sign in sign up
<<
Home , Bagtor, Dendles Wood, East Okement River, Forest of Dartmoor, Holne, Manaton

Deserted Sites and Rabhit-Warrens on , By CATHERINE D. LINEHAN

This paper summarizes the evidence for enclosures and homesteads on Dartmoor from the time if Domesday onward. Part I lists the known deserted sites, including those for which there is asyet no identified name or documentary evidence. A fuller description is given ofthe deserted sites in Park;' of three sites in a rabbit-warren on Common in Widecombe-in-the-Moor; if a group of sites in Corn wood; and if Butterberry in . Maps are given of these four special areas (FIGS. 52, 54, 56, 58), together with a distribution-map C!f sites on Dartmoor as a whole (FIG. 47), and sketch-plans ifa number if individual sites. I Part II contains a summary of the written evidence for rabbit-warrens on Dartmoor, anda list ifI6 sites which contain pillow-mounds. INTRODUCTION HE upland region of Dartmoor in Devon contains some II 1,000 acres (FIG. 46). Of this the owns about 29,000 acres of un­ T enclosed land, together with certain areas of enclosed land. The remainder, estimated at about 59,000 acres, consists of 'venville commons'­ land subject to certain rights by Scale of Miles local villages." o 5 10 20 30 40 Dartmoor has long been Scale of KJlometr~5 known and studied for its pre­ 20 40 60 historic remains and mining activities. Little interest, how­ ever, has been shown in its many deserted homesteads, or in the old rabbit-warrens with their pillow-mounds, known locally as 'burys' or 'buries'. Field-work during the past four years has led to the recording FIG. 46 of I 6 warrenscontainingmounds MAP OF SW. SHOWING POSITION OF (TABLEIII,p. 124) and ofover 100 DARTMOOR

r The scale of all these sketch-plans is 96 ft. to I in. The measurements were taken from the inner faces of walls, and, on the pillow-mounds, from the centres of the surrounding ditches, which are indicated by hachuring. Walls of buildings, where visible, are blocked in, and, where they can only be assumed, indicated by broken lines. Enclosure-walls, hedge-banks, etc., are shown by double lines, and these, as well as streams, pounds, etc., are named. The sketch-plans are so orientated that the houses slope downwards from the upper to the lower end. This indicates the fall of the land except in the few instances in which the houses are built along the contour, where it is shown by an arrow. 'L. Dudley Stamp and W. G. Hoskins, The Common Lands if England and Wales (1963), p. 185. 113 114 CATHERINE D. LINEHAN

","'MPTO 5 5 "1C'pBELST~NE • SOUTH TAW TON 57 , •• 53 J /- · 58·-;~54 ( ,vJ;3 56 ( ~ ) 780 • -SRIDESTOWE 1\ ~ I At (yr·""CH..!!A""G'--'FO<.e~~~ogAMPSTEAP I I ~~":s~~:;':~ 1 0 NORTH ~DFORD ! OVEY ~- 9 ~. 6L. 62 250 50· 48 I 1 W4';'y 400 o. MANA 0 I 3R\ ,~ 37. 42035 39 MARY 630 61 I '1 \lil. J'"i 036

.:> , 88 92) I. 4344-r 3 0 PETER TAVY 89 99 o~, (22· .24 W'4 0 WIDECOMBE- • 86 .82 :i '8J ~IN-THE-MOOR 08' 21.. 2 95 :9'3 1.·8 _,-::90 (- SAMFORD / 30 15 ~ 6'7 23.

). SP1NEY. ~/ )~233 • 19 ;=~~3mND~~2. 1. '/WHITCHURCH 065! 84.. 109 ~ 10; MQQ!l. l-, WALKHA"PTON 083 0/ /Palgnlon" I ASHBURION ? ~ 1) M '1 80/ ! Reservoir 21 V • ~ ~ , / . Burratcr 71 r ~o. Res :~.....-v _5 l' StlfWIOR.''l3 74 I ~ W81g?~ UCKFASTLEIGH elY' ( 720';<'.;' 68 I\ 1080 7ui '/;'" 67 \':J-~9 :;};~~W~ 0 7: • PARISH >. WfJ\ 15•• 14 ~ I' 1- fB:IQB. 66• 16. 012 013 o SITE , , BRENT if: WARREN \ • I r'

1 0 I 2 3 4 5 6 7 8 9 10 Miles IwwwI IwwwI ! IwwwI lweeee! • I"""" ....., ....., I"""" 1 1 ....., I""""i I"""" 1 0 """2 3 4 5 6 7 8 9 10 11 12 13 14- 15 15«me. FIG. 47 MAP OF DARTMOOR, DEVON showing positions of parishes, deserted sites and rabbit-warrens Reproduced by permission ofJohn Bartholomew & Son Ltd.

KEY TO FIG. 47

Site Site no. Parish Name no. Parish Name

1 Ashburton Cold East 9 Ridge Ley Rushlade 10 Stiniel Cottage 2 " 3 Down 1 1 Corn'~ood Ford Barn 4 Lambsdown 12 Ford Waste Down 13 Harrowthorn g Buckl~'nd in-the-Moor Higher Pudsham 14 Hawns 7 Pudsham Down 15 High House 8 Stone 16 Parkland Newtake DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR !I5 17 Dean Moor 61 Peter Tavy Butterberry, II 18 Harford Lower Piles 62" Redford 19 Holne Combestone Wood 63" Wapsworthy 20 Holne Moor 64 "" White Hill 21 " Holne Ridge 65 Hill 22 Ilsington Holwell Tor 66 Blackalder 23 Mountsland 67 Hen Tor 24 " Pinchaford Down 68 " Shavercombe 25 Assycombe 69 " Willings Walls 26 Brimpts, Dolly's Cott 70 She'~pstor Legis Tor 27 " Brownberry 71 Outcombe 28 Believer 72" Ringmoor Down 29 " Broom Park 73" Warren 30 " Cocks Lake 74 Whitten Knowles 31 " 75 Corringdon New- 32 Swincombe, Bishop's lands Brake " House 76 " Gingaford Swincombe Cottage 33 " 77 "" Yalland Warren 34 " Whiteslade 78 Throwleigh Shilstone Combe 35 Beckham, West 79 Ugborough West Glaze River 36 Black Hill 80 Crazywell Farm 37 Blissmoor, Hayne 81 Hucken " Down 82 Merrivale Warren 38 , I 83" Roundy Farm 39 " Challacombe, II 84 Routroundle 40 " Cripdon 85" Yes Tor Farm 41 Great Houndtor 86 Whitchurch Staple Tor Farm 87 Widecombe-in-the-Moor Blackaton 42 Hayne 88 " " "Blackaton Tene- 43 Holwell ments Holwell Cottage 44 " 89 " " Blackmore 45 " Holwell Lawns 90 " Blackslade 6 Houndtor, I 1 " " Bonehill, Lower 4 " 9 " 47 " Houndtor, II 92 Combe 8 Manaton Rocks " " 4 " 93 " Corndon Down 49 Meavy Higher Cadworthy 94 " " Corndonford, Little 50 Lower Hookney 95 " " " Dinna Clerks near Leapra Farm 96 " " " Drywell ~~ Ok~ham'pton Okehampton Park, 97 " " " Eastern Combe Chapel 98 " " Hut Holes 53 Okehampton Park, 99 " " Langworthy " IA,B,C 100 " " Natsworthy, Berry 54 OkehamptonPark, II " " Pound 55 " Okehampton Park, 101 North Rowden " III 102 " " Vaghill 56 Okehampton Park, " " " Yar Tor " IV ~~~ Sha~gh Prior" Shavercombe Foot 57 Okehampton Park, 105 Spanish Lake " V 106" " Trowlesworthy Tor 58 Okehampton Park, 107 Sheepstor Gutter Tor " VI 108 " Legis Lake 59 Okehampton Park, 109 Walkhampton Newleycombe " VII 110 Riddipit 60 Peter Tavy Butterberry, I "

WARRENS Site Site no. Parish Name no. Parish Name W.I Lydford Beardown W.IO Sheepstor Sheepstor W.2 Huntingdon W.II South Brent Yalland W·3 " Wistman's Wood W.12 "" Zeal Burrows W·4 No;:h Bovey Headland W.13 Skaigh W·5 Shaugh Prior Hen Tor W.14 Walkhampton Merrivale W.6 Trowlesworthy W.15 Widecombe-in-the- Vaghill and Eastern W·7 " Willings Walls Moor Combe W.8 Ditsworthy W.16 Ugborough Redlake Tramway Legis Tor W·9 " 9 116 CATHERINE D. LINEHAN abandoned dwelling-sites (TABLE II, p. 124). The majority of the sites appear to be homesteads with one building, as Dinna Clerks, Widecombe-in-the-Moor (FIG. 48, a), Blackslade, Widecombe-in-the-Moor (FIG. 48, b), Blackalder, Shaugh Prior (FIG. 48, c) and Shavercombe, Shaugh Prior (FIG. 48, h); two buildings, as Manaton Rocks, Manaton (FIG. 48, d) and Cold East, Ashburton (FIG. 48, e); or three buildings, as the site near Leapra Farm, North Bovey (FIG. 48,]) and Holne Ridge, Holne (FIG. 48, g), although settlements of up to 13 buildings have been found, as Blackaton, Widecombe-in-the-Moor (FIG. 49).3 The long-houses vary from 18 to 80 ft. in length, and from 9 to 15ft. in width; they usually lie with the long axis at right angles to the contour, and have a cross-passage with opposed entrances towards the lower end. Occasionally, however, they are built with the long axis along the contour, with an entrance on one side only, as, for example, Hawns (FIG. 57, c) and Parkland Newtake (FIG. 57,]) in Cornwood. Nearly all the larger houses, and at least one house in most of the settlements, have a small room at the upper end. Pairs of houses have also been noticed, lying within a few feet of each other and of much the same dimensions, as in Okehampton Park, site 53 A (FIG. 53, d) and site 56 (FIG. 53, c), and also at Yar Tor, Widecombe-in-the-Moor (FIG. 55, a). , whose Guide to Dartmoor, first published in 1909, remains the most accurate survey available, notes several unoccupied farmhouses, and also mentions the ruins of rectangular buildings, sometimes in connexion with tinning works in the valley bottoms. He also warns his readers against confusing rabbit burys with prehistoric barrows, but he dearly did not consider later features of more than passing interest. R. Hansford Worth, who died in 1950, was another man with an intimate knowledge of Dartmoor. He described some of the early stone houses, and noted a number still occupied or only recently abandoned. In his collected works plans are given of cross-passage long-houses of early type; he also gives photographs and details of broad and narrow quoins in ruined buildings at Challacombe (Manaton),' and Yardworthy (Chagford), dating this type of stone work 'not later than 1600'. Worth was also interested in the blowing-houses and their contents; and he gives careful measurements and details of three on the Walkham river above Merrivale bridge. He failed, however, to record a long-house site only a few yards from the most northerly blowing-house on the E. bank (FIG. 50, a), or a number of pillow-mounds near by.'

