¹ This paper is an extended version of the one which I gave at the meeting of the CBA Urban Churches Working Party held in Worcester on 1-2 November 1986. I am very grateful to Dr J. S. Barrow, Dr W. J. Blair, Professor C. N. L. Brooke, Professor N.P. Brooks, Professor M. O. H. Carver, Dr A. S. Esmonde Cleary and Dr R. K. Morris for their many helpful comments on it at various stages in its preparation. I would also like to thank Mr H. Buglass for preparation of figs 1-5, and Mr R. Whittaker for his assistance in respect of plate XLa.

² Bede, , Historia ecclesiastica gentis Anglorum [hereafter H.E.] 1.29Google Scholar, quoted [as hereafter] from Colgrave, B. and Mynors, R. A. B. (eds.), Bede's Ecclesiastical History of the English People (Oxford, 1969), 105Google Scholar. The letter is dated 601.

³ Stenton, F. M., Anglo-Saxon England, 3rd edn. (Oxford, 1971), 109Google Scholar. This view is also taken by Brooks, N. P., The Early History of the Church of Canterbury (Leicester, Stenton, F. M., Anglo-Saxon England, 1984), 10Google Scholar.

⁴ York from 626 ( Fryde, E. B., Greenway, D. E., Porter, S. and Roy, I., Handbook of British Chronology, 3rd edn. (London, 1986), 224Google Scholar); Lin-disfarne from 635 to 833, then Chester-le-Street to 995, then Durham (ibid., 219, 214, 216); Ripon briefly in the seventh century (ibid., 220); Hexham from 678 to 821 (ibid., 217); Abercorn from 681 to 685 (ibid., 301); Whithorn from the 720s to the early ninth century (ibid., 222-3; Thomas, C., ‘Abercorn and the Provincia Pictorum’, in Miket, R. and Burgess, C. (eds.), Between and Beyond the Walls. Essays on the Prehistory and History of North ern Britain in Honour of George Jobey (Edinburgh, 1984), 328)Google Scholar.

⁵ H.E. II.3. On this see Brooks, , op. cit. (note 3), 10–11Google Scholar.

⁶ I owe the suggestion of this possibility to John Blair.

⁷ Mann, J.C., ‘The administration of Roman Britain’, Antiquity 35 (1961), 317–19CrossRefGoogle Scholar; Thomas, C., Christianity in Roman Britain to AD 500 (London, 1981), 193Google Scholar; Hill, R., The Labourers in the Field, Jarrow Lecture for 1984 (Jarrow, n.d.), 3Google Scholar. It would be dangerous to assume that every fourth-century civitas had a bishop: cf. Wightman, E. M., Gallia Belgica (London, 1985), 287Google Scholar.

⁸ Godfrey, C. J., The Church in Anglo-Saxon England (Cambridge, 1962), 69–70Google Scholar; Hill, , op. cit. (note 7), 3–5Google Scholar.

⁹ Though not of course of four metropolitans, an arrangement which the division of late Roman Britain into four provinces seems to have required but which the status of Britain in Gregory's day did not.

¹⁰ Winterbottom, M. (ed. and trans.), Gildas. The Ruin of Britain and Other Works (London and Chichester, 1978), 16, 90Google Scholar. Winterbottom suggests (ibid., 148) that Gildas's xxviii was perhaps a scribal error for xxxviii, the number given by Ptolemy of cities in Britain south of Hadrian's Wall (of a total of 58); but on this see Stevens, C. E., ‘Gildas and the Civitates of Britain’, Engl. Hist. Rev. 52 (1937), 201Google Scholar.

¹¹ Jackson, K., ‘Nennius and the twenty-eight cities of Britain’, Antiquity 12 (1938), 44–55CrossRefGoogle Scholar; Campbell, J., ‘Bede's words for places’ in Sawyer, P.H. (ed.), Names, Words, and Graves: Early Medieval Settlement (Leeds, 1979), 41–2Google Scholar.

¹² Mann, J. C., ‘Civitas: another myth’, Antiquity 34 (1960), 222–3CrossRefGoogle Scholar, for the latter meaning; Frere, S. S., ‘Civitas—a myth’, Antiquity 35 (1961), 29–36CrossRefGoogle Scholar, for the former.

¹³ H.E. 1.1 (Colgrave and Mynors, op. cit. (note 2), 17).

¹⁴ Haverfield, F., The Roman Occupation of Britain, revised by MacDonald, G. (Oxford, 1924), 289–93Google Scholar; Stevens, , op. cit. (note 10), 193–203Google Scholar; Jackson, , op. cit. (note 11)Google Scholar; Jones, A.H.M., The Later Roman Empire, 284-602 (Oxford, 1964), 715Google Scholar. ‘The twenty-eight civitates of Gildas and Nennius may not be so mythical as we sometimes assume’, Rivet, A. L. F., ‘Summing up; some historical aspects of the civitates in Roman Britain’, in Wacher, J. S. (ed.), The Civitas Capitals of Roman Britain (Leicester, 1966), 109Google Scholar.

¹⁵ Stevens, , op. cit. (note 10), 199Google Scholar. His total is accepted by Jackson, , op. cit. (note 11), 53Google Scholar.

¹⁶ The twenty-five places discussed by Stevens, , op. cit. (note 10), 202–3Google Scholar, afe Aldborough, Brough-on-Humber, Caerwent, Caistor-by-Norwich, Canterbury, Carlisle, Carmarthen, Chelmsford, Chi-chester, Cirencester, Colchester, Corbridge, Dorchester, Durobrivae, Exeter, Gloucester, Leicester, Lincoln, London, Mediolanum Ordovicum, St Albans, Silchester, Winchester, Wroxeter and York. Most of diese identifications are not in doubt. Opinions are, however, divided about the status of Brough-on-Humber and Chelmsford, which are seen as ‘civitas capitals’, if failed ones, by some (e.g. Wacher, J., The Tovms of Roman Britain (London, 1974), 22–3Google Scholar) but not by others (e.g. Ordnance Survey Map of Roman Britain (Southampton, 1978)Google Scholar; Salway, P., Roman Britain (Oxford, 1981)Google Scholar, map V after xxi); and about Durobrivae, variously identified as Castor, Chesterton and Water Newton ( Wacher, , op. cit., 22Google Scholar; Johnson, S., Later Roman Britain (London, 1980), 9Google Scholar). Only one of Stevens's suggestions—Ptolemy's Mediolanum Ordovicum—has to be discounted. This is almost certainly Whitchurch (Shropshire), which would have belonged to die Comovii not the Ordovici ( Rivet, A. L. F. and Smith, C., The Place-Names of Roman Britain (London, 1979), 121Google Scholar), judging from its situation; Ptolemy's description of it as a polis carries no weight (ibid., xvii), and it is an otherwise unsupported candidate. In a subsequent paper ( ‘The Roman name of Ilchester’, Proc. Somersetshire Archaeol. Natur. Hist. Soc. 94 (1950), 188–92Google Scholar) Stevens showed why Ilchester should be accepted as another ‘civitas capital': some have accepted this (e.g. Rivet, and Smith, , op. cit., 393Google Scholar; Wacher, , op. cit., 22Google Scholar) but odiers have not (e.g. Bogaers, J. E., J. Roman Stud. 57 (1967), 233CrossRefGoogle Scholar). In addition Johnson (op. cit., 90) has argued that Chester had probably become a ‘civitas capital’ by the end of the third century.

