1. Research on the Compton Census has been enormously advanced by Whiteman, A. (ed), The Compton Census of 1676: A Critical Edition (Oxford, 1986),Google Scholar with the assistance of M. Clapinson. This major and finely documented achievement leaves historians forever indebted to Whiteman, and our work made heavy use of it. See also Whiteman, A., ‘The Compton census of 1676’, in Schurer, K. and Arkell, T. (eds), Surveying the People (Oxford, 1992), pp. 78–96;Google ScholarWhiteman, A. and Clapinson, M., ‘The use of the Compton census for demographic purposes’, Local Population Studies, 50 (Spring, 1993), 61–6.Google Scholar

2. Whiteman, A., ‘Compton census’, in Schurer, and Arkell, , Surveying the People, p. 81.Google Scholar

3. See Whiteman, , Compton Census, p. lxiv,Google Scholar a conclusion she reaches after making comparisons with the Protestation Returns of 1641–2. See also Wrigley, E.A. and Schofield, R.S., The Population History of England, 1541–1871: A Reconstruction (1981), pp. 33–7, 570.Google Scholar They tested the 1676 data in various ways, using it to check the representativeness of their 404 parishes.

4. Whiteman, , Compton Census, p. lxxx,Google Scholar but she doubts that any serious omissions took place.

5. For discussion of the Compton census, see in particular Whiteman's work, noted above, and her select bibliography, Compton Census, pp. 647–55; Cox, J.C., ‘A religious census of Derbyshire, 1676’, Journal of the Derbyshire Archaeological and Natural History Society, VII (1885), 31–6;Google ScholarFletcher, W.G.D., ‘Religious census in Leicestershire in 1676’, Transactions of the Leicestershire Architectural and Archaeological Society, VI (1887), 296–303;Google ScholarFletcher, W.G.D., ‘Religious census of Shropshire in 1676’, Transactions of the Shropshire Archaeological and Natural History Society, I (1889), 75–92;Google ScholarMooney, W., ‘A religious census of Berkshire in 1676’, Berkshire, Buckinghamshires and Oxfordshire Archaeological Journal, IV (1889), 112–5, and V (1900), 55–9;Google ScholarCooper, J.H., ‘A religious census of Sussex in 1676’, Sussex Archaeological Collections, XLV (1902), 142–8;Google ScholarLangley, A.S., ‘A religious census of 1676, A.D.’, Lincolnshire Notes and Queries, XVI:2 (April, 1920), 33–51;Google ScholarGuilford, E.L., ‘Nottinghamshire in 1676’, Transactions of the Thoroton Society, XXVIII (1924), 106–13;Google ScholarRichards, T., ‘The religious census of 1676: an inquiry into its historical value, mainly in reference to Wales’, supplement to the Transactions of the Honourable Society of Cynmrodorion (1925–7), 14–30;Google ScholarPeyton, S.A., ‘The religious census of 1676’, English Historical Review, XLVIII (1933);CrossRefGoogle ScholarJames, F.G., ‘The population of the diocese of Carlisle in 1676’, Transactions of the Cumberland and Westmorland Antiquarian and Archaeological Society, LI (1952), 137–41;Google ScholarStephens, W.B. ‘A seventeenth century census’, Devon and Cornwall Notes and Queries, XXIX (1958);Google ScholarChalkin, C.W. (ed), The Compton Census for the Dioceses of Canterbury and Rochester, Kent Archaeological Society Records, XVII (1960);Google ScholarSpufford, M., ‘The dissenting churches in Cambridgeshire from 1660–1700’, Proceedings of the Cambridgeshire Antiquarian Society, LXI (1968), 67–95;Google ScholarStanes, R., ‘The Compton census for the Diocese of Exeter, 1676’, Devon Historian, 9 (1974), 14–27, and 10 (1975), 4–16;Google ScholarWykes, D., ‘A reappraisal of the reliability of the 1676 ‘Compton Census' with respect to Leicester’, Transactions of the Leicestershire Archaeological and Historical Society, 60 (1980), 72–7;Google ScholarJackson, P., ‘Nonconformity and the Compton census in late seventeenth-century Devon’, in Schurer, K. and Arkell, T. (eds), Surveying the People (Oxford, 1992), 117–129.Google Scholar

