vE«srn OP

kUFORNlA kN DIC*0

Overs i;

THE VICTORIA HISTORY

OF THE COUNTIES OF ENGLAND

A HISTORY OF

DORSET

VOLUME III

THE VICTORIA HISTORY

OF THE COUNTIES OF ENGLAND

EDITED BY R. B. PUGH

THE UNIVERSITY OF LONDON

INSTITUTE OF

HISTORICAL RESEARCH

Oxford University Press, Ely House, J7 Dover Street, London, W.i

GLASGOW NEW YORK TORONTO MELBOURNE WELLLNGTON

BOMBAY CALCUTTA MADRAS KARACHI L.\HORE DACCA

CAPE TOWN SALISBURY NAIROBI IBADAN ACCRA

KUALA LUMPUR HONG KONG TOKYO

© University of London 1968

19 722718 X PRINTED IN GREAT BRITAIN

INSCRIBED TO THE MEMORY OF HER LATE MAJESTY

QUEEN VICTORIA

WHO GRACIOUSLY GAVE THE TITLE TO

AND ACCEPTED THE DEDICATION

OF THIS HISTORY

Knowlton in the Parish ok Woodlands The ruined 12th-century church stands within one of the prehistoric circles of Knowlton Rings; the site is thought to be that of a deserted village and the meeting- place of Knowlton hundred.

A HISTORY OF THE COUNTY OF

DORSET

EDITED BY R. B. PUGH

VOLUME III

WITH INDEX TO VOLUME II

PUBLISHED FOR THE INSTITUTE OF HISTORICAL RESEARCH

BY THE

OXFORD UNIVERSITY PRESS

1968

Distributed by the Oxford University Press until i January 1972 thereafter by Dawsons of Pall Mall

CONTENTS OF VOLUME THREE

Dedication ....

Contents ....

List of Illustrations

Editorial Note ....

Introduction to the Dorset Domesday

Translation of the Text of the Dorset Domesday

Introduction to the Dorset Geld Rolls

Text and Translation of the Dorset Geld Rolls

Summaries of Fiefs in Exon Domesday

Index to the Dorset Domesday and Geld Rolls

Index to Volumes II and III

Corrigenda to Volume II

	PAGE
.	V
.	ix
.	xi
.	xiii
By Ann Williams	I
By Ann Williams	6i
By Ann Williams	IIS
By Ann Williams	124
By Ann Williams	148
By Ann Williams	150
By P. A. Spalding and Ann Williams	159
....	. 189

IX

LIST OF ILLUSTRATIONS

and Celia B. Clarke, and is based on fL OrHn c u 'f^ ^'""^ ^ '^"ft by Ann Williams

H.M. Stationery Office Crown Co;;;gL?eservS^ "''' "''' ^'^ ""'^^'°" "^ *^^ ^-^-''^ ^^

Knowlton in the Parish of Woodlands

„ . ,^ ■ • • • . frontispiece

Domesday Map

between pages 60-61

XI

EDITORIAL NOTE

The Victoria History of Dorset, Volume II, containing most of the 'general' articles for that county, appeared in 1908. Articles on natural history, pre-history, and schools, and the translation, with commentary, of the county section of Domesday Book then remained to be published in order to complete the 'general' volumes. Though a volume to contain those articles was in preparation at the time, it was not proceeded with, and the First World War put a stop to all further activity on Dorset. An opportunity arose in 1965 to publish separately the Domesday section, which had been prepared for another purpose, and it was decided to do so and not to await the completion of any other 'general' articles. The Royal Commission on Historical Monuments are in any case actively engaged in surveying the county's prehistoric monuments and the case for compiling a partially overlapping survey did not seem compelling. There is, moreover, no strong probability that natural history articles, apart from a survey of physique, will now be needed. They have been omitted from the Victoria History scheme in recent years. It is possible that accounts of ancient endowed grammar schools will in Dorset's case eventually be incorporated in the 'topographical' articles.

The structure and aims of the History as a whole are outlined in an article published in the Bulletin of the Institute of Historical Research, Vol. XL (No. loi. May 1967). In preparing the present volume for the press much valuable work has been done by Miss Celia B. Clarke, formerly an Assistant to the Editor.

DOMESDAY SURVEY

I. The procedure of the Domesday survey — the Exchequer text and the Exon. Domesday — the Domesday commissioners and the hearing of claims — assessment of the shire for geld — teamlands and ploughs — land-values, 1066 and 1086 — -the peasants — manorial adjuncts, meadow, pasture, woodland, and others — the Dorset boroughs, pp. 00-00. II. The land of the king, 1086 and 1066 — the pre-Conquest landowners of Dorset — the survival of the English — the lands of the religious houses, in 1086 and before the Conquest, pp. 00-00. III. The lay tenants in 1086 — the king's thegns and the king's Serjeants — the later history of the fiefs, pp. 00-00.

I

DORSET, Wiltshire, Somerset, Devon, and Cornwall, the five counties covered by the Exon. Domesday, probably comprised one of the circuits into which England was divided for the making of the Domesday survey.- Each circuit was assigned its own bodies of commissioners^ and, from a passage in the account of Somerset, it has been assumed that William, Bishop of Durham, headed the group of commissioners for the south-west, but the passage could be otherwise construed.** The commissioners seem to have held special sessions of the shire court, at which the juries of the shire and the hundreds gave sworn evidence, but there is little in the accounts of the south-western shires to illustrate this process. There are several references to the testimony of the English and the thegns of the shire, ^ none of which occurs in the Dorset section, and in Devon there is a single reference to the men of the hundred.*^ In Dorset there are four references to oral testimony, but the hundred juries are not mentioned. ^^ It is noticeable that in Domesday there are no hundred rubrics for any of the five south-western shires, although the rest of the English counties were so rubricated. Two hundreds in Dorset are mentioned inci- dentally, Buckland hundred, where there were 3I virgates attached to the manor of Bingham's Melcombe (no. 30), and Purbeck hundred, where William of Briouze held 7 hides less | virgate (no. 296). It is possible partially to reconstruct the Dorset hundreds by collating the Dorset section of the Domesday survey with the Dorset Geld Rolls. ^ It then emerges that the manors of each tenant-in-chief in Domesday are arranged in a fairly consistent order of hundreds, or rather groups of hundreds. ^ Whether this order indicates that the records of the court proceedings were originally arranged hundred by hundred, as in the Inquisitio Comitatus Cantabrigiensis, is conjectural. It is possible that when the original returns were sent to Winchester, they were already in feudal order. The arguments for this view largely turn on the relationship between the Exchequer Domesday and the Exon. Domesday, preserved in Exeter cathedral library. Exon. Domesday in its original form must have covered all five south-western counties, but the Wiltshire section, with the exception of one manor, and four-fifths

' The author wishes to thank Professor R. R. Darlington ■• Dom. Bk. (Rec. Com.), i, f. 87b; see R. W. Eyton,

for his invaluable assistance in preparing this article, the Domesday Studies: Somerset, i. 12-13; V. H. Galbraith,

ensuing translation, and the text of the Geld Rolls. Making of Dom. Bk. 87, 94, 207 ; V.C.H. Wilts, ii. 42, n. 2.

^ Eyton distinguished 9 circuits in all, which A. Ballard ^ p_ w. Maitland, Dom. Bk. and Beyond, 11, n. i;

(Domesday Inquest (igo6), 12-13) reduced to seven; see Galbraith, op. cit. 70 sqq. For the Wilts, evidence on this

Domesday Re-Bound (H.M.S.O., 1954), App. II. matter, see V.C.H. Wills, ii. 43.

' Robert, Bp. of Hereford, in his contemporary i" Dom. 5/t. (Rec. Com.), i, f. 107; iv. 277.

account of the survey, says that there were 2 sets of ' See nos. 263, 308, 369, 378 and ex.

commissioners, one sent to check on the other: W. H. * See p. 115 sqq.

Stevenson, 'A Contemporary Description of the Domesday ' R. Welldon Finn, 'The Making of the Dorset

Survey', E.H.R. xxii. 74, translated in Eng. Hist. Doc. ii. Domesday', Proc. Dorset Nat. Hist, and Arch. Soc. bcxxi.

851. 150-1.

DO. in I A

A HISTORY OF DORSET

of the Dorset section have not sunived. Apart from the incomplete descriptions of the five counties of Wiltshire, Dorset, Somerset, Devon, and Cornwall, the Exon. Domes- day contains the Geld Rolls for all five counties, including three distinct versions of the Wiltshire Rolls, lists of terre occupate for Somerset, Devon, and Cornwall, and sum- maries of the fiefs of some barons, notably the Abbot of Glastonbury. The descriptions of the manors in Exon. Domesday are fuller than those in the Exchequer text, especially in recording the livestock statistics which the author of the Anglo-Saxon Chronicle found so shocking. '° Many surnames and occupations are recorded in Exon. Domesday and omitted in the Exchequer text. Exon. Domesday regularly distinguishes between the demesne, which the Exchequer text records sporadically, and the land of the t/V/a/;/," which the Exchequer text does not mention at all. The phraseology of Exon. Domesday is very diflFuse, in strong contrast to the brevity of the Exchequer text, and the numerous differences in terminology and the spelling of place and personal names have given rise to the belief that the two versions are 'independent copies of the same original'.'- It was also suggested bv Reichel that the sections for Cornwall, Devon, and Somerset were compiled at Exeter from the original returns, while the sections for Dorset and Wiltshire w^ere made at Winchester, from an Exchequer digest of the ori- ginal returns.'-' This view, which seems on the face of it unlikely, is not borne out by any significant differences between the Dorset section and the rest of Exon. Domesday. More recently, however, the theory has been adopted that the Exchequer text for the south-western counties was derived from Exon. Domesday.'-*

Apart from the question of place and personal names there are discrepancies between the two texts which are difficult to explain if one is based upon the other although on balance the general resemblance of the two texts makes it difficult to believe that they are quite independent of each other. The Exon. Domesdav for Dorset covers the land of the king, with the exception of the two manors formerly held by Countess Goda, the land of the Countess of Boulogne, the lands of Cerne Abbey, Abbotsbur}- Abbey, Athelney Abbey, Tavistock Abbey, and INIilton x^bbey, the lands of William of Moyon, Roger Arundel, Serle of Burcy, the wdfe of Hugh fitz Grip, and Walter de Claville. In all, i6o of the 515 manors recorded in the Exchequer text are also in Exon. Domesday, covering about one-third of the total hidage of the county. As has been said above, Exon. Domesdav contains information not in the Exchequer text; it is also true that the Exchequer text contains items of information which do not appear in Exon. Domesday. At Spetisbury (nos. 274 and Ixxxiv) there were two pieces of pasture, one piece measuring 5^ furlongs by 2 furlongs and in alio loco another piece measuring 7.\ furlongs by i^ furlong. According to the Exchequer text this second piece of land lay super aquain but these words do not appear in the Exon. entrv. There is a more serious omission in the Exon. account of the borough of Shaftesburv. The Exchequer text says that the Abbess of Shaftesbury had there 151 burgesses, 20 mansiones vacuus, and a garden, the whole rendering 65^., but none of these details is in the Exon. account of the borough.

The most serious discrepancy in the arrangement of manors concerns the land of the king. In the Exchequer text the six manors which had belonged to King Edward, beginning with Portland, come first, followed, with a separate heading, by the

'"'... nor indeed (it is a shame to relate but it seemed '^ O. von Feilitzen, Pre-Conquest Personal Sanies of

no sham.e to him to do) one ox nor one cow nor one pig Dom. Rk. g, n. i.

which was there left out, smd not put down in his record' : " \'.C.H. Devon, i. 375-80.

Anglo-Saxon Chron., a revised translation ed. D. White- '■• R. Welldon Finn, 'The Immediate Sources of the

lock and others, 161-2. Exch. Domesday', Bull. John Rylands Libr. xl. 47-78;

" For a discussion of villani and other classes of Ga\hTZ\X\i, Making of Dotn. Bk. cap. WW. peasants, see pp. 14—20.

DOMESDAY SURVEY

manors of Earl Harold and Little Puddle, belonging originally to Earl Harold's mother. In the Exon. Domesday Earl Harold's manors come first, with no heading to distinguish them from the land of King Edward, which follows. Countess Gytha's manor of Little Puddle is included among her son's manors, next to Puddletown, and Portland is placed between the manors of Ibberton and Fleet, both belonging to Earl Harold. In addition the lands of BoUo the priest, Bristuard the priest, and the abbey of St. Wand- rille are interspersed with the king's manors, whereas in the Exchequer text they are entered separately. A virgate of reeveland, held by Aiulf the sheriff, is also entered among the king's manors, but it does not appear at all in the Exchequer text. The manors of Queen Maud are entered in almost the same order in both texts except that one of the manors of Tarente (nos. 26 and xxxv), which in the Exchequer text lies fifth among the manors held by Hugh fitz Grip of the queen, lies eighth in Exon. Domesday. Abbotsbury, the chief manor of the abbey of that name, comes third in the Exon. arrangement of the abbey's fief, and first in the Exchequer arrangement, and Milton Abbas, the caput abbatie of Milton Abbey, lies eighth among the abbey's manors in Exon. Domesday and second in the Exchequer text.'^

Of the discrepancies in actual content the most serious relates to the wood of Hauocumbe, attached to the manor of Burton Bradstock (nos. 2 and x). According to the Exchequer text one-third of the wood was held by Earl Edwin, which Exon. Domesday says belonged to Earl Godwin. The Exon. version is probably correct, since this portion of the wood was appurtenant to Frampton (no. 121) which was held T.R.E. by Countess Gytha, Earl Godwin's widow. The value of Nettlecombe (nos. 88 and li) is also a matter of disagreement between the two texts. According to Exon. Domesday the manor reddit abbati viii libras et prefato militi I solidos et v et quando abbas recepit valebat XX solidos plus. The former value of the manor must therefore have been ^Tii 15^. od. The Exchequer text, however, gives the former value of the manor as j/^12 o^. od. There are numerous other small discrepancies, which could have been due to mistakes in copying. At Child Okeford (nos. 7 and i) the king had 9 (viiii) bordars according to Exon. Domesday but 8 (viii) bordars according to the Exchequer text. At Creech (nos. 412 and cxlv) the pasture measured 7 furlongs by 4 (iiii) furlongs according to Exon. Domesday and 7 furlongs by 3 (iii) furlongs according to the Exchequer text. In all these cases it is easy to see how a misreading may have occurred. Similarly at Hampreston (nos. 19 and xxv) there were 2 villani according to Exon. and 5 villani in the Exchequer text, which could be explained by a misreading of ii as v. At another manor in the same vill of Hampreston (nos. 389 and cxxi) the wife of Hugh had i villanus and i bordar according to Exon. Domesday, but i villanus and 2 bordars according to the Exchequer text. At Cerne (nos. 108 and Ixxxii) 7 bordars in Exon. have become 5 bordars in the Exchequer text. At Winterborne Stickland (nos. 403 and cxxxvi) the former and present values of the manor have become transposed. A mistake like this could arise through the difference in arrangement between the two texts, since Exon. always places the 1086 value first, whereas the Exchequer text gives the former value first.

There are frequent instances of omissions in the Exchequer text. Exon. Domesday records that William of Moyon's manor of Hammoon (nos. 277 and Ixxxvii) was held of him by Torstin, but the Exchequer text omits this and treats the manor as if William

'5 The rearrangement of the lands of Abbotsbury Puddle, in Puddletown hundred, follows the manor of

Abbey and Milton Abbey in the Exch. text has disarranged Puddletown in Exon. Domesday, but is isolated from it in

the hundredal order. Abbotsbury lay in Uggescombe the Exchequer text. This is not the case with the queen's

hundred, along with Portesham and Shihinghampton, manor of Tarente, where the position is reversed; it is in

which it immediately precedes in Exon. Domesday, and the Exon. arrangement that the hundredal order is upset,

Milton lay in //a/tone hundred, with Lyscombe and Wool- and it is rectified in the Exchequer text, land, its neighbours in Exon. Domesday. Similarly Little

A HISTORY OF DORSET

held it in demesne. At Turners Puddle (nos. 391 and cxxiv) there was i hide, 4 acres, and a garden which did not pay geld, which are recorded in Exon. Domesday but not in the Exchequer text, and the same is true of a piece of woodland i league and 8 furlongs long and i league wide in Puddletown (nos. 8 and ii), 15 acres of wood at North Poorton (nos. 329 and ci), 8 cottars at Frome St. Quintin (nos. 15 and xxi), 2 cottars at Chilfrome (nos. 278 and Ixxxviii), and a vUlamis with J virgate at Winterborne Houghton (nos. 275 and Ixxxv). Exon. Domesday reveals that only half of the 2 mills at Child Okeford (nos. 7 and i) was held by the king, and that the wood at Nettlecombe (nos. 88 and li) nullum fnictum fert. At Cruxton (nos. 279 and Ixxxix), Durweston (nos. 401 and cxxxiv), and Ringstead (nos. 409 and cxlii) the Exchequer text does not record the men's ploughs, although they are entered in Exon. Domesday. Other minor omissions include leaving out the words et dimidia in some cases, for instance at Cranborne (nos. 16 and xxii) where there were 2\ leagues of pasture in length according to Exon. Interlineations in Exon. Domesday are not always reproduced in the Ex- chequer text. At Cerne (nos. 108 and Ixxxii) the mill was worth xx(v)</. in Exon. Domesday and xxJ. in the Exchequer text. In the case of North Poorton (nos. 329 and ci) the geld assessment is left out by the Exchequer text, although it is given by Exon. Domesday, but this appears to be due to a scribal error. The Exchequer entry reads Wido teriet de Rogerio POVERTONE. Alwimis et Ulf te?iuerunt [T.R.E.] pro ii hidis. As it stands this makes little sense and it seems plain that the scribe intended to write something like Ahvinus et Ulf tenuerunt pro ii maneriis T.R.E. et geldabat pro ii hidis.^^ At Affpuddle (nos. 80 and xliii), where the Exchequer text breaks off short, the Exon. entry is complete but makes little grammatical sense.''' It seems as if the Exchequer entry, if taken from Exon., was left incomplete until this could be clarified. At Nettle- combe (nos. 88 and li), where Exon. Domesday records a knight with two hides of land, the knight was at first left out in the Exchequer text, but was added in the margin. Several of these marginal additions occur in the Exchequer text, possibly left out in the first place because of haste in the compilation of the Exchequer Domesday, but none of the other entries involving such marginalia survives in the existing Exon. Domesday.'^ Some omissions, such as that of the hundred rubrics mentioned above, occur in both the Exchequer text and Exon. Domesday. Apart from this it is noticeable that in many cases where a gap has been left in the Exchequer text for some item of information to be inserted, the relevant information is missing from Exon. Domesday also. It is not stated how many teamlands there were at Portland (nos. i and vi), Nettlecombe (nos 88 and li), or Winterborne Stickland (nos. 403 and cxxxvi), although spaces have been left in each case for the relevant information which is not recorded in the Exon. entries either. Spaces have also been left for ploughs at Torne (nos. 419 and clii), for the men's ploughs at Tarente (nos. 26 and xxxv) and Renscombe (nos. 91 and liv), and for the number of villani at Morden (nos. 385 and cxvii). At Stafford (nos. 383 and cxv) there is some confusion over the manorial adjuncts. The Exchequer text says that there were 24 acres of meadow, and 16 furlongs of pasture, and 8 acres, leaving a space after acres, which would logically be filled by woodland, since meadow and pasture have already been enumerated. This obscurity also exists in the Exon. text which states that the manor was divided between two men, each of whom held xii agros prati et viii quad- ragenarias pascue et iiii agros, without indicating to what the iiii agros refer. There are

" Cf. the entry for Milborne St. .'Vndrew (no. 477), " The 2 manors referred to in this entry are AfFpuddle

where the hidage is not given, and the entry for Petersham and Bloxworth (nos. 79 and xlii). The Bloxworth entry is

(Farm) (no. 375), where there is no value. In both these complete in both texts.

cases over-compression seems to be the cause. Neither is " See below, covered bv Exon. Domesdav.

DOMESDAY SURVEY

many such entries in the Dorset survey which cannot be checked against Exon. Domes- day, because the relevant portion has not survived. At Cheselbourne (no. 138) a space has been left for the teamlands, at Pulham (no. 146) for the number of ploughs in demesne, at Lulworth (no. 198) for the ploughs belonging to the men, at Wimborne St. Giles (no. 499) for the account of the mill, at Stalbridge (no. 42) for the former value of the manor, and at Knighton (no. 298) there is a space between the bordars and the men's ploughs where one would normally expect to find cotsets or cottars. The entry concerning Kingcombe (no. 485) is unfinished with room left to complete it, and Herston (no. 512), the last manor entered in the Dorset survey, is not completed. The account of Blackmanston (no. 476) is unfinished, since the value is omitted, but, instead of inserting the value in the space provided, the scribe has repeated the whole entry, with the value, later in the text (no. 489). Cases like this seem to indicate haste in the compilation of the Exchequer Domesday. This is borne out by the marginal additions, one of which, the knight with 2 hides of land at Nettlecombe, has already been men- tioned. The other instances are a virgate of land which did not pay geld at Catsley (no. 229), the name of the T.R.E. owner (Burde) at Rushton (no. 292), a burgess at Ware- ham, rendering 2S., attached to Povington (no. 242), and a virgate at Kington Magna (no. 245). The account of the woodland at Iwerne Minster (no. 131) was omitted and entered at the end of the following entry, and the T.R.E. tenure of Bricsrid at Frome Billet (no. 491) was added by interlineation. Apart from these the interlineations are few and confined mostly to titles, like Heraldus {comes), surnames, and the words et dimidia.

Not only small items of information but whole manors, and in some cases groups of manors, have been omitted from their correct position in the Exchequer text, and added in other places. A group of eight manors belonging to William of Moyon was left out and is recorded on the lower part of the dorse of a folio inserted (f. 8ib). In Exon. Domesday Poleham (nos. 276 and Ixxxvi), the first of the manors misplaced in the Exchequer text, is entered on the page which begins with part of the account of Winterborne Houghton (nos. 275 and Ixxxv). Winterborne is entered in the correct place in the Exchequer text, and it cannot be argued that the Exchequer scribe mislaid a sheet or series of sheets of the Exon. Domesday and found them later. The land of the Countess of Boulogne is entered almost as an afterthought in the Exchequer text, after the land of the king's Serjeants. In Exon. Domesday the three manors of the countess are entered on one sheet only (f. 33), the other side of which is blank, and they follow the king's land and immediately precede the land of Cerne Abbey. It is possible that the misplacing of this leaf caused the omission of the countess's manors from the places where it would be more appropriate to find them, that is, with the lands of the Count of Mortain and Earl Hugh, or with the lands of the wife of Hugh fitz Grip. Other dis- placed manors are Iwerne Courtney (no. 316), the single Dorset manor of Baldwin of Exeter which appears on folio 81 inserted in the Exchequer volume and having William of Moyon's manors on the dorse; the king's manor of Hinton Martell (no. 31), inserted on a special sheet (f. 76); Compton Valence (no. 357), the manor of Hugh de Port, entered at the foot of folio 83 ; Kingston (no. 134) and Farnham (no. 135), belonging to Shaftesbury Abbey, added at the foot of folio 78b; North Poorton (no. 249), belonging to Ernulf of Hesdin, added at the foot of folio 80b, and three manors (nos. 510-12), belonging to the king's Serjeants, which were omitted and added after the land of the Countess of Boulogne.

It is noticeable in the Dorset survey, as in other parts of the Exchequer Domesday, that the index given on the first folio of the survey, after the account of the boroughs, does not tally with the headings in the text either in arrangement or in terminology.

DO. Ill C A2

A HISTORY OF DORSET

Aiulf is called vicecomes in the index and camerarius in the heading. Baldwin is Baldwin de Execestre in the index but Baldwin alone in the heading, and similarly Waleran is Waleran J'enator in the index but Waleran alone in the heading. Maci de Moretanie in the index becomes Mathiii de Moretania in the text. The heading in the index Reinbaldus presbyter et alii clerici becomes terra elemosinarioriim regis in the text. In the same way Gudmimd et alii taini and Willebnus Belet et alii servientes regis in the index become terra tainorum regis and terra servientium regis in the text. Alvred of Epaignes has a heading and a number in the text but not in the index, with the result that the numbers do not tally, and the discrepancy was solved only by omitting heading and number from the entry of Iseldis's land in the text, although they appear in the index. The lands of the abbey of St. Wandrille and of Hugh de Boscherbert have no headings in the text, the relevant numbers being inserted in the margin. The heading Hugo de Luri et alii fraud is omitted in the text, the number being inserted in the margin by the land of Hugh de Lure. The index lists first the land of the king (I), then the land of the Bishop of Salis- bury (H), and then the land of the monks of Sherborne (HI) but in the text the lands of the monks are entered as part of the bishop's fief, with the words hec novem descripta maneria sunt de zictu monachorum Scirebiirnensium at the end of the section relating to the monks' land, and the number 'HI' inserted in the margin at the point where the lands of the monks begin, half way through the account of Sherborne itself (no. 37). The manors of the bishop both precede and follow the lands of the monks. According to the index the land of the abbey of Montevilliers precedes that of the canons of Coutances, but in the text the positions are reversed.

Despite the difference in nomenclature and phraseology and the inclusion in the Exchequer text of some items not in Exon. Domesday, the relationship of the two texts seems to be closer than some authorities would allow. With exceptions most of the discrepancies could be put down to the speed at which the Exchequer text was com- piled, and the fact that the Exchequer text often leaves spaces just where the informa- tion is lacking in Exon. Domesday seems to indicate that the Exchequer text was compiled either from Exon. Domesday or a fair copy.