I. DESERTED SITES (TABLE II; PLS. IV-VI; FIGS. 48-59) THE EVIDENCE FROM DOMESDAY A study of Domesday Book (1086) shows 37 manors with their land lying mostly on the moor (TABLE I, p. 117). Ofthese no less than 23 lie on the eastern

3 H. French and C. D. Linehan, 'Abandoned medieval sites in Widecombe-in-the-Moor', Trans. Devon Assoc., xcv (1963), 168. 4 For this site see Appendix, p. 143 f. I R. Hansford Worth, Dartmoor (1953), pp. 309,403. DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 117 or north-eastern side; 10 on the southern slopes, and only two each in the west and north. Three have no entry for villeins, but the other 34 manors together contain 158 villeins, and if we add to these the 37 demesne holdings, a total of 195 holdings is obtained. There are 127 ploughlands recorded, though only 87

TABLE I. DOMESDAY MANORS ON DARTMOOR (from the northern side clockwise around the Moor)

Plough- Manor D.B.Name Parish V.C.H. p. lands Ploughs Villeins

Belstone Bellestam 448 6 5 8 Throwleigh Trula Throwleigh 519 8 8 10 Ghiderleia Gidleigh 436 I I Shapleigh Escapeleia Chagford 453 2 2 4 Shapleigh Escapeleia Chagford 453 3 2' Teigncombe Taincoma Chagford 425 4 2t 6 North Bovey Bovi North Bovey 469 8 6 II Beetor Begatora North Bovey 453 4 3 4 Shapleigh Essapla North Bovey 532 2 2 3 Manaton Manitona Manaton 532 2 2 3 Little Manaton Magnetona Manaton 465 I 1 3 Houndtor Hundatora Manaton 432 4 2 2 Langstone Langhestan Manaton 464 2 2 4 Challacombe (see App. p. I43f.) Manaton 525 I knight Natsworthy Noteswrde Widecombe-in-the-Moor 534 2 2 2 Blackslade Blacheude Widecombe-in-the-Moor 533' 3 2 3 Dunstone Dunestantona Widecombe-in-the-Moor 5333 3 3 2 Dewdon Depdona Widecombe-in-the-Moor 490 Scobitor Scabatora Widecombe-in-the-Moor 428 2 thanes Spitchwick Espicewita Widecombe-in-the-Moor 4II 8 4 8 Buckland-in-the-Moor Bochelanda Buckland-in-the-Moor 525 4 2t 8 Bagathora Ilsington 524 5 5 6 Holne Holla Holne 490 12 6 13 Stoke Estocha Holne 490 2 I 4 Skerraton Siredona Dean Prior 533 2 2 3 Great Aish Brenta South Brent 433 6 3t 8 Peek Pech Ugborough 443 2 1 3 Cornwood Cornehuda Cornwood 443 5 3t 8 Harford Hereforda Harford 443 5 3 6 Baccamore Bachemora St. Mary 477 4 3 4 Loveton Bastard Metwi Meavy 534 2 2 5 Meavy Mewi Meavy 475 2 1 24 Goodameavy Mewi Meavy 475 2 I 3 Church Meavy Mewi Meavy 475 4 It 4 Hoo Meavy Mewi Meavy 476 I t 1 Cudlipptown (seep. 137) Tavistock (now PeterTavy) 429 2 2 25 Willsworthy Wifleurde Peter Tavy 534 4 I

r Text mentions villeins without specifying number, so at least two have been assumed. , Place-Names of Devon (English Place-Name Soc., II, 1931-2), p. 526. 3 Ibid., p. 527. 4 See note I. 5 Ibid.

ploughs were in use at the time of the survey. The excess of ploughlands over plough-teams strengthens the probability of a considerable outfield, where parts of the land were broken up for cultivation and then abandoned again, while a different area was ploughed. The area of a Devon ploughland is not known, but 60 to 80 acres has been suggested." 6 V.C.H. Devonshire, 1(1906),386. 118 CATHERINE D. LINEHAN

a b

HfDGE SANk,.

LYNCHEr

-, -, -, ", -, " -, " -, " -, " -,' -, -, ", -, , ;>.\ '\1'\ \ 1" , 'e:t-', , \ \ \ \\ \ \ \ \ II I I a II I I I I I I h I I I I WUfAL EMMA LEAr I J j

FIG. 48 DESERTED SITES ON DARTMOOR (p. 1I6) a, Dinna Clerks (Widecombe-in-the-Moor); b, Blackslade (Widecombe-in-the-Moor); c, Blackalder (Shaugh Prior); d, Manaton Rocks (Manaton); e, Cold East (Ashburton); j, near Leapra Farm (North Bovey); g, Holne Ridge (Holne); h, Shavercombe (Shaugh Prior) DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 119 To the I95 holdings must be added an unknown number of moorland hold­ ings belonging to manors which lie farther from the moor but which had some land upon it. Such is Lydford, which included all the of Dartmoor, now the Duchy of Cornwall lands, though no mention is made in the Domesday record of any moorland tenements. No deserted sites of early type have been found in the forest area, and it seems possible that Piswell and Babeny (still occupied farms) were the earliest of the ancient tenements allowed in the royal hunting ground. They are first recorded in I260 in the Episcopal Registers. Shaugh Prior, Walkhampton, Sampford Spiney, Whitechurch, and all have common land on the moor, much of which shows signs of past enclosure. Okehampton had 3I villeins and land for 30 ploughs, and must have had holdings on the slopes of Dartmoor, where seven deserted sites have been found (FIG. 52). South Tawton was a large royal manor which included South Zeal, and its holdings of common lands must have been held by some of the 50 villeins recorded. Because of the uncertainties in the Domesday record, it is not possible to calculate exactly the number of holdings or the amount of land under cultivation on Dartmoor at this early period. It is clear, however, from the figures obtained that there were over 200 holdings. Some of these are, of course, represented by farms still in occupation and several remain almost unchanged since the Domesday record was made, as, for example, Stoke (Estocha in Holne), where the record lists the demesne land and four villein holdings, and where five Stoke farms can still be distinguished. W. G. Hoskins suggests that much of the outfield may have been in occupa­ tion before the time of the earliest charters, which date from the I3th century. He gives as example the charter for Cholwich Town in Cornwood parish (FIG. 56), in which parish he finds almost every farm occupied by a free tenant from the I3th century. The charter, dated between I200 and I230, mentions 'all my land of Cholleswyht', showing it as already named and suggesting earlier occupa­ tion/ This early settlement is corroborated at Houndtor village (Manaton) and another near-by site, Houndtor II (FIG. 50, b), and also at Hut Holes, Widecombe-in-the-Moor (FIG. 50, c), where excavation of stone long-houses of the late r zth to the I3th century has revealed the existence of earlier turf houses, and indicated a pre-Conquest date for the beginning of the sites."

FIELD EVIDENCE OF ENCLOSURE AND HOMESTEADS Field-work shows that much of the venville commons, as also the land in the central plains and in the wider river valleys, has been enclosed; only the highest moorland, the rocky hillsides, and the most precipitous wooded valleys remain unenclosed. Superimposed field-systems and boundary-banks are difficult to interpret. Enclosure has taken place at different periods and for different purposes. Remains may be prehistoric, medieval or post-medieval; they may be connected with farmsteads, with tinner's bounds, or with a combination of the 7w. G. Hoskins and H. P. R. Finberg, Devonshire Studies(1952), pp. 316-318, 321-323. 8 Med. Archaeol., VI-VII (1962-3),341 f., pI. xxxii, B, fig. 102; VIII (1964),282 f., figs.90, 91 ;IX (1965), z rof 120 CATHERINE D. LINEHAN

CROFT

LYNCH IT

LYNCHET

= ROAD ATE :: :, ROAD n l~-_..J)

LYNCHEr

FIG. 49 DESERTED SETTLEMENT AT BLACKATON (WIDECOMBE-IN-THE-MOOR), DARTMOOR (p. Il6) two, as for example Ringmoor Down, Sheepstor (FIG. 50, d), where the gully which runs through the site was possibly connected with near-by mining activities. Grants for tinner's rights were sometimes made to local farmers, as is shown in a document of 1759, granting to Roger Hannaford of Dockwell in Widecombe-in­ the-Moor and Robert Mann of the rights in Dockwell in the manor of Dewdon (Widecombe-in-the-Moor). These bounds were 'cut and pitched' in the presence of witnesses, implying the making of a bank or reave.?

