Gv/) 3s oF ps F SALLIMAVIT 6961 “YaaINTOU 1818 LY SV LAUAHS AONVIVA 244 NOTES FOR THE GUIDANCE OF CONTRIBUTORS TO THE WILTSHIRE ARCHAEOLOGICAL MAGAZINE The Editorial Board will be glad to consider for publication articles and notes on the history, architecture, folk-lore, archaeology and natural history of the county. PREPARATION OF COPY Contributions should be typewritten and double-spaced on one side of the paper only, preferably on sheets of quarto size. Margins of not less than 1 inch should be left at the top and left-hand side of each sheet. Clean copy helps to eliminate printing errors and the need for costly corrections at proof stage. The approximate number of words should be indicated on the title page. References to footnotes should be numbered consecutively throughout the text, and the footnotes themselves, correspondingly numbered and double-spaced, should be added on separate sheets at the end. References to books and to articles in journals should take the following forms: Stone, J. F. S., Wessex before the Celts (1958), 68. Thomson, T. R., and Sandell, R. E., The Saxon Land Charters of Wiltshire, Wiltshire Archaeological Magazine, 58 (1963), 442-46. Titles of journals should be given in full; in order to ensure uniformity of practice, abbreviations will be made by the editors. Titles and sub-titles in the main text should be typed in ordinary upper and lower case; they should not be wholly in upper case or underlined. Underlining is to be used only for words in the text that are to be printed in italics. Abbreviations should be avoided in the text. Thus, compass points should be written out as ‘north’, ‘south-east’, etc., and such words as ‘per cent.’, ‘tenth’, should not be shortened, except where they have frequently to be used in subsidiary lists or appendices. ILLUSTRATIONS The space available for line-drawings and half-tone plates is 5} inches wide by 74 inches long. In order to fill this space economically and to ensure that illustrations can be arranged to present a pleasing appearance, drawings and photographs should be propor- tioned accordingly, bearing in mind the need to insert captions beneath. ‘Thus a drawing that is to be reproduced at a scale of one-half of the original size should be 10} inches wide and not more than 144 inches long. Very large drawings (i.e. over 21 by 30 inches, and requiring reduction to less than one-quarter of the original size) cannot be accepted. Folding plans and section-drawings, which add considerably to the cost of publication, should be avoided where possible. Particular attention must be paid to the lettering on line-drawings to ensure that it is large enough to remain legible after reduction. Drawings which do not conform to these standards may be returned to contributors. Two-colour drawings (i.e. black over-printed with another colour) can be accepted only in exceptional circumstances, and the editors should be consulted before such drawings are prepared. Text figures must be numbered consecutively in the text, and these numbers, together with a brief identification, must be written in pencil on the margins of the drawings. Authors will be expected to make a case for the inclusion of half-tone plates. Photographs submitted for this purpose must be prints of good quality on glossy paper of double weight, and should preferably be capable of some reduction. Prints must be 245 lettered serially A, B, C, etc., and the corresponding identification inserted in pencil in the appropriate places in the text. Care must be taken not to mar prints when writing on the backs. A list of line-drawings and photographs, together with the captions, should be