Skip to main content

Full text of "The place-names of Suffolk"

See other formats


UC-NRLF 


B    3    MbE    T30 


L. 


"1 


Cambridge  Antiquarian  Society.     Octavo  Publications.     No.  XLVI. 


THE    PLACE-NAMES 

OF 

SUFFOLK 

BY  THE 

Eev.   WALTER  W.   SKEAT 

LiTT.D.,    D.C.L.,  LL.D.,  Ph.D.,  F.B.A. 

SOMETIME    ELRINGTON   AND    BOSWORTH    PROFESSOR   OF   ANGLO-SAXON 
AND    FELLOW   OF    CHRIST's    COLLEGE,    CAMBRIDGE 


PRINTED    FOR    THE   CAMBRIDGE    ANTIQUARIAN    SOCIETY. 

SOLD  BY  DEIGHTON,  BELL  &  CO. ;   and  B0WES^&  BOWES 

LONDON,  G.  BELL  AND  SONS,  LTD. 

1913 

Price  Five  Shillings  Net 


THE    PLACE-NAMES 


OF 


SUFFOLK 


PUBLICATIONS  :   OCTAVO  SERIES 
No.  XLVI 


THE    PLACE-NAMES 


OF 


SUFFOLK 


BY    THE 


Kev.    WALTER   W.    SKEAT 

Lnr.D.,    D.C.L.,  LL.D.,  Ph.D.,  F.B.A. 

SOMETIME    ELRINGTON    AND    130SW0RTH    PKOFESSOK    OF    ANGIiO-SAXON 
AXU  -FELLOW    OF    CHIUST's    COLLEGE,    CAMLKIDGE 


Ui'ambiiligc  : 

PRINTED    FOR    THE    CAMBRIDGE    ANTIQUARIAN    SUCIETV. 

SOLD  BY  DEIGHTON,  BELL  &  CO.  ;    and  BOWES  &  BOWES 

LONDON,  G.  BELL  AND  SONS,  LTD. 

1913 


CambrtHgr : 

I'RINTEU    BY   JOHN    CLAY,    M.A. 
AT    THE    UNIVEK8ITY    PKESS 


PKEFACE 


rilHE  MS.  copy  of  this  work  was  arranged  by  Professor  Skeat 
but  his  death  occurred  before  the  proof-sheets  could  be 
submitted  to  him. 


The  Council  of  the  Cambridge  Antiquarian  Society  therefore 
publish  the  work  as  Professor  Skeat  left  it. 


5312.6 


PAGE 
1 

4 
4 
5 


CONTENTS 

Prefatory  Remarks 

1.  The  suffix  -acre  :— Benacre 

2.  The  suffix  -bach  : — Debach 

3.  The  suffix  -beck  : — Gosbeck 

4.  The  suffix  -bergh  :  — Babergh  Hundred,  Finborough 

5.  The    suffix    -borough     -bury  :— Aldborough,    Blythburgh, 

Burgh,  Grundisburgh,  Rumburgh,  Bury  St  Edmund's, 
Chedburgh,  Kettleburgh,  Sudbury 6 

6.  The  suffix  -bourx  : — Blackbourn,  Xewbourn,  Sudbourn      .         TO 

7.  The  suffix  -bridge: — Risbridge,  Woodbridge       .  10 

8.  The    suffix  -brook: — Holbrook,    Rushbrooke,    Stradbrok(\ 

Washbrook,  Wickham brook  .         .         .         .  .  11 

9.  The  suffix  -by  : — Ashby,  Barnby,  Risb}',  Wilby  .         .         .         12 

10.  The  suffix  -camp:  Bulcau:ip 14 

11.  The  suffix  -clay: — Hinderclay 14 

12.  The  suffix  -dale: — Botesdale,  Withersdale  .         ...         15 

13.  The  suffix  -dene:— Depden,  Elveden,  Framsden,   Frosten- 

den,  Hundon,  Monewden,  Owsden,  Rattlesden,  Wautisden, 
Wetherden      . .15 

14.  The  suffix  -down:— Ballingdon,  Brandon,  Claydon,  Darrasden, 

Hawkedon,  Raydon  or  Reydon,  Thorndon  18 

15.  The  suffix  -edish  or  -eddish: — Brundish,  Cavendish  .         20 

16.  The  suffix  -ey: — Eye,   Bawdsey,   Bungay,    Campsey   Ashe, 

Kirsey,  Lindsey 21 

17.  The  suffix  -field: — Ashfield,  Bedfield,  Bedingfield,  Bradfield, 

Bramfield,  Bredfield,  Charsfield,  Cockfield,  Cratfield,  Crow- 
field,  Fressingfield,  Homersfield,  Huntingfield,  Laxfield, 
Metfield,  Mickfield,  Pakefield,  Redlingfiekl,  Ringsfield,  Shad- 
ingfield,  Stanningfield,  Stansfield,  Sternfield,  Waldingfield, 
Waldringfield,  Wattisfield,  Westei-field,  Whatfield,  Wingfiekl, 
Withersfield 24 


VIU  CONTENTS 

PAGE 

IS.     The  suffix  -fleet: — Herringfleet  .         .....         30 

19.  The  suffix  -ford:  Battisford,  Blythford,  Boxford,  Bramford, 

Brockford,  Carlford,  Chillesford,  Cosford,  Cransford,  Culford, 
Gleinsford,  Kentford,  Lackford,  Marlesford,  Melford,  Mutford, 
Orford,  Playford,  Poslingford,  Samford,  Stratford,  Thetford, 
Ufford,  Wangford,  Wilford,  Yoxford 31 

20.  The  suffix  -uate  :-  Burgate,  Lidgate,  Plomesgate        .         .         39 

21.  The  suffix  -grave: — Gedgrave,   Hargrave,  Hengrave,  Kes- 

grave,  Palgrave,  Redgrave 40 

22.  The  suffix  -hale  or  -hall  :— Aspall,  Benhall,  Blaxhall,  Bux- 

hall,  Foxhall,  Ilketshall,  Kelsale,  Knettisball,  Knodishall, 
Lawshall,  Mildenhall,  Peasenhall,  Eickinghall,  Ringshall, 
Spexhall,  Stradishall,  Uggeshall,  Westhall  ...         42 

23.  The  suffix  -ham  : — Akenhaui,  Aldham,  Aldringham,  Bading- 

hani,  Barham,  Barnham,  Barningham,  Barsham,  Baylham, 
Blakenham,  Brantham,  Brettenham,  Bucklesham,  Cavenhain, 
Chattiisham,  Coddenham,  Cretingham,  Dalham,  Darsham, 
I)eV)eiiham,  Denham,  (Santon)  Downham,  Elmham,  Faken- 
ham,  Falkenham,  Fariiliam,  Felsham,  Finningbani,  Fornham, 
Framlingham,  Freckenham,  Gislebam,Gislingham,  Glemham, 
Helmingham,  Henham,  Heveninghani,  Highham,  Hintles- 
ham,  Hitcham,  Horham,  Icklingham,  Ingham,  Langham, 
Lavenham,  Layham,  Letheringham,  Martlesham,  Mendhani, 
Mendlesbarn,  ^lettingbam,  Needbam  Market,  Pakenliam, 
Parbain,  Redisham,  Rendbatii,  Rendlesbam,  Rougham, 
Saxham,  Saxmundbam,  Sbottisbam,  Sobam,  Somersbam, 
Stonbam,  Sylebam,  Tbelnetbam,  Tbornbam,  Tuddenbam, 
Walsbarn,  Wattisbam,  Wenbam,  Wbelnetbani,  Wickbam 
Market,  Wickbambrook,  Wickbam  Skeitb,  WilHngbam, 
Wilbsbam,  Witnesbam,  Worlingbam,  Wortbam,  Wi-entbam         47 

24.  The  suffix  -haugh  : — Pettaugb G') 

25.  The  suffix  -heath  : — Horniiigsbeatb,  Lakenbeatb,  Leaven- 

beatb       ...........        66 

26.  The  suffix  -hill: — Haverbill 67 

27.  The  suffix  -hithe: — Covehitbe 67 

28.  The  suffix  -hok  or  -hoo  : — Hoo,  Culpbo,  Dallinglioo,  Tbingoe, 

Wixoe 68 

29.  The  suffix  -holt:— Bergbolt,  Occold,  llamsholt,  Sdutliolt  70 

30.  The  suffix  -hurst  :— Hartest 70 


CONTENTS  IX 

PAGE 

31.  Thk  «l'FF1X  -inu  : — Ashbockiiig,  Barking,  Bealiugs,  Blything, 

Cowlinge,  Greeting,  Exning,  Gedding,  Gipping,  Milden, 
Nedging,  Shimpling,  Sweffling,  Thredling,  Wratting  .  71 

32.  The    suffix    -land: — Kessingland,    LothingLiiid,    Xaylaud, 

Shelland,  Swillaud 7") 

33.  The  suffix  -ley  : — Badley,  Beatley,  Bradley,  Brockley,  Butlcy, 

Cookley,  Eleiglj,  Gazeley,  Hadleigh,  Haughley,  Hemluy, 
Henley,  Hollesley,  Kirkley,  Oakley,  Otley,  Shelley,  Shotley, 
Sotterley,  Trimley,  Westley,  Yaxley 77 

34.  The  suffix  -low  : — Thurlow 81 

35.  The  suffix  -meadow  : — Shipmeadow 82 

36.  The  suffix  -mere  : — Bosmere,  Hartisincre,  Livcnucre,  Rush- 

mere,  Seiner   ..........         82 

37.  The  suffix  -fool  : — Walpole 83 

38.  The    suffix    -set    or    -sett  : — Bricett,    Elmsett,    Hessett, 

AVetheringsett,  Wissett .84 

39.  The  suffix  -stall  : — Eurstall,  Tuustall         ....         86 

4U.  The  suffix  -stead  : — Belstead,  Box.stead,  Harkstead,  Haw- 
.stead,  Heastead,  Linstead,  Xettlestead,  Polstead,  Sax.stead, 
Staustead,  Whepstead,  Wherstead 86 

41.  The    suffix   -.stoke   or  -stock  :— Stoke- by-Clare,   Stoke-by- 

Nayland,  Stoke  Ash,  Tostock 89 

42.  The  suffix  -stoxe  : — Chediston 89 

43.  The   suffix   -stow  : — Stow,  ,West  Stow,   Stowmarket,  Stow 

Upland,  Felixstowe 89 

44.  The  suffix  -thorpe  : — Thorpe-Morieux,  Thorpe,  Thorpe-by- 

Ix worth,  Westhorpe 91 

45.  The  suffix  -toft  : — Lowestoft 92 

46.  The    suffix    -ton: — Acton,    Alderton,    Alpheton,    Ampton, 

Assington,  Athelington,  Bacton,  Barton,  Belton,  Beyton, 
Bildeston,  Blundeston,  Boyton,  Brampton,  Brandeston, 
Browston,  Carlton,  Chelmondi.ston,  Chevington,  Chilton, 
Cloi)ton,  Coney  Weston,  Corton,  Cotton,  Dennington, 
Den.stone,  Drinkstone,  Easton,  Edwardstone,  Erwarton, 
Euston,  Flempton,  Flixton,  Flowton,  Frestou,  Friston,  Fritton, 
Gorleston,  Gunton,  Hacheston,  Harleston,  Hasketon,  Heni- 
ingstone,  Hintou,  Holton,  Honington,  Hopton,  Hunston, 
Kedington,  Kenton,  Kettlebaston,  Kirton,  Leiston,  Leving- 


CONTENTS 


ton,  Market  Weston,  Melton,  ]\Iiddleton,  Moulton,  Xacton, 
Newton,  Norton,  Nowton,  Ofton,  Oiilton,  Preston,  Sapiston, 
Sibton,  Somerleyton,  Sothertou,  Sproughton,  Stanton,  Stuston, 
Stutton,  Svitton,  Tannington,  Tattingstone,  Theberton, 
Thorington,  Thrandeston,  Thurlston,  Thurston,  Troston, 
Ubbeston,  Walton,  Wenhaston,  Westleton,  Weston,  Market 
AVeston,  Whitton,  Winston,  Wiston,  Woolverstone,  Worling- 
ton,  Wyverstone     .........         92 

47.  The  suffix  -tree: — Pettistree,  Thedwestry  .         .         .         .       Ill 

48.  The  suffix  -wade: — Cattawade Ill 

49.  The  suffix  -well  : — Badwell  Ash,  Bardwell,  Bradwell,  Bright- 

well,  Bromeswell,  Elmswell,  Eriswell,  Herringswell,  Orwell, 
Sizewell,  Wordwell 


50.  The  suffix  -wich  or  -wick  : — Dunwich 

Walberswick 

51.  The  suffix  -wold  : — Southwold 

52.  The  suffix  -wood  : — Hazlewood     . 


Hard  wick,  Ipswich, 


112 

114 
115 
116 


53.  The    suffix    -worth  : — Bi'aiseworth,    Chelsworth,    Dunning- 

worth,  Halesworth,  Hepworth,  Ickworth,  Ixworth,  Timworth, 
Worlingworth 116 

54.  The  suffix  -yard: — Bruisyard 118 

55.  Some  other  names  : — Cornard,  Barrow,  Beccles,  Beck  Row, 

Boulge,  Broine,  Bures  St  Mary,  Capel,  Clare,  Colneis,  Combs, 
Copdock,  Cove,  EUough,  Eyke,  Groton,  Hoxne,  Iken,  Land- 
guard,  Loes,  Bound,  Mellis,  Onehouse,  Rede,  Rishangles, 
Snape,  Steven,  Thwaite,  Wey bread,  Woolpit       .         .         .       118 

56.  Concluding  remarks 125 

Index 127 


THE   PLACE-NAMES   OF   SUFFOLK 

Prefatory  Remarks. 

To  the  Cambridge  Antiquarian  Society  will  always  belong 
the  credit  of  initiating  a  series  of  works  upon  the  Place-names 
of  Counties,  founded  upon  strictly  scientific  investigation.  My 
Place-names  of  Cambridgeshire  was  published  by  them  in  1901; 
those  of  Huntingdonshire  in  1903  ;  and  of  Bedfordshire  in 
1906.  My  Place-names  of  Hertfordshire  was  published  by  the 
East  Herts.  Archaeological  Society  in  1904 ;  and  of  Berkshire, 
by  the  Clarendon  Press,  in  1911.  Mr  W.  H.  Duignan,  of 
Walsall,  published  his  Notes  upon  Staffordshire  Place-names 
in  1902 ;  and  upon  those  of  Worcestershire  in  1905,  and  has 
just  given  us  (1912)  an  account  of  those  of  Warwickshire. 
The  West  Riding  Place-names,  by  Prof.  Moorman  of  Leeds, 
was  published  by  the  Thoresby  Society  in  1910 ;  and  the 
Place-names  of  Lancashire,  by  Dr  H.  C.  Wyld  and  Dr  T.  O. 
Hirst,  appeared  in  1911.     This  makes  ten  counties  in  all. 

Being  anxious  to  increase  the  series,  I  now  attempt  to 
give  some  account  of  the  Place-names  of  Suffolk.  I  was  led 
to  select  this  county  because  Dr  Copiuger,  in  his  six  volumes 
of  Collections  for  a  History  of  Suffolk,  has  taken  the  enormous 
trouble  of  collecting  all  the  old  spellings  of  place-names  which 
his  exhaustive  researches  enabled  him  to  discover,  duly  entering 
them  under  their  respective  articles,  in  alphabetical  order ;  and 
since  such  a  collection  of  old  forms  constitutes  no  inconsider- 
able portion  of  the  task  of  the  investigator,  I  was  under  the 
impression  that  all  due  preparation  had  been  made.  I  found 
his  work  of  very  great  service,  but  he  unfortunately  made  the 
regrettable  mistake  of  omitting  to  indicate  his  authorities ! 
The  result  was,  of  course,  that  the  principal  books  of  reference, 
such  as  Domesday  Book,  the  Red  Book  of  the  Exchequer,  the 
Hundred  Rolls,  Testa  de  Nevill,  and  the  Inquisitiones  post 
Mortem  had   to  be  consulted  in  detail  all  over  again.     Still, 

C.  A.  S.  Octavo  Series.     No.  XLVI.  1 


2  THE    PLACE-NAMES   OF   SUFFOLK 

I  did  not  find  it  necessary  to  go  further,  as  the  material  was 
then  ample ;  and  I  now  give  the  references  for  all  the  more 
important  forms. 

Further  information  as  to  the  methods  pursued  and  the 
results  to  be  expected  can  be  found  in  the  introductions  to  my 
previous  essays  of  the  same  character,  and  need  not  be  here 
repeated. 

Abbreviations. 

The  following  is  a  list  of  the  more  important  sources  of 
information,  with  the  abbreviations  that  denote  them. 

D.B. — Domesday  Book  (part  relating  to  Suffolk).  The  page  quoted 
does  not  refer  to  the  Book  as  a  whole,  but  to  the  paging  of  the 
Facsimile  of  the  part  relating  to  Suffolk. 

E.D.D. — English  Dialect  Dictionary. 

F.A. — Feudal  Aids  (Kecord  Series) ;  vol.  i. 

H.R.— Rotuli  Hundredorum,  vol.  i. 

Ipm. — Calendarium  Inquisitionum  post  ^lortem  sive  Escaetarum  ; 
ed.  J.  Caley  ;  vol.  i.  (Record  Series). 

N.E.D.— New  English  Dictionary  (Oxford). 

R.B.— Red  Book  of  the  Exchequer ;  ed.  W.  D.  Selby  (Rolls  Series). 

T.N.— Testa  de  Xevill  (temp.  Henry  III  and  Edw.  I). 

V.E. — Valor  Ecclesiasticus  ;  temp.  Henry  VIII. 

Of  course  I  constantly  refer  to  the  well-known  editions  of 
the  Anglo-Saxon  Charters  by  Kemble  and  Birch,  to  Thorpe's 
Diplomatarium  JEvi  Saxonici,  and  to  Earle's  Select  Charters, 
Also,  to  the  Crawford  Charters,  ed.  Napier  and  Stevenson ; 
and  to  Searle's  Onomasticon,  from  which  I  quote  Anglo-Saxon 
personal  names,  verifying  them  in  many  instances  by  a  reference 
to  the  Charters.  Amongst  numerous  books  of  reference  wliich 
I  have  consulted,  I  may  particularise  the  following : 

Bardsley,  Rev.  C.  W.  A  Dictionary  of  English  and  Welsh  Sur- 
names.    London,  1891. 

Bjorkman,  E.  Nordische  Personennamen  in  England.  Halle  a.  S., 
1910. 

BoswoRTH,  Rev.  J.  and  Toller,  Prof.  T.  N.  An  Anglo-Saxon 
Dictionary.     Oxford,  1882. 

CoplNGER,  W.  A.  The  County  of  Suffolk :  its  History  as  disclosed 
by  Existing  Records.     London.     5  vols.     1904-5. 

DuiGNAX,  W.  H.    Notes  on  Staflfordshire  Place-names.    London,  1902 

.     Worcestershire  Place-names.     London.  1905. 


PREFATORY   REMARKS  O 

M^Clurk,    E.      British     Placo-namos    in    their    Historical    Setting. 

London,  1910. 
MiDDENDORF,  Dr  H.     Altenglisches  Flurnamenbuch,     Halle,  1902. 
Moorman,  F.  W.,  B.A.,  Ph.D.     The  Place-names  of  the  West  Riding 

of  Yorkshire.     Thoresby  Society,  1910. 
NiELSON,  0.,  Ph.D.     Olddansko  Persoiniavne.     Kjobenhavn,  1883. 
Rygh,   O.     (5amle  Personnavne    i    Norske   Stedsnavno.     Kristiania, 

1901. 
Searle,  Rev.  W.  G.,  M.A.     Onomasticon  Anglo-Saxonicum.     Cam- 
bridge, 1897. 
Taylor,  Rev.  J.,  Litt.D.,  LL.D.    Names  and  their  Histories.    London, 

1896. 
Turner,  J.  Horsfall.     Yorkshire  Place-names,  as  recorded  in  the 

Yorkshire  Domesday  Book.     Bingley  ;  printed  for  the  Author. 
Victoria  History  of  the  County  of  Suffolk. 
Wyld,  Prof.  H.  C.  and  Hirst,  T.  0.,  M.A.,  Ph.D.     The  Place-names 

of  Lancashire.     London,  1911. 
Zachrisson,  R.  E.     a  Contribution  to  the  Study  of  Anglo-Norman 

Influence  on  English  Place-names.     Lund,  1909. 

I  have  also  consulted  the  following  works  that  have  special 
reference  to  the  County  : 

KiRBY,  J.  The  Suflfolk  Traveller.  2nd  ed.  London,  1764. 
Raven,  Rev.  J.  J.  The  History  of  Suffolk.  London,  1895. 
Shorberl,  F.     Suftblk;  being  vol.  xiv.  of  The  Beauties  of  England 

and  Wales.     London,  1813. 
Walters,  Cuming.     Bygone  Suffolk.     London,  n.d. 

I  am  grateful  for  several  hints  that  these  local  books  have 
afforded ;  but  cannot  help  regretting  that  (with  the  notable 
exception  of  the  Victoria  History)  they  occasionally  drop  into 
etymology,  with  reprehensible  results.  There  seems  to  be  a 
rather  general  notion,  in  local  works,  that  the  river  Gipping 
gave  its  name  to  Ipswich  ;  with  other  similar  fables. 

The  Place-names  are  arranged  in  alphabetical  order 
according  to  the  suffixes  which  they  contain ;  this  avoids  much 
repetition. 

The  suffixes  found  in  Suffolk  are,  most  of  them,  readily 
intelligible,  and  may  conveniently  be  here  enumerated.     The 

chief  ones  are: acre,  -bach,  -beck,   -bei-gh,  -borough  {-bury), 

-bourn,  -bridge,  -brook,  -by,  -camp,  -clay,  -dale,  -dene  {-den), 
-doivn  {-don),  -edish,  -ey  (-ea),  -field,  -fleet,  -ford,  -gate,  -grave, 
-hale  {-hall),  -ham,  -haugh,  -Jieath,  -hill,  -ho  {-hoe),  -holt,  -hurst, 

1—2 


4  THE   PLACE-NAMES   OF   SUFFOLK 

-ing,  -land,  -ley,  -loiv,  -oneadoiv,  -mere,  -pool,  -set,  -stall,  -stead, 
-stoke,  -stone,  -stow,  -thorpe,  -toft,  -ton,  -tree,  -wade,  -well,  -wich 
(-wick),  -tvold,  -ivood,  -worth,  -yard. 

To  these  I  add  a  few  names  that  cannot  be  included  amongst 
such  compounds. 

The  list  of  names  is  from  Kelly's  Post  Office  Directory  of 
Suffolk. 

The  atlases  consulted  are  Bacon's  County  Atlas,  Philips' 
County  Atlas,  and  Pigot's  County  Atlas  (1831).  The  last  of 
these  gives  the  boundaries  of  the  hundreds,  which  are  very 
clearly  shown  in  the  map  prefixed  to  Kirby's  Suffolk  Traveller. 
I  have  also  made  frequent  use  of  the  ordnance  map  upon  the 
one-inch  scale. 

The  various  suffixes  will  now  be  discussed,  in  alphabetical 
order,  as  given  above. 

1.  Acre. 

The  suffix  -acre  represents  the  A.S.  (ecer,  a  field.  It  only 
occurs  in  Benacre. 

Benacre.  Between  Lowestoft  and  Southwold.  Spelt 
Benakr',  H.E,.;  Benagra,  D.B.,  p.  182\  From  A.S.  hean,  bean  ;  so 
that  the  sense  is  'bean-field.'  The  Supplement  to  Bosworth 
and  Toller  quotes  the  phrase  "secer  beanlandes,"  a  field  of  bean- 
land,  from  Kemble,  CD.  iii.  366.  Ill  s,pelt  bean-eccer  in  Birch, 
C.S.  ii.  18,  1.  18. 

2.  Bach. 

This  interesting  word  only  occurs  as  a  suffix  in  Debach. 
It  is  the  prov.  E.  bach{e),  a  valley  through  which  a  stream 
flows;  M.E.  bache  or  bach,  as  in  Layamon,  and  in  P.  Plowman, 
C.  viii.  159,  discussed,  s.v.  Bache,  in  the  N.E.D.  The  A.S. 
forms  are  bcBC,  bee,  m.  and  n. ;  bcece,  bece,  as  given  in  the 
Supplement  to  Bosworth  and  Toller,  p.  60.  We  may  explain 
it  simply  by  '  valley ' ;  remembering  that  it  is  etymologically 
connected  with  the  prov.  E.  beck,  a  stream. 

Debach.  Spelt  Debeth,  error  for  Debech  (by  the  common 
error  of  writing  t  for  c,  or  of  misreading),  Ipm. ;  Debaht,  error 
for    Debahc   (=  Debach),   H.R.      D.B,    has   Depebecs,   p.    240 ; 

1  The  reference  here,  and  elsewhere,  is  to  the  Part  of  the  Facsimile  edition 
of  Domesday  Book  relating  to  Suffolk,  photozincographed  in  1863. 


NAMES   ENDING   IN    -BACH,    -BECK,   -BEllGH  5 

Depehek,  p.  262 ;  so  that  it  represents  the  A.S.  dat.  form 
deopan  hcuce,  lit.  'deep  valley';  which  explains  the  spelling 
Dehenbeis  (with  n),  in  D.B.,  p.  12G.  See  Birch,  C.S.  iii.  344 
(no.  1111),  where  we  find  : — "  of  tham  diopan  bcece." 

The  present  local  pronunciation  is  Debbidge,  in  strict 
accordance  with   the  usual  popular  sound-changes. 

8.     Beck. 
This   well    known   prov.   E.   word   for  '  a  small   stream '   is 
known   in  Suffolk  and   Norfolk  as  well  as  in  the  North.     It 
occurs  in  Gosbeck. 

GosBECK ;  to  the  E.  of  Needham  Market.  Spelt  Gosebeck, 
Gosebek,  Ipm.  From  A.S.  gds,  a  goose.  It  simply  means 
'goose-brook,'  or  'goose-stream.'  Kemble  has  a  Gosebroc,  lit. 
'  goose-brook.' 

4.     Bergh. 

This  suffix  not  only  occurs  in  Babergh,  the  name  of  a 
hundred,  but  also,  as  the  old  forms  show,  in  Finborough.  The 
confusion  of  bej-gh  with  borough  is  common,  though  they  were 
kept  separate  in  the  older  forms  of  our  language.  Bergh 
represents  the  A.S.  beorh,  a  hill,  a  barrow,  a  mound ;  the 
modern  form  is  barroiu. 

Babergh  hundred.  Spelt  Baben-berga,  D.B.,  p.  12;  but 
the  second  b  is  an  error  for  d,  owing  to  the  influence  of  the 
third  b.  Baberga  also  occurs,  D.B.,  pp.  223,  225,  271.  More 
correctly,  Badbergh  (hundred),  Badberewe  (hundred),  H.R. ; 
Baddebury  hundred,  Ipm.;  so  that  the  D.B.  form  should  have 
been  Baden-berga  (with  a  Latin  -a  suffixed).  Baden  is  the  A.S. 
Badan,  as  in  Badan-den,  Badan-pyt ;  both  in  Kemble ;  and 
Badan  is  the  gen.  of  Bada.  The  sense  is  '  Bada's  hill,'  or 
'  Bada's  barrow.'  The  change  from  db  to  modern  b  is  well 
illustrated  by  Babraham  (Cambs.),  originally  Badburgeham, 
meaning  'Badburh's  home,'  or  'Badburh's  ham.' 

Finborough.  There  is  a  Finborough  to  the  S.W.  of 
Stowmarket,  and  a  Great  Finborough  to  the  W.  of  it.  The 
suffix  -borough  has  been  substituted  for  an  earlier  -bergh.  We 
find  Finebiirge  in  R.B.  and  Finebury  in  Ipm.;  but  Fitieberg  in 


6  THE   PLACE-NAMES    OF    SUFFOLK 

H.R.;  and  Fineberga  in  D.B.,  pp,  9  and  386 ;  Finbergh  in  Ipm., 
p.  55 ;  and  Finbarotue  even  in  V.E.  The  earliest  spelling  is 
Finheorh,  in  a  Wilts,  charter,  dated  957  ;  though  perhaps  in  a 
copy  of  later  date ;  Birch,  C,S.  iii.  186.  The  name  is  a 
compound ;  and  the  former  element  may  be  safely  identified 
with  the  A.S.  wordy^n,  'a  heap,'  which  is  fairly  well  authenti- 
cated, and  not  only  occurs  alone,  but  in  the  compovmds  llm-fin, 
a  lime-heap,  or  heap  of  lime,  and  wudu-fm,  a  heap  of  wood  ; 
see  the  note  in  Napier's  Glosses,  p.  66,  gloss  2456.  The  sense 
is,  accordingly,  '  a  heap-barrow,'  or  an  artificial  mound  made  by 
heaping  up  materials.  Near  Great  Finborough  Hall  there  is  a 
tumulus  named  the  Devil's  Hill. 

5.    Borough,  Bury. 

Borough  is  the  usual  modern  E.  form  of  the  A.S.  hurh,  a 
fort,  borough,  town  ;  and  hury  represents  its  dative  case  hyrig  ; 
so  that  the  two  may  be  taken  together.  Borough  occurs  in 
Burgh,  Aldborough,  Blythburgh,  Grundisburgh  and  Rumburgh. 
Bury  occurs  alone  in  Bury  St  Edmund's  and  in  the  compounds 
Chedburgh,  Kettleburgh,  Sudbury. 

Burgh.  There  are  three  places  with  this  name  referred  to 
in  Domesday  Book.  Burgh.  Three  miles  N.W.  of  Wood- 
bridge.  Spelt  Burc,  D.B.,  p.  25  ;  Burch,  p.  70  ;  Burh,  p.  212  ; 
Burg,  p.  301 ;  which  represent  the  A.S.  hurh,  a  fort ;  mod.  E. 
borough.  According  to  the  map  of  Roman  Suffolk  in  the 
Victoria  County  History  this  is  on  the  line  of  the  Roman  road 
from  Stratford  St  Mary  to  Dunwich  and  is  on  the  site  of  the 
Roman  station  Combretonium, 

Burgh.  On  the  Waveney,  at  the  beginning  of  Breydon 
Water.  Spelt  Burch,  D.B.,  p.  329.  There  is  a  celebrated  castle 
there,  of  Roman  origin.  This  Burgh  Castle  is  that  mentioned 
in  Beda,  Hist.  Eccl.  iii.  19 — "in  castro  quodam  quod  lingua 
Anglorum  Cnobheres  burg,  id  est  urbs  Cnobheri,  uocatur."  So 
that  its  name  in  Beda's  time  meant  '  Cnobhere's  burgh.'  But 
Cnobhere  was  evidently  forgotten  at  the  time  of  the  Norman 
Conquest,  and  it  has  become  simply  Burgh. 

Burgh.  In  Colneis  Hundred.  Spelt  Burch,  D.B.,  pp.  68, 
119  ;  Burg,  D.B.,  p.  286.     This  probably  represents  the  Roman 


NAMES   ENDING   IN   -BURGH  7 

fort,  known  in  later  times  as  Walton  Castle,  near  Felixstowe, 
now  washed  away  by  the  encroachments  of  the  sea. 

Aldborough,  or  (in  Kelly's  Directory)  Aldeburgh.  Spelt 
Aldeburc  in  D.B.,  p.  71.  The  e  may  be  explained  as  the 
termination  of  the  Aveak  feminine  snffix  of  the  nominative 
case  ;  A.S.  seo  ealde  burJt,  Mercian  seo  aide  burh,  i.e.  '  the  old 
borough.'  Prof.  Moorman  thus  explains  Aldborough  in  the 
West  Riding,  and  quotes,  from  a  document  of  the  thirteenth 
century,  the  entry  : — "  Aldeburgh,  Vetus  Burgh."  It  is  clear 
that  the  river  Aide  took  its  name  from  the  town,  and  not 
conversely.  Many  river-names  are  more  modern  than  some 
suppose  them  to  be.     See  Debenham. 

There  is  a  difficulty  about  the  development  of  the  primary 
vowel ;  since  the  0.  Merc,  aid  has  given  us  the  form  old. 
However,  this  modern  adjective  is  really  due  to  Norman 
influence,  which  lengthened  the  a  before  Id,  so  as  to  produce 
an  early  Mid.  E.  form  did.  But  in  the  phrase  seo  aide  burh, 
the  final  e  of  aide  easily  dropped  out ;  and  then  the  old  short 
a  remained  short  (or  was  shortened)  before  the  combination  Idb. 

Blythburgh  ;  or  borough  on  the  river  Blythe.  Spelt 
Bliburg  in  R.B.,  H.R.,  T.N.;  because  the  Norman  scribes  failed 
to  pronounce  the  E.  th  ;  also  Bliburgh,  H.R.  In  D.B.,  the  E. 
voiced  th  was  written  d,  so  that  it  appears  as  Blideburc ;  p.  o. 
The  river-name  Blythe  is  old  ;  there  was  a  river  named  Blithe 
in  Northamptonshire,  mentioned  in  a  charter  dated  944 ;  see 
Birch,  C.S.  iii.  541,  four  lines  from  the  bottom  ;  from  the  A.S. 
blithe,  '  the  blithe,'  or  '  pleasant.'  Hence  also  Blytheford  and 
Blything. 

Grundisburgh.  Spelt  Qrundesburgh,  Grundesburg,  Ipm. ; 
Grundisbur,  H.R. ;  Grundesbiirh,  D.B.,  p.  70;  with  many  slight 
variants  of  no  importance,  but  all  implying  the  same  origin ; 
I  may  notice  GroiULdesburgh  and  Groiindesborough.  There  is 
no  such  name  as  Grund  recorded  in  English,  but  Rygh  gives 
Grundi  as  a  Norse  name,  which  occurs  in  rather  numerous 
place-names  abroad.  Hence  the  sense  may  be  '  Grundi's 
borough ' ;  and  we  may  attribute  the  name  to  Norse  influence. 

RUMBURGH.     I  do  not  know  where  to  find  the  early  forms; 


8  THE   PLACE-NAMES   OF   SUFFOLK 

but  Dr  Copinger  gives  us  ten,  viz.  Rumberwe,  Romboroughe, 
Rumburn  (?),  Rumbiug(?);  and  Romborough,  Romburgh, 
Romburg,  Romborow,  Romborrow,  Romebury,  The  A.S.  form 
may  Avell  have  been  rum-hurh,  i.e.  roomy  or  wide  borough. 
We  may  exemplify  this  by  comparing  it  with  Rum-cofa, 
perhaps  'wide  cove,'  given  in  the  A.S.  Chronicle,  under  the 
date  915,  as  the  old  name  of  Runcorn  (Cheshire);  and  especially 
with  rum-heorgas,  '  wide  barrows,'  in  a  charter  dated  972,  the 
authenticity  of  which  has  been  challenged,  but  apparently  for 
no  good  reason ;  see  Birch,  C  S.,  iii.  589,  last  line.  The  almost 
total  absence  of  a  vowel  after  the  m  is  much  against  the 
explanation  '  Ruma's  borough.'  Dr  Wyld  explains  the  Lanes. 
Rumworth  as  'Ruma's  worth,'  and  at  the  same  time  gives  us 
Rumhurgh  as  the  form  of  Rumburgh,  Suff.,  as  occurring  in  the 
Catalogue  of  Ancient  Deeds,  ii.  A.  3289,  dated  1409.  This  is 
the  best  spelling,  and  supports  my  explanation. 

Bury.  This  form  occurs  alone,  and  represents  the  A.S. 
hyrig,  dat.  case  of  barh,  a  borough.  The  dat.  case  is  common 
in  place-names,  as  the  prep,  cet,  'at,'  was  often  either  ex- 
pressed or  understood  before  them.  Hence,  for  example,  the 
surname  Atterbury,  from  the  A.S.  formula  cet  tJuere  hyrig,  'at 
the  borough.'  At  a  later  time,  when  the  dative  had  been  assimi- 
lated to  the  nominative,  and  the  gender  of  the  substantive  had 
been  changed  from  fem.  to  masc,  we  find  the  formula  at  then 
borough,  which  gave  the  surname  Attenborough.  The  town's 
name  also  appears  as  Bury  St  Edmund's,  'at  the  borough  of 
St  Edmund';  from  the  famous  East  Anglian  king  and  saint 
who  was  buried  there,  and  whose  death  is  fully  described  by 
^Ifric,  in  his  Lives  of  the  Saints,  The  A.S.  Chronicle  gives  his 
slaughter  by  the  Danes  under  the  date  870;  and  the  form  'sancte 
Eadmundes  byrig'  in  1107.  In  an  A.S.  Charter  dated  945,  Bury 
is  alluded  to  as  "in  loco  qui  dicitur  tet  Bwderices  wirthe" ; 
and  again,  in  Ethelweard's  Chronicle,  under  the  date  870,  as 
^'  in..Beadimcesuuyrthe" ',  so  that  the  original  name  of  Bury 
was  really  '  Beaduric's  worth.'     See  the  suffix  WORTH  (below). 

Beaduric  is  compounded  of  beadu,  battle,  and  rw,  dominiou. 
Misspelt  and  explained  'power  in  prayer'  in  Bygone  Suffolk,  p.  7 9. 


NAMES   ENDING    IN   -BURY  ^ 

Chedburgh.  To  the  S.W.  of  Bury.  Despite  the  modern 
form,  I  refer  this  to  the  suffix  hnry,  though  it  makes  no 
difference  to  the  sense.  Spelt  CVier^tt^r^e,  Ipm.;  Cedeheria  {\s\\\\ 
Norman  ce  for  che),  D.B.,  p.  208.  Dr  Copinger  gives  several 
spellings,  of  which  the  best  are : — Cheddebur,  Chedeberi,  Chede- 
berwe,  Ghedebor,  Chedebury.  It  is  hardly  possible  to  say 
whether  the  suffix  was  bergh  or  bury]  I  only  place  it  here  for 
convenience,  though  I  suspect  that  it  once  ended  in  bergh, 
despite  the  spelling  Chedburye  in  Ipm.  (a.d.  1262).  The  prefix 
Chede-  represents  the  A.S.  Ceddan,  gen.  of  the  known  name 
Gedda;  hence  the  sense  is  '  Cedda's  bury,'  or  '  Cedda's  barrow.' 
For  the  suffix  bergh,  'a  barrow,'  see  Babergh  (above). 

Kettleburgh.  Near  Framlingham.  I  suppose  that  the 
suffix  was  at  first  bury,  in  deference  to  the  forms  in  D.B.,  viz. 
Cetelbirig,  p.  27,  Chettlebiriga,  p.  26,  Ketelbiria,  p.  89 ;  but  it 
also  has  Kettleberga,  p.  134,  and  we  find  Ketelebruge  (for 
-burge)  in  R.B.  But  the  suffix  was  certainly  confused  with 
bergh,  'a  barrow'  (see  Babergh),  as  shown  by  Ketelberghe, 
Ipm.,  Keteleber,  H.R.,  Kettleberg,  T.N.  The  prefix  is  exactly 
the  same  as  in  Kettlewell,  in  the  West  Riding,  which,  as  Prof. 
Moorman  explains,  is  not  from  a  genitive  Ketils,  but  from  a 
genitive  Ketilan,  in  which  the  -an  gave  rise  to  the  -e  in 
Ketel-e-  in  some  of  the  forms  above.  Ketila  was  a  pet  name 
for  some  form  beginning  with  Ketel-  or  Ketil-,  such  as  Ketel- 
berht  or  Ketil-frith.  Hence  the  sense  is  '  Ketila's  bury ' ; 
possibly  'Ketila's  barrow.'  Ketil,  Norse  Ketill,  was  a  famous 
Scandinavian  name,  and  appears  in  many  Norwegian  place- 
names.     This  is  a  clear  case  of  Norse  influence. 

Sudbury.  "Anciently  called  South-Burgh,  as  Norwich  is 
said  to  have  been  called  North-Burgh,"  Kirby.  Here  there  is 
no  doubt  as  to  the  suffix,  nor  as  to  the  origin  of  the  name. 
We  find  Sutberie  in  D.B.,  p.  12  ;  Suthbury,  Ipm.;  Sudbyr,  H.R.; 
from  the  dat.  case  Suthbyrig,  which  actually  occurs  (with 
reference  to  Sudbury)  in  a  Suffolk  document  known  as  the 
Will  of  ^Ifflged ;  see  Birch,  C.S.  iii.  603,  1.  7.  Hence  the  sense 
is  'South  bury';  from  the  A.S.  suth,  south. 


10  the  place-names  of  suffolk 

6.    Bourn. 

Bourn  means  a  burn,  or  small  river  (A.S.  ham) ;  and  occurs 
in  Blackbourn,  the  name  of  a  hundred,  and  in  the  place-names 
Newbourn  and  Sudbourn  ;  all  of  obvious  origin. 

Blackbourn.  This  appears  in  D.B.,  p.  11,  as  Blachruna  h'. 
and  Blackebrune,  p.  313,  and  means  'black  bourn.'  Ixworth, 
Bardwell,  and  Fakenham  are  all  in  Blackbourn  hundred  ;  so 
tliat  the  stream  here  intended  is  that  which  flows  by  all  these 
places,  and  enters  the.  Little  Ouse  below  Euston  Park, 

Newbourn.  This  is  the  name  of  a  village,  which  is  so 
called  from  a  stream  that  flows  through  it  southward,  and 
then,  turning  to  the  east,  enters  the  river  Deben  at  Kirton 
Creek,  to  the  N.E.  of  Kirton.  Curiously  enough,  the  name  is 
very  old,  and  occurs  as  Neuhurne  in  R.B. ;  Neubrunna,  D.B., 
p.  178.  Dr  Copinger  gives  several  old  spellings  of  it,  all  without 
any  reference.     However,  the  sense  is  obvious. 

SUDBOURNE.  To  the  S.W.  of  Aldeborough.  Spelt  Sutburna 
in  D.B,,  p.  72 ;  Sudburna,  D.B.,  p.  207 ;  Sutborne  in  a  late  A.S. 
charter,  in  Kemble,  CD.  iv.  245,  1.  7  from  the  bottom.  From 
the  A.S.  suth-biirn,  i.e.  '  South  burn.'  Sudbourne  Marshes  are 
traversed  by  a  maze  of  confluent  streams,  the  waters  of  which 
find  their  way  to  the  river  Aide.  Sudbourne  Park  is  con- 
siderably to  the  south  of  the  village,  and  the  reference  may  be 
to  the  stream  which  flows  from  the  Park  into  the  Butley  river. 

7.     Bridge. 

This  sufiix,  of  obvious  meaning,  only  occurs  in  Risbridge 
and  Woodbridge. 

Risbridge.  This  is  the  name  of  a  hundred  only,  in  the 
extreme  west  and  south-west  of  the  county.  The  chief  river 
hereabouts  is  the  Stour,  and  the  hundred  may  well  have  been 
named  from  a  bridge  over  it,  in  the  days  when  bridges  were 
scarce.  We  find  the  forms  Risbrigg,  H.R.;  Riseh'ige,  T.N.; 
and  Risebruge  in  D.B.,  p.  11.  The  e  in  Rise-  strongly  suggests 
the  genitive  suffix  -en,  later  form  of  -an,  from  a  personal  name 


NAMES   ENDING   IX    -BRIDGE,   -BROOK  11 

iu  -«.  Both  in  Ris-bridge  and  in  Ris-by,  we  should  suppose 
that  the  reference  is  to  the  maker  of  the  bridge  or  to  the 
founder  of  the  town  or  village.  If  the  name  is  that  also  found 
in  Risborough  (Bucks.)  it  certainly  had  an  h  before  the  r. 
With  respect  to  Risborough,  we  find  three  forms  of  various 
dates,  viz.  Hrisan,  gen.  of  Hrisa ;  the  later  form  Hrisen,  for 
Hrisan ;  and  the  form  Hris-,  without  any  suffix  at  all.  The 
references  are  as  follows  : — Hrisan-hyrge,  Thorpe,  Diplonia- 
tarium,  p.  153,  last  line;  Hrisan-heorgan,  id.  p.  .553;  Hrisen- 
beorgas,  id.  p.  549 ;  Hris-beorge,  id.  p.  331,  1.  5.  I  should 
therefore  explain  Risb ridge  as  meaning  'Hrisa's  bridge.'  There 
was  also  a  Norse  name  Hrisi ;  see  Risby.  Hrisa  was  probably 
its  Anglo-Saxon  equivalent. 