9 Grant in the possession of Mr. and Mrs. H. French of Watergate, near Widecombe-in-the-Moor. DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 121

a

.If'" I

d

FIG. 50 DESERTED SITES ON DARTMOOR (pp. I 16, I 19) a, Merrivale warren (Walkhampton); b, Houndtor, site II (Manaton); c, Hut Holes (Widecombe-in­ the-Moor); d, Ringmoor Down (Sheepstor)

Many of the old enclosures have been abandoned to heather, gorse, bilberry and the ever-encroaching bracken; the bracken, often 5 ft. in height, makes field­ work almost impossible from May to December, except on parts of the west side of the moor, where there is some open grass land. In winter, storms and snow are further hazards. Time and patience are needed to trace these signs of past habitation. Evening or early morning light with a view across a valley may show 122 CATHERINE D. LINEHAN up previously undetected enclosures; then a careful study can sometimes locate the smaller enclosures, crofts and yards which usually surround a dwelling house. Aerial photographs and large-scale maps may also help to identify a group of enclosures, usually about 20 to 30 acres, which contain yet another forgotten site. The map may also indicate an unoccupied house in the middle of nowhere, or some apparently incomplete field banks, whose ruins were sufficiently up­ standing to warrant notice by the surveyors, although other buildings and en­ closures close by were not recorded. 'Corn-ditches' can also act as a guide to deserted sites, even though long abandoned. These boundary-walls between cultivated fields and moorland were reinforced by a ditch on the outer or moorland side, to prevent animals, including deer in earlier times, from invading the corn; in the field itself, stones removed to clear and improve the land gradually formed a sloping bank against the inside ofthe wall. Field-work has been most intensive in Widecombe-in-the-Moor and Manaton parishes, where, in consequence, more sites have been found. The Domesday evidence, however, suggests that there were more holdings on the south, east and north-east of Dartmoor, where the rainfall is considerably less than on the west side, and where the hills give protection from the prevailing south-westerly storms.

THE PROCESS OF DESERTION From the bronze-age and iron-age settlements until the continuing drift from the land at the present day many economic factors must have contributed to the fluctuating process of expansion and contraction of population on Dartmoor. Most of these factors remain obscure; others, such as the increase in sheep-farming and the wool trade in medieval times, are well known. The height at which some of the deserted sites lie is surprising, and perhaps was itself a cause for early abandonment. A long-house near Berry Pound, Natsworthy (Widecombe-in-the-Moor) is at 1,350 ft.; that at Dean Moor (Dean Prior), now under the waters ofthe Avon reservoir, is at 1,110 ft.;" Butterberry I (Peter Tavy) is at 1,225 ft., and the Okehampton Park sites lie between 875 and 1,150 ft. Excepting the Dean Moor site, these all show some of the surrounding areas to have been under the plough. At Hen Tor warren, Shaugh Prior (FIG. 51, c), which lies at 1,130 ft., a recollection of 10 ploughing oxen in the r Sth century is recorded by Crossing." The frequent proximity of medieval sites to prehistoric settlement, as at Whitten Knowles, Sheepstor (FIG. 51, a), raises the question of continuity of occupation, although such continuity remains to be proved. The land would obviously be favoured as being already cleared and cultivated; indeed some small rectangular buildings are actually inside prehistoric pounds and enclosures. This can be seen at Willings Walls (Shaugh Prior), and on Holne Moor (Holne) near the reservoir. Medieval enclosures and terracing have in places

10 This site was excavated in 1956. See A. Fox, 'A monastic homestead on Dean Moor, S. Devon', Med. Archaeol., II (1958),141-157.

II W. Crossing, Guide toDartmoor (1909), p. 432. DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 123

HUT -_, ClltCLF.-f )\ " .::=-/ t I

'-JOSSIBlE ItE'USfP o HUT CIRCLE nu _ ~ ~ LE"'T ENCLOSURE WALL LEAT

TERIl....cr -I~) :!OTHER TERRACES"t90ft.£ ~ /1 ,

EN' 0 ItE W",ll

c

FIG. 51 DESERTED SITES ON DARTMOOR (pp. 122, 124) a, Whitten Knowles (Sheepstor); b, Cripdon (Manaton); c, Hen Tor (Shaugh Prior) 124 CATHERINE D. LINEHAN been superimposed on prehistoric field-systems, as for example on Hound Tor (Manaton), and on the slopes of Vaghill (Widecombe-in-the-Moor) overlooking the . In the latter there are several prehistoric enclosure-groups, with their attendant hut-circles, in two ofwhich the field-walls have been reconstructed; no long-house has been found, which suggests the possibility that the huts were reoccupied, or that turfhouses were in use whose remains have totally disappeared. Medieval pottery has been found in many of the hut-circles excavated in the past. Pottery found on the medieval sites mentioned on p. 119 suggests their abandon­ ment in the late 13th century, but the causes for these early desertions of farms and villages remain obscure. An increase in the population on Dartmoor occurred in the rSth and 19th centuries, when a number of quarries and mines were opened. or reopened in various parts. Charles Vancouver expressed his belief that almost the whole of

Dartmoor could be reclaimed by drainage, burning, liming and other means.IO His optimistic dream was not realized. Sir Thomas Tyrwhitt, when Lord Warden ofthe Stannaries, started large reclamation projects at the end ofthe rSth century round his granite quarries near what is now Princetowri." Because of the employ­ ment available, workers were attracted to the district, and a number of dwellings were built at this time. They are of a most primitive type, often little more than a hut some 15 by 20 ft., at most a two-roomed cottage; now only the tumble ofa chimney-stack marks them as later erections. With a small garden and a few very small enclosures, they are usually near mining-sites or quarries, although they can easily be mistaken for medieval sites. Indeed they may lie on older sites, and have, therefore, been included in the list. Examples of these are the Cottage in Swincombe (Lydford) and Ridge Ley (Chagford). Their abandonment followed the decay of the mines and quarries at the end of the rSth and beginning of the 19th centuries. Larger farms were also enclosed during this period. Foxtor Farm at 1,235 ft., Nun's Cross Farm at 1,312 ft., Teignhead Farm at 1,475 ft., and Sheepfold at 1,315 ft., all on Duchy lands, are examples. These farms were occupied for only about a hundred years, conditions no doubt proving too harsh and unproductive at these heights. At a still later date desertions have followed the construction of reservoirs. At Fernworthy and all the farms in the catchment areas have been left to crumble to ruins, or have been submerged. A small acreage of land is still being farmed, but most of the surrounding area is being used for forestry or has reverted to rough pasture. The amalgamation of holdings and consequent desertion of farmhouses, as at Cripdon, Manaton (FIG. 51, b), continues apace, and can be seen as yet another retreat from the least productive land. On all sides of the moor, but especially on the west, many houses stand empty, while the land is farmed from adjacent holdings as sheep-runs and cattle-ranches. With the hedge-banks down and the animals grazing at will, the deserted homesteads present a melancholy appearance

r C. Vancouver, Agricultural Surv~ of Devonshire (1808), pp. 284-285. ') W. G. Hoskins, Devon (1954), p. 429. TABLE II. DESERTED SITES X: Name unknown or uncertain LD: Building showing late development IV: Enclosures in use on present day farm

Site Parish Site name O.S. Nat. Grid Feet No. of LD Enclosures R.A.F. air-photographs no. I in. Ref. SX O.D. buildings