The Domesday commissioners were required to ascertain the name of each manor {mansio), who held it T.R.E., who held it in 1086, how many hides there were, how manv ploughs in demesne and among the men, how many rillaiii, cottars, servi, free men, and sokemen, how much wood, meadow, and pasture, how many mills and fishponds, how much had been added or taken away, how much it used to be worth, and how much it was worth, and how much each freeman and sokeman had. All this information was to be recorded for three different times, scilicet tempore regis Aedzuardt et qiiando rex Willebnus dedit et quomodo sit modo et si potest plus haberi quam habeatur.^^ Where the question of tenure was concerned, the commissioners in their capacity as justices heard conflicting claims. In Dorset the son of Odo the chamberlain claimed the manor of Chelborough (nos. 280 and xc) held by William of Moyon. According to Exon. rex vera iussit ut inde rectum habeat, but William continued to hold the manor. The Abbess of Shaftesbury had been more fortunate in respect of her manors of Cheselbourne and Stour (nos. 127, 138), which Earl Harold had taken T.R.E., for King William eas fecit resaisiri quia in ipsa ecclesia inventus est brevis cum sigillo regis Eduardi precipiens ut ecclesie restituerentur. The writ, however, also ordered the return of Bingham's Melcombe (no. 30), but rex adhuc tenet. At Povington (no. 242), belonging to Robert fitz Ceroid, the mill was claimed ad opus regis, and at Friar Waddon (no. 143),

" The commissioners' terms of reference are preser\ed Com.), iv. 496 ; Inquisitio Comitatus Cantabrigiensis, ed. in the preamble to the Ely Inquest : see Dom. Bk. (Rec. N. E. S. A. Hamilton, 97.

DOMESDAY SURVEY

which Hugh fitz Grip had given to the abbey of MontevilUers, the Abbot of Abbotsbury had been entitled to vi acras messis et Hi circscez de consuetudine sed Hugo nunquam dedit. Two hides in Tatton (nos. 398 and cxxxi), which were de dominio abbatie de Cernel T.R.E., were held by the wife of Hugh fitz Grip in 1086. Hugh is said to have taken them from the church {has Hugo super abbatem accepit)?° Farnham (no. 135), belonging to the Abbess of Shaftesbury, was held in 1086 by the wife of Hugh fitz Grip and Aiulf the chamberlain, and is recorded again among their manors.-' Little Cheselbourne (nos. 378 and ex), held by the wife of Hugh in 1086, had apparently belonged formerly to the Abbot of Abbotsbury. It was claimed that Hugh had held this land of the abbot ut homines eius dicunt sed abbas negat. The inference is that Hugh fitz Grip had taken possession of the property and then alleged that he had been enfeoff'ed by the abbot. At Abbotsbury itself (nos. 109 and Iviii) Hugh held one hide which T.R.E. ad victum monachorum erat and at Portesham (nos. 112 and Ixix) one virgate which erat in victu monachorum T.R.E. Both pieces of land were held by his wife in 1086. At Winterborne Houghton, a divided vill held partly by the wife of Hugh and partly by William of Moyon, the wife of Hugh held one virgate iniuste que pertinet ad Willelmum de Moione (nos. 392 and cxxv). William Rufus {filius regis) had taken 3 virgates at Stalbridge (no. 42) sine consensu episcopi et monachorum and had given them to Manasses.

At Swyre (no. 263) there was a piece of land which did not pay geld sed erat in dominio et infirma regis. The land had been let to Toxus T.R.E. by a king's reeve who later took it back into the king's hand. Toxus vero per regem Edzvardum iterum fuit saisitus sicut dicit et ita tenuit earn in vita et in morte regis Edwardi et tempore Heraldi.^^ A similar entry concerns a piece of land in Gillingham, which Hugh fitz Grip received from the king's farm and gave to Cranborne Abbey. ^-^ This land also was not assessed in hides. Half a hide at Cerneli (no. 212) belonging to the Count of Mortain was de dominica firma Cerne T.R.E.--^ Land held in pledge (vadimonium) is occasionally recorded. At Tarente (nos. 24 and xxx), a manor of Queen Maud, there was a virgate which Alvric, who held the land T.R.E., had in pledge for i mark of gold and necdum est redempta. At Blandford St. Mary (no. 261) William of Eu had i hide which Toli his predecessor had in pledge ef/z/zf adquietata, but Ralph de Limesi had taken it cum ista alia terra. Stock Gaylard (no. 269), another of William of Eu's manors, had been held in pledge by Toli de terra Scireburne. At Silton (no. 271), belonging to William of Falaise, there was a hide and i- virgate which Wulfweard White, the previous holder, had in pledge from one of his reeves, and attached to the same manor was a hide which Wulf- weard had bought from the Bishop of Exeter sed non pertinebat ad ipsum manerium. Eadnoth the staller, the predecessor of Hugh, Earl of Chester, had bought two manors, Catsley (no. 229) and South Perrott (no. 228), from Aelfwold, Bishop of Sherborne, on condition that at his death they should revert to the church, but Earl Hugh held them in 1086. Exchanges of land had also taken place. Five of the Bishop of Salisbury's manors in Dorset and one in Wiltshire were held in exchange for Scipeleia, which cannot be identified. The king had given the church of Gillingham to the Abbess of Shaftesbury in exchange for one hide of Kingston (no. 134) in which to build Corfe Castle. Hugh fitz Grip had given Little Waddon (no. 460) to Brictuin, a thegn, in exchange for a manor worth twice as much {ipsum scambium valet duplum). Although this manor is

^° Another part of Tatton (no. 345) was held by Aiulf he who gave evidence, he presumably held of William of

the chamberlain. T.R.E. i thegn had held it of Cerne Eu in 1086.

Abbey et non poterat ah ea separari. " See no. 70: Hanc terram accepit Hugo de firma regis et

^' See nos. 352 and 396. dedit ecclesie.

'^ The phrase sicut dicit appears to imply that Toxus ■=•» In the Geld Roll for Whitchurch hundred the Count

gave evidence. This seems unlikely, but it is difficult to see of Mortain had 1 hide which was de firma regis, which

who other than Toxus can be the subject of dicit. If it was seems to be this J hide in Cerneli: see pp. 125, 126.

A HISTORY OF DORSET

supposed to have been held by the Count of Mortain in 1086, it cannot be identified among his manors. From the Somerset survey it appears that the Count of Mortain gave to Athelnev Abbey the manor of Purse Caundle (nos. 1 18 and Ixiv) in Dorset in exchange for Bishopston (Montacute) in Somerset." This exchange is not recorded in the Dorset survey. Some of the tenants-in-chief in 1086 had been given their land by the queen. Anschitil fitz Ameline held Tyneham (no. 369) of the queen ut dicit sed post tnortem eius regetn tion requisivit. Dodo held \ hide, in an unspecified localitv, of the queen in alms (no. 444). Torchil held part of Hampreston (no. 443) which Schelin had held of the queen, but which in 1086 the king had in demesne {modo habet rex in dominio).^'' William fitz Osbern had apparently once held land in Dorset, since Waleran the huntsman had held Church Knowle (no. 308) of Earl William but modo ut dicit tenet de rege.

Land disputed between two tenants is sometimes entered under each tenant's name, but this only happens twice in the Dorset survey. Farnham is entered under Shaftes- bury Abbey (no. 135) and under Aiulf (no. 352) and the wife of Hugh fitz Grip (nos. 396 and cxxix), and the disputed virgate at Winterborne Houghton is entered under William of INIoyon (nos. 275 and Ixxxv) and the wife of Hugh fitz Grip (nos. 392 and cxxv). One manor, Blackmanston, belonging to Alvric, seems to be entered twice. One entry (no. 476) gives the name of the manor, the holder T.R.E., the hidage, and the teamland, but is unfinished. The other entr>' (no. 489) omits the hidage but adds the value. The fact that there are so few double entries makes it easier to calculate the assessment of the shire for geld. The total hidage recorded in Domesday amounts to 2,304 hides.-^ This can be compared with the hidage recorded in the Dorset Geld Rolls. There were 39 hundreds and, according to the figures given for the number of hides in each hundred, there should have been 2,298 hides. The details of each hundred account, however, do not always amount to the figure given for the number of hides in the hundred. The figures derived from the actual details of each hundred account yield a total of 2,307 hides, which is much closer to the Domesday figure. ^^ There was in addition a substantial amount of land which was not assessed in hides and not liable to geld. The six manors which had belonged to King Edward, and which were plainly very large, had never paid geld. There were 25I carucates (carucate) at Sherborne (no. 37) which had never paid geld, 16 belonging to the Bishop of Salisbury and g^ to the monks of Sherborne. The bishop had 2 carucates at Beaminster and 2 at Netherbury (nos. 46, 47), and 2 teamlands {qiiatitiim posmnt arare ii cariice) at Charminster (no. 32), 2 more at Alton Pancras (no. 33), 6 at Yetminster (no. 35), and i at Lyme Regis (no. 36). None of this land had ever paid geld. The monks of Sherborne had 2 carucates at Stoke Abbott (no. 45) which did not pay geld. The Abbot of Glastonbur)- had 14 teamlands at Sturminster Ne^\ton (no. 63) and 8 at Buckland Ne\\ton (no. 65), which w^ere exempt from geld. Aiulf the chamberlain had 4 carucates in demesne at Wootton Fitzpaine (no. 347). Some manors were beneficially hidated. Puddletown (nos. 8 and ii) was assessed at \ hide but had land for 15 ploughs. Okeford Fitzpaine (no. 64) was assessed at 8 hides, but had land for 16 ploughs and Stanton St. Gabriel (no. 210) was assessed at \ hide with land for 6 ploughs. Another part of Wootton Fitzpaine (no. 211) was assessed at 2 hides, but had land for 7 ploughs. The king's manor of Wimborne

" Dom. Bk. (Rec. Com.), i, f. 93- Aubrey, which must refer to the manor of Gussage St.

" The queen had enfeoffed Hugh fitz Grip in several Michael, which is in Dorset, but which is treated as part of

manors, all in the king's hand in 1086, and had probably Wilts, in Domesday; and 8 hides in Glochresdone hundred,

given the 2 manors of Edmondsham (nos. 353 and 354) which cannot be identified with any manor recorded in

to Humphrey the chamberlain. Domesday. When these 16 hides, 3 virgates, are deducted

" IMaitland {Dom. Bk. and Beyond, 505) gives this from Eyton's total, 2,304 hides, i virgate, remain,

figure as 2,321 hides, a figure apparently based on Eyton -* Eyton (op. cit. 144) gives these totals as 2,295 hides

(Ke)' to Domesday: Dorset, 144). Eyton's total includes and 2,301 hides respectively. 8 hides, 3 virgates, in Badbur)- hundred, belonging to Earl

DOMESDAY SURVEY

Minster (nos. 21 and xxvii), assessed at h hide, never paid geld although it did not belong to the night's farm of Wimborne, and other exemptions are occasionally recorded.^^ The four Dorset boroughs were assessed at a total of 45 hides, but they contributed to the upkeep of the royal housecarls and were not included in the total assessment of the shire.

There were 39 hundreds in Dorset at this period, which may suggest an original assessment of about 3,900 hides, but this seems unlikely. The hidages recorded in the various earlier charters which can be compared with Domesday assessments do not suggest any reduction in the hidage. The abbey of Shaftesbury, for example, received from King Alfred 100 hides, consisting of Donhead St. Andrew (Wihs.) and the manors which in 1086 formed the hundreds of Handley and Sixpenny. The hundred of Handley consisted solely of the manor of that name, assessed at 20 hides, the hundred of Sixpenny contained 53 hides, and Donhead St. Andrew was assessed at 40 hides, which is rather more than the total hidage given by Alfred, not less. 3° Only three Dorset hundreds contained more than 100 hides, Uggescombe with 104 hides, Beaminster with 105 hides, and CuUifordtree with 109 hides, while the tiny hundred of Redhone contained only 7 hides.

Several hundreds contained approximately 50 hides,^' and it is noticeable that some of these hundreds were later amalgamated. The Domesday hundred of Celberge (51 hides) became part of Winfrith hundred (49 hides), thus forming one unit of 100 hides; Stane (63 hides) was amalgamated with Modbury (63 hides); and Canendone (48 hides) became part of Badbury (32 hides). This suggests that there had been a division of the original hundreds (if indeed the Dorset hundreds ever did approximate to 100 hides) rather than a reduction in the assessment, which in any case is not indi- cated by any earlier evidence. Dorset was not included in the County Hidage, compiled earlier in the nth century, but the earliest text of the Burghal Hidage, dating from the early loth century,^^ includes Wareham, to which it assigns 1,600 hides, and Brydian (which may be identified with either Bridport or Bredy), to which it assigns 760 hides. These figures yield a total of 2,360 hides, some 56 more than the Domesday figure. The four Dorset boroughs in 1086 were assessed at a total of 45 hides, which would largely account for this discrepancy. Later texts of the Burghal Hidage omit Wareham and Brydian, but include Shaftesbury, to which they assign 700 hides. The Domesday hidage can also be compared with the actual amount of geld collected in 1084. At the end of the Dorset Geld Rolls it is stated that the king received ;(^4i5 8^. ()\d., the geld on approximately 1,394 hides. The money recorded in the individual hundred accounts amounts to ^^2,2 6s. ^hd., the geld on approximately 1,407 hides. About 900 hides were therefore exempt in 1084, which can be accounted for by the baronial demesnes, amounting to approximately 750 hides, and by the various exemptions and defaults, amounting to approximately 130 hides.

The system of assessment in Dorset shows traces of artificiality in the number of manors assessed at multiples or fractions of 5 hides. About one-fifth of all the manors in Dorset were assessed on this principle, and this figure can be broken down as follows :

2\ h. 5 h. 10 h. 15 h. 20 h. 25 h. 30 h. 21 44 25 5 5 I 2

In all, 103 out of 515 manors were assessed on a 5-hide basis, and in addition some divided vills add up to 5-hide units as is shown in Table i.

" See pp. 119-20. (49); Celeberge (41); Newton (47); Knowlton (36); Six-

'" See p. 42. penny (50); Brownshall (32); Winfrith (49); Celberge ($1).

3' Yetminster (47 hides) ; Albretesberge (47) ; Canendone " A. S. Robertson, Anglo-Saxon Charters, 246-8 and

(48); Badbury (32); Stane (63); Tollerford (59); Here nn.

A HISTORY OF DORSET

Edmondsham: s hides The king . . . .

Humphrey the chamberlain Eddeva of Humphrey

Farnham: 5 hides Aiulf the chamberlain Wife of Hugh fitz Grip . Odo fitz Eurebold . Aiulf the chamberlain

Nyland: 5 hides

Drew of the Count of Mortain. Ralph of Turstin fitz Rolf Bernard of Turstin fitz Rolf

Tatton: 5 hides

Aiulf

Wife of Hugh fitz Grip .

Mayne: 5 hides

William of Earl Hugh

The same . . . .

Beulf of Waleran .... Walter de Claville .... Walter of William of Briouze . Roger de Beaumont

Mappoivder: 7 hides, li virgate, 7 acres

Count of Mortain

Hugh of William of Eu .

Bollo the priest ....

Hampreston: 10 hides, J virgate The king . . . .

Aiulf the chamberlain William of the wife of Hugh Torchil . . . .

Table i Divided Vills Assessed on a yHide Basis

Shilvinghampton: 5 hides 2 h. Abbotsbury Abbey . . I h. 2 V. Edwin . . I h. 2 V. Count of Mortain .

Creech: 5 hides 2 V. Bretel of the Count of Mortain . 2 V. Roger de Beaumont 2 h. Walter of William of Briouze 2 h. Robert of the wife of Hugh

Warmzvell: 5 hides 2 h. Robert of the Count of Mortain

2 h. William of Earl Hugh

1 h. Turold of the wife of Hugh

Little Windsor: 5 hides

3 h. William of Moyon .

2 h. Hunger fitz Odin

Glanvilles IVootton: 5 hides

3 h. Ralph of William of Briouze

2 h. The same ....

1 h.

2h.

ih. 3 h. 2 V.

.3^v

Milton on Stour (in Gillingham).- 7i hides Roger of William of Falaise Godmund .....

Morden: 10 hides, i virgate Robert of the Count of Mortain Walter de Claville . Aiulf the chamberlain William of the wife of Hugh Ulvric . . . . .

Wife of Ulvric's brother .

Worth Matravers: 17 hides, 3^ virgates 2 h. IV. Roger Arundel ....

6 h. The same ..... I h. Robert of the wife of Hugh 3^v.

1 h. I V.

2 h. 2 V. I h. 1 V.

2h.

2h. 2 V. 2 V.

I h. 2 h. 2 V. I h. 2 V.

4h. I h.

3h. 2h.

7 a. 3 V. 5 h. 3 V.

3h. 4 h. 2 V.

ih.

3h. 2iv. 3 V.

1 h. I V.

2 h. 2 V. I h. oj V.

16 h. 2^ V.

2 V.

3 V.

The divided vill was quite a common feature in Dorset, where manors were mostly small and hamlets more common than vills. Sometimes only one part of a divided vill is recorded. William of Eu had i hide in Hiices (no. 258), but no one else is said to have any land there, and the place is not otherwise mentioned. Similarly, the Count of Mortain had 2 hides in Mannington (no. 186) and a mill and i hide of land in Sto- borough (no. 201), but neither place occurs again. William Malbank had 3 hides in Trill (no. 225), attached to Clifton Maybank, but Trill is not mentioned elsewhere. About a quarter of the manors in Dorset in 1086 were parts of divided vills. A large proportion of the manors amounted to no more than 2 or 3 hides, and even the large manors were not as extensive as those in other south-western counties. Sherborne (no. 37), the largest manor in Dorset, was assessed at 43 hides, Piddletrenthide (no. 69) and Sturminster Marshall (no. 232) were each assessed at 30 hides, and Sydling St. Nicholas (no. 93) was assessed at 29 hides. Apart from these only 11 manors were assessed at 20

10

DOMESDAY SURVEY

hides or over.33 Most of them belonged to ecclesiastical tenants, but Loders (nos. 13 and ix), assessed at 20 hides, had belonged to Earl Harold and was held by the king in 1086, Canford Magna (no. 243), assessed at 25 hides, belonged to Edward of Salisbury, and Broadwindsor (no. 505), assessed at 20 hides, belonged to Hunger fitz Odin.

After the assessment in hides is recorded the number of ploughs which could be employed on the manor. Occasionally it happens that the number of hides and the number of ploughs which could be employed are identical. At Frampton (no. 121) there were 25^ hides and there was land for the same number of ploughs {terra est totidem cariicanim). In a few cases the number of teamlands exceeds the number of hides, usually as a result of beneficial hidation. At Puddletown (nos. 8 and ii) there was land for 15 ploughs, but the manor was assessed at \ hide. Beneficial hidation is more usual in the case of ecclesiastical than lay land but two manors of the Count of Mortain, Stanton St. Gabriel (no. 210), and Wootton Fitzpaine (no. 211), were beneficially hidated. In the majority of cases, however, the number of hides exceeds the number of teamlands. This sometimes appears to affect the value of the manor. Stal- bridge (no. 42) was assessed at 20 hides, but had land for only 16 ploughs. It was worth ^iT,. Similarly Tolpuddle (nos. no and Ixvi), assessed at 18 hides, had land for 12 ploughs and was worth ^(^12 and Stour (no. 127), assessed at 17 hides, had land for 10 ploughs and was worth ^\o. On the other hand Piddletrenthide (no. 69), assessed at 30 hides, but with land for only 17 ploughs, was worth ^^30.

There is no discernible relation between the hidage of a manor, representing the geld assessment, and the number of teamlands, representing an estimate of agricultural capacity. The relation between the teamlands and the number of ploughs actually at work on the manor is likewise not constant. Sometimes their numbers coincide. At Dorchester (nos. 4 and xii) there was land for 56 ploughs and 56 ploughs were actually being used there. In some instances there were more ploughs than teamlands. At Chardstock (no. 49) there was land for 20 ploughs, but 21 ploughs were actually there, and the same figures apply in the case of Cerne Abbas (nos. 76 and xxxix). At Stockland (nos. 106 and Ixxx) there was land for 16 ploughs, but 22 ploughs were actually there, and at Abbotsbury (nos. 109 and Iviii) there was land for 16 ploughs, but 21 ploughs were actually there. It is worth noting that Abbotsbury was assessed at 21 hides. In most cases, however, the number of ploughs falls short of the number of teamlands. Some 180 manors had fewer ploughs than teamlands, as compared with 150 manors where there were equal numbers, and 24 with an excess of ploughs. A considerable number of entries (no), referring to the smaller manors, record teamlands but no ploughs. The values of these manors do not seem to be affected. Woolgarston (no. 297) was assessed at 2 hides and had land for 2 ploughs, and was worth j^z, although no ploughs are actually recorded. The teamlands are given for the manor as a whole, but the ploughs are divided into those in demesne and those held by the peasants. The question whether the villani alone held the men's ploughs or whether they were shared by all the peasants is discussed elsewhere.^*

In the absence of evidence to the contrary it is to be assumed that the ploughs in question were drawn by teams of 8 oxen. There is no mention in the Dorset survey of the number of oxen to a plough-team, although oxen are mentioned in the case of some small manors. Eight small manors, each assessed at i virgate, are said to have land for 2 oxen,35 and Wintreburne (nos. 387 and cxix), assessed at i| virgate, had land for 3

" Frampton (no. I2i); Canford Magna (no. 243); Broadwindsor (no. 505).

Milton Abbas (nos. 94 and Ixxiv) ; Sturminster Newton " See pp. 16-17.

(no. 63); Cerne Abbas (nos. 76 and xxxix); Abbotsbury " B^^'f (no. 348); Brigam (nos. 393 and cxxvi) ; Brige

(nos. 109 and Iviii); Loders (nos. 13 and ix); Stalbridge (no. 465); Rushton (no. 449); Tyneham (no. 473);

no. 42); Netherbury (no. 47); Handley (no. 125) and Woolcombe (no. 474); Wool (no. 487); Worgret (no. 497).

II

A HISTORY OF DORSET

oxen. Woodstreet (no. 508), assessed at 3 virgates, had land for 6 oxen. Brige or Brigam (nos. 348, 393 and cxxvi, 465) was divided into three manors of i virgate each, each of which had land for 2 oxen. But Lewell (no. 492), assessed at 3 virgates, had land for only 2 oxen, and Gillingham (no. 490), assessed at only \ virgate, also had land for 2 oxen. Apart from these two cases the evidence suggests that a manor assessed at i hide would be likely to have land for 8 oxen or i plough.

The value of each manor is generally given twice ; the first value being what it was worth at some time before 1086, the second what it was worth in 1086. In three cases the Exchequer text states that the earlier value relates to the time when the manor was received by the man who held it in 1086 {qtiando recepit). Exon. Domesday shows that this was the case in most entries, since it nearly always says that the manor was worth so much when its present owner received it. Sometimes Exon. departs from this practice. In the case of the manor of Puddletown Exon. in effect gives three values: Ex tempore regis Edzcardi hec mansio cum omnibus appendiciis suis reddidit per annum Ixxiii libras et quando Aiidfus {recepit) reddebat tantundem. In the cases of two manors belonging to the queen the previous values were the values in her lifetime; Cranborne (nos. 16 and xxii) and Ashmore (nos. 17 and xxiii) rendered respectively ;^24 and £1^ vivente regina. Similarly Chaldon (nos. 408 and cxli) valet per annum viii libras et quando Hugo recepit eam zalebat x libras et in zita Hugonis reddidit xi libras. For most of the manors of Cerne Abbey the previous value relates to the time quando abbas recepit, but at Rens- combe (nos. 91 and liv) this was defined as quando abbas W. recepit, and at Littlebredy (nos. 85 and xlviii) the former value was that tempore E. abbatis.

Exon. Domesday reveals that some manors were being held at farm, such as Child Okeford (nos. 7 and i) by Fulcred, and Loders (nos. 13 and ix) by Roger. Both manors had belonged to Earl Harold and were held by the king in 1086. Woodsford (nos. 82 and xlv) was held by Bristuard at farm of Cerne Abbey. The Exchequer text omits these farmers, although it records that 6 men held Ringstead (no. 463) at farm of Brictuin. It is also plain that Osmund the baker's manor of Galton (no. 507) was held for rent by the four men there, who paid 12^. 6d., since no 1086 value is recorded for the manor. Similarly Lyme (no. 36), belonging to the Bishop of Salisbury, was held by fishermen who paid 15*. ad pisces, but no 1086 value is recorded, and Ower (nos. 105 and Ixxix) was held by 13 salt-workers who paid 20^. In all three cases the money paid by the tenants is the only recorded income from the manor, and they must have held the manors at a money rent. The value of the Count of Mortain's manor of Nyland (no. 150) was omitted because it was waste {rasta est) and no value is given for Odo fitz Eurebold's manor of Petersham (no. 375). The values of about one-fifth of the manors seem to have been based on a figure of ^i a hide, and in a few other cases the previous value had been based on this figure, but had changed. The tendency was for values to rise rather than fall.