Woodbridge.  The  sense  is  obvious.  But  it  is  worth 
notice  that  the  Norman  often  pronounced  wood  (A.S.  wiidu) 
as  'ood,  without  the  tv.  Hence  we  find  the  spelling  Udebryge 
in  D.B.,  p.  27 ;  but,  by  the  time  the  scribe  (or  another  scribe) 
arrived  at  p.  90,  he  found  it  better  to  spell  it  Wudebrgge. 

8.    Brook. 

This  well-known  suffix  occurs  in  Holbrook,  Rushbrooke, 
Stradbroke,  Washbrook,  and  Wickhambrook. 

Holbrook.  Spelt  Holebrok,  Ipm.;  Holebroc,  H.R.;  D.B., 
p.  29 ;  A.S.  Holen-broc,  with  reference  to  Suffolk  (according 
to  Kemble,  Cod.  Dipl.  iv.  110,  1.  4);  but  a  better  form  is 
Holan-brdc{e)  for  which  Kemble  gives  many  references.  Here 
holan  is  for  holum,  the  old  dative  form.  The  place-name  is 
best  represented  by  the  nom.  hoi  brdc,  'hollow-brook';  i.e.  a 
brook  whose  banks  rise  above  it.  The  brook  flows  into  the 
Stour  at  Holbrook  Bay. 

Rushbrooke.  Spelt  Rescebroc,  H.R. ;  Ri/scebroc,  D.B., 
p.  166  ;  Ryssebroc,  in  iElfgar's  Will,  in  Thorpe,  Diplom.  p.  508  ; 
better  Riscbrvc,  Birch,  C.S.  ii.  81.  From  A.S.  rise  (also  resce), 
a  rush  ;  and  broc.     The  sense  is  'rush  brook.' 

Stradbroke.  There  is  here  but  a  small  brook,  flowing 
northward  towards  the  W^aveney.     We  find  Stradbroke  inaner 


12  THE    PLACE-NAMES   OF   SUFFOLK 

de  Eia  Jionore,  with  reference  to  Eye;  Ipm.  Ill  spelt  Statebroc, 
D.B.,  p.  137.  In  Matt.  v.  41,  where  the  A.V.  has  "a  mile," 
the  Latin  version  has  mille  passus,  and  the  A.S.  version  has 
thusend  stapa,  lit.  "a  thousand  of  steps";  but  the  Northumbrian 
version  has  thusend  strwdena,  showing  that  there  was  once  a 
word  st7'wda,  m.,  or  strwde,  f ,  meaning  "a  stride"  or  "a  step," 
the  obvious  original  whence  the  verb  to  straddle  is  derived. 
Hence  Strad brook  simply  means  'a  brook  across  which  one  can 
easily  stride'  or  straddle. 

Washbrook.  To  the  S.W.  of  Ipswich.  The  brook  flows 
eastward  into  the  Orwell.  The  etymology  is  by  no  means 
obvious,  as  it  has  only  been  associated  with  the  verb  to  wash 
in  popular  etymology.  There  was  a  personal  name  Wassa,  gen. 
Wassan,  whence  the  tribe  of  Wassings;  as  appears  in  various 
names  in  Kemble's  Index,  such  as  Wassan-burne  (Washbourn), 
Wassan-dun,  Wassan-hdm,  Wassing-burg  (Washingborough, 
Line),  Wassinga-tun  (Washington,  Suss.),  Wassing-welL  The 
original  sense  was  '  Wassa's  brook ' ;  whence  the  old  spelling 
Wassebroc  (Copinger). 

WiCKHAMBROOK.  Compounded  of  Wickham  and  brook. 
Not  connected,  except  by  the  accident  of  name,  with  Wickham 
Market  (see  pp.  63,  64),  which  is  exactly  to  the  east  of  it,  but 
more  than  thirty  miles  away.  The  brook  flows  eastward  into 
the  Glen  and  so  joins  the  Stour. 

9.    By. 

This  suffix  is  of  much  interest,  as  it  is  well  known  to  be  a 
sure  indication  of  Danish  occupation.  We  find  that  such 
occupation  was  after  all,  in  this  county,  really  very  slight. 
There  are  but  four  examples,  viz.  Ashby,  Barnaby  (or  Barnby), 
Risby,  and  Wilby.  The  two  former  are  within  a  few  miles  of 
the  east  coast ;  and  Wilby  is  only  some  twenty  miles  from  it. 
Risby,  however,  is  far  from  the  same,  farther  even  than  Bury. 
By  is  the  modern  Dan.  by,  Old  Dan.  byr,  Icel.  il/r  (more 
commonly  boer),  a  farmhouse,  farm,  or  town  ;  allied  to  Dan.  bo, 
Icel.  bua,  A.S.  buan,  to  build. 


NAMES   ENDING   IN   -BY  13 

AsHBY.  It  is  extremely  unlikely  that  Ash  should  here  be 
a  true  English  word,  as  it  would  hardly  combine  with  a  true 
Norse  suffix  such  as  -hi/.  It  has  been  proved  by  Bjorkman 
that  the  English  often  accommodated  Norse  words  to  their 
own  pronunciation;  and  it  is  to  be  noted  that  Copinger  records, 
as  old  spellings  of  this  name,  not  only  Asheby,  but  Askehy,  a 
spelling  which  frequently  occurs  in  Ipm.  We  even  find  Ashby 
in  Lines.,  where  the  old  form  is  Askeby.  The  prefix  Aske- 
represents  the  form  Asha,  gen.  sing,  of  Aski,  a  Norse  personal 
name.     The  real  meaning  is  'Aski's  town.' 

Barnaby,  Barnby;  to  the  S.W.  of  Lowestoft.  The  same 
name  as  Barnby-on-Don,  in  the  W.  Riding,  which  is  spelt 
Barnaby,  Bai^nehy  in  Ipm.  For  the  Suffolk  Barnaby,  D.B.  has 
Barnebei,  p.  5 ;  Barne-by,  p.  43.  This  is  a  clear  indication,  as 
Prof.  Moorman  points  out,  that  the  former  element  in  the 
name  is  not  from  the  A.S.  Beorna,  but  from  the  Dan.  personal 
name  Barni  (gen.  Barna),  a  name  recorded  by  Nielsen  in  his 
Old  danske  Personnavne.  The  sense  is  'Barni's  farm,'  or  'Barni's 
town.' 

RiSBY.  At  no  great  distance  from  Bury.  Spelt  Riseby, 
T.N. ;  Ryseby,  H.R.;  Risby,  Ipm.  Kemble  also  has  Riseby, 
in  his  Charter  no.  984 ;  but  the  spelling  is  very  late, 
and  of  little  value.  D.B.,  p.  152,  has  Risebi.  I  explain  the 
prefix  as  being  like  that  in  Risbridge  (above),  and  suppose 
that  it  began  with  Hr.  Moreover,  it  was  probably  Danish. 
Hence  Rise-  may  have  represented  the  Icel.  Hrlsa,  gen.  of 
Hrlsi,  which  occurs  as  a  nickname ;  see  Corpus  Poeticum 
Boreale,  ed.  Vigfusson  and  Powell,  vol.  ii.  p.  315, 1.  165  ;  Sigur^r 
Hrisi  Haraldz  sonr\,  i.e.  SigurSr,  nicknamed  Hrisi,  son  of 
Harold.  If  this  be  right,  the  sense  is  'Hrisi's  farm'  or'Hrlsi's 
town.'     So  also  in  Risbridge. 

Wilby.  To  the  E.S.E.  of  Eye.  Spelt  Wileby,  T.N.; 
W ilehegl I,  l\tm..',  Wilebey,^).^.,  p.  97.  The  D.B.  form  in  -bey 
must  be  significant,  since  Dr  Copinger  cites,  from  other  sources, 
such  spellings  as  Wilbeghe,  Wilebeigh,  Wilbey,  Wilbeye, 
Wilbeygh,  Wilebegh,  Wylebeg. 

At  p.  xxi  of  his  Introduction  to  the  W,  Riding  Place-names, 


14  THE   PLACE-NAMES   OF   SUFFOLK 

Prof.  Moorman  has  the  following  note  on  the  termination  -&?/. 
"This  word  existed  in  Old  Danish  in  the  form  -hyr  (Mod.  Dan. 
hy),  and  though  this  form  is  not  unknown  in  Old  Xorse,  the 
usual  O.N.  form  is  hor  [or  h<jer\  Phonology  shows  that  O.N. 
hor  would  have  become  Mid.  Eng.  he,  in  just  the  same  way 
that  O.N,  slogr,  sly,  became  M.E.  sUg."  It  thus  becomes  clear 
that,  in  this  place-name,  we  have  to  do,  not  with  the  usual 
Dan.  hy,  but  with  the  corresponding  0.  Norwegian  hor  (Norw. 
ho  in  Aasen).  This  shows  us  that  the  settler  in  Wilby  was  not 
a  Dane,  but  a  Norwegian, 

We  should  expect  the  prefix  Wil-  to  represent  the  gen.  of  a 
Norse  name.  Egilsson  gives  two  examples  of  the  name  Vili, 
gen.  Vila.  The  O.N.  V  was,  in  A.S.  times,  a  W,  when  Vili 
would  have  been  Wili.  I  therefore  propose  to  explain  Wilby 
as  meaning  'Will's  farm'  or  town.  The  final  -a  in  the  gen. 
Wila  was  easily  lost,  as  in  Ris-by  (above). 

10.  Camp. 

I  have  discussed  this  suffix  in  the  Place-names  of  Cambs., 
showing  that  it  represents  A.S.  camp,  a  field,  not  really  an 
A.S.  word,  but  borrowed  from  the  L,  campus.  The  Supple- 
ment to  Bosworth  and  Toller  now  gives  us  four  good  examples. 

BULCAMP.  A  hamlet  one  mile  N.W.  of  Blythburgh  (Kelly). 
Spelt  holecamp,  H.R. ;  Bidecampe,  D.B,,  p.  10.5.  Copinger  also 
cites  Bidchamp  (with  French  champ),  and  Bulfelda,  where  felda 
(field)  translates  camp.  The  D.B.  prefix  hide-  represents  the 
A.S.  hidan,  gen.  of  hula,  a  bull ;  see  hida  in  the  Supplement  to 
the  A.S,  Diet.  The  sense  is  '  bull's  field.'  Or,  if  Bula  were 
used  as  a  name,  '  Bula's  field,' 

11,  Clay. 

This  is  the  usual  E.  clay ;  it  occurs  in  Hinder-clay. 

HiNDERCLAY,  The  n  in  this  form  is  comparatively  modern ; 
all  the  old  forms  have  I  in  place  of  it,  and  the  form  in  D.B.  is 
Hilderclea,  p.  168.  In  a  late  copy  of  Ulfketel's  bequest  to  the 
Abbey  at  Bury,  it  appears  as  Hildercle;  see  Birch,  C.S.,  iii.  21 G, 
last  line.     It  is  not  from  the  A.S.  hild,  battle  (gen.  Midi),  as 


NAMES    ENDING    IN    -DALE,    -DENK  15 

this  will  not  account  for  the  r;  but  from  the  O.  Norse  feminine 
personal  name  Hildr,  of  which  the  gen.  was  Hildar;  see  Bjork- 
man.  Hence  the  literal  sense  is  '  Hildr's  clay,'  with  reference 
to  a  farm  with  a  clay  soil.  Hildr  (as  said  above)  was  the  name 
of  a  woman. 

12.     Dale. 

This  is  a  well-known  suffix,  of  obvious  meaning;  it  only 
occurs  in  Botesdale  and  Withersdale. 

BoTESDALE.  The  spelling  Botolvesdale  occurs  in  H.R. ; 
Avhich  explains  it  at  once  as  representing  '  B5twulf's  dale.' 
Botwulf,  mod.  E.  Botolph,  is  a  well-known  name,  and  is 
often  shortened  to  Botulf  and  Botolf ;  the  spelling  Botolph 
is  of  course  absurd,  though  perfectly  common.  The  spelling 
with  0  is  Norman ;  they  turned  the  A.S.  wulf  into  tvolf,  but 
they  did  not  thereby  affect  the  sound  of  it. 

Withersdale.  To  the  N.  of  Fressingfield,  and  not  far 
from  the  river  Waveney.  Spelt  Wytherisdal  in  H.R.  But 
Copinger  also  records  the  spellings  Wetheresdale,  Wethersdale, 
which  are  better.  Wethei^es  is  the  gen.  case  of  the  A.S. 
luetlier,  a  young  ram.  The  sense  is  'Wether's  dale';  for 
Wether  must  have  been  a  man's  name,  as  the  -es  suggests. 

18.     Dene. 

The  suffix  dene  represents  the  A.S.  denu,  a  valley.  It  is 
sometimes  confused  in  old  documents  with  -don,  representing 
the  A.S.  dun,  a  down,  and  they  cannot  always  be  distinguished. 
But  they  are  kept  apart  in  Suffolk,  if  we  include,  as  I  believe 
we  should,  the  name  Hundon  among  the  denes  and  Darmsden 
among  the  downs.     I  keep  them  separate. 

Dene  or  -den  occurs  in  Depden,  Elveden,  Framsden,  Fros- 
tenden,  Hundon,  Monewden,  Owsden,  Rattlesden,  Wantisden, 
and  Wetherden. 

Depden.  To  the  S.W.  of  Bury.  The  old  spellings  are  : 
Depeden,  H.R.,  T.N.;  Depedene,  R.B.;  Depdana,  D.B.,  p.  236. 
Cf.  Depeden  (Essex),  Ipm.  And  we  find  '  to  deopan  dsene,' 
Birch,   C.S.  ii.  135.     The  D.B.  often  denotes  the  A.S.  denu, 


16  THE   PLACE-NAMES   OF   SUFFOLK 

a  dene  or  valley,  by  -dana.  The  sense  is  simply  '  deep  valley.' 
The  long  e  has  been  shortened,  by  the  stress,  before  the 
following  pd. 

Elveden,  or  Elden  (Kelly).  To  the  S.W.  of  Thetford. 
Spelt  Elveden,  H.R.;  T.N.;  KB.;  Elvedena,  D.B.,  p.  156.    From 

the  A.S.  ^Ifan  denu,  '  ^Elfa's  valley ' ;  where  iElfa  is  a  pet- 
name  for  a  name  beginning  with  ^Elf-,  such  as  ^Elf-red  or 
iElf-ric.  An  example  of  the  name  .^Ifa  (ill-spelt  ^Ififa)  occurs 
in  Thorpe,  Diplomat.,  p.  562,  1.  2, 

Framsden.  To  the  S.E.  of  Debenham.  Spelt  Framesden, 
T.N.;  Frainisden,  H.R. ;  Framesdena,  D.B,,  p,  36.  From  the 
A.S.  Frames  denu,  lit.  '  Fram's  valley.'  Searle  instances  a 
moneyer  named  Fram.  It  is  simply  the  A.S.  fram,  valiant; 
an  adj.  well  fitted  to  be  used  as  an  epithet. 

Frostenden.  N.  by  W.  from  Southwold.  Spelt  Frosteden, 
Frostenden,  T.N.  Copinger  also  notes  the  forms  Frosenden, 
Frossenden,  as  if  the  t  had  been  inserted ;  and  when  we  com- 
pare these  with  the  form  Fi^oxedena,  in  D.B.,  p.  268,  we  may 
feel  tolerably  sure  that  such  was  really  the  case,  and  that  the 
form  in  D.B.  was  the  original  one  ;  especially  when  we  further 
compare  it  with  the  A.S.  Froxafeld,  which  occurs  in  Birch, 
C.S.,  iii,  432,  1.  22,  and  is  the  modern  Froxfield  (Hants.). 
Froxa  is  the  gen.  pi.  of  frox,  a  frog ;  and  the  sense  is  '  frogs' 
valley,'  The  form  Frossenden  may  be  compared  with  the  A.S. 
variant  froscan,  for  forscan,  gen.  sing. ;  cf.  Forscan-feld,  Birch, 
C.S.,  i.  452. 

Hqndon.  To  the  N.W.  of  Clare.  The  form  is  modern, 
and  it  was  formerly  Hunden.  It  appears  as  Hunden,  Ipm. ; 
Huneden,  H.R. ;  Hunendana,  D.B.,  p.  218.  The  A.S.  prefix 
was  Hunan,  as  in  Hunan-bricg  and  Hunan-weg,  both  in 
Kemble's  Index.  Moreover,  the  ii  was  long;  and  several 
Hunas  are  recorded.  The  sense  is  '  H ana's  valley,'  which 
suits  the  position  of  the  village. 

Monewden.  Sometimes  Monoden  (Kirby).  Five  miles 
S.W.  of  Framlingham  railway-station.  Spelt  Moneivedon, 
Ipm.;  Mungeden,  T.N.  The  w  represents  an  older  g]  Cop- 
inger cites  a  spelling  Monegedene,  and  we    find  Mmiegadena, 


NAMES   ENDING   IN   -DENE  17 

D.B.,  p.  90;  Mungadena,  D.B.,  p.  135;  Mungedena,  D.B,,  p.  184; 
Mangedena,  D.B.,  p.  134.  The  suffix  is  the  A.S.  denu,  a  valley ; 
the  oldest  forms  of  the  prefix  arc,  apparently,  those  in  D.B.,  viz. 
Munega,  Munga,  Mimge,  Manga.  I  find  no  A.S.  forms  which  will 
account  for  these,  and  I  do  not  know  their  origin.  As  a  possible 
guess,  I  suggest  that  the  prefix  may  have  arisen  from  the  0. Norse 
fem,  personal  name  Mundger'^r,  fiom  which,  according  to  Rygh, 
the  Scand.  place-name  Munge-rad  is  derived.  If  this  should  be 
right,  the  sense  would  be  '  Mundgerthr's  valley.' 

OwsDEN  (Kelly),  or  Ousden.  Near  Lidgate,  towards  the 
western  border  of  the  county.  Old  spellings  are :  Ovesden, 
T.N. ;  Ovisdene,  Ipm. ;  Ovesdene,  R.B. ;  Uvesden,  H.R. ;  and 
Vuesdana,  for  Uvesdana,  D.B.,  p.  45.  Kemble  gives  Ufesford 
as  a  place-name  ;  but,  according  to  Birch,  this  is  a  misreading. 
He  also  gives  Ufford  as  the  old  form  of  Ufford  (Northants.), 
but  this  may  be  a  contracted  form ;  see  the  account  of  the 
Suffolk  Ufford  below.  Searle  gives  several  examples  of  the 
personal  name  Ufa  (of  the  weak  declension,  genitive  Ufan) ; 
corresponding  to  which  we  might  expect  to  find  the  form 
Ufe  (of  the  strong  declension,  genitive  Ufes).  Similarly,  cor- 
responding to  the  weak  form  Ofa  (six  examples),  a  strong  form 
Ofe  is  known.  Hence  the  form  Ufe  may  safely  be  assumed ; 
so  that  the  original  sense  of  the  place-name  was  '  Ufe's  valley.' 

Rattlesden.  To  the  W.  of  Stowmarket.  Spelt  Ratles- 
rfewe,  H.R. ;  R.B. ;  Ratilisden,  l^m.\  Ratlesden,  T.N.;  Ratlesdena, 
D.B.,  p.  165.  Rattlesden  is  certainly  referred  to  in  a  Charter 
of  Edward  the  Confessor,  printed  in  Kemble,  CD.,  iv.  245, 
where  we  find : — "  in  comitatu  Sudfolc,  Hertest,  Glemesford, 
Hecham,  Rattesdene,"  &c.  But  the  copy  is  not  very  well  spelt, 
and  we  may  suspect  that  the  form  intended  was  Ratlesdene,  in 
conformity  with  all  the  other  evidence.  An  A.S.  *Ratles  would 
imply  a  nom.  *Ratel ;  or,  if  we  compare  tlie  A.S.  hrcetel-wyrt, 
answering  to  our  modern  rattle-ivort,  we  might  infer  such  a 
form  as  *Hr8etel.  In  any  case,  we  may  assume  that  the  valley 
here  discussed  was  named  from  a  man  whose  personal  designa- 
tion was  *Ratel  or  *Hra3tel.  There  we  must  leave  it,  for  want 
of  evidence. 

C.  A.  S.  Octavo  Series.     No.  XLVI.  2 


18  THE    PLACE-NAMES   OF   SUFFOLK 

Wantisden.  Three  miles  to  the  S.E.  of  Wickham  Market 
railway  station  (Kelly).  Old  spellings  are:  W a ntesden,  1pm.; 
Wantesdena,  D.B.,  pp.  32,  207.  Apparently  the  sense  is 
'  Want's  valley.'  The  personal  name  Want  is  recorded  in 
the  Liber  Vitse  of  Durham. 

Wetherden.  To  the  N.W.  of  Stowmarket.  Spelt  Wether- 
done  (error  for  Wetherdene,  as  other  records  show),  Ipm. ; 
Wetherden,  Weutherden,  cited  by  Copinger ;  Wederdena,  D.B., 
p.  159,  with  d  for  th.  A  simple  compound,  moaning  '  wether- 
valley.' 

14.    Down. 

In  names  that  end  in  -don,  the  suffix  is  the  unstressed  form 
of  down,  A.S.  dtui,  a  hill. 

Examples  occur  in  Balliugdon,  Brandon,  Claydou,  Darmsden 
(for  Darmsdon),  Hawkedon,  Raydon,  Reydon,  and  Thorndon. 

Ballingdon.  a  hamlet  near  Sudbury.  I  find  no  old 
spellings.  If  the  form  is  correct,  it  may  mean  '  dowai  (or 
hill)  of  the  BaBllings,'  or  of  'the  sons  (or  family)  of  Bsell.' 
BaiU  is  recorded  as  a  personal  name.  Cf.  D.B.,  Balles-bi,  i.e. 
Balby  (Yorks.). 

Brandon,  or  Brandon  Ferry  (Kelly).  Spelt  Braundone, 
R.B.;  Brandona,  D.B.,  p.  202  ;  Brandons  (late),  in  Kemble,  CD., 
iv.  245.  I  explain  it  from  the  0.  Norse  Branda,  gen.  of  Brandi, 
a  weak  form  used  beside  the  strong  form  Brandr  (gen.  Brands). 
See  the  account  of  Brandr  in  Rygh,  who  shows  that  Brandr  was 
in  very  common  use,  and  that  Brandi  also  occurs  in  place-names. 
The  sense  is  '  Brandi's  down.'  Brand-  occurs  also  in  English  in 
compounds,  such  as  Brand-wulf.     The  result  is  conjectural. 

Claydon.  To  the  E.  of  the  river  Gipping.  It  occurs  in 
D.B.  as  the  name  of  a  hundred,  distinct  from  Bosmere ;  but 
Bosmere  and  Claydon  are  now  taken  together  as  forming 
but  one  hundred.  Spelt  Cleydon,  H.R. ;  T.N. ;  Ipm.  But 
D.B.  has  Claindune  or  Claindone  frequently,  with  Clain-  as 
the  former  element,  in  which  the  n  must  be  accounted  for. 
It  may  easily  represent  the  adjectival  suffix  -en;  the  form 
clayen,  '  made  of  clay,'  or  '  clayey,'  occurs  in  Wyclif 's  translation 
of  Job  iv.  19  (N.E.D.).     The  sense  is  '  clayey  down.' 


NAMES   ENDING   IN   -DOWN  19 

Darmsden.  To  the  S.E.  of  Ncedham  Market.  The  name 
originally  ended  in  -don  or  -doun,  representing  the  A.S.  dii)), 
a  down.  Copinger  gives  Derniodesdon,  Dormesdon  ;  but  more 
important  spellings  are  those  in  D.B.  and  Ipm.  D.B.  has 
Dermodesduna,  pp.  28,  205  ;  and  Ipm.  has  the  still  fuller 
form  Deorniondesdoiuie,  p.  218 ;  which  explains  it  at  once. 
Deormondes  represents  the  A.S.  Deormundes,  gen.  of  Deormund 
(a  known  name) ;  and  the  original  sense  was  '  Deormund's 
down.' 

Hawkedon.  Nearly  to  the  S.S.E.  of  Bury,  in  the  direction 
of  Clare.  Old  spellings  are  :  ZTftit^o/^,  T.N.,  Ipm. ;  Haukedon, 
II.R. ;  Hauochenduna,  D.B.,  p.  136  (with  che  for  ke)]  Hauokedima, 
D.B.,  p.  232.  At  first  sight,  we  might  suppose  that  Hauochen- 
answers  to  the  A.S.  Heafecan,  given  as  occurring  in  Heafecan- 
berh,  in  Kemble's  Charters,  nos.  291  and  292  ;  but,  according  to 
Birch's  revision  of  these  charters,  the  name  has  here  been  mis- 
read, and  appears  in  the  charters  in  the  forms  Heasecan  (thrice) 
and  Heahsecan  (once) ;  proving  that  the  supposed  Heafecan  has 
no  real  authority.  The  form  in  D.B.  can  hardly  be  correct,  as  the 
O.  Merc,  hafoc,  A.S,  Jieafoc,  is  a  strong  masculine  (gen.  hafoces, 
heafoces),  no  case  of  which  can  end  in  -an.  It  is  therefore 
worth  noting  that  Copinger  also  gives  the  old  spellings  Haukes- 
den  and  Haiukesden,  though  without  a  reference.  Kemble  has 
several  forms  beginning  with  Hafoces-  (followed  by  hlww,  ora, 
pyt,  and  tun),  so  that  the  D.B.  form  should  rather  have  been 
Hauochesduna.  That  Hafoc  was  a  man's  name,  with  a  gen. 
Hafoces,  can  be  safely  concluded  from  its  frequent  occurrence 
in  place-names.  We  find  Hawkesbury  (Glouc.),  Hawksdale 
(Cumb.),  Hawksdown  (Devon),  Hawkshead  (Lanes.),  and 
Hawkesworth  (Notts) ;  besides  Hauxton  (Cambs.),  which  is 
merely  a  form  of  Hawkston.  It  is  known  that  the  genitival 
-s  disappeared,  occasionally,  at  rather  an  early  date,  in  some 
place-names,  whilst  in  others  it  has  remained.  The  probable 
sense,  in  this  case,  is  '  Hawk's  down ' ;  where  Hawk  (0.  Merc. 
Hafoc)  was  a  personal  name.  This  can  be  proved  by  two 
considerations:  (1)  the  occurrence  of  Hafeces  hlTliu,  i.e.  'Hawk's 
burial-mound,'  in  Birch,  C.S.,  ii.  377,  1.  18;   and  (2)  the  fact 

2—2 


20  THE   PLACE-NAMES    OF   SUFFOLK 

that  the  Icel.  haiikr,  'a  hawk,'  was  a  common  personal  name 

(Rygh). 

Raydon  and  Raydon  St  Mary  are  to  the  S.E.  of  Hadleigh ; 
Reydon  lies  to  the  N.W.  of  Southwold,  far  from  the  others. 
Both  were  formerly  spelt  with  ei  or  ey ;  so  that  we  may  select 
Reydon  as  being  the  better  form.  Old  spellings  are  :  Reydon, 
Ipm.;  Reidiine,  H.R.;  Reydon,  T.l^.',  but  D.B.  has  Reinduna, 
p.  194.  Rey  is  an  occasional  form  of  E.  rye,  A.S.  ryge;  and 
the  D.B.  form  rein  represents  the  A.S.  adj.  rygen,  belonging  to 
or  abounding  in  rye.  We  actually  find  the  A.S.  form  of 
Reydon  in  iElfflsed's  Will,  in  which  it  is  spelt  Rigindun  (for 
Rygendun) ;  see  Birch,  C.S.,  iii.  603,  1.  29.  Hence  the  sense 
is  '  down  abounding  in  rye,'  or  '  rye-down.'  For  the  form  rey, 
see  the  N.E.D.,  s.v.  rye. 

Thorndon.  To  the  S.  of  Eye.  Spelt  Thornedon,  R.B. ; 
answering  to  the  A.S.  Thorndun,  of  which  Kemble  has  four 
examples.     The  sense  is  obviously  '  thorn  down.' 

15.    Edish,  or  Eddish. 

The  prov.  E.  eddish  (also  written  edish)  is  in  general  dia- 
lectical use,  with  the  sense  of  '  aftermath,'  or  second  crop  of 
grass  or  clover;  the  A.S.  form  being  edisc.  It  is  the  origin 
of  the  modern  -dish  in  Brundish  and  Cavendish. 

Brundish.  Nearly  to  the  N.  of  Framlingham.  Ipm. 
mentions  a  Burnedishe  in  Staffs.  Copinger  gives  a  number 
of  spellings,  among  which  are  Bornedisce,  Burnedich(e),  Burne- 
dish,  and  Burnedissh(e) ;  all  of  which  would  result  from  an 
A.S.  *burn-edisc.  I  have  no  doubt  that  it  means  'bourn- 
eddish,'  i.e.  a  meadow  beside  a  bourne  or  stream  that  was 
mown  for  aftermath. 

Cavendish.  Nearly  to  the  E.  of  Clare.  Spelt  Kavanedis, 
Cavendish,  Ipm.;  Cauenedis,  T.N.;  Cavenedys,  H.R.;  Kauanadis, 
D.B.,  p.  335.  The  prefix  is  the  same  as  in  Cavenham  (below), 
and  represents  Cafan,  the  gen.  case  of  the  personal  name  Cafa. 
The  sense  is  '  Cafa's  eddish,'  or  '  Cafa's  meadow  for  aftermath.' 


NAMES   ENDING   IN   -EY  21 

16.     Ey. 

This  very  common  suffix  represents  the  Anglian  eg,  A.S.  lec) 
Ig,  an  island.  It  meant  not  only  '  island  '  in  the  modern  sense, 
but  peninsula,  or  any  piece  of  land  wholly  or  partially  sur- 
rounded by  brooks  or  marshy  country.  In  D.B.  and  Latin 
documents  it  is  often  expressed  by  eia.  It  occurs  alone  in  Eye, 
and  in  composition  in  Bawdsey,  Bungay,  Campsey  Ashe,  Kirsey, 
and  Lindsey. 

Eye.  It  is  situate,  says  Kelly,  "  at  the  confluence  of  two 
rivulets,  in  a  low  situation."  One  of  these  streams  is  the  river 
Dove.  Note  the  spellings  Eye,  Eya,  H.R.,  T.N.;  Eya,  Eie,  Eye, 
la  Eye,  R.B.;  Eiam  (ace.  case),  D.B.,  p.  78.  The  final  e  is  due 
to  the  use  of  the  dative  case,  the  prep,  cet  (at)  being  understood, 
as  usual.  Thus,  in  the  A.S.  Chronicle,  an.  855,  we  find  "on 
Sceap-ige,"  i.e.  in  Sheppey.  For  the  A.S.  Ig,  the  Anglian,  Old 
Mercian,  and  Old  Norse  form  was  eg,  which  accounts  for  the 
former  e.  The  A.S.  Ig  is  connected  with  ea,  a  stream ;  the  two 
forms  are  often  ignorantly  confused. 

Bawdsey.  The  name  perhaps  belonged  originally  to  Bawd- 
sey Manor,  which  is  at  some  distance  from  the  present  village, 
and  near  Bawdsey  Point.  The  reference  is  to  its  situation  in 
the  peninsula  between  the  river  Deben  and  the  sea,  which 
terminates  in  Bawdsey  Point.  Old  spellings  are :  Baudeseye, 
Ipm. ;  Balders,  H.R. ;  Baldeseia,  D.B.,  p.  73.  But  Copinger 
gives  other  spellings,  such  as  Balderescia  (with  c  for  e),  Bal- 
dreseia,  Baudersey,  Baudreseye,  Baudrissey ;  which  clearly  show 
that  the  fuller  form  was  Baldereseye ;  and  Baldet-es-  of  course 
represents  the  O.  Merc.  Baldheres,  gen.  of  Baldhere,  the  Mercian 
form  of  the  A.S.  Bealdhere,  a  known  name.  The  sense  is 
'  Baldhere's  island.' 

Bungay.  The  situation  is  remarkable,  as  the  river  Waveney 
is  here  extremely  deflected,  and  forms  a  horse-shoe  bend  round 
the  peninsula  lying  to  the  N.W.  It  is  therefore  on  an  eye,  in 
the  old  sense  of  that  word.  The  old  spellings  are,  accordingly, 
Bungeye,  H.R.;  Bungeia,  R.B.;  Bongeia,  D.B.,  p.  15;  Buugheum 
(ace.  case),  D.B.,  p.  39.     The  prefix  Bung-  is  of  Norse  origin  ; 


22  THE    PLACE-NAMES   OF   SUFFOLK 

from  the  Icel.  hungi,  a  convexity,  elevation  ;  Norweg.  hiaiga, 
a  little  heap ;  closely  allied  to  the  Dan.  bunke,  a  heap,  a  pile, 
and  connected  with  the  E,  bunch.  The  original  sense  of  the 
word,  according  to  Falk  and  Torp's  Dan.  Etym.  Diet.,  Avas 
'  rounded  elevation.'  The  sense  is  '  rounded  elevation  on  a 
peninsula';  just  as  the  old  name  of  Durham  was  Diin-holm, 
i.e.  '  down-island,'  or  '  hill-island.'  It  may  be  added  that  I 
have  already  given  this  explanation  in  my  Place-names  of 
Cambridge,  where  the  pronunciation  of  ng  as  ngg  is  exem- 
plified, as  in  Gamling-ay. 

A  favourite  derivation  of  this  name  was,  once  upon  a  time, 
the  F.  bon  gue,  or  '  good  ford '  (if  there  ^vas  one).  Of  course 
philology  forbids  the  derivation  of  forms  that  occur  in  Domes- 
day Book  from  modern  French ;  and  it  is  well  to  remember 
that  the  Norman  for  '  ford '  was  guet  or  wet,  and  that  the 
Norman  did  not  pronounce  gu  like  the  gu  in  the  F.  gue,  but 
like  the  gu  in  anguish.  The  author  of  this  egregious  fable 
has  not  told  us  liow  to  obtain  the  sound  of  gay  from  that  of 
givet 

Campsea  Ashe,  or  Ashe  by  Campsea  (Kelly).  We  need 
not  trouble  about  Ashe,  which  refers  to  the  familiar  tree-name. 
The  place  lies  between  Saxmundham  and  Woodbridge,  and 
there  are  several  Ashes  in  the  neighbourhood,  viz.  Ash  Corner 
to  the  W. ;  the  remains  of  Mill  Ash  Abbey  to  the  S.S.W. ;  and 
Ash  Green  and  Ash  High  House  to  the  S.E.  The  name  is 
spelt  Campsey  Ash  in  the  Ordnance  Map.  D.B.  has  Camjjes 
ea,  p.  26 ;  some  other  old  spellings  are  quoted  by  Copinger,  viz. 
Ashe  juxta  Campessey,  Ayssh  juxta  Camsey,  Campeseia,  Gamp- 
essey.  The  suffix  appears  to  be  -ey,  island.  As  to  the  prefix, 
I  am  uncertain;  but  Rygh  gives  a  Norse  personal  name  Kampi, 
as  appearing  in  some  place-names,  which  may  have  become 
Camp  in  English  ;  whence  '  Camp's  island.' 

Kersey.  To  the  N.W.  of  Hadleigh.  I  have  already  dis- 
cussed this  name  in  a  paper  for  the  Philological  Society,  printed 
in  the  Transactions  for  1907-10,  at  p.  258.  I  there  show,  in 
opposition  to  the  statement  in  the  N.E.D.  as  to  there  being  no 
known  connexion  between  Kersey  and  Kersej'  cloth,  that  the 


NAMES   ENDING    IN    -EY  23 

Suffolk  cloth-making  is  expressly  mentioned  in  Hall's  Chronicle, 
under  the  date  1526.  Indeed,  the  poet  Skelton,  in  his  piece 
entitled  "  Why  Come  ye  nat  to  Courte,"  refers  to  the  "  cloth- 
making"  of  Sprynge  of  Laniiam,  11.  930 — 2.  Dyce,  in  his  note, 
strangely  explains  "  Lanam  "  as  "  Langham  in  Essex,"  whereas 
it  is  the  usual  pronunciation  of  the  Suffolk  Lavenham,  and 
"  the  coats  of  arms  of  the  Springs,  wealthy  clothiers  in  the 
fifteenth  and  sixteenth  centuries,  and  great  benefactors  of 
the  church"  can  still  be  seen  in  Lavenham  Church;  see 
Bygone  Suffolk,  p.  76.  This  proves  the  point  as  to  the 
naming  of  Kersey  cloth  from  Kersey  in  Suffolk ;  especially 
when  taken  in  conjunction  with  my  note  upon  Lindsey  (below). 
The  old  spellings  are :  Kerseye,  H.R. ;  Kareshey  (error  for 
Karesey),  Ipm. ;  Kereseye,  in  1279  (see  Bardsley);  Careseia, 
D.B.,  p.  217.  All  from  the  A.S.  Gceres-lg,  of  which  the  gen. 
case  Gwres-ige  occurs  in  Birch,  C.S,,  iii.  608,  1.  3  from  the 
bottom.  Here  Cceres  is  the  gen.  of  Ccer,  the  same  name  as 
the  Gar  mentioned  in  Searle.  Hence  Kersey  means  "  Cser's 
island  "  or  "  Car's  island."  This  name  of  Car  looks  like  Celtic. 
Kersey  cloth  was  known  in  1376  (Victoria  Hist,  of  Suffolk) ; 
and  Kersey  Priory  is  as  old  as  1158. 

Lindsey.  Not  far  from  Kersey,  and  associated  with  it  in 
the  old  days  of  the  cloth  manufacture  in  Suffolk.  But  the  old 
name  was  Lellesey  or  Lillesey,  even  as  late  as  the  sixteenth 
century.  Old  spellings  are:  Lelleseye,  Ipm.  (a.d.  1263);  and 
Copinger  gives  Lelessey,  Lellesey,  Lelsey,  as  well  as  Lillesey, 
Lyllesey.  Ipm.  also  has  a  Lylleseye  in  Sussex  (perhaps  an 
error  for  Suffolk).  However,  the  authentic  A.S.  form  was 
rather  Lill  than  Lell ;  Kemble's  index  has  Lilies  beam,  '  Lill's 
tree,'  as  well  as  Lilies  ham,  or  '  Lill's  home.'  Moreover,  the 
related  weak  form  Lilla  also  appears  in  the  same,  as  in  Lillan- 
hrycg,  '  Lilla's  ridge,'  and  Lillanwelle,  '  Lilla's  well.'  We  may 
therefore  well  suppose  that  the  original  name  meant  '  Lill's 
island.'  The  subsequent  change  to  Lindsey  may  have  been 
due  to  confusion  with  that  name,  which  was  better  known. 
A  large  portion  of  Lincolnshire  was  called  Lindsey,  which 
appears    even    in    modern    maps.     I    have    already    shown,    in 


24  THE   PLACE-NAMES   OF   SUFFOLK 

my  paper  on  Kersey,  that  the  material  called  Linsey-wolsey 
certainly  took  its  name  from  the  Suffolk  town ;  for  otherwise 
it  could  never  have  had  the  name  of  Lylse  wulse  in  the  time 
of  Skelton,  who,  in  his  poem  entitled  "Why  Come  ye  nat  to 
Courte,"  at  1.  128,  has  the  expression  :  "  To  weve  al  in  one  lome 
[loom]  A  webbe  of  lylse-iuulse  "  ;  with  a  punning  reference  to 
his  enemy,  the  cardinal,  who  was  one  of  the  most  celebrated 
men  that  Suffolk  has  ever  produced.  (The  spelling  Lynsey 
occurs  in  V.E,) 

17.     Field. 

This  is  a  well-known  suffix  in  place-names,  and  appears  in 
Ashfield,  Bedfield,  Bedingfield,  Bradfield,  Bramfield,  Bredfield, 
Charsfield,  Cockfield,  Cratfield,  Crowlield,  Fressingfield,  Homers- 
field,  Huntingfield,  Laxfield,  Metfield,  Mickfield,  Pakefield, 
Redlingfield,  Ringsfield,  Shadingfield,  Stanningfield,  Stansfield, 
Sternfield,  Waldingfield,  Waldringfield,  Wattisfield,  Westerfield, 
Whatfield,  Wingfield,  and  Withersfield ;  i.e.  just  thirty  times. 

Ashfield.  There  is  an  Ashfield  Magna  to  the  K  of 
Elmswell,  and  another  Ashfield  nearly  to  the  E.  of  Debenham. 
The  meaning  is  obvious,  but  the  Norman  scribes  had  much 
ado  to  spell  it,  as  there  was  no  sh  in  Norman  at  all.  Hence 
we  find  in  D.B.,  at  p.  193,  the  form  Asfelda,  at  p.  29,  Asse- 
felda,  and  at  p.  173,  Eascefelda.  And  they  sometimes  spelt  Ash 
as  Esse. 

Bedfield.  Nearly  to  the  N.E.  of  Debenham,  and  to  the 
N.W.  of  Framlingham. 

Old  spellings  are  scarce ;  but  Copinger  gives  the  form 
Bedefeld,  which  is  quite  satisfactory.  Here  the  prefix  Bede- 
represents  the  A.S.  Bedan,  the  gen.  case  of  Beda,  which  is  a 
famous  name.  The  sense  of  Bedfield  is  'Beda's  field,'  just  as 
that  of  Bedford  is  'Beda's  ford.'  The  name  must  have  been 
common ;  and  no  doubt  the  men  who  gave  their  names  to 
Bedfield  and  to  Bedford  were  not  the  same;  and  the  'venerable 
Bede'  was  a  third.  The  last  mentioned  lived  at  an  early  date, 
when  the  name  was  spelt  Bieda. 


OF  \ 

i  UNI';"- 

NAMES   ENDING   IN   -FIELD\  ,  ,  25 

-   .i-    I  i^:.■■ 

Bedingfield.  To  the  S.E.  of  Eye.  Spelt  Bediiu/feld  in 
Ipm. ;  D.B.  has  Badingafelda,  p.  59,  and  Bedingafelda,  p.  276. 
The  correct  A.S.  form  would  be  Bedinga  feld,  i.e.  '  the  field  of 
the  Bedings'  or  'of  the  sons  of  Beda.'  Beding  is  the  regular 
patronymic  form  from  that  source. 

Bradfield.  Kelly  has  a  Bradfield  St  Clare  aud  a  Bradfield 
St  George ;  as  well  as  Bradfield  Combust  or  Brent  (i.e.  burnt) 
Bradfield,  so  called  because  its  old  hall  was  burnt  in  1327. 
The  old  spellings  are  Bradefel,  T.N, ;  Bradefeld,  H.R. ;  and 
Bradefella,T>.^.,  p.  21.  Bradfield  (Sussex)  appears  as  Bradan- 
feld  in  several  A.S.  charters ;  see  Kemble's  index.  Here 
hrddan  is  the  weak  dat.  of  the  A.S.  hrdd,  broad ;  and  the  sense 
is  'broad  field.'  The  long  a  is  shortened  by  the  stress,  before 
the  df. 

Bramfield.  To  the  S.  of  Halesworth.  Spelt  Bramfeld 
H.R. ;  but  Brunfelda  in  D.B.,  p.  24,  under  the  account  of 
Walepola  (Walpole).  Copinger  also  notes  such  spellings  as 
Bromfield  and  Brumfield,  which  are  of  some  help,  and  show 
that  it  is  quite  different  in  origin  from  Bramfield,  Herts.,  which 
seems  to  have  meant  '  Branda's  field.'  See  further  under 
Bramford. 