typed on a separate sheet of paper. USE OF COPYRIGHT MATERIAL If extracts from the Ordnance Survey maps are to be used, the author must obtain, and produce to the editors, the authority of the Controller of H.M. Stationery Office. Permission must also be obtained by the author for the use of any photographs or other material covered by copyright. PROOFS The editors will agree the final draft of an article with the author. Thereafter, alterations and additions will be permitted in the most exceptional circumstances only. In no case can alterations be made when the article is in page proof. A galley proof will be supplied to authors, who should correct and return it promptly to the editors. A proof is for checking only, and not for making further insertions or large- scale alterations. Corrections must be confined to printer’s errors and deviations from the agreed script. The editors will not be responsible for errors arising from untidy and illegible corrections, and authors should follow the system set out in the B.S.L. Table of Symbols for Printers’ and Authors’ Proof Corrections (British Standards Institution, 1219: 1958; 2s. 6d., obtainable from H.M. Stationery Office). An extract may be obtained from the British Federation of Master Printers, 11 Bedford Row, London, W.C.1. OFFPRINTS Twenty-five offprints of major articles will be supplied free to authors. Orders for additional copies (to be charged to authors) must be placed with the editors when the final version of an article is agreed. Offprints of minor notes are not supplied. The Editorial Board disclaims all responsibility for the opinions expressed by contributors. 246 INDEX TO VOLUME 59 Accessions: to the Library, 196; to the Museum, 205 Algar, D. J., 188 All Cannings: Rybury Camp, 185 Amesbury: Earl’s Farm Down, barrow on, 30 Annable, F. K.: reports on Romano-British pottery from Earl’s Farm Down, 39, and Overton Down, 79, 182, 187, 202 Avebury: excavations, 1960, 28; Northern Inner Circle, 181 Balfour-Browne, F. L.: report on charcoal from Robin Hood’s Ball, 23 ‘Balle’s Place’, Salisbury, 155 Barnes, Ruth G., 212 Benson’s Folly, 183 Bilbury Rings, 186 Black Field: see Mildenhall Bonney, D. J., 185. See also Reviews Bonney, Helen, 155 Boyle, Geoffrey L., 212 Bridgman, C. J., 212 Britford: font in St. Peter’s Church, 168 Bronze Age sites and excavations: barrow at Robin Hood’s Ball, 13; barrow on Earl’s Farm Down, 30; barrows near Clatford, 181; barrows in Savernake Forest, 182; barrows sited near springs, 182; ? ditch, on Totter- down, 186; secondary interments at Wood- ford long barrow, 184 Brothwell, D. R.: report on skeletal material from Overton Down, 82 Cain, A. J.: report on molluscan fauna from Overton Down, 84 Carbonell, Virginia: dental report on skeletons from Winterbourne Gunner, 107 Clatford, barrows near, 181 Christie, Patricia M., 30 Clark, A. J., 187. See also Reviews Cornwall, I. W.: report on soil samples from Robin Hood’s Ball, 23; report on cremation from Earl’s Farm Down, 43 Cricklade: excavation at Parsonage Farm, 188 Crowfoot, Elisabeth: report on textile remains from Winterbourne Gunner, 108 Cunetio: see Mildenhall Davey, N., 116 Downton, 124 Dunning, G. C.: report on marble mortar from Old Sarum, 142 Dunthorn, A. A., 207 Earl’s Farm Down, 30 Ennion, E. A. R., 212 Errington, F. P., 207 Excavation and field work in Wiltshire, 1963, 185 Fowler, E.: report on Roman brooch from Earl’s Farm Down, 39 Fowler, P. J., 46, 181, 185, 190. See also Reviews Frustfield: Saxon boundaries, 110 Fyfield Down: excavations, 185 Gomeldon: excavation of deserted medieval village, 188 Grose, Donald, 228 Grose, J. D., 58 Harper, Rev. S.: see