6. See Whiteman's general introduction and appendices to Compton Census, and her ‘Compton census’; Arkell, T., ‘A method for estimating population totals from the Compton census returns’, in Schurer, and Arkell, , Surveying the People, pp. 97–116;Google ScholarJackson, P., ‘Nonconformity and the Compton census in late seventeenth-century Devon’, also in Schurer, and Arkell, , Surveying the People, pp. 117–24.Google Scholar

7. See e.g. Mann, H., Report on Religious Worship in England and Wales, Parliamentary Papers, lxxxix (1852–3);Google ScholarMann, H., ‘On the statistical position of religious bodies in England and Wales’, Journal of the Statistical Society, XVIII (1855);Google ScholarHume, A., Remarks on the Census of Religious Worship for England and Wales (1860);Google ScholarInglis, K.S., ‘Patterns of religious worship in 1851’ Journal of Ecclesiastical History, XI (1960), 74–86;CrossRefGoogle ScholarRogan, J., ‘The Religious Census of 1851’, Theology (1963), 11–15;Google ScholarPickering, W.S.F., ‘The 1851 Religious Census – a useless experiment?’, British Journal of Sociology, XVIII (1967), 382–407;CrossRefGoogle ScholarThompson, D., ‘The 1851 Religious Census: problems and possibilities’, Victorian Studies, XI (1967), 87–97;Google ScholarAmbler, R.W., ‘The 1851 Census of Religious Worship’, Local Historian, XI (1975), 375–81;Google ScholarThompson, D.M., ‘The Religious Census of 1851’, in Lawton, R. (ed), The Census and Social Structure (1978);Google ScholarField, C.D., ‘The 1851 Religious Census: a select bibliography’, Proceedings of the Wesley Historical Society, XLI (1978);Google ScholarSnell, K.D.M., Church and Chapel in the North Midlands: Religious Observance in the Nineteenth Century (Leicester, 1991), ch. 2;Google ScholarEll, P.S., An Atlas of Religious Worship in England and Wales: An Analysis of the 1851 Census of Religious Worship (unpub. Univ. of Birmingham Ph.D, 1992).Google Scholar

8. There are many ways of handling such data. For parish-level analyses, the most common method is to express the attendance and sittings data in relation to parochial population, as indexes of sittings or attendances. Thus a denominational ‘index of total attendances’ is the sum of attendances for the three possible services during the day, over parish population, multiplied by 100. An ‘index of maximum attendance’ uses only the denomination's maximum attendance for a comparable calculation. An ‘index of sittings’ uses in a similar way the total figure for denominational sittings in the parish. By contrast, ‘percentage share measures’ express a denomination's attendance or sittings data as a percentage of the parish total for all denominations. These and other measures have different properties that need not preoccupy us here. One needs to note that the source did not give numbers of attendants, but rather of attendances.

9. Of the published returns we have used Jones, I.G. and Williams, D. (eds.), The Religious Census of 1851: A Calendar of the Returns Relating to Wales, vol. I: South Wales (Cardiff, 1976);Google ScholarJones, I.G. (ed.), The Religious Census of 1851: A Calendar of the Returns Relating to Wales, vol. II: North Wales (Cardiff, 1981);Google ScholarVickers, J.A. (ed.), The Religious Returns of Sussex, 1851 (Sussex Record Society, 75, 1986–7);Google ScholarBushby, D.W. (ed.), Bedfordshire Ecclesiastical Census, 1851, (Bedfordshire Historical Record Society, volume 54, Luton, 1975).Google Scholar The Derbyshire data was kindly made available to us and subsequently published by Tranter, M. (ed.), The Derbyshire Returns to the 1851 Religious Census (Derbshire Record Society, XXIII, Chesterfield, 1995).Google Scholar Data for our other counties were obtained from the Public Record Office.