Bredfield.  To  the  N.  of  Woodbridge.  Spelt  Bredfelde, 
Bredefeld,  Ipm.;  Bredefelda,  D.B.,  p.  75.  This  is  a  highly 
interesting  example,  as  it  introduces  a  Frisian  form.  The  prefix 
is  not  the  A.S.  hrdd,  broad,  as  in  Bradfield,  but  the  0.  Frisian 
hred,  or  breed,  with  the  same  sense.  That  is,  the  sense  is  '  broad 
field,'  as  in  the  former  case,  but  there  is  a  difference  of  dialect. 

Charsfield.  To  the  W.  of  Wickham  Market.  Spelt 
Charsfeld,  Ipm. ;  Ceresfella,  D.B.,  p.  26 ;  Cerresfella,  D.B., 
p.  186.  D.B.  has  Ce  for  E.  Che,  and  usually  turns  the  A.S. 
feld  (field)  into  fella.  In  the  Crawford  Charters,  ed.  Napier 
and  Stevenson,  p.  33,  1.  28,  we  find  the  spelling  Caresfeld;  a 
form  difficult  to  account  for.  The  D.B.  form  Cerres  may  be 
right.  If  so,  the  sense  is  '  Cerr's  field,'  the  A.S.  Cerr  becoming 
E.  Char,  as  in  Cert,  the  A.S.  form  of  Chart  (Kent).  The  A.S. 
Cerr  is  suggested  by  the  form  Cerringes,  the  A.S.  spelling  of 


26  THE    PLACE-NAMES    OF   SUFFOLK 

Charing  (Kent);  see  Birch,  C.S.  i.  411;  cf.  also  p.  410,  I.  25. 
Cerring  is  the  patronymic  form  of  Cerr,  so  that  there  is 
evidence  for  the  name. 

CocKFiELD.  The  railway  station  is  on  the  line  from  Bury 
to  Lavenham.  Spelt  Cocfelde,  Cokefelde,  RB. ;  Cokefeld,  H.R. 
Oddly  spelt  Cochanfelde  in  Bircli,  C.S.  iii.  603,  1.  1 ;  also 
Cokefelde  in  a  late  hand,  in  the  same,  p.  604,  1.  2.  Compare 
Coccan-hiirh  in  Kemble's  index.  The  sense  appears  to  be 
'Cocca's  field.' 

Cratfield.  Nearly  to  the  E.S.E.  of  Halesworth,  beyond 
Cookley.  Spelt  Cratfeld,  Ipm. ;  Cratefeld,T.'^.;  Ipm.;  Crata- 
felda,  D.B.,  p.  269.  The  prefix  does  not  appear  to  be  English, 
but  rather  Dan.  krat,  a  thicket,  a  copse  ;  Mid.  Dan.  krat,  a 
thorn-bush ;  Swed.  dial,  kratt,  the  same.  The  sense  would  be 
'field  covered  with   brambles.' 

Ckowfield.  To  the  N.E.  of  Needham  Market.  The  name 
has  been  modified,  and  its  original  sense  was  other  than  it 
seems  to  be.  Spelt  Groffeud  (for  Crojfeld),  Ipm.,  p.  55 ; 
Crofelda,  D.B.,  p.  187.  Copinger  also  records  the  forms 
Groffeld  and  Croftfield.  All  of  these  suggest  an  A.S.  form 
croft-feld,  with  the  sense  of 'croft-field';  i.e.  a  small  enclosure. 

Fressingfield.  Otherwise  Fresingfield  (Kirby).  Spelt 
Fresing-feld,  H.R..;  Fresing-feud  (for  -feld),  Ipm.,  p.  161a; 
Fresen-feld,  Ipm.,  p.  1616.  Copinger  also  reports  the  forms 
Fr^esi/ngefeld,  Fresyngfeld,  and  the  like.  But  the  A.S.  form  is 
rightly  Fresena  feld,  i.e.  'field  of  the  Frisians,'  in  agreement 
with  the  form  Fresenfeld.  Here  Fresena  is  the  gen.  pi.  of 
Fresa,  a  Frisian.  This  is  a  very  interesting  result ;  note  that 
there  is  a  similar  allusion  in  the  names  Freston  and  Friston. 

HOMERSFIELD.  Kelly  has  "  Homersfield,  or  St  Mary 
South  Elmham."  Spelt  Humersfeld,  R.B.;  Humeresfeld,  R.B. ; 
Humhresfelda,  D.B.,  p.  197.  In  the  last  form  the  h  is  probably 
intrusive,  as  it  does  nut  appear  in  the  modern  form.  Humeres- 
of  course  represents  the  gen.  case  of  a  masc.  personal  name, 
which    can    hardly    have    been    other    than    *Hiinmger.      For, 


NAMES   ENDING   IN   -FIELD  27 

though  this  name  is  not  on  record,  Hiin-  (never  Hum-)  is  a 
common  prefix  in  such  names,  and  -m.Tn-,  as  in  ^If-mser, 
^thel-mier,  is  a  common  suffix.  I  explain  this  name  as 
'  HunmlBr's  field.' 

HUNTINGFIELD.  Near  Heveningham,  to  the  S.W.  of 
Halesworth.  Spelt  Huntingefeld,  R.B.,  Ipm. ;  Huntyngfeld, 
H.R.;  Huntingafelda,  D.B.,  p.  Gl,  representing  an  A.S.  Huntinga- 
feld,  i.e.  'field  of  the  Huntings.'  Hunting  is  a  tribal  name, 
from  the  personal  name  Hunta,  our  modern  Hunt.  It  does 
not  refer  to  A.S.  huntung,  a  hunting. 

Laxfield.  To  the  W.S.W.  of  Halesworth.  Spelt  Lax/eld, 
H.R.;  Laxafella,  D.B.,  p.  26;  Laxafelda,  D.B.,  p.  27.  Cf. 
LcBxa-dyne,  Birch,  C.S.  iii.  602,  1.  9 ;  and  Leaxan  oc,  in  the 
same,  ii.  510  (in  a  late  copy  of  a  charter).  It  hence  appears 
that  Laxfield  represents  an  A.S.  form  Leaxan  feld;  where 
Lwxan  is  the  gen.  case  of  Lwxa,  a  personal  name.  This  name 
is  quite  un-English,  and  is  obviously  founded  upon  the 
extremely  common  Norse  word  lax,  '  a  salmon.'  It  was 
probably  a  mere  epithet,  though  the  original  sense  may  have 
been  'salmon-er,'  i.e.  a  fisher  for  sahnon.  The  final  -a  is  often 
agential  in  Old  English. 

The  D.B.  form  Laxin-ton  (for  A.S.  La3xan-tun)  occurs  in 
the  D.B.  for  Yorkshire.  And  there  are  Laxtons  in  Northants. 
and  Notts. 

Metfield.  To  the  S.E.  of  Mendham,  which  is  on  the 
Waveney.  I  find  no  old  spellings ;  but  Copinger  records 
Metefeld  and  Medefeld.  The  latter  is  obviously  the  older  form, 
and  shows  that  the  old  sense  was  simply  '  mead-field,'  or 
'meadow-field';  i.e.  a  field  for  mowing. 

MiCKFiELD.  Near  Debenham,  westwards.  Spelt  Mikele- 
feld,  T.N. ;  Mikelfeld,  Ipm. ;  Mucelfelda,  D.B.,  p.  273.  Obviously 
'  mickle  field,'  i.e.  large  field. 

The  A.S.  formula  on  miclan  feld  occurs  in  Birch,  iii.  342 
(no.  1109),  which  accounts  for  the  hard  k  (ck);  its  preservation 
is  due  to  the  contraction  of  micelan  to  miclan  in,  the  dative 
case.     Cf.  Micklefield  in  the  West  Riding. 


28  THE    PLACE-NAMES   OF   SUFFOLK 

Pakefield.  To  the  S.  of  Lowestoft.  Spelt  Paggafella, 
D.B.,  p.  5 ;  with  gg  (hard  g)  for  h ;  but  Copinger  records  the 
forms  Pake/eld  and  Pakelefeld.  The  prefix  Pake-  is  short  for 
Paken-,  as  in  Pakenham,  which  see,  Pakele-  seems  to  be  a 
diminutive,  as  if  for  Pakelen,  for  an  A.S.  *  Pacela,  a  weak  form 
allied  to  the  A.S.  Pacel,  as  seen  in  Pacles-ham  (Kemble). 
I  explain  Pakefield  as  representing  'Paca's  field,'  with  the  same 
prefix  as  in  Pakenham. 

Redlingfield.  To  the  N.  of  Debenham.  Spelt  Ridlingfeld, 
Ipm.;  better  with  e,  as  in  Pedelingf eld,  Redely ngf eld  (Copinger); 
Radinghefelda,  D.B.,  p.  79.  In  the  last  form,  the  ghe  is  for 
A.S.  ga,  and  an  I  or  el  has  been  omitted;  so  that  it  points 
back  to  an  A.S.  Rcedelinga  feld  or  Rddlinga  feld.  The  name 
of  Rcedel  is  on  record ;  hence  we  may  explain  it  as  '  the  field 
of  the  P^delings,'  or  Rc'edlings ;  or  '  the  field  of  the  sons  (or 
tribe)  of  Rtedel.' 

It  is  not  impossible  that  contraction  has  taken  place,  and 
that  the  original  form  was  Riedwulfinga  feld,  or  'the  field  of 
the  R^dwulfings.' 

Ringsfield.  To  the  S.  of  Beccles.  D.B.  has  Ringesfella, 
p.  4.  The  prefix  Hring-  occurs  in  several  A.S.  names,  though 
not  found  alone.  But  we  may  take  it  to  be  Norse.  Rygh 
says  that  the  Norse  Ringr,  originally  Hringr,  was  a  personal 
name,  and  is  preserved  in  a  large  number  of  place-names. 
I  explain  this  as  '  Hring's  field,'  where  Hring  represents  O.N. 
Hringr,  so  that  it  is  really  '  Hringr's  field.'  The  final  -r  is 
merely  the  suffix  of  the  nom.  case,  and  answers  to  the  -us  in 
L.  Marc-US.     The  prefix  Rings-  occurs  also  in  Ringshall. 

Shadingfield.  Spelt  Shaddingfield  by  Kirby.  Between 
Beccles  and  Blythburgh.  The  g  is  modern,  and  a  better 
spelling  would  be  Shadenfield.  Spelt  Shadenfeld,  H.R. ; 
Shadnefeud,  Ipm.;  Scadenafella,  D.B.,  p.  16.  Here  the  -ena 
is  the  mark  of  the  gen.  pi.  of  a  weak  noun,  and  the  nom.  sing, 
would  be  *scada,  or  A.S.  *sceada.  This  exact  form  is  not 
found ;  but  it  evidently  resulted  from  the  form  sceatha,  by  the 
substitution  of  Norman  d  for  the  voiced  tli;  the  gen.  pi.  was 


NAMES   ENDING   IN   -FIELD  29 

sceathena.  Though  sceatha  is  literally  '  one  who  does  scathe 
or  damage,'  it  is  a  fairly  common  word  for  robber  or  thief; 
and  the  sense  is  '  field  of  thieves.'  It  might  even  mean  '  field 
of  pirates,'  as  the  compound  wicing- sceatha  was  used  in  that 
particular  sense.  How  the  field  acquired  its  name,  we  have 
no  means  of  knowing,  though  it  would  probably  be  an  inte- 
resting story,  if  it  could  be  recovered.  As  the  a  was  originally 
short,  the  spelling  Shaddingfield  can  be  justified. 

Stanningfield.  To  the  S.  of  Bury.  Spelt  Stanefeld, 
H.R.,  Ipm. ;  Stanfella,  D.B.,  p.  21.  It  is  clear  that  the  -ing 
is  comparatively  modern.  The  original  was  probably  stdnen 
feld,  i.e.  'stony  field.'  The  usual  adj.  is  stamen  (with  a?),  but 
stdnen  also  occurs ;  as  in  the  dat.  stdnenan  bricge ;  Birch,  C.S. 
iii.  113,  1.  24. 

Stansfield.  To  the  N.  of  Clare,  at  some  distance.  Spelt 
Stanesfeld,  H.R. ;  Ipm.;  Stanefeld,  R.B. ;  Stanesfelda,  D.B., 
p.  182.  Stdnes  is  the  gen.  case  of  Stan,  which  is  here  a  man's 
name,,  as  in  Stansfield  in  the  W.  Riding.  The  sense  is  '  Stan's 
field.'     Stone  is  now  used  as  a  surname. 

Sternfield.  Near  Saxmundham.  Spelt  Sternfeld,  Ipm. ; 
Sternefella,  D.B.,  p.  72.  But  an  es  has  been  lost,  in  a  difficult 
position  between  rn  and  /;  hence  we  also  find  Sternesfella, 
D.B.,  p.  71 ;  Sternesfelda,  D.B.,  pp.  33,  128.  The  apparent 
meaning  is  'Stern's  field.'  This  personal  name  is  not  otherwise 
recorded;  but  cf.  A.S.  styrne,  E.  stern,  adj.  'severe.' 

Waldingfield.  Spelt  TfaWi/i^e/e/c?,  H.R.;  Ipm.;  Waldinge- 
felda,  D.B. ;  p.  159.  A.S.  Wealdingafeld;  Birch,  C.S.  iii.  603; 
O.  Merc.  Waldingafeld.  Meaning :  '  field  of  the  Waldings,'  or 
'  of  the  sons  of  Walda.' 

Waldringfield.  Spelt  Wandringfeld,  misprint  for  Waud- 
ringfeld  (with  u  for  I),  Ipm. ;  Waldringafelda,  D.B.,  pp.  69, 
178.  Meaning:  'field  of  the  Wald(he)rings,'  or  'sons  of 
Waldhere.'     Waldhere  is  a  known  name. 

Wattisfield.  Spelt  Watesfelda,  D.B.,  pp.  37,  100  ;  but 
Watlesfelda,  D,B,,  p.   170.     Copinger   also   gives    Watlesfeld, 


30  THE    PLACE-NAMES   OF    SUFFOLK 

Watelesfeld,  Wattelesfeld.  A.S.  form  *  Wcetles  feld,  where 
Waetles  is  the  gen.  of  *Wa3tel,  the  strong  form  allied  to 
Wffitela,  whence  Wretling  and  Wsetlinga  stn'et  or  Watling 
Street.  Sense :  '  the  field  of  Waetel.'  Though  Wsetel  is  not 
precisely  recorded,  it  is  a  correct  form,  and  is  also  the  obvious 
origin  of  Wateles-tone  in  Ipm.  p.  113. 

Westerfield.  Spelt  Westerfeld,  Ipm.,  p.  97  ;  Westrefelda, 
D.B.,  pp.  28,  29.  Meaning:  'field  more  to  the  west';  cf. 
Icel.  vestari,  vestri,  more  to  the  west.  It  is  more  to  the  west 
than  Bealings  to  any  one  coming  from  Woodbridge  and  the 
river  Deben. 

Whatfield.  Spelt  Whatefeldjpm.;  Quatefeld,U.U{\\ithqu 
for  tvh);  Gawatfelda,  D.B.,  p.  23.  A.S.  form  Hwaite-feld; 
meaning  'wheat-field.'  Though  A.S.  has  not  this  precise  com- 
pound on  record,  we  find  hwwteland,  'wheat-land,'  and  feld  as  a 
suffix.  Whatfield  was  sometimes  called  Wheatfield  (see  The 
Beauties  of  England).  "  This  Town  is  chiefly  remarkable  for 
growing   the   most   excellent   Seed-Wheat";    Kirby. 

WiXGFiELD.  Spelt  ■  Wyngefeld,  H.R. ;  also  Wingefeld  in 
Thurkytel's  will,  in  Thorpe,  Diplomat,  p.  580;  a  rather  late 
document.  Winge  represents  an  earlier  Wingan,  as  in 
Wingan-hdm,  in  Kemble's  index;  from  the  nom.  Winga. 
Meaning:  '  Winga's  field.'  (Distinct  from  Wingfield,  Beds., 
which  was  originally  Winanfeld,  i.e.  'Wina's  field.') 

WiTHERSFiELD.     Spelt  Wetheresfeld,  Ipm.,  T.N.;   Wytheres- 
feld,  H.R.;   Wedresfelda,  D.B.,  p.  233  (with  d  for  th).     Literally 
'wether's    field';    cf.    WiTHERSDALE    (above).     It    is    probable., 
that  Wether  was  a  personal  name. 

18.     Fleet. 

Fleet,  A.S.  fleot,  not  only  means  an  estuary  or  shallow 
channel,  but  also  a  shallow  stream,  or  even  a  drain  or  ditch; 
see  the  E.D.D.     It  only  occurs  in  Herringfleet. 

Herringfleet.  As  Herringfleet  is  some  four  miles  inland, 
it  has  notliing  to  do  with  herrings,  though  herring  is  a  very 


NAMES    ENDING    IN    -FOIJD  31 

familiar  word  in  that  district.  How  old  the  prefix  Herring 
may  be  in  this  case,  I  do  not  exactly  know  ;  but  in  the  four- 
teenth century,  at  any  rate,  the  form  was  Herling.  Old 
spellings  are:  Heiiingflet,  Ipm.  p.  223  (as  late  as  130(i);  H.R.; 
Herlingafiet,  D.B.,  p.  8.  We  find  Herlinga-lidm  in  Thorpe, 
Diplomat.,  p.  563,  a.d.  1046.  The  sense  is:  'fleet  (or  shallow 
stream)  of  the  Herlings,'  or  'of  the  sons  of  Herla.'  Herie 
occurs  in  A.S.  as  a  prefix  in  several  personal  names. 

19.     Ford. 

This  well-known  suffix  occurs  in  Battisford,  Blythford, 
Boxford,  Bramford,  Brockford,  Carlford,  Chillesford,  Cosford, 
Cransford,  Culford,  Glemsford,  Kentford,  Lackford,  Marlesford, 
Melford,  Mutford,  Orford,  Playford,  Poslingford,  Samford, 
Stratford,  Thetford  (mostly  in  Norfolk),  Ufford,  Wangford, 
Wilford,  and  Yoxford. 

Fords  were  once  notable  places,  and  it  is  remarkable  how 
many  of  them  gave  names  to  the  hundreds  into  which  the 
county  was  divided ;  as  will  be  noted  below. 

Battlsford.  Old  spellings;  Batesford,  T.N.;  Batisforde, 
Ipm.,  p.  2.58 ;  Battisforde,  Ipm.,  p.  48 ;  Betesfort,  D.B., 
p.  259.  If  we  could  depend  upon  the  first  vowel  in  the  last 
form,  this  would  be  the  same  name  as  Bettesford,  which  occurs 
in  Birch,  C.S.  iii.  585,  1.  13.  But  it  is  better  to  assume  a  form 
Battesford,  which  would  mean  '  Batt's  ford.'  The  name  Batt 
has  not  been  hitherto  noted,  but  the  allied  weak  form  Bata 
occurs  in  Batancumb,  in  Kemble's  index.     Cf.  Batsford,  Glouc. 

Blythford  or  Blyford.  Spelt  Blideforda,  D.B.,  p.  150. 
The  usual  modern  Blyford  is  due  to  a  Norman  pronunciation, 
and  is  of  rather  early  date.  Compare  Blyford  in  Ipm. ;  which 
probably  refers  to  Blythford,  though  said  to  be  in  Norfolk. 
The  sense  is,  of  course,  '  ford  through  the  river  Blithe.'  In  the 
Will  of  Eadvvine,  dated  1060,  but  extant  in  a  late  and  ill-spelt 
copy,  occurs  the  strange  form  Blitleford,  which  is  probably  an 
error  for  Blitheford ;  see  Thorpe,  Diplomat,  p.  590,  I.  15. 


32  THE   PLACE-NAMES   OF   SUFFOLK 

BoxFORD.  Spelt  Boxford,  H.R.;  Ipm.  The  sense  is  'ford 
near  the  box-tree,'  Fords  are  often  named  from  trees  that 
serve  to  mark  the  spot.  Compare  Boxford  in  Berks.  The 
river  Box  is  named  from  Boxford. 

Bramford.  Spelt  Bramford,  Ipm. ;  Brunfort,  D.B.,  p.  2  ; 
Branfort,  D.B.,  p.  17  (both  incorrect  forms).  Copinger  also 
gives  the  spellings  Braiunforde  and  Bromford. 

Bram  probably  represents  the  prov.  E.  brame,  a  blackberry, 
though  the  name  has  not  been  recorded  earlier  than  1425 
(see  N.E.D.  or  E.D.D.);  but  the  dimin.  hrmmel  occurs  in  A.S., 
and  is  now  the  bramble.  I  suppose  that  the  name  meant  'ford 
near  the  bramble.'  Our  bramble  is  etymologically  connected 
with  broom,  which  may  account  for  the  spelling  Bromford 
(above)  just  as  Bromfield  means  'broom  field.' 

Brockford.  Spelt  Brocford,  H.R.;  T.N. ;  Brock/art,  D.B., 
p.  161.  The  sense  is  'brook-ford'  or  'ford  through  the  brook.' 
The  long  o  in  the  A.S.  broc,  a  brook,  is  shortened  in  the 
stressed  syllable  before  cf.    Compare  Bromfield  for  'broom-field.' 

Carlford.  The  name  of  a  hundred.  Spelt  Carlesford, 
H.R. ;  but  Carleford  in  D.B.,  p.  4.  Carleford  represents  the 
A.S.  cai^la  ford,  where  carla  is  the  gen.  pi.  of  carl,  a  churl,  a 
rustic,  not  an  English  word,  but  borrowed  from  the  Norse  karl, 
a  churl,  a  peasant.  Carlesford  is  an  alternative  form,  from  the 
gen.  sing,  carles.  The  sense  is  'churls'  ford'  or  'churl's  ford'; 
where  churl  means  a  rustic,  a  peasant. 

Chillesford.  Formerly  CAese(/brc^,  H.R. ;  Gesefortda,T)3., 
p.  32  (with  inserted  t  and  lost  I).  Copinger  also  gives  the 
forms  Chesilford,  Chesi/lford,  Clujselford.  All  from  the  A.S. 
ceosel,  cisel,  M.E.  chisel,  gravel ;  showing  that  Chilles  is  a  mere 
perversion  of  Chisel,  as  in  Chiselhurst  (Kent).  The  sense  is 
'gravel-ford'  or  'gravelly  ford.' 

CoSFORD.  The  name  of  a  hundred.  Formerly  Corsford, 
H.R.;  Corsforth,  Ipm.;  Cursforde,  D.B.,  p.  176.  Cors  was  a 
river-name,  also  spelt  Corsa.  Kemble's  index  has  Cors-broc, 
'  Gors-brook' ;     Gorsa-burne,    '  Corsa-bourne' ;      Corsan-stream, 


NAMES   ENDING    IN    -FORI)  33 

'  Corsa's  stream,'  which  was  also  simply  called  Cursa.  Hence 
also  Corsan-tun,  or  'town  on  the  Corsa';  see  Birch,  C.S.  ii. 
498.  The  sense  is  '  ford  thronorh  the  Cors.'  The  meaning  of 
Cors  is  unknown ;  it  looks  like  the  Welsh  cois,  a  fen ;  cf. 
cor  sen,  a  reed. 

Cransford.  Spelt  Cranesford,l^m.\  but  CVane/orc?«,  D.B., 
p.  35.  The  sense  is  '  crane's  ford.'  Cf  Cranford  (crane  ford), 
Middlesex,  and  such  names  as  Ox-ford,  Swin-ford,  Hors-ford. 
When  an  animal's  name  is  prefixed  to  a  ford,  it  roughly 
indicates  the  depth. 

CuLFORD.  Spelt  Guleforda,  D.B.,  p.  167 ;  Culeforde  (dative) 
in  Birch,  C.S.  iii.  219,  in  a  late  A.S.  charter.  We  can  hardly 
dissociate  this  name  from  the  curious  form  Gulum-ford,  which 
occurs  in  Birch,  C.S.  ii.  432  ;  which  must  be  further  compared 
with  the  expression  seven  lines  below,  viz.  "on  tha  lace  adun 
on  culum ;  up  of  cidum  on  tha  ealdan  lace."  Middendorf  (in 
his  Altenglisches  Flurnamenbuch,  Halle,  1902,  p.  33)  explains 
cule  as  a  weak  fem.,  meaning  'a  hole,  depression,  pit,'  like  the 
E.  Fries,  kide,  Du.  kiiil.  If  this  be  right,  Cideforde  represents 
an  earlier  A.S.  Culanforda,  dat.,  meaning  '  ford  near  the  hole, 
or  pit,  or  hollow.'  And  the  above  passage  means :  "along  the 
stream  down  into  the  pits;  up  out  of  the  pits  to  the  old 
stream."     Culum  is  the  dat.  pi. 

Glemsford.  Spelt  Glemesford,  H.R.;  and  in  Kemble,  Cod. 
Dipl.  iv.  245.  Also  Clamesford  (with  G  for  G),  ]).B.,  p.  203. 
Apparently,  for  A.S.  *Gl£emesford,  or  'ford  of  GlsOm.'  But  no 
such  personal  name  is  on  record.     See  Glemham. 

Kentford.  Spelt  Kenteford,  H.R.;  Kentford,  Ipm.,  p.  314. 
Short  for  Kennetford;  'ford  through  the  river  Kennet.'  Kennet 
is  a  known  river-name;  known  to  be  of  Celtic  origin;  spelt 
Cyneta  in  A.S.;  from  a  Celtic  type  "Cunetio.  Cf.  Kintbury 
in  Berks. 

Lackford.  To  the  N.W.  of  Bury.  "Lackford,  the  f.rd 
over  the  Lark,  just  where  that  parish  [i.e.  Lackford]  joins 
Icklingham   All  Saints";    Raven's  Suffolk;    p.  64.     Lackford 

C.  A.  S.  Octavo  Series.     No.  XLVI.  3 


34  THE   PLACE-NAMES   OF   SUFFOLK 

is  also  the  name  of  a  hundred ;  and  it  is  remarkable  that  this 
hundred  took  its  name  from  the  place,  as  the  latter  is  really 
in  Thingoe  hundred,  and  just  outside  Lackford  hundred 
itself.  Spelt  Lakford,  H.R.;  Lacforda,  D.B.,  p.  45.  From  the 
A.S.  lam,  a  running  stream.  [This  word  is  often  confused 
with  the  L.  lacus,  but  the  A.S.  word  cognate  with  lacus  (and 
not  borrowed  from  it)  is  lagu. ;  and  the  E.  lake  is  merely- 
borrowed  from  the  F.  lac]  But  the  native  word  lacu  is  still 
extant  in  the  prov.  E.  lake,  explained  in  the  E.D.D.  as  meaning 
'a  brook,  rivulet,  or  stream,'  very  common  in  S.W.  dialects. 
There  is,  in  fact,  no  lake  at  Lackford,  but  there  is  a  stream. 
The  name  means  'stream-ford,'  or  'ford  through  the  stream.' 
The  modern  name  of  the  stream  is  the  Lark  ;  and  it  is  much 
to  be  suspected  that  this  singular  name  arose  from  the  M.E. 
lake  (from  lacu),  in  which  the  a  was  pronounced  as  in  father 
and  in  lark  (if  the  r  be  suppressed).  Another  name  for  the 
Lark  is  (or  was)  the  Burn,  which  means  bourn  or  stream,  and 
merely  translates  the  A.S.  lacu.  I  further  suspect  that  the 
Linnet,  which  flows  into  the  Lark  at  Bury,  received  its  name 
from  playful  association  with  that  of  the  larger  river.  Both 
names  are  comic. 

Marlesford.  Near  Wickham  Market.  Spelt  Marleford, 
Ipm.;  Marlesforda,  D.B.,  p.  11;  Merlesford,  D.B.,  pp.  12,  27. 
The  sense  is  '  Maerl's  ford '  or  '  Masrle's  fond.'  The  name  Mserl 
(or  Mgerle)  may  safely  be  said  to  be  of  Norse  origin,  as  it  is 
recorded  in  the  compound  Mserle-swegen,  of  which  Searle  gives 
four  examples ;  and  see  Mserleswegen  in  Bjorkman.  The 
suffix  swegen  is  certainly  Norse,  as  it  is  an  A.S.  spelling  of 
O.  Norse  sveinn  (E.  swain). 

Melford,  often  Long  Melford.  Spelt  Meleford,  H.R. ; 
Ipm. ;  Melaforda,  D.B.,  p.  157.  There  are  several  Milfords  in 
other  counties;  but  connexion  with  these  is  doubtful,  as  'mill' 
would  hardly  appear  as  Mela  in  D.B.  The  prefixes  Mela-, 
Mele-  rather  suggest  connexion  with  the  A.S.  Mcelan,  gen.  of 
Mcela,  as  in  Mcelan  heorh,  in  Birch,  C.S.  ii.  291,  1.  3.  If  this 
be  right,  the  sense  is  '  Masla's  ford ' ;    but  it  is  only  a  guess. 


NAMES    ENDING    IN    -FORD  35 

The   Mel-  in  Melton  and  Mellis  appears  to  be  quite  distinct 
from  that  in  Melford. 

MuTFORD.  This  is  the  name  of  a  hundred  as  well  as  of  a 
village.  Spelt  Mutford,  Mutteford,  H.R. ;  Mutford  and  Mntfonla, 
D.B.,  p.  5;  but  Muthford  in  Ipm.,  p.  2G  (a.d.  1263),  which  is  an 
important  variant,  and  may  be  taken  to  represent  an  older  form. 
It  is  not  easy  to  find  the  ford  referred  to,  but  I  think  it  must 
be  near  Mutford  Hall,  where  a  stream  flows  into  the  Hundred 
River,  if  I  rightly  understand  the  ordnance  map.  The  name 
may  signify  as  much,  if  we  connect  it  with  the  A.S.  niutha, 
the  mouth  of  a  river,  the  place  where  one  river  meets  another. 
If  this  be  right,  the  sense  is  '  ford  near  the  junction  of  the 
streams.'  A  similar  explanation  is  given  of  Mitton  in  Worcester- 
shire ;  viz.  from  the  A.S.  variant  mythe,  with  the  same  sense. 
And  the  compound  mythford  is  found  in  Birch,  C.S.  ii.  481, 
1.  21. 

Orford.  This  also  is  the  name  of  a  former  hundred, 
though  no  longer  in  use  for  that  purpose.  The  place  is  on 
the  N.W.  bank  of  the  river  Aide.  Formerly  Oreford,  H.R., 
T.N.,  Ipm.  From  the  A.S.  ora,  a  border,  edge,  bank;  meaning 
'ford  at  the  bank  or  shore.'  Perhaps  it  was  where  the  ordnance 
map  marks  the  Quay  and  the  Ferry.  On  one  side  of  the  Aide 
is  the  mainland  ;  on  the  other,  the  King's  Marshes.  I  find, 
in  the  Victoria  Hist,  of  Suffolk,  i.  57,  that  the  river  Aide  was 
called  the  Ore  below  Aldeburgh ;  and  at  p.  29,  I  find  "  the 
Ore  or  Aide."  This  at  once  suggests  that  Orford  was  taken 
to  mean  the  'ford  through  the  Ore';  but  that  this  is  the  true 
origin  may  be  doubted.  It  is  much  more  likely  that  the  lower 
part  of  the  Aide  was  called  the  Ore  because  Orford  suggested 
such  alteration.  The  fact  remains,  that  the  A.S.  ora  does  not 
mean  a  river,  but  a  river-bank  or  a  brink  of  any  kind. 

Playford.  N.E.  of  Ipswich,  on  the  river  Finn.  Spelt 
Playford,  Ipm.;  H.R.;  Plegeforda,  D.B.,  p.  68.  From  the  A.S. 
plega,  'play,'  with,  in  poetry,  the  occasional  sense  of  'battle.' 
It  may  possibly  commemorate  the  scene  of  a  long-forgotten 
encounter. 

POSLINGFORD.     Copinger  has  collected  22  old  forms  of  this 

3—2 


36  THE    PLACE-NAMES    OF    SUFFOLK 

name,  of  which  only  5  end  m.  ford;  the  majority  of  17  end  in 
luorth ;  and  it  is  certain  that,  as  in  other  cases,  ford  has  been 
substituted  for  ivorth,  which  meant  '  a  farm '  or  '  a  holding ' ; 
see  the  names  under  Worth.  Both  Duxford  and  Pampisford, 
in  Cambs.,  have  suffered  the  same  alteration.  The  old  spellings 
most  worth  notice  are:  PoselingiU7'th,]i.R.,T.'S.;  Poselingetvrth, 
'T.N.;  Poslindewrda,  D.B.,  p.  182;  Poslingewrda,  D.B.  233. 
The  existence  of  the  form  Postlinges  in  R.B.,  and  of  the 
present  Postling  in  Kent,  suggests  that  the  full  form  was 
Postlinga-weorth,  i.e.  the  '  farm  (or  holding)  of  the  Postlings,' 
a  tribe  or  family  otherwise  unknown. 

Samford.  The  name  of  an  old  hundred,  and  still  in  use. 
Spelt  Samford,  Sandford,  H.R. ;  Sanforde,  Sampforde,  R.B. ; 
Sanfort,  D.B.,  p.  12.  Evidently  for  '  sand-ford,'  or  ford  Avith  a 
sandy  bottom.     Cf.  Sandford  in  Oxfordshire. 

Stratford.  Spelt  Strafort,  D.B.,  p.  56  ;  Stratfort,  D.B., 
p.  243.  Like  other  Stratfords,  it  means  a  place  where  '  a  street ' 
or  old  road  is  continued  beyond  a  stream.  The  pi'esent  road 
from  Marlesford  to  Farnham  crosses  the  river  Aide  near  Strat- 
ford St  Andrew.  And  the  road  from  Colchester  to  Ipswich 
crosses  the  Stour  near  Stratford  St  Mary. 

Thetford.  Nearly  all  of  this  town  is  in  Norfolk,  but  it 
is  just  on  the  border  of  the  county,  and  is  worth  notice.  It  is 
situate  on  the  Little  Ouse,  but  a  smaller  river  here  joins  the 
other,  upon  which  modern  ingenuity  has  bestowed  the  name 
of  Thet !  Here,  as  in  other  cases,  the  river  is  named  from  the 
town,  and  not  othei'wise.  For  the  ingenious  people  who 
devised  this  name  evidently  did  not  know  that  the  old  name 
was  really  Thedford,  or  more  strictly  Theedford  ;  for  the  e  was 
once  long.  Indeed,  the  spelling  Theedford  occurs  in  the 
Liber  de  Hyda,  p.  10.  It  is  spelt  Theodford  and  Theotford 
in  the  A.S.  Chronicle  ;  the  more  correct  spelling  Theodford 
appearing  in  the  Laud  MS.,  under  the  dates  870, 1004,  and  1010, 
and  in  the  early  Parker  MS.  under  870.  The  prefix  Theod 
means  nation,  peof)le,  race ;  also,  people  in  general ;  and  in 
composition  it  has  the  sense  of  general,  popular.     Hence  the 


NAMES    ENDING    IN    -FORD  ^7 

sense  is  'popular  ford,'  or  ford  in  frequent  use,  on  the  road 
from  Bury  northwards.  Isaac  Taylor  strangely  denounces  this 
explanation,  but  answers  his  own  objection  by  saying  that  the 
German  Dietfurt  means  'ford  of  the  people.'  There  is  another 
Thetford  in  Cambridgeshire  which  has  the  misfortune  of  not 
possessing  any  Thet  to  derive  itself  from. 

Ufford.  Spelt  Ufford,  H.R.,  Ipm.,  R.B.;  Ufforda,  D.B.,  p.  89. 
Uf-  is  short  for  Uffan,  as  in  Uffan-lege  (dative)  in  Birch,  C.S. 
ii.  175,  last  line ;  and  Ul^an  is  the  gen.  of  Uffa.  The  sense  is 
'  UfFa's  ford.'  Uffa  is  a  known  name,  and  distinct  from  the 
commoner  Offa. 

Wangford.  In  Blything  hundred.  Yet  there  was  also  a 
Wangford  hundred,  by  confusion  of  two  or  three  distinct 
names,  as  will  be  shown.  And  first,  as  to  the  place-name. 
There  are  really  two  such  place-names  ;  for  Kelly  says  there 
is  a  Wangford  near  Southwold,  and  another  near  Brandon. 
This  makes  three  Wangfords ;  and  they  seem  to  be  all  of 
different  origin. 

(1)  Waugford  in  Blything  hundred ;  to  the  N.W.  of 
Southwold.  Cf.  Wangford,  H.R. ;  Wangeford,  Ipm.  Spelt 
Wanheforda  in  D.B.,  p.  268,  where  it  is  associated  with 
Frostenden.  As  nh  occurs  in  Norman  for  the  A.S.  ng,  the 
A.S.  name  must  have  been  Wang-ford,  just  as  it  is  now.  The 
sense  is  obvious,  when  it  is  remembered  that  both  the  A.S. 
wang  and  the  modern  prov.  E.  wang  (also  wong)  mean  a  flat 
field.  The  sense  is  'the  ford  near  the  flat  field.'  "Waugford 
Green  was  all  open  common  till  1817";  Raven,  Hist.  Suffolk, 
p.  31. 

(2)  Wangford  in  Lackford  hundred ;  to  the  S.W.  of 
Brandon.  Probably  so  called  by  confusion  with  the  former ; 
but  really  for  Wainford.  I  find  Waynefoj^d  (Suffolk)  in  T.N. ; 
and  Copinger  notes  such  spellings  as  Waynford  and  Wainford, 
though  these  may  refer  to  the  hundred.  However  this  may 
be,  I  find  in  D.B.,  p.  156,  a  name  which  I  read  as  Wainforda, 
though  in  the  Victoria  County  History,  at  p.  494,  it  is  printed 
Wamford,   and    explained    as   Wangford ;    being    certainly    in 


38  THE   PLACE-NAMES   OF   SUFFOLK 

Blything  Hundred.  I  prefer  my  own  reading,  because  Wamford 
is  nonsense,  and  we  cannot  fairly  connect  such  a  form  with 
Wangford  no.  1  (above).  The  mistake  of  writing  m  for  in  is 
common,  if  a  mistake  it  be.  I  also  find  Wamford  in  R.B., 
but  suspect  that  also  to  be  wrong,  as  there  certainly  was  a 
Wainford  somewhere,  and  we  have  not  yet  come  to  the  hundred, 
which  had  no  more  claim  to  be  called  Wainford  than  it  had 
to  be  called  Wangford.  I  assume  then  that  Wainford  is  here 
the  right  form ;  and  the  sense  is  obvious,  viz.  '  wain-ford,'  or  a 
ford  through  which  a  wain  could  pass,  as  being  but  shallow. 
The  A.S.  form  would  be  lucegn-ford.  There  is  a  Wainfleet  in 
Lines.,  and  fleet  means  a  shallow  stream. 

(3)  Wangford  hundred.  The  original  name  was  neither 
Wangford  nor  Wainford,  but  occurs  in  D.B.  in  another  form. 
Thus,  at  p.  4,  it  is  Wanneforda,  and  so  again  at  pp.  15,  109, 
178 ;  but  Waineforda  at  pp.  35,  94 ;  and  Wenefort  at  p.  39. 
Waineforda  and  Wenefort  may  have  been  due  to  confusion 
with  Wainford  ;  but  the  prevalent  form  Wanneford  requires 
an  explanation  for  which  neither  Wangford  (Norman  Wankford) 
nor  Wainford  will  suffice.  I  am  inclined  to  accept  the  guess 
made  in  Raven's  Hist,  of  Suffolk,  at  p.  3.  He  says  that  the 
Waveney  was  also  called  Wanney  (which  is  likely),  and  suggests 
that  "Wainford"  here  means  "Wanneford."  Of  course  the 
suggestion,  as  so  presented,  is  impossible  ;  no  one  ever  heard 
of  a  ivain  being  called  a  wanney.  But  '  Wanney-ford '  may 
very  well  suggest  an  origin  for  the  form  Wanneford  in  D.B. 
As  thus  presented,  the  guess  seems  reasonable. 

I  therefore  interpret  the  available  evidence  as  showing 
that  Wangford  near  Southwold  was  always  so  called,  and 
meant  '  ford  near  a  wang';  and  it  was  perhaps  the  oldest  name 
of  the  three.  Secondly,  that  Wangford  near  Brandon  was  at 
first  called  Wainford,  or  'ford  for  a  wain.'  Thirdly,  that 
Wanneford  hundred  alluded  to  a  ford  across  the  Waveney, 
which  forms  its  northern  boundary. 

WiLFORD.  The  name  of  a  hundred.  There  is  no  place 
with  this  name  in  Suffolk,  though  there  is  one  in  Notts.  Spelt 
Wyleford,  Willeford,  H.R.;  Wileford,  D.B.,  pp.  76, 186.     In  the 


NAMES   ENDING   IN   -FORD,   -GATE  .SO 

Crawford  Charters,  ed.  Napier  and  Stevenson,  at  p.  33,  1.  3, 
we  find:  "  de  wileford";  and  the  Index  says  that  it  means 
"  Wilford,  CO.  Suffolk";  without  indicating  its  exact  locality. 
If  we  can  rely  upon  the  form  Wileford,  in  which  nearly  all 
the  authorities  agree,  perhaps  we  ma}'  explain  it  as  '  Wili's 
ford';  see  Wilby.  The  sense  'willow-ford'  is  not  impossible, 
but  is  less  likely. 

YoxFORD.  Spelt  Yokisford,  Ipm. ;  loxford,  lokesford, 
H.R.;  Gokesford,  D.B.,  p.  298;  lokesford,  D.B.,  p.  105.  "  The 
sense  is  '  Yoke's  ford,'  where  Yoke  is  used,  apparently,  as  a 
man's  name  or  nickname.  In  the  A.S.  and  Northumbrian 
version  of  the  Gospels,  the  Lindisfarne  MS.  (Northumbrian) 
translates  coniugem  in  Matt.  i.  20  by  gehede  vel  geoc,  lit. 
'bedfellow  or  yoke';  showing  that  geoc  could  have  the  sense 
of  '  spouse.'  The  river  at  Yoxford  is  called  the  Yox ;  but  the 
above  spellings  contradict  the  antiquity  of  its  name.  Some 
further  light  comes  from  comparing  it  with  Yoxall  in  Staffs., 
former  Yokes-hale,  where  Yokes  can  only  be  a  gen.  singular. 
At  any  rate,  this  second  example  shows  that  the  form  Yox  is 
modei-n ;  and  that  the  river-name  is  of  no  value.  Yoxford 
means  'Yoke's  ford';  and  it  is  only  the  interpretation  of  Yoke 
that  is  doubtful.  Another  sense  of  yoke  is  pointed  out  by 
Duignan,  who  refers  us  to  Birch,  C.S.  i.  584,  1.  2,  where  xvi 
give  londes  means  'sixteen  yokes  of  land,'  showing  that  'a  yoke 
of  land'  denoted  a  definite  portion  of  land ;  whence  the  use  of 
prov.  E.  yoke  to  mean  '  a  small  farm.'  If  a  small  farm  bad 
acquired  the  local  name  of  Yoke,  we  might  explain  '  Yoke's 
ford'  as  meaning  a  ford  in  its  immediate  neighbourhood,  or 
one  used  by  the  farm-servants.  This  is  again  one  of  the  cases 
in  which  the  literal  sense  is  obvious,  but  the  exact  inter- 
pretation is  unattainable,  because  we  oannot  tell  to  what  the 
name  refers. 

20.    Gate. 

There  are  two  distinct  words  of  this  form,  viz.  gate,  a  street, 
from  the  O.  Norse  gata,  a  street,  road,  way,  and  gate,  a  movable 
barrier,  answering  to  the  A.S.  geat.     The  suffix  is  only  found 


40  THE    PLACE-NAMES   OF   SUFFOLK 

in  Burgate,  Lidgate,  and  Plomesgate,  which  will  be  considered 
together. 