¹⁷ Rivet, , op. cit. (note 14), 109Google Scholar.

¹⁸ Morris, J. (ed. and trans.), Nennius. British History and the Welsh Annals (London and Chichester, 1980)Google Scholar. For an assessment of it as an historical source, see Dumville, D. N., ‘Nennius and the HistoriaBrittonum’, StudiaCeltica, 10-11 (1975-1976), 78–95Google Scholar.

¹⁹ Morris, , op. cit. (note 18), 40Google Scholar; Haverfield, , op. cit. (note 14), 290Google Scholar; Jackson, , op. cit. (note 11), 46–52Google Scholar.

²⁰ Jackson, ibid., 55.

²¹ Those of Canterbury, Dorchester-on-Thames (if it still was a see then), Dommoc (see below, note 29), London, the Mercians (with the Middle Angle s and Lindsey), Winchester and Rochester, in southern Britain. Fo r northern Britain, see note 4.

²² ‘So Theodore journeye d to every district, consecrating bishops in suitable places’, H.E. iv.2 ( Colgrave, and Mynors, , op. cit. (note 2), 335Google Scholar). He established new sees at Abercorn, Hereford, Hexham, Leicester, Lichfield, North Elmham, Ripon and Worcester, as well as one in Lindsey, and may have trie d to revive die se e at Dorchester-on-Thames: H.E. iv.3, iv.5, iv.12, iv.13, iv.23; Godfrey, , op. cit. (note 8), 131–5Google Scholar.

²³ Bassett, S., ‘Lincoln and the Anglo-Saxon see of Lindsey’, Anglo-Saxon England 18 (1990), 1–32CrossRefGoogle Scholar.

²⁴ Munby, J., ‘Saxon Chichester and its predecessors’ in Haslam, J. (ed.), Anglo-Saxon Towns in Southern England (Chichester, 1984), 317–20Google Scholar. However, the see may have been originally meant to be located elsewhere, conceivably at Bosham whic h also is directly adjacent to Chichester: Kirby, D. P., ‘The Church in Saxon Sussex’, in Brandon, P. (ed.), The South Saxons (London and Chichester, 1978), 168–9, 171Google Scholar; Gem, R. D. H., ‘Holy Trinity Church, Bosham’, Archaeol. J. 142 (1985), 33Google Scholar. On the likelihood of there having been a major Soudi Saxon royal centre in die immediate vicinity of Chichester, now also see Welch, M., ‘The kingdom of the South Saxons: die origins’, in Bassett, S. (ed.), The Origins of Anglo-Saxon Kingdoms (Leicester, 1989), 79Google Scholar.

²⁵ Frere, S. S., ‘Excavations at Dorchester-on-Thames, 1963’, Archaeol. J. 141 (1984), 93, 114–19, 133-5CrossRefGoogle Scholar; Branigan, K., The Catuvellauni (Gloucester, 1985), 82, 92–3Google Scholar, where it is suggested that it was one of several lesser administrative centres in th e civitas. There is, however, no evidence that it became a ‘civitas capital’ itself: Frere, S. S., ‘The end of towns in Roman Britain’ in Wacher, , op. cit. (note 14), 93Google Scholar. It has been proposed, if on the slightest of evidential bases, that the seventh-century cathedral was on the site of a capella memoriae in a former (Roman) cemetery: Doggett, N., ‘The Anglo-Saxon see and cathedral of Dorchester-on-Thames: the evidence reconsidered’, Oxoniensia 51 (1986), 49–61Google Scholar.

²⁶ Rivet, in Wacher, , op. cit. (note 14), 109Google Scholar; Gould, J., ‘Letocetum, Christianity and Lichfield (Staffs.)’, Trans. S. Staffordshire Archaeol. Hist. Soc. 14 (1973), 30–1Google Scholar; Gelling, M., Signposts to the Past (London, 1978), 57Google Scholar.

²⁷ Reynolds, J. M., ‘Legal and constitutional problems’ in Wacher, , op. cit. (note 14), 72Google Scholar and n. 17; Wilmott, A.R., ‘Kenchester (Magnis): a reconsideration’, Trans. WoolhopeNatur. Fid. Club 43.2 (1980), 125–6, 128-30Google Scholar.

²⁸ Wacher, , op. cit. (note 16), 405Google Scholar.

²⁹ Whitelock, D., ‘The pre-Viking age church in East Anglia’, Anglo-Saxon England 1 (1972), 4 n. 2CrossRefGoogle Scholar; Wade-Martins, P., Excavations in North Elmham Park 1967-1972, East Anglian Archaeol. 9 (Dereham, 1980), 4–5Google Scholar; Rigold, S. E., ‘The supposed see of Dunwich’, J. Brit. Archaeol. Ass., 3rd ser., 24 (1961), 55–9CrossRefGoogle Scholar, and ‘Further evidence about the site of “Dommoc”’, J. Brit. Archaeol. Ass., 37 (1974), 97–102CrossRefGoogle Scholar.

³⁰ Hope-Taylor, B., Yeavering. An Anglo-British Centre of Early Northumbria, DoE Archaeol. Rep. 7 (London, 1977), 292, 301-2, 307-8Google Scholar.

³¹ e.g. Exeter, Carlisle and the ones now in Wales.

³² H.E. 1.29, 11.2; Mayr-Harting, H., The Coming of Christianity to Anglo-Saxon England (London, 1972), 71–2Google Scholar.

³³ Some of these lay too close to (or even beyond) the frontiers of the contemporary Anglo-Saxon kingdoms (e.g. Cirencester, St Albans); others wer e well situated, but presumably were less favoured by th e ruling dynasties than the place(s) chosen to be sees (e.g. Gloucester).