The effect of the teamlands on this valuation has already been shown. Other con- siderations also seem to have affected the value. In the case of Puddletown (nos. 8 and ii), assessed at i hide and with land for 15 ploughs, the value of £^2 must have taken into account the income derived from the third penny of the shire which was attached to this manor. Some of the manors which may be supposed to have been heads of hundreds were worth considerably more than one would expect, and it is possible that their values were affected by the profits from the hundred courts. At Sherborne (no. 37) the Bishop of Salisbur\''s demesne, consisting of 12 hides and 16 carucates which never paid geld, was worth £s'^- The bishop's demesne at Beaminster (no. 46) consisted of 6 hides and 2 carucates, and was worth ;Ci6. Loders (nos. 13 and ix), assessed at 18

12

DOMESDAY SURVEY

hides, was worth ^^33, and Frampton (no. 121), assessed at 25^ hides, was worth ^^40. The value seems to have represented the amount at which the manor could be leased. At Wraxall (nos. 328 and c) there were 3 hides worth £2 ^rid comparison with Exon. Domesday shows that 4 villani held this land for ^3 rent. Similarly the 3 thegns with

3 hides at Cranborne (nos. 16 and xxii) paid ^3. On the other hand, Brictuin, who held

4 hides worth ^^5 at Cerne Abbas (nos. 76 and xxxix), paid 30^.

It is difficult to compare earlier values with those of 1086, since there are several manors for which no previous value is recorded. In some cases it is legitimate to assume that the value had not changed, but in others this assumption would be misleading. No previous values are given for any of the estates of the New Minster, Winchester, Abbotsbury Abbey, Horton Abbey, Athelney Abbey, or the Countess of Boulogne. Only one or two of the manors of Milton Abbey, Tavistock Abbey, and Wilton Abbey have their previous values recorded, and previous values are given for only 4 of the Bishop of Salisbury's manors and 3 of the abbey of Glastonbury's manors. The previous value is given for the smaller of Serle of Burcy's manors, but not for the larger, and previous values are given for less than half the manors of William of Briouze. These fiefs have consequently been omitted from Table 2, except for those of the Countess of Boulogne and Tavistock Abbey which are supplied by Exon. Domesday.

Table 2 Comparative Values of Dorset Fiefs

Tenant-in-Chief	Previous Value	1086 Value	Hides	Vir gates
		£	i.	d.	£	J.	d.
Count of Mortain .	f	180	15	8	181	4	6	191	2
	<aii8	1 1	8
	U	62	4	0
Count Alan .		23	0	0	23	0	0	IS
Earl Hugh .	t	31 26 s	15 5 10	0 0 0	27	10	0	35	3
Roger de Beaumont		84	7	6	72	7	6	47	2
Roger de CourseuUes		9	0	0	8	0	0	5
Robert fitz Ceroid .		29	13	0	30	0	0	22	2
Edward of Salisbury		50	0	0	70	0	0	38
Ernulf of Hesdin .	U	9 8 I	10 0 10	6 0 6	12	10	6	15	3
Turstin fitz Rolf .	h	8 7 I	15 5 10	0 0 0	10	0	0	II	I
William of Eu		72	15	0	89	7	0	90	3
William of Moyon .	h	47 40 6	0 10 10	0 0 0	49	0	0	36	3
William of Falaise .		12	0	0	8	0	0	13	2
William of Ecouis .		14	0	0	■3	0	0	1 1
Walscin of Douai .		8	0	0	6	0	0	10
Waleran	t	36 25 II	5 0 5	0 0 0	37	15	0	38
Walter de Claville .		11	10	0	1 1	10	0	13	I
Baldwin		15	0	0	10	0	0	8
Berenger Giffard .		3	0	0	4	0	0	4
Osbern Giffard		I	0	0	I	0	0	2
Alvred of Epaignes		6	0	0	10	0	0	5
Matthew de Moretania		15	0	0	15	0	0	14	3
Roger Arundel	(:	44 40 4	17 7 10	6 6 0	52	17	6	65	2

13

A HISTORY OF DORSET

Table 2 {contd.)

Tenant -in-Chief	Previous	\ 'alue	1086 Value	Hides	Virgates
Aiulf the chamberlain	43 ■■ a 24 b 18	0 5 J5	0 0 0	60	0	8	55	ih
Humphrey the	r .2	0	0	13	10	0	10	■•■
chamberlain	[b 4	10 10	0 0
Hugh de Port	20	0	0	20	0	0	10
Hugh de St. Quintin	2	5	0	2	15	0	4	2i
Wife of Hugh fitz Grip	115 < aio3 [b 12	10 5 5	6 0 6	1 10	15	6	116	3
Bishop of Bayeux .	10	0	0	6	0	0	6	3
Bishop of Coutances	4	10	0	7	10	0	6	2
Bishop of Lisieux .	18	8	0	19	0	0	26
Cranbome Abbey .	26	0	0	23	0	0	21
Ceme Abbey	' 166 < aiig b 46	10 IS 15	0 0 0	166	15	0	121	2
St. Stephen, Caen .	52	0	0	52	0	0	33	2
Shaftesbury Abbey	115 ■ aioo \b 15	10 10 0	0 0 0	.46	0	0	167	I
Holy Trinity, Caen	9	0	0	14	0	0	10
Canons of Coutances	10	0	0	15	0	0	8
Abbey of Montevilliers	10	0	0	10	0	0	6
cTavistock Abbey	6	5	0	8	0	0	5
fCountess of Boulogne	' 6 <a 6 [b	15 0 15	0 0 0	6	15	0

a Previous \alue as given by Domesday, b Previous value deduced by assuming no change. c Supplied from Exon. Domesday.

The Domesday survey does not give any pre-Conquest details for the peasantry in Dorset as it does for some other counties. In 1086 the peasants in Dorset consisted mainly of villani and bordars, with smaller numbers of cottars and cotsets, 33 coliberti, and over 1,000 serzi. There were no sokemen or liberi homines as there were in the eastern counties, and no radknights as in some western counties, but 6 men {homines) held Ringstead (no. 463) at farm and there were 4 men {homines) paying 12s. ^d., apparently as rent, at Galton (no. 507). They may have been the same as the 4 free men {liberi hoinines) who held Galton T.R.E. Rent-paving tenants {censores) are recorded at Askerswell (nos. 1 19 and Ixv) where there were 2 paj'ing 15^., and at Allington (no. 253) where there were 9 paying 1 1^. Two French Serjeants {serrientes francigeid) are recorded at Cerne (no. 157) and 2 free Englishmen {Angli liberi) have 4 hides at Handley (no. 125). A smith {faber) is recorded at Melbury Osmond (no. 183), and two priests are recorded along with the peasants, one at Church Knowle (no. 235) and one at Bleneford (no. 455). Table 3 shows the numbers of all the various classes of peasants in Dorset in 1086, with the corresponding figures as calculated by EUis^^ given in brackets.

In Dorset, as in some other counties, the rillani are outnumbered by the bordars.^^ They are clearly distinguished from the cottager class, and were obviously more prosperous. Robert, Bishop of Hereford, in his description of the Domesday survey,^^ says that it was concerned both with the cottagers {in tuguria tautum habitantibus) and with those who had their homes and a share in the fields {in domos et agros possidentibiis).

" H. Ellis, Gen. Introd. to Dom. Bk. ii. 419-514. The population totals for each county are currently being reckoned afresh in the Domesday Geography series, edited by Prof. Darby: see Domesday Geog. of Eastern Eng. (1952) ; Domesday Geog. of Midland Eng. (1954) ; Domesday Geog. of SE. Eng. (1962) ; and Domesday Geog. of Northern

Eng. (1962). The SSV. volume, covering Dorset, has not yet been published.

■" In Comw., Hants, Worcs., Essex, Norf., and SufF. : Ellis, op. cit. ii. 432, 441, 449-50, 469-70, 488-90, 504-6.

>» E.H.R. xxii. 73-74.

'4

DOMESDAY SURVEY

(I)	(2)
2,947	(2,941)
2,636	(2,613)
1,161	(1,231)
207	(209)
204	(188)
33	(33)
1 1	(II)
10	(4)
3	(3)
2	(2)
2	(5)"
I	(i)

Table 3 Peasantry in Dorset in 1086"

Class Bordarii Villani

Servi .... Cosces .... Cotcirii Coliberti

Censores (gabulalores in Exon.) Homines Ancille

Servientes Francigeni Presbyteri Faber ....

ToT.'iL . . . . . 7,217 (7,241)

a Column (i) lists the totals for each class as calculated by the author from the figures given in the Dorset Domesday survey, and column (2) the totals for the same classes as calculated by Sir Henry Ellis in his General Introduction to Domesdav Book (1833), ii. 438.

b This number appears to include the 3 priests recorded at Hinton (no. 31), who were tenants of the king.

The latter must be the villani. As the more prosperous class, they are usually enumer- ated first. 3'' Their exact status and the actual composition of the class as a whole are difficult to establish, but they were clearly not serfs. Although the word villanus in Domesday may conveniently be translated as 'villein', it had not by that date acquired the connotation of someone unfree that it had in later centuries. In 1086 it meant simply a man who lived in a vill, and was equivalent to the Anglo-Saxon word tunesman.-^° The class of villani must have included men who would have been described as geburs in pre-Conquest documents, but this is not to say that the two classes were coterminous. Men like the 4 villani holding 3 hides of land at Wraxall (nos. 328 and c) for rent could not have been classified as geburs, and among the villani of Domesday there must be included men who at one time had been free ceorls but had become economically dependent on a Norman lord. There is evidence that men whom pre-Conquest docu- ments would have called geneats (and Domesday itself, in some instances, radknights) were sometimes included among the villani.^^ That some similar change had taken place in Dorset is suggested by a comparison of the Domesday description of Iwerne Minster (no. 131), belonging to Shaftesbury Abbey, and the description of the same manor preserved in a survey of the abbey's land about 1130. The survey states that the chaplain of Iwerne Minster had de imoquoque genet i daiwenie ambrani. The Domesday description of the manor records only 29 villani and 20 bordars. The evidence of the later survey suggests that some of these men were, or had been, geneats.-*^ In Dorset many manors were very small and were held by quite large groups of thegns in 1066, and it is plain that some of these thegns can have been hardly more prosperous in economic terms than the villani. In two instances it seems almost as if the pre-Conquest thegns or their heirs had survived as dependents of a Norman lord and were classed as villani in Domesday. Kingcombe (no. 247), assessed at 34 virgates, was held by 5 thegns T.R.E. In 1086 it belonged to Ernulf of Hesdin. He does not seem to have had any demesne there, and the only peasants were 5 villani, holding a plough. Another of Ernulf's manors. North Poorton (no. 249), assessed at \ hide, was held by 7 thegns

^' They do not always come first. At Beaminster (no. 46) there were xix bordarii et ii villani et ii coscez ; at Wai (no. 163) Hi cosces citni uno villano habent imam carucam; and at Spetisburj' (no. 173) ibi est iinus bordarius et iinus villanus. But in general the villani precede the bordars,

cotsets, and cottars.

■•» V.C.H. Hunts, i. 324.

•" F. M. Stenton, Anglo-Saxon Eng. 471.

« B. M. Harl. MS. 61, ff. 4SV-46.

15

A HISTORY OF DORSET

T.R.E. In this manor, too, Ernulf does not seem to have had any demesne, and it was occupied by 7 villani with a plough. It is tempting to see here small, impoverished thegns, seeking the protection of a Norman lord, and being classified by the Domesday commissioners on the basis of their small holdings and not that of their wergild or social status/-^

The Domesday survey of Middlesex gives details of the land held by the peasants, +4 but the Dorset survey is not so informative. Exon. Domesday, however, supplies a little information. At Winterborne Houghton (nos. 275 and bcxxv) a certain villaniis, whom the Exchequer text does not mention, held \ virgate, and at Tarente (nos. 400 and cxxxiii) a villarnis qui niauet held a virgate and 1 plough. At Burcombe (nos. 115 and Ixii) there were 2 villani qui tenent illam terram, assessed at \ hide. Tenements of a virgate and \ virgate were common on the estates of Shaftesbury Abbey in the 1 2th century, and on the estates of Peterborough Abbey at the same period villani with a virgate were called pleni villani and villani with \ virgate dimidii villani or semi villani y^ At Swanage (nos. 515 and xxxviii), belonging to the Countess of Boulogne, a single villaniis appears to have held the whole manor, assessed at i hide and ^ virgate, with a plough. The countess had no demesne in this manor in 1086.+^ Neither Exon. Domesday nor the Exchequer text has anything to say about such services as weekwork or ploughing. Villani paving money are mentioned only once, at Wraxall (nos. 328 and c) belonging to Roger Arundel. William held 3 hides at Wraxall of Roger Arundel which, according to Exon. Domesday, were held by 4 villani for £t, degablo.*'^ In view of the large amount of money involved it seems possible that the villani were holding the land at farm. Although this circumstance is rare, four cases are definitelj' recorded in Domesday, including two in the neighbouring county of Devon, at Herstanhaia and Lympstone.*^ In addition to these instances in Devon, two manors in Hampshire, Alverstoke and Millbrook, belonging to the abbey of Winchester, were held by the villani and may have been at farm, and in Surrey, Clandon, belonging to Chertsey Abbey, was held by the villani for a money rent.-*^

Information about the ploughs held by the villani is even scantier than that about their land. It is not certain whether the men's ploughs were held by the villani alone, or whether they were shared among all the peasants. The evidence on this point is vague and contradictors'. The formula emploved by Exon. Domesday would at first sight imply that only the villani had ploughs. A typical entry, for Abbotsbur}' (nos. 109 and Iviii), runs: habet abbas viii hidas et v carrucas in dotninio et villani xxiii hidas et xvi carrucas. Ibi habet abbas xxxii villanos et xvi bordarios et xiiii servos. This language seems to exclude the bordars from a share in the land and ploughs. But from other entries in Exon. it appears that the term villani was employed in two ways, both to designate the villani themselves, and to mean the whole group of peasants as opposed to the lord. At Cruxton (nos. 279 and Ixxxix) villani (habent) i hidam . . . et i carriicam, but there were no villani, only 9 bordars and a servus. In this case it is obvious that villani means simply the men, and this is probably the meaning throughout. =° The

*> On the likelihood that some such process had taken ■** Dom. Bk. (Rec. Com.), iv. 371, 425. The other cases

place, see F. M. Stenton. 'Eng. Families and the Norman are Oare (Kent) and Willesden (Mdx.): ibid, i, ff. 10,

Conquest', Trans. R. H. S., 4th ser., xxvi. 7-8. 127b. For a full discussion of the question, see R. S. Hoyt,

" See ]'.C.H. Mdx. i. q2. 'The Farm of the Manor and the Community of the Vill",

■"5 B. M. Harl. MS. 61, ff. 37-89 (Shaftesbun.) ; Citron. Speculum, xxx. 147-69; cf. R. Lennard, Rural Eng. 153-4.

Petroburgense (Camd. Soc. xlvii), 157-83 (Peterborough). " Dom. Bk. (Rec. Com.), i, f. 41b; I'.C.H. Hants, i.

*^ Count Eustace was credited with 1 hide and J virgate 442; V.C.H. Suss. i. 367-8; V.C.H. Surr. i. 290-1.

in demesne in Ailetestcode hundred which must refer to so According to Maitland (Dom. Bk. and Beyond, 38-39)

Swanage, but in 1086 it is simply stated that King William 'the term villanus may be used to cover the whole genus as

never had geld from the manor: see pp. 136, 137. well as to designate one of its three species'. Exon. once

■"■ The Exchequer text says merely that the land was uses the word rustic! in this general sense and once uses rus-

worth £3 and that there were 4 villani. ticus where it would normally use lillanus : see pp. 76, 106.

16

DOMESDAY SURVEY

Exchequer terminology is not helpful since it says merely that so many villani and so many bordars had so many ploughs, without indicating how the ploughs were divided. Two entries distinctly say that the ploughs were shared among all the peasants. At Burton Bradstock (nos. 2 and x) there were 41 villani, 30 bordars, 7 coliberti, and 74 cottars, and inter omnes habent xxvii carucas, and at Wai (no. 163) Hi coscez cum uno Tillano habefit imam carucam. On the other hand, two entries specifically assign the plough to the villani alone. At Corscombe (no. 488) there was unus villanus cum i caruca et an bordariis et i servo, and at Stoke Abbott (no. 45) ibi est in dominio i caruca cum i servo et vi coscez. Ibi viii villani habent iiii carucas. At Woolcombe (no. 262) the bordars but not the cottars seem to be sharing in the ploughs {ii villani et viii bordarii cum i caruca et Hi cotarii). With this basic uncertainty it is obviously difficult to decide what fraction of a plough was held by the average villanus. Only the total number of ploughs is given and there is no guarantee that they were shared equally among the villani even if the share of the bordars and cottars could be satisfactorily established. In 19 cases the villani were either the only peasants or can be shown to have held all the ploughs. In 10 of these cases the proportion of villani to ploughs is exactly or approximately 2:1, so that it is possible that each villanus held about J plough. In 5 cases the proportion is exactly or approximately 4:1. There are 3 cases of a villanus with a whole team (nos. 182, 488, and 515) and at Gillingham (no. 250) there is the extremely rare case of a villanus with two teams of which there are only two other known examples, at Haiugurge (Suss.) and Keresforth (Yorks. W.R.).5'

The bordars, cotsets, and cottars, who are enumerated after the villani, together form the cottager class, corresponding to the kotsetlan of the Rectitudines Singularum Personarum. The bordars are by far the most numerous, and usually take precedence over the cotsets, who in turn take precedence over the cottars.^^ Since the Domesday commissioners took the trouble to distinguish three classes of cottagers, there must have been some diflFerence between them, but the nature of this difference is now unknown and there is evidence that the distinction was not clear-cut even in the nth and 1 2th centuries. In the summary of the fief of Glastonbury Abbey in Exon. Domesday it is stated that the abbot had 72 bordars, whereas according to the Ex- chequer text he had 40 bordars and 32 cottars. The Exon. figure thus lumps the bordars and cottars together without distinction. s^ It is noticeable, also, that the Shaftesbury surveys of the 12th century record only cotsets, whereas in all but one case the Domesday account of the manors of that abbey mentions only bordars.'-* Round noted that in the accounts of Surrey and Sussex the cottager class were called bordars in some areas and cottars in others, the terms being mutually exclusive." The Middlesex survey, on the other hand, which is the only one to record individual peasant holdings, suggests that there was a distinction between bordars and cottars based on the size of their holdings. Whereas most bordars in Middlesex had tenements of between 5 and 15 acres, the majority of cottars had no land at all or only their gardens. Those cottars who did have land had as a rule between i acre and 5 acres, only two having more than this. Only 54 out of 342 bordars had less than 5 acres, and only 9 had no land at all.s^ Whether this distinction holds good in Dorset is conjectural, since there are no details of individual holdings, but a distinction of this sort can be seen in the survey of

5' Dom. Bk. (Rec. Com.), i, ff. 2ib, 317; see also R. " Seep. 148.

Lennard, 'The Economic Position of the Domesday " Qn one occasion the Shaftesbury surveys do use the

Villani', Econ. Jnl. Ivi. 260. term 'bordar'. This is in the account of Melbury Abbas

" It is unknown in Dorset for all 3 types to be found (B. M. Harl. MS. 61, f. 48V), which is the only Shaftesbury

together in a single entry, but when bordars are found with manor where cotsets are recorded in Domesday (no. 130).

cotsets or cottars they are enumerated first, and when " V.C.H. Siirr. i. 292; V.C.H. Suss. i. 368.

cotsets are found with cottars the cotsets come first. '^ V.C.H. Mdx. i. 92.

DO. Ill 17 *

A HISTORY OF DORSET

Hinton St. Mary, belonging to Shaftesbury Abbey, in the later I2th century. There the cotsettle, with 5 acres each, are distinguished from the cotarii, who had only their houses and gardens. ^^ The distinction between bordars and cotsets is, if anything, more problematical. It has been suggested that they were alternative names for the same group, bordar (from borde, meaning a wooden hut) being the French equivalent of the Anglo-Saxon kotsetla or cotset, but since bordars and cotsets occur together on two Dorset manors, this cannot be so.^^ Such information as there is about the holdings of the bordars comes from Exon. Domesday. At Cerne (nos. 108 and Ixxxii) 7 bordars had a virgate and 5 acres, at Waia (nos. 380 and cxii) 6 bordars had i hide, and at Nutford (nos. 28 and xxxiii) there were 8 cotsets, two of whom held 8 acres, 5' while the third was apparently landless. There are no entries crediting the cottars with any land. In several cases the statement that the villani have so much land is followed by information about bordars or cotsets only. At Wintrebiirne (nos. 103 and Ixxvii), for example, the rillani are said to have a hide and a virgate, but there were in fact two bordars. ^° There is one striking entry concerning bordars (no. 480) where two of them held a \ virgate which they themselves had held freely T.R.E.: Duo bordarii tenent quartam partem uniiis virgate terre. Valet xv denarios. Ipsi libere temierunt T.R.E. The locality is not given, but it may have been part of Stourton Caundle which is the subject of the preceding entry. These two bordars do not appear to have been holding of any lord other than the king, and though their holding is very small it is not unique in this respect, since Al- ward, a thegn, held J virgate (perhaps part of Wilksworth) rendering 30^/. (no. 446) as a manor. There is nothing to suggest that the two bordars were not holding as freely in 1086 as in 1066.

There is even less information about the rents and services of the bordars. Money rents are recorded twice. At Langton Herring (nos. 23 and xxix) there were eight bordars, one of whom paid 30^., and at Lewell (no. 492) two bordars paid 2od. There is no reference to weekwork, ploughing, or any other services. The question of the bordars' share in the men's ploughs has already been discussed. On manors where there were no villain ploughs were held by bordars and sometimes cotsets, but never by cottars. At Frampton (no. 121) 24 bordars and 7 cottars had 14 ploughs, but whether the cottars shared the ploughs with the bordars or not cannot be determined.^' At Wintrebiirne (no. 179) 7 cotsets had \ plough. At Shilvinghampton (no. 216) there was land for i plough que ibi est cum i coscet, and at Moleham (no. 511) there was land for i plough que ibi est cum i cotario. It is hardly likely in these cases that each held a full team. Probably the ploughs belonged to the lord, and the bordars were merely in charge of them in the absence of servi. There are indications that bordars were some- times attached to the demesne ploughs. At Worth Matravers (nos. 332 and civ) there was land for h plough que ibi est cum Hi bordariis, and the corresponding Exon. entry shows that the plough belonged to the lord and not to the bordars ; ibi habet Rogerus dimidiam carriicam et Hi bordarios. At Creech (nos. 412 and cxlv) there was land for \ plough que ibi est cum iiii bordariis, and the Exon. entry shows that this \ plough was in demesne.

After the villani and bordars the largest group of peasants was that of the servi, of whom there were 1,161 in Dorset in 1086. Large numbers oi servi vftre found in all the

" B. M. Harl. MS. 6i, ff. 65V-66. estates of Shaftesbury Abbey in the 12th cent.

" H. P. R. Finberg, Tavistock Abbey, 61-62; see nos. ''" Forsimilar examples, see nos. 279 and Ixxxix, 329 and

37, 46. The Peterborough survey records cotsets in some ci, 376 and cviii, 401 and cxxxiv, 402 and cxxxv. 405 and

places and bordars in others, but both are recorded at cxx.xviii, 410 and cxliii.

Fiskerton (Lines.), where there is ; plenus cotsetus et in " For examples of bordars holding ploughs, see nos.

bordarii: Chron. Petrohuraense, 164. 158, 169, 253, 272, 279 and Ixxxix, 303, 338, 346, 376 and

" Four acres was the normal holding for a cotset on the cviii, 401 and cxxxiv.

ig

DOMESDAY SURVEY

south-western counties, although this area was not unique in that respect. It is true to say, however, that in 1086 servi were most numerous in the area of the old West Saxon kingdom.^2 Apart from the male servi three female ancille were recorded at Crichel (no. 266).^^ It is probable that most of the servi recorded in Domesday were ploughmen. They are usually recorded in conjunction with the number of ploughs in demesne, and it is not uncommon to find a ratio of 2:1 between the servi and the demesne ploughs. "^-^ Where the number of servi falls short of the amount necessary to yield such a ratio, the deficiency can sometimes be supplied from some other class, such as the bovarii in Warwickshire, whose name indicates that they were ploughmen. ^5 No bovarii are recorded in Dorset in 1086, but in one instance the coliberti seem to occupy a similar position. At Sturminster Newton (no. 63) there was in demesne land for 14 ploughs, although the entry does not say how many ploughs were actually there. There were only 15 servi but in addition there were 13 coliberti, whose number combined with that of the servi yields a ratio of 2 : i with the number of ploughs which could be in demesne. That the coliberti were sometimes linked with the demesne ploughs in the same way as the servi was demonstrated by Round in his introduction to the Somerset Domesday. ^^ Their name suggests that the coliberti were freed servi. They occur only in the area of Wessex and western Mercia and appear in considerable numbers in some counties, though there were only 33 in Dorset. Apart from the 13 already mentioned at Sturminster Newton they were all on manors which had belonged to King Edward, 12 at Dorchester, 7 at Burton Bradstock, and i at Pimperne. At Dorchester and Pimperne they were grouped with the demesne ploughs and the servi, but at Burton Bradstock they were placed between the bordars and the cottars and the wording of the entry does not exclude them from a share in the men's ploughs. ''^ In another instance a cotset seems to be associated with the servi and the demesne ploughs. At Uploders (no. 206) there was land for 2 ploughs que ibi sunt cum i coscet et Hi servis.^^

Comparison between Domesday and the surveys dating from the 12th century indicates a decrease in the number of servi after 1086. The earlier survey of the lands of Holy Trinity, Caen, records servi at Felstead (Essex) and Horstead (Norf.), but at Tarrant Launceston in Dorset, where there were 14 servi in 1086 (no. 141), no servi are recorded at all, and there has been a corresponding increase in the number of bordars. ^^ In 1086 there was only one bordar but in the 12th century there were 13 bordars together with a smith and a shepherd. It has been suggested^o that the successors of the servi were to be found among the bovarii and bubulci who figure so largely in the surveys. The Shaftesbury Abbey surveys^^ do not mention servi at all, with the possible exception of a cliens qui servit in aula at Cheselbourne.^^ There are, however, numerous references

" In 1086, according to Ellis {Gen. Introd. to Dom. Bk. " Ec. H. R. Suppl. ii. 7-8.

ii), there were 995 servi in Norf., 905 in Suff., 1,768 in '''> V.C.H. Som. i. 426.