Obituaries Helbaek, H.: report on grain impressions from Robin Hood’s Ball, 19 Hodges, H. W. M.: report on pottery from Robin Hood’s Ball, 19 Hosier, A. J.: see Obituaries Howard, M. M.: report on animal bones from Robin Hood’s Ball, 20 Idmiston Down: Benson’s Folly, 183 Iron Age sites and excavations: cross-dykes on Ebble-Nadder ridge, 51; Rybury Camp, 185; Bilbury Rings, 186; settlement on Overton Down, 186 Jennings, R. A. U., 211 Jessup, R. F.: report on metal objects from Overton Down, 77 Jewell, P. A.: report on small vertebrates from Earl’s Farm Down, 44 Kerney, M. P.: report on mollusca from Earl’s Farm Down, 43 Kingsbury Square, Wilton, 189 Knipe family, 170 24] Medieval sites and excavations: Downton, 124; Old Sarum, 130; “Balle’s Place’, Salisbury, 155; Raddun (Wroughton Copse), 186; Gomeldon, 188; Wardour, 189; Wick Farm, West Tisbury, 190 Melville, R. V.: report on rocks from Robin Hood’s Ball, 20 Meyrick, O., 181-184. See also Reviews Middleton, T. W. B.: see Obituaries Mildenhall (Cunetio), excavation, 187 Morse, Ven. C. W., 168 Musty, John, 86, 130, 183, 188 Neolithic sites, excavations, finds: Avebury, 28, 181; pottery from Earl’s Farm Down, 37; pit on Overton Down, 82; stone axe from Hamshill Ditches, 181; Robin Hood’s Ball, 1; Rybury Camp excavation, 185; Woodford long Barrow excavation, 185 Obituaries: Rev. S. Harper, A. J. Hosier, T. W. B. Middleton, E. R. Pole, G. Smith, IgI Old Sarum, 130 Overton Down: Roman tombs, 68; prehistoric pit, 82; excavation and field work, 185. See also Reviews Parsonage Farm, Cricklade: excavation, 188 Parsonage Farm, Winsley: Romano-British interments, 182 Piggott, Stuart, 28 Pole, E. R.: see Obituaries Potterne: pre-Conquest church and baptistery, 116 Powers, Rosemary: reports on skeletal remains from Overton Down, 81, and Winterbourne Gunner, 105 Prehistoric plant remains in Wiltshire, 58 Raddun: medieval settlement, 186 Radford, C. A. R., 188 Rahtz, Philip A., 124, 130 Rescue and research work in the Salisbury area, 188 Reviews (the reviewer’s name in brackets): Prehistoric England, by J. G. D. Clark [D. J. Bonney], 192; The Prehistoric Ridge Way, by Patrick Crampton [O. Meyrick], 192; The Experimental Earth- work on Overton Down, Wiltshire, 1960, ed. P. A. Jewell [Anthony Clark], 192; The Scientist and Archaeology, ed. Edward Pyddoke [P. J. Fowler], 194; 248 The Families of Allnutt and Allnatt, by Arthur H. Noble [K. H. Rogers], 195; Staying with the Aunts, by Ida Gandy [E. H. Steele], 195 Robin Hood’s Ball, Shrewton, 1 Rockley Down White Horse, 183 Rogers, K. H.: see Reviews Romano-British sites and excavations: brooch and pottery from Earl’s Farm Down, 39; tombs on Overton Down, 68; roads at Old Sarum, 134; interments at Parsonage Farm, Winsley, 182; field system on Totter- down, 186; excavations at Mildenhall (Cunetio), 187; excavation at Wellhead, Westbury, 187 Rybury Camp: excavation, 185 Salisbury: ‘Balle’s Place’, 155; cathedral font, 168; field work in area, 189; Museum Research Committee, 188 Sandell, R. E., 58, 196 Saunders, Margaret: report on Saxon brooches from Winterbourne Gunner, 100 Savernake Forest, barrows in, 182 Saxon sites and excavations: Cricklade excava- tion, 188; Downton excavation, 124; Frust- field, boundaries of, 110; Overton Down, secondary interments, 73, 76; Potterne, church and _ baptistery, 116; Kingsbury Square, Wilton, excavation, 189; Winter- bourne Gunner, cemetery, 86 Semley, Knipes of, 170 Shaw, W. D., 187 Shortt, H. de S., 168 Shrewton: Robin Hood’s Ball, 1 Simpson, D. D. A., 68 Smith, G.: see Obituaries Smith, I. F., 68, 181 Steele, E. H., 186, 200. See also Reviews Stratton, J. E. D., 86, 135 Taylor, C. C., 110 