³⁴ e.g. Hill, D. H., ‘Continuity from Roman to medieval: Britain’ in Barley, M. W. (ed.), European Towns. Their Archaeology and Early History (London, 1977), 298Google Scholar; Hawkes, S.C., ‘Anglo-Saxon Kent c. 425-725’ in Leach, P. E. (ed.), Archaeology in Kent to AD 1500, CBA Res. Rep. 48 (London, 1982), 76Google Scholar.

³⁵ Biddle, M., ‘Towns’ in Wilson, D. M. (ed.), The Archaeology of Anglo-Saxon England (London, 1976), 103–12Google Scholar and references cited there.

³⁶ Canterbury/Kent: Loyn, H. R., Anglo-Saxon England and the Norman Conquest (London, 1962), 28Google Scholar; Biddle, M., ‘Archaeology and the beginnings of English society’ in Clemoes, P. and Hughes, K. (eds.), England before the Conquest. Studies in Primary Sources presented to Dorothy Whitelock (Cambridge, 1971), 394–6Google Scholar; Brooks, N. P., ‘The creation and early structure of the kingdom of Kent’ in Bassett, , op. cit. (note 24), 55–74Google Scholar, esp. 69. For a different view: Hawkes, in Leach, , op. cit. (note 34), 75–6Google Scholar. Dorchester-on-Thames/Wessex: Kirby, D. P., ‘Problems of early West Saxon history’, Engl. Hist. Review 80 (1965), 23–9Google Scholar; Biddle, M., ‘Hampshire and the origins of Wessex’ in Sieveking, G. de G., Longworth, I. H. and Wilson, K. E. (eds.), Problems in Economic and Social Archaeology (London, 1976), 321–42Google Scholar. York/Deira: Blair, P. H., ‘The origins of Northumbria’, Archaeol. Aeliana, 4th ser., 25 (1947), 41-4, 50Google Scholar; Addyman, P. V., ‘Problems in the archaeology of York: new slants on old Angles’, Interim, 1, no. 4 (York Archaeological Trust, Jan. 1974), 20–1Google Scholar. Lincoln affords an instance of a place's economic importance being influential in its choice as the site of a see: Bassett, , op. cit. (note 23)Google Scholar.

³⁷ Probably in 679 or 680: the first date is given in an appendix to ‘Florence’ (i.e. John) of Worcester's Chronicle: Thorpe, B. (ed.), Florentii Wigomiensis Monachi Chronicon ex Chronicis, English Historical Soc, 2 vols. (London, 1848-1849), 1, 239Google Scholar; the second is in its main text: ibid., 36. But for the possibility of a slightly earlier date, now see Sims-Williams, P., ‘St Wilfrid and two charters dated AD 676 and 680’, J. Eccles. Hist. 39 (1988), 168–9CrossRefGoogle Scholar.

³⁸ The earliest source for the extent of the Anglo-Saxon diocese is Taxatio Ecclesiastica Angliae et Walliae Auctoritate P. Nicholai IV circa A.D. 1291 (Rec. Comm., London, 1802)Google Scholar [hereafter Taxatio], 216-40. Clearly, however, there may have been local adjustments to its boundaries since the seventh century; for discussion of this, see Hooke, D., The Anglo-Saxon Landscape: The Kingdom of the Hwicce (Manchester, 1985), 12–20Google Scholar.

³⁹ The best discussion of the Hwicce is Smith, A. H., ‘The Hwicce’ in Bessinger, J. B. and Creed, R. P. (eds.), Frankiplegius: Medieval and Linguistic Studies, in Honour of Francis Peabody Magoun, Jn. (London, 1965), 55–65Google Scholar.

⁴⁰ H.E. II.2 ( Colgrave, and Mynors, , op. cit. (note 2), 134)Google Scholar.

⁴¹ Plummer, C. (ed.), Two of the Saxon Chronicles Parallel, 2 vols. (Oxford, 1892-1899), 1, 18–19Google Scholar.

⁴² Cook, J.M., ‘An Anglo-Saxon cemetery at Broadway Hill, Broadway, Worcestershire’, Antiq. J. 38 (1958), 76–81CrossRefGoogle Scholar; Meaney, A., A Gazetteer of Early Anglo-Saxon Burial Sites (London, 1964)Google Scholar, passim. For a more recent discovery at Bishops Cleeve, see Wilson, D. M. and Hurst, D. G., ‘Medieval Britain in 1969’, Medieval Archaeol. 14 (1970), 156CrossRefGoogle Scholar.

⁴³ Smith, A. H., The Place-Names of Gloucester-shire (Engl. Place-Name Soc., vols. 38-41, 1964-1965), 41, 38–41Google Scholar.

⁴⁴ Pretty, K., The Welsh Border and the Severn and Avon Valleys in the Fifth and Sixth Centuries A.D.: An Archaeological Survey (unpubl. Ph.D. thesis, Cambridge Univ., 1975)Google Scholar.

⁴⁵ Stenton, , op. cit. (note 3), 45Google Scholar; Finberg, H. P. R., ‘The princes of the Hwicce’ in his The Early Charters of the West Midlands (Leicester, 1961), 171Google Scholar

⁴⁶ Bassett, S. R., ‘A probable Mercian royal mausoleum at Winchcombe, Gloucestershire’, Antiq. J. 65 (1985), 84CrossRefGoogle Scholar; idem, ‘In search of the origins of Anglo-Saxon kingdoms’ in Bassett, (ed.), op. cit. (note 24), 3–27Google Scholar.

⁴⁷ Bassett, S., ‘Church and diocese in the west Midlands: the transition from British to Anglo-Saxon control’ in Blair, J. and Sharpe, R. (eds.) Pastoral Care before the Parish (Leicester University Press, forthcoming)Google Scholar.

⁴⁸ Bassett, , ‘In search of the origins of Anglo-Saxon kingdoms’ in Bassett, (ed.), op. cit. (note 24)Google Scholar. Bede's silence may be due to his inability to get reliable information about a mission to th e Hwicce: Kirby, D. P., ‘Bede's native sources for the Historia Ecclesiastica’, Bull. John Rylands Lib. Univ. Manchester 48 (1965-1966), 368–70Google Scholar. He does tell us that Oftfor had spent a long time in the kingdom, probably before he became the second bishop of the Hwicce, ‘preaching the word of faith [verbum fidei praedicans] and setting an example of holy life to all who saw and heard him’ (H.E. iv.23; Colgrave, and Mynors, , op. cit. (note 2), 411)Google Scholar; this could be taken to mean a mission (a suggestion which I owe to Christopher Brooke), but may be satisfactorily read, in the context of Bede's whole statement about Oftfor, as an appreciation of his having taken on the pastoral work of Bishop Bosel who ‘At that time [Quo tempore]… was greatly troubled by ill-health so that he could not carry out his episcopal duties himself (ibid.).