Essex, and 1,148 in Kent. But the largest figures occur in "" See nos. 2 and x: Ibi sunt xli villani et xxx bordarii et

the SW. : over 3,000 in Devon, over 2,000 in both Som. vii coliberti et Ixxxiiii cotarii. Inter omnes hahent xxzni

and Glos., and over 1,000 in Cornw., Dors., Wilts., and carucas.

Hants. Prof. Darby gives the figures for Norf., Suff., and <"" For cotsets and cottars associated with demesne

Essex as 971, 917, and 1,788 respectively {Domesday Geog. ploughs, see above.

o/i'asfern £«,?. 169, 225), and for Kent as 1,160: Domes(/oj' '"'' Bibliotheque Nationale, MS. Latina 5650, ff. 26v,

Geog. of SE. Eng. 513. 27V, 28V. For an analysis of this MS., see Jean Birdsall,

'•^ Tfiis is the only mention of ancille in the Dorset 'The Eng. Manors of the Abbey of La Trinite at Caen',

Domesday, but in the survey of the manors of Holy Anniversary Essays . . . by Students of C. H. Haskins, ed.

Trinity, Caen, dating from temp. Hen. I, 3 ancille are C. H. Taylor, 25-44. The earlier survey, containing the

recorded at Tarrant Launceston, of whom 2 were dead and account of Tarrant, seems to date from the reign of Hen. L

the third living (harum ii sunt mortue alia vivit) : Biblio- "> Ec. H. R. Suppl. ii. 8 sqq.

thfeque Nationale, MS. Latina 5650, f. 27V. " These 2 I2th-cent. sur%'eys are preserved in the

"■t M. M. Postan, 'The Famulus', £•<:.//. fi. 5»/>/)/. ii. 6. Shaftesbury Abbey cartulary: B.M. Karl. MS. 61, flF.

In Dorset there were 312 manors on which servi were 37-89. The earlier one dates from c. 1130, the later from

recorded, and on 66 of these (c. |) the ratio between them c. 1 175-80.

and the demesne ploughs was 2:1. '^ Ibid. f. 4SV.

19

A HISTORY OF DORSET

to bubulci, who, as the context shows, were ploughmen. At Handley bubiilci qui tenent carucam habent singuU vi acras quietus pro suo servicio. At Tisbury (Wilts.) there were 1 1 bubulci, to whom datur cibus in natale Domini et in die Pasche et quando educant carros. Each of the men held 4 acres and tefiet carucam et de ipsa arat suam terram. At Chesel- bourne the bubulci had 5 acres each (quit of all dues except geld), a piece of land, and a beast quit of herbage ^ro utensilibus caruce quorum parteiii emunt de suo et partem accipiunt in nostram sikamJ^ At Compton Abbas and Melbury Abbas the cotsets appear to have taken part in ploughing the demesne. On both manors the cotsets ad carucas educendas habent dimidiam ambram et dimidiam multonemJ^

After describing the peasants Domesday gives an account of the non-arable appurtenances of the manor, such as meadow, pasture, and wood. Meadow (pratum) was an important adjunct, since, in the absence of root crops, hay was the staple diet of farm animals in wintertime^? Out of 515 manors recorded in the Dorset survey 418 had associated meadow, although only 63 manors had more than 25 acres. Occasionally the meadow lav at some distance from the manor to which it was attached. The Bishop of Salisbury had 130 acres of meadow attached to the manor of Sherborne (no. 37), 3 acres of which were in Somerset tuxta Meleburne. At Bingham's Melcombe (no. 30) 12 acres of meadow had been leased to Wlgar Wit T.R.E. and were held by William Belet in 1086, while at Rushton (nos. 407 and cxl) the wife of Hugh fitz Grip had subinfeu- dated the entire manor to two knights, retaining in demesne only 16 acres of meadow. Meadow was naturally located on or near rivers and streams, the largest concentrations lying along the valley of the Stour. Pasture {pastura, pascua), though less common than meadow, occurs in connexion with 366 manors and was spread over a larger area. One entr\' records an encroachment of arable land on pasture. At Swyre (no. 263) there was a piece of land which prius erat pascualis modo seminabilis. This land had been leased T.R.E. to Toxus the priest by a king's reeve, just as the meadow at Melcombe had been leased to Wlgar Wit. At Spetisbun,- (nos. 274 and Ixxxiv) there was a piece of pasture 2\ furlongs long by 2 furlongs wide, and in alio loco super aquanf^ another piece 2\ furlongs long by i-i furlong wide. A similar entr}', concerning Tarrant Rawston(nos. 404 and cxxxvii), records a piece of pasture 3 furlongs by 2 furlongs and in alio loco another piece, measuring 8 furlongs. At Mapperton (no. 137) there were inter pasturam et silvam xi quarentine lojigitiidine et tantundem latitudine. This may mean a piece of grassland with scattered clumps of trees, or it may refer to the use of woodland for grazing. Woods [sika, ne?nus) were used to pasture pigs^'' as well as to provide fuel and timber for building houses and barns. ''^ At Stoke Wake in the 12th century Wulfric the priest had 10 pigs free of pannage in the Abbess of Shaftesbury's wood.'^' Wood which did not provide acorns and beechmast on which pigs could feed was called silra in- fructuosa, like the wood at Renscombe (nos. 91 and liv). According to Exon. Domesday, the wood at Nettlecombe (nos. 88 and li) nullum fructum fert, although the Exchequer text does not say that it was infructuosa. Silra minuta and silva modica are both recorded occasionally in the Dorset survey. What precisely these terms implied is now difficult

•' B.M. Harl. MS. 6i, ff. 45V, 55; cf. Iweme Minster, MS. 61, f. 46V.

Orchard, and Stoke Wake; ibid. ff. 47, 52, 52V. " In some counties the Domesday suney gives the

'•• Ibid. fF. 48V, 49V-50. extent of woodland in terms of the number of pigs it could

" In the second survey of Shaftesbury .-Abbey's lands suppon: H. C. Darby, 'Domesday Woodland', Ec. H. R.

carting hay was a common obligation of virgaters and half- N.s. iii. 23 ; cf. Laws of Ine, 44, 'a tree that can shelter 30

virgaters. At Iweme Alinster the virgaters carried J cart- swine' : Latcs of the Earliest Eiig. Kings, ed. and translated

load of hay from Combe, and at Fontmell Magna they had F. L. .^ttenborough, 51.

to find 20 mowers for the meadow iiixta Saticliim Ad- "* Thatched with stubble (stipula). In the early 12th

wardiim: B.M. Harl. MS. 61, ff. 65, 67. cent, the peasants of Cheselboume reaped the stubble ad

'"' This reference to pasture on or near water is similar domos cooperindos (sic): B.M. Harl. MS. 61, f. 44V.

to an entry in the earlier Shaftesbury survey which '° Ibid. f. 52. mentions pascua de mareis at Iweme Minster: B.M. Harl.

20

DOMESDAY SURVEY

to determine, but they may both be translated as 'underwood'. ^o Marshland (broca) is recorded at Lytchett Matravers (no. 260) and a league of marshland at Wimborne Minster was attached to the manor of Canford Magna (no. 243). Heathland (bruaria) is recorded at Boveridge (no. 71) in Cranborne. The main concentrations of woodland in 1086 were in the areas of the later forests.^' Although forests were not necessarily wooded areas, they frequently comprised large stretches of woodland. The only Dorset forest mentioned in Domesday is the forest of Wimborne, in which the king held the two best hides (dims meliores hidas) of Horton (no. 1 17) belonging to Horton Abbey. The wood (boscus) of Hauocumbe was attached to the manor of Burton Bradstock (nos. 2 and x), the third oak {qiiercus) being appurtenant to the manor of Frampton (no. 121).

Pasture and woodland were generally measured in leagues and furlongs {leuge, quarentine). At Handley (no. 125), for example, there was woodland i league in length and \ league in width. It is difficult to see exactly what is being measured. The figures cannot in any case be more than a rough estimate of the extent of the wood or pasture, and cannot be taken as a reliable guide to the shape of the land in question. Other units of measurement sometimes used are perches {pertice) and virgates {rirgate). At Poxwell (nos. 81 and xliv) there was pastiira ziii quarentine et xxvi virgate longitudine et in quarentine et xiiii pertice latitudine, and at Symondsbury (nos. 92 and Iv) there was pastura v quarentine longitudine et una quarentina latitudine x virgatas minus. The acre is sometimes used as a linear measurement. At Wootton Fitzpaine (no. 211) there was vii quarentine et iiii acras (sic) pasture. It is also not uncommon to find only one measure- ment given. At Hinton Martell (no. 31) there was una leuga stive, and at Little Puddle (nos. 14 and iii) x quarentine pasture. At Dewlish (no. 148) there was woodland ri quarentine in longitudine et latitudine and pasture xxiii quarentitie inter longitudinem et latitudinem . It is uncertain what these phrases mean. They may be equivalent to the iormvXvi pastura iii leuge longitudine et tantundem latitudine (nos. 6 and xv), or they may represent an attempt to use the furlong as a square measure. A fresh difficulty is created by the fact that there is no indication how many furlongs there were in the league. Round considered that in Worcestershire there were four, since three furlongs was the largest measurement under the league. But at Shillingstone (no. 367) in Dorset there was pasture 42 furlongs long and 8 furlongs wide, and woodland 23 furlongs long and 9 furlongs wide. One is left with the problem of which is the larger; pasture 42 furlongs by 8 furlongs or pasture 4 leagues long and as much in width. This uncertainty makes any comparison of the relative sizes of stretches of pasture or woodland extremely difficult, and the situation is complicated by the fact that both pasture and woodland are sometimes measured in acres, especially in the case of underwood {silva minuta). It is impossible to say whether 140 acres of pasture (no. 223) is more or less than the amounts measured in leagues and furlongs. ^^

Among the other manorial assets mills were the most important. There were 276 mills in Dorset in 1086, attached to 178 manors. Some large concentrations are recorded; 12 at Dorchester (nos. 4 and xii), 12 at Sherborne (no. 37), 8 at Burton Bradstock (nos. 2 and x), and 8 at Wimborne Minster (nos. 3 and xi). In addition, 3 manors had 4 mills each, 10 manors had 3 mills each, and 30 manors had 2 mills each. Fractions of mills are sometimes recorded. There was half a mill at Worgret (nos. 84 and xlvii) belonging to Cerne Abbey. Worgret was a divided vill, and the other half of this mill belonged to the manor held by William of Briouze (no. 293). At Child Okeford (no.

*° Prof. Darby so translates them at Ec. H. R. N.s. iii. ham, Blackmore, and Cranborne Chase.

37, and in the Domesday Geography series. '^ The amount of meadow is almost always expressed

" F. W. Morgan, 'Domesday Woodland in SW. Eng.' in acres, but for meadow measured in furlongs, see nos.

Antiquity, x. 316. The forests were Powerstock, Gilling- 110, 146, and 380.

DO. Ill 21 B2

A HISTORY OF DORSET

I ^2) the Count of Mortain had half of two mills {medietas ii moUnorum) rendering 10s. Part of Child Okeford (nos. 7 and i) was held by the king, who held there two mills rendering 205. of which, according to Exon. Domesday, medietatem partu (sic) habet rex. At Tarrant Crawford (no. 436) Alvric had a \ mill rendering 30^. Three quarters of a mill rendering 95. were recorded at the unnamed manor (no. 494) belonging to William de Dalmar, which is presumably why Eyton identified this manor as part of Tarrant Crawford.**^ Both Alvric's manor of Tarrant Crawford and William de Dalmar's unnamed manor can with some confidence be assigned to the hundred of Celeberge^^ so that Eyton may well have been correct. One third of a mill is recorded at Winbiirne (nos. 388 and cxx) belonging to the wife of Hugh fitz Grip. It is possible that this was part of the mill recorded at Hervey the chamberlain's manor of Wimborne St. Giles (no. 499) where the entry reads in molino ville xxii et dimidia and then breaks off leaving a space. At Morden (no. 473) Ulvric had \id. de parte molini, perhaps part of the mill at Morden (no. 172) held by the Count of Mortain. There was also h mill at Watercombe (no. 29) and another J mill at Ringstead (no. 359), each rendering 45. The other halves cannot be traced. The mill of Stoborough (no. 201), belonging to the Count of Mortain, had h hide of land and 3 bordars attached to it. The mill at Povington (no. 242), belonging to Robert fitz Gerold, was claimed ad opus regis. The renders from mills varied considerably. At Sherborne, where there were 12 mills, four rendered jointly 18^. 6d., three rendered jointly 30^., three others 22^., one 105., and one 55. Where the composite render from a number of mills is given, it is not possible to deduce how much each one was worth since there is no guarantee that the money was divided equally. At Waia (nos. 380 and cxii) there were 3 mills which rendered 355. and Exon. Domesday shows that while one of them rendered 10^. the other two rendered 25^. Of 131 mills for which individual values are given 45 rendered less than 55., 43 rendered from 55. to 10s., and 43 rendered 10^. and over. Renders in kind are mentioned once only, at Tarrant Keyneston (no. 60), where two mills rendered 30^. and 1,000 eels.

The largest numbers of mills were on the upper reaches of the Stour. Mills are sometimes mentioned in the 12th-century surveys of the land of Shaftesbury Abbey. At Compton Abbas, in the earlier survey, the abbess supplied the timber of the mill, the miller himself the mill-stones, and the villani transported the mill-stones to the mill. 85 At Bradford-on-Avon (Wilts.), in the same survey, the miller had timber from the wood yearly, the villani helped in the repair of the mill and the transportation of the mill-stones, and in return the miller did not receive toll from the lord's malt.^*' The mill at Compton was worth 45. 2.d. at the time of Domesday (no. 129) and according to the earlier survey, the miller still rendered 45. 2d. in the 12th century, but this is the only such correspondence between Domesday and the Shaftesbury surveys.

Salt-pans and salt-workers are occasionally m^entioned in Dorset, although the county was not an important centre of the salt industry. The Count of Mortain had 32 salt-pans {saline) at Studland (no. 209) which rendered 405. and 12 at Wai (no. 163), and 16 salt-workers {salinarii) at Charmouth (no. 215). Glastonbury Abbey had 13 salt-workers rendering 13^. at Colway in Lyme Regis (no. 68), Milton Abbey had 13 rendering 20^. at Ower in Purbeck (nos. 105 and Ixxix), and William Belet had 14 at Lyme (no. 504). The production of salt seems to have been confined to three areas, the Isle of Purbeck (Studland and Ower), the mouth of the Wey {Wai), and Lyme Regis

*' R. W. V.yton, Key to Dotnesday : Dorset . 1 17-18. miller debet habere i trunctiim et i pomarium sHrestrem

*■• See p. 136. unoquoque anno ad molendimim facienduni : ibid. f. 75V. At

*' B.M. Had. MS. 61, f. 49V. Hinton at the same date the miller debet habere . . . i

*' Ibid. f. 38: habebit singulis annis i lignum in silva et quercum singulis annis ad reficiendum molendinum (ibid. f.

auxitium de hominibus et carros ad fractum (sic) molendinum 65V), and at Tisburv' debent omnes reparare molendinum sua

et motas adducendas. At Bradford in the second sur\ey the (sic) de bosco domine: ibid. f. 71.

22

DOMESDAY SURVEY

(Colway and Lyme). Later evidence records saltcotes in the same areas. Robert of Lincoln, in his charter founding the priory of Holme as a cell of Montacute, gave to it one tithe of salt from his saltcotes adjoining the manor of Langton Matravers, and Robert's son Alvred in confirming and extending his father's grant mentions a tithe of salt from his saltcotes in Purbeck.^^ in the second survey of Shaftesbury Abbey's lands (dating from the later 12th century) there is an account of Arne in Purbeck, which consisted of a hide of land devoted entirely to the production of salt.^s Since all the places named are on the seaboard, it is evident that the salt was refined from sea water and not from brine-pits, and the account of Arne mentions the plumba, leaden vessels used to collect and boil the sea-water and isolate the salt.^^

Fishermen (piscatores) are recorded in the Lyme Regis and Weymouth areas. The manor of Lyme, belonging to the Bishop of Salisbury (no. 36), was held by an unstated number of fishermen who paid 155. to the monks adpisces.'^° At Brige (no. 348), belong- ing to Aiulf the chamberlain, there were 2 fishermen, and 2 more in the same place belonging to Brictuin (no. 465). This manor lay in the neighbourhood of Weymouth.^' These were obviously sea-fishermen, and it is plain that they cannot have been the total of fishermen in Dorset at this time. The second survey of Shaftesbury Abbey twice refers to the herrings of Wareham, and there must have been a fishing fleet to catch them.92 'Phe Qnly reference to fishing in weirs is the statement that the 2 mills at Tarrant Keyneston (no. 60) rendered i ,000 eels.

Other manorial appurtenances occasionally recorded are vineyards, orchards, gardens, and honey. Two vineyards (vinee) are recorded, both belonging to Aiulf the chamberlain. At Durweston (no. 346) he had ii acre vinee and at Wootton Fitzpaine (no. 347) ii arpetiz vinee. One orchard (virgultiim) is recorded at the appropriately- named manor of Orchard (nos. 422 and civ) belonging to the wife of Hugh fitz Grip.'^ Gardens {orti) were more numerous. William of Eu had two gardens in Wareham attached to his manor of Lytchett Matravers (no. 260) and the Abbess of Shaftesbury had a garden in Shaftesbury itself. William of Moyon had a garden in Wareham attached to his manor of Poleham (nos. 276 and Ixxxvi), which is recorded in Exon. Domesday but not in the Exchequer text. Hugh's wife had a garden at Turners Puddle (nos. 391 and cxxiv), which again is mentioned only by Exon. Domesday. Renders of honey {mellis) are recorded twice: i sester (sextaritim) at Holworth (nos. 104 and Ixxviii) and 4 sesters at Rushton (no. 292). Waste land is rare in Dorset. Nyland (no. 150), held by Drew of the Count of Mortain, was waste {vasta est), and according to Exon. Domesday part of Hurpston (nos. 414 and cxlvii) was laid waste {hec terra omnino devastata est). This manor belonged to the wife of Hugh fitz Grip, whose hus- band was apparently responsible for the wasting of the Dorset boroughs^'* and the diminution of the value of Bloxworth (nos. 79 and xlii) and Affpuddle (nos. 80 and xliii), belonging to Cerne Abbey. At Stourton Caundle (no. 363) Hugh silvestris had a little manor assessed at J hide, with land for \ plough. He had there 2 bordars and 2 acres of meadow and nil amplius. It is not clear why he had no income from the manor.

Exon. Domesday's largest single addition to the Exchequer account is the information about the livestock, which is entirely omitted by the Exchequer scribes. Since the

'' Montacute Cartulary (Som. Rec. .Soc. viii), 160-2. " It is called Briige(s) iuxta Wayinue in 2 charters in

*' B.M. Harl. MS. 61, ff. 6ov-6i. the Montacute cartulary: see p. 56. There is a Bridge

"> A. R. Bridbury, Eng. and the Salt Trade in the Later Farm in the area which may preserve the Domesday name:

Middle Ages, 16-17. The author says (ibid, ig) that there O.S. Map 25,000 SY 67 (1958).

were 32salt-pans in Dorsetin 1086, whichmustbethegroup '" B.M. Harl. MS. 61, ff. 65, 67.

at Studland, but he does not mention the 12 pans at Wai. '' This orchard is probably to be identified with the

'° Lyme was originally given to Sherborne for the garden near Bradle, the tithe of which was given to

taking of salt (see p. 41), but there are no salt-pans Montacute Priory by Alvred of Lincoln (II): see p. 56.

recorded in 1086. There was a house there rendering 6d. '■• See p. 27.

23

A HISTORY OF DORSET

greater part of the Dorset survey is missing from Exon. Domesday all the livestock figures need to be treated with some care and it is doubtful what conclusions can be based on them. It is, however, immediately clear that sheep were overwhelmingly important in the county's economy. There were 22,362 sheep recorded on the lands of the 12 landowners covered by Exon. (see Table 4),^^ or approximately 88 per cent, of the total livestock. ]\Iost of them were ewes [ores), which could provide both wool and milk, but berbices, male sheep kept for mutton, are recorded at Renscombe (nos. 91 and Iv) and Mapperton (nos. 283 and xciii). The numbers of sheep are an indication of the importance of the wool trade even at this date.^^

The largest flocks belonged to the king and were at Cranborne (nos. 16 and xxii), where there were 1,037 sheep, and Puddletown (nos. 8 and ii), where there were 1,600. It is difficult to generalize about distribution from the incomplete data, but there appear to have been more sheep on the western and southern slopes of the North Dorset downs and in the valley of the Piddle than elsewhere. About 3,000 sheep are recorded in this area.97 Exon. Domesdav, however, covers few manors in the north of Dorset, and Cranborne, with its 1,037 sheep, mav have been the nucleus of an equally important sheep-farming area. Over 2,000 sheep were pastured in the coastal areas opposite Chesil Beach. ^^ Sheep provided not only wool and mutton but cheese and milk, which was used much more than cow's milk. She-goats {capre), of which there were 811 in Dorset, were also kept for meat and milk. Four manors which kept goats kept cows also : Cranborne (nos. 16 and xxii); Ibberton (nos. 10 and v); Renscombe (nos. 91 and Iv); and Farnham (nos. 396 and cxxix). Renscombe is the manor where only male sheep {berbices) were recorded. Onlv 58 cows {vacce) are recorded and it is unusual to find more than one or two on any single manor, although there were 4 at Ibberton, 4 at Holworth (nos. 104 and Ixxviii), 5 at Winterborne Monkton (nos. 514 and xxxvii), 6 at Wintre- burne {nos. 384andcxvi), 10 at Cranborne, and 13 at Staff^ord (nos. 383 and cxv). Animalia, usually considered to mean cattle other than the plough-oxen, are more numerous, amounting to 520 in all. Nine oxen {bores) are recorded at Affpuddle (nos. 80 and xliii). The most numerous animals after the sheep were pigs {porci), presumably because they were easy to feed, foraging for themselves in the woods and feeding off acorns and beech mast. There were 1,613 pig^ recorded in Exon. Domesday.

T.\BLE 4

Livestock Recorded in Exon. Domesday

The King

Countess of Boulogne Ceme Abbey Abbotsbup.' Abbey Ta\istock Abbey Milton Abbey William of Moyon . Roger Arundel Serle of Burcv Wife of Hugh Walter de Claville .

Total

Sheep	Pigs	Goals	Oxen	Horses	Cozus
9,132	591	419	147	47	16
100	16				5
2,632	196	68	15	2	6
1.776	90	45	60	14
260	20		9
1.727	S6	87	34	14	4
1,132	140	59	72	12
872	120	48	29	2	4
443			10	I
4,096	357	80	122	15	23
192	27	8	31	6

Mares Asses

13

22,362 1,613

814

529

"3

S8

'5 No livestock are in fact recorded at Purse Caundle, the western slopes of the downs there were 800 sheep at

the single manor belonging to Athelney Abbey, which was I of these 12.

" Eileen Power, Medieval Eng. Wool Trade, 31-32.

" There were 1,600 sheep at Puddletown, 393 at Waterston, 300 at Tolpuddle, 260 at Little Puddle, 115 at Burleston, 80 at Turners Puddle, and 12 at Affpuddle. On

Burton Bradstock, 260 at Askerivell, 200 at Chilfrome, 158 at Powerstock, 150 at Compton Abbas, 108 at X. Poorton, and 93 at Loders.

" There were 600 sheep at Abbotsbury, 550 at Little- bredy, and 900 at Portland (which included W'e>'mouth).

24

DOMESDAY SURVEY It is remarkable that out of 88 manors where pigs were kept, only 48 had associated woodland, and at Renscombe (nos. 91 and Iv), where there were 12 pigs, the wood nullum fnictum fert. In such cases they must have grazed on the pasture, or perhaps in the forests if they were near enough, or on the stubble of the fields. At Cheselbourne, in the 1 2th century, Wulfric the priest hzd pascua suis porcis in stiplam cum porcis abba- tissey^ Cart or pack-horses (runcini) are recorded quite frequently, but in small numbers, usually only one or two to a manor. There were 122 in all. Mares (eque) are rare, being mentioned only three times, and numbering 25 in all. Roger Arundel had 12 unbroken mares {indomite eque) at Chelborough (nos. 324 and xcvi) and a mare at North Poorton (nos. 329 and ci), and at Turners Puddle (nos. 391 and cxxiv) the wife of Hugh fitz Grip had 12 mares with their foals {eque cum suis pullis). A single ass (asinus) is recorded at her manor of Frome Whitfield (nos. 377 and cix).