I Ashburton Cold East (X) 187 743739 1075 2 - yes 58/4424.1"44.0192.25/5/61 I I IV 2 " Rushlade 187 7547 26 635 58/4424.1"44.0194.25/5/61 3 Bovey Tracey Haytor Down (X) 187 782777 850 I yes 58/4424.1"41.01°4.25/5/61 4 Buckfastleigh Lambsdown 187 697 660 1050 2 yes 58/1110.0044.5/5/53 Scorriton Down (X) 18 68 0 5 " 7 3685 95 4 4 yes 58/1110.0023,5/5/53 6 Buckland-in-the-Moor Higher Pudsham 187 718743 9 15 I I IV 58/4424.1"43.0188.25/5/61 7 Pudsham Down (X) 187 734747 1200 I yes 58/4424.1"43.019°.25/5/61 8 " Stone 187 721747 9°0 3 3 IV 58/4424.1"43.0188.25/5/61 9 Chagford Ridge Ley 175 6828 37 1100 I I yes 58/4424.1"44.118.25/5/61 I 10 " Stiniel Cottage 175 706855 850 (in hamlet) 58/4424.1"43.62.25/5/61 II Comwood Ford Barn 187 609618 750 7 3 yes 58/1110.0052.5/5/53 12 Ford Waste 187 607616 825 3 - yes 58/1110.0052.5/5/53 13 Harrowthorn 187 622616 800 3 - yes 58/1110.0052.5/5/53 14 Hawns (X) 187 613626 875 2 - yes 58/1 I 10.0040.5/5/53 15 High House 187 612625 900 4 3 yes 58/1110.0040.5/5/53 16 " Parkland Newtake (X) 187 597 619 9°0 I yes 58/1110.0040.5/5/53 17 Dean Prior Dean Moor (X) 187 679654 1110 2 yes 58/1110.0042.5/5/53 18 Harford Lower Piles (X) 187 6436 10 800 4 yes 58/1110.0052.5/5/53 19 Holne Combestone Wood (X) 187 6737 24 850 I I yes 53/4424.1"44.181.25/5/61 20 Holne Moor (X) 187 683707 1000 I- yes 540/1040.0125.10/4/53 21 " Holne Ridge (X) 187 681688 1075 3 yes 58/1110.0023.5/5/53 22 Ilsington Holwell Tor (X) 187 75 1774 125° 2 ? yes 58/4424.1"41.0 I 08.25/5/61 23 Mountsland 187 75974° 75° I I IV 58/4424.1"43.0194.25/5/61 24 " Pinchaford Down (X) 187 769765 39° 2 yes 58/4424.1"42.0196.25/5/61 25 Lydford Assycombe 175 659827 1320 4 4 yes 58/4424.1"44.120.25/5/61 26 Brimpts, Dolly's Cott 187 673744 850 I I yes 58/4424.1"43.183.25/5/61 27 Brownberry 187 647744 1010 5 I yes 58/4424-1"43.179.25/5/61 28 187 653774 1100 3 3 IV 58/4424.1"41.121.25/5/61 29 Broom Park 187 644739 975 2 2 IV 58/4424.1"43.179.25/5/616547 30 Cocks Lake 187 654743 1055 I I yes 58/4424.1"43.179.25/5/616547 3 1 Hexworthy 187 6557 28 965 3 3 IV 53/4424.1"44.179.25/5/61 32 Swincombe, Bishop's House 187 64 2726 1012 I I yes 58/4424.1"44.179.25/5/6 I 33 Swincombe Cottage (X) 187 6437 25 1035 I I yes 58/4424.1"44.179.25/5/61 34 " Whiteslade 187 662 764 1000 3 3 yes 58/4424.1"42.181.25/5/61 35 Manaton Beckham, West 175 762802 725 I I yes 58/4424.1"42.0 I 06.25/5/6 I 36 Black Hill (X) 187 758793 1075 I yes' 58/4424.1"42.0108.25/5/61 37 Blissmoor, Hayne Down (X) 175 736804 1135 5 yes 58/4424.1"43.0 I 10.25/5/6 I 38 Challacombe I 187 694796 1100 12 9 IV 58/4424.1"42.1 17.25/5/6 1 39 Challacombe II 187 695796 115° I- IV 58/4424.1"42.1 17.25/5/61 40 Cripdon 175 7348 10 1000 4 4 IV 58/4424.1"43.0 I 10.25/5/61 41 Great Houndtor Farm 187 749795 875 I (nowa barn) IV 58/4424.1"42.0108.25/5/61 42 Hayne 175 748806 925 6 5 IV 58/4424.1"43.0108.25/5/61 43 Holwell 187 745777 1075 3 yes 58/4424.1"41.0110.25/5/6 1 44 Holwell Cottage 187 746775 105° 2 2 yes 58/4424.1"41.0 I 08.25/5/61 45 Holwell Lawns 187 743779 1125 3 3 yes 58/4424.1"41.0110.25/5/61 46 Houndtor I 187 747787 1100 II yes 53/4424.1"42.0108.25/5/61 47 Houndtor II (X) 187 745789 1100 6 yes 58/4424.1"42.0 I 08.25/5/61 48 Manaton Rocks (X) 175 7508 17 900 2 yes 58/4424.1"43.0 I 08.25/5/6 I 49 Me;{vy Higher Cadworthy 187 549645 7°0 2 IV 58/1 I 10.0033.5/5/53 50 North Bovey Lower Hookney 175 714825 95° I I IV 53/4424.1"44.114.25/5/61 Near Leapra Farm (X) 0 1827 yes 58/4424.1"44. 116.25/5/6 1 5 I"" 175 7 119° 3 52 Okehampton Okehampton Park Chapel (X) 175 583933 105° I yes 58/4424.1"44.33.25/5/61 53 Okeharnpton Park I A, B, C (X) 175 536934 1°5°-1100 II yes 58/4424.1"44.33.25/5/61 54 Okehampton Park II (X) 175 58493 2 1100 3 yes 58/4424-1"44.33.25/5/61 55 Okehampton Park III (X) 175 580929 1100 6 yes 58/4424.1"44.33.25/5/61 56 Okehampton Park IV (X) 175 5779 27 1125 2 yes 58/4424.1"44.33.25/5/61 57 Okehampton Park V (X) 175 5749 25 1°5° 2 2 yes 58/4424.1"44.33.25/5/61 58 Okehampton Park VI (X) 175 5759 23 115° 4 yes 58/4424.1"44.33.25/5/61 59 " Okehampton Park VII (X) 175 573930 95° 7 yes 58/44.24.1"44-33.25/5/61 60 Peter Tavy Butterberry I 187 549794 1225 6 2 yes 58/4424.1"42.136.25/5/61 61 Butterberry II 187 547797 1125 3 2 yes 58/4424. F42. I 36.25/5/61 62 Redford 175 537834 1085 4 4 yes 53/4424. "'44. I 38.25/5/6 I 63 Wapsworthy 175 537802 855 I I IV 58/4424.1"42. I 38.25/5/61 64 White Hill (X) 175 534832 1025 2 yes 58/4424.F44. I 38.25/5/61 65 Sa~pford Spiney Combe Hill 187 5257 19 59° 3 3 yes 58/4424.1"44. 163.25/5/61 66 Shaugh Prior Blackalder (X) 187 5656 19 75° I I yes 58/1 I 1O.0054.5/5/5~{ 67 Hen Tor 187 590656 II30 5 3 yes 58/1110.0040.5/5/53 68 Shavcrcombe (X) 187 592664 1060 I yes 58/1110.0040.5/5/53 69" " Willings Walls (X) 187 579 658 95° I yes 58/1 I 10.0038.5/5/53 70 Sheepstor Legis Tor (X) 187 570658 94° 3 yes 53/1 I 10.0038.5/553 71 Outcombe 187 581686 875 2 2 yes 58.1 I 10.0029,5/5/53 72 Ringmoor Down (X) 187 568662 935 4 yes 58/1 I 10.0038.5/5/53 73 Sheepstor Warren (X) 187 564680 1075 3 3 yes 58/r I 10.0029.5/5/53 74 " Whitten Knowles (X) 187 586670 1°75 3 yes 53/1 I 10.0038.5/5/53 75 South Brent Corringdon Newlands Brake (X) 187 667 607 875 2 2 yes 58/1 I 10.0050.5/5/53 76 Gingaford 187 699623 75° I IV 58! I I lu.u

SUPPLEMENTARY LIST Site Parish Site name O.S. Nat. Grid Feet No. of LD Enclosures R.A.F. air-photographs no. I in. Ref. SX O.D. buildings

104 Shaugh Prior Shavercombe Foot (X) 187 593666 1025 yes 58/ 1110.0040.5/5/53 105 Spanish Lake (X) 187 583647 1100 yes 58/1110.0038.515/53 106 " Trowlesworthy Tor (X) 187 577636 95° yes 58/ 1I 10.0038'5/5/53 107 Sheepstor Gutter Tor (X) 187 578667 1075 5 yes 58/1 I 10.0029.5/5/53 r08 " Legis Lake (X) 187 567658 850 I yes 58/1 I 10.0029.5/5/53 109 Walkhampton Newleyeombe (X) 187 5877 00 135° 3 3 yes 540/1040/0129.10/4/53 110 Riddipit (X) 187 57 17°2 875 2 I yes 540/1040/0129.10/4/53

TABLE III. RABBIT-WARRENS Nat. Grid Pillow­ Site Parish Warren O.S. Ref. SX Feet mounds R.A.F. air-photographs no. I in. (approx.) O.D. counted

W.I Lydford Beardown (PL. VII, B) 187 600 755 13°0 25 58/4424.1"42. I 73.25/5/61 W.2 Huntingdon 187 665669 1325 20 58/1110.0043.5/5/53 W·3 Wistman's Wood 187 613775 1425 12 58/4424.1"41.127.25/5/61 W·4 No;th Bovey Headland 175 69 28 14 14°° 3 58/4424.1"43.117.25/5/61 W·5 Shaugh Prior Hen Tor 187 590656 113° 9 58/ 1110.0°4°,5/5/53 W.6 Trowlesworthy 187 567647 835 18 58/1110.0038.5/5/53 W·7 Willings Walls 187 582657 960 I 58/1110.0038'5/5/53 W.8 She~pstor" Ditsworthy 187 584663 1000 21 58/1110.0038.5/5/53 W·9 Legis Tor 187 574653 9°0 37 58/1110.0038.5/5/53 W.IO Sheepstor 187 565682 1125 6 58/1110.0029.5/5/53 W.II Sou'th Brent Yalland 187 688632 9°0 6 58/1110.0044.5/5/53 W.12 Zeal Burrows 187 6806 34 1000 4 58/1110.0°44,5/5/53 W.13 So;;th T;~ton Skaigh 175 630935 1000 17 (in 2 groups) 58/4424.1"44.25.25/5/61 W.14 Walkhampton Merrivale 187 554755 1100 12 58/4424.1"42.165.25/5/61 W.15 Widecombe-in-the-Moor Vaghill and Eastern Combe 187 680 725 1000 23 58/4424.1"44. 183.25/5/61 W.16 Vgborough Redlake Tramway 187 654594 1100-1500 6 58/1110.0050.5/5/53

[face page 124 DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 12.') of pending decay. Such farms, although often of medieval origin, mostly have comparatively modern houses; they are clearly marked on the O.S. maps, and have not, therefore, been included in the list. A reasonable living can no longer be made from the small acreage and poor soil of the average Dartmoor farm. For the time being this seems to be the inevit­ able trend, and families which have lived for centuries on the moor may soon be gone, as the younger generation moves away to an easier life. Further work is needed to record the many deserted sites which must still await notice, and work is also needed on the ruined long-houses on existing farms, some of which are still in use as barns or outhouses. A few of these have been listed in this survey, but many farms, particularly in the east and south, have, either in use or in ruins, buildings which clearly represent an older dwelling. Unfortunately they are being swept away at accelerating speed, and time is short in which to record them.

DESERTED SITES IN OKEHAMPTON PARK (PLS. IV, A, B, v, A, B; FIGS. 52, 53, a-j) Okehampton lies at the junction of the East and West Okement rivers, with its castle about half a mile upstream along the latter. The earliest record of the name is in a missal of about 970 where ocmundtune is mentioned as one of the places for the manumission of slaves." The castle, mentioned in Domesday Book together with four burgesses, a market and a mill, is said to have been first built by Baldwin de Brion soon after the Conquest." The park, in which lie the deserted sites, is to the south and south-west of the town and castle, occupying the slopes of the moor between the two Okement rivers. It is bounded on the south by an area of common land belonging to Okehampton, beyond which stretches the forest of Dartmoor. It is not known when the park was formed; it is mentioned in 13th-century charters and was disafforested by Henry VIII at the same time as the castle was demolished, after Henry Courtney, marquis of Exeter, was attainted and beheaded. The medieval manorial records of the manor and the barony of Okehampton, which might have thrown light on the early history of the park, have been lost in the vicissitudes of the Courtney and other later families who held the manor and barony. In the south-west of the park, overlooking the castle and town, seven sites have been found, containing together 35 long-houses; they are all within a mile of a spot marked 'Chapel' on the O.S. map, where there are enclosure-walls and the remains of a building. The small occupied homestead near by is called 'Saxon Gate', a name for which no early documentation has been found, but which may be of significance. The sites are now in the manor of Meldon, a small manor in Okehampton of uncertain origin, recorded from the 13th century.1& An indication of the early occupation of the area is its ancient road-system. The main track, suggested, and discarded, by writers as a Roman road, comes