⁴⁹ Thomas, , op. cit. (note 7), 268–71Google Scholar. Roger of Hoveden (d. ?1201) and Giraldus Cambrensis (d. ?1220) both state that Worcester was one of a number of British sees subject to St David's until it was lost to the Anglo-Saxons: Stubbs, W., Chronica Magistri Rogeri de Houedene, 4 vols., Rolls Ser. 51 (London, 1868-1871), iv, 104Google Scholar; Brewer, J. S. et al. (eds.), Giraldi Cambrensis Opera, 8 vols., Rolls Ser. 21 (London, 1861-1891), III, 54Google Scholar; but Hoveden's list has been called ‘a mere blundering guess’: Haddan, A. W. and Stubbs, W., Councils and Ecclesiastical Documents relating to Great Britain and Ireland, 3 vols. (Oxford, 1869-1878), III, 41 n. cGoogle Scholar. The ‘tradition’ that among the British bishops who met Augustine there was one from Worcester (cf. Bond, C. J., ‘Church and parish in Norman Worcestershire’ in Blair, W. J. (ed.), Minsters and Parish Churches. The Local Church in Transition 950-1200 (Oxford, 1988), 130Google Scholar) is spurious: Haddan, and Stubbs, , op. cit., 1, 148 and n.; in, 41 n. cGoogle Scholar.

⁵⁰ H.E. 1.27, 11.2; Hearne, T. (ed.), Hemingi Chartularium Ecclesiae Wigomiensis, 2 vols. (Oxford, 1723), II, 567Google Scholar( Sawyer, P. H., Anglo-Saxon Charters: an Annotated List and Bibliography (London, 1968)Google Scholar, no. 1822). For archaeological and placename evidence see Morris, R., The Church in British Archaeology, CBA Res. Rep. 47 (London, 1983), ch. 3Google Scholar; for topographical evidence see Bassett, , op. cit. (note 47)Google Scholar.

⁵¹ Gloucester: Bryant, R., ‘Excavations at the church of St Mary de Lode, Gloucester’, Bull. CBA Churches Comm. 13 (1980), 15–18Google Scholar. Much Wenlock: Fernie, E., The Architecture of the Anglo-Saxons (London, 1983), 64–5Google Scholar; Woods, H., ‘Excavations at Wenlock Priory, 1981-6’, J. Brit. Archaeol. Ass. 140 (1987), 36–75CrossRefGoogle Scholar(but now see Biddle, M. and Biddle, B. Kjølbye, ‘The so called Roman building at Much Wenlock’, J. Brit. Archaeol. Ass. 141 (1988), 179–83Google Scholar). Lichfield: Gould, , op. cit. (note 26), 30–1Google Scholar; Bassett, , op. cit. (note 47)Google Scholar.

⁵² Barker, P. A., Cubberley, A. L., Crowfoot, E. and Radford, C. A. R., ‘Two burials under the refectory of Worcester Cathedral’, Medieval Archaeol. 18 (1974), 146–51CrossRefGoogle Scholar. For calibration of th e radiocarbon dates, see below, note 107.

⁵³ Carver, M. O. H. (ed.), ‘Medieval Worcester. An archaeological framework’, Trans. Worcester-shire Archaeol. Soc. 3rd ser., 7 (1980), 2–3, 26Google Scholar.

⁵⁴ Bassett, , op. cit. (note 46), 84 and n. 16Google Scholar.

⁵⁵ Baker, N. J., ‘Churches, parishes and early medieval topography’ in Carver, (ed.), op. cit. (note 53), 31–7Google Scholar.

⁵⁶ Baker, N. J., ‘The urban churches of Worcester—a survey’ in Carver, (ed.), op. cit. (note 53), 115–24Google Scholar. St Helen's church and its pre-Conquest parish are also discussed in a paper first brought to my attention by John Blair: Bond, , op. cit. (note 49), 130–2Google Scholar. I am grateful to James Bond and John Blair for allowing me to see the relevant part of this paper at proof stage in its publication.

⁵⁷ Darlington, R. R., The Cartulary of Worcester Cathedral Priory (Register I), Pipe Roll Soc, new ser., 38 (1968), 32–3, no. 53Google Scholar. On the date see Brett, M., The English Church under Henry I (London, 1975), 180 n. 4.Google Scholar(I owe this reference to Christopher Brooke.)

⁵⁸ All three were on large and important manors, of which the first two had been held in demesne by the Church of Worcester since the middle Saxon period, and the third was royal demesne (see below, p. 238). Wick and Martley both had several chapels in the thirteenth century, and the value of Martley and its chapels c. 1291 was £30 13s. 4d. ( Hale, W. H. (ed.), Registrant Prioratus Beatae Mariae Wigorniensis, Camden Soc, Old Ser., 91 (1865), 35aGoogle Scholar; Taxatio, 216.) By the thirteenth century Claines had superseded North-wick as the name of the bishop's demesne manor on the eastern bank of the Severn. St Helen's held the chapel of Claines c. 1100, one of six said to be in Northwick ( Darlington, , op. cit. (note 57), 33Google Scholar). It is not clear if it was the mother church of Northwick; but it does not appear to have had sepulture until 1400 ( Calendar of Papal Registers. Papal Letters 5 (1904), 374Google Scholar). This area may always have been under the tight control of St Helen's.

⁵⁹ Mr Baker states that St Helen's ‘did not have an extramural parish’ (op. cit. (note 55), 34), while acknowledging that its control over rural churches c. 1100 must have been of ancient origin. His list of the churches omits Hindli p and Martley, and incorrectly includes Oddingley (op. cit. (note 56), 116).

⁶⁰ Taxatio, 216; Thomas, W., ‘An account of the bishops of Worcester’ in his A Survey of the Cathedral-Church of Worcester with an Account of the Bishops., to.. 1660 (London, 1736), 115Google Scholar; and ‘Appendix cartarum originalium’, in ibid., 17-18, no. 29; Habington, T., A Survey of Worcestershire, ed. Amphlett, J., 2 vols., Worcestershire Hist. Soc. 5 (1895-1899). II 12Google Scholar.

