WHITEHAVEN COAST ARCHAEOLOGICAL SURVEY Vol I : Report Text: David Cranstone, Cranstone Consultants Graphics and photography: Simon Roper, Ironbridge Archaeology For The National Trust June 2007 Cranstone Consultants Phone: 0191-482-1037 267 Kells Lane Fax: 0191-487-2343 Low Fell Email: cranconsult@btinternet.com Gateshead Tyne and Wear NE9 5HU Ironbridge Archaeology Phone: 01952 435 945 Ironbridge Gorge Museum Trust Fax: 01952 435 937 Coach Road Email: archaeology@ironbridge.org.uk Coalbrookdale Telford TF8 7DQ CONTENTS VOL I List of Maps/Plans.......................................................................................3 List of Photographs......................................................................................4 Acknowledgements......................................................................................6 Summary.......................................................................................................7 List of Statutory Heritage Designations...........................................8 1. Introduction............................................................................................9 Background to Project.......................................................................9 Geology and Topography..................................................................9 Natural Units.....................................................................................11 2. Methodology.........................................................................................13 Introduction......................................................................................13 Historical Research..........................................................................13 Fieldwork..........................................................................................16 3. Chronological Survey..........................................................................19 Prehistoric.........................................................................................19 Roman................................................................................................21 Later 1st millenium (c 400 to 1150 AD)...........................................22 Medieval............................................................................................24 Medieval Industry.....................................................................28 Post-Medieval to Modern................................................................30 Background...............................................................................30 Saltmaking................................................................................33 Coal Mining..............................................................................36 Roads, waggonways, railways, and water supply.....................54 The Staithes and harbour environs............................................61 Barrowmouth – fishing, alabaster, and gypsum........................69 Quarrying..................................................................................73 Agriculture and Landscape.......................................................75 The Marchon Site......................................................................79 4. Geographical Summary......................................................................83 SW area...........................................................................................83 Barrowmouth.................................................................................83 Marchon/Rhodia Site.....................................................................85 Southeast Area...............................................................................85 The ‘Howgill Ridge’ area..............................................................86 The Harbour environs...................................................................88 5. Management Recommendations........................................................91 6. Conclusion............................................................................................95 Bibliography...............................................................................................97 Selected Photographs...............................................................................105 Appendix 1: Medieval Industry and Agriculture near Whitehaven (Peter King)...............................................................................................135 Appendix 2: Saltom Pit engine house (Simon Chapman)......................142 Appendix 3: Duke Pit fanhouse (Simon Chapman)...............................146 VOL II Appendix 4: Gazetteer of Sites.................................................................... VOL III Maps and Plans.............................................................................................. LIST OF MAPS/PLANS (VOL III) Fig 1. Overall plan of project area, showing map tile areas. Fig 2. Tile A (SW area): site locations (1:5000) Fig 3. Tile A: 1st edition OS 25” survey, reproduced at 1:5000 Fig 4. Tile A: 2nd edition OS 25” survey, reproduced at 1:5000 Fig 5. Tile A: 3rd edition OS 25” survey, reproduced at 1:5000 Fig 6. Tile E: (Barrowmouth mine area): site locations (1:2000) Fig 7. Tile E: 1st edition OS 25” survey, reproduced at 1:2000 Fig 8. Tile E: 2nd edition OS 25” survey, reproduced at 1:2000 Fig 9. Tile E: 3rd edition OS 25” survey, reproduced at 1:2000 Fig 10. Tile B (SE area): site locations (1:5000) Fig 11. Tile B: 1st edition OS 25” survey, reproduced at 1:5000 Fig 12. Tile B: 2nd edition OS 25” survey, reproduced at 1:5000 Fig 13. Tile B: 3rd edition OS 25” survey, reproduced at 1:5000 Fig 14. Tile C (central area): site locations (1:5000) Fig 15. Tile C: 1st edition OS 25” survey, reproduced at 1:5000 Fig 16. Tile C: 2nd edition OS 25” survey, reproduced at 1:5000 Fig 17. Tile C: 3rd edition OS 25” survey, reproduced at 1:5000 Fig 18. Tile D (northern area): site locations (1:2000) Fig 19. Tile D: 1st edition OS 25” survey, reproduced at 1:2000 Fig 20. Tile D: 2nd edition OS 25” survey, reproduced at 1:2000 Fig 21. Tile D: 3rd edition OS 25” survey, reproduced at 1:2000 Fig 22. Tile D (part): 1790 Whitehave town plan, reproduced at 1:2000 Fig 23. Tile D (part): 1st edition OS 1:500 Town Plan (1863), reproduced at 1:2000 LIST OF PHOTOGRAPHS Plate 1. Carlisle Spedding’s Plan 1752 (CROW: TNCB 24/4), north end. (DC) Plate 2. Carlisle Spedding’s Plan 1752 (CROW: TNCB 24/4), south end. (DC) Plate 3. Plan of Whitehaven, n.d.[1760s?], showing field system and waggonways in north end of project area (CROC: D/Lons/W7 Engineering Drawings, ‘Sundry Old Collieries’, p 25). (DC) Plate 4. South end of project area (Sandwith township): ancient landscape (SR) Plate 5. North end of project area (Preston Quarter township) from south: 19th century planned landscape, with 20th century urbanisation (SR) Plate 6. Wall 28925, showing ‘Cornish hedge’ construction (SR) Plate 7. Quarry 28936 (SR) Plate 8. Quarry 28936: graffito of carved head, possibly Stalin (SR) Plate 9. Anomalous wall 28942: boundary between waggonway 28945 and field system 28943 (SR) Plate 10. Barrowmouth mine (28950): lower area (features 28967-28983) from clifftop (SR) Plate 11. Barrowmouth mine: engine house 28952, door jamb (SR) Plate 12. Barrowmouth mine: incline 28955, showing dog-leg produced by landslipping (SR) Plate 13. Barrowmouth mine: building 28958 (SR) Plate 14. Barrowmouth mine: building 28960 (SR) Plate 15. Barrowmouth mine: building 28963 (SR) Plate 16. Barrowmouth mine: enclosure 28968 (SR) Plate 17. Barrowmouth mine: structure 28971 (SR) Plate 18. Barrowmouth mine: brake-drum mounting 28976 (SR) Plate 19. Tip 28984 from clifftop (SR) Plate 20. Structure 28986 (seaward terminal of aerial ropeway) (DC) Plate 21. Sandwith anhydrite mine (28999; foreground) and Marchon site (29900) during demolition (June 2006) (SR) Plate 22. County pit (29011) from W (SR) Plate 23. Croft Incline: view uphill from N, Kells Square on left (SR) Plate 24. Waggonway 29030 (low bank across centre) (DC) Plate 25. Ravenhill Colliery (29035): structure on cliff-edge (SR) Plate 26. Saltom Pit (29036) from E (SR) Plate 27. Saltom Pit (29036); design drawing for vertical winder engine in surviving engine house (The Beacon) Plate 28. Saltom Pit: rock-cut tank for saltworks 29038 (?) (SR) Plate 29. Saltom Pit: rock-cut postholes 29040 (SR) Plate 30. Haig Colliery (29044): headgear from N (SR) Plate 31. Haig Colliery (29044): east elevation (SR) Plate 32. Haig Colliery, c 1930s (The Beacon: 1986.152.27) Plate 33. Howgill Incline (29055), looking uphill (SR) Plate 34. Cast iron lamp-post 29059 (DC) Plate 35. Bowling Green area, overall from S, showing possible skittle alley to right of Jonathan Swift’s House, with Candlestick Chimney and Whitehaven Harbour in background (DC) Plate 36. From right: Jonathan Swift’s House (29060; Old Bowling Green (29061); building 29063; Bowling Green Battery (29062) (DC) Plate 37. Bowling Green Battery (29062), gun emplacement from E (DC) Plate 38. Bowling Green Battery (29062), boundary marker (DC) Plate 39. Wellington Pit (29067): view from north, late 19th century (The Beacon: 1994.461.77) Plate 40. Wellington Pit (29067): view from South Beach, 1911 (The Beacon: 1994.349.6) Plate 41. Wellington Pit: Candlestick Chimney (29069) (SR) Plate 42. ‘Harbour Incline’ – inclined viaduct 29071 (SR) Plate 43. Revetments 29075 and steps 29076 (SR) Plate 44. Revetments 29078 (to right) and 29080 (SR) Plate 45. Trackway 29084 (across centre) from S (DC) Plate 46. Duke Pit fanhouse (29086) from SW (SR) Plate 47. Duke Pit fanhouse (29086): evasee from SE (SR) Plate 48. Duke Pit fanhouse (29086) from NW (SR) Plate 49. Duke Pit fanhouse (29086): fan mounting from SW (SR) Photographers: DC = David Cranstone, Cranstone Consultants SR = Simon Roper, Ironbridge Archaeology ACKNOWLEDGEMENTS This study has been commissioned by the National Trust with funding support from West Lakes Renaissance, English Partnerships, The Land Restoration Trust, and the Cleaner Safer Greener Fund (Copeland Borough Council and Cumbria County Council). The project was undertaken by David Cranstone of Cranstone Consultants (historical research, field descriptions, gazetteer and report text) and Simon Roper of Ironbridge Archaeology (photography, GPS survey, and digital mapping). Our first acknowledgement must be to Jamie Lund (Regional Archaeologist) and Jeremy Barlow (Property Manager) of the National Trust (Northwest Region) for commissioning the project and for friendly assistance and support throughout, and to Paul Belford (Senior Archaeologist, Ironbridge Archaeology) for his helpful and efficient management of the Ironbridge elements of the project. We are also very grateful to Dr Peter King and to Simon Chapman for contributing their specialist expertise, respectively on the Medieval and legal records in The National Archives, and on the interpretation of the Saltom Pit enginehouse and the Duke Pit fanhouse. We are also grateful to Robert Baxter, Tom Robson, David Bowcock and other staff of Cumbria Record Office (Carlisle and Whitehaven), to Michelle Kelly (Museum Collections Officer, The Beacon), Pamela Telford and Toni Desovskie (Haig Colliery Mining Museum), Caroline Rhodes and Rosemary Preece (National Coal Mining Museum), David Clarke and other staff (Mining Records, Coal Authority), to Dr Richard Newman and Jo Mackintosh (Cumbria County Council historic environment section), Peter Brash (National Trust ecologist), Wayne Cocroft (English Heritage), Norman Gray, Andy Guy, Alan Routledge, Ben Russell (Science Museum), and John Todd, for their helpful assistance and discussion of various aspects of the project. In addition, Steve and Janet Pearson of Abbey Farmhouse, St Bees, provided warm, comfortable, and friendly accommodation during fieldwork. In a long-running project, it is all too likely that other help has been missed from this list; we can only apologise to anyone whose help has not been acknowledged, and assure them that it is no less appreciated. SUMMARY The Whitehaven Coast study area consists of a Y-shaped strip of land extending 4km along the coast from Whitehaven Harbour south and SW to the base of St Bees Head, with an extension to the SE from the centre of this area. A historical and field survey was carried out by Cranstone Consultants and Ironbridge Archaeology in 2006, on behalf of the National Trust. The bulk of the study area lies on the West Cumberland Coalfield, but the SW part includes an escarpment and coastal cliffs of St Bees Sandstone, underlain by seams of alabaster/gypsum and anhydrite. The survey has revealed little positive evidence of prehistoric or Roman activity, although there is potential for the discovery of Mesolithic and/or Roman military archaeology, especially on the sandstone uplands. The cultural history of the area from the Roman period until final incorporation into England in the 12th century was complex, and from the 12th century until the Dissolution the study area lay within the estates of St Bees Priory. Elements of the Medieval field pattern survive at the southern end, but not elsewhere, reflecting the different Post-Medieval histories of the two townships involved. The Priory developed very early coal-fuelled saltmaking within the study area, though the precise site has not been located. From the 1630s, much of the study area came progressively into the hands of the Lowther family, later Earls of Lonsdale. They developed a major coal-mining industry, and by the late 17th century the Howgill and Greenbank collieries were at the forefront of progress nationally. Early mining was predominantly to the east of the project area, but spread progressively westwards; in 1729 Saltom Pit was sunk on the coast, and from this period until the 20th century the area was dominated by coal- mining and associated waggonways. Major sites include: Saltom Pit, with 18th century shaft, horse-gin circle, and seawall, and 1820s vertical-winder enginehouse; Wellington Pit, sunk c 1840 with an impressive surviving chimney and gate-lodge; Duke Pit, with a surviving 1870 Guibal fanhouse incorporating remains of an 1840 experimental fan ventilator; and Haig Pit, sunk 1916 and retaining its engine house and horizontal winding engine and headgear. A waggonway system developed from the 1730s onwards and elements survive, including the impressive 1813 Howgill Incline running down to the remains of important coal staithes beside the Harbour. The south side of the Harbour is dominated by an impressive castellated landscape from the staithes to Wellington Pit, designed by Robert Smirke for the Earl of Lonsdale in the 1840s; the area also contains complex military, urban, salt-making, and railway-related features, including the 18th century Old Fort, and a later battery built into an old bowling green. The study area also includes considerable remains of the Barrowmouth alabaster/ gypsum mine, active from at least the 1730s to 1907. To the south, the crest of the sandstone scarp and cliffs contains an impressive group of quarries, of 18th-20th date with possible earlier origins. The final major industrial activity was the Marchon Chemical Works, active from 1943 to 2005 The report identifies priorities for conservation, research, and practical management of the archaeological resource and broader historic environment of the study area. List of Statutory Heritage Designations Scheduled Monuments Gaz No. SM No. Name 28950 35009 Barrowmouth Gypsum Mine 29036 27801 Saltom Pit 29044 27800 Haig Colliery 29072 34982 Old Fort Listed Buildings Gaz No. LB Name 29044 26332 