Burgate.  Spelt  Burgata,  D.B.,  pp.  276,  277,  A.S. 
Burhgat,  occurring  in  the  dat.  case  as  burhgate ;  Birch,  C.S. 
i.  8,  1.  4  from  bottom;  variant  oihurhgeat,  a  borough-gate.  Of 
course  the  exact  reason  for  the  name  is  lost. 

Lidgate.  The  birthplace  of  Lydgate  the  poet.  Spelt 
Lidgate,  H.R.;  Lydegate,  Litgate,  Ipm.;  Litgata,  D.B.,  p.  809. 
For  A.S.  hlidgeat,  explained  as  'a  swing-gate'  in  the  Diet., 
but  I  suspect  that  it  was  rather  a  clapper-gate,  i.e.  an  old- 
fashioned  kind  of  stile,  one  end  of  which  falls  when  pressed 
down,  but  rises  again  when  the  pressure  is  removed  ;  cf.  A.S. 
Idid,  a  lid  of  a  box.  The  dat.  Jdidgeate  occurs  in  Birch,  A.S. 
ii.  284;  spelt  hlidgate  in  the  same,  p.  164. 

Plomesgate.  The  name  of  a  hundred.  Spelt  Plumesgate, 
H.R. ;  Plumesgata,  D.B.,  p.  32.  The  exact  origin  of  the  name 
is  necessarily  lost.  It  is  remarkable  that  the  A.S.  plume,  a 
plum,  was  feminine,  with  a  genitive  in  -an  ;  but  here  we  have 
to  deal  with  a  masc.  sb.  *Pluvi,  gen.  *  Plumes ;  whence  the 
sense  '  Plum's  gate.'  The  plum  is  referred  to  in  the  place- 
names  Plumstead,  Plumpton,  and  Plumtree. 


21.    Grave. 

The  A.S.  grcsf,  a  grave,  also  means  a  ditch,  a  trench,  a 
cutting  or  entrenchment.  This  suffix  occurs  in  Gedgrave, 
Hargrave,  Hengrave,  Kesgrave,  Palgrave,  and  Redgrave. 

Gedgrave.  The  parish  (says  Kelly)  is  a  new  one. 
Gedgrave  Hall  lies  to  the  S.W.  of  Orford,  and  near  it  are  the 
extensive  Gedgrave  Marshes.  Spelt  Gategrave,  H.R.;  Gatagraua, 
D.B,  p.  27;  Gategraua,  D.B.,  p.  93.  Copiuger  also  gives  the 
spellings  Gadegrave,  Gadgrave,  and  Gedgrave.  The  original 
prefix  was  obviously  Gata-,  as  in  Gata-ford,  Gata-tun,  Gata-wic; 
all  in  Kemble's  index.  As  Gatton  is  in  Surrey,  it  is  unlikely  that 
Gata-  is  of  Norse  origin.     It  would  therefore  seem  to  be  the 


NAMES   ENDING    IN    -GRAVE  41 

A.S.  gdta,  gen.  pi.  of  gdt,  a  goat.  The  literal  sense  is  '  burial- 
place  of  goats ' ;  though  grave  might  merely  mean  trench,  or 
even  enclosure.  It  is  obviously  impossible  to  learn  the  cir- 
cumstances of  the  case. 

Hargrave.  Spelt  Hai'egrave,  Ipm. ;  Haragraua,  D.B., 
p.  300.  The  A.S.  form  is  Haraii-grafa ;  in  Birch,  C.S.  iii.  492, 
1.  15.  Here  grafa  is  a  weak  sb.,  closely  related  to  grcef,  a 
grave,  and  no  doubt  had  the  same  sense  of  trench.  Haran 
is  the  gen.  of  hara,  a  hare.  The  sense  is  '  hare's  trench ' ; 
or  '  Hare's  grave,'  taking  Hare  as  a  man's  name.  We  cannot 
tell. 

Hengrave.  Spelt  Hemgrave,  Ipm. ;  Hemegrave,  H.R. ; 
Hemegretha,  D.B.,  p.  1.54.  The  last  spelling  seems  to  be  due 
to  some  mistake.  The  exact  form  of  the  prefix  is  not  recorded; 
but  it  may  have  been  *H8ema.  The  word  is  not  in  the 
dictionaries,  but  is  found  as  a  suffix  in  some  place-names ;  and 
it  is  a  derivative  of  ham,  a  home.  Thus  Kemble's  index  has 
Niwen-htema  gemero,  or  '  the  boundaries  of  the  dwellers  in 
Newnham ' ;  literally,  'the  boundaries  of  the  new-homers.' 
Hence  a  possible  sense  is  'grave  of  the  dwellers  in  the  home'; 
possibly  '  a  family  burial-place.'  We  cannot  ascertain  the 
circumstances. 

Kesgrave.  To  the  E.  of  Ipswich.  Copinger  records  the 
early  forms  Kessegrave  and  Kekesgrave,  of  which  the  latter 
must  be  the  older.  Prof.  Moorman  shows  that  Kexmoor  in 
the  W.  Riding  was  originally  spelt  Ketelsmore,  and  Kex  is 
the  natural  contraction  of  Kekes.  The  name  Ketel  is  Norse, 
originally  spelt  Ketill;  and  the  occurrence  of  the  hard  K  before 
the  e  is  in  itself  an  indication  of  Norse  origin ;  since  the  A.S. 
Ce  became  Clie.  The  name  has  been  much  corrupted,  probably 
because  the  prefix  was  un-English.  The  succesive  changes 
must  have  been  from  KetilLs  or  Ketels  to  Ketes ;  then  to 
Kekes,  Kex.'and  Kes.  The  original  sense  was  probably  'Ketill's 
grave.' 

Palgrave.  Spelt  Palegraue,  H.R.;  Palegraua,  D.B.,  p.  161; 
Palegrave  (in  the  dat.  case),  in  Birch,  C.S.  iii.  314;  in  a  grant 


4'2  THE    PLACE-NAMES   OF   SUFFOLK 

dated  962.  Perhaps  from  the  A.S.  pal,  a  pole,  a  pale,  a  stake; 
a  word  borrowed  from  the  L.  pdliis.  If  so,  the  sense  may  be 
'grave  enclosed  with  palings.' 

Redgrave,  Spelt  Redgrave,  H.R.  When  we  compare  it 
with  Redditch  in  Worcs.,  there  seems  to  be  no  reason  why  it 
may  not  have  meant  '  red  trench,'  or  a  trench  cut  through  red 
soil. 

22.    Hale,  Hall. 

The  suffix  hale  is  of  much  importance,  as  it  is  in  common 
use  in  many  counties,  and  frequently  appears  in  disguised 
spelling,  usually  assuming  the  form  of  liall. 

It  has  become  -all  in  Aspall,  and  -ale  in  Kelsale ;  and  has 
been  changed  into  -hall  in  Benhall,  Blaxhall,  Buxhall,  Ilkets- 
hall,  Knettishall,  Knodishall,  Mildenhall,Peasenhall,RickiDghall, 
Ringshall,  Spexhall,  Uggeshall,  and  Westhall.  The  suffix  in 
Foxhall  was  originally  -hole,  while  that  in  Lawshall  and  Stradis- 
hall  was  -sele.  But  all  the  words  that  now  end  in  -hall,  -all,  or 
-ale,  will  be  taken  together,  for  convenience.  Not  one  of  them 
originally  ended  in  -hall ;  whereas  fifteen  of  them  once  ended 
in  -hale. 

The  0.  Merc,  hale,  A.S.  heale,  only  appears  in  the  dative 
case  ;  the  nom.  ended  in  It,  the  O.  Merc,  form  being  halh,  and 
the  A.S.  healh.  Halh  has  given  us  the  modern  haugh,  which 
is  explained  in  the  E.D.D.  as  meaning  '  low-lying,  level  ground 
by  the  side  of  a  river';  while  the  pro  v.  E.  hale  (from  the  above 
dative  case)  is  similarly  defined  as  'a  piece  of  flat  alluvial  land 
by  the  side  of  a  river.'  The  old  sense  of  halh  or  healh  seems 
to  have  been  a  corner,  nook,  or  sheltered  place ;  it  seems  safe 
to  define  it  as  'a  sheltered  spot,  beside  a  river';  pei'haps  we 
may  call  it  'a  nook'  for  the  sake  of  brevity. 

Aspall.  Spelt  Aspenhalle,  Ipm. ;  Espala,  D.B.,  p.  196 ; 
Aspella,  D.B.,  p.  275;  Aspala,  D.B.,  p.  339.  Copinger  gives 
many  spellings,  of  which  the  best  are  Aspale,  Asphale,  Asphal. 
The  etymology  is  evidently  from  the  A.S.  cesp,  an  asp-tree  or 
aspen-tree,  and  liale,  as  explained  above.  The  variant  Aspenhalle 


NAMES    ENDING    IN    -HATJO,    -HALL  43 

is  valuable,  as  giving  aspen,  which  is  really  an  adjectival  form, 
made  by  adding  -en  (as  in  gold-en,  tuood-en)  to  the  A.S.  cesp. 
The  sense  is  'aspen-nook.' 

Benhall.  Spelt  Benhall,  Ipm.,  p.  161 ;  but  BenJiale  earlier, 
Ipm.,  p.  121.  D.B.  has  Benhala,  pp.  57,  128;  Benehcda,  pp.  57, 
130;  Benenhala,  pp.56,  130;  Benehalla,^.  34.  The  right  form, 
amongst  these,  is  Benenhala ;  where  Benen  represents  A.S. 
Beonan,  gen.  of  Beona ;  a  personal  name  occurring  in  Beonan- 
feld,  in  Kemble's  index.     The  sense  is  '  Beona's  nook.' 

Blaxhall.  Spelt  Blakeshal,  H.R. ;  Ipm. ;  D.B.  has  Blaches- 
sala,  pp.  31,  53 ;  with  che  for  ke,  and  s  wrongly  repeated ; 
Blaccheshala,  D.B.,  p.  53.  Copinger  also  gives  Blacheshala, 
Blakeshale.  The  prefix  represents  A.S.  Bheces,  gen.  of  Blcec, 
lit.  '  black,'  used  as  a  personal  name,  like  Black  at  the  present 
day.     The  sense  is  '  Black's  nook.' 

Buxhall.  Spelt  Bakeshale,  R.B.,  T.N. ;  Buckeshale,  T.N. ; 
Buckeshala,  D.B.,  p.  139.  It  appears  as  Bucyshealm,  in  the 
dat.  case,  in  ^IfiBaed's  Will ;  in  Birch,  CS.  iii.  602.  Bucys  is 
an  inferior  spelling  of  Bucces,  gen.  of  Bucc,  lit.  '  a  buck,'  but 
here  used  as  a  man's  name.     The  sense  is  '  Buck's  nook.' 

Foxhall.  To  the  E.  of  Ipswich,  and  S.  of  Kesgrave.  There 
is  also  a  Foxhall  Hall,  which  is  not  tautological,  since  it  stands 
for  Foxhole  Hall ;  as  old  spellings  show.  Spelt  Foxehole,  H.R. ; 
Foxeliola,  D.B.,  p.  212.  The  form  Foxe-  shows  that  the  prefix 
represents  the  A.S.  gen.  pi.  foxa,  '  of  foxes ' ;  and  the  whole 
word  represents  foxa  holu,  '  holes  of  foxes,'  or  '  foxholes,'  which 
has  been  turned  into  '  foxhole,'  by  neglecting  the  pi.  suffix  -u. 

Ilketshall.  Spelt  Ilketeleshale,  H.R. ;  shortened  to 
Ilketeshale,  Ipm.  D.B.  has  Ilcheteleshala,  p.  40 ;  shortened  to 
Ilcheteshala,  pp.  40,  151 ;  with  che  for  ke.  From  a  Norse  name, 
as  the  use  of  ke  shows.  The  sense  seems  to  be  '  Ilketill's  nook.' 
Ketill  is  very  common  in  Norse  names,  or  in  names  adapted 
from  them ;  but  I  find  no  compound  with  the  prefix  II-.  There 
can  hardly  be  any  doubt  that  Ilketill  is  a  reduced  form  of  the 
known  name  Ulfketill ;  so  that  the  original  sense  was  really 


44  THE    PLACE-NAMES   OF   SUFFOLK 

'  Ulfketill's  nook.'  Indeed,  it  is  likely  enough  that  the 
reference  is  to  a  famous  Ulfeytel  who  was  alderman  of  East 
Anglia,  and  inflicted  a  serious  defeat  upon  the  Danes  in  the 
year  1004;  see  the  A.S.  Chronicle.  Nevertheless,  his  name 
was  of  Norse  origin ;  Ulfcetel  was  a  very  common  Norse  name, 
and  Bjorkman  (p.  169)  gives  the  shortened  form  Ulketel,  for 
which  D.B.  has  substituted  Ilketel,  spelt  Ilchetel  because  D.B. 
has  che  for  ke  regularly. 

Kelsale.  Spelt  Keleshulle,  R.B.  (wrongly) ;  but  Keleshale, 
H.R. ;  Ipm. ;  and  Keleshala,  D.B.,  p.  59.  Copinger  has  many 
other  forms,  giving  the  prefix  as  Gheles  (in  Norman  spelling, 
with  che  for  ke),  Kales,  Kelis,  Keils,  Kels  (very  rarely  with  II)', 
so  that  the  vowel  was  long.  Perhaps  the  prefix  was  Ceoles, 
gen.  of  Ceol,  a  known  name ;  for  though  Ceol  would  normally 
be  palatalised  to  Chele,  this  process  was  sometimes  arrested  by 
Danish  influence,  as  in  the  case  of  Kellington  in  the  West 
Riding,  which  is  from  Ceolinga-tun ;  see  Prof.  Moorman's 
explanation  of  this  name.  The  very  same  thing  seems  to 
have  occurred  again  in  the  case  of  Kelshall  (Herts.),  which 
has  the  same  prefix,  though  the  suffix  -hall  has  there  been 
substituted  for  '  hill ' ;  see  my  Place-names  of  Herts.,  p.  34. 
Thus  the  name  probably  means  '  Ceol's  nook.' 

Knettishall.  Also  sometimes  Knattishall,  as  in  Philips' 
County  Atlas.  Kirby  calls  it  Knattishall  or  Gnattshall.  Spelt 
Gnateshal,  T.N.;  also  Ghenetessala,  D.B.,  p.  81,  with  ss  for  sh; 
Gnedeshalla,  D.B.,  p.  174;  Gnedassala,  D.B.,  336.  The  use  of 
a  or  e  in  the  first  syllable,  and  the  spelling  with  gn,  suggest 
that  the  prefix  was  associated  with  the  A.S.  gncettes,  gen.  of 
gncett  or  gncet,  a  gnat ;  but  it  probably  represents  the  Norse 
name  Knottr  (gen.  Knattar),  given  by  Rygh.  If  so,  the  sense 
is  '  Knottr's  nook.'     The  English  turned  Knattar  into  Knattes. 

Knodishall.  Spelt  Knoteshal,  H.R.;  Cnotesheala,  D.B., 
p.  106;  Chenotessala,  D.B.,  p.  116  (with  Chen  for  Kn,  and  ss 
for  sh).  Also  Knoteshalle,  Ipm.  Apparently  from  a  personal 
name  *Cnot:  but  I  can  only  find  Cnott,  with  a  short  vowel,  as 


NAMES    ENDING    IN    -HALE,    -HALL  45 

in  Gnottis  rode,  in  Kemble,  vi.  217,  1.  10.  Tiiis  would  give  us 
the  sense  of  '  Cnott's  nook,'  or  '  Knott's  nook.'  Prof.  Moorman 
refers  Knottingley  in  the  W.  Riding  to  an  O.  Norse  personal 
name  Knottr ;  but  the  form  given  by  Rygh  is  KnOttr  (gen. 
Knattar) ;  as  in  Knettishall  above. 

Lawshall.  In  this  case,  the  evidence  shows  that  the 
suffix  was  neither  hall  nor  hale,  but  sele.  This  sele  is  the  A.S. 
sele,  m.  (gen.  seles),  also  found  as  seel,  n.  (gen.  swles,  seles),  a 
habitation,  dwelling-place,  house.  The  form  in  D.B.  is  Laives- 
selam,  p.  196.  Copinger  gives  many  other  forms  (without 
noting  the  sources) ;  the  chief  are  Lausel,  Lausele,  Lawcell, 
Lawsele,  Lawsell.  The  word  appears  to  be  a  compound  sb. ; 
and,  as  aw  usually  corresponds  to  an  A.S.  ag,  it  could  easily 
be  derived  from  the  A.S.  lagu,  a  lake,  and  sele,  a  dwelling ; 
meaning  '  a  dwelling-place  near  a  lake.'  If  this  be  so,  the 
situation  of  this  dwelling-place  would  not  be  near  the  present 
Hall,  but  near  the  Hall  in  Chadacre  Park,  about  a  mile  to  the 
S.W.,  where  a  small  lake  is  marked  upon  the  ordnance  map, 
as  being  an  expansion  of  the  Chad  Brook.  See  lay,  sb.  (1), 
by-form  laiu,  a  lake,  a  pool,  in  the  N.E.D. 

Mildenhall.  Spelt  Mildenhal,  H.R.;  miswritten  Mitdene- 
halla,  B.D.,  p.  16.  But  fortunately,  the  true  dat.  form  Milden- 
hale  occurs  in  a  charter  of  Edward  the  Confessor ;  see  Thorpe, 
Diplomat.,  p.  418,  1.  13.  Milden  is  a  late  spelling  of  Mildan, 
gen.  of  Milda,  which  represents  some  name  beginning  with 
Mild,  such  as  Mildred,  which  was  formerly  masculine.  The 
sense  is  '  Milda's  nook.'  For  the  prefix,  cf.  Mildan-hald ;  Birch, 
C.S.  i.  452. 

Peasenhall.  Spelt  Pesenhale,  H.R. ;  Ipm. ;  Pesenhala, 
D.B.,  p.  102 ;  Fisehalla,  D.B.,  p.  64.  The  prefix  is  the  A.S. 
pisena,  gen.  of  pisan,  which  is  the  pi.  of  pisa,  a  pea.  This  pi. 
pisan  became  pesen  in  Mid.  Eng.,  and  peasen  in  the  sixteenth 
century.  The  sense  is  '  peas-nook ' ;  or  a  sheltered  spot  where 
peas  were  grown.  Or  the  prefix  may  represent  pisan-,  com- 
bining form  of  pisa  when  forming  a  compound. 


46  THE    PLACE-NAMES   OF    SUFFOLK 

RiCKiNGHALL.  Spelt  Rykingliole,  Ipm.  D.B.  has  Rikin- 
chala,  p.  95;  Richingehala,  p.  161;  Rikinghala,  p.  58.  From 
A.S.  Rlcinga,  which  occurs  in  Rlcinga-hmn,  in  Kemble's  index. 
Ricinga  is  the  gen.  pi.  of  Ricing,  a  son  of  Rica.  The  sense  is 
'  nook  of  the  sons  (or  family)  of  Rica.' 

RiNGSHALL.  Spelt  Ringeshale,  Ipm. ;  Ringeshal,  T.N. ; 
Ringeshala,  D.B.,  p.  249.  The  prefix  is  the  A.S.  Hringes,  gen. 
of  Hring  (borrowed  from  the  0.  Norse  Hringr),  and  the  sense 
is  'Hringr's  nook.'     See  RiNGSFiELD. 

Spexhall.  Copinger  gives  as  old  spellings  such  forms  as 
Speccyshale,  Spectyshale  (obvious  error  for  Speccyshale),  Spet- 
teshale  (error  for  Specceshale),  Speckshall,  &c.  The  suffix  is 
clearly  hale.  The  prefix  can  only  take  the  form  of  Specces, 
gen.  of  an  A.S.  *Specc,  which  is  unknown.  If  it  were  a 
name,  we  should  then  have  'Speck's  nook'  as  the  sense.  The 
E.D.D.  says  that  Speck  is  the  Norf.  word  for  a  wood-pecker, 
which  would  represent  an  A.S.  *specc,  and  would  be  cognate 
with  the  G.  8pech-t.  Kluge  says  that  the  E.  speight,  a  wood- 
pecker, is  borrowed  from  German,  but  thinks  that  the  G. 
Specht  may  be  allied  to  the  A.S.  specca,  a  speck  ;  with  reference 
to  the  parti-coloured  plumage  of  the  bird.  My  guess  is  that 
the  name  means  'wood-pecker's  nook.'  Compare  Yaxley,  i.e. 
'  cuckoo's  lea.' 

Stradishall.  Here  the  suffix  is  not  hale,  but  sele,  a 
dwelling-place  ;  as  in  Lawshall  (above).  D.B.  has  Stratesella, 
p.  233.  Copinger  gives  such  forms  as  Stradesel,  Stradesele, 
Stradesyll ;  and  H.R,  has  8t7xitesele,  Strattesele.  I  suppose  the 
original  form  to  have  been  the  A.S.  stra't-sele,  i.e.  'dwelling 
near  a  street'  or  old  road.  The  A.S.  strwt  is  frequently 
represented  by  Strat-  in  Strattons  and  Stratfords.  The  medial 
-is-  or  -es-  was  easily  introduced  as  a  fictitious  genitive  suffix, 
as  it  is  common  in  many  place-names,  and  was  suggested  by 
the  s  in  the  suffix  -sele.  About  a  mile  and  a  half  from  the 
present  Stradishall  church  we  find  a  Wickham  Street  marked 
on  the  ordnance  map,  which  leads  directly  to  Wickhambrook. 
The  present  road  from  Stradishall  to  Clare  passes  through  a 


NAMES    ENDING    IN    -HALL,    -HAM  47 

place  where  the  map  has  a  Chilton  Street ;  and  a  road  joining 
Wickham  Street  to  Chilton  Street  would  pass  through  or  near 
Stradishall. 

Uggeshall.  Spelt  Huggethale,  error  for  Uggec{e)hale,  T.N. 
D.B.  has  Uggiceheala,  p.  38 ;  Wggessala,  p.  337 ;  Ulkesala, 
p.  102.  All  these  are  bad  spellings,  but  they  lead  back  to 
the  form  Ugges-hale,  evidently  compounded  of  the  suffix  liale 
and  of  Ugges,  gen.  of  Ugg,  not  an  English  name,  but  adapted 
from  the  Dan.  Uggi,  allied  to  the  Icel.  uggi-,  fear,  which  has 
given  us  the  adj.  ugly.  The  sense  is  '  Uggi's  nook.'  (Uggr 
is  one  of  the  names  of  the  god  Odin  in  the  Edda.) 

Westhall.  Spelt  Westhale,  H.R.;  R.B.  The  sense  is 
simply  'west  nook.' 

23.     Ham. 

This  is  an  extremely  common  suffix,  and  arises  from  two 
distinct  sources,  which  cannot  in  many  cases  be  separated  ;  so 
that  all  the  names  in  -ham  must  be  considered  together.  The 
modern  -ham  represents  either  (1)  A.S.  ham,  a  home,  or 
village,  or  village  community,  shortened  to  ham  in  an  un- 
stressed position;  or  (2)  the  A.S.  hamm,  also  Jtam,  meaning 
an  'enclosure'  or  'a  place  fenced  in,'  connected  with  the  modern 
English  to  hem  in.  In  the  few  cases  in  which  the  ultimate 
origin  can  be  ascertained,  the  fact  will  be  noted. 

This  common  suffix  occurs  in  the  following,  viz.  Akenham, 
Aldham,  Aldringham,  Badingham,  Barham,  Barnham,  Barning- 
ham,  Barsham,  Baylham,  Blakenham,  Brantham,  Brettenham, 
Bucklesham,  Cavenham,  Chattisham,  Coddenham,  Cretingham, 
Dalham,  Darsham,  Debenham,  Denham,  (Santon)  Downham, 
Elmham,  Fakenhara,  Falkenham,  Farnham,  Felsham,  Finning- 
ham,  Fornham,  Framlingham,  Freckenham,  Gisleham,  Gisling- 
ham,  Glemham,  Helmingham,  Henham,  Heveningham,  Higham, 
Hintlesham,  Hitcham,  Horham,  Icklingham,  Ingham,  Langham, 
Lavenham,  Layham,  Letheringham,  Martlesham,  Mendham, 
Mendlesham,  Mettingham,  Needham  Market,  Pakenham, 
Parham,  Redisham,  Reudham,  Rendlesham,  Rougham,  Saxham, 


48  THE    PLACE-NAMES    OB^   SUFFOLK 

Saxinundhani,  Shottisham,  Soham,  Somersham,  Stonham,  Syle- 
ham,  Thelnetham,  Thoraham,  Tuddenham,  Walsham,  Wattis- 
ham,  Wenham,  Whelnetham,  Wickham  Market,  Wickhambrook, 
Willingham,  Willisham,  Witnesham,  Worlingham,  Wortham, 
and  Wrentham  ;  more  than  eighty  in  number. 

Akenham.  Ill  spelt  Acreham,  D.B.,  p.  17.  Aken  represents 
the  A.S.  Acan,  gen.  of  Aca,  a  known  name.  Cf.  Acan-tiln,  in 
Birch,  iii.  603.  We  may  generally  take  -ham  to  mean  '  home ' 
after  a  personal  name  in  the  genitive,  unless  there  is  evidence 
to  the  contrary.     The  probable  sense  is  '  Aca's  home.' 

Aldham.  Spelt  Aldham,  Ipm.;  Aldeham,  D.B.,  p.  14. 
The  e  in  Aide-  indicates  the  use  of  the  definite  form  of  the 
adjective.  For  O.  Merc,  se  alda  ham,  nom.,  or  cet  tham  aldan 
hdme,  dative.     It  means  '  the  old  home.' 

Aldringham.  Spelt  Aldringham,  H.R.;  but  Alrincham  in 
D.B.,  p.  59.  The  A.S.  personal  name  Eallrinc  occurs  in  Birch, 
C.S.  ii.  45  as  a  witness;  answering  to  0.  Merc.  Allrinc.  The 
gen.  suffix  -es  has  been  lost,  as  occasionally  happens.  The  sense 
is  '  Allrinc's  home.' 

Badingham,  or  Baddingham.  Near  Framlingham.  Spelt 
Badingham,  H.R.;  Badincham,  D.B.,  p.  96.  The  name  Bada 
occurs  in  the  Liber  Vitse  of  Durham,  and  elsewhere.  The 
sense  is  '  home  of  the  Badings,'  or  '  of  the  sons  of  Bada.'  (A.S. 
Badinga  ham.)     Compare  Badley,  Bad  well. 

Barham.  Spelt  Bergham,  Ipm.,  p.  241  ;  Berhani,  D.B., 
p.  236 ;  Bercham,  p.  49.  The  prefix  is  the  O.  Merc,  herh, 
A.S.  heorh,  a  hill,  a  barrow ;  and  the  suffix  is,  in  this  case, 
probably  hamm,  an  enclosure.  The  sense  is  '  hill-enclosure ' ; 
or  '  enclosure  beside  a  hill.'  There  is  a  small  hill  near  it. 
(N.B.  Barham,  Kent,  is  A.S.  Beoraham.) 

Barnham.  Spelt  Bernham,  T.N.;  D.B.,  p.  37.  From  A.S, 
bern,  herern,  a  barn.  The  sense  is  '  barn-enclosure ' ;  or  '  en- 
closure with  a  barn.' 


NAMES    ENDING    IN    -HAM  49 

Barningham.  Spelt  Berningham,  KB. ;  D.B.,  p.  170 ; 
Berni7icham,  D.B,,  p.  147.  For  O.  Merc.  Berninga  ham,  A.S. 
Beoringa  ham.  The  sense  is  'home  of  the  Bernings'  (A.S. 
Beornings),  or  '  of  the  sons  of  Bern '  (A.S.  Beorn). 

Barsham.  Spelt  Barsham,  T.N. ;  Barshmn,  D.B.,  p.  94 ; 
Bersham,  D.B.,  pp.  109,  110.  From  the  A.S.  Bwre,gen.  Bwres. 
The  sense  is  '  Baere's  home.'     Baere  is  a  known  name. 

Baylham.  Spelt  Beilhant,  Ipm.;  Beylham,  Ipm.;  Baylham, 
Ipm.;  Beleham,  D.B.,  p.  112.  The  diphthong  ei  suggests  a 
Scandinavian  origin ;  and  as  the  Mid.  E.  slei,  '  sly,'  is  from 
the  0.  Norse  sla^gr  (in  Zoega's  O.  Icel.  Diet.),  so  I  suppose 
beil-  may  be  from  the  O.  Icel.  boeli,  a  farm,  dwelling.  The 
sense  may  be  '  farm-enclosure.' 

Blakenham.  Spelt  Blakenham,  H.R.;  Blacheham,  D.B., 
p.  142  (with  che  for  ken).  The  corresponding  A.S.  form  is 
Blacan-hdm,  i.e.  '  Blaca's  home.'     Blaca  is  a  known  name. 

Brantham.  Spelt  Brantham,  Ipm.;  D.B.,  p.  31  ;  Brantestana, 
D.B.,  p.  30.  Copinger  also  gives  the  spelling  Brentham.  It 
seems  to  be  derived  directly  from  the  A.S.  brant,  'steep';  if 
so,  the  sense  is  '  the  steep  enclosure,'  or  '  enclosure  near  the 
steep  slope.' 

Brettenham.  Spelt  Bretenham,  H.R.;  T.N.;  Ipm. ;  Bretham, 
D.B.,  p.  22  ;  Bretenhama,  D.B.,  p.  177.  Not  to  be  connected 
with  the  Britons ;  because  the  A.S.  Brytt,  a  Briton,  is  a  strong 
sb.,  with  the  gen.  sing.  Bryttes  and  gen.  pi.  Brytta  (without  n). 
The  origin  is  rather  from  the  A.S.  brettan,  bryttan,  gen.  of 
bretta,  brytta,  mostly  in  the  sense  of  'lord,'  or  'prince';  though 
the  literal  meaning  is  'distributor';  cf,  Icel.  bryti,  a  steward. 
The  probable  meaning  is  'prince's  home.'  It  is  clear  that 
Brettenham  suggested  the  name  of  Breton  (as  it  is  spelt  in 
Kirby,  p.  270)  for  the  river  that  rises  near  it,  though  the  name 
is  now  shortened  to  Bret,  which  disguises  the  connexion. 

BucKLESHAM.  Spelt  Bokelesham,  H.R.;  Bukelesham,  D.B., 
p.  23.  These  forms  suggest  an  A.S.  form  *Bucles  (or  *Buccles) 
ham,  i.e.  *Bucers  (or  *Buccers)  home  ;  but  the  forms  Bucel, 

C.  A.  S.  Octavo  Series.     No.  XLVI.  4 


50  •  THE   PLACE-NAMES   OF    SUFFOLK 

Buccel  are  not  recorded.      They  look  like  a  diminutive  from 
the  known  name  Bucca.     Compare  Buckle  Brook,  Lanes. 

Cavenham.  "For  shortness  called  Canham  " ;  Kirby(1813). 
Spelt  Gauenham,  H.R. ;  Ipm.  (printed  Cavenham);  Kauanaham, 
D.B.,  p.  245,  where  it  is  apparently  miswritten  as  Kanaaaham. 
The  third  a  is  superfluous,  and  may  have  been  due  to  confusion 
with  Kauanadis,  which  is  the  misspelling  in  D.B.,  at  p.  335,  of 
Kauanedis  (Cavendish).  The  right  form  is  Cauanham,  where 
Cauan  is  the  gen.  of  Caua  (in  Searle's  Onomasticon),  which 
should  rather  be  written  as  Gafa ;  for  /  is  the  right  symbol 
for  V  between  two  vowels.  The  sense  is  '  Cafa's  home.'  See 
Cavendish. 

Chattisham.  Spelt  Chatisham,  H.R.;  Ghatesham,  Ipm.; 
Getessam,  D.B.,  p.  14.  Ghates  represents  an  A.S.  *Geattes,  gen. 
of  *Geatt.  This  exact  form  does  not  occur ;  but  the  correspond- 
ing weak  form  Geatta  is  found  in  Geattan-broc ;  in  Kemble's 
index.  The  sense  is  '  Ceatt's  home';  or  '  Ceat's  home.'  Com- 
pare Chettisham,  Cambs. 

CoDDENHAM.  Spelt  Godeiikam,  Ipm.;  D.B.,  pp.  9,  115.  For 
A.S.  Godan  ham,  where  Codan  is  the  gen.  of  Goda,  as  in  Godan- 
ford,  Godan-clibe  (Birch,  C.S.  i.  295).  The  sense  is  '  Coda's 
home.'     See  Cotton. 

Cretingham.  Spelt  Greting,  H.R. ;  Gretinges,  T.N. ; 
Gretingaham  (for  Greting  a  J  tarn),  D.B.,  pp.  39,  305.  It  seems 
safest  to  refer  these  forms  to  the  A.S.  Gretta,  as  this  is  a  known 
name.  Then  Greting  (H.R.)  would  represent  the  gen.  pi. 
Grettinga,  'of  the  sons  of  Cretta';  Gretinges  (T.N.)  would 
represent  the  nom.  pi.  Grettingas;  and  the  place-name  will 
mean  '  home  of  the  sons  (or  family)  of  Cretta.' 

Dalham.  Spelt  Dalham,  H.R. ;  R.B. ;  D.B. ;  p.  219; 
Daelham,  Birch,  C.S.  iii.  612,  last  line.  From  A.S.  dcel,  a  dale. 
The  sense  is  'dale-enclosure,'  or  'enclosure  in  the  dale.'  'Dale- 
home'  is  less  probable;  but  quite  possible. 

Darsham.    Spelt  Z>er5Aa7>t,  Ipm.;  H.R.;  D.B.  p.  4;  Dersam, 


NAMES    ENDING    IN    -HAM  51 

D.B.,  p.  24 ;  Diresham,  D.B.,  p.  108.     For  A.S.  Deoies  ham  ; 
i.e.  '  Deor's  home.'     The  literal  sense  oi  deor  is  'deer.' 

Debenham.  Spelt  Deheham,  H.R. ;  R.B. ;  Depham,  D.B., 
p.  192;  Depbeham,  on  the  same  page;  Dephenham,  D.B., 
pp.  49,  50.  Also  Depham,  Ipm. ;  Dehham,  in  a  late  copy  of  a 
charter,  in  Kemble,  iv.  245. 

The  spellings  DepbenJiam,  Depbeham  are  only  variants  of 
Debbenham,  Debbeham,  as  shown  by  the  D.B.  spelling  of  Up- 
bestuna  for  Ubbeston.  The  original  form  was  certainly  the 
adj.  deop,  'deep,'  in  the  dative  case  deopan;  in  the  phrase  cet 
thcim  deopan  hamme,  'at  the  deep  enclosure';  or,  less  probably, 
wt  thdm  deopan  hdme,  '  at  the  deep  home.'  Under  the  stress, 
the  eo  was  shortened,  giving  Deppenham  and  Debbenham ; 
Depham  and  Debham  resulted  immediately  from  the  nom. 
deop  hamm  (or  ham).  There  is  a  Deopham  in  Norfolk ;  and 
the  modern  Deptford  is  spelt  Depeford  in  Chaucer;  meaning 
'deep  ford.' 

It  follows  that  it  is  wholly  impossible  even  to  imagine  that 
Debenham  took  its  name  from  the  river  Deben  ;  on  the  con- 
trary, the  river  was  named  from  the  place,  because  it  there 
takes  its  rise.  "  The  country  round  this  Town  is  very  deep 
and  dirty,  but  the  Town  itself  is  clean,  standing  on  a  rising 
Hill";  Kirby. 

Denham.  There  are  two  places  of  this  name ;  one  near 
Bury,  and  one  near  Eye  (Kelly).  Spelt  Denham,  H.R. ;  T.N. ; 
D.B.,  p.  59  ;  Deneham,  Thorpe,  Diplomat.,  p.  408.  In  the  last, 
the  prefix  Dene-  represents  the  A.S.  denu,  a  valley,  rather  than 
Dena/of  the  Danes.'  The  sense  may  well  be  'valley-enclosure,' 
or  '  enclosure  in  the  valley.' 

Downham,  or  Santon  Downham,  as  distinguished  from 
Santon  in  Norfolk,  on  the  other  side  of  the  Little  Ouse 
(Kelly).  The  soil  is  of  light  sand ;  and  Santon  means  Sand- 
town.  i)«/iAam,  D.B.,  pp.  157,  203.  There  is  another  Downham 
in  Norfolk,  at  some  distance  to  the  N.W.  Downham  is  probably 
the  A.S.  Dun-ham  {=D€in-hamm),  i.e.  '  hill-enclosure.'  Kemble 
and  Thorpe  mark  the  a  in  ham  as  long,  but  without  authority ; 

4—2 


52  THE   PLACE-NAMES   OF   SUB'FOLK 

according  to  Thorpe,  at  pp.  383,  422,  and  424  of  his  Diplo- 
matarium,  the  MS.  has  'Dunham'  in  each  passage  ;  and  all  the 
passages  occur  in  quite  late  charters.  Were  the  a  long  (which 
I  doubt)  the  sense  would  be  '  hill-home.' 

Elmham.  The  South  Elmhams  are  parishes  in  the  Northern 
division  of  the  county  ;  viz.  South  E.  All  Saints,  South  E. 
St  George,  St  James,  St  Margaret,  St  Michael,  and  St  Peter. 
Spelt  Elmham,  H.R. :  T.N.  ;  Elmeham,  D.B.,  p.  94.  The  sense 
is  'enclosure  (or  home)  near  the  elm.' 

Fakenham.  Spelt  Fakeham,  R.B.,  T.N. ;  Fachenham,  D.B., 
p.  174 ;  Fakenham,  in  a  late  copy  of  an  A.S.  document,  in 
Thorpe,  Diplomat.,  p.  594.  The  prefix  represents  an  A.S. 
*Facan,  gen.  of  *Faca ;  cf.  the  recorded  names  Fac-ualdus  and 
Facca.     The  sense  is  '  Faca's  home'  or  '  enclosure.' 

Falkenham.  Spelt  Falkeham,  T.N. ;  Faltenham,  error  for 
Falcenham,  D.B.,  p.  120;  cf.  Falchenham  in  Birch,  CS.  iii. 
659.  Nearly  the  same  name  as  that  of  Fawkham,  Kent, 
which  appears  in  a  Latin  charter  as  Falcheham,  in  Birch,  C.S. 
iii.  375  (1.  6  from  bottom);  but  in  a  better  A.S.  charter  as 
Fealcna-ham,  in  the  same,  iii.  374,  1.  5.  Here  Fealcna  must 
be  the  gen.  pi.  of  Fealca,  which  can  only  be  the  L.  falco,  a 
falcon,  done  into  English  spelling ;  whence  also  the  gen.  sing. 
Fealcan.  This  A.S.  form  is  otherwise  unknown,  but  it  may 
have  been  in  occasional  use,  as  the  Lat.  y>^.  fal cones  appears  in 
the  Epinal  and  Corpus  Glossaries,  both  of  the  eighth  century. 
The  sense  is  '  the  enclosure  of  the  falcon ' ;  or  '  Falcon's  enclo- 
sure.'    Compare  Hawkedon,  above. 

Farnham.  Spelt  Farnham,  D.B.,  p.  72 ;  Ferneham,  D.B., 
p.  128.  The  A.S.  form  is  Fearnham,  in  Kemble's  index.  The 
sense  is,  probably,  '  fern-enclosure '  rather  than  '  fern-home.' 

Felsham.  Spelt  Felisham,  Ipm. ;  Fealsham,  D.B.,  p.  164. 
Copinger  also  notes  the  form  Faleshani.  The  prefix  answers  to 
the  A.S.  Fades,  which  appears  in  Fddes-gnefe,  in  Birch,  iii. 
587,  1.  3  from  bottom  (twice).  The  nom.  case  is  Fcele ;  cf.  A.S. 
fwle,  adj.,  faithful,  good.     The  sense  is  '  Ftele's  home.' 


NAMES    ENDING    IN    -HAM  53 

FiNNiNGHAM.  Spelt  Feningham,  Ipm. ;  Finingaham,  D.B., 
p.  58.  Copinger  also  gives  Finingham.  The  spelling  with  en 
ought  to  be  significant,  as  en  usually  becomes  in ;  but  not 
conversely ;  and  fen-ing  might  mean  a  fen-man.  The  sense  is 
either  '  home  of  the  fen-men,'  or  '  home  of  the  sons  of  Finn.' 
But  in  the  latter  case  we  should  expect  to  find  nn  in  the  old 
spellings. 

FoRNHAM.  There  are  three  places  of  this  name  near 
together,  viz.  F.  All-Saints,  F.  St  Martin,  and  F.  St  Genevieve. 
Spelt  Fornham,  Ipm.:  D.B.,  p.  162.  Forna  was  a  somewhat 
common  name,  as  there  are  half-a-dozen  examples  of  it;  we 
may  conclude  that  the  original  form  was  Fornanhdin,  i.e. 
'  Forna's  home.'  The  syllable  an  would  be  very  easily  lost, 
owing  to  the  repetition  of  n.  Rygh  gives  Forni  as  a  Norse 
name ;  whence  (says  Bjorkraan)  the  A.vS.  Forna  was  borrowed. 

Framlingham.  Spelt  Framling ham,  Ipm.;  Framelingham, 
H.R. ;  Framelingham,  D.B.,  p.  44  :  Framelingaham,  D.B.,  pp.  90, 
297.  The  last  suggests  '  home  of  the  Framelings  ' ;  but  whether 
that  is  quite  the  correct  form,  we  have  no  further  evidence. 

Freckenham.  Spelt  Frekenham,  H.R.;  Frakena.ham,'D.'B., 
p.  201.  Also  Freheham  (twice)  in  a  twelfth  century  copy  of 
an  A.S.  charter  originally  dated  895  ;  in  Birch,  C.S.  ii.  212, 
213.  We  also  find  Frecan-thorn  in  Birch,  C.S.  ii.  270.  Frecan 
is  the  gen.  of  freca,  a  bold  man,  a  warrior.  The  form  in  D.B. 
suggests  the  gen.  pi.  frecena,  as  if  it  were  'home  of  the 
warriors ' ;  rather  than  the  gen.  sing,  frecan,  which  would  give 
'  home  of  the  warrior.'  Cf.  the  patronymic  Frsecing ;  Birch, 
C.S.  i.  474. 

GiSLEHAM.  Spelt  Gisleham,  D.B.,  pp.  5,  43.  The  form 
Oislan-ford  occurs  in  an  A.S.  charter ;  Birch,  C.S.  iii.  588. 
Here  Glslan  is  the  gen.  of  Glsla ;  compare  the  names  Glsl- 
heald,  Glslbeorht,  &c.  But  Gtsli  (says  Rygh)  was  a  common 
Nonse  name,  and  Gisle-  well  represents  its  genitive  Glsla ; 
moreover  the  Norse  initial  g  remained  hard,  whilst  the  A.S.  g 
usually  became  y  and  then  disappeared.  The  sense  is  '  Gisli's 
home.' 


54  ■  THE   PLACE-NAMES   OF   SUFFOLK 

GiSLiNGHAM.  Spelt  Gyslyngham,  H.R.  ;  Gislingaham, 
D.B.,  p.  11 ;  Gislingheham,  D.B.,  p.  83.  The  sense  is  'home  of 
the  Gislings,'  or  'of  the  sons  of  Gisli.'  See  the  preceding 
name. 

Glemham.  Sipeh  Olemham,l\:im.;  D.B.,  p.  245;  Gliemham, 
D.B.,  pp.  33,  129  ;  Glaimham,  D.B.,  p.  56.  The  forms  Gliem-, 
Glaim-,  show  that  the  vowel  was  formerly  long.  Perhaps  for 
*Gl^m-hamm  ;  from  the  A.S.  glw7n,  gleam,  brightness.  This 
suggests  the  sense  '  gleam-enclosure  ' ;  as  if  it  were  in  a  sunny 
situation.  Of.  Glemsford.  This  solution  is,  of  course,  con- 
jectural. The  name  of  the  river  Glem  is  probably  unoriginal ; 
for  otherwise,  we  should  expect  the  form  Glemford.  Glemham 
and  Glemsford  are  a  long  way  apart. 