The four Dorset boroughs, Dorchester, Wareham, Bridport, and Shaftesbury, con- form to the usual type of borough found in south-western England in 1086, small and not fully developed, but clearly distinguished from their agricultural surroundings. Of the boroughs three (Dorchester, Wareham, and Bridport) contributed to the firma unius noctis. Dorchester the borough was presumably connected with the group of manors headed by Dorchester (nos. 4 and xii) and Bridport with the group headed by Burton Bradstock (nos. 2 and x). Wareham was probably associated with Winfrith Newburgh (nos. 6 and xv), which is geographically nearest. A fourth group of royal manors, the Wimborne Minster group, seems to have had burgesses at its centre in 1086, although it was not at that time classed as a borough. A priest with land at Hinton Martell (no. 31) had 11 houses in Wimborne, and at Hinton also the church of Wimborne had li hide and | virgate, and 8 burgesses. It is not impossible that the burgesses were at Hinton, but they may equally well have been at Wimborne.' The abbey of Horton (no. 117) had a chapel (ecclesiola) and the land of two houses {terra duabus domibus) at Wimborne. The two best hides of the manor of Horton lay in the forest of Wimborne, and it is possible that the chapel and the land were given to the abbey by the king in exchange for these two hides just as the church of Gillingham was given to Shaftesbury Abbey in exchange for a hide at Kingston (no. 134). Edward of Salisbury had two bordars and a house in Wimborne attached to his manor of Canford Magna (no. 243), and also a league of marshland. The fact that four persons had land in Wimborne provides one characteristic (though not invariable) feature of a borough, 'tenurial heterogeneity'. ^ This characteristic is exhibited in varying degrees by the four boroughs. In Dorchester in 1086 there were 88 houses standing and 100 destroyed, but the account does not state to whom they belonged. One house in Dorchester belonged to the abbey of Horton (no. 117) and the Bishop of Salisbury had one burgess and 10 acres of land in Dorchester attached to his manor of Charminster (no. 32). The same bishop had J acre of land in Bridport (no. 48), where in 1086 there were 100 houses standing and 20 ruined but still inhabited. Their owners are not specified. Shaftesbury was divided between the king and the Abbess of Shaftesbury. In 1066 the king had held 104 houses and the abbess 153 houses. In 1086 the king had 66 houses standing and 38 destroyed and the abbess 1 1 1 houses standing and 43 destroyed. She had there 151 bur- gesses, 20 empty mansiones, and a garden, which together were worth ;^3 55. od. In Ware- ham T.R.E. there were 143 houses in dominio regis. In 1086 the king had 70 houses standing and 73 destroyed, the abbey of St. Wandrille had 45 houses standing and 17 destroyed, and de partibus aliorum baronum there were 20 houses standing and 60

»' B.M. Had. MS. 61, f. 44V. Hinton entry.

' Ellis (Gen. Introd. to Dom. Bk. ii. 438) lists 8 burgesses ^ F. W. Maitland, Dom. Bk. and Beyond, 178.

in Wimborne who are presumably those mentioned in the

25

A HISTORY OF DORSET

destroyed. The abbeyof St. Wandrille also held a church inWareham (no. 124). A church and i; houses in Wareham belonged to the abbey of Horton (no. 117). This church was probably that of St. Martin, which dates from the nth centur}\^ A house in Wareham was attached to the manor of Creech (no. 202), held by the Count of Mortain, and a house in Wareham belonged to the manor of Broadmayne (no. 223) held by Earl Hugh. Two burgesses and 12 acres belonging to the Bishop of Salisbur}', and a burgess in Wareham, were attached to the manor of Povington (no. 242) held by Robert fitz Ceroid. William of Eu had a bordar and two gardens in Wareham attached to his manor of Ljtchett Matravers (no. 260) and William of Moyon had a garden in Ware- ham attached to his manor of Polehatn (nos. 276 and Ixxxvi). At Hurpston (nos. 413 and cxlvi), belonging to the wife of Hugh fitz Grip, there was a burgess rendering 8^. In view of the proximity of Hurpston to Wareham, it seems not unlikely that he was in Wareham ."^

Although some burgesses are said to render money, there is nothing to throw light on their form of tenure, or the rate at which they paid rent. The burgess attached to Povington rendered 2S., the one at Hurpston 8d. The Abbess of Shaftesbury had 65^. from 151 burgesses, 20 empty tnansiones, and a garden. Assuming the empty mansiones to be unproductive and discounting the garden, the burgesses may have paid about 2^. each. The values of property in the boroughs are only incidentally mentioned. The church and five houses attached to Horton rendered b^d., about i^d. a house. The house attached to Broadmayne rendered ^d. The Bishop of Salisbury had a house at Lyme rendering 6d., and his \ acre in Bridport rendered the same amount. William of IMoyon's garden in Wareham rendered 3^.

Each of the four boroughs was assessed for geld. Bridport was assessed at 5 hides, Dorchester and Wareham at 10 hides each, and Shaftesbury at twenty. They all dis- charged this obligation by contributing to the support of the king's housecarls. Bridport rendered \ mark of silver {bs. 8d.), Dorchester and Wareham i mark (135. 4^.) each, and Shaftesbur}' 2 marks (265. 8^.). The boroughs of Devon and the Wiltshire borough of Malmesbury did a similar service. In Devon the borough of Exeter paid geld only when London, York, and Winchester paid et hoc erat dimidia marka argenti ad opus militum. Exeter was assessed at five hides: quando expeditio ibat per terram aut per maretn serviebat hec civitas quantum v hide terre.^ Bridport, assessed at 5 hides, also rendered \ mark ad opus hiiscarlium. The three other Devon boroughs — Barnstaple, Lydford, and Totnes — did jointly the same service as Exeter. Totnes paid geld when Exeter did, and then it rendered 3^. 4^.^ At Malmesburj' (Wilts.) quando rex ibat in expeditionem vel terra vel mari habebat de hoc burgo aut xx solidos ad pascendos suos buzecarlos aut unum honiineni ducebat secum pro honore v hidarum. By analogy with the Dorset and Devon boroughs 205. (li mark) was the amount one would expect a borough of 15 hides to pay.''

All four boroughs had moneyers in 1066 but none of these was said to be there in 1086. There had been one at Bridport, two each at Dorchester and Wareham, and three at Shaftesbur\'. Each rendered one mark of silver to the king and 20^. quando moneta zertebatur, that is when fresh dies were issued. The mints of Wareham and Shaftesburj- are mentioned by name in Athelstan's mint Law^ when there were two mints at each borough. Coins struck at Dorchester and Wareham under Athelstan have survived,'

' G. Baldwin Brown, Arts in Early Eng. (1925), ii. 484. ' F. L. Attenborough, Latcs of the Earliest Etig. Kings,

* For this burgess, see also M. Bateson, 'The Burgesses 134. The I2th-cent. Quadripartitus version lists a mint at

of Domesday and the Malmesburj' Wall', E.H.R. xxi. 710. Dorchester.

5 Dom. Bk. (Rec. Com.), i, f. 100. " G. C. Brooke, Eng. Coins from the yth Cent, to the

' Ibid. f. io8b. ' V.C.H. Wilts, ii. 23. Present Day (1955), 59.

26

DOMESDAY SURVEY

and the Bridport mint is first recorded in the time of Aethelred II.'° Although Domes- day says only that the mints were there T.R.E. and does not say what the position was in 1086, coins from all four boroughs were struck in the reigns of William I and William II."

It seems unlikely that at this date these little boroughs had courts of their own, but there is a possibility that Dorchester did have its own court. There was a hundred of Dorchester as well as a borough and a vill, and it seems quite likely that the hundred court met in the borough of Dorchester. The boroughs seem to have been extra- hundredal themselves, like the king's manors to which they were attached, and this fact probably contributed to the development of separate courts. Professor Tait observed that the hundred of Dorchester was later known as the hundred of St. George, to whom the parish church of Fordington was dedicated, Fordington being a suburb of Dorchester. He considered it possible that the hundred of Dorchester had split into two, one half covering the borough and the other half covering the geldable portion. It is possible that this division had taken place some time previously, and that consequently the borough of Dorchester had its own jurisdiction perhaps even before the Conquest. '^

In all four boroughs a considerable number of houses had been destroyed after the Conquest. In Dorchester there were 172 houses T.R.E. ; in 1086 there were 88 houses standing and 100 destroyed. In Wareham there were 143 houses T.R.E.; in 1086 the king had 70 houses standing and 73 destroyed, the abbey of St. Wandrille had 45 houses standing and 17 destroyed, and the other barons had 20 houses standing and 60 destroyed. In Shaftesbury T.R.E. the Abbess of Shaftesbury had 153 houses and the king had 104 houses; in 1086 the abbess had 11 1 houses standing and 43 destroyed and the king had 66 houses standing and 18 destroyed. In Bridport T.R.E. there were 120 houses. In 1086, according to the Exchequer text, there were 100 houses and 20 sunt ita destitute quod qui in eis manent geldum solvere non valent. The Exon. text implies that 20 houses must have been destroyed as well, by showing that the 20 impoverished houses were numbered among the 100 houses still standing: xx ex his c domibus ita sunt adnichilate quod homines qui intus manent non habent unde reddent nullmn^^ geldum. There is no apparent reason for this destruction. The Chronicle does not record any disturbance in the area which could have led to such systematic wasting. Eyton's suggestion that it was caused by 'internal conflicts between the Anglican and Norman burgesses' may be the correct one,''* but in view of his other depredations, '•■> it is possible that the boroughs came into conflict with Hugh fitz Grip. In each account the devastation is said to have taken place a tempore Hugonis vicecomitis usque nunc.

II

King William had received in Dorset four groups of lands, the lands of King Edward, the lands of Earl Harold, the lands of Queen Maud, which had reverted to her husband on her death, and two manors which had belonged to Countess Goda, King Edward's sister. The lands of King Edward consisted of Portland and five large groups of manors. Portland (nos. i and vi) was not assessed in hides, and there is a blank space in the Exchequer entry where one would expect to find the number of teamlands. It rendered ^^65 blanched {Ixv libras albas) a year, and did not pay geld. The five groups of manors consisted of Burton Bradstock, Bere Regis, Colesberie or Coles- breia, Shipton Gorge, Bradpole, and Chideock (nos. 2 and x); Wimborne Minster, Shapwick, Crichel, and Opewinburne or Obpe Winborna (nos. 3 and xi); Dorchester,

'° Ibid. 70. " ni/Z/wm is underlined in the Exon. text as if for erasure.

" Ibid. 83-85. '■' Eyton, Key to Domesday: Dorset, 72.

'2 J. Tait, Medieval Eng. Boro. 52-53. " See pp. 23, 46.

27

A HISTORY OF DORSET

Fordington, Sutton Poyntz, Gillingham, and Frome (nos. 4 and xii); Pimperne and Charlton Marshall (nos. 5 and xiv); and Winfrith Newburgh, Lulworth, Wintreborne , and Knowlton (nos. 6 and xv). None of them was assessed in hides, nor had they ever paid geld. The three groups headed respectively by Burton Bradstock, Dorchester, and Wimborne Minster each rendered one night's farm (firma iinius noctis) and the two remaining groups headed by Pimperne and Winfrith Newburgh each rendered \ night's farm {dimidia firma iinius noctis). The night's farm was an ancient food-rent, dating from the time when the king continually travelled about with his court, and representing the supplies needed to support the king and his retinue for one day. It survived in 1086 largely in the area of the old West Saxon kingdom, and is found in Hampshire (where it is called the firma uniiis diei), Wiltshire, and Somerset as well as in Dorset, but traces of it also survive elsewhere. In Dorset it was not commuted to a money rent in 1086, but money values are given for the farm in the other three western counties. In Hampshire, the only county where the value was in any way standardized, the farm was worth ^jb \bs. M. T.R.E. and £\o^ 12s. zd. in 1086. No T.R.E. value is given for either Wiltshire or Somerset. In Somerset in 1086 two groups of manors rendered ;(^io6 c*. lod., a third group rendered ^^loo 10^. gld., and a fourth group ;^I05 lys. ^\d. In Wiltshire in 1086 one manor paid i^ioo and another /^i 10, while the remaining four manors liable to the night's farm were not valued in money."'

In Dorset the manors liable to the night's farm are grouped together in such a way that the only obvious reason for their association is that of size, although this is unlikely to have been the only consideration. Certainly they do not form compact geographical entities. Gillingham, in the northern tip of Dorset, is associated with Dorchester and its suburb of Fordington, in the south of the shire, and Bere Regis in the east is associated with Burton Bradstock, Chideock, and Shipton Gorge, which form a compact group in the west. If the number of teamlands can be taken as a rough guide to the comparative sizes of the manors, then the three groups rendering a full night's farm were approximately equal, if not in size, at least in agricultural capacity, having respectively 56 teamlands (Dorchester), 55 teamlands (Burton Bradstock), and 45 teamlands (Wimborne Minster). The two groups rendering \ night's farm were about half the size, with 20 teamlands (Pimperne) and 24 teamlands (Winfrith Newburgh). Three of the four Dorset boroughs contributed to the night's farm in the manner already described, and some of the Dorset hundreds — Dorchester, Gillingham, Bere Regis, Pimperne, Winfrith Newburgh, and Knowlton — bear the names of royal manors, which must have been their heads.

Of the manors which had belonged to Earl Harold the most important was Puddle- town (nos. 8 and ii). Since the third penny of the whole shire was attached to this manor it was clearly part of the official endowment of the earldom, and it was in the hands of Aiulf, Sheriff of Dorset, in 1086. With its adjuncts, consisting of li hide in Purbeck and \ hide in Mapperton, Puddletown was worth £']-^, and was beneficially hidated, being assessed at | hide, with land for 15 ploughs. It must also have been the head of the hundred of the same name. Of Earl Harold's other manors, Charborough {Celeberge) was also the head of a hundred, and Loders, assessed at 20 hides, was a hundred in itself.

The land of Queen Maud consisted of nearly 42 hides: 31 hides which had belonged to Beorhtric son of Aelfgar, and a further 1 8 hides which Hugh fitz Grip had held of her. According to Exon. Domesday Schelin held two of her manors, Witchampton (nos. 20

"■ For a full discussion of the night's farm and its Poole, Exch. in the i2th Cent. 27-30; Dialogus de Scaccario, commutation, see J. H. Round, Feudal Eng. 109-15 ; R. L. ed. C. Johnson, pp. xxxviii, 40-41.

28

DOMESDAY SURVEY

and xxvi) and Edmondsham (nos. i8 and xxiv). Schelin also had held part of Ham- preston (no. 443) of the queen, but in 1086 it was held by Torchil of the king. Another part of Hampreston (nos. 19 and xxv) was held of the queen by William Belet, according to Exon. Domesday. He also held 12 acres of meadow at Bingham's Melcombe (no. 30) which formerly had been leased to Wlgar Wit. Melcombe and Hinton Martell (no. 31) had formerly belonged to Countess Goda. Melcombe is said to have been taken by Earl Harold from Shaftesbury Abbey and the fact that the abbey had once held this manor is confirmed by the entry relating to the abbey's manor of Cheselbourne (no. 138) where it is stated that Earl Harold had seized the abbey's manors of Stour and Cheselbourne. King William had ordered them to be returned to the abbey in accordance with a writ of King Edward, but he himself still retained Melcombe. Since Goda was dead by 1056 it seems likely that Melcombe passed to Shaftesbury Abbey as a bequest, and that subsequently Harold committed the trespass. What became of Hinton in the interval between the death of Goda and the arrival of King William is unknown. Nearly half the manor was subinfeudated, mostly to ecclesiastics. A priest had held i hide of thegnland, which the king had in demesne in 1086. Another priest held 2| hides T.R.E., of which he retained i| hide in 1086, the other hide being held by the Bishop of Lisieux. A third priest living in Tarente held i\ hide, and Ulvric held i virgate of land. Lastly, i\ hide and \ virgate belonged to the church of Wimborne Minster, and were held by Maurice, Bishop of London. Attached to the manor of Melcombe were 3^ virgates in Buckland hundred which three free thegns had held T.R.E., and which, according to Domesday, Countess Goda had added to her manor. The Geld Roll for Buckland hundred gives a different account, attributing their acquisition to Robert de Oilly, who seems, from an entry in the Geld Roll for Canendofie hundred, to have held Countess Goda's two manors at farm of the king."'' Exon. Domesday records that Fulcred held the manor of Child Okeford (nos. 7 and i) ad firmam de rege. Child Okeford had belonged to Earl Harold, and Fulcred also held all the other manors of Earl Harold, with the exceptions of Loders (nos. 13 and ix) which was held at farm by Roger, and Puddletown which was held by Aiulf, Sheriff of Dorset, who also held the queen's manors of Frome St. Quintin (nos. 15 and xxi) and Wimborne (nos. 21 and xxvii). Fulcred appears in the Geld Roll for Uggescombe hundred, accounting for the geld on i| hide of Harold's land which can be identified as part of Fleet, and it seems likely that he held the other manors at farm also. Roger was perhaps Roger Arundel, a considerable landowner in the area, who had held the manor of Piddletrenthide (no. 69) before it passed to the New Minster at Winchester.

It is uncertain whether all King William's manors were exempt from geld or whether such exemption was confined to those of them which rendered the night's farm.'^ In the Geld Rolls the demesne of the king is exempt in the same way as the baronial demesnes, but in some cases, notably in the manors of Earl Harold, the villani had not paid geld either. '^ It is not clear whether these were defaults or whether the land in question was exempt. In the case of Ibberton in Haltone hundred there certainly seems to have been a default. It was a manor of 5 hides, zl hides in demesne and 2\ hides belonging to the villa?ii. In Haltone hundred the king received ;(^i2 15*. on 45 hides. At 6s. on the hide he should have received ,^13 10^. and in fact the account concluded restant xv solidi de terra Heroldi que est terra villanorum, i^s. being the geld on 2\ hides. This certainly implies that the land was geldable and should have paid with the rest. Fleet (nos. 1 1 and vii), in Uggescombe hundred, is said to have paid geld in another

" See pp. 129, 146. Rob. de Oilly was Sheriff of Warws. Celeberge (Charborough) hundred; Chaldon, in Winfrith " See V.C.H. Wilts, ii. 176-7. hundred; and Loders, in Loders hundred: see pp. 136,

" Okeford, in Ferendone hundred; Charborough, in 143,146,148.

29

A HISTORY OF DORSET

hundred, though there is no record of this. Of the manors belonging to the queen two, Cranborne, in Albretesberge hundred, and Ashmore, in Langeberge hundred, had not paid geld. Hampreston, in Canetidone hundred, had not paid geld. It was assessed at 2 hides and i virgate with i hide in demesne, and according to Exon. Domesday William Belet had held it of the queen. The Geld Roll for Canendone hundred states that de it hidis et i virga quas tenet i tagtiiis adfirmam de rege non habuit rex geldiim. The rest of the queen's manors appear to have paid geld in the usual way. The two manors which had belonged to Countess Goda, and which the king held in 1086, were certainly liable to geld. At Hinton there were 6 hides and i virgate in demesne, duly recorded as exempt in Canendone hundred, where the manor lay, and de v hidis de terra Gode quam tenet Rotbertns de Oilleio ad firmavi de rege habuit rex geldum post Pascha. Hinton was assessed at 14 hides and i virgate so that 3 hides must have paid geld normally. The geld due from Melcombe was also withheld until after Easter, and the king had still not received it. The manor was assessed at 10 hides and lay in Haltone hundred, 3^ virgates in Buckland hundred being attached to it. The king's demesne consisted of 7 hides and 3 virgates, which were exempt, and Robert de Oilly retinuit inde xv solidos iisqiie post Pascha quos nundum habet rex. In the Geld Roll for Buckland hundred it is further stated that de dimidia hida et dimidia zirga quas Rotbertns de Oilleio abstulit i tagno et posuit intra firmam regis in Melecoma non habuit rex geldum. Thus no part of Melcombe had paid geld.

Alienations of royal land are occasionally recorded. A king's reeve had leased to Toxus a piece of land in Sw^^re (no. 263) que nunquam geldavit T.R.E. sed erat in dominio et in firma regis, and which in 1086 was held by William of Eu. The abbey of Cranborne held a piece of land in Gillingham (no. 70) which Hugh fitz Grip had given to the abbey, having taken it de firma regis. Half a hide attached to Cerneli (no. 212) fuit de dominica firma Cerne T.R.E. and an entry in the Geld Roll for Whitchurch hundred, which seems to refer to this land, says that it belonged to the king.^° Apart from these pieces of land it seems likely that King Edward had already given Portland to the Old Minster at Winchester. A writ exists which purports to be a record of this grant, but it is of doubtful authenticity.^' The grant, if genuine, may have been intended to take effect after the king's death, and it is noticeable that the king held Portland in vita sua or, as Exon. Domesday expresses it, ea die qua ipse fuit virus, omitting the customary et mortuus. In any event the writ seems to reflect an actual grant, since in his charter to the Old Minster King Henry granted it Portland and its appurtenances as the gift of King Edward.22 Soon to be alienated was Burton Bradstock, which in i loi was given by King Henry to the abbey of St. Stephen, Caen, in exchange for the crown and other regalia which William I had left to that abbey. ^^ Richard de Rivers (or Redvers) gave Loders to the abbey of Montebourg, with the king's permission, and its assessment was reduced from 20 hides to five.--* Two of King Edward's manors, Winfrith Newburgh and Bradpole, were later held by serjeanty.

After King Edward the richest man in Dorset before the Conquest was Earl Harold, whose earldom of Wessex included Dorset. As well as the manors held in 1086 by King William, Harold had possessed Bincombe (no. 122), assessed at 8 hides, Waterston (nos. 334 and cvi), assessed at 10 hides, and Shillingstone (no. 367), assessed at 16 hides. He held in all 87I hides, or about one-eighth of the land in Dorset held by

^° See pp. 125, 126. In 1212 the Prior of Winchester held Portland de antiqiio

" This writ is printed in F. E. Harmer, Anglo-Saxon fefemento region Anglorum: Bk. of Fees, 90.

Writs, no. 112. For a discussion of it, see ibid. 385-7. -^ Regesta Regum Anglo-Nonnannorum, ed. C. Johnson

^^ This charter is printed in V. H. Galbraith, 'Royal and H. .\. Cronne, ii, nos. 601, 1575.

Charters to Winchester', E.H.R. xxxv. 390 (no. xviii). " Ibid. no. 1018.

30

DOMESDAY SURVEY

laymen excluding the king. He had taken four manors from Shaftesbury Abbey: Stour (no. 127), assessed at 17 hides, Cheselbourne (no. 138), assessed at 16 hides, Melcombe (no. 30), assessed at 10 hides, and a manor called Pidele.-^ He had also taken a manor belonging to a certain clerk (quidam clericiis) and given it to Eadnoth the staller. This manor, Ilsington (no. 221), was held by Earl Hugh in 1086. Harold's mother, Countess Gytha, held two manors in Dorset T.R.E., Little Puddle (nos. 14 and iii) and Frampton (no. 121), assessed at 25 i hides, a total hidage of 3oi hides. To the manor of Frampton was attached the third share of the wood of Ilauocumbe, belonging to Burton Bradstock.26 Queen Edith, Earl Harold's sister and the widow of King Edward, had held the manor of Sherborne (no. 37), assessed at 43 hides. She presumably held it for life, since Bishop Aelfwold of Sherborne had held it before her, and in 1086 it belonged to the Bishop of Salisbury and the monks of Sherborne. Countess Goda had 24 hides in Dorset, consisting of the manors of Melcombe (no. 30) and Hinton (no. 31). She died before 1056 and on her death Melcombe appears to have passed to Shaftesbury Abbey.^^ Archbishop Stigand held one manor in Dorset, the large and valuable Sturminster Marshall (no. 232), assessed at 30 hides and worth J^bb.

The richest thegn in Dorset in 1086 was Beorhtric son of Aelfgar whose lands were given to Queen Maud and later formed the nucleus of the honor of Gloucester. Of Beorhtric's Dorset manors Cranborne, Ashmore, and Frome St. Quintin belonged to the king in 1086. The queen had given two other manors away, Tarrant Launceston (no. 141) to Holy Trinity, Caen, and Tyneham (no. 369) to Anschitil fitz Ameline. Boveridge, in Cranborne (no. 71), which belonged to the abbey of Cranborne in 1086, was held by Brictric T.R.E. who is probably Beorhtric son of Aelfgar. The latter may well be identified also with the T.R.E. holder of Dewlish (no. 148) which Count Alan held in 1086, since this manor was later part of the honor of Gloucester,-^ but whether he was the Brictric who held Tarente (no. 370) or the Brictric who preceded the Count of Mortain at Mappowder (no. 171) and Uploders (no. 206) is less certain. He certainly held 59 hides in Dorset T.R.E. and may have held 73 hides. Toll, whose name suggests that he was of Scandinavian origin, held 490 hides in Dorset. He was a prosperous local thegn with 34 hides in Hampshire and land in Wiltshire and Devon, all of which was held by William of Eu in 1086. William's other predecessor was Aelfstan of Boscombe, whose land in 8 counties belonged to William in 1086. Most of Aelfstan's land lay in Wiltshire and Gloucestershire, but he was one of the larger landowners in Dorset where he held 34 hides. Ailvert (Aethelfrith) and Ailmar (Aelmer), the most important of Roger Arundel's predecessors, were quite wealthy thegns, with land in Somerset as well as Dorset which also passed to Roger. Aethelfrith has been identified as Ailferth mmister who witnessed Edward the Confessor's grant to Bath Abbey in 1061.^9 He held 26i hides in Dorset and Aelmer 25! hides. They can be identified as the two men who held Piddletrenthide (no. 69), as two manors, of King Edward, since this manor was held by Roger Arundel before it passed to the New Minster, Winchester. It was assessed at 30 hides, and, assuming that it was roughly divided in half between the two men, they must have held in all about 40 hides apiece.