Haig Colliery 29060 429157 Jonathan Swift’s House 29072 26166 Old Fort 1. INTRODUCTION Background to project The Whitehaven Coast survey was performed in accordance with a Project Design by Cranstone Consultants and Ironbridge Archaeology, in response to a Brief and Invitation to Tender issued by The National Trust in March 2006. Fieldwork and historical research were undertaken between May and October 2006, and the gazetteer and report were written over the winter of 2006-7. The study area consists of a block of land which is the focus of the Whitehaven Coast Project, part of the broader Whitehaven Regeneration Programme. Partners in the project include Copeland Borough Council, English Partnerships, The Land Restoration Trust, West Lakes Renaissance, the Haig Mining Museum and the National Trust. With funding support from these partners, the National Trust has been tasked with producing a Development Plan for the Whitehaven Coast which includes detailed research into the natural and cultural sigificance of the areas. This report forms a key part of that work. The study area consists of an inverted-Y shaped strip of land, running for c 2.5 km south along the coast from immediately behind (but not including) the south side of Whitehaven Harbour (NX 967 182) to the edge of the St Bees Head sandstone escarpment at NX 961 157, where it divides into lobes extending c 1.5 km respectively SW along the north coast of St Bees Head to NX 947 150 (c 1 km short of North Head), and SE (inland) to the Greenbank area at NX 975 153 (Fig 1). Much of the eastern boundary is formed by the edge of the modern housing of the Arrowthwaite, Kells, and Woodhouse areas of Whitehaven; remaining boundaries are formed by field boundaries of varying antiquity, though not normally marking any major historic ownership or tenurial boundaries - the study area is therefore arbitrary in terms of historic property and administrative units. The three ‘arms’ of the study area have very different topographic and archaeological characters, and are referred to respectively as the Northern, Southwestern, and Southeastern areas. In addition, a large area extending from the junction of the ‘arms’ into the Southeastern area was occupied by a large chemical factory (successively the Marchon, Rhodia, and Huntsman Works) from c 1950 until closure in 2005; this area is very different from the remainder in terms of its later-20th-century history and current state, and is referred to as the ‘Marchon Site’ (in conformity to normal local usage) where appropriate in this report. Geology and Topography The solid geology (British Geological Survey 1998, 2004) of the Northern and much of the Southeastern areas consists of Coal Measures sandstones, shales, and coal seams, dipping to the west (typically at c 1 in 10). Topographically this area consists of a ridge with a flat plateau-like top at c 70-90m OD (broadening and rising gradually to the south), bounded by coastal cliffs to the west and by the steep slope of the Pow Beck valley to the east; this valley forms a low-level through route from Whitehaven to St Bees, separating the high ground to its west from the remainder of west Cumberland. Much of the crest of the ridge is formed by outcrops of the Whitehaven Sandstone, a thick bed in the top of the Coal Measures; its often-purple colour results from oxidation when it lay close beneath the surface of the post- Carboniferous desert. The main workable coal seams (the Bannock, Prior or Main, and Six Quarters Bands) outcrop along the lower part of the Pow Beck slope (outside the National Trust area, though only just outside the end of the Southeast lobe); due to the dip they are below sea level at the coast, and extend for several miles under the sea. Thinner seams are also present, including several above the Bannock Band which do outcrop locally in the coastal cliff and have apparently been worked locally at outcrop. However the Coal Measures deposits are broken up by frequent small faults, so the detailed geological structure is considerably more complex than this summary may suggest; in particular, a pair of east-west faults have lifted a block of strata extending east from Saltom Pit, bringing the main coal seams here closer to sea level, exposing the thin upper seams in the cliffs behind Saltom Pit, and interrupting the outcrop of the Whitehaven Sandstone across the plateau; the re-entrant of the cliffs behind Saltom Pit reflects this geological feature. The Pow Beck valley follows a more substantial fault, and (except under Whitehaven town) cuts through the Coal Measures to below the productive seams; as a result the landward coal seams form effectively a self-contained coalfield, only connected to the rest of the Cumberland Coalfield beneath the sea and Whitehaven town. The Coal Measures strata are overlain by boulder clay and other glacial deposits over much of the area, with the exception of the coastal cliffs, the north end end of the ridge from Saltom to Whitehaven Harbour, and the steeper parts of the Pow Beck slope; consequently exposed surface outcrops of the coal seams will have been very limited. To the south, the Coal Measures are unconformably overlain by Permian beds, dipping to the south and consisting successively of the Brockram, the Magnesium Limestone, the St Bees Shales, and the St Bees Sandstone. The first three of these are soft and/or thin deposits, and their outcrop is marked by a slight hollow running across the south end of the ridge, from the coast at Barrowmouth to the Marchon site, and thence southeast outside the National Trust area; the northeast boundary of the Permian runs from the north end of the Marchon site to the southeast, just to the east of High Road then just to the west of Wilson Pit Lane. The Brockram is an unusual formation, of limestone and volcanic rock fragments cemented with haematite and baryte; its outcrop on the coast at NX 958 159 is of geological interest. The Magnesian Limestone appears to be very thin, and contains workable beds of gypsum/alabaster and anhydrite as well as limestones and shales (gypsum and alabaster are hydrated calcium sulphate, referred to as gypsum when it is used as a chemical raw material (mainly for making plaster) and as alabaster when it is extracted in blocks for use as a monumental stone; anhydrite is anhydrous calcium sulphate, used entirely as a chemical raw material). The St Bees Shales are a much thicker formation, though poorly exposed due to the their soft nature; they also contain some anhydrite. All three deposits are largely masked by boulder clay, except on their coastal outcrop at Barrowmouth; here however they are involved in major active landslipping, and are also partly overlain by massive (peri-glacial?) scree deposits from the St Bees Sandstone cliffs which tower over them to the south. The presence of limestone (or, strictly, dolomite) in the lower Permian is barely apparent in the present landscape, but the presence of quarries and limekilns on the 19th century OS editions indicates that it was exploited to some extent, and it may have had an influence on earlier vegetation and agriculture. In contrast, the fourth Permian bed, the St Bees Sandstone, is a thick and hard bed of red sandstone, and forms a north-facing escarpment, running ENE from North Head to above Barrowmouth (where it reaches a maximum height of over 130 m OD), and then SE along the west edge of the Marchon site to Townhead and beyond. The western part of the scarp forms part of the St Bees Head coastal cliffs (with the landslipped undercliff at Barrowmouth already referred to), while the SE part forms a steep slope some 30m high, above the plateau of the Northern Area; the dip slope drops gradually to St Bees and Rottington, dissected by several small valleys. Some beds of the St Bees Sandstone form high-quality building stone, which has been extensively quarried along the scarp crest. The southern boundary of the National Trust area runs along or just behind the scarp crest, incorporating most of the historic quarrying and the northernmost fields of the very different dip-slope landscape of Sandwith, Rottington, and St Bees, most of which lies outside the study area – the project has indicated that this dramatic change in landscape owes more to historical ownership and administrative factors than to geology, though this is also a factor. As already noted, over most of the National Trust area the solid geology is masked by glacial till (‘boulder clay’), which has presumably dominated soil formation. The topography within the study area is mostly flat or gently-sloping, with the exception of the St Bees Sandstone escarpment and the coastal cliffs and slopes. The latter are vertical and dramatic from Barrowmouth westwards, and landslipped and actively eroding at Barrowmouth and from Saltom Pit northwards to Whitehaven South Beach. Between Barrowmouth and Saltom, the base of the main coastal cliff is separated by a narrow terrace, indicated as a raised beach on the 1:50,000 geological map (though not on the more detailed but slightly earlier 1:10,000 sheet). Given the very active coatal erosion, this coastal shelf may have been both wider and more extensive in pre- modern periods, forming a coastal environment very different from what now survives. Natural Units One point arising from the preceding section should be stressed and amplified at this stage. The National Trust area forms a wholly-new land unit, and is arbitrary in terms of the natural factors of geology and topography, and the human factors of historical land units (both administrative and tenurial) and the historical and archaeological landscape (at least until the later 20th century, when the eastern boundary of much of the NT area formed as the western boundary of the Whitehaven suburbs). However, the Trust area forms part of two very ‘real’ land units, in all of these terms. The south end (the Southwest area and the western hinterland of the Marchon site) forms part of the natural landscape of the St Bees Sandstone hills, and the human landscape of Sandwith township, Rottington, and St Bees, centring on St Bees village. The remainder of the area forms the west half of the natural landscape of the Coal Measures Arrowthwaite/Kells ridge, the historic township of Preston Quarter, and the 17th-19th century ‘Howgill Colliery’, all bounded to the east by the Pow Beck, and centred on Whitehaven. The latter is of particular importance – the bulk of the study area forms the longitudinal west half of a very distinct, unified, and clearly-bounded natural and historical unit (referred to below as the ‘Howgill Colliery’ in specifically industrial contexts, and the ‘Howgill Ridge’ in more general contexts), and its history and archaeology cannot be understood without considerable reference to its broader context; its management also must take major consideration of this broader conservation unit. 2. METHODOLOGY Introduction The research requirements of the project divided into two major elements: historical research and fieldwork. Initial reconnaissance indicated that, despite generally good field conditions for observation, many historically-known sites did not survive as obvious field monuments, and a decision was therefore taken to undertake the historical research first, with a large element of the fieldwork being to inspect historically-identified site locations, in order to identify any subtle surviving field evidence, and to record the current state and assess the prospects for below-ground preservation even where no visible field evidence could be identified. The wisdom of this approach was confirmed by the one exception, Barrowmouth, where fieldwork was undertaken early in the project due to the known presence of dense bracken cover in summer; in practice the vegetation proved to include extensive bramble and scrub such that initial inspection in late spring did not yield adequately-interpretable results until revisit with historic OS mapping. Historical Research Prior knowledge and research for the project design indicated that the secondary literature relating to the site and its environs was very extensive, and that the primary documentation was also very extensive, and was also widely scattered between a range of repositories, and in the case of the single main relevant collection (the Lonsdale Papers in Cumbria Record Office, Carlisle (CROC)), both massive in extent and drastically under-catalogued, to the extent that only limited consultation would be realistic within project budgets and timetables (although it was clear that much of the effectively-inaccessible material had been used (albeit often very selectively) by the authors of the secondary literature). The decision was therefore taken to start with a comprehensive read of the secondary literature, including compilation of as complete a bibliography as reasonably achievable; the Bibliography for this report is therefore thought to be a nearly- comprehensive list of the published material relating to the study area and its environs, and includes numerous sources not cited specifically in the text (either because their information proved to be irrelevant in detail, or repetitive from one publication to another, or because the primary source material was located and consulted, and therefore cited in preference to secondary sources). At the same time, potential library and Record Office collections were contacted, in order to identify relevant holdings, and assess their importance and accessibility; Winstanley and David 2006 formed a useful guide to collections within Cumbria, though it does not include out-of-county collections. This desk-based research was followed by research visits to the main libraries and document repositories identified as containing relevant material. These are: • The Beacon, Whitehaven. Extensive library of published books, research files (including copies of primary documents held elsewhere), maps and plans (mainly copies, and only partially catalogued at the time of visit), and historic photographs (including both catalogued boxes and uncatalogued albums), together with the important 1738 Matthias Read painting. As the most comprehensive, accessible, and easily-used collection, this was used as the prime research venue • Cumbria Record Office, Whitehaven (CROW). Extensive local studies library of published works and research files (both partially duplicating The Beacon, but with substantial additional material), and major primary document collections relating to the Whitehaven area; the colliery abandonment plans (TNCB) and St Bees School and manorial records (YDS 60) proved to be particularly valuable • Cumbria Record Office, Carlisle (CROC). The main repository collection for the county. Holdings include the Preston Quarter and Sandwith tithe maps (not held at Whitehaven), and the massive and crucial Lonsdale Papers, the records of the Lowther family (D/Lons/W for the Whitehaven estate). Unfortunately much of the D/Lons/W collection (including many papers used by previous researchers, and/or listed in the former (now withdrawn) Record Office catalogue of the collection) is effectively inaccessible due to storage and cataloguing problems; use of this collection was therefore limited, and it is clear that an enormous amount of relevant, and doubtless in many cases highly important, material remains unconsulted. • Haig Colliery Mining Museum (HCMM). Extensive collections of artefacts (part on display), books and mining-related journals (currently not sorted or catalogued, so not consulted) maps (mainly copies of former-NCB colliery plans, some now in CROW or CAAM; uncatalogued), and