⁶¹ Farley, A. (ed.), Domesday Book seu Liber Censualis Willelmi Primi Regis Angliae [cited hereafter as D.B.], 2 vols. (London, 1783), I, fol. 173cGoogle Scholar; The Victoria History of the Counties of England [cited hereafter as VCH], Worcestershire, III (London, 1913), 400Google Scholar.

⁶² Haines, R. M. (ed.), Calendar of the Register of Wolstan de Bransford, Worcestershire Hist. Soc, new ser., 4 (1966), 368, 383, 386, 397.Google Scholar‘Inductor, R. of St Helen, Worcester, as custodian of the peculiar jurisdiction to which the church belongs’: Marett, W. P. (ed.), Calendar of the Register of Henry Wakefield, Worcestershire Hist. Soc, new ser., 7 (1972), 105Google Scholar.

⁶³ Valor Ecclesiasticus Temp. Henr. VIII Auctoritate Regia Institutus, 6 vols. (Rec. Comm., London, 1810-1834) [hereafter Valor], III, 232Google Scholar. There are grounds for suggesting that Oddingley had belonged to the minster church of Hanbury before becoming a manor of the Church of Worcester.

⁶⁴ It can be shown to have been a township in Himbleton in the late ninth century, which in turn was part of the land belonging to Hanbury church in 836 ( Birch, W. de G., Cartularium Saxonicum, 3 Vols. (London, 1885-1893), II no. 552 and 1Google Scholar, no. 416 respectively; Sawyer, , op. cit. (note 50), nos. 219 and 190)Google Scholar. The argument will be made in full in a discussion of the origins and development of the medieval landscape of Hanbury, Worcestershire (in preparation).

⁶⁵ Huddington was regularly recorded as Hodin[g]ton[a] in the twelfth century: Mawer, A. and Stenton, F. M., The Place-Names of Worcester-shire, Eng. Place-Name Soc. 4 (1927), 142Google Scholar.

⁶⁶ D.B. 1, fol. 173c; Valor, III, 235.

⁶⁷ Probably the estate held of Kempsey in 1086: D.B. I, fol. 172d.

⁶⁸ Probably the estate held of Northwick in 1086, which wa s a township in Whittington earlier called Battenhall: D.B. I, fol. 173c.

⁶⁹ Birch, , op. cit. (note 64), 1Google Scholar, no. 295 ( Sawyer, , op. cit. (note 50), no. 154Google Scholar); Dugdale, W., Monasticon Anglicanum, ed. Caley, J., Ellis, H. and Bandinel, B., 6 vols. in 8 (London, 1817-1830), 1, 608Google Scholar; VCH Worcs., III, 456, 53&7.

⁷⁰ Valor, III, 221, 224-6, 235, 512; Hale, , op cit. (note 58), 32b, 44a, 50aGoogle Scholar.

⁷¹ Nash, T., Collections for the History of Worcestershire, 2 vols. (London, 1781-1782), 1, 259Google Scholar(Cotheridge), and 11, 364 (Spetchley); and in ibid. (2nd ed. with additions, London, 1799), 1, 126-7 (Bredicot).

⁷² D.B. I, fol. 174a. A priest is recorded, and it s i stated that ‘Fro m this estate 8d. is paid annually to the Church of Worcester for churchscot [cirsette] and acknowledgement of the land’.

⁷³ Bund, J. W. Willis (ed.), Register of Bishop Godfrey Giffard, Worcestershire Hist. Soc. 15 (1902), 106, 428Google Scholar; Valor, III, 225. The other estates were Cudley and Perry.

⁷⁴ The church of St Michael, Worcester—formerly the parish church of the cathedral, and now lost ( Baker, , op cit. (note 56), 120Google Scholar. —may well have been founded to look after these chapels of the cathedral priory, i.e. as a counterpart to St Helen's church in the latter's contemporary role.

⁷⁵ Darlington, , op cit. (note 57), 33, no. 54Google Scholar.

⁷⁶ Ibid., 31-2, no. 52; now also printed, with corrections, in Whitelock, D., Brett, M. and Brooke, C. N. L. (eds.), Councils and Synods with other documents relating to the English Church, 2 vols. (Oxford, 1981), 1, pt. 2, 636–9Google Scholar.

⁷⁷ On the probable antiquity of St Alban's, the other church involved in the dispute, see below, pp.244-6.

⁷⁸ Its most recent editors state, ‘We assume that it is genuine, though this is perhaps not beyond doubt.’ ( Whitelock, et al., op. cit. (note 76), 635 n. 5)Google Scholar. However, Christopher Brooke reports (pers. comm.) that he is now more doubtful about its authenticity. Julia Barrow (pers. comm.) has kindly offered the following comment on the document: ‘It is not in Hemming's cartulary. This is certainly not a conclusive argument against its validity, but effectively this document survives in no manuscript earlier than the 1240 one (Worcester Cath. Lib. MS Reg. 1 (A.4), fols. 7^v-8). And other charters of the 1090s are in Hemming (cf. Darlington, , op. cit. (note 57), nos. 2–53Google Scholar). Secondly, the style of the piece protests too much, e.g. in its reference to the changeover in the cathedral from clerks to monks. Going into long historical excursus does happen in genuine documents of this period, but it is more commonly a feature of forgeries. It is, moreover, not written like a document, but like a story: it does not even bother to turn the details it is mentioning into a compact text—cf. short sentences such as ‘Hec sinodus habita est anno dominice Incarna-tionis m^o. xc^o. ii^o. indictione xv.’ and ‘Horum pres-byterorum altercatio sanctam sinodum multum detinuit.’ ( Whitelock, el al., op. cit. (note 76), 636–7Google Scholar). But this too is not necessarily proof of forgery, as charter form in this period was not absolutely fixed. Finally, it contains some obvious anachronisms: (1) vicarages, vicars (ibid., 638): surely not as early as 1092; but certainly by the mid twelfth century (possibly by the 1130S-40S). (2) Getting the monks of a community to act as deans over land or other property (ibid., 639) is a Cluniac phenomenon. The term would not have come into common usage in England as early as Oswald's time. (3) ‘nullus decanus nullus archidiaconus’ (ibid., 639): here the dean referred to is not a monastic one but a rural dean, i.e. someone under the control not of Worcester Cathedral priory but of the bishop and the archdeacon. Rural deans do not occur in England, at least not in the west Midlands, much before the 1130s ( Brett, , op. cit. (note 57), 211–15Google Scholar). There may be an interesting parallel for this document in the charter of Osmund for Salis-bury, which was composed in the mid(?) twelfth century to pretend to be from the early 1090s ( Greenway, D., ‘The false Institutio of St Osmund’ in Greenway, D., Holdsworth, C. and Sayers, J. (eds.), Tradition and Change: Essays in Honour of Marjorie Chibnall (Cambridge, 1985), 77–101).’Google Scholar

⁷⁹ ‘Hii omnes… affirmaverunt nulla m esse parrochiam in tota urbe Wigrac’ nisi tantum matris ecclesie’, Whitelock, el al., op. cit. (note 76), 637–8Google Scholar.