10. ‘Introducing the documents’ in Schurer and Arkell, Surveying the People, p. 34.Google Scholar See also Spufford, M., ‘The dissenting churches in Cambridgeshire from 1660 to 1700’, Proceedings of the Cambridge Antiquarian Society, LXI (1968), 95.Google Scholar

11. See Whiteman, , Compton Census, pp. lix–lxxvi.Google Scholar

12. On Easter books, see Wright, S., ‘Easter books and parish rate books: a new source for the urban historian’, Urban History Yearbook (1985);Google Scholar and her ‘A guide to Easter Books and related parish listings’, Local Population Studies, pt. I, 42, pt. II, 43 (1989).Google Scholar On the demographic relation between 1676 and 1811, see Whiteman, , ‘Compton census’, p. 88Google ScholarWrigley, E.A. and Schofield, R.S., The Population History of England, 1541–1871: A Reconstruction (1981), pp. 33–7.Google Scholar These historians were not of course concerned with appraising religious continuities as manifest from the seventeenth- and nineteenth-century sources.

13. An earlier description of this geography may be found in Gay, J.D., The Geography of Religion in England (1971).Google Scholar For fuller and more detailed cartographic analysis, see Ell, Atlas of Religious Worship.

14. Our wider research analyses all the English and Welsh registration-district data; and secondly the parish-level documentation for fifteen counties, twelve of which are used for this article. The other three counties (omitted here because they lack adequate Compton data) are Dorset, Lancashire and Northumberland. This forms part of an extensive project examining the 1851 religious and Sunday school data in conjunction with a wide array of socio-economic and demographic variables, mainly for the mid nineteenth century, but with some earlier religious sources.

15. On the Hearth Tax data, see in particular Alldridge, N. (ed.), The Hearth Tax: Problems and Possibilities (Hull, 1983);Google ScholarPatten, J., ‘The Hearth Taxes, 1662–89’, Local Population Studies, 7 (1971), 14–27.Google Scholar

16. See her discussion in Compton Census, pp. lii-liv, lxxxvi-xci.

17. Additional tests were carried out to check for rounding based on multiples of twelve. The results were all negative (i.e. very close to the 8.3% expected by chance alone), confirming Whiteman's judgement on the possibility of this form of tabulation.

18. Probabilities (p) of 0.0000 are of course highly significant results, (p = 0.05 corresponds to a 95% confidence level).

19. The index of total attendances was also used (as well as the percentage share), but the results were almost identical. For reasons of consistency with the Compton ratio measure, only the percentage share measure is reported here.

20. Rank correlation was the preferable method for this data, which is often non-normally distributed statistically.

21. This is probably the best summary measure, given the lack of absolute comparability across all parishes of the Compton data.

22. Spearman and Pearson correlations were also conducted for these religious groups using the 1851 index of attendances against the 1676 data, and the results were very similar to those reported here. In both these exercises, all Compton-documented parishes were included in the analysis, and this involved making use of many places which in 1676 had no nonconformist or Papist presence. Correlations were also undertaken by only including parishes which had Compton figures greater than zero for nonconformists or Papists. One could debate the methodology and historical interpretation of these differing methods, but the overall coefficients resulting were alike for all procedures.

23. The computerised cartography was carried out using ESRI's ARCVIEW software package linked to the SPSS dataset, the parish boundaries having been originally digitised using the GIMMS software package.

24. Jenkins, P., ‘“A Welsh Lancashire”? Monmouthshire Catholics in the eighteenth century’, Recusant History, 15 (1979–81), 176.CrossRefGoogle Scholar

25. Watkin, E.I., Roman Catholicism in England from the Reformation to 1950 (1950), p. 110.Google Scholar

26. See also Attwater, D., The Catholic Church in Modern Wales (1935);Google ScholarHowell, P., ‘Church and chapel in Wales’, in Brooks, C. and Saint, A. (eds.), The Victorian Church: Architecture and Society (Manchester, 1995), p. 128.Google Scholar

27. Jenkins, , ‘“A Welsh Lancashire”?’, 177.Google Scholar

28. Hemphill, Dom B., The Early Vicars Apostolic of England, 1685–1750 (1954), pp. 143–5.Google Scholar

29. And see also Matthews, J.H., The Vaughans of Courtfield (1912),Google Scholar on a further key family with interests in north Monmouthshire.