photographs (with an emphasis on the personal and social, rather than on sites; currently partially catalogued and not rapidly consultable). The primary document collection is limited, but includes the important Bateman letter books (first two volumes now transcribed and printed). • Helena Thompson Museum, Workington. Winstanley and David (2006, 77) report that the museum holds the Earl of Lonsdale’s replies to Bateman’s letters in the HCMM letter-books, but they cannot be traced in the museum catalogue (Nicci Tofts, pers. com) and are probably within the CROC Lonsdale Papers; not visited. • Coal Authority Archives, Mansfield (CAAM). Extensive collection of mine abandonment plans, complementing the CROW: TNCB collection; the origin of the separation is not clear. Most plans have been digitised, but in some instances a combination of large size and poor visual quality caused problems in obtaining print-outs. • National Coal Mining Museum, Wakefield (NCMM). A small collection of relevant artefacts and historic photographs (notably the pre-closure Cornwell photos of Haig Colliery). • Northumberland Record Office (Mining Institute collection) (NRO). Contains extensive relevant material, mainly in the North of England Institute of Mining and Mechanical Engineers collection (NRO 3410), notably colliery viewers’ reports and correspondence. Unfortunately NRO was closed for relocation at the time of research, so consultation has not been possible; the catalogue is however available online via www.a2a.org.uk . • The National Archives (TNA). Research was undertaken for the project by Dr Peter King, a historian and trained solicitor familiar with legal records, Medieval Latin, and the TNA archiving and cataloguing system; it concentrated on records relating to Medieval land-holding and agricultural organisation (largely in records relating to St Bees Priory). It was intended that Dr King would also consult records of legal actions by the Lowther family/Earls of Lonsdale relating to the project study area. However preliminary investigation indicated that TNA held records of a very large number of cases relating to the Lowthers/Lonsdales, with no means of readily determining which might be relevant to the project area; this element of research was therefore abandoned. The relationships between these collections are sometimes complex, and ‘secondary’ collections such as The Beacon and Haig include copies and extracts, of varying quality, from original material known or presumed to be in CROC or CROW, including currently-inaccessible parts of the Lonsdale MSS. In particular, the machinery drawings in the CROC Portfolios of Engineering Drawings (currently catalogued in D/Lons/W7, though formerly in D/Lons/W10), themselves a mid-late 19th century bound collection of originals of various earlier dates, are the primary source for most (probably all) of the drawings in CROW TNCB 28/13 ( a set of re- drawings for R W Moore’s 1898 paper), and one of these collections appears to be the source for the various copy drawings of the same engines in the Beacon, Haig, and Cumbria HER collections – these are much more accessible than the originals, but do appear to have been slightly simplified in some cases. Copies of relevant maps and other documents were obtained wherever possible, by photocopying or hand-held digital photography. Unfortunately Health and Safety requirements in CROW and CROC required the use of a rather low and unstable kick- stool for photography of documents laid out on the desk; the height from document to camera proved to be just inside the minimum focal length of the camera, and the resulting poor quality of many images is regretted. It should also be noted that many of the colliery-related plans, both in CROW: TNCB and in CAAM, are faded, stained, and dirty, and consequently difficult to use or copy; their state does however form an interesting sidelight on conditions in the former Whitehaven colliery office, where they were presumably stored. For these reasons, only very limited historical material is presented as illustrations to this report, the bulk being presented as photocopies or digital images in the project archive. A full set of OS 1st-3rd edition 1:2500 was obtained, as A3 photocopies from CROW, and subsequently digitised. This forms the basis of much of the project mapping. Subsequent OS editions are still in copyright, and therefore could not be copied. However an annotated copy of the 4th edition 6” map was consulted in The Beacon; although published in 1957, the relevant area was last revised in 1930-1945 (with ‘major changes only’ in 1951), and the map is therefore referred to as ‘OS (1930s revision)’ in this report. As noted in the previous chapter, the bulk of the study area forms the west half of a very distinct historic land unit, defined industrially as the ‘Howgill Colliery’ and referred to more generally in this report as the ‘Howgill Ridge’. Except for the most site-specific items, most of the primary and secondary literature, and therefore the Bibliography in this report, relate to this broader unit rather than solely to the National Trust area. Fieldwork Fieldwork was undertaken (by DC and SR) primarily in two sessions, in June and October 2006, though observations were also made by DC at other times, during reconnaissance and during visits primarily for historical research. It was initially intended to use Global Positioning System (GPS) extensively for site mapping and planning, using WAAS-enabled Garmin GPSMAP 60 hand-held equipment. However, pilot usage of this equipment in the Barrowmouth area (sites 28950 et seq) gave very poor results in terms of accuracy, with displayed error margins of up to +-17m. This unexpected result may result from limited sky visibility in this undercliff area (coupled perhaps with unfortunate timing when satellites were concentrated in the obscured southern sector of the sky), and might have proved atypical for flatter parts of the study area. However, increasing familiarity with the area indicated that in practice most archaeological features were either marked on current OS map base, or could easily be related by visual means or pacing to mapped features, or were not visible on the ground and were only plottable by overlay from historic OS and other mapping. The use of GPS was therefore discontinued. Fieldwork was therefore undertaken using print-outs of digital overlays of the 1st-3rd OS map editions, together with the 1790 Whitehaven plan and the waggonway routes and colliery shafts from the 1752 Spedding plan (see below), onto current 1:10,000 digital OS map base (supplied by The National Trust). The locations of all sites known from this historic evidence were visited (except within the Marchon site – see below), the field evidence (if any) recorded, and the prospects for below-ground survival of non-upstanding sites assessed by careful visual examination. The whole project area was also carefully inspected for field evidence of further sites not known from the historical evidence. Field boundaries and systems were included in this recording, and the current agricultural status noted. In practice, virtually the whole area could be closely inspected, either by direct on-site walking or by fairly close and unimpeded observation from roads, paths, and pasture fields. Pasture and stubble fields were entered, by permission of the tenant farmers. Field visibility was generally good. There were four partial exceptions to this level of coverage:- • The Marchon site. This was excluded from fieldwork by the Brief, for Health and Safety reasons; the major industrial complex was in process of demolition during the project. By the final stint of fieldwork, most of this site (except for its NE quadrant) had been cleared, and observation from outside the perimeter fence indicated no upstanding pre-Marchon features, in an extensive relatively-flat area variably covered by surviving concrete foundations and surfaces. • The Barrowmouth area (including the areas of quarrying on the clifftop above). Fieldwork here was impeded by difficult field conditions, of dense undergrowth (including brambles and scrub), broken and uneven ground (due to very steep natural topography coupled with active landslipping), and very limited paths – off-path exploration was also inhibited by a strong local reputation for adders (not positively confirmed by personal observation, though the habitat does appear highly suitable). Some minor sites known from historic mapping could not be visited, and it is inevitable that some field detail will not have been observed. • The foreshore and coastal terrace. While the foreshore from Saltom Pit northwards to South Beach was fieldwalked, and the foreshore below Barrowmouth visited and closely observed, the foreshore between Saltom and Barrowmouth was not visited due to a total lack of known sites coupled with lack of access to the coastal terrace and safety considerations on the foreshore (which showed a disturbing combination of rocks and large boulders with exceptionally-slippery surfaces). The area was however closely scanned from the cliff-top, without positive results. • The area of ‘Jonathan Swift’s House’ and the Old Bowling Green. This area was excluded from the Brief for fieldwork since it was in private occupation. In practice, the owner invited the surveyor (DC) in to look at the remains of the 19th century fort, and outline recording was therefore possible, but was less detailed than would have been the case with unrestricted access. Written recording was undertaken by DC, using paper pro formas designed for compatibility with the NTSMR fields, and sites were numbered directly into a block of NTSMR numbers (28925-29100), with no separate permanent project numbers. This sequence includes sites within the Old Bowling Green area (29060-29063), and also Duke Pit (29085-29086) (outside the National Trust boundary, but a visually- important feature on a main access route, with considerable display potential – this site was therefore included in the project coverage at the Trust’s request). These numbers are arranged in geographical order, from SW to north, and are used for reference in the following sections where appropriate, with full details in the Site Gazetteer (Appendix 4). A few further sites, immediately outside the project area but important to its interpretation or management, were also inspected and recorded; these have been allocated 2-digit project numbers (01-08), and are included at the end of the Site Gazetteer. These sites have generally been recorded in less detail. It should be noted that the context of the NT area also includes Whitehaven Harbour and town; no attampt has been made to record these, due to their complexity, although their understanding is clearly important to the understanding and management of the project area. Formal record photography was undertaken by Simon Roper (3-digit frame numbers), though some additional hand-held photography was undertaken by DC (four-digit image numbers), for features observed during fieldwork after the main photographic recording. 3. CHRONOLOGICAL SURVEY The results of the project are presented as a narrative survey, integrating the historical and archaeological evidence. This is structured chronologically for the earlier periods (for which evidence is limited), but is divided by site and topic for the Post-Medieval period (for which the evidence is extremely rich and varied). Individual sites are referred to where appropriate by their 5-digit NTSMR number (or 2-digit project number for sites outside the National Trust boundary). The history and archaeology of West Cumberland, in its broader regional context, are surveyed by Higham 1986, McCord and Thompson 1998, Winchester 2000, and the various papers in Brooks et al 2002 and Brennand (ed) 2006; these provide the broader context within which the archaeology of the NT area must be assessed. McCarthy 2002 provides a more local overview, though with a focus on Carlisle and the Solway Plain, and on the Roman period. However it should be noted that the West Cumberland coastal area forms a very distinct geological and topographic unit, isolated from the rest of the region (except the Solway Plain) by the Lake District fells, and potentially more closely-connected by sea transport (and by view-shed, with all its psychological implications) to Galloway and the Isle of Man, and little-studied archaeologically in its own right. The northwest-facing ‘coalfield coast’ from Maryport to Barrowmouth also differs substantially (both in its natural background and in its archaeology and history) from the ‘southwest coast’ from St Bees Head to Haverigg. The ‘North West England’ regional context should therefore be applied with caution. Prehistoric There is no positive evidence for human occupation anywhere in the region during the bulk of the Palaeolithic period, and the whole area was certainly ice-covered and uninhabitable during the peak of the Devensian (final) glaciation. During the final waning stages of the glaciation, there is now evidence for Cresswellian (Late Upper Palaeolithic) occupation in some of the Furness and Cartmel caves, with hints that the Solway Plain may also have been inhabited (Brennand (ed) 2006, 24-25; Young 2002, 20-22). There is no evidence for or against human activity within the study area during this period. During the succeeding Mesolithic period (c 8000-4000 BC), the climate warmed rapidly to broady-modern temperatures, and the vegetation recovered more slowly to become dominated by deciduous woodland. Sea-level recovered more slowly, since the ‘eustatic’ rise in global sea-levels was partially offset regionally by an ‘isostatic’ rise of the land (which still continues), compensating for the removed weight of the Devensian ice cover. Data for Northwest England in the national assessment of coastal archaeology (Fulford et al 1997) are dominated by Liverpool Bay, where isostatic recovery was less pronounced than in Cumbria; the most recent discussion (Brennand (ed) 2006, 23-26, 30-31) suggests that sea-levels rose from c -20m OD in the earliest Mesolithic to c -2m OD around 5000 BC, and to high-tide levels of up to +8m OD at the Mesolithic-Neolithic transition (since when sea-levels have oscillated close to modern levels with relatively slight transgressions and regressions whose effects may have been cancelled-out locally, by continuing isostatic recovery and by continuing erosion of the more exposed and/or soft coastal cliffs respectively). In the earlier Mesolithic, therefore, the NT area will have formed the edge of an abrupt upland massif overlooking a coastal plain. By the end of the period, however, this plain was reduced to, at most, a terrace, and may have been replaced by active sea- cliffs. It is not clear whether the apparent raised beach between Saltom and Barrowmouth is part of the late Mesolithic sea-level maximum; if so, late Mesolithic coastal activity may conceivably be preserved beneath beneath later screes and sediments formed by erosion from the fossil cliff behind, though if not, any Mesolithic coastal activity will have occurred on destroyed or submerged coastlines to seawards of the modern coast. The known Mesolithic sites within the Whitehaven-St Bees area consist of a major cluster of Later Mesolithic flint scatters, recovered by fieldwalking in the St Bees area (Cherry and Cherry 1983, 1996, 2002); the nearest of these lies just SW of the SW end of the National Trust area (Cherry and Cherry 1983, 2; Cumbria HER 3678). No Mesolithic finds are known from within the study area, but it is not known whether any of