Helmingham.  So  spelt  in  Ipm.,  T.N. ;  Helmingheham, 
D.B.,  p.  22.  Compare  Helmyngton  in  Kemble's  index.  For  A.S. 
Helming a-hdm;  'home  of  the  Helmings,'  or  'of  the  sons  of 
Helm.' 

Henham.  Near  Wangford.  So  spelt  in  H.R.,  T.N. ;  also 
Heneham,  T.N.  Copinger  also  gives  Heenham.  Spelt  Hen- 
ham,  D.B.,  p.  268.  The  dat.  case  Hean-hammce  occurs  in 
Birch,  C.S.,  iii.  649.  Here  htan  is  the  dat.  of  heah,  high  ; 
and  hammce  is  the  dat.  of  hamm.  The  sense  is  '  at  the  high 
enclosure.'     See  Higham. 

Heveningham.  To  the  S.W.  of  Hales  worth.  Spelt  Heve- 
ningham,  Ipm.;  Heueningham,  H.R;  Heueniggeham,  D.B.,  p.  107. 
The  form  Hefan-croft  occurs  in  Kemble's  index  ;  where  Hefan 
is  the  gen.  of  Hefa,  a  known  name.  Thus  the  A.S.  form  would 
be  *Hefaninga-ham,  i.e.  'home  of  the  Hefanings'  or  'of  the 
sons  of  Hefa.' 

Higham.  Near  the  Stour,  to  the  W.  of  E.  Bergholt.  Spelt 
Heham,  Ipm.;  Heyham,  H.R.;  Heiham,  D.B.,  p.  285.  A  charter 
relating  to  Higham  (Kent)  is  endorsed  '  boc  to  heh-ham '  in  a 
hand  of  the  eleventh  century ;  Birch,  C.S.,  i.  301.  Heh  is  the 
O.  Merc,  form  of  A.S.  heah,  high.  The  sense  is  'high  enclosure.' 
The  dat.  case  occurs  in  Henham. 


NAMES   ENDING    IN    -HAM  55 

HiNTLESHAM,  Sipelt  Hijntlesham,  iTpm.]  Hintlesham.T.'N.; 
D.B.,  p.  17 ;  and  in  a  late  A.S.  charter,  in  Thorpe,  Diplomat., 
p.  569.  Also  Hintelesham,  R.B.,  H.R. ;  Hentlesham,  Huntles- 
ham,  T.N.  The  vai'iation  in  the  sound  of  the  first  vowel,  which 
appears  as  y,  i,  e,  and  x,  can  be  accounted  for  if  we  assume  it 
to  have  been  originally  y.  Hence  the  sense  is  'home  of  Hyntel'; 
where  Hyntel  is  a  name  not  yet  recorded.  But  it  is  a  regular 
diminutive  of  A.S.  Hunta,  which  would  give  *Huntila,  *Hyntel. 

HiTCHAM.  Spelt  Hicham,  Ipm.;  Hecham,  H.R.;  D.B.,  p.  208; 
Hetcham,  D.B.,  p.  221.  The  dat.  Hecan-ige  occurs  in  an  A.S. 
charter;  Kemble,  CD.,  vi.  221;  and  the  name  Heca  occurs 
again  in  the  A.S.  Chronicle.  The  A.S.  form  would  be  Hecan- 
hdm,  i.e.  '  home  of  Heca.'  Copinger  records  the  spelling 
Heacham ;   so  that  Heacham  (Norf )  is  the  same  name. 

HoRHAM.  Spelt  Horham,  H.R.;  Hoi^am,  D.B.,  p.  97.  The 
spelling  Horham  occurs  in  Bp  Theodred's  Will ;  see  Thorpe, 
Diplomat.,  p.  513.  From  the  A.S.  horu,  mud.  The  sense  is 
'  mud-enclosure,'  or  '  muddy  enclosure.'  The  same  prefix  occurs 
in  Horbury  and  Hoiton,  in  the  West  Riding  of  Yorkshire. 

ICKLINGHAM.  Spelt  7/;e^z»^Aam,  H.R. ;  T.N.;  Ecclingaham, 
D.B.,  pp.  16,  271.  The  prefix  is  the  same  as  in  Ickleton, 
formerly  Icklington,  Cambs. ;  see  the  A.S.  spelling  Icelingtim 
in  ^If  helm's  Will ;  in  Birch,  C.S.,  ii.  630.  The  sense  is  '  home 
of  the  Iclings '  or  '  of  the  sons  of  Icel.'  Icel  and  Iceling  (of 
which  Icling  is  an  abbreviation)  both  occur  in  the  A.S. 
Chronicle.  The  Iclingas  or  Iclings  were  a  Mercian  family. 
See  my  accounts  of  Ickleford,  Cambs.,  and  Icleford,  Herts. 
Of  course  none  of  these  names  is  in  any  way  connected  with 
the  Icenhild  Way,  as  the  antiquaries  so  often  delight  in  saying, 
in  contempt  of  phonetic  considerations.  Sometimes  they  invoke 
the  Iceni ! 

Ingham.  Spelt  Yngeham,  R.B. ;  Ingham,  D.B.,  p.  167 ; 
Incham,  D.B.,  p.  135.  In  an  Oxfordshire  charter,  dated  880, 
there  is  mention  of  "  Incghsema  gem^re,"  or  '  boundary  of  the 
men  of  Incgham  ' ;  where  the  occurrence  of  cd  shows  that  the  a 
in  liam  was  long,  and  the  sense  '  home.'     Incgham,  variant  of 


56  THE    PLACE-NAMES   OF   SUFFOLK 

Ingham,  is  a  compound  word,  and  the  prefix  ing  represents  the 
O.  Norse  eng,  a  meadow;  as  in  Ingbirchworth  in  the  W.  Riding 
of  Yorkshire.  The  prefix  is  certainly  Norse ;  we  do  not  find 
any  trace  of  it  in  such  counties  as  Beds.,  Berks.,  Carabs.,  Herts., 
Hunts.,  which  show  scarcely  any  sign  of  Scandinavian  influence. 
(Inkpen,  Berks.,  means  '  Inga's  pen,'  where  Inga  is  a  personal 
name.)  The  sense  of  Ingham  is  '  meadow-home.'  There  is 
another  Ingham  in  Norfolk,  and  a  third  in  Lines. 

Langham,  Four  miles  N.  [by  W.]  of  Elmswell  station 
(Kelly).  Spelt  Langeham,  R.B. ;  Langham,  D.B.,  p.  173.  I 
suppose  it  means  '  long  enclosure  ' ;  from  the  A.S,  lang,  long. 

Lavenham.  Spelt  Lauenham,  H.R. ;  D.B.,  pp.  149,  275; 
Lauanham,  in  Thorpe,  Diplomat.,  p.  521,  1.  2.  Here  Lauan 
represents  the  A.S.  Lafan,  gen.  of  Lafa,  a  name  recorded  in 
the  Liber  Vitae  of  Durham.  The  sense  is  'Lafa's  home.'  Often 
shortened  to  Lauham,  and  actually  spelt  Lanam  by  Skelton,  in 
bis  Why  Come  ye  nat  to  Courte,  1.  930.  Dyce  wrongly^  explains 
it  as  meaning  Langham  (Essex), 

Layham..  To  the  S.  of  Hadleigh.  Spelt  Leyham,  T.N. ; 
H.R.;  Ipm,;  Leiham,  D.B.,  p.  246.  But  it  is  spelt  Hligham 
(in  connexion  with  Hadleigh)  in  iElfflSd's  Will ;  in  Thorpe, 
Diplomat.,  p.  520,  1.  10.  I  cannot  explain  this  form  hltg,  in 
which  the  final  g  was,  however,  a  mere  glide,  except  by  con- 
necting it  with  the  O.  Friesic  hli,  O.  Norse  hly,  warmth,  Dan. 
ly,  shelter,  cover,  and  the  O.N.  hlyi^  warm,  A.S.  hleoiv,  shelter, 
protection;  prov.  E.  lew,  shelter,  and  E.  lee.  Cf.  A.S.  Iilywan, 
to  keep  warm ;  in  Napier's  Glosses,  i.  252.  The  modern  form 
also  presents  difficulty ;  there  may  have  been  confusion  with 
the  verb  to  lay.  See  Lay,  verb,  in  the  E.D.D.,  sect.  24 ;  where 
lay  is  given  as  a  sb.,  meaning  '  shelter  for  wild  fowl.'  There  is 
a  probability  that  the  right  sense  is  'sheltered  enclosure';  with 
reference  to  protection  from  cold. 

Letheringham.  Spelt  Letheringham,  Ipm. ;  Letheringa- 
ham,  D.B.,  p.  216  ;  Ledringaham,  D.B.,  p.  135.  I  think  the 
A.S.  form  of  the  prefix  must  have  been  Leoderinga,  gen.  pi. 
of  Leodering,  for  Leod(h)er-ing,  i.e.  son  of  Lead-here,  a  known 
name,  formed  from  two  very  common  elements.     If  so,  the  sense 


NAMES   ENDING    IN    -HAM  57 

is  'home  (or  enclosure)  of  the  sons  of  Leodhcre.'     Compare 
Letheringsett,  Norf 

Martlesham.  To  the  S.W.  of  Woodbridge.  Spelt  Mar- 
cles/iam  (with  c  for  t),  Ipm.,  p.  218;  but  Merlesham  (without  t), 
D.B.,  p.  287,  which  can  hardly  be  right.  Copinger  also  gives 
the  forms  Martelisham,  Marthelisliam,  Martilsham,  Mertlesham, 
Me7'tlisham ;  all  from  a  base  Mart-,  Marth-,  Mert-.  The  only 
name  with  a  similar  base  is  Mart-ley,  Wore. ;  and  the  only 
Teutonic  word  at  all  resembling  this  is  the  A.S.  mearth,  a 
marten ;  0.  Norse  mofSr.  A  base  *Mart-  might  give  a  dimin. 
*Mertila,  A.S.  *iyiertel ;  whence  we  might  obtain  the  sense  of 
'  Mertel's  home.'  But  we  have  no  sufficient  evidence  of  this; 
so  that  the  name  remains  unsolved. 

Mendham.  On  the  Waveney;  to  the  S.W.  of  Bungay.  So 
spelt  in  T.N. ;  R.B. ;  Ipm. ;  D.B.,  p.  175  ;  also  in  a  late  copy  of 
Bp  Theodred's  Will,  in  Thorpe,  Diplomat,  p.  513;  and  Mi/nd- 
ham,  on  the  same  page;  also  Myndaham,  Birch,  iii.  210,  1.  17  ; 
see  below. 

Mendlesham.  Spelt  Mendlesham,  R.B.;  Mendelisham,  Ipm., 
p.  8.  D.B.  has  the  forms  Melnessam,  p.  10;  Munlesham,  p.  11 ; 
Mundlesham,  same  page.  The  variation  of  the  vowel,  from  e 
to  u,  suggests  that  the  original  form  had  the  A.S.  y,  and  that 
the  prefix  represents  an  A.S.  *Myndel,  formed  from  an  earlier 
^Mund-il,  due  to  adding  the  dimin.  suffix  -il  to  the  A.S.  Mund. 
Cf.  G.  milndel,  a  pupil.  Both  Mund  and  Munda  occur  as  A.S. 
personal  names.  If  this  be  right,  the  original  sense  was 
'  Myudel's  home '  or  '  Myndel's  enclosure.'  And  a  possible 
sense  of  Mendham  (above)  is  '  Mynda's  home '  or  '  Mynda's 
enclosure';  where  *Mynda  is  from  a  stem  *Mund-jon-,  formed, 
in  the  usual  way,  as  a  weak  masculine  ;  from  the  base  Mund. 

Mettingham.  E.  by  N.  from  Bimgay.  Spelt  Metiugham, 
H.R. ;  R.B. ;  Metingaham,  D.B.,  p.  40.  The  last  form  implies 
'  the  home  (or  enclosure)  of  the  sons  of  Ma3te ' ;  supposing  the 
A.S.  adj.  7n(ete  to  be  used  as  a  proper  name.  The  usual  sense 
of  m(^te  is  '  moderate,  small,  poor,'  which  would  easily  become 
an  epithet  and  give  a  name.  Cf.  the  form  MiPtelm  (for  *MiSt- 
helm)  in  Birch,  C.S.,  ii.  469,  1.  20. 


58  THE    PLACE-NAMES   OF    SUFFOLK 

Needham.  Spelt  Nedham,  H.R.  Copinger  also  records 
the  form  Nedeham.  Cf.  A.S.  mod,  nied,  need,  necessity  ;  used 
in  numerous  compounds.  The  sense  is  'a  home  in  need,'  a 
home  which  one  is  driven  to  occupy ;  a  place  of  refuge. 

Pakenham.  Spelt  Pakenham,  Ipm.;  Pachenham  (with  che 
for  ke),  D.B.,  p.  162.  Also  Pakenham,  in  a  late  copy  of  Bp 
Theodred's  Will ;  in  Thorpe,  Diplomat., p.  514.  The  form  implies 
an  A.S.  *Pacan,  gen.  of  *Paca ;  a  name  not  otherwise  known  ; 
but  also  required  to  explain  Pakefield.  The  sense  is  '  home 
(or  enclosure)  of  Paca.' 

Parham.  Spelt  Parham,  Ipm.,  p.  161 ;  Perreham,  D.B., 
pp.  9,  27.  Copinger  gives  Parkham  as  a  variant,  which  has 
the  same  sense.  For  park  (F.  2)arc)  goes  back  to  a  Teut.  form 
*parr-uc  {A.S.  pearr-oc),  where  -uc  is  a  dimin.  suffix  ;  from  an 
older  form  *parr,  which  doubtless  meant  '  an  enclosure.'  Cf. 
prov.  E.  par,  an  enclosed  place  for  domestic  animals ;  from  the 
verb  parren,  to  enclose  or  bar  in.  The  pp.  parred,  confined, 
occurs  in  1400,  and  represents  an  unrecorded  A.S.  *pearran, 
or  *parr{an,  which  I  take  to  be  a  shortened  form  of  A.S. 
sparrian,  to  bar  in  or  fasten  in  with  spars  or  bars.  Thus 
Parham  properly  signifies  '  an  enclosure  made  with  bars,'  or 
'  a  railed-in  enclosure.'     See  Park  in  my  Etym.  Diet. 

Kedisham.  To  the  S.  of  Beccles.  Spelt  Redesham,  Ipm. ; 
H.R. ;  D.B.,  p.  111.  Redes  represents  the  A.S.  Reades,  gen.  of 
Read,  lit.  '  red.'  Though  Read  is  not  recorded  as  a  personal 
name,  it  is  very  clearly  implied  in  the  A.S.  Readingas  (now 
Reading,  Berks.),  lit.  '  sons  of  Read.'  Indeed,  the  forms  Read, 
Reade,  Reid  (all  from  A.S.  Read)  are  still  in  common  use  as 
surnames.     The  sense  is  '  Read's  home,'  or  '  Read's  enclosure.' 

Rendham.  To  the  N.W.  of  Saxmundham.  Spelt  Rend- 
ham,  Ipm.;  but  Rindham,  D.B.,  p.  127;  Rindeham,  D.B.,  p.  128; 
and  Rimdham  (for  Rindham  ?),  D.B.,  p.  54.  The  A.S.  rinde, 
the  rind  or  bark  of  a  tree,  is  here  unsuitable.  We  should 
rather  compare  Rinde-ham  with  the  A.S.  hrinda,  which  occurs 
in  the  phrase  "  oth  hrindan  broc  "  in  Birch,  C.S.,  ii.  60 ;  where 
broc  means  '  brook.'     The  sense  of  hrindan  is  not  known  ;   if  it 


NAMES   ENDING    IN    -HAM  59 

is  the  gen.  of  Hrinda  as  a  personal  name,  the  sense  of  Rendham 
may  have  been  '  Hrinda's  home '  or  '  enclosure.'  But  this  must 
remain  a  conjecture.  (There  is  a  Norse  female  name  Rindv,  but 
the  genitive  is  Rindar.) 

Rendlesham.  Spelt  Rendlesham,  Ipm. ;  D.B.,  p.  26.  But 
the  name  is  very  old,  and  is  explained  by  Beda,  in  his  Eccle- 
siastical History,  bk  iii.  ch.  22,  where  he  speaks  of  "  Rendlaes- 
ham,  id  est,  mansio  Rendili " ;  so  that  the  sense  is  '  Rendil's 
home.'  Rendlaes  is  an  old  form  of  Rendles,  the  regular  genitive 
of  Rendil,  which  drops  the  *  when  the  form  is  lengthened  by  a 
syllable. 

Rougham.  To  the  E.S.E.  of  Bury,  Spelt  Rougham,  Ipm. ; 
Ruhham,  D.B.,  p.  163.  We  find  on  rinoan  Jiammas,  'to  the 
rough  enclosures,'  in  Birch,  C.S.,  ii.  492 ;  where  ruwan  is  the 
ace.  pi.  of  the  A.S.  ruh,  rough,  uncultivated.  Hence  Ruh-ham 
here  means  'rough  or  uncultivated  enclosure.' 

Saxham.  Spelt  Saxham,  Ipm. ;  Saxam,  D.B.,  p.  9  ;  Saxham, 
D.B.,  p,  222.  The  prefix  Sax-  here  represents  the  O.  Merc. 
Saxan,  gen.  of  Saxa  (A.S.  Seaxa),  a  personal  name.  The  sense 
is  '  Saxa's  home,'  or  '  Saxa's  enclosure.'  The  O.  Norse  Saxi  is 
also  common  in  place-names. 

Saxmundham.  Spelt  Saxmundeham,  H.R.;  Saxmondehayji, 
D.B.,  p.  116.  An  s  has  been  dropped;  the  original  form  must 
have  been  Saxmundesbam,  where  Saxmundes  is  the  gen.  case 
of  Saxmund,  an  0.  Merc.  form.  Though  Saxmund  is  not  in 
Searle's  list,  it  is  perfectly  regular ;  since  Sax-  is  a  common 
prefix,  and  -mund  a  common  suffix.  The  sense  is  '  Saxmund's 
home  '  or  '  enclosure.' 

Shottisham.  To  the  S.E.  of  Woodbridge ;  the  same  name 
as  Shottesham,  Norfolk.  Spelt  Shotesham,  Ipm. ;  Scotesham, 
D.B.,  p.  75.  The  same  prefix  occurs  in  Scottes-healh,  in  Birch, 
C.S.,  iii.  240,  1.  2.  Here  Scottes  is  the  gen.  of  Scot,  used  as  a 
personal  name,  and  pronounced  Shot  in  later  A.S.,  and  still  in 
use.  Compare  Shotley  (Suff.),  Shottesbrook  (Berks.),  &c. 
Whether   this    Scot   is    the    same   as   the   A.S.   Scot,   meaning 


(JO  THE    PLACE-NAMES    OF   SUFFOLK 

(1)  a  Scot  of  Ireland,  and  (2)  a  Scot  of  Scotland,  can  hardly 
be  determined. 

SoHAM.  There  is  an  Earl  Soham,  so  called  from  the  Earls 
of  Norfolk;  also  a  Monk  Soham,  called  Soham  Monachorum  in 
Ipm.,  p.  295;  so  called  because  the  Monks  of  Bury  were  patrons 
of  the  Rectory  (Kirby).  The  same  name  as  Soham  in  Cambs. 
Spelt  Saham,  Ipm.,  p.  218  ;  R.B. ;  H.R. ;  D.B.,  p.  26  ;  so  that 
the  0  represents,  as  usual,  an  A.S.  d.  And  this  Saham  is  short 
for  Sag-ham,  as  shown  by  the  variant  form  Saeg-ham,  in  a 
charter  of  the  twelfth  century;  see  Earle,  Land  Charters,  p.  368, 
1.  8.  I  explain  the  A.S.  sag  as  meaning  '  a  depression '  or 
'  hollow ' ;  from  sigaii  (pt.  t.  sag),  to  sink  down.  See  my 
Place-Names  of  Cambs.,  p.  23.  The  sense  is  (probably)  '  en- 
closure near  a  hollow.' 

SoMERSHAM.  Spelt  Somersham,  Ipm. ;  Sumersham,  R.B. ; 
D.B.,  p.  247;  Sumersam,  D.B.,  p.  113.  There  is  another 
Somersham  in  Hunts.,  which  (in  my  Place-Naraes  of  Hunts.) 
I  explain  as  'summer's  enclosure,'  or  'enclosure  for  the  summer.' 
The  A.S.  gen.  sing,  smneres  is  sometimes  thus  used  adverbially, 
meaning  '  in  the  summer.' 

South  Elmham.  See  Elmham  (above). 

Stonham.  Stonham  Aspall  is  to  the  N.E.  of  Needham 
Market.  Near  it  are  Earl  Stonham,  called  Stonham  Comitis 
in  Ipm.,  with  reference,  says  Kirby,  to  Thomas  Brotherton,  Earl 
of  Norfolk  ;  and  Little  Stonham  or  Stonham  Parva,  also  called 
Stonham  Jerningham,  from  the  family  of  that  name.  Stonham 
Aspall  is  so  named  from  the  family  of  that  name,  but  the  family 
was  named  alter  the  place  called  Aspall  (above).  Perhaps  it 
is  worth  while  to  note  that  the  surname  Jerningham  is  an 
example  of  attempting  to  give  an  English  look  to  a  French 
name.  As  Bardsley  points  out,  the  original  form  was  Gernagan; 
which,  with  an  initial  J  for  G,  became  Jernagan.  Then  the 
suffix  -agan  was  "Englished"  by  turning  it  into  -ingham,  which 
is  so  common  an  ending  in  English  place-names.  All  the  same, 
the  initial  J  is  quite  enough  to  detect  its  French  origin. 


NAMES    ENDING    IN    -MAM  61 

Stonluim  is  spelt  Stcmham  in  Iprn.,  R.B.,  H.R.,  and  in  D.B,, 
p.  140.  All  from  the  A.S.  stan,  a  stone.  It  meant  '  stone 
enclosure,'  with  reference  either  to  a  wall  or  to  st(my  soil;  we 
can  hardly  say  which. 

Syleham.  On  tlie  Waveney  ;  N.VV.  of  Wingficld.  Spelt 
Silham,  H.R. ;  also  in  a  late  copy  of  Bp  Theodred's  Will,  in 
Thorpe,  Diplomat.,  p.  513;  Seilam,  D.B.,  p.  298.  Copinger 
also  notes  the  spelling  Sulham.  The  vowels  y,  i,  a,  and  e 
(wrongly  ei  in  D.B.)  all  point  back  to  an  A.S.  y\  and  the 
etymology  (like  that  of  Sulhamstead  in  Berks.)  is  from  the  A.S. 
sylu,  a  miry  place.  The  sense  is  '  enclosure  in  (or  near)  a  miry 
place.' 

Thelnetham.  On  the  Little  Ouse,  to  the  N.W.  of  Botes- 
dale.  There  is  some  strange  mystery  about  this  name  and 
that  of  Whelnetham.  In  both  cases,  there  is  evidence  that 
the  n  is  unoriginal,  and  has  taken  the  place  of  a  v.  In  the 
present  case,  the  spellings  are :  Telneteluivi  or  TelueteJiam,  D.B., 
p.  21;  Thelueteham,  D.B.,  p.  94;  Teluetteham,  D.B.,  p.  148; 
Teolftham,  D.B.,  p.  172.  The  last  of  these  is  very  striking; 
it  is  impossible  that  /  should  be  miswritten  for  n  ;  it  must 
rather  refer  to  a  sound  related  to  v.  It  should  also  be  noticed 
that  there  is  a  Thelveton  in  Norfolk,  and  that  Copinger  notes 
the  spelling  Telvetham.  For  further  discussion  of  this  difficult 
name,  see  under  Whelnetham. 

Thornham.  Spelt  Thornham,  H.R.,  Ipm.,  R.B. ;  Tornham 
(with  T  for  Th\  D.B.,  p.  10;  Thornham,  D.B.,  pp.  80,  83. 
Also  Thornham  in  an  A.S.  charter;  Kemble,  CD.,  iv.  110,  1.  2. 
Kemble's  index  also  has  Thornhwma  die,  with  reference  to 
Worcs. ;  where  the  w  shows  that,  in  some  instances,  the  word 
was  Thornham,  with  long  a.  This  gives  the  sense  '  Thorn- 
home  ' ;  or  a  dwelling-place  near  thorn-trees.  The  sense '  thorn- 
enclosure  '  is  also  admissible,  and  may  here  be  meant. 

TuDDENHAM.  There  are  two  places  of  this  name  ;  Tudden- 
ham  St  Martin,  near  Ipswich,  and  Tuddenham  St  Mary,  near 
Mildenhall  (Kelly).     Spelt  Tudenham,  H.R.;  T.N.;  D.B.,  p.  25 


62  THE    PLACE-NAMES    OF    SUFFOLK 

Tudeham,  R.B.;  D.B.,  p.  823.  Tuddenham  near  Mildenhall 
appears  in  an  A.S.  charter,  dated  854,  as  Tuddan  ham;  Birch, 
C.S.,  ii.  81 ;  the  a  being  marked  as  long.  The  sense  is  '  Tudda  s 
home,'     Tudda  is  a  known  name. 

Walsham,  or  Walsham-le- Willows.  To  the  E.  of  Ix- 
worth.  N.  and  S.  Walsham  are  in  Norfolk.  The  A.F.  le  was, 
no  doubt,  believed  to  be  the  definite  article  at  an  early  date ; 
but  it  is  clear  that  it  had  originally  the  form  les  (as  in  French) 
and  was  a  preposition,  meaning  'near';  being  derived  from  the 
Lat.  latus,  side  ;  whence  the  sense  of  '  beside.'  We  find  Wales- 
ham,  R.B.,  T.N.;  Walsam,  D.B.,  p.  94.  Kemble's  index  has 
Wales-ho,  Wiales-flet,  and  Weales-hic6 ;  so  that  the  most 
probable  original  form  (as  indicated  by  Wales-ham  in  R.B. 
and  T.N.)  was  Weales-hdm  (or  -hamm).  Weales  is  the  gen. 
of  Wealh,  '  a  stranger,  a  foreigner,'  usually  '  a  Briton.'  The 
sense  is  '  stranger's  home,'  or  '  stranger's  enclosure,  near  the 
willows.' 

Wattisham.  Near  Bildeston.  Spelt  Watesham,  R.B, 
Copinger  also  notes  the  forms  Wathesham,  Wathisham,  evi- 
dently with  A.F.  th  for  t;  also  Wachesham,  Wachisham,  with 
th  miswritten  as  ch.  All  the  forms  are  equivalent,  and  can 
be  reduced  to  Watesham  ;  and  there  is  no  evidence  to  connect 
this  immediately  with  the  prefix  in  Wattisfield  (above). 
The  form  Wates  requires  an  A.S.  nom.  Wcet ;  and  though 
we  have  no  record  of  this  except  in  Latin,  we  find  the  allied 
weak  masc.  Wata,  and  the  dimin,  Wwtel,  as  in  Watling  Street, 
and  in  Wattisfield,  The  sense  is  '  Wast's  home '  or  '  Wset's 
enclosure.'  The  Latinised  form  is  Wattus,  spelt  "  Uuattus 
rex"  in  Birch,  C.S.,  i,  113. 

Wenham.  Great  Wenhaim  was  also  known  as  Brent  Wen- 
ham,  i.e.  Burnt  Wenham ;  also  called  Wenham  Gombusta,  Ipm., 
p.  93.  Little  Wenham  is  two  miles  S.E.  of  Raydon  station 
(Kelly).  Spelt  Wenham,  H.R. ;  T.N. ;  D.B.,  p.  29.  We  find 
"terram  de  Wenintone"  in  Birch,  C.S.,  iii.  281  (no.  1061). 
Here  Wenin  is  for  Wenan,  gen.  case  of  a  personal  name  Wena, 


NAMES    ENDING    IN    -HAM  (>:{ 

not   otherwise   known,  though   the   compounds   Wenburh   iuid 
Wenbeorht  occur.     The  sense  is  '  Weua's  home  '  or  '  ench)sure.' 

WheLNETHAM.  Great  and  Little  Whehietham  lie  to  the 
S.E.  of  Bury.  Here,  as  in  the  case  of  Thelnetham,  we  have 
evidence  that  the  n  is  uncniginal.  In  D.B.  we  find  the  extra- 
ordinary form  HueljiJiain,  p.  165.  .  In  Ipni.,  p.  116  (no.  89), 
we  find  Parva  WhelnethciDi ;  but  at  p.  24"J  (no.  34),  the  same 
place  is  referred  to  as  Whelwitham,  which  looks  like  the  original 
from  which  Hiielfiham  was  made,  by  the  turning  of  an  E.  w 
into  a  Norman  v  (here  written  as/).  It  is  clear  that  Thelnet- 
ham and  Whelnetham  must  be  explained  together.  I  can  only 
guess  at  this  riddle,  and  the  theory  I  propose  is  the  following, 
viz.  that  there  were  already,  before  the  Conquest,  two  places  in 
Suffolk  named  Witham  (probably  contracted  from  an  earlier 
Witanham  or  '  home  of  Wita');  and  that  these  places  (which 
are  little  more  than  15  miles  apart)  were  distinguished  by 
the  prefixes  Thel-  and  Hwel  (=  Hweol).  They  thus  became, 
respectively,  Thelwitham  and  Whelwitham ;  or,  with  v  for  w, 
Thelvithan  and  Whelvitham  ;  or,  in  the  spelling  of  D.B.,  Teol- 
f{i)thain  and  Huelfi{t)hain.  How  the  sound  of  v  was  afterwards 
exchanged  for  that  of  n  in  both  cases  (showing  that  they 
certainly  affected  each  other),  it  is  hard  to  say ;  but  we  have 
positive  evidence  that  such  a  change  really  took  place.  It 
remains  to  explain  the  prefixes  Thel-  and  Hwel-.  Thel  is  the 
A.S.  thel,  a  plank,  particularly  one  used  to  form  a  wooden 
bridge  over  a  stream,  as  in  the  case  of  Theale  (Berks.)  and 
of  Thelbridge  (Devon).  Hwel-  is  the  A.S.  hiueol,  mod.  E.  wheel. 
The  A.S.  hweol  is  also  used  in  the  sense  of  '  circle,'  and  may 
denote  that  this  Witham  was  of  circular  form.  This  is  the 
best  I  can  make  of  this  extraordinary  pair  of  names,  both  of 
which  present  very  unusual  features.  I  am  informed  that,  not 
long  since,  the  pronunciation  of  the  latter  place  was  Wheltham, 
a  shortened  form  which  ignored  the  middle  syllable. 

Wickham  Market.  Spelt  Wichant,  R.B. ;  D.B.,  p.  11  ; 
Wycham,  H.R. ;  Wiccham,  D.B.,  p.  10 ;  Wikham,  p.  26.  The 
A.S.  form  is  Ww-ha7n ;  see  Kemble,  CD.,  vi.  98, 1.  6.  From  A.S. 
wic,  a  village  ;   and  ham,  a  home.     The  a  was  long,  because  we 


64  THE    PLACE-NAMES    OF    SUFFOLK 

find  W'lc-hwma  in  Kemble,  v.  243,  1.  8.  The  sense  is  '  village- 
home.'  But  there  was  also  a  Wic-hamm,  or  '  village-enclosure ' ; 
Birch,  C.S.,  iii.  610. 

WiCKHAMBROOK.  The  same  as  the  above;  with  the  addition 
of  bj'uok,  A.S.  broc.  There  is  also  a  Wickham  SkeitS,  near 
Finningham;  from  the  family  name  Skeith,  which  is  obviously 
of  Norse  origin.  The  Icel.  skeith  means  a  space,  a  certain 
length  in  a  course;  and  Vigfusson  notes  that  it  occurs  in  place- 
names. 

Willingham.  Spelt  Wilingham,  T.N. ;  Willing aham,  D.B., 
pp.  6,  109,  which  may  be  the  original  form.  If  it  be  so,  the 
sense  is  '  home  (or  enclosure)  of  the  Willings,'  or  '  of  the  sons 
of  Willa.'  Willa  is  a  known  name.  But  Willingham  in  Cambs. 
is  differently  spelt  in  D.B.,  and  means  '  home  of  the  Wifelings,' 
or  '  of  the  sons  of  Wifel.' 

WiLLiSHAM.  Spelt  Wylavesham,  Ipm. ;  Willauesham,  T.N. ; 
Wylevjesham,  H.R.;  Willaluesham  (error  for  Willauesham),  D.B., 
p.  141.  Also  Willauesham  in  Leofgifu's  Will  (a.d.  1045);  in 
Thorpe,  Diplomat.,  p.  570.  All  from  Willaues,  representing 
the  A.S.  Wigldfes,  or  Wildfes,  gen.  of  Wigldf  or  WUaf,  a  well- 
known  name.  The  g  is  here  a  mere  glide,  and  the  difference 
in  sound  between  the  two  forms  was  very  slight  in  late  A.S. 
The  name  Wtglaf  occurs  in  Beowulf,  11.  2602,  2631,  2745.  The 
sense  is  '  Wiglaf's  home.' 

Witnesham.  Near  Ipswich.  Spelt  Witnesham,  Ipm.,  p.  241, 
col.  1 ;  Witlesham,  col.  2.  But  Copinger  also  gives  such  forms 
as  Wittelesham,  Wyttylisham,  answering  to  Wittlesham  in  D.B., 
p.  294;  so  that  the  n  was  once  I.  The  change  is  not  uncommon. 
Further,  Copinger  gives  the  form  Whitnesham,  with  initial  Wh; 
and  a  comparison  with  Whittlesford  (Cambs.)  and  Whittlesea 
(Cambs.)  tends  to  confirm  this.  I  would  therefore  explain  it 
from  the  name  *HwUel,  the  only  original  form  w^hich  will 
explain  those  names,  being  itself  a  derivative  from  Hwit,  i.e. 
white.  Thus  the  original  sense  was,  probably,  '  Hwitel's  home ' 
or  '  Hwitel's  enclosure.' 


NAMES    ENDING    IN    -HAM,    -HAUGH  65 

WoRLiNGHAM.  Spelt  Werlinghani,  H.R.;  T.N. ;  Warlinga- 
ham,  D.B.,  p.  253  ;  WerlingaJiam,  D.B.,  p.  4.  In  Birch,  C.S.,  ii. 
295,  1.  5  from  bottom,  we  find  "werseles  wellae,"  suggesting  that 
there  was  once  a  name  spelt  Wersel,  equivalent  to  Werel,  whence 
a  derivative  Werling  (shortened  from  Wereling)  would  easily 
result.  However,  it  seems  safe  to  explain  the  above  name  as 
meaning  'home  (or  enclosure)  of  the  Werlings';  which  may 
have  meant  'sons  of  Werel.'  The  prefixes  in  Worlingham, 
Worlington,  and  Worlingworth  are  all  different. 

WouTHAM.  So  in  T.N. ;  spelt  Wordham  (with  d  for  th)  in 
D.B.,  p.  80 ;  Wortham,  D.B.,  pp.  84,  148.  Spelt  Wrtham  (for 
Wurthaiii)  in  Kemble,  CD.,  iv.  293.  From  A.S.  luorth,  an 
enclosed  homestead ;  and  (probably)  haniut,  an  enclosure  ;  the 
compound  having  the  sense  of  '  farm-enclosure.'  See  Bosworth 
and  Toller's  A.S.  Diet.,  p.  1267. 

Wrentham.  To  the  N.  of  Southwold.  Spelt  Wrentham, 
Ipm.;  H.R.;  Wretham  (probably  an  error  for  Wretham=  Wrent- 
ham), D.B.,  p.  237.  Copinger  also  gives  the  form  Wrantham. 
Not  A.S.,  but  Friesic.  Koolman  gives  the  E.  Friesic  wranten, 
to  grumble,  and  ivrante,  sb.,  a  grumbler.  Hexham's  Mid.  Dutch 
Diet,  has  tvranten,  to  wrangle,  to  quarrel,  or  to  chide ;  and  wrant, 
a  wrangling,  or  a  quarrelsome  man.  Outzen  gives  the  N.  Fries. 
wrante,  to  whimper ;  cf  Dan.  vrante,  to  be  peevish.  From  the 
base  wrant  would  be  formed  an  A.S.  ^wrantian,  *wrentan,  to 
grumble ;  and  hence  *iurenta,  a  grumbler,  which  could  be  used 
as  a  nick-name.  Hence  Wrentham  would  mean  '  Wrenta's 
home '  or  '  Wrenta's  enclosure.' 

24.     Haugh. 

The  mod.  E.  haugh  is  from  the  O.  Merc,  halh,  A.S.  healh, 
a  sheltered  place,  hence,  low-lying  land  beside  a  stream;  as  has 
already  been  explained  under  Hale,  which  is,  grammatically, 
the  dat.  case  of  haugh.     The  sole  example  is  Pettaugh. 

Pettaugh.  To  the  S.  of  Debenham.  Spelt  Pethage,  T.N.; 
Pethagh,  Ipm. ;  Petehaga,  D.B.,  p.  194  ;  Pettehaga,  D.B.,  p.  320. 
Copinger  also  notes  the  forms  Pethaugh  and  Pethale,  which  are 
important  as  showing  that  we  are  here  dealing  with  haugh  and 

C.  A.  S.  Octavo  Series.     No.  XLVI.  o 


66  •  THE   PLACE-NAMES   OF   SUFFOLK 

hale,  not  with  hem,  a  hedge;  though  the  spelling  Fethaiue 
occurs  also.  The  forms  in  D.B.  show  four  syllables,  so  that 
the  word  is  not  a  mere  compound  with  pet,  'a  pit.'  The  prefix 
Pete-  represents  the  A.S.  Peotan,  gen.  of  Peota,  which  occurs 
in  a  Worcester  charter  dated  851,  as  the  name  of  a  witness  ; 
see  Birch,  C.S.,  ii.  56.  I  explain  Pettaugh  as  '  Peota's  haugh ' ; 
or  rather  (with  a  simple  vowel)  as  '  Peta's  haugh.'  And  see 
Pettistree. 

25.  Heath. 
The  mod.  E.  heath  appears  in  A.S.  as  hwth.  There  are 
three  places  that  end  in  -heath  at  the  present  day,  but  in  only 
one  of  these,  viz.  Leavenheath,  is  the  suffix  original.  The  other 
two  are  Horningsheath  and  Lakenheath.  But  I  shall  take  all 
three  together,  for  practical  convenience. 

Horningsheath.  Near  Bury.  Often  called  Horringer, 
and  marked  as  Horningsheath  or  Horringer  on  Bacon's  map. 
This  remarkable  variation  is  due  to  the  fact  that  the  suffix 
-heath  was  substituted,  at  an  early  date,  for  another  suffix  that 
contained  -er-,  which  is  even  now  not  quite  forgotten.  Spelt 
Horning esherth,  H.R.;  Ipm.  Copinger  records  numerous  forms, 
but  in  all  of  them  the  prefix  is  equivalent  to  Horninges  ;  whilst 
the  old  suffix  appears  as  herih  (once  harth),  erth,  herde,  herd, 
erda,  erd ;  and  (by  mistake)  as  worda.  D.B.  has  Horningesworda, 
p.  152,  and  Horningeserda,  p.  222,  with  the  usual  Norman 
neglect  of  initial  h,  and  change  of  th  to  d.  The  suffix  repre- 
sented by  the  Middle  English  herth  (whence  the  other  forms 
result)  is  the  A.S.  lieorth,  mod.  E.  hearth,  which  was  sometimes, 
though  rarely,  used  to  denote  'a  dwelling,'  or  'house';  see  the 
A.S.  Diet.  The  original  sense  was  '  Horning's  hearth,'  which 
was  afterwards  turned  into  '  Horning's  heath,'  Horning  means 
'son  of  Horn,'  which  is  a  known  personal  name.  The  form 
'  Horningges  ha^S,'  i.e.  Horning's  heath,  occurs  in  a  late  copy  of 
Bp  Theodred's  Will ;  in  Thorpe,  Diplomat.,  p.  514.  But  this 
copy  abounds  in  quite  late  spellings,  so  that  it  proves  but  little. 

Lakenheath.  Spelt  Lakingheth,  H.R.;  Lakinghethe,  H.R.; 
Lakenham  (by  some  mistake),  T.N.;  Lakingahethe,  D.B.,  p.  223  ; 


NAMES    ENDING    IN    -HEATH,   -HILL     -HITHE  67 

Lal'inrjh lithe,  Ipm.,  p.  24 ;  Lakinghith,  Ipm.,  p.  221.  Also 
Lacinga-hi^,  Birch,  C.S.,  ii.  567  ;  Lakinghe^e,  Kemble,  CD,, 
iv.  18,  1.  7 ;  Lakijnge-hJ^e,  Thorpe,  Diplomat.,  p.  308,  1.  4. 
The  spellings  huthe,  hythe  are  absolutely  inconsistent  with 
'  heath,'  and  can  only  represent  the  A.S.  hyth  (dat.  hythe),  a 
hithe,  a  landing-place.  The  prefix  Lacinga  is  the  gen.  pi. 
of  Lacing,  which  is  not  a  patronymic  (as  in  the  case  of 
Lockinge,  Berks.,  where  the  form  is  Lacing,  with  long  a), 
but  a  derivative  from  the  A.S.  lacu,  a  stream ;  see  Earle,  Land 
Charters,  p.  465.  Lacing  meant  '  a  stream-dweller,'  or  dweller 
beside  a  stream,  viz.  a  southern  affluent  of  the  Little  Ouse, 
at  a  time  before  the  present  lodes  or  drains  were  made.  The 
sense  is  '  hithe  of  the  dwellers  beside  the  stream.'  Lakenheath 
had  much  more  water  near  it  in  early  days  than  it  has  now. 
At  a  later  period,  the  suffix  hithe  was  exchanged  for  heath. 

Leavenheath.  To  the  N.W.  of  Nayland.  Copinger  gives, 
without  references,  the  old  spellings  Levenesheath,  Levenes- 
heth,  Leveney.  Here  Levenes  is  certainly  a  late  form  of 
Leo/wines,  gen.  of  Leofiuine,  an  extremely  common  name 
(whence  our  modern  Levin,  Lewin,  &c.).  Kemble  has  Leof- 
wines  dene,  Cod.  Dipl.  iv.  68.     The  sense  is  '  Leofwine's  heath.' 

26.     Hill. 

A  well-known  suffix.     The  A.S.  form  was  hylle,  whence  the 
varying  Mid.  English  forms  hille,  helle,  hulle,  hylle. 
There  is  but  one  example,  viz.  Haverhill. 

Haverhill.  Spelt  Hauerhill,  H.R. ;  T.N. ;  Haverhulle, 
Ipm. ;  Haverhille,  Ipm. ;  Hauerhella,  D.B.,  p.  185.  Not  from 
haver,  '  oats,'  because  that  word  is  unknown  in  English  till 
after  1300;  but  from  A.S.  hoifer,  a  he-goat.  The  sense  is 
'  goat  hill.' 

27.     Hithe. 

From  the  A.S.  hy^  (gen.  and  dat.  hy^e),  a  hithe,  a  landing- 
place,  a  haven.  The  only  apparent  example  is  Covehithe  ;  but 
Lakenheath  was  once  Lakenhithe,  as  shown  above. 

5—2 


68  THE    PLACE-NAMES    OF    SUFFOLK 

COVEHITHE.  Near  the  coast,  to  the  N.  of  Southwold.  The 
village  is  a  little  way  inland.  The  hithe  must  have  been  where 
Covehithe  Broad  now  appears  on  the  ordnance  map.  The  sense 
is  '  cove-hithe,'  or  '  cove-landing-place.'     See  Cove. 