30. Or in Llanarth ‘there are as near as I can guess 110 [?] Papists – consisting of poor and wealthy Inhabitants. There are occasional perversions: brought about I believe through the influence of the wealthy Proprietors’. For these and other similar cases, see National Library of Wales, Visitation Returns: Deaneries of Abergavenny and Monmouth. Clergy 1848. LL/QA/36. Or see Hemphill, , Early Vicars Apostolic, pp. 103–4,Google Scholar quoting Joseph Berington in 1780: ‘Excepting in the towns and out of Lancashire, the chief situation of Catholics is in the neighbourhood of the old families of that persuasion. They are the servants, or the children of servants who have married from those families’.

31. Guy, J.R., ‘The Anglican Patronage of Monmouthshire recusants in the seventeenth and eighteenth centuries’, Recusant History, 15 (1979–81), 453;Google Scholar and his ‘Eighteenth-century Gwent Catholics’, Recusant History, 16 (1982–3), 78–88.Google Scholar

32. Guy, , ‘Anglican patronage’, 82.Google Scholar

33. Elliott, B., ‘A Leicestershire recusant family: the Nevills of Nevill Holt’, Recusant History, 1st part, 17:2 (1984), 173–80;CrossRefGoogle Scholar2nd part, Recusant History, 17:4 (1985), 374–85;Google Scholar3rd part, Recusant History, 18:2 (1986), 220–24;Google ScholarElliott, B., ‘The history of Catholicism in Market Harborough’, Harborough Historian, 2: (1985), 2–3;Google ScholarBroughton, H.E. (ed), Nevill Holt: Studies of a Leicestershire Estate (Leicester, 1985).Google Scholar

34. Husband's Bosworth was the seat of the Turville Constable Maxwell family. There was Catholic influence here from the early seventeenth century, well documented by family papers in Bosworth Hall Library. See Leics. C.R.O., Catalogue DG 39. 54 on the family MSS. See also Leics. C.R.O., QS 45/1/23; QS 45/3/2 (1748) (Papist estates). Resident Roman Catholic priests at the Hall were buried in the parish churchyard. The parish had 80 Catholic ‘followers’ recorded in 1829, the highest in Leicestershire. (Leics. C.R.O., QS 95/2/2). The Pugininfluenced Catholic Church of St. Mary was built in the grounds of Bosworth Hall in 1873–4. See also Pugh, R.B. (ed), The Victoria County History of the Counties of England (1964), vol. 5, p. 36.Google Scholar

35. Elliott, B., ‘The return of the Cistercians to the Midlands’, Recusant History, 16 (1982–3), 99–104;CrossRefGoogle ScholarJewitt, L., Guide to the Abbey of Mount St. Bernard (4th edn, 1897);Google ScholarLacey, A.C., The Second Spring in Charnwood Forest (Loughborough, 1985).Google Scholar The literature on their later reformatory covers aspects of their earlier history. See Elliott, B., ‘Mount St. Bernard's Reformatory, Leicestershire, 1856–81’, Recusant History, 15:1 (1979), 15–22;CrossRefGoogle Scholar his ‘Mount St. Bernard's Reformatory: a reply’, Recusant History, 15 (1979–81), 302–4;Google ScholarTucker, J.L.G., ‘Mount St. Bernard's Reformatory, 1856–81: a correction’, Recusant History, 15 (1979–81), 213–7.CrossRefGoogle Scholar

36. On the Dominican Order in Hinckley, see Leics. C.R.O., QS 45/8/1–4, giving the returns made under 10 Geo IV, c. 7, s. 28 and 30, and the oaths required under 31 Geo III, c. 32. Catholicism is also documented in the town in Leics. C.R.O., QS 45/7/2 (1791, a chapel in a dwelling house) QS 45/7/3 (1793, a ‘new erected building’) and QS 45/7/4 (1825). For declarations of loyalty to the Crown from Hinckley, Eastwell, Burbatch, Burbage and Eaton Catholic priests and residents, see Leics. C.R.O., QS 45/6/1. The named occupations were widow, gent, labourer, hosier and farmer.