⁸⁰ Brooke, C. N. L., ‘The missionary at home: the church in the towns, 1000-1250’ in Cuming, G. J. (ed.), The Mission of the Church and the Propagation of the Faith, Studies in Church History 6 (London, 1970), 64Google Scholar.

⁸¹ Darlington, , op. cit. (note 57), 31Google Scholar.

⁸² Baker, , op. cit. (note 55), 34–5Google Scholar.

⁸³ i.e. by the early eleventh century: Dyer, C. C., Lords and Peasants in a Changing Society (Cambridge, 1980), 35–6, 40Google Scholar.

⁸⁴ D.B. 1, fols. 172d, 173c, d.

⁸⁵ Ibid., fols. 173d-174a. For the separate provisions made for the monks, see John, E., Land Tenure in Early England (Leicester, 1964), 113–15Google Scholar.

⁸⁶ The large estate of Wick was given by Offa in 757×775: Birch, , op. cit. (note 64), 1Google Scholar, no. 21 9 ( Sawyer, , op. cit. (note 50), 142Google Scholar). For its extent, see Grundy, G. B., ‘Saxon charters of Worcestershire’, Trans. Birmingham Archaeol. Soc. 52 (1927), 13–18Google Scholar; it is mapped in Hooke, D., Anglo-Saxon Landscapes of the West Midlands: the Charter Evidence, Brit. Archaeol. Rep. 95 (Oxford, 1981), 337Google Scholar. Broadwas was given by Offa in 789 or 790 ( Robertson, A. J. (ed.), Anglo-Saxon Charters, 2nd ed. (Cambridge, 1956), 2–5Google Scholar, no. 2), and Grimle y by Beorhtwulf in 851 according to a spurious charter ( Birch, , op. cit., 11, no. 462Google Scholar( Sawyer, , op. cit., no.201)Google Scholar) which may nevertheless be correct in the name of the (otherwise unlikely) donor. No charter of donation is known for the land east of the Severn (except Kempsey: see above, note 69); the earliest reference t o any of it is in a gran t of Coenwulf in 816 (Birch, ibid., 1, no. 357 (Sawyer, ibid., no. 180)). It may therefore have been given to the minster church of St Peter at its foundation, presumably by the Hwiccian ruler(s) of the day.

⁸⁷ See above, note 69.

⁸⁸ D.B. I, fol. 172d.

⁸⁹ Dugdale, , op. cit. (note 69), 1, 609Google Scholar; Birch, , op. cit. (note 64), III, no. 937Google Scholar (Sawyer, , op. cit. (note 50), no. 633Google Scholar), by which King Eadwig granted five hides at Fepsetnatune to the Churc h of Worcester in 956. This estate can be shown to have included both Himbleton and Huddington. For Oddingley see Finberg, , op. cit. (note 45), 187Google Scholar.

⁹⁰ Apart from Knightwick's inclusion in the spurious charter by which King Edgar is alleged to have set up the triple hundred of Oswaldslow in 964 ( Birch, , op. cit. (note 64), III, no. 1135Google Scholar( Sawyer, , op. cit. (note 50), 731Google Scholar); Dyer, , op. cit. (note 83), 35–6Google Scholar), the earliest reference to it is in a marriage agreement of 1014X1016 ( Robertson, , op. cit. (note 86), 148–9Google Scholar, no. 76). At that time it was held by the monastic community of Winchcombe, but it was later acquired for the sister of Archbishop Wulfstan. By 1086 the monks of Worcester held it of their manor of Grimley (D.B. 1, fol. 173d). It was, however, always ecclesiastically linked to Doddenham and Mardey, which suggests that it had remained in royal demesne for much longer than most of the Church of Worcester's estates to the west of the Severn.

⁹¹ Martley: D.B. I, fols. 177a, 178a, 180c; Doddenham: ibid., fol. 176d. For the latter's ecclesiastical links with Martley, see above, p. 232 and note 60.

⁹² Especially the Codicellus possessionum, fols. 118^r-132^r of Brit. Lib. MS Cotton Transcript A.xiii ( Hearne, (ed.), op. cit. (note 50), 1, 248–86)Google Scholar.

⁹³ See above, p. 238 and note 86.

⁹⁴ Barker, P. A., ‘The Roman town’Google Scholar in idem (ed.), ‘The origins of Worcester’, Trans. Worcester-shire Archaeol. Soc. 3rd ser., 2 (1968-1969), 15–19Google Scholar; Carver, , op. cit. (note 53), 1–3, 25-6Google Scholar.

⁹⁵ Ibid., 2.

⁹⁶ The evidence will be set out in full in a discussion of Droitwich and Hanbury (in preparation). The area to the west of the Severn has not yet been investigated. For other similarly early systems in the west Midlands and elsewhere, see Bassett, S.R., ‘Medieval Lichfield: a topographical review’, Trans. S. Staffordshire Archaeol. Hist. Soc. 22 (1982), 95–8Google Scholar and refs. there; idem, ‘The Roman and medieval landscape of Wroxeter’ in Barker, P. (ed.), From Roman Viroconium to Medieval Wroxeter (Worcester, 1990), 10–12Google Scholar; Williamson, T., ‘Paris h boundaries and early fields: continuity and discontinuity’, J. Hist. Geogr. 12 (1986), 241–8CrossRefGoogle Scholar.

⁹⁷ Barker, , op. cit. (note 94), frontispiece and 40Google Scholar; Carver, , op. cit. (note 53), 2, 19–21Google Scholar.

⁹⁸ Hooke, D., ‘The hinterland and routeways of medieval Worcester: the charter evidence ‘ in Carver, , op. cit. (note 53), 43–7Google Scholar.

⁹⁹ Barker, , op. cit. (note 94), 16–18Google Scholar and refs. there; idem, ‘Excavations on the Lich Street Development Site, 1965-66’ in Trans. Worcester-shire Archaeol. Soc. 2 (1968-1969), 50–6Google Scholar; Baker, , op. cit. (note 55), 34–5Google Scholar.