Thomas, Nicholas, 1 Tisbury: Wardour medieval settlement, 189 Tisbury (West): Wick Farm medieval settle- ment, 190 Vatcher, Major and Mrs. H. F. W. L., 185 Wallis, F. S.: report on shell, etc., in pottery from Robin Hood’s Ball, 27 Wardour: see Tisbury Webber, G. L., 227 Weddell, B. W., 232 Westbury: Wellhead excavation, 187 Westley, B.: reports on animal bones from Earl’s Farm Down, 45, and from Old Sarum, 152 West Tisbury: see Tisbury (West) Wick Farm: see Tisbury (West) Williams, J. Anthony, 170 Wilson, D. M.: report on chalice and paten from Old Sarum, 142 Wilton: Kingsbury Square, 189 Winsley: Romano-British interments at Par- sonage Farm, 182 Winterbourne Gunner: Saxon cemetery, 86 Woodford: long barrow, 185 Wylye: Bilbury Rings, 186 249 Fic. Fic. Fic. Fic. Fic. oP OO ND no PO NHN LIST OF ILLUSTRATIONS IN VOLUME 59 LINE DRAWINGS Tue NEOLITHIC CAUSEWAYED CAmp AT Rosin Hoop’s Batt, SHREWTON Locality map = = = = 2 - P - 2 Plan - = = = = = = : Z = é Sections - - - = = : = = - : Pottery - = - = é : e = - Flint implements - - = = = - > 2 : EXCAVATIONS AT AVEBURY, 1960 Plan showing areas excavated - - - - - . A Bronze AGE Rounp BARROW ON EARL’s FARM Down, AMESBURY Location map - - - - “ S Z . = Plan of barrow - - - 2 5 7 3 J : Sections of barrow - - - - - - between pp Section through central post-hole - - - - 2 ‘ Diagrammatic section through cremation pit - - - - Finds from the barrow - - : : - Z : Cross-DyKEs ON THE EBBLE-NADDER RIDGE Map showing dykes and settlements —- - - between pp EXCAVATION OF THREE ROMAN 'TomMBS AND A PREHISTORIC Pir NI DMO HP WO ND ON OvERTON Down Location map - - - = : £ : : Plan of site G.7. = - - - : : d = : Plan of site G.6a_ - = = = z = z = - Sections of sites G.6a and G.7 - - = = = 3 Plan and sections of site G.6 - = = S - = : Metal objects from sites G.7 and G.6a_ - - - - - Prehistoric pottery from pit and mounds - - = = Page 29 30 32 - 34°35 35 36 Sif . 48-49 69 71 72 74 75 78 83 251 A SAxon CEMETERY AT WINTERBOURNE GUNNER, NEAR SALISBURY Page Fics. 1-3 Location maps - - - - - - - - - 87 4. Plan of site - - - - - - - - - - 88 5 Objects from Graves I, IV and V - - - - - - go 6 Objects from Graves VI and VII - - - - - - 92 7 Objects from Grave VIII ~ - - - - - - - 04 8 Objects from Graves IX and X_ - - - - - - 96 g Plans of Graves 1, 1V, V, VII, TX and X_ - - - - 98 10 Plans of Graves VI and VIII - - - - - - 100 THE SAxoN BOUNDARIES OF FRUSTFIELD Fic. 1 Bounds of the three hides at Frustfield - - - - = 2 2 A PrE-ConQuEst CHURCH AND BAPTISTERY AT POTTERNE Fic. 1 Site of the pre-Conquest church and its relation to adjoining tithe plots - - - - - - - - my Vly Plan of the pre-Conquest church and stages of development - 118 Plan of foundation matrix cut in Greensand - - - - 120 4 Arrangement of foundation pads and posts - - - eee iol. SAXON AND MeEprEvVAL FEATURES AT DOWNTON Fic. 1 Section of Saxon gravel pit - - - - - - - 126 2 Section of trial hole at Old Court - - - - - = [2126 3 Saxon pottery - - - - - - - - - 126 THE SUBURBS OF OLD SARUM Fic. 1 General plan of excavations and features - - - - 133 2 Detailed plan of excavations and features - - between pp. 136-137 3 Section of Road C - - - - - - between pp. 136-137 4 Plan of medieval building - - - - - between pp. 136-137 5 Objects of stone, metal and bone - - - - - - 14.4 6 Early medieval pottery from Pit 1 - - - - - - 148 7 Medieval pottery - - - - - - - - - I5I ‘BALLE’s PLACE’, SALISBURY Fic. 1 Details of hammer-beam roof in No. 27 Winchester Street - 157 2 Sketch plan of the site - - - - - - - = 163 Roap CAsuALTIES AMONG MAMMALS, REPTILES AND AMPHIBIANS IN WILTSHIRE Fic. 1 Histogram showing seasonal variations in casualties - - 209 252 PLATES (at end of volume) I. Air photograph of the Neolithic causewayed camp at Robin Hood’s Ball II. Potsherds illustrating the chief wares at Robin Hood’s Ball III. Round barrow on Earl’s Farm Down, Amesbury IV. Cross-dykes near Chiselbury V. Roman tomb on Overton Down, site No. 7 VI. Roman tombs on Overton Down, sites Nos. 6a and 6 VII. Pre-Conquest church and baptistery at Potterne VIII. Saxon pottery from Downton IX. Chalice from Old Sarum X. ‘Balle’s Place’, Salisbury XI. ‘Balle’s Place’, Salisbury XII. Fonts from Salisbury cathedral and St. Peter’s, Britford XIII. Fonts from St. Peter’s, Britford, Salisbury cathedral, and West Knoyle XIV. Rockley Down white horse 255 17a PLATE I Air photograph of the Neolithic causewayed camp at Robin Hood’s Ball. Crown Copyright reserved By courtesy of the Cambridge University Press PLATE II CENTIMETRES —~ Say 3 ~~ a rry cP Potsherds illustrating the chief wares at Robin Hood’s Ball. 1-2: burnished vessel of West Country ware (p. 19; FIG. 4: 17-18; Hodges No. 70) ; 3-4: oolite-gritted ware (p. 19; Hodges No. 73); 5: flint-gritted ware; 6: shell-gritted ware (p. 15; FIG. 4: 3). PLATE III a. Remains of chalk ‘ring’ on north-east. b. Turf stack, detail. c. Central post-hole, partly sectioned. d. Tool marks on wall of cremation pit. ROUND BARROW ON EARL’S FARM DOWN, AMESBURY PLATE IV Cross-dykes. The atypical south end of D.75, running south from Chiselbury, which is just visible at top right. Many dykes end about the break of slope, where this one is blocked before re-emerging and dropping almost to the combe bottom. Beyond, bank between two ditches, associated with 18th-century turnpike. From the east. Photograph by P. J. Fowler PLATE V a. Site No. 7 after excavation. b. Site No. 7: detail of post-sockets in ditch. ROMAN TOMBS ON OVERTON DOWN PLATE VI a. Site No. 6a after excavation; in lower left corner the prehistoric pit with fragment of sarsen in situ. b. Site No. 6 prior to excavation of ditch. ROMAN TOMBS ON OVERTON DOWN PLATE VII a. Slot cut in bench of greensand for the east wall of the baptistery. b. Site of the font in the baptistery. c and d. General views of the site, showing shallow slots and holes in PRE-CONQUEST CHURCH AND BAPTISTERY AT POTTERNE the greensand. PLATE VIII inch Saxon pottery from Downton. PLATE IX Old Sarum. Drawing and photograph of the chalice from Akerman’s excavation. Scale of drawing: 1:1. PLATE X b a. The site, looking north-east; Nos. 25-29 Winchester Street. b. Hammer-beam truss in No. 27 Winchester Street. Lower north half, looking east. Photographs by Royal Commission on Historical Monuments } (England), Crown Copyright reserved “BALLE’S PLACE’, SALISBURY P2AkIasok JYBIMJOD Ugio4Ly -(punjszusy) NAD ce Td S.a S/U2UNUGTY JDILIOISIFY UO Uorssiumoy yokoy Cq sydvss0j0yq SSENAIEIS ORAS SION ESE 4SOM SULOO'T +3999 Iajsoyour\A 63 “oN ur Joor ysod-uMmoary -q ysom Suryooy ‘yey raddy ‘39011g JojsoyourAA Lo “ON UI ssnajz Ueaq-ioUlLUeyy “ve noi ae PLATE XI PLATE XII at LON TAA SALISBURY a TED eed Ce a. Font from Salisbury Cathedral, now in Christ Church, Yankalilla. b. Font from St. Peter’s, Britford, now in St. James’s, Delamere. c. Engraving of the Cathedral font by James Biddlecombe. PLATE XIII a. Font now in St. Peter’s, Britford. b. Font now in Salisbury Cathedral. c. Classical font, West Knoyle. d. Victorian Gothic font, West Knoyle. parsasat VyBLiCdor) ydasof "IS "MY o£ Rats oP iM ‘asaoy ayiyM uMOCT A © I[ PIO, AIX ALV'Id The Bronze Age Round Barrow in Britain PAUL ASHBEE The ‘British Bronze Age barrow-digger’s handbook’ (Man), ‘a standard work for many years to come’ (The Times Literary Supplement). By the Secretary to the Council for British Archaeology’s Neolithic and Bronze Age Research Committee. 