37. Analysis of the Leicestershire Catholic Baptism Registers suggests the extent and chronology of Irish immigration after the Great Famine. In some parishes which had earlier indicated a Catholic presence, the inflow of Irish surnames was small (e.g. Hinckley St. Peter, Shepshed St. Winifred, Melton Mowbray St. John, and perhaps Measham St. Charles, although there were evidently Irish there). But in the fast expanding mining town of Whitwick there was a huge increase in the Irish from the late 1840s, with well over 50 Irish surnames present by the early-mid 1850s which had not appeared earlier. (Leics. C.R.O., Catholic Baptism Registers for the above churches, and for Whitwick Holy Cross, 1843–66). The mining rather than hosiery towns appear to have had greatest inflow. Other Irish at this time and in later decades are documented as on tramp near these settlements, Irish Catholic mothers bringing their children for baptism, ‘of course pro Deo, they being deadly poor!!’, one priest commented. (Leics. C.R.O., Melton Mowbray St. John, Catholic Baptisms, 1843–97).

38. Pevsner, N., Yorkshire: York and the East Riding (Harmondsworth, 1972), p. 226.Google Scholar However, most Catholic architecture was the fashionable Gothic, by architects like Hansom and Pugin, as at Sicklinghall or Leeds.

39. Aveling, H., Post Reformation Catholicism in East Yorkshire, 1558–1790 (York, 1960), p. 47.Google Scholar And see Linker, R.W., ‘English Catholics in the eighteenth century’, Church History, 35 (1966), 288–310;CrossRefGoogle ScholarHemphill, , Early Vicars Apostolic, p. 83.Google Scholar

40. Longley, K.M., Heir of Two Traditions: the Catholic Church of St. John the Baptist, Holme-on-Spalding-Moor, 1766–1966 (1966).Google Scholar

41. Aveling, , Post Reformation Catholicism, p. 46.Google Scholar

42. Aveling, J.C.H., The Handle and the Axe: the Catholic Recusants in England from Reformation to Emancipation (1976), p. 286, and 301–2,Google Scholar where he points to the difficulties in holding the rural poor to Catholic rituals, Latin prayers and irregular masses, especially with the decline in resident Catholic gentry. One might add that the agricultural labouring poor in eastern regions were often strongly anti-Irish, accustomed as they were to harvest and other competition from Catholic Irish. There was by this time little of that sense of loss which had characterised rural areas in earlier centuries, when the cycle of festivals and other Catholic attributes were abandoned. Some of our other census counties, discussed in less detail here, show considerable discontinuitiesof Catholicism. In Rutland, a noticeable Catholic presence in 1676 was replaced by no documentation whatever of Catholics in 1851. In Anglesey and Caernarvonshire, a scattering of Papists in Llandudno, Eglwys Rhos and surrounding parishes in 1676, with a few others dispersed elsewhere, had given way to an 1851 congregation only in Bangor. In Suffolk, there was only one parochial example of continuity between the two sources (Stoke-by-Nayland), although there had been 39 parishes with Papists in 1676. None of the Suffolk parishes with over 10 per cent Papist in 1676 (Long Melford, Flempton, Bulmer, Stanningfield and Wetterden) had any Catholic places of worship in 1851. In Bedfordshire, hardly a Catholic stronghold, there was not a single case of continuity between the two dates. In Sussex, of 44 parishes with Papists in 1676, only two had a Catholic place of worship documented in 1851 (Arundel and Slindon). None of the Sussex parishes with over 10 per cent Papist in 1676 (Burton, Clapham, Coates, Midhurst, Racton, Shipley or Westfire) had Catholic places of worship in 1851. Derbyshire had six out of 38 parishes showing continuity between 1676 and 1851 (Bakewell, Chesterfield, Eckington, Glossop, Hathersage and Tideswell). Hathersage was very strongly Catholic at both dates, but other parishes which had conspicuous Catholicism in 1676 (notably Carlton and West Hallam) had no Catholic venues in 1851.