¹⁰⁰ ibid., 35. Mr Baker assumes tha t the non-survival of this roa d within the Cathedral close is a strong argument in favour of total discontinuity between the Roman town and th e creation of the Anglo-Saxon see (ibid.); but that need not be so. It is much more likely to have been shut off as a through-route by the community of St Peter's, once its continuing use proved a nuisance to them; for analogues, see Bassett, ‘Medieval Lichfield’ (note 96), 106 and n. 19.

¹⁰¹ Barker, , op. cit. (note 94), 16, 18, 48, 50 and figs. 2, 9, 10Google Scholar; for a reassessment of their date, see Baker, , op. cit. (note 55), 34Google Scholar. Martin Carver accepts Barker's proposal of a first-century AD defensive circuit (i.e. a fort) based on ditch a on the Lich Street site and o n an apparently first-century ditch excavated on the site of 23-9 Sidbury (J. Sawle, ‘HWCM 117 Worcester Sidbury excavation—the second season’, Worcs. Archaeol. Local Hist. Newsletter 21 (1978), 1, 4); but he does not follow Barker's suggestion, taken up by Baker, that the much larger ditchb on the Lich Street site, whic h cut ditch a, belongs to a second, considerably later, defensive circuit ( Carver, , op. cit. (note 53), 1–2Google Scholar, and pers. comm.).

¹⁰² Birch, , op. cit. (note 64), 1, no. 60Google Scholar(Sawyer, , op. cit. (note 50), no. 70)Google Scholar; Finberg, H.P.R., ‘The genesis of the Gloucestershire towns’ in Finberg, (ed.), Gloucestershire Studies (Leicester, 1957), 55 and n. 2Google Scholar; Heighway, C. M., ‘Excavations at Gloucester. Fifth interim report: St Oswald's Priory, 1977-8’, Antiq. J. 60 (1980), 217–20CrossRefGoogle Scholar.

¹⁰³ Simon Esmonde Cleary comments that no Roman unit of administration is known which was dependent on a ‘small’ town, so that the land unit at Worcester is likely to be post-Roman (pers. comm.).

¹⁰⁴ Gelling, M., ‘A note on the name Worcester’ in Barker, , op. cit. (note 94), 26Google Scholar.

¹⁰⁵ Baker, , op. cit. (note 55), 34 and fig. 10Google Scholar.

¹⁰⁶ Ibid., 35; see also Carver, , op. cit. (note 53), 3Google Scholar.

¹⁰⁷ Barker, et al., op. cit. (note 52)Google Scholar, where raw radiocarbon date s from two samples of bone [hereafter called A and B] from Grave 1 are given as BP 1414 + 107 = ad 536 (ad 429-643) for A, and BP 13651102 = ad 585 (ad 483-687) for B. Use of the latest calibration curve ( Stuiver, M. and Pearson, G. W., ‘High-precision calibration of the radiocarbon time scale, AD 1950-500 BC’, Radiocarbon 28, no. 2B (1986), 805–38, esp. fig. ic on 812)CrossRefGoogle Scholar yields a calibrated range for sample A of AD 549-680, and for sample B of AD 602-(731, 764) (both based on one standard deviation, i.e. ia). At 2 a (giving a 95 per cent chance of certainty) the range for sample A is AD 416-(811, 852), and for sample B it is AD 449-886. (I am grateful to Susan Limbrey for her advice in this matter.) The obvious historical implication of the calibrated dates is that the burial in Grave 1, and presumably the one in Grave 2 as well, can no longer be safely said to be British: they may wel l postdate the creation of the Anglo-Saxo n see.

¹⁰⁸ In which case the Anglo-Saxon cathedral would have been built on a religious site even older than the site of St Helen's.

¹⁰⁹ Burials date d by their excavator to the late Anglo-Saxon period were found a short distance to the east: Clarke, H., ‘Excavations at Worcester Cathedral 1970-71’ in Carver, , op. cit. (note 53), 133–110Google ScholarBaker, , op. cit. (note 55), 35, 37Google Scholar.

¹¹¹ Whitelock, et al., op. cit. (note 76), 637Google Scholar. On the question of its authenticity see note 78.

¹¹² Macray, W. D. (ed.), Chronicon Abbatiae de Evesham ad annum 1418, Rolls Series 29 (1863), 211, 223Google Scholar.

¹¹³ e.g. Bund, Willis, op. cit. (note 73), 5, 455Google Scholar; Bund, J. W. Willis, The Register of the Diocese of Worcester during the Vacancy of the See, usually called ‘Registrum sede vacante’, Worcestershire Hist. Soc. 8 (1897), 428Google Scholar; Valor, III, 253b.

¹¹⁴ St Margaret's is said to have been granted, with Evesham's consent, to the rector of St Andrew's, Worcester, c. 1066, and to have stood on the latter church's land: VCH Worcs., iv (1924), 411 and n. 59 (citing B. L. Cotton MS Vespasian B. xxiv, fol. iib, 13, which I have not been able to see).

¹¹⁵ Macray, , op. cit. (note 112), 73Google Scholar.

¹¹⁶ I can find nothin g in the alleged notification of the 1092 synod to show that the priest of St Alban's was there claiming to have precedenc e over St Helen's. This often repeated misreading originates in I. Atkins, ‘The church of Worcester from the eighth to the twelfth century, part II’, Antiq. J. 20 (1940), 206. For the rivalry between Evesham and Worcester, see Darlington, R. R., ‘ALthelwig, Abbot of Evesham’, Engl. Hist. Rev. 48 (1933), 1-22, 177–98CrossRefGoogle Scholar.

¹¹⁷ The account was the work of Thomas de Marleberge, prior, then abbot, of Evesham. ‘Where he can be checked, he will be found trustworthy, and his pages give an impression of honesty and precise statement’: Knowles, D., The Monastic Order in England (Cambridge, 1949), 331Google Scholar n. 1. ‘He narrates with great care the past history of the abbey, quoting, as seems certain, from earlier documents of some fullness’: ibid., 506.

¹¹⁸ H. M. and Taylor, J., Anglo-Saxon Architecture, 3 vols. (Cambridge, 1965-1978), in, 1020–1Google Scholar.

¹¹⁹ As opposed to its eventually serving as a parish church for the minster if the latter became properly monastic: Bassett, , op. cit. (note 46), 88–9Google Scholar. St Alban's had a cemetery by 1630 (the date of its first register: VCH Worcs., iv, 409), but it is not known when it originated.