the fields have been walked (by Cherry or any other fieldworkers; Cherry’s study area certainly did not extend north of the St Bees Head area, though it may have included the southwestern arm of the Trust area), so this absence may be apparent rather than real. Many of the sites are close behind the cliffs, and Cherry and Cherry (2002, 3) suggest that this may reflect use of the seabird colonies as a food source; if so, further sites may well exist in the fields along the southern boundary of the NT area, as far as Birkhams Quarry. The Cherrys’ distibution does not extend onto the Coal Measures coast of the Howgill Ridge, but it is not clear whether any fields here were fieldwalked. Given the appearance of the important St Bees group of sites nearby, every effort should be made to develop a programme of fieldwalking after ploughing within the project area. As already noted, the coastal terrace between Saltom and Barrowmouth forms another potential locus for Mesolithic activity. In addition, the Pow Beck valley between Whitehaven and St Bees forms a potential low-level animal migration route along the coastal lowlands (especially after rising sea-levels had reached the base of the St Bees Head cliffs, cutting off any route round the end of this upland), and hunting sites along the sides of this valley would appear possible; potential locations within the NT area are the crest of the Howgill Ridge above Harbour View, and the E end of the Southeastern area around Greenbank. The St Bees cluster of flint-scatter sites appear to have continued from the Mesolithic into the Neolithic (Cherry and Cherry 2002, 6-7), and the presence of important Neolithic sites at Ehenside Tarn, Williamson’s Moss, and Monk’s Moor indicates substantial activity on the coastal strip south from St Bees Head (Brennand 2006, 33). However there appears to be little evidence as to whether this activity was matched in the coalfield area. The only evidence from the study area consists of two records (Cumbria HER 1177, 1190) of polished stone axe finds from Barrowmouth, one recorded as from the gypsum mine (28950); these are not detailed, and may both refer to the same find. They do at least indicate some utilisation of the study area, conceivably timber-gathering or coppicing in Barrowmouth Wood (as this area was still labelled on the 1st edition OS). This lack of local evidence continues throughout later prehistory, though the important Ewanrigg site near Maryport does at least indicate Late Neolithic-Early Bronze Age activity within the north end of the coalfield area (Brennand (ed) 2006, 42, 48-50). Given the virtually aceramic nature of the later Bronze Age and the Iron Age within Cumbria, and the poor potential for cropmark formation, sites of this period are very hard to locate. An undated oval cropmark enclosure (HER 4701) at Hannah Moor, on the Red Sandstone hills just to the SW of the project area, may possibly be of Bronze Age date, and an apparent kite-shaped enclosure (29088) within the SE corner of the NT area, with other possible cropmark features associated, can plausibly be interpreted as an Iron Age (or Roman-British) settlement – given the rarity of later prehistoric sites in the area, this would be of some importance if confirmed. The place-name ‘Castle Rigg’, documented from 1694 to 1808 as an enclosure ‘close to Saltom Engine’ (CROC: D/Lons/W6/1-4) may refer to an Iron Age embanked settlement or even hillfort (conceivably a promontory fort on the Ravenhill Pit site, or on the cliff edge NE of Saltom Pit), but is not strong evidence in isolation. Roman The Roman army conquered Cumbria in the 70s AD, though the degree of early Roman penetration into the western coastal strip is unclear. Hadrian’s Wall was constructed in the 120s, and was continued along the Solway coast by a series of forts and milecastles, at least as far as Risehow south of Maryport, with further coastal forts at Burrow Walls (Workington) and Moresby (Wilson 2004, Wilson and Caruana (eds) 2004, 67). A further fort, perhaps entirely of 3rd-4th century rather than Hadrianic date, has recently been suggested on the basis of coin finds recorded from Whitehaven Castle (Caruana and Shotter 2005). However, while this site may well have close to the Roman-period shoreline of Whitehaven Harbour, it is not obviously defensible (the name Whitehaven Castle dates only from its rebuilding by Sir James Lowther in the 1760s), and it seems at least as likely that these finds represent a discarded Lowther antiquarian collection. The apparent ‘fizzling out’ of the Solway frontier at Risehow has puzzled many workers on Hadrian’s Wall, and it has often been suggested that St Bees Head would have formed a logical terminus for the system (though Collingwood’s suggested evidence for a fortlet or signal station at North Head has not met with later acceptance (HER 1187)). This raises the possibility of a fort or signal station within the NT area. The crest of the Howgill Ridge above Harbour View offers a defensible location commanding Whitehaven Harbour and offering extensive views across the Solway, though the absence of any recorded Roman finds from this much-disturbed area perhaps argues against this. The ‘Castle Rigg’ field-name near Saltom could also possibly refer to a Roman fort. Observation during project fieldwork also suggests that the St Bees Sandstone scarp crest from North Head to the hill north of Sandwith at NX 961 157 would also offer suitable locations, with very wide views across the Solway and towards the Isle of Man, and in places also sight-lines south along the coast (though not specifically to the Ravenglass fort site, so far as could be ascertained). No field evidence was observed during the project, but since all the relevant fields are either arable or formerly-ploughed long-ley pasture this is unsurprising; neither fieldwalking nor targetted aerial photography are known to have been attempted in this area. A single Hadrianic coin-find from School Croft, Sandwith (HER 13692) does offer slight support for Roman activity in this area. The possibility of a Roman military site on the high ground along the southern boundary of the NT area does therefore appear to merit investigation, particularly by aerial photography under drought conditions. A faint but very neat rectilinear cropmark (29087) within the NT area just west of the Croft Incline, though not fort-like, could conceivably be the robbing-trenches of a major Roman building. However it is suspiciously-parallel to the adjacent field boundary, of early 19th century origin, and it is thought more likely to be a 19th or 20th century feature, or an artefact of agricultural practice at the time of photography. If it should be Roman, it would be great interest – fieldwalking of this area is desirable after any future ploughing, in order to check for any artifact scatters. Pollen evidence from the western Lake District shows considerable forest clearance starting in the later Iron Age, and intensifying in the later Roman period, perhaps reflecting a slightly warmer and drier climatic fluctuation (Wells 2003, 73, 78). Within the study area, the possibly-Iron Age cropmark sites referred to above could equally be of Romano-British or Roman Iron Age date (the latter term is perhaps more appropriate in view of the very limited Romanisation of rural Cumbria); Roman- period agricultural use of the area seems likely, whether or not there was ‘native’ settlement and/or any military activity within the Trust area. Later 1st millenium (c 400 to 1150 AD) The cultural and political affiliations of Cumbria, and particularly of Copeland, from the withdrawal of Roman imperial control in c 400 AD to the final establishment of English rule in the 1150s, were complex; the archaeological evidence is hard to identify or date, and difficult to relate to the cultural and political picture developed from the (very limited) historical evidence (Higham 1986; Loveluck 2002; McCarthy 2002; McCord and Thompson 1998, Newman R M 2006). In the 5th century, ‘Romanised’ occupation clearly continued in at least some of the forts and vici, with increasing evidence for Christianity and continued use of Latin for inscriptions, though the bulk of the population probably spoke an early form of Cumbric (a Brittonic language closely allied to Welsh). The story of St Patrick (abducted, almost certainly from Cumbria, by Irish pirates) indicates substantial contact with Ireland, if not always friendly. Interesting, both of the only two diagnostic small finds of this period from Cumbria are of Irish stylistic affinities – one comes from Moresby, just up the coast from the study area, and the other from Mealsgate, on the boundary between the West Cumberland coastal strip and the Solway Plain (Loveluck 2002, 144). By the later 6th century, the area almost certainly formed the core of the kingdom of Rheged, a Brittonic kingdom whose rulers came from Strathclyde, were or became Christian, and whose legends and poetry survive in part in the later Welsh manuscript of The Gododdin. However during the following century Rheged was absorbed into the growing and broadly Anglo-Saxon kingdom of Northumbria, though this may represent inter-marriage and acculturation by the elite, rather than any broader population movement. There is pollen evidence for a marked reduction in cultivation and regrowth of woodlands for a period of several centuries perhaps starting in the 6th century, though the precise dating is unclear and may have varied widely within the region (Wells 2003, 81-2). At its 7th-century peak, Northumbria controlled the whole of Cumbria, the Isle of Man, and at least as far as Whithorn in Galloway (Higham 1986, 250-263). After AD 685, Northumbrian control shrank, but still included Cumbria. Norse raids began from the 790s on the east coast; there is no documentation from Cumbria at this time. Much of Northumbria was overwhelmed by a Danish army in the 860s and 870s and its estates passed into the hands of Danish settlers, but it is unclear how far this affected Cumbria; Higham (1986, 316-322) argues that this marked a re-establishment of British-speaking communities, developing into the state of Strathclyde/Cumbria, and the diocese of Glasgow. He also argues that ‘Cumbria’ was centred on the northeast of the modern county, and may have actively acquiesced in Scandinavian setlement of peripheral areas, including the western coastal strip, in the early 10th century; this settlement formed a Hiberno-Norse ‘reflux’ from Dublin and other Norse settlements in Ireland, and its language (and no doubt culture) contained a substantial Gaelic element – the placenames show particularly close similarities with the Isle of Man. By the 11th century, Cumbria/Strathclyde had been absorbed, at least politically, into the kingdom of Scotland. For most of the 11th century, Cumbria remained part of Scotland, with some interludes of control by the Earls of Northumbria (at least notionally English); however it is not clear that ‘Cumbria’ included Copeland, and the Scottish border may have followed the Derwent (McCord and Thompson 1998, 18); the barony of Copeland appears to have been under English suzerainty by 1086 (Sharpe 2005, 38-9), although it does not appear in the Domesday Book. Cumbria was taken into English control in 1092, when William Rufus founded Carlisle castle as a border fortress, but Cumberland was not constituted as a county until (probably) 1133, and did not include Copeland (Sharpe 2005). The diocese of Carlisle was also split off from that of Glasgow at this time; it excluded Copeland, which formed part of the diocese of York. Cumberland, and seemingly also Copeland, returned to Scottish control in 1135; Carlisle was a major royal town of David I. Both were returned to English control in 1157, and with hindsight this was permanent; Copeland was incorporated into Cumberland in the 1170s. The status of Copeland before this date, as a barony under English suzerainty (except during the reign of David I) but outside any formal administrative system, is unclear; it seems to bear resemblances to the status of the quasi-independent Welsh Marcher lordships (eg Whittington - Brown et al 2004), though these remained outside the English legal system from the Norman period until the 16th century. The political affiliations of what is now West Cumberland between the end of the Roman occupation and the final incorporation into England in the mid 12th century were therefore complex and kalaidoscopic. Linguistically, Anglian (ie Northumbrian), Scandinavian, British, a few Gaelic, and 12th century English names are all present - the first and last of these can be hard to distinguish. Cultural developments within the Whitehaven area presumably broadly followed the political and linguistic pictures, though geography suggests that the Irish Sea elements (British, Norse, and Gaelic) may have been particularly strong; Rachel Newman (2006, 103) argues that the Cumbrian coast had particular links to Chester. In considering the archaeology of the area, therefore, it is important not to uncritically assume that ‘English’ models will have applied. This applies also to at least the earlier part of the Medieval period – cross-Solway links appear to have remained strong until the onset of the Border Wars in the 1290s, and the inhabitants of the Whitehaven-St Bees area may well have had closer contact with Galloway and the Isle of Man than with non-Cumbrian England. Within the study area and its immediate context, placenames include Scandinavian (Arrowthwaite, Whitehaven, Sandwith), Anglian/English (Rottington, Preston, Saltom), and British (Higham 1986, 319 – it is not clear which specific names the relevant dots refer to). By the end of the period (and perhaps very much earlier), the study area formed part of the massive parish of St Bees (Todd 1980, 2003). The ecclesiastical origin of this are unknown. The earlier historical literature claims that St Bega fled from Ireland in the 7th century and founded a nunnery, but Todd has shown that much of this legend was of 17th century origin; St Bega is clearly referred to in the earliest references as living alone (ie a hermit), and the alleged context, of fleeing to avoid forcible marriage to a Norse prince, would fit with the late 9th century situation (if it had any historical reality). Todd (2003) argues that St Bees may have been a ‘minster’ church, most probably of 9th-10th century origin and non-monastic, with an original parish covering much of Copeland; this latter seems likely, though not necessarily within the specifically Anglo-Saxon framework assumed by Todd (the large multi- township parishes described by Winchester (1987, 2000) for the northern uplands, or the ‘multiple estates’ argued by Glanville Jones and many others within a Welsh context, would seem equally applicable). (The placename ‘Eaglesfield’, close to the church, may possibly be an ‘eccles’ name, indicating a ‘mother’ church with wide jurisdiction and early origins (Gelling 1978, 82-3, 96-8), but is more likely to be of recent origin (J Todd, pers.com)). The field evidence of high-quality Hiberno-Norse sculpture associated with the church confirms an important status by the 10th-11th centuries. There is therefore a strong case for arguing that the study area formed part of a major early estate and parish centred on St Bees. The placenames Saltom (‘salt- ton’?) and Preston (priest’s-ton) in or adjacent to the study area may indicate that the area already performed specific roles within this estate, and their Anglo-Saxon nature suggests that they, and therefore the ecclesiastical estate, dated from (or