32 pp. plates; backed frontispiece; 61 figs in text. 50s Beginning in Archaeology KATHLEEN M. KENYON Initiates the beginner into recording of digs and the interpretation, and preservation of finds. Photos and diagrams lead the beginner from fundamentals to specialized methods. ‘.. . full of reliable information’ (The Times Literary Supplement). Aldine Paperback 7s 6d; cloth 18s The Scientist and Archaeology EDWARD PYDDOKE Nine experts demonstrate how specialist disciplines, such as soil-science, microscopy, carbon dating, artefact analyses, or resistivity surveying, augment traditional methods. ‘.. . a good, solid, factual account’ (Discovery). With 32 text illustrations and 24 pp. plates. 244 pp. 30s Stratification for the Archaeologist EDWARD PYDDOKE Describes the great variety of deposits in which antiquities are discovered or the processes which lead to their becoming buried and preserved. Foreword by S. W. Wooldridge. 9 pp. photos, 18 line illustrations. 30s The Runes of Sweden SVEN B. F. JANSSON The world’s foremost runologue casts a fascinating light on Norse culture. Translated by Professor P. G. Foote. 76 photos, taken for the Swedish edition published by the Royal Academy, Stockholm. 25s From booksellers Complete list of books on Archaeology on request to Aldine House Published from Aldine House, Bedford Street, London W.C.2 Established over 100 Years F. RENDELL & SONS — LIMITED — DEVIZES WILTS BUILDING AND CIVIL ENGINEERING CONTRACTORS HEATING AND ELECTRICAL ENGINEERS Telephone: Devizes 157 (4 lines) The Magazine The Magazine is at present issued once a year. It is issued free to members of the Society. Contributions, editorial correspondence and books for review should be sent to the Editor at The Museum, 41 Long Street, Devizes. Back numbers of Magazines can be obtained from the Honorary Librarian, with the other publications listed below. Publications to be obtained from the Librarian, The Museum, Devizes A GuIDE CATALOGUE OF THE NEOLITHIC AND BRONZE AGE COLLECTIONS AT DEviIzEs Museum, by F. K. Annable and D. D. A. Simpson. 1964. Post free, 26s. 6d. A BIBLIOGRAPHY OF THE GREAT STONE MONUMENTS OF WILTSHIRE: STONEHENGE AND Avesury, by W. J. Harrison. No. LXX XIX (1901) of W.A.M. 5s. 6d. WILTSHIRE INQuisiTIONES Post Mortem: Henry III, Epwarp I anp Epwarp II. 13s. Dirto. Epwarp III. 133s. THe CyurcH Betts oF WILTSHIRE, THEIR INSCRIPTIONS AND History, by H. B. Walters. Parts II and III only. 14s. RECORDS OF THE County oF WiILTs, EXTRACTS FROM THE QUARTER SESSIONS GREAT ROLLs OF THE 17TH CENTURY, by B. H. Cunnington. 12s. 6d. Devizes BoroucH ANNALS, EXTRACTS FROM THE CORPORATION ReEcorps, by B. H. Cunnington. Vol. II, 1792 to 1835. 15s. (Vol. I is out of print.) THe WILTSHIRE ARCHAEOLOGICAL AND NATURAL History Society, 1853-1953. A Cen- tenary History. 3s. 6d. BRADON Forest, by T. R. Thomson. 1953. tos. 6d. _ Tue Fiora or WIxTsuireE, by Donald Grose. Pp. iv, 824. 1957. 428. _ Wictsuire Birps, by L. G. Peirson. 4s. THE MACROLEPIDOPTERA OF WILTSHIRE, by Baron de Worms. Pp. xv, 177. 1962. 25s. (Members 20s.) PARLIAMENTARY History OF CRICKLADE, by W. B. Crouch. 1961. 12s. A History oF SAVERNAKE Forest, by The Marquess of Ailesbury. 1962. 