'4 W. de G. Birch, Cartularium Saxonicum, no. 1245. '5 Op,cit. in note 13, p. 447. ,6 Information from Mr. R. L. Taverner ofOkehampton. --:::::::==::::- ---- IV O'l Okehampton )~~I§ to FATHERFORD Castle -,,\\' / / I

East Hill

~\.i

(:-, (Ce o",e -:f.e,,' o l' > elY' >-3 ::r: M I ....~ Z M tJ ....r- Z M ::r: ~"- z> I' Black Down " "II Rough Tor

a 1000 3000 51XXJ 'eet 8M ••• 1 !! I

r - ~O Idoo ISba metres

FIG. 52 MAP SHOWING POSITIONS OF DESERTED SITES IN OKEHAMPTON PARK, DARTMOOR (pp. 125, 127 f.) Based on the 6-in. O.S. map with the sanctionof the Controller ofH.M. Stationery Office. Crown copyright reserved DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 127 up the slope diagonally from Fatherford on the East Okement river in the north­ east corner of the park, and runs directly south-west through the 'Chapel' en­ closures towards Meldon; it ends abruptly where the Meldon quarry works have destroyed it. This main way is now blocked in places by massive rqth-century hedge-banks, which divide the area into large fields. Branch tracks lead from this way into the valley from site 52, the 'Chapel', and to site 59 to the west. An upper, south, branch of the old road from Fatherford follows the present road for a short distance by Fitzwell and then, branching from the tracks to Halstock, the army camp and the moor, runs approximately parallel to the lower track. It passes through sites 53, 54 and 55, but is lost in places. It becomes very clear between sites 55 and 56, and continues to site 57 as a sunken grass-covered way, until it, too, ends abruptly over the edge of the Meldon quarry. Several other tracks, also sunken in parts and grass-covered, lead from this route to the common lands above. The two main tracks are shown in parts on the 6-in. O.S. map. A continuation of these ancient roads to Meldon and Tavistock passed Higher Bowden and Prewley Farms (where a good track still exists) over the shoulder of the moor, where the way is known today in one part as Kingsway or King­ wall, and is so marked on the O.S. map." There are a number of springs in the park close to and below the sites. One, beside the road from the town, is known as FitzwelI. Some authorities mention a legend that the stone cross standing beside it was brought from Halstock Chapel, which lay about one mile to the east." Halstock, now only a single farm, was once a considerable village, where six tenements were recorded in 1586. The chapel of St. Michael of Halgestoke is named in the account of a perambulation of made in 1240.'9 The sites in the park were noticed by the Rev. H. G. Fothergill. After describing the hill-fort on the east side of the park, which lies on the spur of East Hill between the East Okement river and its tributary, the Moorland brook, he mentions 'the numerous traces of houses, near any spring of water, also of roads and enclosed patches ofland, to be seen in the Park-that lead me to imagine that this was the site of the original colony, that peopled, in later and more civilized times, the old town, now situated near the source of the Okement, previous to the Norman Conquest, when this spot was converted into a Park, being well stocked with wood.?? WimboIt and Ward mention a 'Saxon village and chapel in the east side of the Park'." As no ruins have been found in the east, these authors probably refer to the sites under discussion. There is a hint that the park was extended over these sites at the time of a charter dated between 1216 and 1272, by which the lord of the manor confirmed

17 W. Crossing, op, cit. in note II, p. 59, and id., A Hundred Years on Dartmoor (1902), p. 30. 18 Devonshire Association: Parochial Histories, Okehampton, p. 15. 19 Ibid., p. 60. 20 Cited in W. H. K. Wright, History of Okehampton (republished 1889), p. 179. ar Wimbolt and Ward, Pocket Guide toDevon (1929), p. 93. 128 CATHERINE D. LINEHAN the rights of the burgesses of Okehampton. This charter was transcribed by Richard Shebbeare ofOkehampton in 1671. He adds: 'I gather by this Charter of the Borough that at the tyme of this graunt, and at all tymes before, either the Lord of Okehampton had noe park within the lordship, or else after this graunt hee did much enlarge his parke; for all that wood and soyle on the south part of the Castle and borough, wherein by the instructions hereafter issuing they did release their comons of pasture now inclosed in the parke, which all that time it should seeme, lay open and in comon. Mr. Humberston's opinion as presented by Shebbeare.':" If the lord of the manor extended his park at this time over the sites of the deserted homesteads, as well as over the common lands of the burgesses in this area, presumably the settlement was already abandoned, although the process may have been assisted. Once the castle, mill and borough had been established in the valley, a transference of population might well have taken place from the old hillside settlement as the lowland woods were thinned and cultivation extended on more fertile land. Fothergill suggests, indeed, that this transference took place from the deserted sites in the park, which he regarded as the original colony that peopled the town, now called Okehampton. Ifthis large scattered settlement, with 35 houses and possibly a chapel to be distinguished at the present day, represents the original Okehampton, the growth of the new town in the valley beneath might well have taken the name as well as the population, thus account­ ing for the fact that the deserted sites have no recorded name. This suggestion of Fothergill's is, however, contradicted by local historians, who consider that the Saxon Okehampton existed on the opposite side of the valley around the present church, which stands in isolation on rising ground halfa mile to the west ofthe town. The church is first mentioned in 1239, together with the chapel of the castle, in the appropriation of both by Bishop William Briwer to the monks of Cowick. No early font or Saxon features exist in the present church, which was rebuilt in 1843-4 after a fire; only the tower remains of the rebuilding of 1447. The existence of the iron-age hill-fort and the large number of long-houses and enclosures on the opposite side of the valley suggests that the site of one Saxon Okehampton lies along the edge of the uplands south of the present town. If, however, the tradition of a Saxon settlement around the church is correct, then the existence of another Okehampton is indicated; a North and South, or Higher and Lower Okehampton, perhaps. As the visible remains of the deserted sites are of stone, occupation in the 13th century, at least, is probable. The post-Conquest migration from the church area to the present town site, nearer to the castle and Okement river, would have preceded that of the scattered upland homesteads, which lingered on into medieval times.

Site 52, the Chapel (PL. IV, A; FIG. 53, a). The remains of a building lie within sub­ stantial enclosure-banks to the W. of site 53 and N. of site 54. The W. and S. sides of the ruins are the most upstanding, each about 22 ft. long by 19 ft. wide; there are several dressed stones lying within the ruins, which appear to be parts of a window or door. " Cited by Wright, op.cit. in note 20, p. 160 (note). DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 129

~Nk. ======!-I) o--D------.---1 ~====='=_:'O===~ TRACK TIl",,, nu nu a b

I .Iln n nnn vr u uuuU -J,- 0 c LYNCHEr \HU I c8) d .Ai' ~ 8

, I , " ""I I / m oJf? ! oJ.-- MELt'ON ~' J~,~u,lTi nQU"RRY / II 1,1 .,r U .y f 9 D DrFTI U Li nA n UU I~!! HEpGE B4Nk.

FIG. 53 DESERTED SITES IN OKEHAMPTON PARK, DARTMOOR (pp. 128 fr.) a, Chapel, site 52; b, site 53 B; c, site 56; d, site 53 A; e, site 53 C;f, site 57; g, site 54; h, site 55; i, site 58;j, site 59 The tithe map shows the main enclosure and marks it as 'Hospital'; no records have been found to indicate the history or purpose ofthe site. Site 53 is the largest, both in area and in the number of long-houses it contains. Eleven have been counted in three groups, and others may remain unnoticed in the CATHERINE D. LINEHAN uneven ground and the confused remains of enclosure-walls half hidden in the bracken which covers most ofthe settlement. The first group, site 53 A (PL. IV, A; FIG. 53, d), can be found close to the N. fence of the army camp. Two houses, one much ruined, lie side by side, close to one of the circular enclosures or pounds which are a feature of the Okehampton Park home­ steads. The second group, site 53 B (PL. IV, A; FIG. 53, b), a little to the N. and cut by the 19th-century field-bank, consists of four long-houses. The largest, over 72 ft. long, lies on the W. side of the bank. It has a cross-passage and small room at the upper end. Near by is one of the small rounded enclosures, and also other enclosure-walls. The S. track running through the park from Fatherford to Meldon passes this group of houses a short distance to the S., and two more houses can be seen on the N. side of the track, E. of the others and some distance apart; they can barely be distinguished in the under­ growth, as the remains are slight. The third group, site 53 C (PL. IV, A; FIG. 53, e), is situated on the brow of the hill, N. of the track. There are three buildings lying parallel to one another, and close together. A good spring, which would supply water for all the homesteads, rises a little below group B. A side track from the main through way indicates the old track on to the commons. The buildings ofthe army camp may cover the remains ofother medieval sites; indeed a farm is said to have existed where the camp now stands. Site 54 (PL. IV, A; FIG. 53, g). The three houses lie S. of the 'Chapel' site, near the modern 'Saxon Gate' farm. The longest building, 63 by 14 ft., has a cross-passage and a room at the upper end, where the width tapers to r z ft. The second building adjoins the first at right angles to its upper end, and the third lies 45 ft. to the west. A spring rises below the buildings. Site 55 (PL. IV, B; FIG. 53, h). One long-house, 52 by 12 ft., with a small room, 12 ft. square, at the upper end, lies immediately below and to the N. ofthe track. A small building lies S. of the track, and three others above it are concealed in the gorse. The scanty remains of another house lie close to one of the rqth-century field-banks. There are a number ofcrofts and enclosures, and below, to the NE., a spring. Site 56 (PL. IV, B; FIG. 53, c). In the corner of the field, N. of a modern barn, are two long-houses, parallel to each other, with a small circular pound at the N. end of the E. house. The main S. track passes alongside the hedge-bank from site 55, and continues W. across the open field, as a sunken grassy way, towards site 57. A branch track leads S. from this site towards the moor. A stream runs W. of the site from a spring on the S. side. Site 57 (PL. V, A; FIG. 53, f) is marked 'cottage and garden' on the tithe map, where it is the only house-site shown; the ruins are clearly oflater date, although the enclosures themselves may easily belong to an older homestead. There are two buildings, one ofwhich shows the tumbled stones ofa chimney-stack; the other building could have been a barn, or used in connexion with the Meldon quarries, whose workings extend close to it. A stream runs through the site from a spring on the S. side. Site 58 (PL. V, A; FIG. 53, i). These remains are difficult to interpret. Three buildings lie close to and S. of a hedge-bank, with crofts on its lower side. Disturbances of the ground, possibly caused by mining operations, have confused the area. Parts of the walls of a fourth and larger building exist to the E. This has a room at the upper end and a cross-passage, but the lower halfofthe house has been almost completely destroyed. Site 59 (PL. V, B; FIG. 53, j). This site lies below the others, well over the brow of the hill, and comprises seven long-houses in two groups 2 I ft. apart. A sunken way curves down to the site from the lower, N., trackway through the park. Some of the buildings have been used as a dump for stones removed from the field during recent land improvement. Three of the houses, which lie close to the hedge-bank, are hidden in a spinney ofunfriendly thorn bushes. The largest, 59 by 12 ft., has a room at the upper DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 131