before) the period of Northumbrian domination, rather than from the period of Norse place- names. Medieval St Bees Priory was founded between 1120 and 1135, probably in the early 1130s (Sharpe 2005, 64-5), during the first period of clear English control, by William le Meschin, lord of Copeland (and based at Egremont), as a Benedictine cell of St Mary’s Abbey, York; the signatories to its inaugural charters included ‘Coremac’ and ‘Gille Becoc [servant of St Bega]’, both Gaelic names and possibly the priests of the pre-priory church (Todd 2003, 101). Its history has been discussed by Wilson (1905, 1915), who also published its Register or cartulary (in Latin; Wilson (ed) 1915). There is no modern overall study, though John Todd (1980, 2003) has done valuable work on specific aspects; in particular, the landscape and land management of the monastic estate, which included the whole of the National Trust area, has never been investigated. The historical resources probably exist to do this, in the Cartulary, The National Archives, the Lonsdale MSS (CROC C/Lons), the St Bees School MSS (CROW YDS 60), and other papers in CROC and CROW, and this would undoubtedly contribute considerably to our understanding of the study area in its local context. For the present project, however, research on this aspect has been limited to the secondary literature, and to translation and study of the most relevant entries in the Register by Dr Peter King (Appendix 1). St Bees Priory had extensive land-holdings on the Isle of Man (Wilson 1915, xiv- xviii), and the c 1230 grant (Appendix 1) was from Maurice of Man. The Gaelic signatories to the inaugural charters may therefore have been of Manx rather than Irish (or Scottish) origin; the possibility of Manx influence on the culture and archaeology of the study area is therefore real. The Priory also had saltworks, and probably other interests, at Colvend and near Southernness in Galloway and Redkirk Point in Dumfriesshire (Cranstone 2006, 48, 50, 59), as well as near Millom and Burgh-by-Sands in Cumberland. Medieval and later land-holding and administration in west Cumberland was complex; the best description for the present area remains Angus Winchester’s Landscape and Society in Medieval Cumbria (1987, esp. 27-33; Winchester 1978 contains additional detail), since his more recent The Harvest of the Hills (2000) concentrates firmly on the uplands rather than the coastal lowlands. For ecclesiastical (and some civil administrative) purposes, the main unit of administration nationally was the parish; all the agricultural land of the parish paid tithes to support the parish priest. However in much of northern and western England, many parishes (including St Bees) were very large and were divided into multiple ‘townships’ or ‘vills’ (often corresponding to the lands of specific hamlets, though in some the settlement consisted entirely of scattered individual farms with no known nucleation at all), and the township rather than the parish was the fundamental administrative unit. To complicate matters further, in Cumbria (especially Copeland) hamlet-sized early Medieval ‘vills’ seem to have later been grouped at least for some purposes as ‘members’ or ‘hamlets’ of larger ‘entire vills’; for example, Whitehaven was a vill in its own right c 1270, but was treated as part of ‘Kirkeby’ (St Bees) in 1324 (Winchester 1978, 66). In 1662, townships became the basic units of Poor Law administration; normally these ‘poor law townships’ corresponded to the pre-existing ‘vills’, but in some cases the earlier units were amalgamated or divided. Preston Quarter and Sandwith both formed poor law townships within the parish of St Bees (Winchester 1978, 57). Township boundaries were first systematically mapped by the 1840s Tithe Maps and the 1860s (in Cumbria) 1st Edition Ordnance Survey (OS); their boundaries (barring amalagations or sub-divisions) are generally assumed to have remained constant from the earlier Middle Ages, though this may not always be the case. For agricultural purposes, however, the basic unit was the manor, with its manor court controlled by the lord of the manor (who could be an institution such as a Priory, as well as an individual); the manor was normally a single estate. A township could contain several manors, boundaries of manor and township could differ, and at worst (for the historian) manorial and township boundaries could completely cross-cut; in west Cumbria, manor and township boundaries did however tend to coincide. Within a township, lord, freeholders, tenants, and peasants had often-complex rights and obligations; among these, the lord of the manor held the mineral rights (including rights of access to land for mining and transport of minerals). On a broader scale, the whole of western Cumbria from the Derwent to the Duddon formed the Barony of Copeland (Winchester 1987, 19-21). Within the Barony, the northeastern and southern areas formed the estates of Cockermouth and Millom respectively, the remaining area from the Cocker and Derwent to the Esk forming the lordship of Egremont, with its lordly focus at Egremont and its religious focus at St Bees. The parish of St Bees may originally have occupied the whole of the lordship; by the time parish boundaries fossilised in the 12th-13th centuries most of the larger settlements had attained their own parochial status, but St Bees remained a very large multi-parish township (including the uplands of Copeland forest as a large detached area, as well as the main coastal area). The study area and its immediate context lay entirely within the parish of St Bees. In terms of townships (as mapped in the 19th century), the Southwest arm of the NT area lay within Sandwith (which was a separate township to Rottington, wholly outside the study area); the boundary followed the trough of the St Bees Shale outcrop, running across unenclosed land from the coast north of the Barrowmouth gypsum mine (28950), east (largely beneath 20th century tip 28984), then south on a more complex course through enclosed fields within what is now the west side of the Marchon site, to leave the study area near Townhead. The remainder of the area was within Preston Quarter township, which also extended south as a narrow strip to include the Priory hamlet at St Bees (though not the main village on the other side of the Pow Beck valley), and NE to include the Brackenthwaite area (though not the town of Whitehaven itself, seemingly carved-out as a separate township at an early date). The various land-grants relating to the study area in the records of St Bees Priory, translated and discussed by Peter King (Appendix 1), presumably relate to manors or smaller estates, whose boundaries do not necessarily conform to the known townships (in their early Medieval form, let alone any later amalgamations). However, some interpretations can be attempted. It would appear that Whitehaven was separated-out as an estate or manor from at least the 13th century, although its development subsequently spread into Arrowthwaite manor. The township of Preston Quarter (possibly a ‘poor law township’ as described above – the name Quarter sounds like a late coining) appears to have consisted of at least two earlier manors or estates - Arrowthwaite to the north and Preston to the south. The hamlet of Arrowthwaite itself lay just to the east of the study area; the estate or manor appears to have covered the whole northern part of the study area, including the coast from the south side of Whitehaven Harbour (probably with a boundary to Whitehaven township between The Beacon and the head of the present harbour) to the boundary with Sandwith township just north of Barrowmouth; its inland boundary may have run from How Gill (now occupied by the Corkickle Incline), along the crest of the Howgill Ridge and close to the NW side of the Marchon site (in modern terms). The area SE of this (including the Southeast arm of the study area) appears to have formed the manor or estate of Preston (‘Priest-ton’). The Priory demesne lands (lands farmed ‘in-house’) appear to have occupied the whole of Preston, perhaps extending into Arrowthwaite to include Monkwray. This strongly suggests that the name ‘Preston’ reflects the pre- monastic endowment of St Bees church, passed to the priory at its foundation. It is not clear whether a lost hamlet of ‘Preston’ ever existed; settlement within the estate may have ben entirely dispersed, or (perhaps the most likely) the name may have referred to the settlement round the church at St Bees (which, unlike the main village on the other side of the Pow Beck valley, was within Preston Quarter township). In this case the name may have been an Anglo-Saxon alternative (precursor?) to the Scandinavian ‘Kirkby’. To the SW, Sandwith formed a separate township and manor, under much less direct monastic control. Rottington formed a further separate poor-law township, wholly outside the study area; however some of the more tentative place-name correlations would place the bounds of Rottington as perambulated in the Register as running along the east side of Sandwith rather than between Sandwith and Rottington, in which case they describe part of the south edge of the study area (including a monastic capraria, probably a specialist goat-farm by analogy with the better-known vaccaries and bercaries). Within Preston Quarter, the Medieval landcape has been swept away without visible trace by Lord Lonsdale’s remodelling of the landscape in the 1830s (see below). Two 18th century maps (CROW: TNCB 24/4 and CROC: D/Lons W/7 Portfolios of engineering drawings; ‘Sundry Old Collieries’, f 25), and Matthew Read’s 1738 painting, show an enclosed strip-field system around Arrowthwaite hamlet, with the adjacent study area partly occupied by rather irregular hedged enclosures and partly unenclosed. Within Sandwith, however, the outlines of the Medieval agricultural system still survive within the modern field system, with crofts or enclosed strips running east and west from Sandwith village (outside the study area), and a major ringfenced enclosure (28929) centred on Tarnflatt Hall and extending from Fleswick Bay to enter the extreme SW end of the study area – this enclosure may well be of Medieval origin. The correlation of these features to any of the named features in the Register documents translated by King, and the location of these on the ground, is uncertain; substantial information on the Medieval landscape of the Sandwith part of the study area is probably available if and when a full documentary study is possible. The importance of Whitehaven during the Medieval period is uncertain. Most studies of the town (eg Collier 1991, 7-8) assume that it was negligible, on the basis of a survey of 1566 (Fox 1921), which reported that Whitehaven was only a small village of six householders, with no licence for loading or unloading, and one nine-ton vessel crewed by hired fishermen. However this survey was prepared by a commission of Cumberland landowners, as the (belated) contribution to a national survey by the Crown, aimed at least partially to curb smuggling – it is not inconceivable that they had their motives for minimising the importance of any local trade. The limited to the 14Medieval documentation (Appendix 1; Hay 1979, 11-19) is also at odds with the survey. The 13th century separation of Whitehaven as a township (Reg 372) implies a substantial hamlet, and in 1359 it had a chaplain (presumably implying a chapel-of- ease); it furnished shipping to the Crown in 1172, and had three ship-masters in 1299. The early 14th-century leases of Whitehaven properties also imply substantial settlement, the settlement seemingly having overflowed its township boundary into Arrowthwaite manor. However by 1517, St Bees Priory derived only minor income from quayage (implying the existence of at least a small built quay) and tollage, on ships trading largely with the Isle of Man (Tyson 1985, 173); a later-17th-century Lowther description seemingly referring to the 16th century claims that Whitehaven had only 3-4 cottages, a small ‘Peer’ of wooden piles and stones serving 3-4 ‘barks’, and a ruined chapel. It therefore seems that Whitehaven was an appreciable settlement at least from the 12thth centuries, perhaps mainly a fishing village but also with some shipping trade. Whether it genuinely shrank to six (or three to four) householders by the 16th century, or whether the 1566 survey was economical with the truth, is less clear; there does seem to be some evidence for decline. Nor is it clear whether the settlement extended far enough along the south shore of Whitehaven Harbour, beyond the township boundary, to enter the project area. Medieval Industry Apart from agriculture, the only Medieval activities clearly documented within (or at very close to) the study area are coal-mining and salt-making. Of the several known processes for extracting salt from seawater (Cranstone 2006, 4-5). the dominant earlier Medieval method (in Cumbria as in most of England and probably Scotland) was ‘sleeching’. In this process, salt-encrusted silts (formed between spring tides under dry summer conditions) were scraped up, taken to a saltworks above high-tide level, placed into filter pits, and the salt leached out using seawater, to form a strong brine that was then boiled in huts called ‘saltcotes’, normally using small lead pans over fires of peat or wood. The process required a silty (and therefore marshy or estuarine) coast. Coal was not normally used as fuel, perhaps because the heat and/or sulphurous fumes corroded the lead pans. In the late Middle Ages, this process began to be replaced by ‘direct boiling’, in which seawater (collected in cisterns (‘sumps’) and settling tanks on the foreshore) was simply boiled, in large iron pans over coal fires. This process is conventionally considered to have been developed on Tyneside in the early 15th century. In contrast to sleeching saltworks, direct-boiling sites (‘pans’) were normally located on rocky coasts (perhaps to obtain cleaner less silty seawater, or because rock-cut cisterns and tanks were more robust than built stone or timber ones). As well as its direct importance for salt-making, this process was probably the first appreciable coal-fuelled industry, and was therefore of wider importance for the development of coal-fuelled technology more generally (coal has very different properties as a fuel than peat, wood, or even charcoal, and its successful use requires the development of new types of furnace and firegrate). The 13th-century references to coal-fuelled saltmaking in Arrowthwaite are therefore extremely interesting, especially on a coast that (at least in its modern condition) is far more suitable for direct-boiling than sleeching. The combination of coal fuel and a rocky coast does strongly suggest a form of direct-boiling process, although probably not in the ‘classic’ form that took off in the 15th century; it may well have used lead pans (or even ceramic pans or vessels), and the coal-burning may well have been inept by later standards. The documentary evidence indicates coal mining in/under a cliff or rock, with an adjacent saltworks, with reasonable access (though a road had to be built), and within Arrowthwaite manor. On geological evidence, the two areas that match these criteria and have outcropping coal seams (though thin upper seams only workable at outcrop, rather than the thick seams that were the mainstay of later mining) in the coastal cliffs are the south side of Whitehaven Harbour, and Saltom (where faulting exposes upper seams which are below sea level on the reminder of