One of the Count of Mortain's predecessors was called Edmar. He held about 35 hides, consisting of the manors of Gussage All Saints (no. 192), Blaneford (no. 194), and some smaller manors, one of which, Wootton Fitzpaine (no. 211), appears to have been beneficially hidated.^o Edmar is probably to be identified as the man of the same name

" See pp. 43, 83. ^f" See pp. 3, 65. ^* Bk. of Fees, 93 (1212).

" See p. 29. For a full account of Goda's life and " V.C.H. Som. 419; Eyton, Domesday Studies: Som. i.

marriages, see J. H. Round, Studies in Peerage and Family 155-6.

Hist. 147-51. '" It was assessed at 2 hides but had land for 7 ploughs.

31

A HISTORY OF DORSET

whose manors in Devon and Somerset passed to the count. In the Devon survey he is called Edmer Atre, and in the Exon. Domesday for Somerset, Edmeratorius. He also held land in Cornwall.^' Eadnoth the staller, whose lands passed to Hugh, Earl of Chester, held 25 hides in Dorset, and in King William's time had taken the manor of Burstock (no. 230), assessed at three hides, from a certain thegn. He is addressed in a writ of William relating to Bath Abbey.^' He was killed in 1067, leading the militia against the sons of Harold. It was Harold from whom he had received the manor of Ilsington (no. 221). Two of his manors, South Perrott (no. 228) and Catsley (no. 229), he had bought from Bishop Aelfwold of Sherborne, on condition that at his death they should return to the church, but Earl Hugh held them in 1086. Two other manors held by Eadnoth had at one time belonged to Sherborne Abbey." Eadnoth's son, Harding, also survived the Conquest, and became the ancestor of the Berkeleys.^* He may be identical with Harding who held Bredy (Farm) (no. 317) which passed to Berenger Giffard. According to the Geld Roll for Godderthorn hundred, where Bredy (Farm) lay, Berenger's predecessor continued to hold of him at farm.^s Wulfwynn, a wealthy English lady, with about 100 hides in six counties, held two manors in Dorset, Canford Magna and Kinson (nos. 243, 244), assessed jointly at 38 hides. Like the rest of her land, these two manors were held in 1086 by Edward of Salisbury. In the Middlesex survey she is called IVlwene homo regis and in the Buckinghamshire survey Whven homo regis Edzvardi. In Buckinghamshire she is also called Wlwene de Cresselai, which appears to refer to her tenure of the manor of Creslow (Bucks. ).36

The prevalence of names like Alvric and Ahvard makes identification uncertain, but it is at least likely that the Alvric who held five of the manors belonging to the wife of Hugh fitz Grip in 1086 is identical with the Alvric who T.R.E. had been the tenant of three of the manors which Hugh fitz Grip held of the queen. If this is the case he had 24! hides. Alward Colin(c), who T.R.E. had held Langton Herring (nos. 23 and xxix), another of the manors held by Hugh fitz Grip of the queen, and who still held Thorn- combe (no. 439) in 1086, was probably the man who held five manors belonging to the wife of Hugh fitz Grip in 1086 and also the manor of Little Waddon (no. 460) which Hugh gave to Brictuin. If so, he had 16 hides. It is unsafe to identify him with the Alward who held 14.I hides which passed to the Count of Mortain, or the Alward who held 152 hides which passed to William of Moyon.

Bondi is a common Scandinavian name, but it occurs only twice in Dorset and it is probably safe to identify the man who held Broadwindsor (no. 505) with the man who held Compton Valence (no. 357), giving him a total of 30 hides. John, who held the two manors which passed to Matthew de Moretania (nos. 320 and 321), is probably to be identified with John the Dane {daniis), a predecessor of Matthew in Gloucestershire and Somerset." He had about 15 hides in Dorset. Godric held 12 hides as the pre- decessor of William of Moyon. Alfred the sheriff held Lulworth (no. 350), assessed at 8 hides, 3 virgates, which in 1086 was held by Aiulf the chamberlain, then Sheriff of Dorset. Since Alfred is a common Saxon name, it is unsafe to identify Alfred the sheriff with the pre-Conquest holders of Stour Provost (no. 231) and Wintreburne (no. 305), or with the man who held two manors belonging to the Count of Mortain in 1086.38 Bricsi (Beorhtsige), miles regis Edwardi, held Wootton Fitzpaine (no. 347), assessed at

'' Dom. Bk. (Rec. Com.), i, f. 104b; iv. 190, 191. '- Regesta Regum Anglo-Normannorum, i. no. y.

Round (V.C.H. Som. i. 418) identifies him with Edmer " See p. 41.

attile or atiile, who held land in Herts., Md.\., and Berks., ■'■' V.C.H. Som. i. 417-18; see also p. 57.

■which passed to the Count of Mortain. Feilitzen {Pre- " See pp. 36, 131.

Conquest Personal Names of Dom. Bk. 232 n.) regards this " V.C.H. Wilts, ii. 99.

identification as no more than a possibility'. Edmer attile " See p. 49.

■was also called teigrius Heraldi comitis and teigmis R.E. : 5* For a writ addressed to Alfred the sheriff, see F. E.

Dom. Bk. (Rec. Com.), i, ff. 129b, 136b, 146. Harmer, Anglo-Saxon Writs, no. i.

32

DOMESDAY SURVEY 12 hides, which Aiulf the chamberlain held in 1086. He also held Keevil (Wilts.) and about 20 hides in Somerset.^? Ode the treasurer held i hide in Wimborne Minster (nos. 21 and xxvii). Round identified him with Odo of Winchester who held land in Hamp- shire, Berkshire, and Wiltshire, as a thegn in 1086.40 Rainbald of Cirencester, who had the office if not the name of chancellor under Edward the Confessor, held Pulham (no. 146) in Dorset,4i which he still held in 1086.

Wulfweard White, who held Pentridge (no. 67) and Silton (no. 271), had land in eleven counties. He may be identical with the Wulfweard who held Tarrant Crawford (no. 58), and Eyton identified Wlgar Wit, who T.R.E. held 12 acres of meadow at Bingham's Melcombe (no. 30), with Wulfweard White.42 Other landowners with little land in Dorset but larger estates elsewhere were Saeweard, Alwi, and Wulfgifu. Sae- weard, the predecessor of Baldwin at Iwerne Courtney (no. 316), held five manors in Somerset and five in Devon, all of which had passed to Baldwin. Alwi held Turnworth (no. 319) which in 1086 was held by Alvred of Epaignes. He was Alvred's predecessor in Wiltshire, Devon, and Somerset as well, and is called Alwi Banneson in Exon. Domesday. Wulfgifu, who held 1 1 hides in Dorset as the predecessor of the Countess of Boulogne, had 65 hides in Hampshire, where she was called Wulfgifu Beteslau.^^ In addition, she held 5 hides in Wiltshire and a manor in Somerset. Other women besides Wulfwynn and Wulfgifu held land in Dorset, although only in small amounts. Aelfrun held part of Afflington (no. 236), to which she gave her name.44 Aethelflaed held 2 hides at Hethfelton (no. 294) and Leofrun held two manors, both in Stourton Caundle (nos. 363, 478), amounting to i| hide. Leofgifu held Blandford (no. 336) and Aelgifu held Morden (no. 337). The latter is probably to be identified with the woman of the same name who held Lyme Regis (no. 504) and Stourton Caundle (no. 219).

The relative rarity of the name Beorhtnod permits the identification of Britnod, who held Stafford (no. 155), with Bricnod, who held Melbury Bubb (no. 323), and Brisnod who held West Parley (no. 37 1)."*' Other rare names occur in Dorset. Sared (Saered) is found only in Dorset, where he and his brother held Blandford St. Mary (no. 185), and in Somerset, and Watman (Hwaetman), who held Waia (no. 364), occurs only in Dorset and Herefordshire. Burde, who held the little manor of Rushton (no. 292), is only found in Dorset. His name is apparently derived from the French burdel meaning mule. Her, who gave his name to Herston (nos. 333 and cv), is also peculiar to Dorset. Herston was a divided vill in 1086, part of it (no. 512) being held by Godfrey the scullion whose father held it T.R.E. His father's name is not given. Toxus the priest who held part of Swyre (no. 263) is not found anywhere else but Dorset. Aldebert (Ealdbeorht) and Wicnod (Wihtnoth), who held Cerneli (no. 212) and Milton on Stour (no. 272) respect- ively, were peculiar to Dorset. Bern (Beorn), who held part of Church Knowle (no. 312), has a name found otherwise only in East Anglia and Northamptonshire, and Turmund (Thormund) who held Wintreburne (no. 56) is otherwise only found in Somerset. Herling, who held Tarrant Keyneston (no. 60), is found only in Dorset and Berkshire, and Trawin, who held Lulworth (no. 199), Dachelin, who held Nyland (no. 251) in company with Edric and Alward, and Gerling, who held Turners Puddle (nos. 391 and cxxiv), are all peculiar to Dorset. Trasmund, who held Hill (no. 318), which passed to Osbern Giffard, is found elsewhere as Osbern's predecessor in Wilt- shire, and also held Manston (no. 302), which passed to Waleran. He is not otherwise recorded. Herling, Trawin, Dachelin, Gerling, and Trasmund all bear names of Germanic origin, while Bern, Brune (Bruno, no. 149), Turmund, Her, Toli, Bondi,

" V.C.H. Wilts, ii. 66. " ie. Beslow (Salop) : see V.C.H. Hants, i. 429.

■"■ V.C.H. Hants, i. 427. ■" See p. 45. '•'' A. Fagersten, Place-Names of Dorset, 117.

■•^ Eyton, Key to Domesday: Dorset, 112 n. '•^ O. von Feilitzen, Pre-Conquest Personal Names, 196.

DO. in 33 **

A HISTORY OF DORSET

Askell (Anschil), and Azor are all of Scandinavian derivation. Askell (Anschil) occurs twice in Dorset (nos. i8o, 490). Azor appears three times. One of his three manors, Aihvood (no. 482), had passed to Swain in 1086. Swain's three other manors had been held by his father T.R.E., and it is possible that he is to be identified with Swain son of Azor, who held land in Northamptonshire, and that Azor can be identified as Swain's father, the holder of some 20 hides in Dorset T.R.E.^^ The pre-Conquest holders of three manors are described as free men {Uberi homines). Godwin who held Cernel (no. 147) is described as a liber homo. Four free men held the manor of Galton T.R.E. (no. 507) and may be identical with the 4 men holding it for rent in 1086. Three free men held Woodstreet (no. 508). Both Galton, assessed at i hide and i virgate, and Woodstreet, assessed at 3 virgates, had passed to Osmund the baker in 1086. Cenicl was assessed at 3 hides and was held by Walter the deacon, an almsman of the king in 1086. Apart from these free men several manors were held by 'free thegns' T.R.E. A free thegn {liber tainus) held Hemsworth (no. 355). Three free thegns {Uberi taini) held the 2>\ virgates in Buckland hundred which were added to the manor of ]Melcombe. Mappowder (no. 431), which Bollo the priest held in 1086, was held T.R.E. by the same Bollo cum aliis vii liberis tainis. The thegn who held Church Knowle (no. 308) was free with his land {liber erat cum hac terra), which presumably means that he was not commended to any lord. Freedom to commend oneself to any lord is quite frequently mentioned, in a haphazard fashion which suggests that it was so common that it was not thought necessar}' to mention it in all cases. Five thegns who held three virgates attached to Rampisham (no. 55) could quo zolebant se vertere. Alward, who held Little Windsor (nos. 282 and xcii), potuit ire ad quemlibet dominiim volebat. Wulfgifu, the T.R.E. holder of the Countess of Boulogne's manors, poterat ire cum terra sua quo zolebat.^~ Dodo who held Edmondsham (nos. 18 and xxiv), Saul who held Hampreston (nos. 19 and xxv), and the two thegns who held Witchampton (nos. 20 and xxvi) could all go with their land to any lord they liked. The men who held the manors which in 1086 were held by the wife of Hugh fitz Grip were all free to go to any lord, as appears from a note appended at the end of the account of her manors : Omiies taini qui has terras tenebant poterant ire ad quern dominum zolebant. A similar statement is appended to the land of the king's Serjeants: Qui has terras tenebant T.R.E. poterant ire quo Tolebaiit, and at the end of the account of the land of the Count of IMortain there is a note that omnes qui has terras tenebant T.R.E. libere tenebant. References to commenda- tion are rare. The three thegns holding 3 hides at Cranborne of Beorhtric son of Aelfgar non potuerant separari ab eo and Alnod held Stourpaine de Edzcardo Lipe et nan poterat separari a dominio suo. Aelmer and Aethelfrith who held Piddletrenthide as two manors of King Edward non poterant cum terra ista ire ad quemlibet dominum. Aethelfrith also held Worth iNIatravers (no. 330) of King Edward et non potuit separari a seriicio regis. Beorhtsige, who held Wootton Fitzpaine (no. 347), is described as miles regis Eduardi.

There are frequent references to land held in parage {in paragio, pariter), which consisted in the joint tenure of an estate bv co-heirs. The merit of this system was that it prevented the fragmentation of manors which would otherwise have occurred in a society which did not recognize primogeniture. Sared and his brother held Blandford (no. 185) in parage and two brothers held Ranston (no. 241) in parage. Scirewold and Ulward, who held U'ai (no. 162), and Edric, Dachelin, and Alward, who held Xyland (no. 251), are said to hold in parage. The necessity of such a system is obvious when the

■•* For these rare names, see Feilitzen, op cit. /)aj«';/i. For ■" Seep. 114.

the identification of Swain as Swain son of Azor, see p. 52.

34

DOMESDAY SURVEY

proportion of manors held by large numbers of unnamed thegns is taken into con- sideration. An extreme instance is the manor of Poleham (nos. 276 and Ixxxvi), which was held by 21 thegns T.R.E., and which was assessed at 10 hides. The Exon. version shows that these thegns held in parage. If the manor had been divided between them the fragmentation would have produced a large number of tiny, uneconomic units. Even so, it is plain from such examples that many pre-Conquest thegns were scarcely more prosperous, in economic terms, than the rillani. Poleham is unusual, but large groups of thegns are not exceptional. Twelve thegns held 7 hides in Purbeck and poterant ire quo volebant. Ten thegns held 3 hides in Cerne (no. 157). Four manors were each held by a group of 9 thegns,-*^ two more were held by groups of 7 thegns,'*^ and Stinsford (no. 358), assessed at zh hides, was held by 6 thegns. Kingcombe (no. 485), assessed at i hide and |- virgate, was held by 10 thegns pro iino mmierio and they still held it in 1086. A certain amount of consolidation had taken place after the Conquest. There are cases where the lands of several thegns had been made into one holding for a single Norman lord. William of Ecouis held the land of 5 thegns in Stourton Caundle (no. 299), assessed at 5 hides, as one manor. Hugh de Lure held land in 3 places [terras in tribus locis) which had belonged to 11 thegns T.R.E. and which was assessed at 5 hides. William de Dalmar had the lands of 3 thegns, assessed at 3 hides, 3 virgates. Some manors had been held as two manors {pro ii maneriis) T.R.E. Piddlehinton (no. 168), which the abbey of Marmoutier held of the Count of Mortain in 1086, had been held T.R.E. by two thegns ^ro ii maneriis. Three thegns had held Wool (no. 208), which Bretel and Malger held of the Count of Mortain in 1086 /)ro ii maneriis. Wintrehiirne (no. 300), held by Walcher of Walscin of Douai in 1086, was held T.R.E. by Alward and x\lwin pro ii maneriis. Roger Arundel's manor of North Poorton (nos. 329 and ci) was held T.R.E. by Alwin and Ulf^ro iihidis (sic). Comparison with Exon. Domesday shows that there were in fact two manors T.R.E., one of i^ hide, held by Alwin, and one of \ hide, held by Ulf. Stafford (nos. 383 and cxv) was held by 3 thegns in parage T.R.E. pro ii maneriis. At first sight this seems like a contradiction in terms, but Exon. Domesday shows that there were in fact 2 manors, one held by two thegns in parage and one held by Leving. At Hurpston (nos. 413 and cxlvi) the wife of Hugh had 3 hides which Alward held T.R.E. and h hide which Sawin held^ro manerio T.R.E. Buckhorn Weston (no. 149) was held by Godric and Bruno in parage pro ii maneriis. Like the Stafford entry this seems to contradict itself. A similar position is revealed by the Exon. entry for Chilfrome (nos. 278 and Ixxxviii) belonging to William of Moyon. The Exchequer text says that 3 thegns held the manor in parage T.R.E. Exon. Domesday repeats this, and adds has Hi mansiones clamat Willelmiis pro ii. This entry also suggests that whereas two or more manors might be given to one lord, and listed as a single manor in Domesday, they were still reckoned as two manors or more. Chilfrome was held by Dodoman and Niel (the Exchequer text says merely duo homines) of William of Moyon, and this is not the only manor of this sort to be held by more than one post- Conquest mesne tenant. Stafford was held by William and Hugh, and Wool by Bretel and Malger.

Some Englishmen can be shown to have survived the Conquest still in possession of at least some of their land, or to have received land which had belonged to other thegns T.R.E. Most of these men are numbered among the king's thegns in 1086, and are

■*' Waia (nos. 380 and cxii), assessed at 4 hides, E. manor in 1086: see pp. 36, gg.

Chaldon (nos. 408 and cxli), assessed at 5 hides, Martins- '" Mappowder (no. 431) and Mapperton (no. 248).

town (nos. 376 and cviii), assessed at 6 hides, and Rolling- Mappowder was assessed at 5 hides, 3 virgates, and

ton (nos. 331 and ciii), assessed at 2^ hides. According to Mapperton at 3 hides, 3 virgates. Exon. Domesday the Q thegns at RoUington still held the

35

A HISTORY OF DORSET

discussed elsewhere.^o Some Englishmen appear as tenants of Norman lords in 1 086,5 ' and a few appear to have become economically dependent upon a Norman lord to the extent of being numbered among the villani in 1086.5^ Apart from these Regenbald was still in possession of his manor of Pulham (no. 146) in 1086, and Eadnoth the staller survived until his death in 1067 and in King William's time obtained Burstock (no. 230) from the thegn to whom it belonged T.R.E. Other thegns continued to hold their land, but as tenants of Norman lords. According to Exon. Domesday the 9 thegns who held Rollington (nos. 331 and ciii) T.R.E. still held it in 1086, although it had passed to Roger Arundel, and was subinfeudated to Robert Attlet. The Exchequer text does not mention the tenure of these 9 thegns. It is probable that the 4 men at Galton (no. 507) who rendered 12s. 4J. to Osmund the baker, to whom the manor belonged, were identical with the 4 free men who held the manor T.R.E. The Geld Rolls reveal other cases in which the English owners of a manor were still holding their land of a Norman. In Cullifordtree hundred a thegn cuius ipsa terra fuit held i hide and i virgate of William Belet. This land can be identified as part of Winterborne Belet or Cripton (no. 493), which 2 thegns held T.R.E. In Godderthorn hundred Berenger Giffard had a piece of land which his predecessor held of him at farm {banc tenet antecessor Berengerii de eo ad fir mam). Berenger held only one manor in Dorset, Bredy (Farm) (no. 317), which was held T.R.E. by Harding. It is possible that he is to be identified as Harding son of Eadnoth the staller, who held several manors in this area both before and after the Conquest. In Uggescombe hundred a thegn held 2\ hides of Aiulf the chamberlain. These 2\ hides were probably part of Aiulf's manor of Tatton (no. 345) held by a thegn of Cerne Abbey T.R.E., who may also have held it of Aiulf in 1084. Exon. Domesday reveals that the 3 thegns holding 3 hides of Cranborne (nos. 16 and xxii) of the king in 1086, held them of Beorhtric son of Aelfgar T.R.E. Brictuin, who held 4 hides of Cerne Abbas of the abbot in 1086, held it likewise T.R.E. Tenants with English names are quite common on ecclesiastical land in 1086. Chetel held Fifehead St. Quintin (no. 133) of Shaftesbur}^ Abbey. Algar and Brictuin appear among the Bishop of Salisbury's tenants at Beaminster (no. 46). Chetel appears as a tenant of Glastonbury Abbey at Sturminster Newton (no. 63) and at Okeford Fitzpaine (no. 64), and Warmund held land at Buckland Ne\\ton (no. 65). ^3 Two widows are mentioned in connexion with ecclesiastical land, one at Piddletrenthide (no. 69) and one at Atrim (nos. 116 and Ixiii).

In contrast to the lay lands the possessions of the religious houses in Dorset suffered no major upset during the transition to Norman rule. Where the names of bishops or abbots are given, they are generally Norman, like Osmund de Seez, Bishop of Salisbury', Maurice, Bishop of London, and Geoffrey, Abbot of Tavistock. Various foreign ecclesiastics and religious houses had received small amounts of land. Odo, Bishop of Bayeux, the king's half-brother, had received Rampisham (no. 55) and Geoffrey, Bishop of Coutances, two small manors called Wintreburne (nos. 56, 57). The canons of Coutances also held a small manor called Winterborne Stickland (no. 142), the only land they had in England. Gilbert Maminot, Bishop of Lisieux, held four Dorset manors, amounting to 26 hides, and Maurice, Bishop of London, held \ hide in Odehatn (no. 62) and i i hide at Hinton Martell (no. 31) which belonged to the church of Wimborne. Another hide of land at Hinton, which had belonged to a priest T.R.E., was held in 1086 by the Bishop of Lisieux. The abbey of St. Stephen, Caen, had 35^ hides in Dorset, consisting of Frampton and Bincombe, which had belonged to Countess

5° See pp. 51-53. " For a further discussion of ecclesiastical tenants and

5' See p. 49. s^ See pp. 15-16. the question of thegnland, see pp. 39-40.

36

DOMESDAY SURVEY

Gytha and Earl Harold respectively.54 The sister house of Holy Trinity, Caen, held the manor of Tarrant Launceston (no. 141) which was the gift of Queen Maud. 55 The abbey of Montevilliers held Friar Waddon (no. 143), given to it by Hugh fitz Grip; the abbey of St. Wandrille held the churches of Bridport, Burton Bradstock, Whitchurch Canoni- corum, and Wareham (nos. 123 and xviii, 124 and xx),^^ and the abbey of Marmoutier held Piddlehinton (no. 168) of the Count of Mortain.57 According to the Geld Rolls the abbey of St. Leger, Preaux, held land in Dorset, and, although Domesday does not mention the fact, the land in question was probably the manor of Stour Provost (no. 231), held by Roger de Beaumont, whose father founded the two abbeys at Preaux. ss

Of the losses sustained by the English houses some dated from before 1066. Earl Harold had taken four manors from Shaftesbury Abbey, two of which were restored by King William in accordance with a writ of King Edward. Queen Edith, Earl Harold's sister, had held Sherborne (no. 37) but Bishop Aelfwold (i 045/6-1 058) had held it previously, and it belonged to the Bishop of Salisbury in 1086. It is possible that Queen Edith had held the manor on a life-lease. One hide at Sherborne had been held by Alward of King Edward T.R.E. sed prius erat de episcopatu. In 1086 it had reverted to the bishopric. Sometimes the churches had lost lands because they had been leased to thegns whose estates had passed to Norman lords. Wulfweard White had held Pentridge (no. 66) of Glastonbury Abbey T.R.E., but in 1086 the manor was held by the king who had taken possession of most of Wulfweard's lands. Clifton and Trill (no. 225) had at one time belonged to the Bishop of Sherborne, but were held T.R.E. by Eadnoth the staller and passed to Hugh, Earl of Chester, with the rest of Eadnoth's lands. 59 Eadnoth had also bought South Perrott (no. 228) and Catsley (no. 229) from Bishop Aelfwold for his own lifetime, on condition that at his death they should revert to the church, but both these manors were held by Earl Hugh in 1086. Stock Gaylard (no. 269), which T.R.E. was held by Toli in pledge de terra Scireburne, had passed to William of Eu who received the rest of Toll's land. Attached to the manor of Silton, belonging to William of Falaise, was one hide which Wulfweard White had bought from the Bishop of Exeter. Some losses, usually involving smaller amounts of land, were due to deliberate seizure for which Hugh fitz Grip, the former sheriff, was largely responsible. From Abbotsbury Abbey he had taken a hide at Abbotsbury (nos. 109 and Iviii) and a virgate at Portesham (nos. 112 and Ixix), which his wife retained by force.^° Tatton, which had belonged to Cerne Abbey T.R.E., was in 1086 held partly by Aiulf the chamberlain, then sheriff (no. 345), and partly by Hugh's wife (nos. 398 and cxxxi), and according to Exon. Domesday two other manors of this abbey, Bloxworth (nos. 79 and xlii) and Affpuddle (nos. 80 and xliii), had been devastated by Hugh. Farnham (no. 135), which T.R.E. belonged to Shaftesbury Abbey, was held in 1086 by Aiulf (no. 352) and Hugh's wife (nos. 396 and cxxix) and a virgate at Kingston (no. 134), belonging to the same abbey T.R.E., was held in 1086 by William of Briouze. Manasses held 3 virgates at Stalbridge (no. 42), belonging to Sherborne Abbey, which William the king's son had given him sine consensu episcopi et monachorum, and the two best hides of the manor of Horton (no. 1 17), belonging to Horton Abbey, had been taken into the king's forest of Wimborne.