-',- ,.~,*Cairns "'I "'. Corndon Down "" '-',' " Yarlor Down " ""II " ~ Corndon Tor ". Cairn Var Tor " " " 'I "II

Hut ctrctes o~~'-.~.::-_ . ~toA5HBURT6N l/IU ----~ , 1~~_,~~~~: m .. ~LJ -_ 0 II '" Hut Circle ~ ~, :~~7c;"

I:J ~ :l

t .Ar I River Dart

o 1000 3000 •..... ' !

1000 FIG,54 MAP SHOWING POSITIONS OF DESERTED SITES AND PILLOW-MOUNDS on Yar Tor, Eastern Combe and Vaghill (Widecombe-in-the-Moor), Dartmoor (p. 131 f.), Pillow-mounds are indicated by hatching. Based on the 6-in. O.S. map with the sanction of the Controller ofH.M. Stationery Office. Crown copyright reserved end which tapers to an unusually narrow width of9 ft. There is an oval-shaped pound against the bank below the first group, and a spring above, which runs through the site. Surrounding crofts and enclosures must have been destroyed by recent ploughing, although some can still be seen to the N., beyond and below the hedge-bank, and among the trees.

DESERTED SITES ON YAR TOR, EASTERN COMBE AND VAGHILL IN WIDECOMBE-IN-THE­ MOOR (PLS. VI, B, VII, A; FIGS. 54, 55, a-c) The area under discussion is contained by the river Dart on the south and west and on the north by the main -Ashburton road from

10 CATHERINE D. LINEHAN to Eastern Combe, which itself forms the eastern boundary. The head of the combe, which lies a short distance above and to the north of the road, between Corndown and Yar Tor, is also included in the area. The Dart was the route of penetration for the first inhabitants of this area of early settlement, as is shown by the many prehistoric remains. Three large cairns dominate the heights of Corndown; there is another on Yar Tor, on the eastern slopes of which are the remains of a cist and a stone row. Extensive pre­ historic enclosures with hut-circles and pounds exist on these hillsides; some appear to have been reconstructed in later times, and reused in connexion with the medieval homesteads. Of the three deserted sites recorded, Eastern Combe and Vaghill may have been the homes of warreners, since they lie near some of the pillow-mounds of the warren. In such a large area the warren could have been shared by two men, as, indeed, is suggested by the lease to two separate holders cited below (p. 139). No other documentary evidence has been found for these sites, but there is a local recollection that the building on Vaghill used to be known as 'Warreners' House'. All other names have been forgotten, and the settlements are now part of the open common lands ofSpitchwick Manor. Domesday shows Spitchwick Manor as held by the king; there were 8 villeins and 8 ploughlands, although only 4 ploughs were in use. It was a large manor, bounded on the south and west by the Dart and its tributary the , and on the north and east by Challacombe (Manaton) and Blackaton and Dewdon (Widecombe-in-the-Moor).

Site 103, Tar Tor (PL. VI, B; FIG. 55, a), lies close to the main road on the N. side, where it curves round to cross the head of Eastern Combe. Two long-houses lie side by side, both with cross-passages, and part of another building can be seen a little above, to the N.; it is likely that road-widening has covered other remains. There is extensive terracing on the hillsides, as well as the walls of smaller enclosures near the houses. Two sunken ways lead towards Corndown and Yar Tor. Site 97, Eastern Combe (PL. VI, B; FIG. 55, b), a little way down the combe from the Yar Tor site, consists of two buildings, each 24 by II ft., which appear to be of rSth­ century date, but parts of other walls near by suggest an earlier medieval settlement. There are three small enclosures, and three pillow-mounds outside them, two of which have been partially destroyed by the hedge-banks of the lower enclosure. A spring rising near the houses is joined by the stream from another spring which rises a little below the Yar Tor site. Both streams have become choked, and much of the ground around the Eastern Combe site is boggy. The higher stream has changed its course to scour out the old sunken way leading from the road. Site 102, Vaghill (PL. VII, A; FIG. 55, c), lies almost hidden in a dip on the steep S. side of Vaghill above an island in the river Dart. There is one building, 20 by 15 ft., and some tumbled enclosure-walls round the rocky knoll which shelters the small valley on the S. side. The five pillow-mounds near by include a circular one. The Warren (W. 15; PL. VII, A). Other pillow-mounds of the old warren extend over the S. and E. sides of Vaghill and into Eastern Combe. There are five on the brow of the hill overlooking the combe, and these may have belonged to the Eastern Combe part of the warren. On the S. slope are six others, including two circular ones, and another can be seen from the aerial photograph to lie on the spur of Logator Rocks DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 133 (misnamed Lucky Tor on the O.S. map), close to the Dart. Others may exist in the very rough and overgrown land on the lower slopestowards the river. ~ EARUER HOUSE

~ \ b

c mtV o

FIG. 55 DESERTED SITES IN WIDECOMBE-IN-THE-MOOR, DARTMOOR (p. 132) a, Yar Tor; b, Eastern Combe; c, Vaghill

DESERTED SITES IN CORNWOOD (FIGS. 56, 57, a-j') Six deserted sites have been found in Cornwood parish in the upper valleys of the Yealm and its tributaries. This is a large parish of over 10,000 acres, of which more than 6,000 acres are woodland and common, stretching into the depths of the moor at Ermehead. The Domesday entry for Cornwood shows it held of the count of Mortain; there were 8 villeins and 5 ploughlands, although only st ploughs were in use. Most of the Cornwood homesteads were held by ...... <.>0 >.f>.

o >...., II: ttl ~ ...... Z ttl Rook Tor t:::l

l:"' ...... Z ttl II: > Z t .N' I q 1°.00 " ~. 500 1000 1500metres ""

FIG. 56 MAP SHOWING POSITIONS OF DESERTED SITES IN CORNWOOD, DARTMOOR (pp. 133, 135) Based on the 6-in. O.S. map with the sanctionof the Controller of the Stationery Office. Crowncopyright reserved DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 135 free tenants from the 13th century, and it is likely that the sites under review were among them (see p. 119). High House, Ford and Harrowthorn are named in the Land Tax list of 1824, but it is probable that the houses were no longer occu­ pied, as in each case the occupier mentioned also held near-by farms. Parkland Newtake is perhaps represented by the Little Parkland in a Land Tax list in 1781, held by the occupier of Cholwich, whose lands adjoined it. Place-Names of Devon gives references for some of the sites. Harrowthorn is mentioned in 1249; Ford and Hawns in 1330, and Parkland in 1333.23 No other documentary evidence has yet been found for any ofthe six sites.

Site I I, Ford, or Ford Barn (FIG. 57, a), is the largest in the group. Seven buildings have been counted; the whole settlement area lies in a large enclosure, and the uneven ground, with a confusion of banks and walls, indicates numerous rebuildings and a considerable period of occupation. One long-house on the SW. side, lying on a different alignment and showing only slight remains, suggests an earlier phase of occupation. The largest house appears to be of typical 17th-century construction; two other buildings are of still later date, showing the use of mortar. It has been noticed that places named 'barn' represent the final phase of an ancient holding. In Cornwood parish, for instance, Wide's Barn and Brook Barn, although now only ruined modern farm buildings, seem to have been ancient holdings, and their land is still being cultivated by neighbouring farms. At Ford Barn desertion is complete. Site 12, Ford Waste (FIG. 57, b), lies in the N. part of the enclosures of this name, close to the 'corn-ditch' which divides them from the open Penn Moor. There are two long-houses with a room at the upper end, one of which is over 70 ft. long; a small building lies at the E. end of the enclosures immediately surrounding the homestead. No documentary evidence has been found for this site, which could, perhaps, have been an off-shoot from the Ford Barn settlement. Site 13, Harrowthorn (FIG. 57, d), is in the S. end of Harrowthorn Plantation. There appears to have been considerable quarrying here and a number of trees have been uprooted by gales. As a result, this is a very confused site of tumbled banks which are difficult to interpret. Of the three buildings, all near the S. wall of the plantation, the middle one is the largest. Site 14, Hawns (FIG. 57, c). The buildings, which probably represent the site of ancient Hawns, lie on the opposite side of the valley of Broadall Lake from High House and a little further upstream. They are built along the contour, each with one entrance on the lower side, and are constructed ofexceptionally large stones. There are a number of enclosures, and also pits and mounds from later mining operations. The natural forest of Hawns Wood, which has encroached on the site after its desertion, has recently been felled and rows of fir trees planted, and these, before long, will further conceal the remains. Site 15, High House or High House Waste (FIG. 57, e). This site, together with its enclosures, has recently been purchased by the Dartmoor Preservation Society. It lies on the hillside above the W. bank of the Broadall Lake, a term indicating a stream or brook on Dartmoor. This is a tributary of the Yealm, which it joins in Dendles Wood about half a mile downstream. The main building shows later development and seems to have been a cottage with three ground-floor rooms. There is a collapsed chimney between the two rooms at the upper end, and a tumbled porch on the S. side leading to the central room. Near by are two small buildings, one of which may have been constructed from a hut-circle. The earliest building lies farther up the slope, against the bank of the sunken way which leads down to the cottage. The walls of this house are only just visible, but the N. end is marked by a distinct kink in the lane-bank into

>3 Place-Names of Devon (English Place-Name Soc., I, 1931-2), pp. 270-1. CATHERINE D. LINEHAN which it is built. On the steep slopes are extensive enclosures and massive terracing, and lower down a good spring rises between the buildings and the brook. A small rectangular

ENCLOSURE WALL. -t ,Ai' I I'i:" ::" " D" ~DJ':" r'::

ENTRA"N.::C:E====~======~ ~AT'" b a

nu d

!I JIII J TRACK.

e f

FIG. 57 DESERTED SITES IN CORNWOOD, DARTMOOR, DEVON (pp. 135 If.) a, Ford Bam; b, Ford Waste; c, Hawns; d, Harrowthorn; e, High House; j, Parkland Newtake

enclosure lying to the SE. appears to contain another building, of which only one side can be seen. Site [6, Parkland Neiotake (FIG. 57,!), lies a little away from the valleys and is in the middle enclosure of the. three which represent Parkland Newtake. Parkland itself is DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 137 about half a mile to the west. There.is one house only, and it is a good example of the single-house homestead. It is built along the bank and has two entrances on the lower side and a room at the upper end. Adjoining are small enclosures, and a number of hut-circles with their enclosures lie near by.