the open coast). The latter certainly had 17th-century saltpans, behind South Beach (29074) within the study area, and also further SE. However the place-name Saltom is almost certainly significant, and the Register grants seem to connect the saltworks with the coastal strip of waste, though with provision for access across the cultivated land of Arrowthwaite (whereas access to South Beach would probably have been along the coast from Whitehaven). There is therefore a strong circumstantial case for locating this saltworks at Saltom, and a funnel-shaped trackway (CROW: TNCB 24/4 and CROC: D/Lons W/7 Portfolios of engineering drawings; ‘Sundry Old Collieries’, f 25) from Arrowthwaite hamlet to the clifftop above Saltom may well be the access route prescribed by the charters. If so, no visible evidence survives (unless the rock- cut cistern 29038 derives from this works, rather than the 18th century saltpans associated with Saltom Pit); the site may have been destroyed by erosion or by the construction of Saltom Pit, but may survive either beneath the Pit or along the adjacent coast. In addition to the saltworks of St Bees Priory, and the adjacent Wetheral Priory works, Calder Abbey also had two saltworks in Whitehaven, first documented in 1152-3 (Thorley 2004, 135, 139). These must have been on Whitehaven harbour, probably to the east of the National Trust boundary. The difference in place-name suggests that they were not immediately adjacent to the St Bees and Wetheral saltworks, and the probable location plus the lack of any reference to coal suggests that they may have been sleeching works. Other names in the document do not all appear to be townships, so the implications for the 12th century status of Whitehaven as a settlement are not immediately clear. Most of the known Medieval evidence for coal-mining is in relation to saltmaking, and must refer to the coastal cliffs. However it does seem likely that the outcrops of the main coal seams, along the lower west side of the Pow Beck valley from Whitehaven to Greenbank, were already known and exploited to some degree, especially at the later nuclei of Howgill and Greenbank where ravines cut through any drift cover and exposed the seams. The other industry within the study area that may have Medieval origins is quarrying, for sandstone and/or limestone, though neither is documented in the documents studied. The origin of the substantial freestone quarries along the clifftops of the SW end of the study area, and extending along the St Bees Sandstone to Aikbank, is not known – it may or may not have started within the Medieval period. The outcrop of the Magnesian Limestone, from Barrowmouth gypsum mine east to the Marchon works, then SE along High Road and Wilson Pit Lane, would have formed the nearest source of lime for mortar in the construction of St Bees Priory (and other mortared Medieval buildings in the area), though the most likely location for any exploitation is perhaps the area from Wilson Pit Road to Demesne, closest to the Priory, within its demesne land, and just outside the study area. Post-Medieval to Modern For the Medieval and earlier periods, both archaeological and historical evidence relating to the study area have been very limited, and the purpose of this report has been to set the local and regional scene, to raise rather than answer questions, and to stress the need and potential for future research that may produce the ‘hard’ evidence so lacking at present. For the Post-Medieval period, and especially from the 17th century onwards, the problem is the reverse; there is a wealth of historical and archaeological evidence (much of the former unstudied and at present virtually unstudiable), and the problem is to summarise the main evidence directly relevant to the sites and landscapes of the study area, within a briefly-sketched wider background. The use of the study area, and the surviving archaeological field evidence, are overwhelmingly ‘industrial’, and are dominated by coal-mining and its infrastructure – even the agriculture, at least within Preston Quarter, is arguably part of the infrastructure of mining. The section therefore starts with a brief survey of the background (land-ownership, and the development of Whitehaven town and harbour), before an industry-by-industry survey of the (mainly) 18th-20th century evidence, finishing with a discussion of the broader landscape which these industries made up. Background St Bees Priory was dissolved in 1538. As noted by Peter King (Appendix 1), the priory lands were initially leased by the Crown to Thomas Leigh. However the lands, rectory, and manor were sold in 1553 to Thomas Chaloner, whose son sold it in 1599 to Thomas Wybergh (Wilson 1905, 359). In 1600, Wybergh mortgaged it to George Lowther; the Lowthers were an old gentry family, and already major landowners around Lowther and Penrith. In 1630, Sir John Lowther bought a moiety (half-share) of the manor, and settled it on his younger son Christopher, thus setting up the Whitehaven branch of the family (Hay 1979, 19-20; Beckett 1981, 14; Collier 1991, 2-3). The Wybergh family remained active in the area; Lowther foreclosed the mortgage on their remaining moiety in 1663, but did not gain full control and possession until 1671 (Tyson 1985, 174, 201). Very little is known about the development of the estates, or of Whitehaven town, during the Leigh, Chaloner, or Wybergh ownerships, and nothing specific to the study area. This may however reflect absence of evidence, rather than absence of activity. It does appear that by 1630 the centre of gravity of the estate was already swinging from St Bees to Whitehaven. The Whitehaven-St Bees estate descended from Sir Christopher (died 1644) to Sir John (active 1663-1706), then to Sir James (active 1706-1755) (Beckett 1981, 14-19; Phillips 1979, ix-xvii). Sir James bequeathed it to his cousin Sir William Lowther of Holker; on his death in 1756, it passed to Sir James Lowther of Lowther, thus reuniting the Whitehaven and Westmoreland estates of the family. Sir James was created Earl of Lonsdale in 1782, and the estate passed through successive generations of Lonsdales; the Lonsdale family withdrew from direct management of their collieries and other industries in the later 19th century, and the Whitehaven estate was finally sold off in the early 20th century. The first Sir Christopher lived at Old Hall, Whitehaven, but succeeding generations lived in London and ran their Whitehaven estates via agents (normally separate estate and colliery stewards). Almost all the letter-books of correspondence between the Lowthers/Lonsdales and these agents, from the later-17th to the mid-19th centuries, survive. These, with the rest of the Lonsdale Papers, form a vast source of historical information, much of it relating to the study area and individual sites within it, but systematic use of this material has been far beyond the resources of this project; in practice use of this material has been restricted to the published or typescript material for 1632-1644 (Hainsworth 1977), 1617-1675 (estate memoranda books; Phillips 1979), 1693-1698 (Hainsworth 1983), plus the many secondary resources that have used the material (tending to concentrate on the time of Carlisle Spedding as colliery agent (1730- 1755)). Just as the study area in the Medieval period can only be understood as part of St Bees parish and Priory estates, the archaeology and history of the area from the 17th century onwards cannot be understood without some reference to the development of Whitehaven town and harbour. The main published surveys for the development of the town are Hay 1979 (and other editions), Beckett 1981, and Collier 1991; the development of the harbour in the 18th and 19th centuries is discussed more specifically by Scott-Hindson 1994. By the end of the 16th century, Whitehaven contained eight tenements and four ‘camerae’ for bakehouses, and by 1631 there were 24 customary tenants and one ancient freeholder, mainly located on the west side of Quay Street in the ‘Old Town’ area (Collier 1991, 8-9). These included Old Hall, a name which confirms an earlier manorial origin, and the settlement was clearly more substantial than indicated in the 1566 survey (though whether this reflects growth during the Chaloner and Wybergh ownerships, or economical truth in the 1566 survey, is less clear). Sir Christopher Lowther started the active development of the town, centring on the Market Place (though the market charter was not granted until 1656) and beginning the grid-pattern development of the area east of Pow Beck (the modern town centre), though this only took off under Sir John Lowther after 1663. Collier does not discuss the origins of the ‘New Town’ seemingly a fairly small southward extension of Old Town; it is hard to see the name being applied after the far more major developments east of Pow Burn had started, so the development of the town under Sir Christopher (or perhaps under the Chaloners and/or Wyberghs) may have been rather more substantial than she indicates. Whitehaven then developed into a substantial town, with major coal-using industries (mainly along the strip southwards from Old and New Towns towards the Ginns and Howgill collieries), and substantial warehouses and other port-related industries within the town itself. In many respects, the importance of the town peaked around the 1740s, when it was a major transatlantic port, with all the wider connections (including those to colonialism and slavery) that this implies. As already noted, Whitehaven existed as a port from at least the late 12th century, and there are grounds for suspecting that the 1566 survey (much relied on by all the modern writers) understated its importance. However it was clearly a minor port, and there is no evidence for any constructed harbour installations, or for any maritime activity extending into the Trust holding; ships and boats were probably beached on the head of the natural (progressively silting) inlet, beneath the seaward side of the modern town centre. As soon as Christopher (later Sir Christopher) Lowther obtained practical control of the St Bees/Whitehaven estate in 1630, he commenced active development of the harbour, as an outlet for his coal and salt; the core of Whitehaven’s prosperity in the later 17th century was its virtual monopoly of the coal supply to Dublin, though initially salt (see below) may have been as important. The first ‘peer’ (now the Old Quay) was under construction in 1632-4, but was being altered or extended in 1636, by which time there was also a ballast quay (Hainsworth 1977, 9, 13, 64, 194). However ballast dumping and storms were causing problems in the mid 17th century (Tyson 1985, 175-7). The Old Peer was extended in 1679-81, and the harbour behind it deepened by quarrying of sandstone from its bed. This improved loading from a set of iron ore staithes, already in existence on West Strand on part of the area of the later coal staithes (below); the location appears to have been close to The Beacon, possibly extending just into the project area (Tyson 1985, 183). A ‘breastwork’ here may have been either military or a sea defence. There were also an anchor smithy and a blockmakers’ shop further NW, near the base of Old Quay. Beyond Old Quay, the cliffs extending to Tom Hurd rock were already unstable, with a fall onto the saltpans in 1676, and debris from cliff falls washing into the harbour in 1683; these were caused by quarrying, and small-scale unofficial coal mining by local inhabitants (of outcropping thin seams above the main coal, never exploited by the large-scale Lowther collieries). The state of the harbour at the end of the 17th century is shown by the various Pellin plans in CROW; the 1695 plan in particular (CROW: Plan of Whitehaven, 1695; Collier 1991, 16) shows the Old Pier with the Quay and iron ore staithes to its SE, while the c 1705 plan (Collier 1991, 17) shows a breakwater or harbour wall to the W of Old Pier, beginning the westward extension of the harbour. The 1695 plan is particularly interesting in that it shows Tom Hurd Rock (now a relatively minor feature on the South Beach foreshore) as a major rock shelf, extending north to almost opposite the (then) end of Old Pier, and enclosing a narrow inlet of deeper water running from the west side of Old Pier south beneath what is now the Old Fort and Wellington Pit area. The ‘Old Salt Pans’ were beside this inlet; its infill deposits (assuming they survive) could well also contain remains of some of the wrecks for which Tom Hurd Rock was notorious. Despite concern from Lowther in 1672 and 1680 (Tyson 1985, 177, 188), the Rock appears to have been quarried by 1687-8 (Winchester and Wane (eds) 2003, 104). Lowther granted a lease for quarrying on the Tom Hurd foreshore in 1692; by 1698 he was again concerned that the quarrying (partly for grindstones) ‘layes the peer bare’ (Hainsworth 1983, 215-6, 597). To judge by later maps, the original Tom Hurd Rock was indeed largely destroyed around this date, no doubt with considerable effects on the coastal erosion of the South Beach area, as well as on Whitehaven Harbour. Whitehaven Harbour developed very considerably in the 18th century, as described by Beckett, Hay, Collier and Scott-Hindson. The enclosure of the South Beach area began with the construction of New Pier (now Old New Pier) in 1742; this ran west then north from the contemporary highwater line at Old Fort, the east-west part being now incorporated into the south wall of the harbour (Beckett 1981, 163). In 1755, a ‘bulwark’ was also built from the west end of South Beach to Tom Hurd Rock, this is shown on several later 18th century plans (eg this report Plate 3, Collier 1991, 22-23), and seems to have been finally destroyed by the severe storm of 1796 (Scott-Hindson 1994, 35). The exclosure of South Beach from Whitehaven Harbour was completed by the construction (by Rennie) of the New West Pier from 1824-30; the limekiln built into Old Fort was constructed for this (Scott-Hindson 1994, 94-100). Saltmaking Since salt is the earliest Lowther industry to be documented in any detail, and since, unlike the other industries to be discussed, its floruit was in the 17th century, it is logical to discuss it first. It is not known whether the Medieval salt industry, at Saltom and/or on Whitehaven harbour, survived to the Dissolution, or continued throughout the Leigh, Chaloner, or Wybergh ownerships, though in January 1634 Wybergh sold saltpans and coals in ‘Henrie Davies groundes’ to the Lowthers for £88 (one year’s rent or profit) (Phillips 1979, 38). These pans may have been on the edge of Whitehaven town, and therefore just outside the study area. The South Beach saltpans (29074) Sir Christopher Lowther was already building two saltpans near the Quay by October 1632; these seem to have been quite separate from the Wybergh saltpans. One of these was in production by (probably) February 1633, though the other was still under construction. The panhouses contained pans of plate iron, and were clearly direct-boiling works (Hainsworth 1977, 9, 13, 60-61), though in March 1633 Lowther appears to have been experimenting with salt refining (Hainsworth 1977, 60) – in this process, impure imported salt (or, later, Cheshire rocksalt) was dissolved in seawater, the impurities settled out, and the brine boiled to produce clean salt. In December 1633, the ‘salt pannes & house’ were worth (or had cost) £280, as compared to £100 each for for the Pier and coal works; by June 1636 Sir Christopher valued the pier at £303, and the salt pans at £467, and around this time the pans were using 