According  to  Copinger,  Covehithe  is  the  place  sometimes 
alluded  to  by  the  name  of  North  Hales.  North  refers  to  its 
position  as  regards  South  Cove,  which  is  not  far  off.  Hales 
is  an  old  word,  now  obsolete,  in  use  in  English  from  1330  to 
1606,  meaning  ' tents,  booths,  huts,  or  temporary  structures'; 
from  the  Old  French  hale,  mod.  F.  halle,  a  covered  market- 
place ;  from  the  0.  High  German  halla,  which  is  cognate  with 
E,  hall.  It  occurs  in  D.B.,  p.  15,  ill  spelt  as  Northals,  as  a 
solitary  example  of  the  use  of  the  O.  Fr.  hales,  long  before  its 
general  introduction  into  English. 

28.     Hoe,  or  Hoo. 

The  A.S.  hoh  signities  '  the  heel,'  and  is  the  parent  of  the 
modern  E.  hough  and  hock;  but  it  is  also  common  in  place- 
names,  with  the  sense  of  '  spur  of  a  hill '  or  '  projection  on  a 
hill-side.'  With  loss  of  the  final  li,  and  change  of  the  A.S.  o 
into  00  (as  in  A.S.  col,  mod.  E.  cool),  it  has  become  Hoo.  Some- 
times it  appears  as  Hoe  or  Ho,  which  preserves  an  older 
pronunciation.  It  appears  in  the  name  Hoo ;  and  as  a  suffix 
in  Culpho,  Dallinghoo,  and  Wixoe.  The  suffix  in  Thingoe  is 
different. 

Hoo.  To  the  S.W.  of  Framlingham.  Spelt  Hoe,  T.N. ; 
Hou,  R.B.;  D.B.,  p.  74;  Hov  (for  Hou),  D.B.,  p.  215.  The 
spellings  Ho,  Hor/h,  Hohg,  occur  in  Kemble's  index,  with 
reference  to  Hoo  in  Kent ;  and  Hohg,  with  reference  to  Hoo, 
Suff.  From  the  A.S.  hoh,  'spur  of  a  hill,'  as  said  above.  The 
hill-spur,  above  the  river  Deben,  is  clearly  shown  in  the 
ordnance  map  by  the  contour-line  marked   100. 

Culpho.  To  the  N.E.  of  Ipswich.  Spelt  Culpho,  H.R. ; 
T.N. ;  Culphowe,  R.B.  Also  Culfho,  H.R. ;  R.B. ;  Culfo,  T.N. ; 
Quilfo,  T.N.  D.B.  has  Culfole,  Gulfola,  p.  131 ;  where  perhaps 
the  addition  means  '  lea.'     The  suffix  is  plainly  the  A.S.  hoh, 


NAMES    ENDING    TX    -UOE   OR   -HOO  Gf) 

'  spur  of  a  hill ' ;  it  is  on  high  ground.  The  original  prefix 
seems  to  have  been  Gulf;  and  Gulfo  (for  Gidf-hd)  was  respelt, 
with  a  Norman  p/<  for  /.  Gimlf  is  noticed  by  Searle  as  oc- 
curring in  Ellis's  lists  of  land-owners  and  tenants  in  D.B.,  so 
that  it  is  a  late  form,  and  a  manifest  contraction  for  the  very 
common  name  Guthwulf,  also  Guthulf,  in  which  the  th  would 
readily  disappear.  The  gen.  -es  is  also  lost,  as  is  not  un- 
common in  early  names.  There  can  hardly  be  a  doubt  that 
the  name  meant  '  Ciithwulf's  hoh,'  or  a  hill-spur  named  after  a 
Cuthwulf. 

Dallinghoo.  Between  Ipswich  and  Framlinghf\m.  Spelt 
Dallingahou,  D.B.,  p.  27 ;  Dalingahou,  D.B.,  p.  90  ;  Daliiigeho, 
in  the  Crawford  Charters,  ed.  Napier  and  Stevenson,  p.  33,  1.  2. 
For  A.S.  Dalinga  hoh,  '  hill-spur  of  the  Dalings.'  I  understand 
Dalings  to  mean  dal-ings,  or  dalemen,  or  '  dwellers  in  a  dale  ' ; 
just  as  Centingas  means  Kent-ings,  or  men  of  Kent.  We  may 
suppose  them  to  have  moved  uphill  out  of  a  valley. 

Thingoe.  Not  the  name  of  an  existing  village,  but  of  a 
hundred  ;  a  fact  which  suits  the  name.  Spelt  Thinglioive,  H.R, 
D.B.  has  TJiingehov,  p.  221  ;  Tinchou,  p.  18  ;  and  Thingohov, 
p.  202  (with  needless  repetition).  Also  Thinghowe;  in  Thorpe, 
Diplomat,  p.  418. 

The  suffix  has  evidently  been  assimilated  to  that  in  Culpho, 
Dallinghoo,  and  Wixoe.  But  the  spelling  hoiue  (in  H.R.  and 
Thorpe)  suggests  that  it  was  once  the  prov.  E.  how,  howe,  a 
small  detached  hill  or  mound,  from  the  Icel.  haugr,  a  mound. 
The  prefix  Thing,  from  the  A.S.  thing,  0.  Norse  thing,  meant 
'  an  assembly,  a  meeting  for  consultation  or  deliberation,'  a 
meeting  of  the  men  of  the  hundred ;  so  that  the  sense  is 
'  meeting-mound.'  In  Gage's  Suffolk,  p.  x,  it  is  said  that  the 
Thing-how  was  an  'aceruus'  or  artificial  mound  just  outside 
the  North  gate  of  Bur3^ 

Wixoe,  or  Whixoe.  On  the  Stour,  to  the  S.E.  of  Haver- 
hill. Spelt  Wixoe  in  Bacon's  map,  but  Whixoe  in  his  index. 
Spelt  F^Yes^^eoM(!),  D.B.,  p.  267;  Wydekesho,  R.R.;  T.N.  The 
prefix  answers  to  the  A.S.  Hvnttuces,  which  occurs  in  Hiuit- 
tuces   hlcewe;    in    Birch,    C.S.,    iii.    70,    1.    11.      Hwittuc   is   a 


70  THE    PLACE-NAMES   OF   SUFFOLK 

diminutive    from    HwUa,  lit.  '  white  one,'    from    hivit,  white. 

The   sense   is  '  Hwittuc's   hill-spur.'  The   shape   of  the    hill 

can  be  traced  on  the  ordnance  map  by  help  of  the  contour- 
line  marked  200. 

29.    Holt. 

The  A.S.  and  mod.  E.  holt  means  a  small  wood  or  copse. 
It  occurs  in  Bergholt,  Occold,  Ramsholt,  and  Southolt. 

East  Bergholt.  Near  the  mouth  of  the  Stour.  Spelt 
Berkholt,  H.R. ;  Bergholte,  Ipm.;  Bercolt,  T.N. ;  D.B.,  p.  13. 
Copinger  gives  five  other  forms  beginning  with  Berc-  or  Berk- ; 
and  one  example  of  Bircholt.  The  prefix  is  certainly  the  O. 
Merc,  here,  A.S.  heorc,  a  birch ;  and  the  sense  is  '  birch-copse.' 
The  modern  form  is  not  a  happy  one. 

Occold.  Near  Eye.  Formerly  Occolt,  which  is  a  better 
spelling.  Spelt  Acolt,  D.B.,  p.  10;  for  Acholt.  Spelt  Acholt, 
in  Kemble,  CD.,  iv.  245 ;  for  A.S.  dc-holt ;  from  ac,  an  oak. 
The  sense  is  '  oak-copse.' 

Ramsholt.  S.S.E.  of  Woodbridge.  Spelt  Ramisholt,  Ipm.; 
Rammesholt,  D.B.,  p.  76.  Apparently  '  Ram's  holt ' ;  Ram  or 
Ramm  might  be  a  personal  name.  More  probably,  as  in 
Ramsey,  it  stands  for  the  A.S.  hrcem,  variant  of  hrcemn,  a 
raven ;  and  the  sense  was  '  Raven's  holt.'  Raven  could  also 
be  a  personal  name. 

Southolt.     The  sense  is  '  south  copse.' 

30.     Hurst. 

A  hu7'st  means  '  a  wooded  eminence,'  or  '  a  small  wood.'  It 
occurs,  much  disguised,  in  Hartest.     The  A.S.  form  is  hyrst 

Hartest.  N.N.W.  of  Long  Melford.  Spelt  Hertherste, 
Ipm. ;  Hertherst,  T.N. ;  Herterst,  D.B.,  p.  224  ;  ill  spelt  Hertest, 
D.B.,  p.  203;  and  in  a  late  copy  of  a  charter,  in  Kemble,  CD., 
iv.  245.  Also  Harthurst,  Herthurst  (Copinger).  The  sense  is 
'  harthurst.' 


NAMES   ENDING   IN    -lN(i  71 

31.      -ING. 

In  the  case  of  Ingham,  the  prefix  is  Norse,  and  means  '  a 
meadow ' ;  but  there  is  no  other  example  of  it  in  Suffolk.  As 
a  suffix,  it  occurs,  in  this  county,  only  as  a  patronymic,  or  with 
the  signification  of  'dweller  in'  or  'dweller  near';  for  which 
see  Blything.  In  the  plural,  it  refers  to  a  tribe  or  family. 
It  occurs  in  Ash  Booking,  Barking,  Bealings,  Blything,  Cowl- 
inge.  Greeting,  Exning,  Gedding,  Gipping,  Milden,  Nedging, 
Shimpling,  Sweffling,  Thredling,  Wratting. 

ASHBOCKING  or  Ash  Booking.  E.  of  Needham  Market. 
The  prefix  Ash  merely  means  '  ash-tree ' ;  but  the  place  was 
often  called  Ash  simply,  in  olden  times ;  and  D.B.  has  Essa, 
p.  9 ;  which  is  a  Latinised  form  of  Esse ;  and  Esse  is  the 
Norman  spelling  of  A.S.  cesce,  dat.  of  cesc,  an  ash.  It  was 
afterwards  called  Ash  Booking,  because  it  was  in  the  possession 
of  the  Booking  family  for  some  centuries  (Kelly).  This  family 
had  its  name  from  Booking  in  Essex;  called  'aet  Boccinge,'  in 
Thorpe,  Diplomat.,  p.  539.  And  Boccing  here  meant,  originally, 
'  a  son  of  Bocc'  Bocc  occurs  in  Bocches-hale  (for  Bocceshale), 
in  Thorpe,  Diplomat.,  p.  583,  note  3. 

Barking.  The  same  as  Barking  in  Essex  ;  which  is  spelt 
Berking  in  Ipm.,  p.  100.  The  Suffolk  Barking  was  also  for- 
merly Berking  (Copinger).  Spelt  Berchingas,  nom,  pi.,  in  D.B., 
p.  204 ;  Beorcingan,  dat.  pi.  (for  Beorcingum),  in  Birch,  CD., 
iii.  602.  As  the  A.S.  beoix  means  '  birch-tree,'  and  does  not 
appear  as  a  proper  name,  we  may  explain  Beorcingas,  nom.  pi., 
to  mean  'dwellers  beside  a  birch-tree  or  birch-grove.'  Note 
that  Berking  is  from  the  O.  Merc,  form  berc  rather  than  the 
A.S.  beorc. 

Bealings.  Great  Bealings  and  Little  Bealings  are  near 
Woodbridge.  Spelt  Beling,  T.N. ;  Belinges,  H.R. ;  D.B.,  p.  70  ; 
Belings,  Ipm.  The  name  Beola  is  noted  by  Searle  as  being 
that  of  a  moneyer,  in  the  time  of  Cnut,  but  I  suspect  that  the 
name  was  Norse ;  from  Bele  or  Beli,  a  name  which  occurs 
several  times  in  the  Edda.  In  any  case,  we  may  explain 
Belinges  (for  A.S.  *Belingas  or  *Beolingas)  as  a  tribal  name. 


72  THE  PLACE-NAMES   OF   SUFFOLK 

Blything.  The  name  of  a  hundred.  Spelt  Blidinga  (with 
d  for  S),  D.B.,  p.  3.  Blything  is  a  hundred  on  the  E.  coast, 
containing  Southwold  and  Dunwich,  and  the  river  Blythe  flows 
through  tlie  midst  of  it.  The  river-name  (meaning  'blithe'  or 
'pleasant')  is  an  old  one,  and  another  river  of  the  same  name 
is  mentioned  in  a  Northants.  charter,  dated  944 ;  see  Earle, 
Land  Charters,  p.  179,  1.  3.  I  understand  Blithinga,  gen.  pi., 
to  be  the  gen.  of  Blithingas,  or  '  the  dwellers  beside  the  Blythe.' 

CowLiNGE.  N.N.E.  of  Haverhill.  Spelt  Culing,  B..U.:  Gul- 
inges,  T.N.,  Ipm.;  Cvlinge,  D.B.,  p.  24.  It  evidently  represents 
an  A.S.  Culinga,  gen.  pi.  (as  in  Culinga  gemsere,  Birch,  C.S., 
i.  318);  and  '  Gules  feUV  occurs  in  a  Hants,  charter,  dated  909, 
in  Birch,  C.S.,  ii.  284.  Hence  the  Culingas  were  'the  sons  of 
Cril.'  It  seems  to  be  not  quite  the  same  name  as  Cowling 
in  the  W.  Riding  of  Yks.,  which  refers  to  '  the  sons  of  Coll ' ; 
as  Prof.  Moorman  shows.  The  form  Gidinges  in  T.N.  repre- 
sents the  nom.  pi.  Ciilingas. 

Creeting.  West  Creeting  and  Creeting  St  Mary  lie  to 
the  N.  of  Needham  Market.  Spelt  Gretinges,  T.N. ;  Greting, 
H.R. ;  Gretinge,  R.B.,  Ipm.  But  D.B.  has  Gratingas,  p.  22 ; 
Gratinga,  p.  47  ;  in  Gratingis,  p.  48.  Probably  fiom  the  A.S. 
Gretta,  a  name  of  which  two  instances  are  known.  Creeting 
may  represent  a  gen.  pi.  Grettinga ;  and  the  nora.  Grettingas 
means  '  the  sons  of  Cretta,'  or  '  the  family  of  Cretta.' 

ExNiNG.  To  the  N.W.  of  Newmarket.  The  spelling  I wning 
is  commoner,  at  an  early  date ;  and  the  place  was  once  con- 
sidered to  be  in  Cambridgeshire,  as  its  position  suggests. 
Spelt  Ixninge,  locnynge,  and  even  Ixningliam,  Ipm. ;  Ixninges, 
R.B.;  Yxninge,  R.B.;  Ixnwg  and  Exning,  H.R.;  Ixning,  Yxning, 
T.N.  No  doubt  the  prefix  is  related  to  that  of  Ixworth,  which 
is  also  in  Suffolk  ;  and  the  latter  is  known.  Ixworth  appears 
in  a  very  late  charter  (no.  1019)  as  Ixeworthe ;  and  in  another 
(no.  1018)  as  Gyxeweorde  (with  d  for  tS);  both  in  Birch,  iii.  219. 
But  Gyxe-  stands  for  Gixe-,  with  i,  not  y;  because  before  y  a  G 
remains  hard,  whilst  before  i  it  disappears.  And  Gixe  is  a  very 
late  form  of  the  true  gen.  Gixan,  from  a  nom.  Gixa.     It  is  now 


NAMES    ENDING    IN    -ING  78 

clear  that  Ixning  was  a  later  form  of  Gixan-ing',  or  rather  of 
the  gen.  pi.  Gixaninga;  which  meant  'of  or  belonging  to  the 
Gixanings '  or  *  of  the  sons  (or  family)  of  Gixa.'  Note  parti- 
ciihirly  the  spelling  Ixenyng  in  Ipm.  (Index  Nominum).  The 
longer  form  Ixningham  is  quite  legitimate;  it  means  'home  (or 
enclosure)  of  the  Gixanings.'     Gixa  =  Gisca  ;  see  IxwoRTH. 

Gedding.  Spelt  Geddinge,  Geddinges,  R.B.  But  also 
known  as  Giddiug,  in  which  form  it  appears  in  Pigot's  Atlas, 
1831.  Spelt  Oedinga,  D.B.,  p.  235  ;  Geldinga  (error  for  Gedd- 
inga),  D.B.,  p.  165.  In  Thorpe,  Diplomat.,  p.  526,  Giddincg- 
forda  (dat.)  is  mentioned  in  connexion  with  Kersey  and  Hadley, 
and  must  refer  to  Gedding.  Note  that  in  Ipm.,  p.  48,  Geddinge 
refers  to  Gedding  (Suff.)  ;  whilst  in  the  same,  p.  25,  Gedding 
refers  to  Gidding  (Hunts.).  The  phonology  is  not  difficult. 
Both  these  places  really  had  once  the  same  name,  and  this 
name  must  have  begun  with  an  A.S.  Gy,  as  otherwise  the  G 
would  not  have  remained  hard.  This  Gy  is  often  spelt  Gi 
in  later  times,  and  it  often  iippears  as  Ge  (with  hard  g) 
dialecticall}^  It  is  therefore  certain  that  Gedding  once  referred 
to  a  settlement  of  '  the  sons  of  Gydda.'  The  personal  name 
Gydda  occurs  in  Gyddan-den ;  in  Kemble,  CD.,  v.  289. 

GiPPiNG.  About  2^  miles  E.  by  N.  from  Haughley  railway 
station  ;  a  small  hamlet.  I  find  no  old  spelling,  but  Copinger 
records  Gipping,  Gypping,  Gippyngge,  and  Gyppinges ;  all  un- 
dated. There  is  also  a  river  Gipping,  and  I  have  seen  it 
suggested  that  Ipswich  (formerly  Gippes  wic)  took  its  name 
from  the  river!  But  I  presume  that  the  G  in  Gipping  is  hard, 
and  therefore  wholly  unconnected  with  the  A.S.  Gippes,  in 
which  the  G  was  sounded  like  the  y  in  yield.  The  map  shows 
that  the  river  Gipping,  before  it  joins  another  stream  above 
Stowmarket,  comes  down  from  the  direction  of  Gipping,  and 
whilst  still  small,  flows  past  the  end  of  Gipping  Great  Wood. 
It  is  therefore  fairly  certain  that  the  river  took  its  name  from 
the  place,  and  not  otherwise ;  just  as  the  Deben  comes  from 
the  neighbourhood  of  Deben  ham.  I  suppose  that  the  oldest 
spelling  was  Gypping,  and  that  it  represents  a  tribal  name, 
from   a  personal  name    Gypp  or  Gyppa  ;    but  of  such   names 


74  THE    PLACE-NAMES   OF   SUFFOLK 

no  trace  seems  to  exist,  unless  we  can  compare  Geppa  (twice) 
in  Searle's  list. 

MiLDEN.  Called  Milding*  by  Kirby  (1764).  S.W.  of 
Bildeston.  This  must  be  included  among  the  names  in  -ing, 
on  account  of  the  older  forms.  Spelt  Meldinge,  Ipm.,  p.  198 ; 
Mellinga,  D.B.,  159  ;  Mildinges,  F.A.,  v.  43.  I  find  no  other 
old  spelling ;  but  Copinger  recites  some  seventeen,  of  which 
all  but  two  end  in  -ing,  -ingg,  -inge,  -ingge,  or  -yng,  -yngg, 
-inga.  Eleven  of  them  begin  with  Me- ;  so  that  the  oldest 
type  seems  to  be  Meldinga,  which  we  may  associate  with  the 
personal  name  Melda,  whence  are  derived  the  names  Meldreth 
and  Melbourn,  both  in  Cambs.  The  gen.  pi.  Meldinga  refers 
to  a  settlement  of '  the  sons  (or  family)  of  Melda.' 

Nedging.  Near  Bildeston.  Spelt  Nedding,  H.R. ;  Nedd- 
inge,  R.B. ;  so  that  the  sound  of  dg  is  unoriginal.  D.B.  has 
Niedinga,  p.  209.  In  a  Suffolk  charter  relating  particularly 
to  Cockfield,  in  Birch,  C.S.,  iii.  603,  1.  3,  we  find  "  cet  {H)nydd- 
inge,"  which  Kemble  and  Thorpe  explain  as  Nedging,  with 
obvious  correctness.  This  furnishes  another  instance  in  which 
the  A.S.  y  is  locally  rendered  by  e.  The  initial  H  is  printed  by 
Birch  between  two  marks,  to  show  that  it  has  been  supplied 
afterwards.  We  also  find  Neddinge  in  a  late  copy  of  a  charter 
in  Kemble,  CD.,  iv.  245  ;  it  is  of  no  great  value.  The  evidence 
shows  that  the  name  refers  to  a  settlement  of  Hnyddings  or  of 
Nyddings ;  but  there  is  nothing  to  help  us  any  further.'  In 
the  form  Neddinge,  the  final  -ge  must  have  been  palatalised, 
or  sounded  as  j,  giving  Neddinj  ;  after  which  the  J-sound  was 
passed  back  into  the  former  syllable,  and  so  it  became  Nedjing, 
or  Nedging. 

Shimpling.  To  the  W.  of  Lavenham.  There  is  another 
Shimpling  in  Norfolk.  Kelly  says  it  is  also  known  as  Shim- 
plingthorne  ;  where  '  thorne '  is  the  mod.  E.  '  thorn.'  Spelt 
Simpling,T.'M.;  Shimpling,  Ipm.;  Simplinga,  D.B.,  p.  270  (with 
Norman  8  for  A.S.  Sc) ;  Scimpling,  H.R.  The  A.S.  form  must 
have  been  Scimplinga,  or  settlement  '  of  the  Scimplings,'  mod. 
E.  Shimplings ;   i.e.  '  of  the  sons  of  Scimpel,'  mod,  E.  Shimpel ; 


NAMES    ENDING    IN    -ING,    -LAND  75 

a  name   not   recorded.     But   it   probably   meant   'jester';    cf. 
mod.  Dii.  schimpen,  to  scoff  at. 

SwEFFLiNG  or  SwEFLiNG.  Near  Saxmundham.  Spelt 
Sivifiinge,  Ipm. ;  Sueflinga,  D.B.,  p.  35  ;  Sueftlinga,  D.B,,  p.  34. 
A.S.  Sueftlinges ;  in  the  Crawford  Charters,  ed.  Napier  and 
Stevenson,  p.  33,  1.  2.  The  name  records  a  settlement  '  of 
Sweftlings,'  or  '  of  the  sons  of  *Sweftel ' ;  a  name  not  otherwise 
known.  Probably  it  was  once  *Sw}eftel ;  and  it  may  be  allied 
to  E,  siuift.     Kemble's  index  has  Swiftan-beorh. 

Thredling.  The  name  of  a  hundred.  Spelt  Tredelinge, 
Ipm.  (with  T  for  Th).  I  find  no  other  old  form;  Copinger  gives 
Thrydelingge,  Thridelingge.  There  is  no  further  clue.  The  e 
(i,  y)  may  have  been  due  to  an  A.S.  y,  as  in  other  cases.  It 
appears  to  refer  to  a  settlement  of  *Thrydelings ;  but  no  such 
form  appears.  Possibly  for  *Thrythhildings ;  since  Thrythhild 
is  a  known  female  name. 

Wratting.  Spelt  Wratting  in  T.N. ;  Wrotinge,  R.B. ; 
Wratinga,  D.B.,  p.  220.  The  same  name  as  Wratting  in 
Cambs.,  which  appears  as  A.S.  Wrcettincge  in  the  dat.  case, 
in  vElfhelm's  Will ;  see  Birch,  C.S.,  iii.  629  ;  also  Wrcattinge, 
on  the  preceding  page.  The  reference  is  to  a  tribe  of  Wrsettings 
or  to  a  man  named  Wrsetting.  A  man  may  have  been  so  named 
from  a  wart  upon  his  face ;  since  ivret  was  the  E.  Anglian  form 
of  wart  in  the  fifteenth  century  (Prompt.  Parv.).  Cf.  Du.  wrat, 
a  wart. 

32.    Land. 

Land  is  well  knoAvn  in  mod.  E.,  A.S.,  and  Old  Norse.  It 
means  tract  of  country,  region,  &c.  It  occurs  in  Kessingland, 
Lothingland,  Nayland,  Shelland,  and  Swilland. 

Kessingland.  On  the  E.  coast;  S.  of  Lowestoft.  Spelt 
Kessinglond,  H.R. ;  Ipm.  ;  Kessingland,  T.N. ;  Kessingekinda, 
D.B.,  p.  5.  But  the  A.S.  form  could  not  have  begun  with  Ce, 
or  the  C  would  have  become  Ch;  and  we  find,  in  fact,  the  alter- 
native spelling  Cassingland.  It  is  twice  spelt  Cassingland  in 
Ipm.,  p.  55 ;  and  Copinger  quotes  six  examples  in  which  the 


76  THE   PLACE-NAMES   OF   SUFFOLK 

name  begins  with  Cass-.  It  therefore  probably  has  the  same 
prefix  as  Kasing-hurna;  in  Birch,  i.  477.  Cf.  Casincg -street 
and  Gasan-thorn  in  Kemble's  index.  The  sense  is  therefore, 
in  all  probability,  'land  of  the  Casings'  or  'of  the  sons  of 
Casa.' 

LoTHiNGLAND.  The  name  of  a  hundred.  This  hundred 
contains  Lowestoft  and  Lake  Lothing  or  Lothing  Lake,  from 
which  the  hundred  took  its  name.  The  o  was  short,  and  must 
be  the  o  which  the  Normans  frequently  substituted  for  short 
u ;  cf  the  spellings  Luddingland,  Ipm. ;  Ludingland,  H.R. ; 
Luthingland,  H.R. ;  Ludingaland,  D.B.,  p.  329.  We  also  find 
Lii^inglond  in  a  late  copy  of  Bp  Theodred's  Will ;  in  Thorpe, 
Diplomat.,  p.  513.  The  full  form  was  Luthinga-land,  or  '  land 
of  the  Luthings';  but  we  have  no  further  information  as  to 
this  tribe  or  family.  The  name  may  perhaps  be  connected 
with  the  personal  name  Luda,  which  occurs  in  Lvdan-heorh ; 
in  Birch,  C.S.  iii.  204. 

Nayland.  This  is  a  very  interesting  name,  as  it  is  an 
example  of  a  spelling  in  which  a  //  has  been  prefixed.  It  is 
spelt  Eylaiid,  Eylaund,  T.N. ;  but  Neyland  in  Ipm.,  p.  16 
(A.D.  1257);  Neylond,  H.R.;  Eilanda,  D.B.,  p.  242;  rather 
from  the  O.  Norse  eyland,  an  island,  than  from  the  A.S. 
leg-land,  %g-land,  with  the  same  sense.  The  situation  of 
Nayland  is  low,  and  it  is  "subject  to  occasional  inundations"  ; 
see  The  Beauties  of  England,  xiv.  158.  Its  river  is  the  Stour.  We 
can  easily  explain  how  the  n  was  prefixed.  It  arose  from  the 
fact  that  there  are  two  places  in  S.W.  Suffolk  called  Stoke ; 
one  was  called  Stoke-juxta-Clare,  or  Stoke  near  Clare,  and  the 
other  Stoke  near  Eyland,  described  in  1327  (in  the  Index  to 
the  Charters)  as  Stoke-atte-Neilond,  of  which  an  older  form 
must  necessarily  have  been  Stoke-atten-Eilond ;  as  the  Mid. 
Eng.  atten  represents  the  A.S.  cet  thani,  i.e.  '  at  the.'  The  date 
when  the  n  was  prefixed  seems  to  have  been  somewhat  earlier 
than  A.D.  1250.     Observe  that  the  name  is  Norse. 

Shelland.  To  the  W.N.W.  of  Stowmarket.  Spelt  Shel- 
laund,  Shellonde,  Ipm. ;  Sellanda,  D.B.,  p.  224  (with  Norman  S 
for  A.S.  Sc).     I  think   the   prefix    was   not   the  A.S.  scell,  '  a 


NAMES    ENDING    IN    -LAND,    -LEY  77 

shell,'  but  see  If,  '  a.  shelf;  note  Shelton,  Beds.,  of  which  the 
A.S.  form  was  Scelf-tim,  and  Shelley,  Suff,  (below),  of  which 
the  A.S.  form  was  Scelf-leah,  affording  good  reason  for  the 
change  from  Ji  to  II.  The  sense  is  '  shelf-land,'  which  is  pre- 
cisely the  right  sense.  A  shelf  is  a  high  terrace  of  land  or 
ledge ;  and  Shelland  Green  is  more  than  200  feet  above  the 
sea-level.     Cf  Shelfhanger,  Norf  ;  Shelve,  Salop. 

SwiLLAND.  N.  of  Ipswich,  i^pelt  Swynlaund,  Ipm.;  Swine- 
londe,  R.B. ;  Suinlanda,  D.B.,  p.  291.  From  A.S.  swin,  swine; 
literally  '  swine-land.' 

33.     Ley. 

Ley  is  a  common  suffix  in  many  counties  and  represents 
the  A.S.  leak,  a  lea,  a  meadow  ;  the  sense  is  rather  vague.  It 
occurs  in  Badley,  Bentley,  Bradley,  Brockley,  Butley,  Cookley, 
Eleigh,  Gazeley,  Hadleigh,  Haughley,  Hemley,  Henley,  Hol- 
lesley,  Kirkley,  Oakley,  Otley,  Shelley,  Shotley,  Sotterley, 
Trimley,  Westley,  and  Yaxley  ;  twenty-two  examples. 

Badley.  Near  Needham  Market;  westward.  Spelt Badele, 
T.N. ;  Badeir,  Ipm.;  Badelea,  D.B.,  p.  9.  For  A.S.  Badan- 
leah;  cf  Badan-pyt  (Kemble);  Badan-dene  (Birch,  C.S.  i.  304). 
The  sense  is  '  Bada's  lea.' 

BENTLEY^  Spelt  i?e//e^/e^y,  T.N. ;  Benetleye,Y{.'R.,  Benetleia, 
D.B.,  p.  13.  The  AS.  form  is  Beonetleah  (Kemble).  The  A.S. 
heonet,  prov.  E.  hennet,  means  'long  coarse  grass,'  or  'bent-grass.' 
The  sense  is  '  bentgrass  lea.'     See  Bent  in  my  Etym.  Diet. 

Bradley.  A  common  name.  Spelt  Bradeley,  H.R. ; 
Bradeleye,  Bradelegh,  Ipm.;  Bradeleia,  D.B.,  p.  182.  For  A.S. 
brcidan  leage;  in  Birch,  C.S.  iii.  29,  which  is  a  weak  dative  case. 
From  A.S.  brad,  broad ;  and  leak.  The  sense  is  '  at  the  broad 
lea,'  or  simply  '  broad  lea.' 

Brockley.  Spelt  Brokley,  B7-ockele,  Ipm.,  p.  270;  Brokle, 
D.B.,  p.  138;  Broclega,  D.B,  p.  155.  Cf  A.S.  Broc  lea  ford; 
Birch,  C.S.  iii.  288,  1.  7  from  bottom.  Either  from  A.S.  hroc,  a 
badger  ;  or  from  hroc,  a  brook ;  the  long  o  in  the  latter  case 
would  be  shortened  before  cl.    There  is  nothing  to  discriminate 


78  THE    PLACE-NAMES   OF   SUFFOLK 

between  the  senses  of  '  badger  lea '  and  '  brook  lea.'  Pateley, 
in  Yorks.,  means  '  badger  lea/  because  the  Yorks.  pate  means  a 
oadger.     This  favours  the  former  explanation. 

BuTLEY.  Hence  was  named  the  Butley  river,  which  joins 
the  Ore,  and  flows  into  the  sea  near  Hollesley.  Spelt  Buttele, 
T.N.,  Butelai,  D.B.,  p.  27;  Butelea,  D.B.,  p.  94.  The  patro- 
nymic Butting  occurs  in  Buttingc-graf ;  Birch,  C.S.  i.  307,  last 
line.  The  name  Butti  (also  Botti)  is  Norse  (Rygh).  The 
sense  is  '  Butti's  lea.'  The  gen.  of  Butti  was  Butta,  which 
would  give  the  Butte-  in  T.N.,  and  the  Bute-  in  D.B. 

CooKLEY.  Near  Halesworth.  STpelt  Gokeleye,il.^.;  Gokelei, 
D.B.,  p.  106 ;  Cukeleye,  Ipm.  The  prefix  seems  to  be  the  A.S. 
Cucan,  gen.  of  Cuca;  a  name  which  appears  in  Cucanhealas 
(Birch,  C.S.  iii.  113)  and  in  Cucandun  (Birch,  C.S.  iii.  140). 
Thus  the  sense  is  '  Cuca's  lea.'  Prof.  Moorman  explains  Cook- 
ridge,  in  the  West  Riding,  in  the  same  way. 

Eleigh.  There  is  Monks'  Eleigh  (for  which  Copinger 
quotes  lUea  Monachorum),  and  Brent  Eleigh  (for  which  he 
quotes  lUea  Combusta,  and  Illegh  Ars,  where  ars  is  Norman 
for  'burnt').  Spelt  Illea  -drs,  H.R. ;  llleya,  H.R. ;  Illeleia, 
D.B.,  p.  185.  The  A.S.  form  is  Illan-leah,  of  which  the  dat. 
case  Illan-lege  occurs  in  Birch,  C.S.  iii.  602.  The  sense  is 
'  Ilia's  lea.'  But  Ilsley  (Berks.)  has  lost  initial  H,  and  stands 
for  '  Hild's  lea.' 

Gazeley.  Otherwise  Gaiesley ;  as  in  Kirby.  E.  of  New- 
market. Spelt  Gasele,  Gaisle,  Gaysle,  H.R.  ;  Gaysley,  Ipm. ; 
Geisley,  Ipm.  The  A.S.  form  would  be  *Ga?ges-leah.  The 
name  *Gseg  is  not  found,  but  can  be  inferred  from  the  patro- 
nymic Gwging,  in  Birch,  C.S.  iii.  257;  whence,  as  I  have  shown, 
is  derived  the  name  of  Ginge,  in  Berks.  The  weak  form 
*Gsega  appears  as  Gega  in  Geganden  (Kemble),  and  as  Gage 
in  Gageleah  (also  Gagenleah),  also  in  Kemble's  Index ;  also  in 
Gaydon,  Gayton  and  Gaywood  in  modern  names.  The  sense  is 
'  Gseg's  lea/  or  in  later  spelling  '  Gay's  lea.' 

Hadleigh.  Spelt  Badlega,  R.B. ;  Badleigh,  Ipm. ;  Hced- 
leage,  in   a    late    charter,  Thorpe,  Diplomat.   527;    Headlega, 


NAMES    ENDING    IN    -LEY  79 

Annals  of  St  Neot,  (juoted  in  Phnnmer's  ed.  of  the  A.S. 
Chronicle,  ii.  102;  Hetlega,  D.B.,  p.  184.  In  D.B.  the  t  stands 
for  th  ;  and  the  true  A.S.  form  appears  in  a  Worcs.  charter, 
dated  849,  as  lice^leage  (gen.)  with  reference  to  Headley  Heath 
(a  tautological  name)  in  Birch,  C.S.  ii.  40 ;  see  Duignan,  Place- 
names  of  Worcs.  The  sense  is  '  heath-lea.'  In  a  similar  way 
the  A.S.  S  has  become  t  in  Hatfield  (Herts.)  which  means 
'  heath-field.' 

Haughley.  Spelt  Haiueleye,  Ipm. ;  Hagala,  D.B.,  p.  256. 
The  A.S.  form  is  Hagan-leah,  in  a  Wore,  charter;  in  Birch,  iii. 
587,1.  11  from  bottom.  The  sense  is  'haw-lea/  or  'enclosed 
lea.'  The  old  spellings  clearly  connect  the  prefix  with  haw 
rather  than  the  mod.  E.  haugh,  O.  Merc,  halh,  A.S.  healh. 
Copinger  gives  many  old  forms,  of  which  the  most  intelligible 
are  Haghele,  Haghle,  Haghlegh,  Halley,  Haugle,  Haule, 
Hawele,  Hawelee,  Hawelege,  Hawleigh. 

Helmley.  Near  Waldringfield ;  misprinted  Henley  in 
Bacon's  map,  though  given  as  Hemley  in  the  Index.  (Henley 
is  due  N.  of  Ipswich.)  Hemley  is  short  for  Helmley,  Spelt 
Halmeleia,  D.B.,  p.  138;  Halmelega,  p.  287;  but  Helmelea, 
p.  120.  The  A.S.  form  should  be  Helman-leah;  compare 
Helmaii-hyrst  in  Kemble's  index,  p.  297,  col.  2.  The  sense  is 
'  Helma's  lea.'  Helma  is  a  pet  name  for  a  name  beginning 
with  Helm-,  such  as  Helmbeald,  Helmbeorht,  &c. ;  which  are 
numerous. 

Henley.  N.  of  Ipswich.  Spelt  Henleye,  T.N.;  Hanle, 
H.R. ;  Henleia,  D.B.,  p.  50.  There  are  several  places  of  this 
name,  answering  to  A.S.  Heanleage;  in  Birch,  C.S.  iii.  519;  of 
which  a  later  spelling  is  Henlea;  in  Birch,  C.S.  i.  64.  This 
Heanleage  is  the  dat.  case  of  Heahleah,  meaning  '  high  lea.' 

Hollesley.  Near  the  mouth  of  the  river  Aide.  Spelt 
Holesle,  H.R. ;  Holeslee,  Ipm.;  Holeslegh,  Ipm.;  Holeslea,  D.B., 
p.  78.  Lit.  '  Hoi's  lea.'  The  name  Hoi  does  not  appear  by 
itself,  but  it  occurs  in  the  derivative  Hol-ing,  or  '  son  of  Hoi,' 
in  the  A.S.  Holinga  human  ;  Kemble,  CD.  iv.  232 ;  and  in 
the  modern  names  Hollingbourn,  HoUington,  and  Holling- 
worth. 


80  ■  THE    PLACE-NAMES    OF   SUFFOLK 

KiRKLEY  ;  forming  part  of  Lowestoft.  Spelt  Kyrkele,  H.R. ; 
Kirkelea,  D.B.,  p.  5.  The  prefix  is  Norse ;  from  O.N.  kirkja,  a 
church.     The  sense  is  'kirk-lea';  i.e.  'chnrch-lea.' 

Oakley.  To  the  N.  of  Eye.  Spelt  Acle,  R.B. ;  D.B., 
p.  180.  From  A  S.  dc,  an  oak ;  and  leak.  Lit.  '  oak  lea.' 
Written  dc-lea;  Birch,  C.S.  ii.  291. 

Otley.  Spelt  Oteleye,  Ipm.,  H.R. ;  Otteleye,  H.R. ;  Otelega, 
D.B.,  p.  133.  The  prefix  is  the  same  as  in  Otan-hyrst ;  in 
Kemble's   index.     The  sense  is  '  Ota's  lea.' 

Shelley.  Spelt  Selleye,  H.R. ;  Selflega,  T.N. ;  Sceiieleia, 
D.B.,  pp.  13,  14.  The  A.S.  dat.  case  is  Scelfleage ;  Thorpe, 
Diplomat.,  p.  52.5.  The  sense  is 'shelf  lea.'  /S/ie//" may  answer 
to  the  A.S.  scylf,  a  crag,  a  rock,  a  tor;  but  the  prov.  E.  shelf  also 
means  a  shoal,  a  ford  ;  or  sJielf  may  mean  a  high  terrace  or  ledge, 

Shotley.  Spelt  Schottele,  T.N. ;  Scoteleia,  D.B.,  p.  13. 
Apparently  trisyllabic ;  compare  Scotehu,  Scotta  ^JceS,  Scotta- 
7'%^,  in  Kemble's  index.  The  sense  is  doubtful.  I  incline  to 
consider  it  as  containing  the  A.S.  scot,  a  building,  as  in  the 
compound  sele-scot;  and  to  look  upon  the  original  form  as 
having  been  scota  leak,  '  lea  of  huts '  or  of  small  buildings. 
Of.  the  M.Du.  scJiot,  'a  closure  of  boards,'  in  Hexham.  See 
Scot  in  the  A.S.  Dictionary.  The  word  sele-scot  is  rendered 
tahernaculum  in  Matt.  xvii.  4  (Old  Mercian  Version)  and  in  the 
Vespasian  Psalter  Ps.  xiv.  1. 

Sotterley.  Spelt  Soterle,  H.R. ;  T.N. ;  Ipm.,  p.  249 ; 
Soterleghe,  R.B. ;  Soterlega,  D.B,,  p.  41.  The  same  old  spelling 
of  the  prefix  occurs  in  Sotet'ton,  Ipm.,  p.  203,  which  represents 
Sutterton,  Lines.;  so  that  Sotterley  might  have  become  Sutterley. 
The  meaning  of  the  prefix  in  Sutterton  is  easily  ascertained ; 
since  v^^e  find  the  spelling  Sutterton  in  Birch,  C.S.  ii.  53,  but 
Sutherton  in  the  same,  ii.  137.  It  thus  appears  that  Sotter- 
has  the  same  sense  as  the  Souther-  in  Southerton,  q.v.  The 
sense  is  'lea  more  to  the  south';  possibly  because  it  is  to  the 
South  of  the  Hundred  River,  but  a  mile  away  from  it. 

N.B.  Bjorkman  notes  a  Norse  name  Sdti,  O.  Danish  Soti, 
but  the  gen.  is  Sota,  not  Sotar ;  so  that  it  will  not  account  for 
the  prefix  Setter- ;  and  still  less  for  Sutter-. 


NAMES   ENDING   IN   -LEY,    -LOW  81 

Trimley.  Certainly  for  Tremley,  with  im  for  em,  a  common 
chaDge.  Spelt  Treinlye,  Ipm.;  Tremlega,  D.B.,  p.  124;  Treinelaia, 
D.B.,  p.  286.  Copinger  also  records  the  forms  Tremley,  Tremleye, 
Tremele,  Tremeley.  The  difficulty  is  to  know  whether  the  middle 
e  in  the  last  three  forms  is  significant.  Perhaps  it  is  best  so  to 
consider  it,  as  it  is  hard  to  see  why  it  should  have  been  inserted. 
I  can  only  conjecture  that  it  represents  a  form  Treman,  gen.  of 
Trema,  which  is  a  variant  of  Tryma,  with  the  Suffolk  e  for  the 
A.S.  y,  as  in  a  few  other  cases.  Though  Tryma  does  not  occur, 
it  is  easily  associated,  as  an  agential  masc.  in  -a,  with  trymian, 
trymman,  also  found  as  tremman,  to  confirm,  strengthen,  set  in 
order;  whence  it  might  well  become  a  name,  as  signifying  'one 
who  strengthens.'  The  base  Trum-  appears  in  such  names  as 
Trum-beorht,  Trum-here,  &c.  The  A.S.  trymian  has  given  us 
the  modern  E.  to  trim.  Trimley  can  hardly  mean  'trim  lea,' 
because  this  adjective  is  comparatively  modern,  and  due  to  the 
old  verb.  Id  fact,  the  A.S.  adj.  is  not  try^n,  but  trum,  i.e. 
strong,  firm,  excellent.  The  sense  probably  would  be  '  Tryma's 
lea.' 

Westley.  Spelt  Westlega,  R.B. ;  Westle,  H.R. ;  Westlea, 
D.B.,  p.  156.  Meaning  '  west  lea.'  There  is  another  Westley 
in  Cambs. 

Yaxley.  Spelt  Yakesley,  Ipm.;  lacheslea,  D.B.,  p.  201. 
There  is  another  Yaxley  in  Hunts.  The  A.S.  form  is  Geacesled ; 
in  Kemble,  CD.,  v.  342 ;  from  geaces,  gen.  of  geac,  a  cuckoo. 
The  sense  is  '  cuckoo's  lea.'  Fully  discussed  in  my  Place- 
names  of  Hunts. 