DESERTED SITES AT BUTTERBERRY IN PETER TAVY (PL. VI, A; FIGS. 58, 59, a-b) Two deserted sites, still known locally as Butterberry, lie about a mile south-east of Wapsworthy, and south of the Youldon, or Wapsworthy brook. An amphitheatre ofhills encircles the valley ofthe brook and its small tributaries. All round the slope of the hill extensive terracing and enclosures can be seen, suggesting an area of early settlement, as does the name Youldon, or 'old down'. Today, upstream from Wapsworthy and its still cultivated enclosures, only the empty house of Longbetor farm and the two deserted sites at Butterberry show the remains of homesteads, although it is said locally that 'several cottages' once existed north of Longbetor. The area is now given over to ranching and extensive sheep-runs.

Sites 60 and 6I, Butterberry I and II, lie at heights of 1,225 and I, I 25 ft. respectively. The higher site (FIG. 59, a) consists of six buildings. One, with a cross-passage and a room at the upper end, shows considerable tapering at both ends of the house; two buildings have small annexes, and another long-house appears to have been rebuilt in part to make a shed or barn. There are crofts adjoining two of the houses, and a track enters the site through the old enclosures from Wapsworthy. A spring rises a short distance above the site, and forms a small stream which runs down past the lower site to join the Youldon brook below it. The lower site (FIG. 59, b) comprises three build­ ings, with yards and enclosures; here, also, two of the buildings show later development.

Butterberry was the demesne land of the manor of Cudlipptown to the west. 24 Cudlipptown at the time of Domesday was included in the manor of Tavistock (Taustocha), which was the principal demesne and seat of Tavistock. The abbot had to contribute IS knights, a duty imposed by William the Conqueror. Domes­ day Book shows 6 knights holding lands in Tavistock, and others were added to insure against possible defaults. One such was endowed during the abbacy of Wymund (c. 1091-1 I02) and consisted of Roweburgh, now Morwell, and Cud­ lippe, an isolated holding on the edge of the Moor. The connexion between these two manors lasted for centuries; in I 194 they were seized by Richard I together with all lands held by those accused of siding with John Lackland. The Pipe Roll for 1194 shows the revenue as I4s. 7d. from Morwell and qs, IId. from Cudlipp; the following year Ralph Archdeacon proffered IO marks to recover his inheritance. In 1488 Morwell and Cudlipp were held together for one knight's fee." In 1252, the abbot of Tavistock granted a tenement at But­ worthy, reserving the right to enclose let or otherwise use the whole waste of Butworthy, except the common pasture on Langstone Moor (misnamed Laun­ ceston Moor on the O.S. map);" Butworthy was leased for twenty years from

'4 H. P. R. Finberg, Tavistock Abbey(1951), p. 242. '5 Op.cit. in note 7, pp. 155-162. ,6 Op.cit.in note 24, p. 33. ... t.>O CO

00"1­ 'Of \Of ~qO ",0" t eN' o > >-3 I :::t: t'1 ...:;d Z t'1 tj ...t'"' Z Cudlipplown Down t'1 :::t: > Z Langslone Moor ~ a 1000 3000 5000 feel Iw •••• ! ! I

r-----, i i n 500 1000 1500 metres

FIG. 58 MAP SHOWING POSITIONS OF DESERTED SITES at Wapsworthy and Butterberry (Peter Tavy), Dartmoor (p. 137) Based on the 6-in. O.S. map with the sanctionof the Controller of H.M. Stationery Office. Crown copyright reserved DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 139 1301.27Other variants of the name are shown in Place-Names qfDevon: Boterworthi in 1330, and Boterworthie in 1408.28 The Land Tax list of 1780 shows two holdings at Butaford alias Butworthy owned by Stephen Madaford for I4S. and I8s. respectively. This is a comparatively small sum for a relatively large area of land and may represent the final phase of the farms, when only a barn or outhouse remained in use, buildings, perhaps, which show signs of later redevelopment.

II. RABBIT-WARRENS (TABLE III, opp. p. 124; PLS. VII, VIII; FIG. 59, c) Sixteen rabbit-warrens have been found on Dartmoor. The greatest concen­ tration is in the upper Plym valley, where 86 pillow-mounds have been counted in five warrens, Legis Tor and Ditsworthy (Sheepstor), Hen Tor, Willings Walls and Trowlesworthy (Shaugh Prior). It is interesting to note that 14 of the warrens are in or close to areas of pre­ historic settlement, the exceptions being Skaigh, South Tawton (PL. VIII, A), in the north, where there are few early remains, and Redlake Tramway (Ugborough), which was formed in recent times. Documentary evidence for the warrens is sparse. A lease concerning the making of Vaghill warren (Widecombe-in-the-Moor) is dated 23 April 1613 and is by William earl of Bath to Richard Reynell esq. and Walter Fursland of gent., 'concerning waste ground called Spitchwick Common lying between the river Dart on the west and south east, and from thence to Yartor on the north and from Yartor to Corndon Tor on the north and east to the west of Rowbrook hedge and so on to Logator on the East and so on to the river Dart with free liberty to make a warren there for the keeping breeding and killing of rabbits. And also if any rabbits go over the Dart to the commons there called Holne Commons ... the said Richard and Walter may kill them. Rent lOS.',"9 There is, however, no mention of the making of any mounds, and the problem of dating them remains to be solved. Both William Crossing and R. Hansford Worth by inference imply that rabbit burys were well-known objects. Referring to Zeal Burrows (South Brent) Crossing warns the visitor against mistaking such mounds for prehistoric barrows. 'They are really old rabbit shelters ... and mark the site of a former warren.i'" He also mentions that Trowlesworthy warren (Shaugh Prior) on the Plym is recorded in the 13th century and that in old documents relating to the forest the burys are recorded as a form of burrows. Unfortunately he gives no reference for these statements and so far they have not been traced to their source."

'7 Ibid., p. 242. ,8 Op. cit. in note 23, p. 232. '9 Exeter City Library Muniments, ref. 48/14/40/9. 3° Op. cit. in note I I, p. 369. 3' Ibid., p. 13. CATHERINE D. LINEHAN The inhabitants of Dartmoor today certainly recognize the pillow-mounds as indicating the existence of rabbit-warrens, and indeed most of the mounds show clear signs of occupation by rabbits; they were no doubt in use by the rabbits until myxomatosis almost wiped them out. The farming of rabbits was at

EHC~REWALL

::.:J

b

FIG. 59 DESERTED SITES AND PILLOW-MOUND ON DARTMOOR (pp. 137, 141 f.) a, Butterberry, site I (Peter Tavy); b, Butterberry, site II (Peter Tavy); c, Yalland warren (South Brent) times an industry of considerable importance both for the food they provided and for their skins. Huntingdon warren (Lydford) on the Avon, and Ditsworthy warren, Sheepstor (PL. VIII, B) on the Plym have houses only recently abandoned; Ditsworthy latterly included the old warrens of Hen Tor and Willings Walls (Shaugh Prior) and Legis Tor (Sheepstor). Trowlesworthy (Shaugh Prior) is DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 141 still occupied as a farm, although the rabbits have gone. The final blow to the ancient trade came as recently as 2 February 1956, when Devon was declared a 'rabbit clearance area' by the County Council and the industry became illegal. In an interesting article by Eric Hemery much information is given. 3" He says that the industry was carried on until after the second world war. He mentions that rabbits were fetching 9d. or IOd. each in the early years of this century in Devonport and markets, and he tells how 3-cwt. loads were carried down by pack-horse. During the second world war rabbit skins were selling for IS. each; the warrener would deftly skin the rabbits for his customers, and then carry the pelts back home to be dried and tied into bundles of two dozen for collection by the furrier. Later in this century carts were used, but the trips to market remained a day of hard toil for the warrener and his family, as well as the high spot of the week. Hemery gives more details of the earliest documentary evidence for Trowles­ worthy warren (Shaugh Prior), although again no exact reference is stated. The date is 1292 and the document tells that Samson de Traylesworthie some time before that date had become lessee ofland near the Plym; the grant was for additional land for a rabbit-warren." We cannot be sure that this Samson threw up mounds for his newly-introduced stock, but it seems likely that some such encouragement would be needed, especially in the boggy areas otherwise quite unsuited to burrowing animals. The mounds vary greatly in size and shape. On Legis Tor (Sheepstor) there are huge mounds up to 130 ft. long by 24 ft. wide, measuring from the middle of the ditches which surround them. In places an old boundary-ditch and bank seem to have been reused to form a succession of long mounds with only a few feet between them to form a gap. On Vaghill, Widecombe-in-the-Moor (PL. VII, A), there are four circular mounds, 30 to 40 ft. in diameter, which are the only circular pillow-mounds so far found on Dartmoor. Examples on Sheepstor warren (Sheepstor) have revetted stonework 2 or 3 ft. high with a mound above; others with similar revetting have been recorded at Willings Walls (Shaugh Prior) and Ditsworthy, Sheepstor (PL. VIII, B), both in the upper Plym valley. Some can barely be seen above the undergrowth, or only in the right light; others are quite upstanding, such as those on Sheepstor and in the Merrivale warren (Walkhampton) on the bank of the Walkham river, where their height is 4 or 5 ft. Some at Merrivale are marked, wrongly I believe, as tumuli on the O.S. map. Several appear to have been dug into from the top. By their scattered distribution in an area of prehistoric and medieval settlement, they are more probably pillow­ mounds. They are of an unusual oval shape, about 30 ft. long by 20 ft. wide, which perhaps explains their being mistaken for barrows. It is tempting to suggest an early date for this warren, as also for that at Yalland, South Brent (FIG. 59, c), because there are deserted long-houses close to some of the pillow­ mounds. This could, however, be illusory, as the medieval buildings may belong