43. Bossy, , ‘Four Catholic congregations’, 110.Google Scholar

44. Hemphill, , Early Vicars Apostolic, p. 78.Google Scholar For a good example of this in East Lulworth, Dorset, where a third of the inhabitants were Catholic in 1766, see Biggs, B.J., The Wesleys and the Early Dorset Methodists (Gillingham, Dorset, 1987), p. 18.Google Scholar With regard to the effects of charity, dryly, R. Southey wrote that ‘proselytes always abound in the neighbourhood of a wealthy Catholic family’. Letters from England (1807, Gloucester, 1984 edn.), p. 157.Google Scholar

45. Watkin, E.I., Roman Catholicism in England from the Reformation to 1950 (1957), p. 115.Google Scholar

46. See e.g. Holt, T.G., ‘An eighteenth century Chaplain: John Champion at Sawston Hall’, Recusant History, 17:2 (1984), 181–7;CrossRefGoogle ScholarPevsner, , Cambridgeshire, p. 368;Google ScholarHistory, Gazetteer, and Directory of Cambridgeshire (no author, Peterborough, 1851), p. 266.Google Scholar The Hall, with its secretive priest hole, had been burnt after Queen Mary spent a night there in 1553.

47. Bossy, J., ‘More Northumbrian congregations’, Recusant History, 10:1 (1969), 12–13.CrossRefGoogle Scholar

48. Bossy, , ‘More Northumbrian congregations’, p. 24;Google Scholar and see his ‘Four Catholic congregations in rural Northumberland, 1750–1850’, Recusant History, 9 (1967), 88–119.CrossRefGoogle Scholar

49. Aveling, , Post Reformation Catholicism, p. 52.Google Scholar

50. See e.g. Bellenger, D., ‘The English Catholics and the French exiled clergy’, Recusant History, 15 (1979–81), 433–51;CrossRefGoogle ScholarIsherwood, F.P., Banished by the Revolution (Jersey, 1972);Google ScholarAveling, , Handle and Axe, pp. 308–18.Google Scholar The priests began to come over after the Civil Constitution of 1791, as did many English Catholics living in France. Robert Southey, no friend to Catholicism, commented on how ‘The English clergy, trembling for their own benefices, welcomed the emigrant priests as brethren … the Catholic priests obtained access everywhere’, Letters from England (1807, Gloucester, 1984 edn), p. 155.Google Scholar Nearly 900 Catholic chapels opened between 1791 and 1814, notably in the north. (Watkin, Roman Catholicismin England, p. 158.) These priests certainly played a role in places like Arundel, Slindon, Abergavenny, or some of the coastal settlements of the East Riding, although their confident, state-centred style of Catholicism was very different to that found in England.

51. Aveling, , Handle and Axe, pp. 358–9.Google Scholar

52. The 1829 returns originated from a resolution of the House of Commons on 19 June 1829 calling for ‘A Return of the number of Places of Worship not of the Church of England in each Parish, distinguishing as far as possible of what sect or persuasion, and the total number of each sect in England and Wales’. Many of the returns were burnt in the 1834 conflagration at the Houses of Parliament, but copies survive for some counties in Quarter Session records. In effect they record places of worship and the numbers of ‘followers’, ‘adherents’ or ‘members’ reported by each parish, there being some doubt over the precise categories returned and how distinct those were. (The forms often used the term ‘adherents'). And so problems of comparability arise once again when one relates these returns to the other two sources. For example, for Leicestershire Catholicism in 1829 one has Husband's Bosworth (80), Eastwell (25), Eaton (4), Hinckley (40), Holt (48), Hose (5) and an unspecified figure for Leicester. Bracketed figures are for numbers of adherents (subject to the above doubt). This format is quite different to that of 1676 or 1851, but it still describes a Catholic picture immediately recognisable in 1851. The 1829 returns may be found in Leics. C.R.O., QS 95/2/2. For further documentation on their purpose and interpretation, see Leics. C.R.O., QS 95/2/3/4/2, and QS 95/2/3/1–2. Many of the returns (like Hose or Eaton above) document only a handful of adherents, and the numbers are in some cases so small that they must describe house meetings rather than chapels or purpose-built places of worship. It is unclear whether the numbered adherents were only of those resident in the returning parish, and there may be some parochial variation in this. On the 1829 returns, see Ambler, R.W., ‘A lost source? The 1829 returns of non-Anglican places of worship’, The Local Historian, 17 (1987), 483–89,Google Scholar and see his references; Ambler, R.W., ‘Religious life in Kesteven – a return of the number of places of worship not of the Church of England, 1829’, Lincolnshire History and Archaeology, 220 (1985), 59–64;Google ScholarAmbler, R.W., ‘Social Change and Religious Experience: Aspects of Rural Society in South Lincolnshire with Special Reference to Primitive Methodism, 1815–1875’ (unpub. University of Hull Ph.D., 1984), 148–53;Google ScholarCaplan, N., ‘Sussex religious dissent, c. 1830’, Sussex Archaeological Collections, 120 (1982), 193–203;Google ScholarTranter, M., “Many and diverse dissenters‘ – the 1829 religious returns for Derbyshire’, The Local Historian, 18:4 (1988), 162–7;Google ScholarGill, R., The Myth ofthe Empty Church (1993), pp. 96, 107.Google Scholar