¹²⁰ Baker, , op. cit. (note 55), 33–4, 37Google Scholar; Bond, , op. cit. (note 49), 130–2Google Scholar(where Helena and Margaret are mistakenly identified as Romano-British saints). Arnold-Forster, F., Studies in Church Dedications, 3 vols. (London, 1899), 1, 131–3, 181-9Google Scholar; 11, 294-9. Levison, however, listed no dedications to Helena attested for seventh- and eighth-century English churches: Levison, W., England and the Continent in the Eighth Century (Oxford, 1946), 259–65Google Scholar.

¹²¹ Hunt, E. D., Holy Land Pilgrimage in the Later Roman Empire AD 312-460 (Oxford, 1982), 41–2.Google Scholar(This work provides an excellent accoun t of the historical (Flavia) Helena.) For the development of the specifically British elements of her legend, see Thomas, , op. cit. (note 7), 41Google Scholar; Matthews, J.F., ‘Macsen, Maximus and Constantine’, Welsh Hist. Rev. 11 (1982-1983), 445–7Google Scholar(I owe this reference to John Blair).

¹²² H.E. 1.18 ( Colgrave, and Mynors, , op. cit. (note 2), 58–60)Google Scholar.

¹²³ Brown, P., The Cult of the Saints. Its Rise and Function in Latin Christianity (Chicago and London, 1981), 6–8, 50-68Google Scholar. But Richard Morris comments (pers. comm.), ‘I doubt if there would have been dedications to Helena in the fourth or fifth century, since Constantine and Helena were something of an embarrassment to the fourth-century church on the continent, and I am not aware of any continental interest in her cult at this period.’ 124 Baker, op. cit. (note 56), 119, citing pers. comm. from Christophe r Dyer; Arnold-Forster, op. cit. (note 120), 1,148-51. For Felicita s (not to b e confused with the second-century marty r Felicita s of Rome), seeibid., 150. The two appear together in the Old English Martyrology as th e saints commemorated on 7 March: Kotzor, G., Das Altenglische Martyrologium, Bayerische Akad. der Wissenschaften, Phil.-Hist. Klasse, Abh., N.F. 88, 2 vols. (Munich, 1981), II, 29–30Google Scholar(I a m grateful to Donald Bullough for this reference). The reference of 969 to St Peter [the Great] is in Birch, , op. cit. (note 64), in, no. 1240Google Scholar( Sawyer, , op. cit. (note 50), no. 1327)Google Scholar.

¹²⁵ Baker, , op. cit. (note 55), 37Google Scholar, and pers. comm. 126 Arnold-Forster, , op. cit. (note 120), 11, 297; 111, 321, 352Google Scholar.

¹²⁷ Bassett, , op. cit. (note 46), 93–4Google Scholar.

¹²⁸ e.g. Abingdon: Biddle, M., Lambrick, H. T. and Myres, J. N. L., ‘The early history of Abingdon, Berkshire, and its abbey’, Medieval Archaeol. 12 (1968), 27–9CrossRefGoogle Scholar; Colchester: Rodwell, W. J., Historic Churches—a Wasting Asset, CBA Res. Rep. 19 (London, 1977), 37–8Google Scholar; York: Magilton, J. R., The Church ofSt Helen-on-the-Walls, Aldwark, York Archaeol. Trust, fasc. 10/1 (York, 1980), 16–18Google Scholar.

¹²⁹ Arnold-Forster perpetuates the notion that Margaret is named in ‘English litanie s of the seventh century’ (op. cit. (note 120), 1, 131). Recent writers have take n this to show that a dedication to her is potentially pre-Anglo-Saxon. No seventh-century litanies from England survive, however, so that the notion may have been originally based on an attempt to uncover ‘layers’ in the Old English Martyrology (Donald Bullough, pers. comm.), of which the earliest MS to include Marina/Margarita of Antioch is of tenth-/eleventh-century date: Kotzor, , op. cit. (note 124), 11, vii, 141–4Google Scholar.

¹³⁰ James Bond suggests that St Helen's ‘bears all the hallmarks of a clas church, the mother-church in the Celtic system’ (op. cit. (note 49), 130, following Whitehead, D., The Book of Worcester. The Story of the City's Past (Chesham, 1976), 18Google Scholar). Besides its extensive parish, he cites as evidence its possession of nine houses at the start of the twelfth century: ‘In burgo… ad ipsam ecclesiam pertinent nouem mansure circa ecclesiam’ ( Darlington, , op. cit. (note 57), 33Google Scholar). But neither he nor Whitehead mentions that St Alban’s and St Margaret's were also said to have houses around them, allegedly in 721: ‘… capellas sancti Albani et sanctae Margaretae in Wigornia, cum pluribus domibus circumjacen-tibus’ ( Macray, , op. cit. (note 112), 73Google Scholar). Surely not all three of them were clas churches: it is much likelier that each held a number of town houses in its vicinity in the later medieval period, in common with many other urban churches. (Evesham Abbey had 28 masuras in Worcester in 1086 (D.B. 1, fol. I75d), some of which were no doubt attached to its two churches there.)

¹³¹ Thomas, , op. cit. (note 7), 157–8Google Scholar.

¹³² Gilmour, B., ‘The Anglo-Saxon church at St Paul-in-the-Bail, Lincoln’, Medieval Archaeol. 23 (1979), 214–18CrossRefGoogle Scholar; Rodwell, W. J., ‘Churches in the landscape’ in Faull, M. L. (ed.), Studies in Late Anglo-Saxon Settlement (Oxford, 1984), 4Google Scholar.

¹³³ e.g. Canterbury (Christchurch): H.E. 1.33 ( Colgrave, and Mynors, , op. cit. (note 2), 114Google Scholar); Colchester (St Martin's): Rodwell, , op. cit. (note 128), 29–30Google Scholar; Exeter: Bidwell, P. T., The Legionary BathHouse and Basilica and Forum at Exeter, with a Summary Account of the Legionary Fortress, Exeter Archaeol. Rep. 1 (Exeter, 1979), 110–14Google Scholar; Thomas, , op. cit. (note 7), 168Google Scholar; and not inconceivably London (St Peter Cornhill): Wheeler, R. M., London and the Saxons, London Museu m Catalogues no. 6 (London, 1935), 102–3Google Scholar.

¹³⁴ Postscript. Dr R. A. Holt has shown me, and very kindly permitted me to mention here, evidence from original deed s in Worcester Cathedral Library (D and C muniments) in respect of the location of St Margaret's church. This demonstrates that it lay, not where mapped in fig. 6, but to the west-southwest of St Alban's (where Baker, op. cit. (note 55), fig. 9, shows an unnamed chapel).