J' E. Hemery, 'Rent Act has spelled doom to an ancient Dartmoor trade', Western Morning News, 3 Sept. 1956. CATHERINE D. LINEHAN to a period before the warren existed. Excavation might assist in clarifying the Issue. Examples of warrens near quarrying and mining works are the Plym valley group, Merrivale warren (Walkhampton) near the Walkham valley tin-workings, Huntingdon warren (Lydford) on the Avon, where there has been considerable mining, and Headland warren (North Bovey), near the Vitifer mines. It is known that quite recently mounds were thrown up for food and sport for the workers at the Redlake and Leftlake Clay Works that lie between the heads of the Erme and Avon valleys. Crossing describes the works as long-deserted but they were reopened in this century, and the tramway, alongside which the mounds can be seen, was completed in 191 1. The men stayed on the moor for weeks together in the summer, and no doubt the rabbits were an important asset. These modern mounds are of very rough construction and seem to be formed of waste material brought down the line from the workings; they show no surround­ ing ditch and are of varying sizes. One, near Spurrells Cross, has more the shape of a field silo, sloping down towards the end from which the material was dumped on the mound. There are five of these mounds near the track below Hangershell Rock and others along the way towards the clay works. The project failed after little more than 20 years." It is interesting to note that the tinner's emblem was three rabbits, arranged in a triangle, with only three ears between them. This can be seen in the medieval carved roof-bosses in several of the moorland churches, Widecombe-in-the-Moor, North Bovey and Chagford amongst them. Although the rabbit, together with the hare, was an earlier pagan fertility symbol, its adoption by the tin miners as their device suggests its importance to their economy in earlier times. Worth mentions a species of rush, ]. conglomeratus, as being especially luxuriant 'on the summits of the rabbit burries, which are mounds of peat soil raised to give dry conditions for the rabbit burrows. Similar instances occur elsewhere on the moor, so that it would appear that the opportunity afforded by light, well broken, peat soil, even when aerated by rabbit burrows, may outweigh the loss of moisture involved in such a roothold'.34 These rushes grow on the mounds in several of the warrens, including Ditsworthy (Sheepstor), Hentor and Trowles­ worthy (Shaugh Prior). They stand up above the other herbage, and thus aid recognition from afar. Stone-built vermin-traps have been recorded in several warrens. Worth notes a number of these," and more recently R. M. L. Cook gives details of such traps in south-west Dartmoor, mostly in the upper Plym valley. He has listed 58 traps and suggests a date 'not earlier than 1700' for their structure, because of their construction in the walls of tin-works of the r Sth century." Worth suggests that the traps were already a thing of the past by the middle of the r qth century. He records a conversation with Mr. Richard Lavers, the warrener at Trowles- 33 Information supplied by Mr. H. L. Watkins, , from his personal contact with some of the workers concerned. 34 Op. cit. in note 5, p. 70. 35 Ibid., p. 157. 36 R. M. L. Cook, 'Vermin traps on south-west Dartmoor', Trans. Devon Assoc., XCVI (1964), 190-201. DESERTED SITES AND RABBIT-WARRENS ON DARTMOOR 143 worthy, who died in 1914 aged 94, who 'had heard the structures were vermin traps for the elimination of stoats, weasels and other small animals of prey; but he did not know how the traps were set';" No doubt the traps were no longer necessary when the shot-gun became a common acquisition, and the barbarous gin was used. The method of catching the rabbits was always by the use of a long net, with help from ferrets and dogs. From all the evidence available locally, therefore, it can be confidently stated that, on Dartmoor, the pillow-mounds were constructed for the encourage­ ment and habitation of rabbits in the warrens devoted to their breeding.

APPENDIX CHALLACOMBE (MANATON) The Challacombe villagc-" and lynchets are well known; the lynchets were studied and planned by Dr. A. H. Shorter before the last war.t? A few of the houses appear to have been long deserted, but others are quite upstanding, including one which is said to have been an inn. There was a considerable increase in population in the rSth and t qth centuries, when the Vitifer mines at Headland (North Bovey), further up the valley, were being worked. Today the modern farmhouse and two cottages which have been converted into another dwelling are all that remain of the village. Including these, there are 12 houses and a number ofcrofts on either side ofan old track which goes north to Headland and beyond. A cross track in the centre of the village goes west on to Challacombe Down, and east over the stream towards another small site in front of a modern barn.t? Here one wall of a building and a croft remain, although it is possible that other houses lay under the barn. A track, known as the Challacombe Churchway, goes from this side of the valley across Blackaton land above Blackaton deserted village (Widecombe-in-the-Moor), beyond which it joins the Blackaton Churchway to pass over the shoulder of Hameldown to Widecombe-in-the-MoorY Although in Manaton parish, the people of Challacombe seem to have used the shorter route to Widecombe church, some two miles away; this is confirmed in the parish registers of Widecombe, in which there are numerous entries from Challacombe in the rSth and r qth centuries. The site has probably never been altogether deserted. The remains of the earlier houses indicate a considerable population, and a contraction, possibly at the same period as at sites already excavated (p. 119), can be suggested. A renewal of occupation occurred in connexion with the mining operations near by, followed by a partial abandonment when mining ceased. Challacombe is probably the manor represented by the one virgate held 'by a certain knight Roger' from Nicholas the Crossbowman who held Buckland-in-the-Moor. Testa de Nevil, dated about 1244, shows Cherlecombe again held with Buckland-in­ the-Moor for one feeY It has been suggested that this Cherlecombe was Challamoor in Buckland-in-the-Moor parish.s" but this seems unlikely, not only because of the difference in name, but because of its situation. In a document dated 1303, there is mention of 'one messuage one mill and three furlongs of land' in Churlecombe.w 37 Op, cit. in note 5, p. 158. 38 TABLE I (p. 117); TABLE II, no. 38 (opp. p. 124). 39 A. H. Shorter, 'Ancient fields in Manaton parish, Dartmoor', Antiquity, XII (1938), 183-9. 4° TABLE II, no. 39. 4' Op, cit. in note 3, fig. I. 4' O. J. Reichel, 'The Hundreds of Haytor in the time of "Testa de Nevil", A.D. 1244', Trans. Devon Assoc.,XL (1908),120. 43 Gp. cit. in note 6, p. 525. 44 Devon Feet of Fines, II, no. 899. 144 CATHERINE D. LINEHAN Challamoor farm is on level ground where there is only a small stream. It appears a most unlikely place for a mill, whereas at Challacombe there is a good flow of water, where the remains of a blowing-mill can be seen above the village on the bank of the brook. This could have been the site of the mill mentioned in 1303. Challacombe has clearly been a place of some standing, whereas Challamoor seems to be a single home­ stead. The confusion may have occurred because Challacombe, though held with Buckland-in-the-Moor, was in the parish ofManaton. Challacombe is also mentioned in a list of Fines Villarum dated 1504-5 headed Foreign Rents, which includes an entry 'Villat de Chalnecombe in parochia de Moneton VI d.'.45

POSTSCRIPT Since the completion of this paper, Mr. R. G. Haynes of Plympton has brought to my notice a further seven deserted sites. These have been added to the gazetteer as a supplement (opp, p. 124), and the positions indicated on the distribution map (FIG. 47). In his unpublished 'Warrens on Dartmoor's" Mr. Haynes has dealt with the warrens in detail and provided information on the field-systems and vermin-traps in the areas. In the Plym valley group, which is considered to comprise the oldest warrens, 160 pillow-mounds have been mapped, and in Huntingdon warren 80 mounds have been estimated. Evidence is given for regarding Merrivale and Wistman's warrens as com­ paratively recent, and mention is also made of the Trowlesworthy document (pp. 139 and 141 f.), giving Risdon's Survey of the County of Devon as the reference.P Professor E. Carus-Wilson has drawn my attention to references to rabbits on Lundy Island and the Isles ofScilly in the r zth century.s"

ACKNOWLEDGEMENTS I am indebted to Mrs. E. M. Minter for drawing the maps and plans, and to my sister, Miss E. H. Whetham, for her assistance with the text. Thanks are due to Miss R. Everard, Miss H. E. Gawne, Mr. H. French, Mr. K. Isham, Mr. D. J. Seymour, Dr. A. H. Shorter, Mr. J. V. Somers Cocks, Mr. R. L. Taverner and Mr. H. L. Watkins, for bringing a number of sites to my notice and for supplying information of value in the preparation of this paper; also to Mr. R. G. Haynes for the additional sites and for his permission to refer to his paper. I am also grateful to Mr.]. G. Hurst for his considerable interest in the work of recording the deserted sites on Dartmoor.

45 Dartmoor Preservation Assoc. (1890), Dartmoor, p. 40. 46 The manuscript is in the Plymouth City Library. 47 Tristram Risdon, Survey of the County of Devon, published 181 I from an earlier manuscript copy, P·392 • 48 Elspeth M. Veale, The English Fur Trade in the Later Middle Ages (1966), Appendix B, p. 209.