34.     Low. 

The  suffix  -loiv  is  not  uncommon ;  it  represents  the  A.S. 
hldw,  a  mound ;  sometimes  a  burial  mound  or  barrow.  The 
only  Suffolk  example  is  Thurlow. 

Thurlow.  Spelt  Thrilloive  Magna,  Ipm. ;  i.e.  Great 
Thurlow.  There  is  also  a  Little  Thurlow.  Spelt  Trillawe, 
H.R.;  Trillowe,  H.R.  Thrillauura,  D.B.,  p.  182;  Tridlauua, 
D.B.,  p.  233;  Tritlaiua,  D.B.,  p.  11.  The  final  t,  d  in  the  D.B. 
Trit-,  Trid-,  point  to  an  A.S.  th  ;  and  the  true  initial  was  Tk 

C.  A.  S.  Octavo  Series.     No.  XLVI.  6 


82  THE    PLACE-NAMES    OF    SUFFOLK 

also.  Hence  it  represents  an  A.S.  Thrythe-hlmu ;  where 
Thryth  (gen.  Thrythe)  is  a  known  female  name.  The  sense 
is  'burial-mound  of  Thryth.' 

35.    Meadow. 

Meadow  represents  the  A.S.  mwdive,  dat.  of  inwd,  a  mead  ; 
so  that  meadow  is,  in  fact,  merely  the  dative  of  mead,  without 
any  variation  of  sense.     The  only  example  is  Shipmeadow. 

Shipmeadow.  Spelt  Scipmedu,  D.B.,  p.  41.  Copinger 
notes  the  forms  Shepmedive,  Shepmed ;  where  the  e  is  older 
than  i.  From  A.S.  sceap,  a  sheep.  The  sense  is  '  sheep- 
meadow.'  In  Shropshire,  a  sheep  is  always  a  shij) ;  and  the 
pi.  is  ships.     Cf.  Ship-ton. 

36.     Mere. 

The  term  mere  was  applied  to  a  lake  or  pool  of  any  size. 
In  some  counties,  another  mei^e  is  used  as  a  suffix  also,  with 
the  sense  of  '  boundary.'  But  I  think  this  second  suffix  does 
not  appear  in  Suffolk  names.  The  examples  are  five,  viz. 
Bosmere,  Hartismere,  Livermere,  Rushmere,  Semer. 

Bosmere.  This  is  the  name  of  a  hundred.  Spelt  Bosemere, 
H.R.  ;  Bosemera,  D.B.,  p.  9.  The  prefix  is  from  the  A.S.  Bosan, 
gen.  of  Bosa,  a  name  of  which  there  are  several  examples ;  as 
in  Bosen-hangran,  in  Birch,  C.S.  ii.  492.  The  sense  is  '  Bosa's 
mere.'  I  find,  in  the  Ordnance  map,  that  there  is  still  a  pool 
or  small  mere,  beside  the  river  Gipping,  less  than  a  mile  below 
Needham  Market.  It  is  situate  within  grounds  belonging  to  a 
hall  named  Bosmere  Hall.  This  pool  is  called  Bosmere  in  one 
of  Kirby's  maps;  it  was  probably  larger  in  days  when  little 
care  was  taken  of  the  waterways.  Moreover,  it  is  situate  very 
near  the  centre  of  Bosmere  Hundred.  In  the  Beauties  of 
England,  p.  217,  Bosmere  lake  is  called  "a  lake  of  30  or 
40  acres." 

Hartismere.  The  name  of  a  hundred.  ^^e\t  Hertesmere, 
H.R.:  T.N.;  D.B.,  p.  260;  Hertesmera,  D.B.,  p.  3.    The  O.  Mercian 


NAMES    ENDING    IN    -MKRE,    -POOL  .S3 

equivalent  is  Herotes  mere,  later  form  Hertes  mere  ;  where  Herat, 
Hert,  answer  to  the  A.S.  Ileorot,  Heart,  lit.  '  a  hart.'  But  it  is 
here  a  name.     The  sense  is  '  Hart's  mere.' 

LiVERMERE.  Sometimes  miscalled  Livermore  ;  as  in  Kirby. 
To  the  N.N.E.  of  Bury  is  Livermere  Park,  containing  a  lake  of 
considerable  length.  Spelt  Livermere,  H.R. ;  Livremere,  R.B. ; 
Liuermera,  D.B.,  p.  166.  Spelt  Leuiiremer  in  a  charter  of 
Edward  the  Confessor,  in  Kemble,  CD.  iv.  245  ;  but  the  copy 
is  a  late  one,  and  the  spelling  is  that  of  a  Norman  scribe.  The 
name  seems  to  be  a  mere  compound,  and  is  easily  explained  by 
the  A.S.  Icvfer,  lefr,  a  yellow  flag,  the  plant  Iris  Pseudacorus, 
still  called  levers  (sometimes  livers)  in  prov.  E. ;  see  the  E.D.D. 
The  sense  is  '  flag-mere.'     See  Rushmere  below. 

Perhaps  it  is  well  to  note  here  that  the  same  explanation 
may  not  apply  to  Liverpool ;  which  is  supposed  to  have  been 
'  Leofhere's  pool.'     See  Wyld,  Lanes.  Place-names. 

Rushmere,  Considered  by  Kelly  to  be  a  part  of  Ipswich. 
It  lies  to  the  N.E.  of  the  town.  Spelt  Rushemore  (with  o  for  e, 
a  common  error),  Ipm. ;  Riscemara,  D.B.,  p.  4.  The  latter 
spelling  is  explained  by  the  fact  that  the  usual  A.S.  form  of 
'  rush  '  is  Wsc  ;  whence  Rise-mere  ;  Birch,  i.  83,  1.  3.  The  sense 
is  simply  'rush-mere.' 

Semer.  On  the  river  Brett,  some  distance  above  Hadleigh. 
The  Ordnance  map  shows  a  small  pool  near  the  church.  Spelt 
>S*eme?^e,  H.R.;  this  seems  to  be  the  oldest  spelling;  Seamera, 
D.B.,  p.  176.  It  is  seldom  mentioned ;  Copinger  has  also  Soi/mer, 
Seamer.  It  appears  to  be  simply  compounded  of  the  A.S.  sre, 
a  sea,  lake,  pool,  pond ;  and  mere,  a  mere.  The  A.S.  scd  was 
applied  to  a  pool  of  any  size  as  well  as  to  the  sea ;  and  a  pool 
would  thus  be  called  simply  '  sea' ;  very  likely,  mere  was  a  later 
explanatory  addition  ;  its  sense  was  very  nearly  as  vague. 

37.     Pool. 

The  A.S.  pol,  a  pool,  sometimes  appears  as  -pale  in  the 
suffix  of  a  place-name.     It  occurs  in  Walpole.     Cf.  PoLSTEAD. 

6—2 


84  THE    PLACE-NAMES    OF    SUFFOLK 

Walpole.  To  the  S.W.  of  Halesworth.  Spelt  Walepol, 
R.B. ;  H.R. ;  Walepola,  D.B.,  p.  24.  It  appears  as  Walepol  in 
a  late  charter  with  Norman  spellings;  in  Kemble,  CD.  iv.  245; 
with  reference  to  Walpole  in  Norfolk.  The  prefix  Wale- 
re presents  the  A.S.  Weala,  gen.  pi.  of  Wealh,  a  foreigner,  a 
Briton.  The  sense  is  '  Britons'  pool ' ;  or  '  Welshmen's  pool' 
Cf.  Welshpool. 

38.     Set,  Sett. 

The  suffix  -set,  as  in  Somer-set,  represents  the  A.S.  -swte,  a 
pi.  form  signifying  settlers,  residents,  or  inhabitants.  There  is 
also  a  by-form  -sStan,  pi.,  with  the  same  sense.  It  occurs  in 
Bricett,  Elmsett,  Hessett,  Wetheringsett,  and,  etymologically, 
in  Wissett. 

Bricett.  Great  Bricett  is  to  the  S.W.  of  Needham  Market ; 
the  i  is  long,  as  in  E.  ice.  Spelt  Bresete,  Ipm. ;  H.R. ;  Brisete, 
T.N.,  Ipm.;  D.B.  has  Brieseta,  pp.  226,  273;  Bricseta,  p.  248 
(printed  Brieseta  in  the  Victoria  County  History  of  Suffolk).  It 
is  in  the  hundred  of  Bosmere,  and  we  find  in  D.B.,  p.  306,  the 
following  note,  which  seems  to  allude  to  Bricett. — "  Hundret  de 
Bosemera.  In  Brictices-haga  est  silua  qua  poterant  pasci  xvi. 
pore'."  Again,  in  D.B.,  p.  12,  is  another  note  : — "  Herchesteda: 
ten'  Harold  t[empore]  r[egis]  e[dwardi]  V.  car'  terre :  p7-o 
berewica  in  brictesceseia  in  comitatu  de  exsessa  [Essex]."  It  is 
certain  that  the  latter  note  refers  to  Brightlingsea  in  Essex, 
lit.  '  Brightling's  island';  but  instead  of  -ling's  we  here  have 
-esces,  for  -isces,  i.e.  mod.  E.  -isKs.  That  is,  instead  of  '  Bright- 
ling's  island '  it  is  here  called  '  Brightish's  island,'  which 
apparently  expresses  the  same  thing ;  the  suffixes  being- 
adjectival  and  equivalent.  In  the  same  way,  Brictices  haga 
(better  spelt  Brictisces  haga)  means  '  Brightish  haw ' ;  and  we 
may  fairly  conclude  that  Bricett  was  originally  Brictsete  in 
Norman,  or  Beorht-soete  in  A.S.  spelling.  The  A.S.  heorht  is 
not  only  an  adj.,  meaning  '  bright,'  but  also  a  neut.  sb.,  meaning 
'  brightness '  or  '  clear  light.'  I  explain  Beorkt-swte  as  '  settlers 
in  a  bright  spot.'  This  may  seem  a  somewhat  strange  formation, 
but  it  is  exemplified  and  justified  by  the  notorious  form  Burner- 


NAMES   ENDINr,   IN    -SET   01{   -SETT  85 

sd'te,  or  '  summer-settlers/  which  exists  to-day  as  Somerset, 
That  Brightset  could  pass  into  Bricett  is  sufficiently  obvious; 
but  it  would  not  be  easy  to  assign  any  other  form  which  would 
give  the  same  result.  The  intermediate  form  would  be  Brighset, 
which  (if  the  gh  be  kept  silent)  fairly  accounts  for  the  modern 
pronunciation. 

Elmsett.  Spelt  Elmesete,  T.N. ;  Elmesset,  H.R. ;  Elmeseta, 
D.B.,  p.  249.  We  find  the  gen.  pi.  Elmesetene  in  Birch,  CD., 
i.  502,  1.  11.  The  correct  nom.  pi.  of  this  is  Elmescvtan,  i.e. 
'settlers  at  the  elm,'  or  'beside  the  elm.'  Elme  is  in  the  dat. 
case,  the  prep,  wt,  '  at,'  being  understood. 

Hessett  ;  also  Hedgsett  (in  1813).  Spelt  Hegesset,  H.R. ; 
misspelt  Heteseta  (probably  for  Heceseta,  with  Norman  c  for  g), 
D.B.,  p.  149.  Copinger  also  gives  Heggesete,  Heggesett,  Hegsete, 
Hegyssete.  The  g  must  have  been  single,  because  the  M.E.  gg 
became  dg.  The  derivation  is  not  exactly  from  the  A.S.  hecg, 
'hedge,'  but  from  the  allied  form  hege,  a  'hay'  or  fence.  The 
sense  is  '  hay-settlers ' ;  where  hay  is  the  prov.  E.  liay,  '  a  hedge, 
a  fence,  a  boundary';  E.D.D.     The  A.S.  form  is  hege-swte. 

Wetheringsett.  Spelt  Wederingesete,'^.^.\  Wederingaseta, 
D.B.,  p.  179 ;  Weringheseta  (a  contracted  form),  D.B.,  p.  10. 
We  find  the  A.S.  form  Wederingesete  in  Kemble,  CD.  iv.  245 ; 
but  the  copy  is  in  late  and  Norman  spelling.  The  th  is  un- 
original, as  in  father,  mother  (A.S.  feeder ^  modor) ;  several 
words  ending  in  -der  were  altered  so  as  to  end  in  -ther  in  the 
fifteenth  century.  No  name  begins  with  Wether  in  Old 
English ;  but  the  Wederas,  or  tribe  of  Weders,  are  mentioned 
repeatedly  in  Beowulf.  They  were  a  tribe  of  Geats,  and  their 
province  was  called  Weder-mearc  or  '  Weder-mark.'  No  doubt 
the  Wederings  belonged  to  this  tribe.  The  sense  is  'settlers 
belonging  to  the  tribe  of  the  Weders.' 

WISSETT.  Spelt  Wyssete,  iTpm.;  Wisete,R.B.;  Wysete,  U.R. ; 
Wiseta,  D.B.,  p.  337  ;  Wisseta,  D.B.,  p.  25.  The  corresponding 
A.S.  form  should  be  Wi-s^te ;  and  I  take  wi  to  represent  the 
A.S.  ^u^h,  'an  idol,'  of  which  the  original  sense  was  really  'a 
heathen  temple,'  like  that  of  the  O.  Sax.  wlh,  Icel.  ve.     The 


80  THE    PLACE-NAMES   OF   SUFFOLK 

sense  is  *  settlers  beside  or  near  a  temple.'  A  corresponding 
name  Veseti  occurs  in  O.  Norse,  and  was  used  as  a  personal 
name,  though  the  original  sense  was  'settler  near  a  temple'; 
see  Bjorkman  and  Rygh.  The  name  may  have  been  merely 
borrowed  from  Norse ;  in  which  case  it  must  be  remembered 
that,  as  far  as  Suffolk  was  concerned,  Wlseti  was  merely  a 
personal  name,  the  origin  of  which  may  have  been  but  dimly 
remembered.  R-ygh  gives  no  less  than  seven  place-names  in 
Avhich  the  Norse  name  is  preserved ;  such  as  Veset-rud, 
Veset-vik. 

39.  Stall. 

This  is  the  same  word  as  the  modern  E.  stall  in  cattle-stall ; 
though  the  sense  somewhat  varies.  It  occurs  in  Burstall  and 
Tunstall. 

Burstall.  To  the  W.  of  Ipswich.  Spelt  Burstall,  H.K. ; 
Ipm. ;  Burgestala,  D.B.,  p.  189  ;  Burghestala,  D.B.,  pp.  193,  229. 
The  A.S.  form  is  hiirg-steall,  lit.  'position  for  a  fort';  not  a 
common  word.  In  the  A.S.  poem  called  '  The  Ruins,'  ed.  Grein, 
1.  29,  hrosnade  hurgsteal  means  '  the  foundation  of  the  fort  has 
crumbled  to  pieces.'  In  Wright  and  Wlilker's  Vocabularies, 
205.  30,  burhsteal  has  the  curious  sense  of  '  a  path  down  a  hill ' ; 
and  the  prov.  E.  borstaU  means  '  a  path  up  a  steep  hill ' ;  or,  in 
Kent,  'any  seat  on  the  side  of  a  hill.' 

Tunstall.  Tunstall-cum-Dunningworth  is  nearly  due  W. 
of  Aldeburgh.  Spelt  Tunstall,  H.R. ;  T.N. ;  Tonestala,  D.B., 
p.  236 ;  Tunestal,  D.B.,  p.  .53.  A.S.  tun-steall,  a  farmstead ; 
from  t€m,  a  '  town,'  i.e.  a  farm  ;  and  steall,  a  stall,  position, 
place,  stead. 

40.  Stead. 

Stead,  a  place,  position,  is  the  A.S.  stede.  It  occurs  in 
Belstead,  Boxstead,  Harkstead,  Hawstead,  Henstead,  Lin  stead, 
Nettlestead,  Polstead,  Saxstead,  Stanstead,  Whepstead,  Wher- 
stead. 

Belstead,  S.W,  of  Ipswich.  Spelt  Belstede,  H.R. ;  Bele- 
steda,  D.B.,  p.  51.     I  am  in  doubt  as  to  the  prefix ;  but  think 


NAMES    ENDING    IN    -STEAD  87 

it  may  be  Norse.  There  is  a  Norse  name  Beli,  occurring  in 
the  Edda ;  so  that  it  might  mean  '  Beli's  stead.'  The  gen.  of 
Beli  was  Belja.  Searle  gives  an  E.  name  Beola,  a  moneyer  in 
the  time  of  yEtheh-ed  II.  and  Cnut.  This  may  be  merely  the 
same  name  in  E.  spelling ;  and  Belstead  may  represent  Beolan- 
stede.     See  Bealings  (above). 

BoxsTEAD,  or  BoxTED  (Kelly).  Spelt  Boxsted,  H.R. ;  Tpm. ; 
Boxstede,  T.N. ;  R.B. ;  Boesteda,  D.B.,  pp.  138,  139.  From  A.S. 
box,  a  box-tree ;  and  stede,  a  place.  '  A  place  where  box  trees 
grow.' 

Harkstead.  Spelt  Herkested,  H.R. ;  Herkestede,  T.N., 
Herchesteda  (with  che  =  ke),  D.B.,  pp.  12,  280.  A  similar  prefix 
occurs  in  Herces-nws,  Herces-dlc,  and  Herces-get ;  all  in  Birch, 
C.S.  iii,  103.  Compare  also  Hwrices-hamm;  in  Birch,  C.S. 
ii.  298.  Here  may  be  a  shortened  form  of  Hceric.  A  likely 
sense  is  '  Hseric's  place  '  or  '  Here's  place  '  or  '  stead.'  A  weak 
form  Hereca  is  given  in  Searle,  and  is  probably  an  allied  name. 
But  all  these  forms,  Hcuric,  Here,  Hereca,  are  probably  un- 
English,  and  are  really  due  to  an  O.  Norse  Herekr,  or  Ha'7^ekr, 
explained  by  Rygh  as  a  name  which  is  only  found  in  place- 
names,  such  as  Herikstad  (sometimes  abbreviated  to  Herstad), 
which  is  obviously  a  Norse  form  of  Harkstead.  I  regard 
Harkstead,  accordingly,  as  due  to  Scandinavian  influence. 

Hawstead.  Spelt  Hausted,  H.R. ;  but  Halsteda,  D.B., 
p.  155.  Copinger  also  gives  Halstead,  Hcdsted,  Halstede,  from 
other  sources ;  so  that  an  Id  has  been  lost.  I  take  the  prefix 
to  be  the  0.  Merc,  hald,  A.S.  Jieald,  sloping ;  which  sometimes 
appears  in  place-names.  See  healdan  graf,  sloping  ditch ; 
Birch,  C.S.  ii.  382  ;  healdan  weg,  sloping  way ;  id,  524  ;  healdan 
hlince,  sloping  linch,  id.  iii.  33.  The  sense  is  '  sloping  stead '  or 
'  sloping  place.'     The  form  hald  is  also  found  in  Old  Frisian. 

Henstead.  S.W.  of  Lowestoft.  Spelt  Henestede,  H.R. ; 
D.B.,  p.  238;  Henstede,  R.B.;  Henested,  Hensted,  T.N.  I  regard 
it  as  a  parallel  formation  to  Henley ;  and  explain  it  as  repre- 
senting A.S.  a3t  tham  hean  stede,  '  at  the  high  stead '  or  position. 
It  is  not  very  high,  though  above  the  Hundred  river. 


88  THE    PLACE-NAMES   OF   SUFFOLK 

LiNSTEAD.  Spelt  Linsted,  Ipm. ;  Linestede,  D.B.,  p.  61 . 
The  e  before  the  stede  in  the  latter  form  may  be  due  to  the  st 
following,  and  need  not  be  considered.  From  A,S,  Im,  flax ; 
and  stede,  stead.  The  same  prefix  occurs  in  Linton,  Cambs. 
The  sense  is  '  flax-stead ' ;  or  place  where  flax  was  grown, 

Nettlestead.  N.W.  of  Ipswich.  Spelt  Netlested,  T.N. ; 
Netlestede,  R.B. ;  Netlestedam,  D.B.,  p.  28.  From  A.S.  netele, 
netle,  a  nettle.  The  sense  is  'nettle-place.'  There  is  another 
Nettlestead  in  Kent;  spelt  Netlestede  in  Birch,  C.S.  iii.  659, 
1.  19. 

Polstead.  S.W.  of  Hadleigh.  Spelt  Foisted,  R.B. ;  Pol- 
stede,  T.N. ;  Polesteda,  D.B.,  p.  241.  It  occurs  as  Polstede  in 
Thorpe,  Diplomat.,  p.  525.  From  A.S.  pol,  a  pool;  and  stede,  a 
stead,  place.     Lit.  '  pool-stead.' 

Saxstead.  Spelt  Saxstede,  Ipm. ;  Saxteda,  D.B.,  p.  37. 
For  O.  Merc.  Saxan  stede,  A.S.  Seaxan  stede.  Lit.  '  Saxa's 
stead.'     Saxa  is  a  known  name. 

Stanstead.  Spelt  >S'^awste(i,  Ipm. ;  H.R.;  Stanesteda,  D.B., 
p.  255.     For  A.S.  stdnstede  ;  lit.  '  stone  stead.' 

Whepstead.  Spelt  Whepstede,  T.N. ;  Ipm. ;  Huepestede, 
D.B.,  p.  152.  The  A.S.  form  Hwipstede  occurs  in  -^If helm's 
Will ;  in  Thorpe,  Diplomat.,  p.  596.  The  j)refix  must  denote  a 
personal  name  of  the  form  *Hwipa  or  *Hwepa  (gen.  *Hwipan, 
*Hwepan);  and  the  sense  must  be  '  *Hwipa's  (or  *Whepa's) 
stead.'  There  certainly  was  a  base  *hwip-,  probably  with  the 
sense  of  '  to  move  quickly '  or  '  to  bend  easily ' ;  the  A.S. 
hwip-er  means  unstable,  infirm  ;  Wright's  Vocab.  245.  25. 

Wherstead.  To  the  W.  of  the  river  Orwell,  below  Ipswich. 
Spelt  Whersted,  Ipm. ;  Wer-uesteda,  D.B.,  p.  30.  Copinger  also 
quotes  Wefniestede,  from  another  source.  The  prefix  Werue-, 
for  Wherue-,  suggests  the  A.S.  hwearf,  hwerf,  spelt  hwerf  in 
Thorpe,  Diplomat.,  p.  841,  1.  7  ;  where  it  seems  to  mean  a 
protecting  bank.  It  further  suggests  that  there  was  a  wharf, 
bank,  or  landing-place  on  the  Orwell,  near  Wherstead  ;  and  it 
is  remarkable  that  the  Ordnance  map  marks  a  "  Wharf"  not  far 


NAMES    ENDING    IN    -STOKE,    -STONE,    -STOW  89 

off,  on  the  bank  of  the  Orwell  between  Wherstead  and  Freston. 
It  seems  probable,  accordingly,  that  Wherstead  means  '  wharf- 
stead.'  The  church  is  quite  half  a  mile  from  the  river,  but  the 
Hall  is  nearer. 

41.    Stoke,  Stock. 

The  A.S.  stoc  meant,  in  the  first  instance,  a  stock  or  log ; 
but  is  evidently  used  also  in  the  sense  of  habitation  or  settle- 
ment ;  perhaps  one  protected  by  stocks  or  stakes,  and  so  fenced 
in.  It  is  safest  to  explain  it  by  '  settlement.'  The  length  of 
the  0  is  doubtful;  I  see  no  special  reason  for  supposing  that  it 
was  long  (stoc),  though  it  is  sometimes  so  marked.  The  modern 
English  stoke  will  best  answer  to  A.S.  stoce,  dative,  with  short  o; 
just  as  the  mod.  E.  broken  is  the  A.S.  brdcen. 

It  seems  only  necessary  to  add  that  there  is  a  Stoke-by- 
Clare,  i.e.  near  Clare ;  a  Stoke-by-Nayland ;  and  a  place  called 
Stoke  Ash,  to  the  S.W.  of  Eye.  The  A.S.  stoc  occurs  as  a  suffix 
in  Tostock. 

Tostock.  To  the  E.  of  Bury.  Spelt  Tostoke,  H.R. ; 
Totstocha  (with  ch  for  k),  D.B.,  p.  166  ;  Totestoc,  D.B.,  p.  8. 
The  last  is  the  fullest  form.  For  A.S.  Tottan-stoc ;  a  form 
which  occurs  in  Kemble,  CD.  ii.  872.  The  sense  is  '  Totta's 
settlement.' 

42.  Stone. 

Chediston.  Near  Halesworth.  The  suffix  is  not  -to7i,  but 
-stone,  as  the  old  forms  show.  Spelt  Ghedeston,  Chedestan,  H.R.; 
Gedestan  (with  ce  for  che),  D.B.,  p.  328 ;  ill-spelt  Cidestan,  D.B.. 
p.  103;  misspelt  Sedestana  (with  se  for  ce),  D.B.,  p.  25.  The 
A.S.  Ce  becomes  M.E.  Che;  and  the  corresponding  A.S.  form  is 
Ceddes  stan,  i.e.  '  Cedd's  stone.'  This  is  more  likely  than 
Ceddan  stdn,  i.e.  '  Cedda's  stone.'  Cedd  and  Cedda  are  both 
real  names,  and  are  closely  allied. 

43.  Stow. 

Stow  is  the  A.S.  stow,  a  place ;  whence  the  phrase  to  stow 
aiuay.  There  is  a  hundred  named  Stow  ;  West  Stow,  two  and 
a  half  miles  W.  of  Ingham  (N.  of  Bury)  ;  and  Stowmarket,  near 


90  THE    PLACE-NAMES  OF   SUFFOLK 

which  is  Stow  Upland.  The  hundred  may  have  been  named 
from  Stow,  which  was  doubtless  the  old  name  of  Stowmarket ; 
since  market  is  a  word  of  Picard-French  origin.  Stowmarket  is 
somewhere  near  the  centre  of  Stow  hundred.  And  stotu  appears 
as  a  suffix  in  Felixstowe. 

Felixstowe.  Copinger  gives  eleven  spellings  of  this  name. 
One  of  them  is  Felixstow.  But  the  other  ten  are  very  different, 
viz.  Filthestowe,  Fillthustowe ;  also  (with  c  for  t,  erroneously) 
Felchestoiue,  Filchestowe,  Fylchestow;  Felyestoiue  (with  y  for  the 
A.S.  ]>  =  th) ;  and  (in  a  contracted  form)  Felstoiv,  Fylstowe, 
Fylstoe,  Filston  (error  for  Filstou).  It  is  quite  certain  that, 
not  long  after  the  Conquest,  the  prevalent  form  was  Filthestow, 
and  that  it  was  afterwards  shortened  to  Filthstow  and  Filstow. 
The  name  Felixstow  is,  in  fact,  not  the  original  one,  and  does 
not  occur  in  early  documents  ;  but  it  was  known  in  the  time  of 
Henry  VIIL,  as  it  occurs  in  the  Valor  Ecclesiasticus,  made 
during  that  reign.  A  distinction  is  there  drawn  between 
"  prior'  de  Filstoive  modo  subpress' ''  (vol.  iii.  p.  447)  and 
"  monasterio  de  Felyxstowe "  (vol.  iii.  p.  449) ;  so  that  the 
names  did  not  then  refer  to  the  same  place.  In  Tanner's 
Notitia  Monastica  (ed.  1787)  there  is  mention  of  the  monastery 
of  Walton  St  Felix ;  and  he  speaks  of  a  MS.  that  refers  to  "  the 
priory  of  Felixstowe,  alias  Fylchestowe  in  Walton."  It  appears, 
in  fact,  that  Walton  St  Felix  was  a  priory  founded  by  Roger 
Bigod,  about  1105;  that  it  was  called  by  the  name  of  Felix; 
and  that  it  was  situate  at  Walton,  about  a  mile  from  Filstow, 
properly  so  called/.  The  names  were  confused,  and  so  the 
name  of  Filstow  was  sometimes  changed  to  Felixstow,  though 
many  still  held  to  the  older  name.  Thus  we  find  in  Raven's 
Suffolk,  p.  150—"  Felixtow  alias  Fylstou."  But  in  the  present 
enlightened  days,  the  more  attractive  name  has  prevailed, 
owing  to  such  advancement  of  knowledge  as  has  enabled  the 
antiquaries  to  discover  that  the  historian  Beda  mentions  the 
labours  of  St  Felix  in  Suffolk,  who  "had  the  see  of  his  bishopric 
appointed  him  in  the  city  Dommoc,"  which  was  the  old  name 
of  Dunwich    (Hist.   Eccl.  ii.  15).     Dunwich,  after   all,  is   not 

1  "  The  church  of  St  Felix  in  Walton  "  ;  Kirby,  p.  88. 


NAMES   ENDING   IN    -STOW,   -THORPE  91 

Felixstow,  but  it  is  in  the  same  county  ;  which  is  held  to  be 
good  enough  for  corroboration  of  a  blunder.  One  hardy  fable 
is  that  St  Felix  landed  near  Felixstowe  when  he  came  from 
Burgundy  ! 

It  is  more  to  the  purpose  to  discover  the  origin  of  the  true 
name  Filstow.  On  this  subject,  we  read  in  the  Crawford 
Charters,  ed.  Napier  and  Stevenson,  p.  71  : — "  Felixstowe, 
CO.  Suffolk,  apparently  represents  a  Fileth-stow,  as  it  is  ciiUed 
Filthstowe  in  1316  (Nomina  Villarum,  319  a)."  They  explain 
that  "  Fileth  is  an  unexplained  word  that  occurs  several  times 
in  local  names  "  ;  and  they  give  examples,  in  which  occur  the 
spellings  fileth,  fdet,  filed,  /(/let,  and  once  fcelet  (probably  by 
error),  in  compounds  ;  and  uncompounded,  in  the  expression  on 
filetha]  see  Birch,  C.S.  iii.  494,  589,  &c.;  ii.  519.  Such  nouns 
are  often  made  from  verbs,  and  the  right  form  may  be  fyleth 
(ioT  fylleth),  from  fyllan,  to  fell  trees.  It  would  then  correspond 
to  the  prov.  E.  fellet,  'the  portion  of  a  wood  felled  annually,  a 
portion  of  felled  wood.'  The  original  sense  may  bave  been, 
simply,  '  a  place  of  felled  trees,'  a  place  where  trees  have  been 
felled  to  make  a  clearing.  It  is  compounded  with  leaJi,  a  lea, 
cumb,  a  combe,  hanim,  an  enclosure,  and  stoiv,  a  place  ;  all 
implying  some  special  locality,  and  all  suitable.  But  whatever 
the  right  sense  may  be,  we  at  any  rate  learn  that  Felixstow  is 
an  ingenious  '  learned '  alteration  of  a  place  that  was  once 
named  Filethstow,  afterwards  shortened  to  Filthstow  and 
Filstow.  (I  find  that  Middendorf  derives  fileth  from  A.S.  fillan, 
to  flay,  skin,  with  reference  to  the  removal  of  turf;  but  no  such 
verb  is  found  in  A.S.,  and  the  sense  is  very  forced.  On  the 
other  hand,  the  verb  '  to  fell '  appears  as  fille  in  the  best 
(Ellesmere)  MS.  of  Chaucer :  "  It  semed  as  it  wolde  fille  an 
ook"  ;  Knightes  Tale,  A  1702.) 

44.    Thorpe. 

Thorpe,  more  correctly  thorp,  is  the  O.  Frisian  and  A.S. 
thorp,  'a  village';  cognate  with  Du.  dorp,  G.  dorf.  We  find 
(in  Suffolk)  Thorpe-Morieux,  a  hamlet  called  Thorpe  near 
Aldriugham,  and  Thorpe-by-Ixworth  ;  and  also  Westhorpe. 
The  first  takes  the  name  of  Morieux  from  a  Norman  family  so 


92  •  THE    PLACE-NAMES   OF   SUFFOLK 

called.  The  0.  Fr.  moriel  (of  which  morieux  is  the  plural)  is  a 
variant  oi  morel,  meaning  '  mnlberry-coloured';  the  name  Morel 
was  often  given  to  a  horse.  Westhorpe  is  merely  compounded 
of  West  and  Thorpe.  Thorp  is  not  exclusively  Scandinavian, 
as  some  say, 

4.5.     Toft. 

Toft  meant  a  green  knoll,  open  ground,  or  homestead ;  see 
my  Etym.  Diet.  It  occurs  in  Lowestoft  and  in  Stowlangtoft. 
The  latter  is  a  mere  compound,  containing  Stow,  a  place,  lang, 
i.e.  long ;  and  toft ;  and  Kirby  thinks  it  was  named  from  a 
family  of  Langtofts. 

Lowestoft.  Copinger  gives  many  forms,  including  late 
spellings  such  as  Laistoft,  Leistoft,  and  Lestoff;  the  last  of 
these  represents  a  common  pronunciation  of  it.  An  early 
spelling  is  Lowistoft,  H.R. ;  T.N.  D.B.,  p.  5,  has  Lothuwistofte. 
Here  Lothuwis  represents  an  A.S.  Hlothuwlges,  gen.  of 
Hlothuwig,  usually  spelt  Hlothewig  or  Hlodwig;  the  form 
Hlothewlg  occurs  in  Birch,  C.S.  iii.  491,  1.  1.  This  A.S.  name 
is  cognate  with  the  Old  High  German  Hluduwig  or  Hludwig, 
Mid.  High  G.  Ludewic,  G.  Ludwig ;  a  famous  name.  Latinised 
as  Chlodovichus  and  Chlodovius,  which  produced  the  F.  Louis. 
Thus  the  sense  is  '  Hlothuwig's  toft '  or  homestead. 

46.     Ton,  unstressed  form  of  Town. 

Acton.  N.  by  E.  from  Sudbury.  Spelt  A  ketona,  D.B.,  p.  71, 
and  on  the  same  page,  Achetuna.  Copinger  gives  several  forms, 
of  which  the  most  important  are  Acketon,  Aketon,  Aketone, 
Aketune ;  showing  (by  the  e  before  the  t)  that  it  is  not  derived 
from  A.S.  ac-tun,  i.e. '  oak-town,'  as  is  the  case  with  some  of  the 
Actons  elsewhere.  Moreover,  this  Suffolk  Acton  appears  as 
Acantun  in  a  list  of  boundaries  of  land  at  Bildeston,  which  is 
in  the  neighbourhood ;  and  this  A.S.  form  is  consistent  with  the 
forms  given  above.  Hence  the  meaning  of  this  Acton  is  '  Aca's 
town,'  Aca  is  a  known  name.  See  Birch,  C.S.  iii.  603.  The 
A.S.  tun,  lit.  '  town,'  had,  practically,  the  sense  of '  farm.' 


NAMES    ENDING    IN    -TON  93 

Alderton.  Near  the  E.  coast,  to  the  S.  of  the  mouth  of 
the  Aide.  Spelt  Alderton,  Ipm. ;  but  Alretuna  in  D.B.,  p.  74; 
and  Alreton  in  the  Crawford  Charters,  p.  33,  The  prefix 
represents  A.S.  alra,  gen.  pi.  of  air,  an  alder-tree.  The  sense 
is  'alders'  town,'  or  *  farm  by  the  alder-trees.' 

Alpheton.  W.  by  N.  from  Lavenham.  Also  spelt 
Alpheaton  (Kelly).  Spelt  Alfeton,  Alffleton,  H.R.  The  latter 
spelling  answers  to  the  A.S.  form  ^IJlwdetun  in  a  Worcester- 
shire Charter ;  in  Birch,  iii.  586.  jElflccde  is  better  spelt 
^Ifflrrde,  the  gen.  of  JSlffla'd,  a  female  name,  which  accounts 
for  the  gen.  in  -e  and  the  spelling  Alffleton  (with  the  double/). 
The  sense  is  '  town  (or  farm)  of  ^Elfflied.'  She  was  probably  a 
widow. 

Ampton.  N.  of  Bury,  near  Ingham  station.  Copinger  re- 
cords the  old  forms  Ameton,  Ametone,  Ametun  ;  spelt  Hametuna 
(with  H  wrongly  prefixed)  in  D.B.,  p.  165.  The  prefix  repre- 
sents the  A.S.  Amman,  gen.  oi  Amma ;  cf.  Amman-hroc,  Amman- 
ivell  (both  in  Kemble's  Index).     The  sense  is  '  Amma's  farm.' 

AssiNGTON.  N.W.  of  Nay  ton.  Spelt  Asetime,  KB.; 
Asington,  T.N. ;  Asinton,  Asington,  H.R.;  Asetona,  D.B.,  p.  271. 
Copinger  also  records  Asentune,  Assinton,  Asynton.  The  A.S. 
form  is  certainly  Asantun ;  and  the  -an  has  been  turned  into  -en, 
-in,  and  -ing.  We  must  neglect  the  g  in  this  case.  The  sense 
is  '  Asa's  farm.'     Asa  is  a  known  E.  name. 

Athelington.  E.  of  Occold,  which  is  S.  by  E.  from  Eye. 
Spelt  Athelington,  Ipm. ;  answering  to  A.S.  ^thelinga-tun ; 
lit.  '  town  (or  farm)  of  the  ^thelings  or  nobles.'  The  A.S. 
cetheling  means  a  prince  or  nobleman.  Cf.  Athelney,  of  which 
the  old  form  was  .^thelinga-ig,  or  '  isle  of  nobles.'  Perhaps  I 
ought  to  add  that  to  translate  cetheling  by  'noble'  in  these 
instances  is  by  no  means  certain  ;  since  Ji^theling  might  equally 
well  have  the  simpler  sense  of  '  son  of  ^thela,'  i.e.  '  son  of  any 
one  whose  name  began  with  vEthel ' ;  and  such  names  are 
numeroils. 

Bacton.  N.  of  Stowmarket.  Copinger  gives  the  forms 
Baketon,  Bakenton;  D.B.  has  Bachetuna,  p.  292  (with  che  for  ke). 


94  THE    PLACE-NAMES    OF    SUFFOLK 

The  late  A.S.  Baketun  occurs  in  the  Crawford  Charters,  p.  33. 
For  A.S.  Bacan-tun.  The  sense  is  '  Baca's  farm.'  Baca  is  a 
known  name. 

Barton.  Great  Barton ;  N.E.  of  Bury.  Spelt  Bertuna, 
D.B.,  p.  162.  A.S.  beretun,  lit.  'corn-farm,'  or  barley-enclosure; 
from  here,  barley.     There  are  many  Bartons. 

Belton.  S.W.  of  Yarmouth.  Spelt  Belton,  H.R.;  Beltone, 
R.B. ;  Beletuna,  D.B.,  pp.  6,  8.     For  the  prefix,  see  Belstead. 

Beyton.  Spelt  Beyton,  Ipm. ;  Beytone,  R.B.  ;  Begatona, 
D.B.,  p.  259.  It  answers  to  A.S.  Bcvgan-tim.  The  prefix 
Bwgan  occurs  in  Bwgan-iuyrth ;  Birch,  C.S.  iii.  96,  I.  29.  The 
sense  is  '  Bgega's  farm.'  The  prefix  in  Bayford  (Herts.)  is  not 
quite  the  same,  as  the  form  in  D.B.  is  Begesford ;  see  my 
Place-names  of  Herts.,  p.  27. 

Bildeston.  N.  of  Hadleigh.  Spelt  BUdeston,  H.R. ;  T.N. ; 
Bildestone,  R.B. ;  Bilestuna,  D.B.,  p.  291.  The  d  is  intrusive, 
as  the  last  form  shows.  Alluded  to  in  the  form  Byliges-dyne; 
Birch,  C.S.  iii.  603 ;  where  dyne  is  a  derivative  from  dun.  The 
same  prefix  occurs  in  Bylges-Uge,  in  the  A.S.  Chron.,  an.  10.55. 
The  sense  is  '  Bylig's  farm.' 

Blundeston.  N.W.  of  Lowestoft.  Spelt  Bhmdeston,  H.R, 
Blundes  is  the  gen.  of  Blund,  which  occurs  as  a  personal  name 
in  the  name-list  given  in  Ipm.,  vol.  i.  Of  Norse  origin  ;  Zoega 
has  :  Icel.  "  blundr,  m.  dozing,  slumber  ;  occurs  as  a  nickname." 
The  sense  is  '  Blund's  farm '  ;  Blund  (orig.  Blundr)  being 
Norse. 

Boyton.  S.  of  Butley,  and  near  the  Butley  river.  Spelt 
Bointone,  R.B. ;  Boituna,  D.B.,  p.  81.  Copinger  also  has  the 
form  Boynton.  The  prefix  Boin-  is  short  for  A.S.  Boian,  late 
form  of  Bogan,  gen.  of  Boga,  a  known  name.  The  sense  is 
'  Bosra's  farm.' 

Brampton.  N.E.  of  Halesworth.  Spelt  Brampton,  T.N. ; 
but  Brarntuna,  D.B.,  p.  15  ;  BranUma,  D.B.,  p.  102.  Copinger 
also  gives  the  form  Bramton ;  and  the  p  is  certainly  unoriginal. 


NAMES   ENDING    IN    -TON  95 

The  dat.  form  Bramtune  occurs  iii  the  A.S.  Chronicle,  an.  1121. 
The  form  with  mt  must  be  older  than  that  with  nt ;  the  change 
from  mt  to  nt  is  easy,  but  that  from  nt  to  nit  is  abnormal.  The 
same  prefix  Brani-  occurs  again  in  Bramcestria  (Birch,  C.S.  iii. 
280),  which  is  Brancaster  in  Norfolk  ;  showing  that  the  original 
form  of  Brancaster  was  Bramceaster  in  Anglo-Saxon.  The 
meaning  of  Bram  (which  can  hardly,  in  the  latter  case,  repre- 
sent a  personal  name)  is  unknown;  but  it  may  be  related  to 
the  A.S.  broyn,  a  broom,  with  which  the  mod.  E.  bramble 
is  etymologically  connected.  The  sense  of  'bramble-farm' 
seems  possible  here.  (If  a  personal  name,  it  is  from  Brama, 
gen.  of  Brami,  a  Norse  name ;  see  Nielsen.) 

Brandeston.  Brandeston  (Leic.)  appears  in  Ipm.  We  also 
find  Branteston,  T.N. ;  Braundestone,  in  the  Liber  Custumarum; 
Brantestuna,  D.B.,  p.  302 ;  but  Brandestuna,  D.B.,  p.  216. 
Brandes  is  the  gen.  of  the  known  name  Brand ;  and  the  sense 
is  'Brand's  farm.' 

Browston,  in  Belton ;  a  hamlet  one  mile  S.E.  of  Belton. 
Copinger  gives,  as  old  spellings,  the  forms  Broweston  and 
Broxton.  It  answers,  by  position,  to  Brochestuna,  D.B.,  p.  7 ; 
and  the  last  two  forms  suggest  that  a  guttural  sound  has 
been  lost,  and  that  the  original  form  was  Brocces-tun ;  with 
which  compare  Brocces-ham,  -hlcew,  and  -slwd,  in  Kemble's 
Index.     The  sense  may  have  been  'Brocc's  farm.' 

Carlton.  Carlton  Colville  is  to  the  S.W.  of  Lowestoft; 
and  Colville  is  the  name  of  a  Norman  family  connected  with  it. 
Spelt  Carleton,  T.N.,  H.R. ;  Carletuna,  D.B.,  p.  254  ;  Kaiietuna, 
D.B.,  p.  43.  For  A.S,  Carla  tun,  'farm  of  the  churls'  or 
husbandmen.  Carla  is  the  gen.  pi.  of  carl,  a  churl,  a  husband- 
man ;  where  carl  is  not  the  true  native  word,  but  borrowed 
from  the  O.  Norse  karl,  a  man,  rustic,  carle  ;  the  A.S.  related 
word  is  ceorl,  mod.  E.  churl;  as  in  Chelsvvorth. 