50 tons of coal per week, and Lowther was thinking of leasing them out (Hainsworth 1977, 107, 187, 191-2). These pans were presumably those for which design drawings were prepared by Rowland Jackson in 1631 (Tyson 1999, 215-6), though it is not clear whether these designs were actually executed; their references to ‘French salt’ imply salt refining as well as primary production. On 20th May 1637, the Lowthers leased out the pans, sumps, and tools to two salters (implying two pans?), David Bibby and Thomas Younghusband, Lowther guaranteeing the supply of coal. However this agreement was dissolved on 31st May, the pans being leased to a partnership of Richard Wybergh and Thomas Lowman on the same terms. This agreement in turn must have been terminated by 18th February 1638, when Christopher Lowther engaged three salters to work three pans; Robert Stockdell had Pan A and was allowed 28 loads (of 32 gallons) of coal per day on a 6-day week; David Bibby had Pan B with 26 loads of coal; and Patrick Card had Pan C with 13 loads. The salters had the use of the pans, sumps, and tools. Robert Stockdell was contracted to deliver 14 bushels of salt ‘with one shake’ per working day, and to pay two-fifths of the cost of ‘bryne, winding blowers [damper plates for the furnaces], dressing shovells, creeper and battle shaftes & coles’; David Bibby was to deliver 13 bushels daily on the same terms; Patrick Card was to deliver 34 bushels of salt, and pay one-fifth of the costs. Patrick Card may have been engaged on a different basis to Bibby and Stockdell, in which case the relative size of their pans is unclear; alternatively, Pan C may have been a salt refinery (Hainsworth 1977, 213-217). The pans clearly remained in use through the mid 17th century, with occasional known references (Tyson 1985, 1999). Detailed accounts survive for 1675, by which time the pans were taking nearly half the coal output from the Three Quarters Band (seam), the works included at least one substantial pump (for pumping brine from the foreshore sump to cisterns within the works), and seems still to have had three pans. However this was precisely the time at which the coastal salt industry, especially round the Irish Sea, was coming under under intense competition from Cheshire, where the discovery of rocksalt had led to greatly increased production and decreased costs. At least one pan was damaged by a cliff fall in November 1676; part of the site was rented out to merchants, but at least one pan was brought back into use, only to close by the early 1680s (when Lowther rented the Bransty pans on the other side of Whitehaven Harbour); Denton’s 1687-8 Perambulation only mentions the Bransty pans (Winchester and Wane (eds), 2003, 105). The location of these pans requires some discussion, since Tyson (1999) places them actually on the Old Pier. However he himself notes that the Pier would not accommodate the pans designed by Jackson, and the reference that he regards as conclusive refers to pans on the Key rather than the pier (in 1669); Tickell’s 1675 sketch (Tyson 1999, 201) shows two panhouses, one to the W of the base of the pier, the other very close to the base but not necessarily on it. Pellin’s 1695 plan (CROW: Plan of Whitehaven, 1695) clearly shows ‘The Old Salt Pans’ tucked in below the (then) seacliff just west of the Pier and on a narrow inlet between the main coast and Tom Hurd Rock(at the location of gazetteer site 29074), and the 1676 cliff fall also indicates this location rather than the pier itself. However, Pellin only shows two panhouses, so it remains possible that a third panhouse, and/or the brine pump for the works, was actually on the pier, or on the Quay to its SE. The technology of the site also requires brief discussion. Tyson (1999, 202) assumes that the site was a sleeching works; however all the evidence indicates that the saltworks was in fact a direct-boiling works (and perhaps also a salt refinery), as would be normal for a new works of this date. However, Jackson’s 1631 plans show at least four (in one plan six) individual pans within a single panhouse, and Tickell’s sketch appears to indicate a similar arrangement. It appears therefore that the 17th century panhouses contained several relatively- small pans and furnaces, rather than the single large pan and furnace, occupying much of the building, shown historically by Brownrigg (1748; see below), and archaeologically by the excavated 18th century Scottish sites at Preston Island and St Monans (Ewart et al 1996; Ewart and Dunn 1997; Lewis et al 1999) – at the only other excavated British direct-boiling works (the 16th-17th century Port Eynon ‘salthouse’ in Wales) the panhouse(s) did not survive (Wilkinson et al 1998). In modern terms, the site of the saltpans as mapped by Pellin lies beneath a small carpark between Old Fort and the revetments and former railways of Wellington Pit (along Pellin’s seacliff); however the works, and particularly the third pan and the brine-collecting and -pumping arrangments, may have extended substantially beyond this plot, both along the (then) coastline and down the foreshore. Although the site was re-used for housing and commercial premises (29075) through the 18th and 19th centuries, archaeological remains may well survive beneath later stratigraphy; by analogy with Port Eynon, the foreshore features could be very substantial. Since no direct-boiling saltworks has yet been excavated in England, and the works appears to have differed substantially from the excavated but later Scottish works (and from the unexcavated Scheduled 18th century site at Crosscanonby), its below-ground archaeology has considerable potential importance. The Saltom Pit saltpans (29038) So far as is known, saltmaking within the study area ceased for about 50 years with the closure of the South Beach works. In 1734-5, however, Carlisle Spedding (Sir James Lowther’s colliery agent) built two saltpans at the new Saltom Pit development, the seawater being pumped by one of the colliery Newcomen engines; the saltpans achieved considerable local success for a while (Beckett 1981, 135-6). The pumping arrangement was illustrated by Angerstein in 1755 (Berg and Berg 2001, 285-6); he indicates a substantial masonry sump. The salt pans were still working in 1760 (CROC: D/Lons/W7/1/16), but had closed by the 1780s, when one of the panhouses was converted into a foundry (Dixon 1801). By far the most detailed known historical description of the direct-boiling process, and its works, is that by William Brownrigg (1748, 49-72, 287-292). Since Brownrigg was a Whitehaven doctor, and friend of Spedding, his descriptions and plates may well draw on the Saltom works, though there is nothing in the text to confirm this. The panhouses of the works were presumably within the main Saltom Pit compound, close to the Newcomen engines; this suggests the north side of the compound, within the area now destroyed or deeply buried by recent landslipping. Prospects for archaeological survival are therefore rather poor. However the sump and other brine-collecting and –transporting channels will have been on the foreshore, where some survival is likely beneath shingle. A rock-cut pit on a foreshore outcrop to the NW of the pithead (29038) may well be part of these arrangements, since it has similarities to elements of the much more complex rock-cut pits and channels at Bank End, Maryport (Cranstone 2006, 21). However it is rather further from the pithead than might be expected, and may have other explanations. Coal Mining Coal mining was the core of the industrial and economic development of Whitehaven, and of the archaeology of the study area. The national development of the industry, and thus the technological and economic context of the Whitehaven industry, is treated in full in the recent five-volume History of the British Coal Industry (in chronlogical order: Hatcher 1993, Flinn 1984, Church 1986, Supple 1987, and Ashworth 1986). This contains only limited, and non-original, coverage of the local industry. While this may correctly reflect the limited overall economic importance of the West Cumberland coalfield in terms of national production, it does not reflect the considerable importance of the Whitehaven industry from the 17th to the early 19th century in terms of technological innovation. Nor does it reflect the quality of the surviving field remains of the Whitehaven industry (largely within the NT area), which is remarkable by national standards. The history of the West Cumberland industry is discussed in detail by Wood (1988); however Wood’s coverage contains surprisingly little site-specific detail, and it is clear that enormous amounts of site information remain to be extracted from the Lonsdale MSS, especially from the agents’ letterbooks. There is also a considerable historical literature of shorter papers etc, ranging in date from the late 19th century to the present, and largely drawing again on the Lonsdale MSS (with a tendency to make repetitive use of the same very limited selection from within the Lonsdale MSS). Coverage in this report draws largely on the published sources, plus maps and documents preserved in non- Lonsdale collections, and material already transcribed into research files at The Beacon, CROW, and Haig Colliery Museum. It should be noted that almost all this material derives ultimately from Lowther/Lonsdale management sources, and is inevitably one-sided; among other biasses, it will systematically ignore any non- Lowther mining, which may have been appreciable in the early part of the period (when the Lowthers already held the mineral rights as Lords of the Manor, but did not yet own all, or initially perhaps most, of the freehold land within the Manor). The limited evidence for pre-Lowther mining may therefore understate the reality. In 1560, Sir Thomas Chaloner granted 50-year leases to 90 of his tenants, including the right to mine coal, and in 1586 he granted St Bees School to take forty loads of coal from his pits in the parish; this indicates a degree of active mining by the landowner (Wood 1988, 3). The pits, both of Chaloner and his tenants, were presumably along the outcrops of the seams, including those along the west side of the Pow Beck valley. Christopher Lowther was already exporting Whitehaven coal to Ireland in 1632 (Hainsworth 1977, 58-60); this may imply that he had taken over pre-existing mines. In 1633, he valued his coal works at £100, a relatively small part of his assets (Hainsworth 1977, 107). By 1637, his intention seems to have been to concentrate on pits in Davis Field and Flatt Field (now under Whitehaven town) to minimise transport costs, though he was also mining at Woodagreen (where a level was being driven) and Greenbank (Hainsworth 1977, 192, 214-215). At some point, this policy of concentrating on mines close to the harbour seems to have included substantial mining in the scarp behind Old Pier, since in the 1670s the ‘olde mines next ye peere’ were ‘poor and almost spent’ (Makey 1952, 201); these mines must have been within or immediately adjacent to the Trust area. Serious planned development of the collieries seems only to have started under Sir John Lowther, from the 1660s; Sir John started a policy of active acquisition of coal- bearing land (finally completed, for Preston Quarter, by the 1820s), and of systematic exploitation of the coal (initially the top, Bannock Band, seam) along the west side of the Pow Beck valley (Wood 1988, 6-7). According to Makey (1952, 199-200), Lowther successfully dispossessed freeholder miners at Greenbank before 1675 (when the colliery was producing c 4000 tons/year), in order to develop the area as one of his integrated gravity-drained collieries; the Lowther-derived documentation may well therefore under-indicate the degree of earlier mining here. The working of these collieries is most clearly described by Makey (1952, 195-249). Since the outcropping seams dipped slightly to the north as well as towards the west, a level (the ‘Ginns Bannock Band Level on a plan in CROW: TNCB 28/13) was driven from near Pow Beck (ie below the outcrop of the seam) at the northern end of the ‘colliery’ until it met the dipping seam; it was then carried south in a strict horizontal course along the seam, thus draining a block of coal between level and outcrop. Man, horse, and materials access was by ‘bearmouth’ (the local term for a tunnel within the seam, opening to surface at the outcrop, and dipping into the mine down the dip of the seam). Coal might also be brought out through the bearmouth, but this involved carrying or dragging the various containers uphill through the galleries, so more often the coal was wound to surface by shallow shafts; the winding being performed by hand-powered windlasses or (from the 1680s) horse-powered ‘gins’. The shafts were carefully located slightly up-dip (ie to the east) from the level, so that if they were sunk further to a deeper seam, these deeper workings were not at risk of catastrophic flooding in the event of any back-up of water in the level. In many respects, this system of operating continued until the 1730s. The Main (or Prior) Band was discovered at Howgill in the 1680s c 100ft below the Bannock Band. Again, a level (the ‘Ginns Prior Band Level’ in CROW: TNCB 28/13) was driven west then south to drain the seam, though in this case the level started from the base of a shallow shaft at The Ginns in order to be deep enough to drain the Main Band; this level was therefore not gravity-draining, and was pumped to surface by a horse- powered ‘gin’. Even so, the levels were not deep enough to drain large reserves of coal, and workings soon spread down-dip of the levels, using barrels and underground ‘gins’ to raise water to the levels. In 1700, a further level (plan in CROW: TNCB 28/13) was driven west then south from Thickett (NX 977 162), its pits including Baxter (just outside the Southeast arm of the NT area, north of High House); a branch level was driven south from near High House, draining Country Pit. The Ginns water level was also extended southwards (presumably at a deeper level than the Thickett Level), eventually passing Gameriggs and Fox Pits and extending as far as Wilson Pit (Moore 1898, 4). As the amount of water to be raised from the Ginns level grew, the horse-gin became inadequate, and in 1716 the world’s sixth Newcomen engine was erected at Stone Pit (to the west of the previous water-gin pit) (Allen 1975). While this was an innovative and broadly successful solution, workings in the Ginns/Howgill area had already extended to the dip of Stone Pit, so that arrangements for raising water to the base of the pit were complex and inelegant from the outset (Allen 1975, 239). It is not clear how soon the Greenbank pits were connected to the Ginns water level; in 1715 Fox Pit was only sunk to the Bannock Band, and appears to have had no connection to the Ginns system (Allen 1975, 239). Although the Greenbank pits were working in 1675, and throughout the 1693-8 period of published Lowther correspondence (Hainsworth 1983), the published material contains frustratingly little site-specific detail. Whenever production or employment figures are quoted (eg Hainsworth 1983, 694-6), it is clear that Greenbank was very much secondary to Howgill in importance. Fires and small explosions were common; the tone of the agents’ comments to Lowther suggests that this was a fairly recent development, presumably as mining penetrated further from the surface. One accident, in October 1693, involved Thomas Fox, ‘our old sinker’(Hainsworth 1983, 71, 81) – since pits were often named after their sinker, this