53. The Mormons and the 1851 census category of ‘other isolated congregations’ were excluded from this calculation.

54. This strongly dissenting parish was perched on the border, in the historical county of Warwickshire, but in the registration-district county of Leicestershire: an example perhaps of strategic use of the county boundary to help avoid local authorities' jurisdiction.

55. Parishes like Scalford, Goady Marwood, Waltham on the Wolds, Thorpe Arnold, Saltby, Sproxton, Coston, Edmondthorpe, Saxby or Wymondham.

56. The parish of Eastwell had 17 per cent of its Compton population as Papist, and an exceptionally high percentage share for the Roman Catholics in 1851 of 87.8%. A Catholic chapel had been built around 1806 (another source gives 1798), apparently in lieu of one at the Hall which had been destroyed. The parish had been owned by the Eyres from 1631, and was purchased by the Duke of Rutland in the very early nineteenth century. See Kelly's Directory of Leicester and Rutland (1922 edn), pp. 76–7;Google ScholarWright's Directory of Leicestershireand Rutland (Leicester, 1896), p. 544;Google ScholarWhite, W., History, Gazetteer, and Directory of Leicestershire (Sheffield, 1846), pp. 233–4.Google Scholar The 1851 religious data suggest very ineffective Anglican competition in this parish. The Catholic traditions of Eastwell can be traced through the eighteenth century in the Leics. C.R.O. See QS 45/1/4, QS 45/1/15, QS 45/3/4 and QS 45/1/21 (1737, 1739, 1750 and 1752 registrations of Catholic estates by Roland Eyre of Eastwell Hall) QS 45/1/25/1 and QS 45/2/52 (1777); QS 45/1/26 and QS 45/2/53 (1785); and QS 45/7/1 (1791 registration of a chapel in a dwelling house, by Robert Beeston).

57. On the large increase of Catholics in England and Wales in the half century after 1851, from around 700,000 to over 1,500,000 by 1900, see Currie, R., Gilbert, A. and Horsley, L., Churchesand Churchgoers: Patterns of Church Growth in the British Isles since 1700 (Oxford, 1977), p. 153;Google ScholarNorman, E., The English Catholic Church in the Nineteenth Century (Oxford, 1984), pp. 205–6.Google Scholar On the urban impact of the Irish Catholics after the 1840s, see e.g. Lees, L.H., Exiles of Erin: Irish Migrants in Victorian London (Manchester, 1979);Google ScholarHickey, J., Urban Catholics: Urban Catholicism in England and Wales from 1829 to the Present Day (1967);Google ScholarJackson, J.A., The Irish in Britain (1963);Google ScholarO'Tuathaigh, M.A.G., ‘The Irish in nineteenth-century Britain: problems of integration’, Transactions of the Royal Historical Society, series 5, 3 (1981), 149–73;CrossRefGoogle ScholarGilley, S., ‘The Roman Catholic mission to the Irish in London, 1840–1860’, Recusant History, 10:3 (1969), 123–5.CrossRefGoogle Scholar

58. All calculations in this section only use the 1,463 parishes which had data for 1676, so as to keep the parochial basis for comparison over time identical. If however one takes all 1,990 parishes in the same counties for which data is available in 1851, the equivalent figure here would be even higher, at 19.3 per cent for ‘old dissent’.