The Exchequer text gives the hides in demesne on the ecclesiastical estates, though only rarely on lay estates, whereas Exon. Domesday regularly gives the hides in demesne and the hides held by the villani for both lay and ecclesiastical estates. It is

'* Regesta Regimi Anglo-Normannorum, i, no. 105. " See p. 141.

55 Ibid. no. 149. 56 Jbid. no. no. " For Clifton and Trill, see p. 41.

5' Piddlehinton had belonged to Countess Maud, '"' Adhuc uxor eius vi detinet. The Abbotsbury entr\'

Count Robert's wife, and was given by him to the abbey has given rise to an error in V.C.H. Dors. ii. 49, where vi

after her death: Cal. Doc. France, ed. Round, 435. ('by force') is translated as 'six'.

DO. Ill 27 C2

A HISTORY OF DORSET

noticeable that the term 'demesne' is used in three distinct senses, to designate the portion of a manor held b}^ the lord (the home farm) as opposed to that held by the villani; the demesne (in this sense) with the terra viUanoriun, as opposed to the land subinfeudated to knights or thegns; and lastly, a whole manor which was or should have been held by the lord himself, and not by a tenant. The manor of Stockland (nos. 1 06 and Ixxx) was held of Milton Abbey by Hervey fitz Ansger, but nevertheless was de dominio monachoriim ad victutn et zestitum eorum. Similarly the manor of Little Puddle (nos. 77 and xl), held of Cerne Abbey by William de moiiasteriis, was de propria terra ecdesie, and the 3 hides at Poxwell (nos. 81 and xliv), held by Hugh's wife of the same abbey, were de dominica firma tnonachorum T.R.E. The last case seems to imply an encroachment on the part of Hugh's wife. At Cerneli (no. 212) William held of the Count of Mortain i hide que fuit de dominica firma CERNE T.R.E. , which seems to be a similar use of 'demesne'. Hampreston (no. 443), which Torchil, a king's thegn, held in 1086, had been held by Schelin of the queen but mode tenet rex in dominio. It is probable that Schelin had held some of the queen's manors at farm*' which in 1086 were in the king's hand, and the use of 'demesne' in this context is very unusual. It may be that Torchil held the manor at farm in 1086 or for some service.^^ Other ecclesiastical tenants include Wadard, who held Rampisham of the Bishop of Bayeux, as well as land in other counties of the same lord, and who is mentioned by name in the Bayeux Tapestry.*'^ The two manors of the Bishop of Coutances were held by Osbern. Domesday does not mention any tenants on the land of Gilbert Maminot, Bishop of Lisieux, but the Geld Rolls name Hugh Maminot as his tenant in connexion with land which can be identified with Tarrant Crawford and Preston (nos. 58, 59), and the bishop's other two Dorset manors, Tarrant Keyneston and Coombe Keynes, passed to Hugh Maminot's daughter with the Wiltshire manor of Somerford Keynes.^"*^ The Bishop of Salisbury had subinfeudated several manors. Robert held Up Cerne (no. 34), the wife of Hugh fitz Grip held Bardolfeston (no. 51), and Otbold held Athelhamp- ton (no. 52). Cernel (no. 50) was held by an unnamed woman, and it is worth noting that a manor of the same name (no. 153), belonging to the Count of Mortain, was also held by a woman {qiiedam femina). Hugh's wife held Woodyates (no. 65) of Glastonbury Abbey. John held Leftisford (no. 73) of Cranborne Abbey, and Chetel held Fifehead St. Quintin (no. 133) of Shaftesbury Abbey. BoUo the priest held Shilvinghampton (nos. 113 and Ix) of Cerne Abbey, and a hide at Atrim (nos. 116 and Ixiii), the other hide of which was held by a widow. Aiulf the sheriff held Cerne (nos. 108 and Ixxxii) of the abbey of Milton, which T.R.E. was held by Edric, who could not be separated from the church with this land. Ulviet (Wulfgeat) held Colway (no. 68) of Glastonbury Abbey both T.R.E. and in 1086 and could not be separated from the church. According to Exon. Domesday Bristuin held Woodsford (nos. 82 and xlv) of Cerne Abbey at farm, although the Exchequer text does not mention his tenure. The Bishop of Salisbury's manor of Lyme Regis (no. 36) was held by fishermen (piscatores) who rendered 15s. ad pisces, and Ower (nos. 105 and Ixxix), belonging to Milton Abbey, was held by salt- workers (salinarii) who rendered 20s. a year. Burcombe (nos. 1 1 5 and Ixii) was held by the rillani of Abbotsbury Abbev.

The enfeoffment of knights was frequent on the ecclesiastical estates, especially in the case of the richer houses, that is, Sherborne and Glastonbury. At Alton Pancras (no. 33), belonging to the Bishop of Salisbury, Edward and Pain held zh hides each and at Sherborne (no. 37) the knights of the bishop held 22 i hides and included

'" See p. iig. <" Bayeux Tapestry, ed. F. M. Stenton, 21.

'^ For the king's thegns and their tenure, see p. 52. '* See pp. 60, 129.

38

DOMESDAY SURVEY

Otbold, Sinod, Ingelbert, Ralph, Waleran, and the wife of Hugh. At Stalbridge (no. 42) Lambert held 2 hides, and at Beaminster (no. 46) the knights held 10 hides and a virgate. Two of them were obviously French (H. de Cartrai and Sinod) but the names Algar and Brictuin show that two others were English. At Netherbury (no. 47) Tezelin had 5 hides, 3 virgates, William and Godfrey 2 hides each, and Serle li hide. Walter held Buckham (no. 54) and two knights, Walter and William, held Chardstock (no. 49). Three knights, Godfrey, Osmar, and Elfric, held Bowood (no. 53), which three thegns held T.R.E. Two of these knights also must have been English. Of the Glastonbury manors Okeford Fitzpaine was held by knights, namely the wife of Hugh, who held 4 hides, and Alvred of Epaignes and Chetel who held 2 hides each. Four thegns had held the manor T.R.E. At Sturminster Newton (no. 63) Waleran held 6 hides, Roger i hide, and Chetel i hide, and at Buckland Newton (no. 65) the wife of Hugh had 7 hides, li virgate, and Warmund 2 hides. A knight and a widow held 3 hides at Piddletrenthide (no. 6g) belonging to the New Minster, Winchester.

The meaning of the term 'thegnland' which occurs in connexion with some ecclesiastical land is obscure. It does not seem to have been held by military service, although a French knight {miles francigenus) held 2 hides of thegnland at Nettlecombe (nos. 88 and li). In some cases it seems to have been liable to some kind of service. At Cerne Abbas (nos. 76 and xxxix) Brictuin held 4 hides of land, which he also held T.R.E. et non potuit recedere ab ecclesia. Exon. Domesday records that the land was thegnland and that Brictuin rendered 305. to the church excepto servitio. A similar entry is that of Cranborne (nos. 16 and xxii), where 3 thegns held 3 hides of land (not specifically said to be thegnland) for which they rendered £2i excepto servitio. Exon. Domesday adds that they held the same land T.R.E. of Beorhtric and non poterant separari ab eo. It is not altogether clear from these entries whether the thegns both performed service and rendered money, or rendered money instead of doing service. Other evidence favours the latter interpretation. Durnford (Wilts.) was held T.R.E. by 3 Englishmen, two of whom paid 5^. while the third serviebat sicut tainusfi^ At Winsford (Som.) there was | hide which 3 thegns held T.R.E. et serviebant preposito manerii per consiietiidinem absque omnifirma donante. If these thegns did not contribute to the farm because they performed some service, then presumably the Dorset thegns who paid money renders did not serve. The \ hide attached to Winsford is entered twice, appearing again under the name of Robert de Odburville, who held it in 1086. Accord- ing to the second entry the land was judged to be thegnland {modo diratiocinata est in tainland).^^ This second entry implies that the thegns rendered service in the capacity of foresters, and suggests that the thegnland was set aside for men serving in a ministerial capacity. But from other entries it is plain that the term could be used simply to describe land once held by thegns. At Loders (nos. 13 and ix) there were 2 hides of thegnland que non ibi pertinent , which 2 thegns held T.R.E. In the Geld Roll for Loders hundred (which consisted solely of the manor of Loders), it is stated that ii hide quas tenuerunt tagni tempore regis Edwardi sunt addite hide mansioni. There is no reason to suppose that these thegns owed either service or money to the manor of Loders. At Hinton a priest held a hide of thegnland T.R.E. et poterat ire quo volebat. This terminology does not suggest land owing a service to the holder of Hinton.^^ Jn 1086 the land was in the king's

<" Dom. Bk. (Rec. Com.), i, f. 67b; V.C.H. Wilts, ii. "' The statement that a manor owed service is in itself

81. There is a similar entry in Dorset relating to Tatton ambiguous. It could mean service like that described in the

(nos. 345, 398 and cxxxi). Part of Tatton had been held Wmsford entries, or it could mean the rent in money or

T.R.E. by a thegn of Cerne Abbey, who won poterat ab ea kind derived from the manor. When the Som. Domesday

separari, while the other portion was held by 2 thegns of states, in enumerating the manors in Som. taken from the

the same abbey for rent (prestito). abbey of Glastonbury, that the church had lost the

** Dom. Bk. (Rec. Com.), i, flf. 86b, 98b. service (ecclesia servitium inde non habet, ecclesia servitium

39

A HISTORY OF DORSET

demesne. Most of the thegns holding land of the ecclesiastical tenants in 1086 are not said to hold thegnland, and they were presumably just men with a thegn's wergild. At Long Bredy (nos. 87 and 1) an English thegn {teignus anglicus) had a hide of land worth ;^3. Six thegns held 8i hides at Sherborne (no. 37) and 2 thegns held 2\ hides at Stoke Abbott (no. 45). The thegn who held part of Farnham (no. 352) of Shaftesbury Abbey T.R.E. et tion poterat ab ea separari must have been commended to the church and may have owed some service like that of the 3 thegns at Winsford. He is probably identical with Alwin, who held the other part of Farnham (nos. 396 and cxxix) of the abbey T.R.E. et tion poterat ab ea separari. Two free Englishmen {angli liberi) held 4 hides at Handley (no. 125) of Shaftesbury Abbey. Three thegns held Bowood (no. 53) T.R.E., which was held by the same number of knights in 1086, 2 of them, Osmar and Elfric, apparently being English.

There are a few references to churches in the Dorset survey. Bristuard the priest held the churches of Bere Regis and Dorchester, with the tithes and i hide and 20 acres of land (nos. 144 and xiii). The church of Gillingham was given to Shaftesbury Abbey in exchange for a hide of land at Kingston (no. 134) in which Corfe Castle was to be built.^* The church of Winfrith Newburgh, with a virgate of land, and the churches of Puddletown, Chaldon, and Fleet, with i| hide of land, were held by Bollo the priest (nos. 145 and xvi, 145a and xix). The churches of Burton Bradstock, Bridport, and Whitchurch Canonicorum belonged to the abbey of St. Wandrille (nos. 123 and xviii) and so did the church of Wareham (nos. 124 and xx). Another church in Wareham, pro- bably the 11th-century church of St. Martin, ^^ and a chapel {ecclesiola) in Wimborne Minster belonged to Horton Abbey (no. 117). To Wimborne belonged i^ hide and | virgate in Hinton (no. 31) which Bishop Maurice held in 1086. It is evident that this is not a comprehensive list of all the churches in Dorset in 1086. There must have been a Saxon church at Sherborne, and this is confirmed by the survival of a Saxon doorway in the west wall of the present building. ^o As their names suggest there must also have been Saxon churches at Yetminster (no. 35), Charminster (no. 32), Beaminster (no. 46), which belonged to the Bishop of Salisbury, Iwerne Minster (no. 131), belonging to Shaftes- bury Abbey, and Sturminster (no. 232) belonging to Roger de Beaumont. Three priests are recorded at Hinton (no. 31), two of whom still held land in 1086. One lived at Tarente. There was a priest at Church Knowle (no. 235) and another at Bleneford (no. 455), both of them being enumerated in conjunction with the peasants. Bristuard the priest and Bollo the priest have already been mentioned and Godric the priest occurs among the king's thegns. Walter the deacon (diaconus) held Cernel (no. 147) as an almsman.

The most prosperous of the ecclesiastical landowners and after the king the wealthiest man in Dorset was the Bishop of Salisbury. The ancient see of Sherborne, founded by Ine in 705, and numbering Aldhelm and Asser among its bishops, had been restricted to the county of Dorset since the reign of Edward the Elder.^' In 1058 Bishop Herman united the sees of Sherborne and Ramsbury and between 1075 and 1078 the episcopal seat was transferred to Salisbury.^- This amalgamation of Sherborne and Ramsbury explains the size of Bishop Osmund's fief, which consisted of

perdit) this could be taken to mean the income from the rather large.

manor, not any specific service: Dom. Bk. (Rec. Com.), i, '* castellum WARHAM in Domesday, but later

f. 91. Abbotsbury Abbey was entitled to 6 a. of crops and evidence indicates that Corfe Castle is meant: see p. 83.

3 church-scots de consuetudine from Friar Waddon (no. '"> G. Baldwin Brown, x'Jr^s i« i'ar/)' £';i^. (1925), ii. 484.

143) and this in turn could be described as service. In '" Ibid. 477-8.

Som. Brictric and Ulward held Buckland (Dom. Bk. (Rec. " F. M. Stenton, Anglo-Saxon Eng. 433.

Com.), i, f. 98b) as king's thegns. They had held the land '^ W. Stubbs, Registrum Sacrum Anglicamim, 35. The

of Bp. Peter et reddebant ei x solidos, but the king had had transfer was sanctioned by the Council of London in 1075

nothing since the bp.'s death. It is possible that they paid and the removal took place between that date and Her-

this sum instead of doing service, although the amount is man's death in 1078.

40

DOMESDAY SURVEY

267 hides in Wiltshire, 8 hides in Somerset, and the two large manors of Sonning in Berkshire73 (assessed at 60 hides T.R.E. but reduced to 24 hides in 1086) and Dunsden in Oxfordshire (assessed at 20 hides), with just under 100 hides in Dorset, where the monks of Sherborne held 119^ hides. In addition the bishop and the monks held between them about 40 carucates of land in Dorset which never paid geld. The original endowment of the Bishop of Sherborne seems to have been 300 hides. A letter from Bishop Aethelric to Aethelmaer, dating from the early nth century, complains that he is not receiving ship-scot from 33 out of the 300 hides which his predecessors had for their diocese.74 He itemized the deficit as i hide at Btibbanciimbe, 2 hides at Alton Pancras, 7 hides at Up Cerne, 6 hides at Clifton, 5 hides at Hewish, 2 hides at Trill, i hide at Wyllon, 5 hides at Buchaematime, 3 hides at Dibberford, and i hide at Peder. Alton Pancras and Up Cerne (nos. 33, 34) were both held by the bishop in 1086. Clifton Maybank and Trill (no. 225) were held by Earl Hugh as the successor of Eadnoth the staller. Before the Conquest Eadnoth had bought two manors, Catsley and South Perrott (nos. 228, 229), of Bishop Aelfwold, on condition that at his death the manors should revert to the church; Earl Hugh, however, held them in 1086. It seems possible that South Perrott is the Peder of Bishop Aethelric's letter, but this has been disputed.75 Dibberford lies in Dorset, but does not appear in the Dorset survey.'^

Sherborne was originally a house of secular canons, but Bishop Wulfsige (992-1001) expelled the clerks and replaced them by monks.'? A dubious charter of Aethelred II, dated 998,^8 confirms to the church the manors of Bradford Abbas, Over Compton, Oborne, Stalbridge, Stalbridge Weston, Thornford, and Lyme Regis, all of which, with the exception of Lyme, were de victii monachorum Scireburne in 1086. Lyme was held by the Bishop of Salisbury in 1086. It had never paid geld and was held by an unspecified number of fishermen who rendered 155. for the fish (adpisces). The bishop had a house there worth 6d. In a charter dated 774 Cynewulf, King of the West Saxons, gave Lyme to the church of Sherborne for a salt-pan. 79 Bradford Abbas and Stalbridge were given to the church by Aethelstan^" and Oborne by Edgar.^' Aethelstan's charters give the extent of Bradford as 10 hides and of Stalbridge Weston as 8 hides and these reckonings agree with the Domesday assessment of the manors, but in Aethelred's confirmation charter Bradford is reckoned as 7 hides and Stalbridge Weston as five. In Edgar's charter Oborne is reckoned as 5 hides, as in Domesday, but in the confirmation charter it is reckoned as ten. Eadred is supposed to have given 8 cassati in Thornford to Wulf- sige II in 951, with a reversion to the church. 8- Thornford was reckoned as 15 cassati in the confirmation charter and in Domesday was assessed at 8 hides. Another charter of Aethelred II, dated 10 14, gave 13 hides in Corscombe to Sherborne,^^ and in 1035 Cnut

'3 Potterne, Cannings, Ramsbury, and Salisbury salt-workers are recorded at the Bp. of Salisbury's manor

(Wilts.), and Sonning (Berks.) were the endowment of the of Lyme, there were 1 3 salt-workers at the manor of Colway

Bp. of Ramsbury. (no. 68) held by the abbey of Glastonbury, and 14 salt-

'* F. E. Harmer, Anglo-Saxon Writs, no. 63 (dated workers at the manor of Lyme held by William Belet (no.

1001/2-1009/12), where it is pointed out that the Bp. of 504).

Salisbury held 3 hundreds in Dorset, Yetminster, *" Cart. Sax. nos. 695, 696. Stevenson appears to

Beaminster, and Sherborne. accept them as genuine (Asset, Life of King Alfred, ed.

'5 For the alternative identifications of Peder, see W. H. Stevenson, 148 n.). Cart. Sax. no. 695 calls Sher-

Harmer, op. cit. 485. The Domesday form of the name borne a monasteriuni, but this could mean a minster.

S. Perrott is Pedret. *' A. J. Robertson, Anglo-Saxon Charters, no. i ; Cart.

'"' Dibberford lay in Dorset in 1252, when it was held Sax. no. 1308.

by Grece de Mucegros ; there is no evidence that it belonged *^ Cart. Sax. no. 894. The phrase ad refectionem familie

to the Bp. of Salisbury at that date: Bk. of Fees, 1267. Scireburnensis ecclesie suggests a community, and Eadred

" Stenton, Anglo-Saxon Eng. 450. styles himself Occidentaliiim Saxonum rex, whereas his

" Cod. Dipl. no. 701. Stevenson seems to accept this normal style is rex Anglorum. Stevenson regards this

charter (E.H.R. xxix. 689), but Miss Harmer {Anglo- charter as 'doubtful or spurious' : E.H.R. xxix. 692 n.

Saxon Writs, 485) does not regard it as genuine. '^ Cod. Dipl. no. 1309. Both Miss Harmer (Anglo-

'« Cart. Sax. no. 224. Stenton accepts it as a genuine Saxon Writs, 553) and Miss Robertson (Anglo-Saxon

charter : E.H.R. xxxiii. 443 n. Although no salt-pans or Charters, 387) seem to accept it as genuine.

A HISTORY OF DORSET

gave 1 6 niatise in the same place to the monastery.^^ In io86 Corscombe (no. 44) was assessed at 10 hides less a virgate.

The last five manors entered in the bishop's fief are preceded by the heading Has terras que stibterscribiintur habet episcopiis pro excambio de Scipeleia. He also held Chaddemvick in Mere (Wilts.) in exchange for Scipeleia. There is no indication where Scipeleia was or to whom the exchanged manors had belonged. The Dorset manors amount to over 20 hides and Chaddemvick is assessed at 5 hides, so Scipeleia should have been a sizeable manor. This consideration eliminates Shipley in Yorkshire, assessed at 3 carucates,^^ and Shipley in Derbyshire, assessed at 2 carucates.^^ William of Briouze held a manor called Shiplev in Sussex, but the fact emerges only obliquely in the account of Fulking and this Shipley is omitted from the Sussex Domesday. ^^ Xo other manor of this name occurs in Domesday and the exchange remains a myster}'.

If Sherborne was the largest of the Dorset monasteries, the smallest was Horton, a poorlv endowed little house with only the manor of Horton (no. 117) in Dorset and 3 hides in Devon, a total of 10 hides in all. It may conveniently be discussed here since in 1 122 it was amalgamated with Sherborne and became a cell of that abbey. ^^ It seems to have been founded between 1033, when Cnut gave 7 manse in Horton to his minister Bovi,^^ and 1061, when Edward the Confessor freed the monks of Horton from all duties except geld, the repair of fortifications, and the building of bridges.'" According to William of Malmesbury Horton was founded by Ordwulf son of Ordgar, who also founded Tavistock Abbey, but he is last recorded in looS.^i It has been noted that Littleham (Devon), belonging to Horton Abbey in 1086, had been given by Edward to his minister Ordgar in 1042, and suggested that this Ordgar was a descendant of Ordwulf son of Ordgar and that the founder of Horton Abbey was a member of the same family.'- William of Malmesbury stated that the land which Ordwulf left to Horton Abbey was seized by the Abbot of Tavistock, and in 1086 the abbey of Horton was claiming the manor of Antony (Cornw.) from Tavistock Abbey .'^

The largest nunner}- in Dorset, and indeed in the whole of England, was Shaftesbury, which possessed about 360 hides, including 172 hides in Wiltshire, 161 hides in Dorset, the manor of Falcheham (Suss.), and 10 hides in Somerset.''* It was founded by King Alfred, whose daughter Aethelgifu was its first abbess.'^ Alfred left her a bequest of 100 hides"' of which 40 hides were at Donhead St. Andrew (Wilts.) and Compton Abbas, 20 hides at Handley and Gussage St. Andrew, 10 at Tarrant, 15 at Iwerne Minster, and 15 at Fontmell INIagna, all in Dorset. All these manors were in the possession of the abbey in 1086. Donhead St. Andrew was assessed at 40 hides, Compton Abbas at 10, and Iwerne Minster at 18, but the other hidages were the same as in Alfred's bequest. With the exception of Donhead St. Andrew these manors comprised the Domesday hundreds of Sexpene and Handley (later amalgamated under the name Sixpenny Handley). According to two charters preserved in the abbey's cartulary Mapperton (no. 137) was given by Edmund to Eadric the ealdorman in 943, when it was reckoned

*■* Cod. Dipl. no. 1322. granted by Aethelberht, King of Wessex, to Sherborne in

85 Doni. Bk. (Rec. Com.), i, f. 318. 864.

** Ibid. f. 277b. " Wm. of Malmesbun.-, Gesta Pontificum Anglorum

" Ibid. f. 29b. Round suggested that Shipley (Suss.) (Rolls Ser.), 203.

was included in William of Briouze's manor of Thakeham, '^ H. P. R. Finberg, 'The House of Ordgar and the

assessed at 20 hides and 3 virgates: I'.C.H. Suss. i. 440 n. Foundation of Tavistock Abbey', E.H.R. Iviii. 190-201.

'* Regesta Regttm .-Inglo-N ormamwrum , ii, no. 1325. " Doni. Bk. (Rec. Com.), i, f. 121.

*» Cod. Dipl. no. 1318. Miss Harmer (Anglo-Sa.xon •>* Kilmington (Som.) was given to the abbey by Serle

Writs, 576) seems to accept it as genuine. Urk, who attests of Burcy when his daughter became a nun there : Dom. Bk.

it, was the founder of Abbotsbur\-. (Rec. Com.), i, f. 98.

''"Cod. Dipl. no. 1341 ; Robertson, Anglo-Saxon '>^ Asser, Life of King Alfred, S5.

Charters, no. cxx. IMiss Robertson points out that this " Cart. Sa.x-. no. 531; Cod. Dipl. no. 310; Robertson,

charter, which is incomplete, is almost identical with one Anglo-So-von Charters, no. xiii.

42

DOMESDAY SURVEY as II manse,'^^ and Hinton St. Mary {Hamtiine) (no. 126) was given by Edmund to Wulfgar in 944, when it was reckoned as 5 manse.'>^ Mapperton was assessed at 1 1 hides in 1086 and Hinton St. Mary at eight. Another charter of Edmund, dated 942, concerns the manor of Cheselbourne. According to this charter Edmund restored 7 manse of land at Cheselbourne to Wynflaed, a rehgious woman, with an additional grant of 8 manse in the same place. ^9 According to Domesday Cheselbourne was one of the manors taken by Earl Harold from the abbey and restored by King William in accordance with a writ of King Edward. It was then assessed at 16 hides. Earl Harold had also taken another manor, Pidele, from the abbey, which was not returned and was held by the Count of Mortain in 1086. According to a charter in the abbey's cartulary Edgar in 966 restored to the church 10 cassati of land at Uppidelen, which had originally been given by Wynflaed, described as his grandmother {ava), whose charter had been lost through carelessness.' Of the four manors called Pidele held in 1086 by the Count of Mortain, only one, Piddlehinton (no. 168), assessed at 10 hides, is large enough to be identifiable with Uppidelen. It had been held T.R.E. by two thegns^ro ii maneriis, and it is not said ever to have belonged to Shaftesbury Abbey. Uppidelen has in fact been identified as part of Piddletrenthide (no. 69), 2 held T.R.E. by Almar and Alverd of King Edward, which belonged in 1086 to the New Minster, Winchester. Neither Pidele (whether it be identified with Piddlehinton or Piddletrenthide) nor Melcombe (no. 30) was ever returned to the abbey, but half the hide in Farnham (no. 135), taken from the abbey by Aiulf and the wife of Hugh, was returned. Aiulf the chamberlain restored it to the abbey when his daughter became a nun there, and added the manor of Blandford (no. 336) for the soul of his wife. Drew of Montacute's daughter also became a nun at Shaftesbury, and on this occasion he gave to the abbey his manor of Nyland (no.