12s. 6d. The Society’s Museum and Library, Long Street, Devizes The Curator is Mr. F. K. Annable; the Honorary Librarian, Mr. R. E. Sandell All members of the Society are asked to give an annual subscription towards the upkeep of the Museum and Library collections. The Museum contains many objects of great local interest; and the Library a rich collection of books, articles and notes about the history of Wiltshire. Subscriptions should be sent to The Treasurer, The Museum, Devizes. Old printed material and photographs of Wiltshire buildings or other objects of interest will be welcomed by the Librarian at the Museum. The repository for records, e.g. old deeds, maps, plans, etc., is the Wiltshire Record Office, County Hall, Trowbridge. co The Records Branch Founded in 1937 for the publication of original documents relating to the histo the County. The subscription is £2 yearly. New members are urgently needed. Honor: Secretary, Miss T. E. Vernon, Dyer’s Leaze, Lacock, Wilts. The Branch has issued the following. Prices on ‘application. I ApsTRAcTs OF FEET OF Fines RELATING TO WILTSHIRE FOR THE REIGNS io) Epwarp I anp Epwarp II. Edited by R. B. Pugh. 1939. Out of print. Il ACCOUNTS OF THE PARLIAMENTARY GARRISON OF GREAT CHALFIELD AND Bee BURY, 1645 To 1646. Edited by J. H. P. Pafford. Out of print. Ill CALENDAR or ANTROBUS DEEDS BEFORE 1625. Edited by R. B. Pugh. 1947. IV WILTsHIRE County Recorps. MinuTEs OF PROCEEDINGS IN SESSIONS 1563 AND 157 .TO 1592. Edited by H. C. Johnson. 19409. Vv List oF WitTsHirE BoroucH Recorps HaRLieR In DATE THAN 1836. Edited b Maurice G. Rathbone. With an introduction by R. B. Pugh. 1951. VI Tue Trowsrmce Wooiien Inpusrry as ILLusrRATED BY THE Stock Boo oF JoHN AND THomas CLARK, 1804 To 1824. Edited by R. P. Beckinsale. 1951. VII Guitp STEWARDS’ Book or THE BorouGH oF CALNE, 1561 To 1688. Edited b A. W. Mabbs. 1953. ; VIII AnpRews’ anD Dury’s Map or Witrsuire, 1773. A reduced facsimile. Wit a an introduction by Elizabeth Crittall. 1952. 7 Ix SURVEYS OF THE Manors oF Puinip, Firsr HARL OF PEMBROKE AND MONTGOMER’ 1631 To 1632. Edited by Eric Kerridge. 1953. x Two r6rH-century Taxation Lists, 1545 AND ae. Edited o G. Dy Rar 1954. : XI ss WILTSHIRE Conon SESSIONS AND AssizEs, 1736. Edited by Je: Po Me Fowle 1955: XII =—- Cotrectansa. Edited by N. J. Williams, with a foreword by T. F. ‘T. Pluckn XIII Procress Nores of WaRDEN WoopwaArD FOR THE WiLTsHIRE Estates or N COLLEGE, OxrorD, 1659 To 1675. Edited by R. L. Rickard. 1957. XIV Accounts AND SURVEYS OF THE WiLTsmRE LANDS Or ADAM DE STRA Edited by M. F. Farr. 1959. XV TRapEsMEN In Earty Stuart Wi tsuire. A Miscertany. Edited by N. Williams. 1960. XVI Crown Puieass oF THE WitTsaiRE Eyre. Edited by C. A. F. Meekings. 1961. XVII WriitsHirE APPRENTICES AND THEIR Masters, Edited by Christabel Dale, w. an introduction by N. J. Williams. 1961. XVIII Hemincsy’s Recister. Edited by Helena Chew. 1963. XIX Documents ILLustRATING THE WILTSHIRE TEXTILE TRADES IN THE 18TH Cenrvar Edited by Julia de L. Mann. 1964. ; Natural History Section The object of this Section is to promote the study of all branches of Natural Hi in the County by encouraging. field observations, maintaining records, arranging fi and other meetings and by putting observers in touch with each other. Members others who wish for particulars of the Section and its activities should write to the Honor, -Treasurer of the Section: Mr. Arnold Smith, 49 Clarendon Avenue, Trowbridge. Membership of the Section does not entail any further subscription from those are already members of the Society. Observations should be sent to the recorders: Birps: Mrs. Egbert Barnes, Hungerdown, Seagry, Chippenham, Wilts. FLowers: Mr. J. D. Grose, Downs Edge, Liddington, near Swindon. LepipopreraA: Mr. B. W. Weddell, ti The Halve, Trowbridge. MamMaAtLs, REPTILES AND AmpHIBIANs: Miss B. Gillam, 19 Roundway Gardens, Devi Back numbers of the Report of the Section can be obtained from Mrs. Egbert B : NES, Prices on application. . sane eitiesyietel el pases pets 2 Leeteuinoskscsetns anne ns