suggests an approximate date for Fox Pit; both Fox and Gameriggs were certainly in production by 1709, when they produced 141 tons and 169 tons respectively in a week (Fletcher 1878, 277). The best-known figure in the development of the Whitehaven coal industry was Carlisle Spedding, many of whose innovations were of national importance (Beckett 1983). Spedding entered the colliery in 1710, at the age of fifteen, became chief colliery agent in 1730 (succeeding his brother Edward, who had held the post since 1707), and remained in post until his death in an explosion in 1755 (Wood 1988, 21- 3). Spedding’s best-known achievement was the sinking of Saltom Pit, from 1729 to 1732 (Moore 1898, 4-5; Ward 1991). The pit (Gazetteer 29036-29041) was sunk on the shoreline west of Arrowthwaite, reaching the Main Band at a depth of c 480 ft. It formed the first undersea colliery in England (though not in Britain – Culross colliery in Fife was working beneath the sea from the late 16th century (Hatcher 1993, 99)). Perhaps more importantly, it also formed a remarkably deep central pumping shaft for the whole Howgill Colliery area (including Greenbank), allowing mining of a wide strip of coal from the previous depth limit (roughly below the centre of the Howgill Ridge) to a line well out to sea. The pit was pumped by a Newcomen engine; an initial 17” cylinder in 1731 was rapidly replaced by a 35½” in 1732, then by a 42” cylinder in 1737, and a second 42” engine was added in 1740 (Beckett 1979). The Newcomen engine, although effective for pumping, could not easily be adapted to provide rotary motion for winding; winding at Saltom was therefore performed by a very large horse-gin, of Scottish rather than Newcastle design (Prevost 1965, 308), whose emplacement can still be seen. The other major surviving feature of the initial construction is the impressive seawall. Although Saltom Pit was very successful as a pumping station, transport of its coal production was more problematic. The only land access to the pithead was by a steep zigzag track down the cliff (29041). It appears that Spedding initially intended the coal production to be exported directly from a small harbour at the pit, with transport from pithead to pier being by a short length of waggonway (still depicted on the 1752 plan); this avoided Whitehaven harbour dues, as well as minimising handling and transport costs. The harbour was constructed in 1731-2, and exported 861 tons in 1733 and 1607 tons in 1734 (Ward 1991, 137), but it was repeatedly damaged by storms, and was barely used after 1734. Instead, the coal was led into an adit from the rear of the pithead, to the base of a shaft from Ravenhill Pit (29035) on the clifftop above, where it was wound up by a further large horse-gin, and then transported to Whitehaven Harbour by waggonway (see below). Since Ravenhill Pit was constructed solely to wind the coal from Saltom (its shaft was never sunk below the Saltom adit to communicate directly with the coal seams (CROW: TNCB 28-21)), it formed a very expensive and inefficient solution to the transport problem; if it was sunk in response to the failure of the harbour rather than as part of the initial plan (which is unclear from Ward’s text, since this seems to conflate the driving of the adit with the construction of the surface track down the cliff), it can be seen as a bold if expensive piece of disaster recovery. The construction of saltpans at the pit, in 1734- 5 (see above) may also be seen as a response to the failure of the initial coal-handling arrangements, in this case by using coal on-site to avoid the need for transport. As already noted, Saltom Pit provided deep drainage for the whole of the Howgill Colliery, which was worked by Spedding as an integrated operation over an area of at least two square miles. The Ginns engines remained in use, and the main entry to the mines for miners and horses (definitely in use underground by 1739) was also on the eastern outcrop SE of Whitehaven town (probably at Howgill bearmouth), by an inclined tunnel largely within the Main Band seam to Saltom and other working areas (Prevost 1965, 311). In 1739 Saltom appears to have been the main winding pit. However Spedding was also responsible for the sinking of Thwaite, King, Duke, Moss, and Kells Pits within or adjacent to the NT holding, together with other pits further east (Wood 1988, 23 – the Lonsdale MSS used by Wood also credits Spedding with the sinking of Fox and Country Pits, but as noted above these are in fact documented before Carlisle Spedding’s time as agent). These pits were all linked underground, and could be used for pumping, coal-winding, man-access (routine or emergency), ventilation (as either upcast or downcast shafts) or any combination of these, in the Bannock, Main, and/or other seams; the detailed operation of the system was therefore extremely complex, and the function of individual pits could and did change quite rapidly. Although some of the eastern pits appear to have been fairly short-lived, the more western pits such as Thwaite, King, Duke, Kells, Ravenhill and Saltom, as well as Wilson and Croft Pits sunk later in the 18th century, remained in use until the late 19th or in some cases 20th centuries. In 1742, the Trustees of St Bees School leased their coal royalties in Sandwith township to Lowther, in very dubious circumstances; the rent was far below the going rate, and the term (867 years) bizarrely long (Wood 1988, 26; Todd 1983). In practice, the Lowthers never tested this lease to the extent of sinking collieries within Sandwith, but mined the Sandwith coal entirely from pits within Preston Quarter. It is not clear whether Fox, Country, or Moss Pits (the nearest pits to the Sandwith boundary in the 1740s) worked the leased coal at this time; from the later 18th century Wilson and Croft Pits (both strategically placed just on the Preston Quarter side of the township boundary) worked the Sandwith coals extensively. The dubious lease came to light in 1814, and the resulting scandal was a factor in the setting-up of the Charities Commission and the beginnings of 19th century educational reform; after protracted litigation the Earl of Lonsdale was forced to pay massive back-rent and compensation, which allowed the development of St Bees School into a major public school. As well as the complex underground connections between the pits (and bearmouths), Spedding was also responsible for the development of a substantial surface waggonway system; since this was used entirely for coal transport, the pits served by the system on the 1752 map (Plates 1, 2) (Saltom/Ravenhill, King, Duke, and Country within the study area) were presumably the main coal-winding pits at this time, the layout of the system suggesting that Fox Pit had also been a major winding pit (see below). The development of the Whitehaven collieries under Spedding contrasts interestingly with the development of the Tyneside coal industry through the earlier 18th century. On Tyneside, with a topography and geology that produced large expanses of Coal Measures above sea level, and therefore amenable to gravity drainage and to downhill waggonway transport to the Tyne (and other rivers and harbours), the coal industry expanded sideways, relying on surface transport by increasingly long and sophisticated waggonway systems (Lewis 1970; Bennett et al 1990), and the major development of deep shaft-mining, made possible by steam pumping and (slightly later) winding, did not occur until the later 18th century. At Whitehaven, Spedding’s strategy pioneered the use of deep shafts, steam pumping, and centralised pumping shafts, to expand the industry downwards; this, rather than undersea mining per se, was probably his major contribution to the development of coal mining nationally. His vision was also made achievable by more detailed innovations in underground mining: early use of gunpowder, bratticing of shafts, ‘coursing the air’ as a means of underground ventilation, the steel mill as a (relatively) safe form of underground illumination in gassy conditions, and ‘methane drainage’ to vent gas blowers to surface (including the concept, though not the practical reality, of gas lighting). The Whitehaven collieries had an international reputation, and were visited and reported- on by overseas industrial reporters such as the Swedish Angerstein in 1755 (Berg and Berg 2002) and the French Gabriel Jars in 1765 (Jars 1774, 238-244; unfortunately there is no English translation of this work). This allows a more balanced appreciation of Carlisle Spedding than the slightly hagiographic tone of much of the literature. Spedding’s main contribution appears to have been the concept of the extensive deep-shaft colliery, draining a wide area for mining in an integrated manner, and the suite of detailed innovations in underground technology that made this vision achievable. However his judgement on surface issues appears to have been less sound; the Saltom harbour was clearly a failure (and a very expensive failure if Ravenhill Pit, the Saltom waggonway, and the multiple- handling that they required, were all a response), and the short working life and apparent rapid alterations to the eastern route waggonways (see below) suggests that these may also have been less than cost-effective. The development of the individual pits of the Howgill Colliery through the 18th century cannot be traced in detail from the published summaries (eg Wood 1988), though a detailed picture, of considerable importance for the history and archaeology of coal mining nationally, could undoubtedly be prepared from detailed study of the Lonsdale MSS and other sources. Gameriggs and Thwaite suffered from explosions and underground fires in 1743, and King Pit was the deepest in the country when it was sunk in 1750 (Wood 1988, 37, 40, 49). In 1752 ‘leaders’ were transporting coal from King, Ravenhill, Thwaite, Kells, Fox and Gameriggs Pits to Whitehaven harbour (Wood 1988, 46); Wood assumes that all this transport was by waggonway, but this is hard to reconcile with the waggonway system as shown on the contemporary Spedding plan (see below). Duke Pit appears to have been sunk at about this time; it is clearly depicted on Angerstein’s sketch in 1755, with a large conical horse-gin house, perhaps connected to the shaft by a short rope-race; the presence of a substantial smoke-trail from this shaft suggests it may already have been used for forced ventilation, by a coal-burning furnace either on the surface or at the foot of the shaft (Berg and Berg 2002, 282). After Spedding’s death in 1755 (in an underground explosion, said to have been in Saltom Pit), the Whitehaven collieries moved from a period of radical innovation into one of steady production and more piecemeal development, though in 1793 Howgill Colliery was still both the deepest and the most extensive in Britain (Wood 1988, 83; Fisher 1793), King Pit being 160 fathoms deep. By 1765, the Saltom Pit workings already extended three-quarters of a mile under the sea (Wood 1988, 70: Jars 1774, 238-244). Croft Pit was sunk in 1774 (Wood 1988, 70), and Wilson Pit (immediately outside the southern boundary of the NT holding) had also been sunk by 1779, when an explosion killed seven miners (Wood 1988, 99-100); this may be why it was ‘not working now’ in 1781. Moss Pit was also sunk at some point between 1755 and 1802 (Fletcher 1878, 288-9). By 1781 Croft had been connected to the waggonway system; in this year the coal-drawing pits were Duke, King, Kells, and Croft, and all sent coal to the staithe by waggonway (Wood 1988, 101); Saltom was only working in the Bannock Band, and Fox and Wilson, although listed, were not in production (CROC: D/Lons/W7/1/18). By 1816, Thwaite Pit was the deepest, at 150 fathoms (Lysons and Lysons 1816, cxxii). At Saltom, a massive new atmospheric (ie Newcomen-style) engine was erected in 1782, replacing at least one of the earlier engines and greatly increasing the pumping capacity (Wood 1988, 84); contrary to some published opinion, this was not housed in the surviving engine house. A mid-19th century photo (Routledge files, print labelled W T & N PIC 39a) shows a roofless engine house to the north of the currently- surviving winding engine house (see below), with gabled east and west walls, a stepped chimney at its NE corner and seemingly a bob opening in the centre of its south (long) wall. This must have housed one of the atmospheric engines; its late survival suggests the 1782 engine, and the bob opening in the side, rather than end, wall is unusual and similar to the 1795 atmospheric engine house at Elsecar (South Yorkshire), strengthening the suggestion. Although no longer cutting-edge technology, the erection of an atmospheric engine for colliery pumping was by no means old-fashioned; although the new Boulton and Watt engines were much more fuel-efficient, and much more easily utilised for winding and other rotary functions, the atmospheric engine was much cheaper to purchase and maintain, and fuel economy was not important on collieries where small coal would otherwise have gone to waste. The Saltom engine remained in use until 1866, so it was clearly a successful development. Unlike the Curwen collieries at Workington, Whitehaven did not pioneer the use of Boulton and Watt engines for winding in the 1780s. However a Heslop winding engine was erected at Kells Pit in 1793. Heslop engines were a local variant on the Boulton and Watt principle, having a separate condenser in the form of a second cylinder, at the opposite end of the beam from the steam cylinder Fletcher 1878, 292-5). They were manufactured at Seaton ironworks near Workington, and were widely used in Cumbria. The Kells Pit engine was later moved to Low Wreah Pit (outside the NT area) (Moon 1973, 15), and eventually to the Science Museum as the last surviving Heslop engine; it is currently in the Science Museum stores. A second Heslop engine, built for the Lady Pit at Whingill, was moved to Wilson Pit in 1841 (CROW: TNCB 28/13); its substructures may survive on the site. The saltworks at Saltom Pit was closed in the early 1780s (if not before), and in 1786 one of the panhouses was converted into an iron foundry to supply wagon wheels, furnace bars and other cast-iron goods to the collieries. The iron was melted in a cupola furnace holding over ½ ton, and blown by bellows operated by the (1782?) engine; this remained in use until at least 1809 (Dixon 1801, 112; CROC: D/Lons/W7 Portfolios of engineering drawings: ‘steam engines’ portfolio, reverse of index). This was a very early use of the cupola remelting furnace, and a possibly-unique use of an atmospheric engine to operate bellows. In 1802, Howgill colliery had 453 employees, of whom 149 were immigrants from outside the county. The largest numbers (44 and 41 respectively) were from Ireland and Scotland, the remainder being largely from the other counties of northern England. This workforce also included 124 women, many of them in underground jobs (though there were no female hewers – this appears to have been an exclusively male occupation) (Wood 1988, 87-88, 97-99). A phase of more systematic updating began with John Bateman’s re-appointment as colliery agent in 1802 (he had previously served in this role from 1781 to 1791), and continued under his successor John Peile (agent 1811-1847). The wooden underground railways were replaced with cast iron between 1802 and 1813. These were cast in a Lonsdale foundry, but it is not clear whether this was still at Saltom, or at one of the foundries within Whitehaven town – by 1815 the main foundry was at Newtown, and was st