I50).3

Apart from the two great houses of Sherborne and Shaftesbury the Dorset abbeys were quite small. Cerne Abbey was the foundation of Aethelmaer son of Aethelweard, patron of Aelfric the homilist. Aelfric was responsible for teaching at Cerne, and later became Abbot of Eynsham. Aethelmaer has been identified with the earl of the western provinces to whom Bishop Aethelric addressed his complaint about ship-scot inen- tioned above."* His foundation charter of 987^ gave to the abbey Cerne Abbas itself, with ID mafise in Winterborne, the two manors of Littlebredy and Long Bredy, reckoned at 12 and 16 manse respectively, and 3 manse in Renscombe. Leofric the clerk of Poxwell gave Poxwell, and Aelfrith, a relative of Aethelmaer, gave 4 cassati at Puddle. Alfwold gave 5 manse at Bloxworth. All these manors belonged to the abbey in 1086. Winter- bourne Abbas was still assessed at 10 hides, Littlebredy and Long Bredy were assessed at II and 9 hides respectively, and Renscombe at 5 hides, i virgate. Poxwell was a manor of 6 hides and Bloxworth of five and a half. Two manors called Puddle were

" Cart. Sax. no. 781. Eadric minister attests nos. 763, Saxon Charters, 281-2. 765, and 767, and no. 769 is a grant to Eadric vassalus, ' Cart. Sax.no. 11 86. The Wynflaed of this charter may- dated 941, of Beechingstoke (Wilts.) by Edmund, which be the same woman as the Wynflaed of Edmund's charter Stevenson considered 'may be genuine' : Asser, Life of relating to Cheselbourne. The Wynflaed who received the King Alfred, 255. Eadric attests Cart. Sax. no. 775 (dated grant of Cheselbourne has been identified with the woman 942) as dux. who bequeathed Chinnock (Som.) to the abbey c. 950:

'* Carf. 5ax. no. 793. For the identification of //awdoie D. Whitelock, .4nglo-Sa.xon Wills, no. iii and nn. ; cf.

with Hinton St. Mary, see A. Fagersten, Place-Names of Robertson, Anglo-Saxon Charters, 379-80.

Dorset, 41 and n. ^ Saxon Charters of Dorset (Proc. Dorset Nat. Hist, and

" Cart. Sax. no. 775. Two charters of Aethelred I are Arch. Soc. lix), 107.

preserved, granting 7 manse of land at Cheselbourne to ' Regesta Region Anglo-Normannorum, ii. 346-7.

Earl Aelfstan: Car/. Sa.v. nos. 52s, 526. One is printed in •'See p. 41. For the identification, see Robertson,

Robertson, Anglo-Saxon Charters, no. xii. Stevenson Anglo-Saxon Charters, 386-7; Whitelock, Anglo-Saxon

regarded no. 525 as doubtful or spurious: E.H.R. xxix. Wills, 144-5; Harmer, Anglo-Saxon Writs, 553.

692 n., 698 n. Cnut gave i6 hides at Cheselbourne to ' Cod. Dipl. no. 656. Dr. Whitelock (Anglo-Saxon

Agemund in 1019: Cod. Dipl. no. 730. These 3 charters Wills, 145) seems to accept it as genuine, and that of Edmund are discussed in Robertson, Anglo-

43

A HISTORY OF DORSET

held by the abbey in 1086, Affpuddle assessed at 9 hides, and Little Puddle assessed at tvvo and a half. In all Cerne Abbey had 121 i hides in 1086, all in Dorset.

Milton Abbey had i2oi hides in Dorset and two manors in Glanvilles Wootton (nos. 284, 285) had also belonged to this abbey T.R.E. Its register was destroyed by fire in 1309, but there are in existence two versions of a charter attributed to Athelstan, one in Latin and the other in English.^ If these documents represent a genuine charter of Athelstan, he gave to the church 26 hides at Milborne, 5 at Woolland, three at the mouth of the Frome 'on the island, two on sea and one on land, that is to say at Ower', 3 at Clyffe, 3! at Lyscombe, i at Burleston, i at Little Puddle, 5 at Cattistock, 6 at Compton, 2 at Whitcombe, 5 at Osmington, and 6 at Holworth. In addition, he gave 30 hides at Sydling St. Nicholas for victuals, 2 hides at Chelmington and 6 at Hillfield, 10 hides at Ercecombe 'to timberlond', and a weir on the Avon at T\\yneham, with 12 acres to support it. With the exception of Chelmington and Hillfield these lands all belonged to the abbey in 1086. Milborne was presumably Milton Abbas itself, assessed at 24 hides. Woolland in 1086 was still assessed at 5 hides and Ower at 3,'' but the assessments of the other manors had changed from the earlier reckonings. Clyfi"e was assessed at 2 hides, Lyscombe at 3, Burleston (Puddle Burston) at 3, Little Puddle at 2, Cattistock at 10, Compton at 5, Whitcombe at 6, Osmington at 10, and Holworth at five. Sydling St. Nicholas was assessed at 29 hides. Ercecombe appears in Domesday as Ertacomestoche (nos. 106 and Ixxx), and can be identified as Stockland (Devon), which lay in Dorset at that date. In 1086 it was worth ^^9 and ftiit semper de dominio monachorum ad victum et vesttim eorum. The abbey held 12 acres on the Avon in 1086 and there had once been a fisher}' there. ^ Since it is known that in 964 Edgar expelled the clerks from Milton Abbey and replaced them with monks under Abbot Cyneweard,^ Athelstan's grant, if genuine, must have been made to a community of clerks.

Abbotsbury Abbey, with 75 hides in Dorset, was founded by Urk, who had been a housecarl of both Cnut and Edward the Confessor, and his wife Tole.'° In 1024 Urk received 7 manse in Portesham from Cnut, and in 1044 5 perticas in Abbott's Wootton from Edward the Confessor." Both these manors belonged to the abbey in 1086, when Portesham was assessed at 12 hides and Abbott's W^ootton at two and a half. In a writ dating from between 1053 and 1058 Edward the Confessor commanded that his house- carl Urk should have his shore, with right of wreck. This is presumably a reference to Chesil Beach. A second writ of Edward, dating from between 1058 and 1066, gives permission to his 7naim Tole, Urks's widow, to bequeath her land to the abbey of Abbotsbur}% which he takes under his protection." One of the manors which the abbey derived from Tole must have been Tolpuddle, which bears her name.'^ William I issued two writs concerning the land and rights of the abbey, both addressed to Hugh fitz Grip, whose encroachments on the land of this and other abbeys have already been mentioned. ■"*

' Robertson, Anglo-Saxon Charters, no. xxiii and nn. Tole to the abbey still exists, but in a ven.' mutilated

The Eng. version is printed in Car/. 5a.v. no. 738 and Corf, condition, which makes it impossible to read: O.S.

Dipt. no. 1119, and the Latin version in Cart. Sax. no. 739 Facsimiles of Anglo-Saxon MSS. Pt. II (1881), p. xv, and

and Cod. Dipl. no. 375. 'Earl of Ilchester', no. \'. The original charter, with some

' Ower could not be ploughed in 1086, and was held by others relating to Abbotsbury-, is at Count>- Hall, Dor- salt-workers. Chester.

' Dom. Bk. (Rec. Com.), i, f. 43b. " Cod. Dipl. nos. 741, 772. Both charters seem to be

' Anglo-Saxon Citron., a revised translation ed. D. accepted as genuine by Miss Harmer: .(4ng/o-iSajfon PfVitf,

Whitelock and others, 76. 576.

'" Regesia Regum Anglo-Nonnannonim, i, no. 108. '^ Printed in Harmer, op. cit. nos. 1-2.

Urk's charter founding the guild of Abbotsbury is trans- "In 1212 the abbey held Abbotsbur>', Portesham,

lated by D. Whitelock, in Eng. Hist. Doc. i, no. 139. The Hilton, Tolpuddle, and Abbott's Wootton, qiie data

Latin text is printed in Cod. Dipl. no. 942. It is one of the fuerant per Oro (recte Ore) et Tolam uxorem suam: Bk. of

few known examples of guild statutes, providing for the Fees, 92.

needs of the guildsmen and the minster. A charter of '■• Regesta Regum Anglo-Normatmorum, i, nos. 109, 203.

44

DOMESDAY SURVEY

Cranborne Abbey was poorly endowed, having only 21 hides in Dorset and the same number in Wiltshire. Hugh fitz Grip gave to this abbey a piece of land in Gillingham, which he received from the king's farm, and i hide at Orchard (nos. 422 and civ) pro anima sua. The manor of Cranborne belonged to the king in 1086, having been one of the manors which passed to Queen Maud from Beorhtric son of Aelfgar.'s When Robert fitz Hamon received the land which had once belonged to Beorhtric, he became the patron of Cranborne Abbey, and in 1102 made it a cell of Tewkesbury."^

Of the houses not situated in Dorset but holding land there, the most important was the abbey of Glastonbury, the richest house in England. The 52 hides held by the abbey in Dorset were only a fraction of its huge fief, totalling about 800 hides. The largest manors of the abbey in Dorset were Sturminster Newton and Buckland Newton. Each of these manors was the head of a hundred, and the two hundreds themselves were later amalgamated to form Buckland Newton hundred. Sturminster Newton had been bequeathed by Alfred to his younger son Aethelweard,'^ and according to a charter preserved in the Glastonbury cartulary it was given to the abbey by Edgar in 968. '^ Among the other religious houses with land in Dorset were the New Minster at Winchester (Hyde Abbey), with Piddletrenthide (no. 69) which had belonged to Roger Arundel ; Athelney Abbey, which held Purse Caundle by an exchange with the Count of Mortain, who received Bishopston (Montacute) in return ; Tavistock Abbey, with two small manors totalling 5 hides; and Wilton Abbey with Didlington and Philipston {Winburne).

The land of the king's almsmen follows the account of the bishops' and abbeys' land. Bristuard the priest held the churches of Dorchester and Bere Regis (nos. 144 and xiii) with their tithes and i hide, 20 acres, of land. BoUo the priest held the churches of Winfrith Newburgh, with a virgate of land, and the churches of Puddletown, Chaldon, and Fleet, with i\ hide of land (nos. 145a and xvi, 145b and xix). He held land as a king's thegn as well, and was a tenant of Abbotsbury Abbey at Atrim. Walter the deacon (diacomis) held Cernel (no. 147), and Bernard held of him. But the most important of the king's almsmen was Rainbald (Regenbald) the priest, who held the manor of Pulham (no. 146). He is undoubtedly to be identified as Rainbald of Cirencester,'^ who held the post, if not the name, of chancellor under Edward the Confessor. He is called Rainbald canceler in the Herefordshire survey.^" He had held Pulham T.R.E., assessed at 10 hides. William I confirmed his lands to him,^' and in 1086 he held 67 hides in 5 counties, besides 8 carucates in Somerset. Some of this land had belonged to him T.R.E. and he had obtained some of it after the Conquest. ^-

III

In 1086 the greatest lay landowner in Dorset after the king was Robert, Count of Mortain, the king's half-brother, with 190 hides. His Dorset lands were a mere appendage of his vast estates in Cornwall, where he held virtually the whole county. He was probably the richest man in England apart from the king, with lands scattered in many areas, particularly the south-west, Yorkshire, Northamptonshire, and Sussex.

■5 There is a tradition that the abbeys of Tewkesbury Glastonbury, vol. Hi (Som. Rec. Soc. Ixiv), 592.

and Cranborne were founded by Aelfweard, said to be the " He is called Rainbald of Cirencester in the account of

grandfather of Beorhtric : Dugdale, Man. iv. 465; V.C.H. Berks.: Dom. Bk. (Rec. Com.), i, f. 63. In 11 30 Alvred of

Dors. ii. 70. Lincoln paid 60 silver marks to have the manor of Pulham

■* Ann. Mon. (Rolls Ser.), i. 44; D. Knowles and R. N. de honore Cirecestr' : Pipe R. 11 30 (Rec. Com.), i6.

Hadcock, Medieval Religious Houses, 63. ^^ Dom. Bk. (Rec. Com.), i, f. i8ob.

" Select Eng. Hist. Doc. ed. F. E. Harmer, 17. For the ^' Regesta Regum Anglo-Normannorum, i, no. 19.

identification, see A. Fagersten, P/ace-Afames 0/ Dorset, 47. ^' For a full account of Rainbald, see Round, Feudal

'* Cart. Sax. no. 1214; The Great Chartulary of Eng. 421-30.

45

A HISTORY OF DORSET

In all, he held over 790 manors in 20 counties. One of his Dorset manors, Piddlehinton, had apparently belonged to his wife, Maud.-' He derived his lands from a number of small thegns, but Edmar, who held his more important manors in Dorset, seems to have been a man of considerable wealth. He is probably identical with the Edmer who held land in Somerset and Devon which later passed to the count, and possibly identical with Edmer attile or atule who held land in Hertfordshire, Micidlesex, and Berkshire which belonged to the Count of Alortain in 1086.--* Count Alan, lord of Richmond in Yorkshire, held the 15-hide manor of Dewlish in Dorset. He held no other land in the area, and his possession of this solitary' manor is rendered more inexplicable by the fact that it had formerlv belonged to Beorhtric son of Aelfgar, whose lands generally passed to Queen jMaud. Hugh, Earl of Chester, held 35 hides in Dorset in consequence of his acquisition of the lands of Eadnoth the staller, who had held all but two of the Dorset manors held by Earl Hugh in 1086. Aubrey de Couci, sometime Earl of Northumbria, had held Gussage St. Michael in Dorset which in io85 was in the king's hand.--'' He had held several manors in Wiltshire and Gussage appears in the Wiltshire survey with the rest of his land. The Countess of Boulogne held 3 manors in Dorset, Winterborne Monkton, Bockhampton, and Swanage, all of which had been held by Wulfgifu T.R.E. The Geld Roll for Ailezes-zfode hundred refers to the tenure by Count Eustace (of Boulogne) of a manor which can only be Swanage.-^

The wife of Hugh fitz Grip held 116 hides in Dorset and 3 hides at Damerham (Hants), the latter as tenant of Glastonbur}^ Abbey. Her name is not recorded in Domesday, but in an inspeximns of Philip IV, dated 1305, concerning the land of the abbey of IMontevilliers, there is a charter by \\-hich Iladuidis, filia Xicolai de Baschel- rilla, uxor Hugoiiis de \ arhaii (Wareham) filii Griponis, ga\e the manor of Waddon to the abbey. Friar Waddon (no. 143) belonged to this abbey in 1086 by the gift of Hugh fitz Grip. It seems likely that the charter is genuine, and that Hugh's wife was Hadwidis or Hawise de Baschelville.-" Hugh fitz Grip, late Sheriff of Dorset, had held 18A hides of the queen, which had reverted to the king. The Domesdav survev itself supplies most of what is known about him, but two writs, both concerning the land of Abbots- bury Abbey, are addressed to him as sheriff.-^ Three Dorset abbeys, Shaftesbur}% Abbotsbury, and Cerne, had suffered losses at his hands, and he seems to have been responsible for the devastation of the Dorset boroughs. He also appropriated a virgate of \\'illiam of INIoyon's manor of Winterborne Houghton (nos. 275 and Ixxxv) and gave to Brictuin the manor of Little Waddon (no. 460) in exchange for a manor worth twice as much. He was dead by 1084, since Aiulf the chamberlain appears as sheriff both in the Geld Rolls and in Domesday. The wife of Hugh held 28 hides as a mesne tenant, 6 hides of the Bishop of Salisbury, 15 hides of Glastonbury Abbey (excluding Damer- ham), a piece of land in Purbeck of William of Briouze, and Ailwood (no. 482) of Swain.

Aiulf the chamberlain held just over 55 hides in Dorset, 6 in Wiltshire, and 10 in Berkshire. Chine (no. 351) belonged to him as long as he was sheriff {qitaiudiii erit zicecomes) and Lulworth (no. 350) had belonged to Alfred, the Saxon sheriff. Aiulf's largest manor, Wootton Fitzpaine (no. 347), was held T.R.E. by Beorhtsige, a thegn of King Edward {miles regis Edzvardi).-^ None of his other Saxon predecessors was a man of any importance. By 1091 Aiulf was Sheriff of Somerset, and held both offices in the

" See p. 130. =■• See pp. 31-32. Dorset Xat. Hist, and Antic/. Field Club, xiv. 1 15-16.

^* See p. 129. For .Aubrey de Couci, see Stenton, ^^ Regesta Regiim Anglo-Normannorum, i, nos. 109,

Anglo-Saxon Eng. 606. 203.

^' See pp. 136, 137. ^^ Eyton (Key to Domesday: Dorset, 141-2) identified

-' GnWm C/iriin'ana (1874), xi, App. col. 329E; T. Bond, this manor as Marshwood, Fagersten (Place-Names of

'On the Barony of the Wife of Hugh fitz Grip', Proc. Dorset, 298) as Wootton Fitzpaine.

46

DOMESDAY SURVEY

reign of Henry I, perhaps until about i izo.^o He may have been aUve in 1 130, since he appears in the Pipe Roll for that year.3' A daughter of his became a nun at Shaftesbury, whereupon he returned to that abbey the part of Farnham (no. 352) which he held in 1086 but which had belonged to the abbey T.R.E. Aiulf's brother, Humphrey the chamberlain, held just over 10 hides in Dorset. He held land in 8 counties, and seems to have been a protege of Queen Maud. In Surrey he held the manor of Combe whose previous owner, a woman, placed herself under the queen's protections^ and two of his manors in Gloucestershire had been given to him by the queen." Queen Maud pre- sumably gave him his two manors at Edmondsham (nos. 353, 354) since she had held the remaining portion of this vill herself (nos. 18 and xxiv). Eddeva, who held one of these two manors, is probably to be identified with the widow mentioned in the Geld Roll for Albretesberge hundred, who held a hide at farm of Humphrey the chamberlain which did not pay geld because Aiulfiis dicit reginam perdonasse pro anima Ricardi filii sui. Humphrey seems to have held some oflncial position in East Anglia under William Rufus, either as sheriff or as local justiciar. -'-^ William of Eu held 90 hides in Dorset. Over half of his entire fief (consisting of 336 hides in eight counties) lay in Wiltshire and Dorset. His lands were derived largely from Aelfstan of Boscombe, who held 36 hides in Dorset, and Toli, who, though less wealthy than Aelfstan, held a considerable amount of land in Dorset and the south-west. William's father. Count Robert, held land in Essex and Huntingdon, but the bulk of his fief lay in Sussex. The mother of William of Eu occurs once in the Dorset Geld Rolls, holding a manor which can only be Crichel (no. 266) which her son held in 1086.35 Eyton suggested that William's mother was a relative of Ralph de Limesi, who had once held land in Dorset and Gloucestershire which belonged to William in 1086.3^ In Dorset Ralph de Limesi had held Blandford St. Mary (no. 261) and in Gloucestershire 34 carucates of the honor of Strigoil (later Chepstow) and several manors. ^^ William of Eu was a rich and powerful baron but he did not retain his position long. In 1088 he took part in the rebellion against William Rufus, and in 1094 was involved in the plot against the king's life. In 1096 he was unable to clear himself of a charge of treason, and in consequence was blinded and mutilated, probably dying soon afterwards. ^^ His steward, William de Aldrie, was involved in his downfall and hanged. He held land of William of Eu in Wiltshire, and in the Dorset Geld Rolls appears as the holder of a manor which is probably Blandford (no. 261).

Roger Arundel held a considerable amount of land in the south-west which later formed the honor of Powerstock. He had 65 hides in Dorset and 78^ in Somerset, most of which he derived from two English thegns, Aethelfrith and Aelmer. He had at one time held the 30-hide manor of Piddletrenthide (no. 69) which in 1086 belonged to the New Minster, Winchester. Roger's surname appears to be a corruption of rhirondelle, and has no connexion with Arundel in Sussex. ^^ As Roger Derundel he witnessed a charter to the Bishop of Wells in 1068, but is otherwise unknown.-*^ A much more famous figure was Roger de Beaumont,"*' who held 47^ hides in Dorset and the manor of

" W. A. Morris, Medieval Eiig. Sheriff, 47, n. 48 " Dom. Bk. (Rec. Com.), i, ff- 162, i66b~i67.

(reprinted from W. A. Morris, 'The Office of Sheriff in "* Anglo-Saxon Cliron., ed. D. Whitelock and others,

the Early Norman Period', E.H.R. xxxiii. 151, n.); 173 ; Ordericus Vitalis, H/s/. Sec/., ed. A. Le Prevost, iii.

Regesta Regiim .■inglo-Nonnannoriim, ii, no. 1367 and n. 411. According to the Chron. the accusation was brought

5' Pipe R. 1 130 {Rec. Com.), 14. by Geoffrey Bainard. Orderic says that the accuser was

" T.R.W. femina que banc terram tenehat misit se cum ea Earl Hugh, whose sister William had married. Le Provost

in manu regine: Dom. Bk. (Rec. Com.), i, f. 36b. calls William the Count of Eu, but Dr. Whitelock rejects

" Ibid. f. 170. this.

^* Regesta Regum Anglo-Normannormii, i, p. xxv. ■''' Eyton, Domesday Studies: Soni. i. 62-63.

" See pp. 138, 139. '"' Regesta Regum Anglo-Normannorum, i, no. 23.

3* Eyton, Key to Domesday: Dorset, -^ii; Eyton, Domes- ■" Name derived from Beaumont-le-Roger; Eure, arr.

day Studies: Som. i. 64. Bernay, cant. Beaumont.

47

A HISTORY OF DORSET

Dorsington (Glos.), assessed at lo hides. Eyton remarked that 'the appearance of this name on any page of Domesday is a marvel'-*- since Roger must have been of an advanced age in 1086. He was the son of Humphrey de Vieilles, who died before i047.'*3 Roger furnished 60 ships for the invasion of England, but was already too old to fight in the battle of Hastings, and was represented by his elder son Robert.'^* Nevertheless, Roger remained active throughout his long life. As late as 1090 he supported his son Robert in a quarrel with the Duke of Normandy*' and about 1095 he entered the monastery of St. Pierre, Preaux, where he died as a monk, som^ years later {post aliquot aiinos (sic) comer sionis suae bono fine quievit)^^

Some of the men holding smaller amounts of land in Dorset were powerful barons in neighbouring counties. Edward of Salisbur}', who held two manors assessed at 38 hides in Dorset, was Sheriff of Wiltshire, where he held 193 hides. Much of his land in 8 counties, including his Dorset manors, had belonged to the English ladv Wulfwynn. Waleran the huntsman {venator), who held 38 hides in Dorset, had extensive lands in Wiltshire also. One of his Domesday manors. Church Knowle (no. 308), was given to him by William fitz Osbern. Robert fitz Ceroid held 22 hides in Dorset and 55 in Wiltshire. Ernulf of Hesdin, a Fleming from the Pas de Calais,-*" had land in Wiltshire and Gloucestershire, as well as his 15 hides in Dorset. William of Moyon,-*^ Sheriff of Somerset, had 36 hides in Dorset, and 75 in Somerset, including the manor of Dunster, where he built his castle. Turstin fitz Rolf, another Somerset landowner, held 1 1 hides in Dorset. He is perhaps to be identified with Turstin son of Rollo, who is said to have borne the Norman banner at Hastings.-*^ Two other Somerset barons, Serle of Burcy and William of Falaise, had 13 hides each in Dorset. William of Falaise was Serle's son-in-law, having married his daughter Geva.so Another of Serle's daughters was a nun at Shaftesbury.'' Walter or Walscin of Douai, whose lands lay mainly in Devon, held ID hides in Dorset, and Walter de Claville, who had 13 hides in Dorset, also held land in Devon. Unlike the foregoing barons, the bulk of the land of William of Briouze'^ lay at a distance, in Sussex, where he had over 400 hides. His 26 hides in Dorset were an insignificant part of his fief, later known as the honor of Bramber. A less important person was William of Ecouis,'' who held 1 1 hides in Dorset, and also held land at Caerleon, then part of Herefordshire. Most of his land lay in East Anglia and Essex. Hugh de St. Quintin, who had 4^ hides in Dorset, held 3 manors in Essex, and was a tenant of Hugh de Port in Hampshire. Hugh de Boscherbert who held iii hides, is unknown outside Dorset, but appears also as a tenant of the wife of Hugh fitz Grip.

Some men who held isolated manors had received the estates of Englishmen whose lands were scattered over several shires. Baldwin of Exeter, Sheriff of Devon, held Iwerne Courtney (no. 316) as a result of his acquisition of the lands of Seward, who held this manor T.R.E. and appears as a predecessor of Baldwin in Devon and Somerset. It acquired its alternative name Shroton (i.e. sheriff's town) from Baldwin. The pre- decessor of Alvred of Epaignes'^ at Turnworth (no. 319) was Alwi, whose lands in Devon and Somerset had also passed to Alvred. Exon. Domesday for these two counties

" Eyton, Key to Domesday: Dorset, 76. ^o £)o»i. Bk. (Rec. Com.), i, f- 96b. For a discussion of

*' D. C. Douglas, 'Companions of the Conqueror', the 2 famihes and their descendants, see H. Maxwell-

History, xxviii. 136. Humphrey founded the 2 mon- Lyte, 'Burci, Falaise and Martin', Proc. Sam. Arch. Soc.

asteries at Preaux, St. Pierre for monks and St. Leger for Ixv. 1-27.

nuns: Ordericus Vitalis,