59. For example, in considering the relation between population growth and denominations, we did not make use of demographic growth rates between 1676 and 1851, as calculated from these sources, although they gave similar results to those of table 5. We do not wish to preclude the demographic use of the 1676 data in any way, and some experts have suggested that it is best suited for such purposes. See in particular Whiteman, , Compton Census, p. lxxxii;Google ScholarWhiteman, A. and Clapinson, M., ‘The use of the Compton census for demographic purposes’, Local Population Studies, 50 (Spring 1993), 61–6.Google Scholar

60. Spufford, M. (ed), The World of Rural Dissenters, 1520–1725 (Cambridge, 1995), pp. 200–4. See also pp. 179–180.Google Scholar

61. Spufford, , ‘The dissenting churches in Cambridgeshire’, 81.Google Scholar

62. Whiteman, , ‘Compton census’, p. 92Google Scholar and see Spufford's view that ‘great caution must be employed in using the census, and that it is a difficult or impossible source on which to base estimates of total population, even if not of dissenters’. (Spufford, ‘The dissenting churches in Cambridgeshire’, p. 95.)

63. Similar problems have much preoccupied historians of the period since 1851. See e.g. Gill, R., Competing Convictions (1989);Google ScholarGill, R., The Myth of the Empty Church (1993).Google Scholar

64. On the changes in traditional patterns of Irish migration, see Kerr, B.M., ‘Irish seasonal migration to Great Britain, 1800–1838’, Irish Historical Studies, 3 (1942–3), 365–80;CrossRefGoogle ScholarLawton, R., ‘Irish immigration to England and Wales in the mid-nineteenth century’, Irish Geography, 4 (1959);CrossRefGoogle ScholarJohnson, J.H., ‘Harvest migration from nineteenth-century Ireland’, Transactions of the Institute of British Geographers, XLI (1967), 97–112;CrossRefGoogle ScholarO'Dowd, A., Spalpeens and Tattie Hokers: the History and Folklore of the Irish Migratory Agricultural Worker in Ireland and Britain (Dublin, 1991).Google Scholar

65. See e.g. Brooks, C. and Saint, A. (eds), The Victorian Church: Architecture and Society (Manchester, 1995), p. 13.Google Scholar Even in 1851, almost two-thirds of Catholics in England had been born in Ireland. See Aveling, , Handle and Axe, p. 19.Google Scholar

66. For example, see the emphases on continuity in earlier centuries of local dissenting traditions in the Chilterns and Cambridgeshire, in Spufford, World of Rural Dissenters.

67. Whiteman, , Compton Census, pp. lxxvi-lxxvii.Google Scholar

68. Whiteman pointed out with regard to the 1669 Conventicles Return that ‘Once the distribution of Dissent has been put on a geographical basis, independent of parochial, archidiaconal and diocesan confines, the discrepancies between the 1669 and 1676 figures are much reduced’, Compton Census, p. lxxviii. Or see Carlson's, E. comments, made for south-east Cambridgeshire, on the ‘close correspondence between the Compton Census and church court records’, in Spufford, World of Rural Dissenters, pp. 178–9.Google Scholar Spufford argued elsewhere, with reference to parts of Cambridgeshire, that the Compton census recorded more dissenters than the visitation returns nearest to it in date, and more indeed than ‘impeccable’ nonconformist sources of the same time. (Spufford, ‘The dissenting churches in Cambridgeshire’, 94.)

69. Yates, N., Hume, R. and Hastings, P., Religion and Society in Kent, 1640–1914 (Woodbridge, 1994), p. 15.Google Scholar

70. Whiteman, , Compton Census, p. lxxix.Google Scholar

71. Laslett, P., ‘Clay worth and Cogenhoe’, in his Family Life and Illicit Love in Earlier Generations (Cambridge, 1977).CrossRefGoogle Scholar

72. See Snell, Church and Chapel, ch. 1.