Chelmondiston.  Commonly  called  Chemton  (Kirby).  Near 
the  S.W.  bank  of  the  Orwell.  Spelt  Chelmundeston,  H.R. ; 
Chelmondeston,  T.N.  The  A.S.  form  is  Ceolminides  tun,  i.e. 
'  Ceolmund's  farm.'     Ceolmund  was  once  a  very  common  name. 


96  THE    PLACE-NAMES    OF    SUFFOLK 

Chevington.  S.W.  of  Bury.  The  </  is  a  late  insertion  ; 
it  should  have  been  Cheventon.  Spelt  Cheventon,  H.R.;  Ceven- 
tiina  (with  Ge  for  Che),  D.B.,  p.  153.  The  prefix  is  the  A.S, 
Ceo/an,  gen.  of  Ceo/a  ;  and  the  sense  is  '  Ceofa's  farm.' 

Chilton.  Near  Sudbury.  Spelt  Chilton,  Ipm. ;  Ciltona 
(with  Ci  for  Chi),  D.B.,  p.  47.  Copiuger  also  gives  the  spellings 
Cheletuna,  Chelton,  without  references.  But  these  forms  are 
doubtless  right,  and  show  that  e  was  the  older  vowel.  The 
prefix  Cliele-  represents  the  A.S.  Ceolan,  as  in  Ceolan-hyrst ; 
Birch,  C.S.,  ii.  458.  Ceolan  is  the  gen.  of  Ceola ;  and  the  sense 
is  '  Ceola's  farm.'  Ceola  would  become  Clieel,  easily  shortened 
to  Chil. 

Clopton.  Spelt  Clopton,  H.R. ;  Clopetuna,  D.B.,  p.  70. 
The  prefix  Clope-  answers  to  the  A.S.  Cloppa  in  Cloppa-hCim  ; 
see  Sw^eet,  E.  Eng.  Texts,  p.  451.  This  Cloppa  looks  like  a 
gen.  pi.  from  a  nom.  clop,  as  in  clop-wcer,  clop-hyrst,  in  Birch, 
C.S.,  iii.  589,  590.  In  this  case,  it  is  difficult  to  assign  the 
origin ;  unless  we  ally  it  to  the  O.H.G.  claph,  '  a  boulder.' 
Otherwise,  Clope  may  answer  to  the  A.S.  Cloppan,  gen.  of 
Cloppa,  a  personal  name,  not  exactly  found  ;  but  it  may  be 
the  equivalent  of  Clappa,  the  name  of  a  king  of  Bernicia  in 
Florence  of  Worcester's  Chronicle,  i.  0 ;  according  to  Prof. 
Moorman,  in  his  account  of  Clapham  in  the  W.  Riding,  Clap- 
ham  in  Surrey  is  called  Cloppaham  in  Kemble,  CD.,  no.  317  ; 
so  that  Clap  and  Clop  seem  to  have  been  convertible. 

Coney  Weston.  N,  of  Ixworth ;  not  far  from  the  Little 
Ouse.  Spelt  Cunegestuna,  D.B.,y>.  169;  showing  that  the  name 
has  suffered  some  alteration.  The  original  prefix  must  have 
been  the  0.  Norse  konungs,  a  form  of  the  gen.  of  konung,  a 
king.  The  second  n  being  lost,  this  took  the  form  konugs  ; 
or  (with  the  English  suffix  -es  in  place  of  the  Norse  s)  konuges, 
fairly  well  represented  by  the  Cuneges  in  D.B. ;  after  which 
the  -es  was  dropped,  and  the  prefix  became  Coney.  We  thus 
see  that  the  original  sense  was  '  king's  town  or  farm ' ;  which 
would  regularly  have  given  a  later  form  Coneyton ;  but  the 
latter  part  of  the  word  was  changed  from  ton  to  Weston  (west 


NAMES   ENDING   IN   -TON  97 

town) ;  evidently  by  association  with  the  neighbouring  village 
called  Market  Weston,  the  prefix  Coney  being  contrasted  with 
that  of  Market.  Compare  Coneythorpe  in  the  W.  Riding, 
which  means  '  king's  thorpe '  or  '  king's  village.' 

CoRTON.  On  the  E.  coast,  N.  of  Lowestoft.  Spelt  Cortone, 
R.B. ;  Gorton,  H.R. ;  Karetuna,  D.B.,  p.  6  (which  can  hardly  be 
quite  correct,  but  must  stand  for  Koretuna).  The  sense  of  the 
prefix  is  unknown  and  uncertain;  but  Kore-  suggests  an  A.S. 
*Corau,  gen.  of  *Cora,  used  as  a  personal  name.  That  such  a 
name  was  in  use  is  suggested  by  the  occurrence  of  Cores,  the 
gen.  of  a  strong  form  Cor.  Cores  occurs  in  Coreshroc ;  in 
Kemble,  CD.,  no.  632.  Hence  a  possible  sense  is  'Cora's 
farm.' 

Cotton.  Near  Mendlesham.  Spelt  Coton,  H.R. ;  but  D.B. 
has  Cottuna,  pp.  10,  11  ;  Cotetuna,  p.  84;  and  Codetuna,  p.  58. 
It  is  clear  that,  of  these  forms,  Codetuna  is  the  older ;  afterwards 
d  became  t,  and  the  two  ^'s  were  united.  Codentuna  at  once 
suggests  an  equivalent  A.S.  form  Codan-tun.  The  prefix  Codan 
occurs  in  Codan-ford ;  Birch,  C.S.,  ii.  224,  1.  3.  The  sense  is 
'Coda's  farm.'     See  Coddenham. 

Bennington.  N.  of  Framlingham.  Spelt  Dinyeueton,  H.R.; 
Dingiuetuna,  D.B.,  p.  90;  Dingiuetona,  D.B.,  p.  91;  Binneuetuna 
(where  B  is  an  error  for  D),  D.B.,  p.  95.  Copinger  gives  a 
large  number  of  later  forms,  among  which  may  be  particularly 
noticed  these:  Digneveton,  Dingneueton,  Bingniueton,  Binieueton, 
Byneyeueton.  It  is  clear  that  the  original  suffix  was  by  no 
means  -ington;  but  that  -ington  was  substituted  for  something 
far  less  usual.  All  the  forms  can,  without  much  difficulty,  be 
deduced  from  A.S.  *Denegife,  gen.  of  a  female  name  *Denegifu. 
Though  this  name  does  not  happen  to  occur,  it  is  regular  and 
probable,  as  it  is  compounded  of  the  common  prefix  Bene- 
and  the  common  suffix  -gifit.  Moreover,  we  find  a  similar  form 
Benegyth,  also  a  female  name,  with  a  gen.  Benegythe.  The 
compound  place-name  Bengithe-graf  (with  i  for  y)  occurs  in 
Birch,  C.S.,  ii.  419,  1.  3.  The  g  in  gifu  was  pronounced  as  y, 
and  the  form  Deneyiveton  easily  became  Denyivton  ;  and  this 

C.  A.  S.  Octivo  Series.     No.  XLVI.  7 


98  THE   PLACE-NAMES   OF   SUFFOLK 

unusual  and  difficult  form  was  changed  to  Bennington,  because 
the  suffix  -ington  was  common.  I  have  little  hesitation  m 
explaining  this  unusual  name  to  mean  '  Denegifu's  farm';  and 
we  must  remember  that  this  is  a  female  name,  with  a  genitive 
in  -e. 

Denstone  ;  otherwise  Denardiston  (Kelly).  Spelt  Denar- 
deston,  H.E,.;  Danerdestuna,  D.B.,  p.  219.  Evidently  for  Dene- 
heardes  tun  (O.  Merc.  Denehardes  tun).  The  gen.  occurs  (with 
DcBne-  for  the  more  usual  Dene-)  in  Dceneheardes  hegeneive; 
Birch,  C.S.,  ii.  81.     The  sense  is  '  Denehard's  farm.' 

Drinkstone.  E.  by  S.  from  Bury.  Spelt  Drencheston, 
Ipm. ;  H.K ;  Drencestuna,  D.B.,  p.  164.  Other  spellings  (in 
Copinger)  are  Drengstone,  Dryngeston,  which  are,  practically, 
better.  The  form  Rengestuna  occurs  in  D.B.,  p.  21,  with  initial 
D  omitted  by  mistake.  A  more  correct  form  would  be 
Drenges-tun,  where  Drenges  is  the  gen.  of  the  A.S.  dreng,  a 
warrior,  soldier ;  not  an  English  word,  and  only  occurring 
once,  as  it  was  borrowed  from  O.  Norse.  The  O.  Norse  word 
was  drengr,  a  valiant  man,  strong  young  fellow.  The  sense 
is 'soldier's  farm.'  The  same  prefix  occurs  in  Dringhouse,  in 
the  W.  Riding ;  see  the  discussion  of  the  social  position  of  the 
drengr  in  Prof.  Moorman's  W.  Riding  Names,  p.  xxiii. 

N.B. — Dreng  was  also  a  personal  name  (Nielsen). 

Easton.  There  are  two  Eastonsj  one  near  Southwold,  and 
another  on  the  Deben.     It  means  '  east  town  '  or  '  east  farm.' 

Edwardstone.  Between  Sudbury  and  Hadleigh.  Spelt 
Edivardeston,  T.N. ;  Ipm. ;  Eduardestuna,  D.B.,  p.  47.  For 
A.S.  Eadtueardes  tun ;  0.  Merc.  Eadwardes  tun ;  mod.  E.  '  Ed- 
ward's farm.' 

Erwarton.  N.  of  the  Stour,  near  its  mouth.  Spelt  Euer- 
luardton,  H.R. ;  Eureiuardestuna,  D.B.,  p.  229.  For  O.  Merc. 
Eforwardes  tun,  A.S.  Eqforweardes  tun.  The  sense  is  'Efor- 
ward's  farm.' 

EusTON.  S.E.  of  Thetford.  Spelt  Eustone,  R.B. ;  Eueston, 
H.R. ;  Euestuna,  D.B.,  p.  174.     In  the  last  two  examples,  u  is 


NAMES   ENDING   IN   -TON  99 

for  v;  so  that  an  older  name  was  Eveston.  The  {arcfix  is  the 
same  as  in  Evesham,  viz.  the  A.S.  Eofes,  gen.  of  Eof,  a  known 
name.     The  sense  is  '  Eofs  farm.' 

Flempton.  N.W.  of  Bury.  The  spelling  in  D.B.  is  Fleming- 
tuna,^.  154.  This  answers  to  A.S.  fleaniinga  tun  or  O.  Merc. 
fleminga  tiln.  where  fleaininga  is  the  gen.  pi.  of  Jleaming,  a 
fugitive.     The  sense  is  '  farm  of  the  fugitives.' 

Flixton.  There  are  two  Flixtons ;  one  near  Bungay,  and 
one  near  Lowestoft.  Spelt  i^^z^^o??,  H.R.;  Ipm.;  Flixtuna,  D.B., 
p.  6.  Perhaps  of  Norse  origin.  Nielsen  says  that  Flik  was  a 
Danish  name,  known  in  the  thirteenth  century.  This  form 
would  suit  very  well,  as  it  would  take  the  f^rm  of  Flikkes 
when  declined  as  an  E.  name.  It  is  quite  likely  that  Flixton 
meant  '  Flik's  farm.'  In  Raven's  Hist,  of  Suffolk,  p.  44,  it  is 
said  that  "  the  two  Flixtons  preserve  the  name  of  Felix."  Of 
course  this  is  wholly  impossible,  and  shows  what  comes  of  ne- 
glecting phonetic  laws.  The  stressed  vowel  in  Felix  cannot 
disappear ;  a  shortened  form  would  become  Fell  or  Fele. 

Flowton.  Between  Bury  and  Bildestou.  The  name  (like 
Browston)  has  lost  a  guttural ;  spelt  Flokton  in  Ipm. ;  Floclie- 
tiina  (with  die  for  ke),  D.B.,  pp.  114,  226.  The  A.S.  ct  becomes 
ht ;  and  this  ht  becomes  M.E.  gh,  and  then  lo  ;  the  form  Floctun 
would  regularly  become  Flohtun,  Floghton,  Flowton.  The 
present  form  shows  that  the  e  in  the  D.B.  form  Flochetuna  was 
falsely  inserted  ;  and  that  the  A.S.  form  was  Floctun.,  with  the 
c  and  t  in  contact.  Floctim  is  iox  jlocc-tun  ;  from  _^occ,  a  flock 
of  sheep.     The  sense  is  '  flock-farm  '  or  '  sheep-farm.' 

Freston.  Near  the  S.W.  bank  of  the  Orwell.  Spelt 
Frestune,  H.R. ;  Fresetuna,  D.B.,  p.  230.  A.S.  Fresan  tun ; 
Birch,  C.S.,  iii.  602.  Fresan  is  the  gen.  of  Fresa,  a  Frisian. 
The  sense  is  '  farm  (or  town)  of  the  Frisian.'  This  is  an  in- 
teresting result.     See  Friston. 

Friston.  S.W,  of  Saxmundham.  This  name  is  a  mere 
variant  of  Freston  (above),  and  has  the  same  sense.  And  see 
Fressingfield. 

7—2 


100  THE   PLACE-NAMES   OF   SUFFOLK 

Fritton.  N.W.  of  Lowestoft.  Spelt  Freton,  H.R. ;  Fride- 
tuna,  D.B.,  p.  7  (with  d  for  th).  For  A.S.  Frith-tun-,  see 
Thorpe,  Diplomat.,  p.  564,  I.  7.  Cf.  A.S.  frith-hurh,  a  town 
in  which  /rzY^  or  peace  between  two  parties  was  secured. 
Hence  the  sense  is  '  frith  farm,'  or  '  a  farm  to  which  security 
was  assured.'  The  security  may  have  been  due  to  situation. 
For  the  development  of  the  senses  of  frith,  see  N.E.D. 

GORLESTON.  It  forms  a  part  of  Great  Yarmouth.  Spelt 
Gorleston,  T.N.,  H.R. ;  Gurleston,  T.N. ;  Gorlestuna,  D.B.,  p.  6. 
The  prefix  evidently  represents  the  geu.  case  of  a  name  which 
has  been  cut  down  to  Gorl.  I  can  think  of  no  form  that  would 
admit  of  this  except  *Gorwulf,  made  (like  Garwulf)  with  the 
suffix  widf.  There  is  such  a  prefix  as  Gov-,  in  limited  use  ; 
it  occurs  in  Gor-mund  and  Gor-noth.  Garwulf  became  Garulf, 
and  might  (in  D.B.)  have  been  shortened  to  Gar'l,  but  would 
hardly  have  given  the  required  vowel  o  at  so  early  a  date  as 
1086.  As  a  rather  likely  guess,  I  suggest  the  sense  of  '  Gor- 
wulf's  farm.' 

GuNTON.  Near  Lowestoft.  Spelt  Gunetone,  R.B.;  Guneton, 
H.R.  Evidently  Norse.  From  the  O.  Norse  (r^mjia-^im ;  where 
Gunna  is  the  O.N.  gen.  of  Gunni,  a  known  masc.  name,  which 
is  common  in  place-names  (Rygh).  The  sense  is  '  Gunni's 
farm.' 

Hacheston.  Spelt  Hacheston,  Haccheston,  Ipm. ;  Races- 
tvna,  D.B.,  p.  26 ;  Hecestuna,  D.B.,  p.  12.  The  palatalised  form 
suggests  an  English  origin,  rather  than  Norse.  The  patronymic 
Hcecing  occurs  in  Birch,  C.S.,  ii.  403,  1.  27,  from  a  form  *Hcec 
(gen.  Hseces).  The  sense  is  probably  '  Hjbc's  farm.'  The  weak 
form  Haca  is  known. 

Harleston.  Three  miles  N.W.  of  Stowmarket  Station. 
Spelt  Herleston,  H.R.;  Heroluestima,  D.B.,  p.  159.  A.S.  Heo- 
7-ulfes  tUn;  Kemble,  CD.,  no.  722.  Heorulfes  is  the  gen.  of 
Heorulf,  shortened  form  of  Heormuidf.  The  sense  is  '  Heoru- 
wulf's  farm.' 

Hasketon.  Near  Woodbridge.  Spelt  Hasketon,  H.R. ; 
Haschetuna,   D.B.,   p.    69;    Hascetuna,   D.B,,  p.   70.     The   sk 


NAMES   ENDING   IN   -TON  TOl 

shows  that  the  name  is  of  Norse  origin.  Rygh  gives  O.  Norse 
Hoskuldr,  a  Norse  personal  name  which  appears  in  place- 
names  as  Haskel.  Hence  *Haskeltun,  shortened  to  Hasketun. 
The  sense  is  therefore  '  Hoskuldr's  farm.' 

Hemingstone.  E.  by  S.  of  Needham  Market.  Spelt 
Hemingeston,  T.N. ;  Hemingestone,  Ipm. ;  Hainingestuna,  D.B,, 
pp.  29,  115.  But  Copinger  also  gives  the  fuller  forms  Hel- 
mingstone,  Hemelingeston,  showing  that  an  el  has  been  lost 
after  the  m.  The  original  form  was  therefore  Hemelinges 
tun ;  and  the  sense  is  '  Hemeling's  farm,'  or  '  farm  of  the  son 
of  Hemele.'  Hemele  is  a  known  name,  of  which  Searle  sives 
six  examples. 

HiNTON.  A  mile  and  a  half  S.W.  of  Blythburgh.  Spelt 
Hinetuna,  D.B.,  p.  101.  There  are  several  Hintons;  and 
Hinton  (Dors.)  is  spelt  Hineton  in  Ipm.,  p.  20.  These  spellings 
show  that  Hinton  represents  the  A.S.  hlna  tun  where  hina  is 
the  gen.  of  hlwan,  a  pi.  sb.  meaning  'domestic  servants,'  and 
allied  to  mod.  E.  hind,  an  agricultural  labourer.  The  sense 
is  '  farm  of  the  labourers ' ;  a  farm  held  by  the  labourers  upon 
it,  of  which  there  were  examples  in  a  few  places.  Cf  Carlton 
(above). 

Holton.  Near  Halesworth.  Spelt  Holeton,  Ipm. ;  H.R. ; 
Holiton,  H.R. ;  Holetuna,  D.B.,  pp.  67,  102.  Not  from  A.S. 
Iialig,  holy,  because  the  A.S.  a  is  represented  by  a  in  D.B.,  as 
in  A.S.  stantun,  D.B.  Stanton,  &c.  In  some  cases  Hoi-  repre- 
sents A.S.  holan,  '  hollow,'  but  this  is  scarcely  applicable  here. 
Rather,  in  this  case,  the  corresponding  A.S.  Holan-tun  must  be 
due  to  a  personal  name  Hola  (gen.  Holan),  as  also  in  such  a 
compound  as  Holan-heorh,  Hola's  hill  or  barrow.  The  sense  is 
'  Hola's  farm.'     And  see  Hollesley. 

HoNiNGTON.  N.  by  W.  from  Ixworth.  Spelt  Honeiueton, 
Ipm. ;  Hunegtuna,  D.B.,  p.  171.  Copinger  also  has  Honyton, 
Hunegetune.  Hiineg  represents  the  A.S.  hunig,  honey.  Kemble 
has  several  place-names  beginning  with  Hunig,  as  Hunig-broc, 
-hurne,  -ham,  -hyrst,  &c. ;  and  there  seems  to  be  no  reason 
against  its  occurrence  here  also.     The  sense  is  '  honey-farm,' 


l02'     '  THE   PtAtiE-N AMES   OF   SUFFOLK 

or  a  farm  where  bees  were  kept.  It  might  be  supposed  that 
Honiton  became  Honington  by  confusion  with  the  numerous 
places  ending  in  -ington,  as  in  other  cases.  But  in  the  present 
case  a  simpler  solution  presents  itself.  The  Danish  for  honey 
is  honning,  and  the  Swedish  is  honing;  so  that  Honing  was 
merely  due  to  the  Scandinavian  way  of  pronouncing  honey. 

HoPTON.  N.  of  Lowestoft.  There  is  another  Hopton,  near 
Thetford.  Spelt  Hopeton,  H.R.;  T.N.;  Hopetuna,  D.B.,  p.  170  ; 
Hoppetuna,  D.B.,  p.  66 ;  Hopestuna,  .D.B.,  p.  69.  The  last  form 
contains  an  s,  which  can  hardly  be  original.  There  is  another 
Hopton  in  the  W.  Riding ;  and  it  is  supposed  that  the  prefix  is 
the  A.S.  hop,  mod.  prov.  E.  hope,  '  a  small  enclosed  valley,'  '  a 
recess  in  a  valley,'  or  '  a  piece  of  enclosed  land.'  The  sense  is 
'hope  farm';  where  hope  is  to  be  thus  explained.  The  insertion 
of  the  e  seems  to  have  no  meaning  in  this  case. 

HuNSTON.  S.E.  of  Ixworth.  Formerly  Hunterstuna,  D.B., 
p.  11.  Copinger  also  gives  the  forms  Honterston,  Huntereston, 
Hunteriston,  and  the  like ;  all  answering  to  A.S.  Hunteres  tun, 
i.e.  '  Hunter's  town  '  or  '  farm.' 

Kedington.  Near  Haverhill.  Spelt  Ketton  in  1813,  in 
Beauties  of  England,  xiv.  142;  Kediton,  T.N.;  H.R.;  Kidituna, 
-D.B.,  p.  220.  Copinger  also  gives  Kedintun,  Kedynton;  showing 
that  an  n  must  be  supplied  before  t.  The  name  does  not  seem 
to  be  Norse ;  nor  can  it  be  from  an  A.S.  base  Ced-,  because  that 
would  have  given  us  Ched-.  The  use  of  i  in  Kid-  in  D.B.,  and 
of  e  in  Ked-  in  other  records,  suggests  that  the  A.S.  vowel  was 
y,  before  which  the  G  would  remain  hard.  The  A.S.  form 
was  probably  Gydan-tun ;  where  Cydan  is  the  gen.  of  Cyda. 
The  name  Cyda  occurs  in  the  Durham  Liber  Vitae.  If  this 
be  right,  the  sense  is  '  Cyda's  farm.'  The  A.S.  suffix  -an  be- 
came M.E.  -en  (regularly),  and  might  easily  pass  into  -in.  The 
late  form  in  -ington  arose  from  confusion  with  the  suffix  -ington, 
as  correctly  used  in  other  instances. 

Kenton.  N.E.  of  Debenham.  Spelt /jTme^o/^  Ipm. ;  Chene- 
tiina  (with  Che  for  Ke),  D.B.,  p.  49  ;  Kenetuna,  D.B.,  p.  91.  The 
A.S.  c  becomes  ch  before  short  e,  but  remains  hard  before  a 


NAMES   ENDING   IN   -TON  103 

long  one,  as  in  Kenelm,  from  A.S.  CenJielm,  Hence  Kenc- 
represents  A.S,  Genan,  gen.  of  Gena,  a  known  name.  The  sense 
is  '  Cena's  farm.'     Cf.  mod.  E.  Keene. 

Kettlebaston.  Near  Bildeston.  Spelt  Kettilharston  in 
1813;  in  Beauties  of  England,  xiv.  210;  Kettleherstone,  Ipm. ; 
and  (in  Copinger)  Cutelherston,  Ketelherston ;  so  that  r  must 
be  supplied  before  the  s.  D.B.,  p.  177,  gives  Kitelheornastnna, 
which  agrees  with  the  known  personal  name  Cytelbearn,  Gytel- 
barn,  Ketelharn ;  lit.  '  Cytel's  or  Cetel's  son,'  from  A.S.  hearn, 
0.  Merc,  ham,  O.  Norse  ham, '  child.'  The  sense  is  '  Cytelbarn's 
farm,'  or  '  farm  of  Cytel's  son.'  Cytel  is  the  same  as  the 
Norse  name  Ketill ;  and  Ketelbern  occurs  as  a  Norse  name 
(Bjorkman). 

KiKTON,  or  Kirkton.  N.W.  of  Felixstowe.  Spelt  Kirke- 
tone,  R..B. ;  Kirhetune,  H.R. ;  Gherchetuna,  D.B,,  p.  229,  which 
is  an  Englished  form.  But  Kirkton  is  of  Norse  origin  ;  from 
the  0.  Norse  kirkja,  church.  The  sense  is  '  church-town '  or 
'  church-farm.' 

Leiston.  To  the  E.  of  Saxmundham,  Spelt  Leyston,  H.R.; 
but  this  is  a  contracted  form,  since  it  appears  as  Lestaneston  in 
Ipm.,  p.  218,  col.  1  ;  Leiston,  D.B.,  p.  59.  It  does  not  seem  to 
correspond  in  position  to  Leofstanestuna  in  D.B.,  pp.  121,  123; 
but  it  represents  the  same  name.  The  A.S.  form  was  Leofstanes 
tun,  or  '  Leofstan's  farm.'  It  is  quite  likely  that  Leiston  arose 
from  the  shortening  of  -stonston  to  -stoji ;  to  avoid  repetition. 

Levington.  S.E.  of  Ipswich.  Spelt  Levington,  Ipm. ;  Le- 
vintone,  R.B. ;  Leuentona,  D.B.,  p.  251;  Leuetuna,  D.B.,  p.  121. 
It  is  clear  that  the  insertion  of  g  is  comparatively  late.  It 
represents  the  A.S.  Leo/an  tun ;  i.e.  '  Leofa's  farm.'  Leofa  was 
a  pet-name  for  Leofric,  or  for  other  names  beginning  with 
Leof. 

Market  Weston,  i.e.  west  town ;  see  Weston. 

Melton.  Near  Woodbridge.  Spelt  Melton,  T.N.;  Meltuna, 
D.B.,  p.  25.    [N.B. — The  reference  in  Kemble's  Index  to  Meltun 


104  THE   PLACE-NAMES   OF   SUFFOLK 

is  wrong ;  the  charter  relates  to  Kent,  and  has  the  forms  Mele- 
tun,  Melentun,  but  not  Meltun.]  Copinger  also  gives  the  forms 
Meltoune  and  Mcdton ;  the  last  is  doubtful.  The  forms  suggest 
a  compound  word,  not  a  name  followed  by  -ton ;  because  I  and  t 
are  in  immediate  contact.  Perhaps  we  may  here  take  mel  as  a 
Norse  word,  as  in  the  place-name  Mellis  (see  p.  124).  The 
sense  will  be  then  given  by  the  Norw.  ynel,  a  sandbank  along 
a  river-course,  which  in  the  present  case  is  the  Orwell.  That 
is,  it  will  mean  '  farm  near  a  sandbank.'  Note :  in  the  Glossary 
to  Thorpe's  Diplomatorium,  he  explains  Methelton  as  meaning 
Melton  (Suff.).  This  can  hardly  be  right,  and  it  contradicts  his 
own  suggestion  at  p.  591,  that  Methelton  means  Middleton. 

Middleton.  Near  Westleton ;  Kelly  describes  it  as  3  miles 
S.  of  Darsham  station,  on  the  river  Minsmere.  There  are  many 
Middletons,  and  some  of  them  are  strangely  shortened  to  Milton, 
as  good  evidence  proves.  Spelt  Mideltuna,  D.B.,  p.  24  ;  Middel- 
tuna,  D.B.,  p.  63 ;  and  Copinger  cites  the  forms  Medilton, 
Middilton,  Midelton.  It  means '  middle  farm,'  as  in  other  cases, 
though  the  point  of  the  application  is  not  obvious.  Thorpe,  in 
his  Diplomatorium,  p.  591,  equates  Middleton  with  the  A.S. 
Metheltun,  which  I  believe  to  be  wrong. 

MoULTON.  E.  of  Newmarket.  Spelt  Midetone,  R.B. ;  Mule- 
ton,  T.N.;  Midetuna,  D.B.,  p.  184.  A.S.  Mulan  tun;  in  Thorpe, 
Diplomat.,  p.  508.     The  sense  is  '  Mula's  farm.' 

Nacton.  S.E.  of  Ipswich.  Spelt  Naketon,  H.R. ;  Naketune, 
R.B. ;  Nachetuna  (with  che  for  ke),  D.B.,  p.  251.  The  prefix  is 
not  English,  but  Norse ;  from  the  O.  Norse  Nakki,  gen.  Nakka 
(Rygh).     The  sense  is  '  Nakki's  farm.' 

Newton,  near  Sudbury.  Old  Newton,  near  Haughley. 
Newton  means  'new  town,'  or  'new  farm';  but  the  older  one 
of  the  pair  has  the  extraordinary  name  of  Old  Newton,  as  it 
is  older  than  the  Conquest,  and  is  called  Niwetuna  in  D.B., 
p.  159,  and  Neiuetuna,  D.B.,  p.  140;  from  A.S.  nuve,  new.  It 
was  called  Newton  Vetus  in  1278;  see  Ipm, 


NAMES   ENDING   IN    -TON  105 

Norton.  S.E.  of  Ixworth.  Spelt  Nortuna,  D.B.,  p.  8.  For 
A.S.  north  tun.  The  sense  is  '  North  form.'  It  lies  to  the  north 
of  Tostock,  and  may  have  been  named  from  that  circumstance. 

NowTON.  S.  of  Bury.  SjDelt  iVo^io>;e,  R.B. ;  i\^aw^o?i,  V.E. ; 
Neotuna,  D.B.,  p.  153.  The  form  of  the  word  suggests  that  it 
is  rather  Norse  than  English.  The  prefix  appears  to  be  the 
O.  Norse  naut,  cattle  (E.  neat,  A.S.  neat).  The  sense  is  'cattle 
farm.'  Cf.  M.E.  7wwt,  neat  cattle.  Bjorkman,  p.  99,  gives  an 
example  in  which  nouthird  means  '  neat-herd.'  The  ow  in 
Nowton  is  pronounced  like  the  ow  in  noiu. 

Ofton,  or  Offton.  N.W.  of  Ipswich,  but  at  some  distance 
from  it.  Spelt  Offintone,  R.B. ;  Offeton,  Ipm. ;  H.R. ;  Offinton, 
T.N. ;  Offetuna,  D.B.,  p.  9.  For  A.S.  Offan  tun ;  meaning 
'  Offa's  farm.' 

Oulton.  W.  of  Lowestoft.  Spelt  Oulton,  T.N. ;  Olton, 
H.R.  Simply  for  old  town,  or  '  old  farm.'  Not,  however,  in 
D.B. 

Preston.  Near  Lavenham.  Spelt  Prestone,  R.B. ;  but 
Prestetona,  Prestetune,  D.B.,  pp.  139,  158.  Thorpe,  Diplomat., 
p.  583,  has  the  dat.  Prestone,  but  only  in  a  late  copy  of  an 
A.S.  charter.  The  form  in  D.B.  suggests  the  A.S.  preosta  tan, 
'  town  (or  farm)  of  the  priests  ' ;  from  preosta,  gen.  pi.  Hardly 
a  compound,  as  preost-tiui,  i.e.  '  priest  farm.' 

Sapiston.  N.  by  E.  from  Ixworth.  Spelt  Sapston,  Ipm. ; 
Sapiston,  H.R. ;  Sapestuna,  R.B. ;  D.B.,  p.  171.  The  forms  are 
all,  unmistakeably,  genitives  singular ;  from  an  unknown  per- 
sonal name  *Sap  or  *Saip.  The  A.S.  seep  (gen.  scopes)  means 
'  sap ' ;  and  sceppe  (gen.  sceppan)  is  '  a  spruce-fir.'  Sap- cote  is 
in  Leics.,  and  Kemble's  Index  has  Sap-cmnh;  but  all  these 
throw  no  real  light  on  the  personal  name.  The  sense  seems 
to  be  '  Sap's  farm.' 

Sibton.  S.  by  W.  from  Halesworth.  Spelt  Sibheton,  Ipm.; 
Sihhetone,  R.B.;  Sibeton,  H.R.;  Sibbetuna,  D.B.,  p.  G4  ;  Sibetuna, 
D.B.,  p.  24.  For  A.S.  Sibban  tun;  meaning  '  Sibba's  farm.' 
Sibba  is  a  known  name. 


106  THE   PLACE-NAMES   OF   SUFFOLK 

SoMERLEYTON.  N.W.  of  Lowestoft.  "  For  shortness  called 
Somerley  "  (Kirby).  Spelt  Somerleton,  H.R. ;  Sumerledetuna, 
D.B.,  pp.  6,  7.  The  form  Somerledeton  is  in  Thorpe,  Diplo- 
mat., p.  583.  From  A.S.  Sumerlidan  tun.  Sumerlidan  is  the 
gen.  of  Sumerlida,  occurring  in  the  A.S.  Chronicle  to  mean 
a  summer-expedition,  or  a  band  of  Danes  who  landed  in  the 
summer  for  plunder.  But  it  properly  refers  to  an  individual 
member  of  such  a  band,  and  it  is  best  to  consider  it  so  here. 
That  is,  sumerlida  means  a  sailor,  mariner,  or  one  who  sails 
over  sea  for  plunder ;  such  a  one  might  afterwards  settle  down. 
The  sense  is,  practically,  that  of  pirate ;  and  we  might  here 
explain  the  place-name  as  '  pirate's  farm  ' ;  meaning  by  '  pirate  ' 
one  who  had  once  been  a  rover. 

SoTHERTON.  Two  miles  to  the  S.  of  Brampton.  Spelt 
Suthei-ton,  H.R,;  Sudretuna,  D.B.,  p.  303.  For  A.S.  suthra 
tun ;  where  suthra  is  the  comparative  of  suth,  south,  and  means 
'  more  to  the  south.'  The  sense  is  '  farm  more  to  the  south ' ; 
perhaps  with  reference  to  Brampton,  which  is  due  N.  of  it,  and 
is  a  larger  place.     See  Sotterley. 

Sproughton.  W.  of  Ipswich.  Spelt  Sproutou,  Ipm.,  T.N. ; 
Sproutune,  H.R.  It  has  lost  an  s ;  for  Copinger  also  cites  the 
forms  Spi'oiiston,  Sproustun,  Sproxton,  Sproxtun.  From  A.S. 
Sprotues  tun ;  the  sense  being  '  Sprow's  farm.'  Sprow  is  a 
known  name. 

Stanton.  Stanton  St  John's  and  Stanton  All  Saints  are 
to  the  N.E.  of  Ix worth.  Spelt  Stanton,  H.R. ;  Stantuna,  D.B., 
p.  94.     A.S.  stantun,  i.e.  '  stone  farm.' 

Stuston.  Spelt  Stufton  (error  for  Stiistor^),  H.R. ;  Stutes- 
tuna,  D.B.,  p.  180.  Copinger  also  gives  Stouston,  Stutestun. 
For  A.S.  Stutes  tun ;  i.e.  '  Stut's  farm.'  The  A.S.  stut, 
prov.  E.  stout,  means  '  a  gnat,  a  midge ' ;  but  it  is  here  a 
personal  name. 

Stutton.  Near  the  N.  bank  of  the  Stour.  Spelt  Stutton, 
H.R. ;  Stuton,  H.R. ;  Stutone,  Ipm.;  Stuttima,  D.B.,  p.  279; 
Stottuna   (with    o    for   u),  D.B.,  p.  31.      A   compound   word; 


NAMES   ENDING   IN   -TON  107 

answering  to  A.S.  stut-tun,  lit.  '  ,stout-f;xrm.'  The  prov.  E. 
stout  means  '  a  gnat,  a  midge ' ;  as  if  it  were  a  farm  infested 
with  midges.  Prof.  Moorman  gives  the  same  explanation  of 
Stutton  in  the  W.  Riding,  and  explains  Midgley  as  '  midge- 
lea.'     See  Stuston. 

Sutton.  S.  by  E.  from  Woodbridge.  Spelt  Suttuna,  D.B., 
p.  25 ;  Suthtuna,  D.B.,  p.  78.  A.S.  Suth-tun ;  the  sense  is 
'  south  farm  ' ;  perhaps  with  reference  to  Woodbridge. 

Tannington.  N.W.  of  Framlingham.  An  altered  form ; 
for  Tattenton.  Spelt  Tatingtoii,  H.R. ;  but  Tatintuna,  D.B., 
p.  95.  The  latter  represents  A.S.  Tatan-tuii]  the  sense  is 
'  Tata's  farm.'     Of  the  name  Tata  there  are  eleven  examples. 

Tattingstone.  S.  of  Ipswich.  Spelt  Tattingeston,  Ipm. ; 
Tatingeston,  T.N. ;  Tatingstun,  H.R. ;  Tatyngeston,  Ipm.  Copinger 
gives  many  spellings,  but  it  is  difficult  to  know  whether  they 
belong  to  this  place  or  to  Tanniugton.  The  spellings  above 
answer  to  an  A.S.  form  Tatinges  tun ;  lit.  '  farm  of  Tating,'  or 
'  of  the  son  of  Tata.'  There  is  also  a  form  Tatting,  as  in  Tatting- 
snad  ;  Birch,  C.S.,  i.  295. 

Theberton.  Near  the  E.  coast;  N.  of  Leiston.  Spelt 
Theherton,  H.R. ;  Ipm.  Copinger  notes  the  form  Thehaston  ; 
and  no  doubt  an  s  (or  -es)  has  been  lost,  so  that  it  represents 
Theberteston.  And  Thebert  is  a  very  late  form  of  the  A.S. 
name  Theodbeorht,  0.  Merc.  Theodberht  or  Thedhert.  The  sense 
is  '  Theodbeorht's  farm.' 

Thorington.  S.E.  of  Halesworth.  Spelt  Thoriton,  H.R. ; 
Thuritune,  H.R. ;  Torentuna  (with  T  for  Th),  D.B.,  p.  24.  For 
A.S.  Thoran-tun ;  as  the  g  is  evidently  a  later  insertion.  The 
name  Thora  is  recorded  as  being  that  of  a  daughter  of  Thor- 
berg,  and  wife  of  Harald  Hardrada ;  but  the  A.S.  form  would 
be  There,  with  the  fem.  nom.  ending.  The  name  Thora  (gen. 
Thoran)  would  be  the  corresponding  masculine.  We  may  as- 
sume the  masc.  form  as  being  more  likely  here ;  and  the  sense 
is  then  'Thora's  farm.'  But  an  A.S.  form  Thora  does  not  occur; 
and  it  can  only  be  regarded  as  an  Anglicised  form  of  the  O. 


108  THE   PLACE-NAMES   OF   SUFFOLK 

Norse  Thori,  variant  of  Thuri,  a  common  Scandinavian  name, 
of  which  there  is  an  example  in  Kemble,  Cod.  Dipl.,  iv.  71, 
1.  14 ;  and  Thuri  seems  to  be  merely  a  reduced  form  of  Thurir, 
which  is  a  very  old  Scandinavian  name  and  very  common.  See 
Bjorkman,  p.  158;  Rygh,  p.  259.  Hence  Thorington  is  ulti- 
mately of  Norse  origin.  In  Bardsley's  Surnames,  s.v.  Thor,  is 
an  instance  of  a  late  form  of  Thori,  in  the  entry  "  Orm  iil. 
Thore';  dated  1179  (from  the  Pipe  Rolls). 

Thrandeston.  N.  of  Eye.  Spelt  Thrandestuna,  D.B.,  p.  135. 
The  late  A.S.  form  is  Thrandeston;  in  Thorpe,  Diplomat.,  p.  580. 
Evidently  the  prefix  is  the  gen.  of  Thrand,  a  name  not  recorded 
as  being  of  English  origin,  though  Thrond  is  given  as  being 
that  of  a  Dane..  Of  Norse  origin ;  Rygh  gives  the  0.  Norse 
name  Throndr  or  TJirandr  as  occurring  in  numerous  place-names. 
We  may  drop  the  formative  r  of  the  masc.  nom. ;  this  gives,  as 
the  sense,  '  Thrand's  farm.' 

Thurlston,  Thurleston.  The  church  of  Whittou-cum- 
Thurlston  is  between  Whitton  and  Thurlston  (or  Thurleston) ; 
to  the  N.  by  W.  from  Ipswich.  Spelt  Thurliston,  Ipm.,  p.  258 ; 
but  Turoluestuna,  D.B.,  p.  29  (with  T  for  Th).  For  the  A.S. 
Thurulfes  tun,  also  appearing  as  Turolfes  tun.  Really  of  Norse 
origin;  the  prefix  is  due  to  the  O.  Norse  personal  name  Thuriilf, 
Thorolf,  or  Thorulf]  the  original  O.  Norse  nominative  of  which 
appears  as  Thorolfr  or  Thuridfr ;  corresponding  to  the  A.S. 
Thurwidf.  Hence  the  sense  is  '  Thorolfr's  farm ' ;  and  we  have 
here  a  clear  example  of  Norse  influence.    See  Bjorkman,  p.  102. 

Thurston.  To  the  E.  of  Bury.  Spelt  Thurston,  T.N. ; 
Torstuna,  D.B.,  p.  8.  For  A.S.  Thures  tun ;  i.e.  '  Thur's  farm.' 
Thur  (as  in  Thursday)  is  a  form  of  Thor ;  and  Thur-  occurs,  as 
the  former  element,  in  many  A.S.  names. 

There  was  also  a  Thurstanestun  or  '  Thurstan's  farm '  in 
Suffolk.  But  there  is  nothing  to  show  whether  it  was  Thurston. 
They  seem  to  differ. 

Troston.  N.W.  of  Ixworth.  Spelt  Troston,  Ipm.;  Trostuna, 
D.B.,  p.  172.  Copinger  records  the  form  Throston,  which  is 
evidently  nearer  to  the  original.    But  further,  the  syllable  -ing- 


NAMES   ENDING   IN    -TON  109 

must  be  restored,  as  the  dat.  form  Trostingtune  occurs  in  Birch, 
C.S.,  iii.  630,  in  ^Ifhelm's  Will.  This  takes  ns  back  to  the 
original  form  Throstinga  tun.  From  the  Norse  personal  name 
ThrOstr  (lit.  '  thrush '),  which  yielded  several  place-names  be- 
ginning with  Trost-  or  Troste- ;  see  Rygh.  The  sense  is  '  farm 
of  the  sons  of  Throstr.' 

Ubbeston.  S.W.  of  Halesworth.  Spelt  Uhheston,  H.R. ; 
Huhheston  (with  H  wrongly  prefixed),  T.N. ;  Upbestuna  (with 
pb  for  bb),  D.B.,  p.  269.  Tlie  gen.  Ubbes  suggests  a  nom.  Ubb, 
not  an  A.S.  name,  but  borrowed  from  the  Danish  Ubbi,  recorded 
by  Rygh  as  being  specifically  Danish,  not  Icelandic.  The  sense 
is  '  Ubbi's  farm.' 

Walton.  N.  of  Felixstowe.  Also  (later)  known  as  Walton 
St  Felix,  because  a  monastery  of  St  Felix  was  built  there ;  see 
Felixstowe  (above).  Spelt  Waletoii,  H.R. ;  Waletima,  D.B., 
p.  118.  The  corresponding  A.S.  form  is  Weala  tun,  the  'farm 
(or  enclosure)  of  the  Welshmen ' ;  from  the  nom.  sing.  Wealh, 
a  foreigner,  a  Welshman.  The  same  explanation  applies  to  the 
two  Waltons  in  the  W.  Riding.  In  no  way  connected  with  E. 
wall. 

Wenhaston.  Between  Halesworth  and  Blythburgh.  Spelt 
Wenhaston,  H.R.,  Ipm. ;  but  Wenadestuna,  D.B.,  p.  24.  The 
prefix  occurs  in  an  A.S.  charter,  in  Birch,  C.S.,  ii.  529,  as 
Wikneardes  and  Weneardes ;  it  is  spelt  both  ways  in  the  same 
line.  In  both  cases  an  h  has  been  dropped,  and  the  right  form 
of  the  nom.  is  Wenheard.  The  sense  is  '  Wenheard's  ferm  ' ;  or, 
in  Mercian  spelling,  '  Wenhard's  farm.' 

Westleton.  Spelt  TFes^^e^o/i,  H.R.;  Ipm.;  Westletuna,T>.B., 
p.  66;  but  Westlentuna,  D.B.,  p.  63;  Westledestuna,  D.B.,  p.  319. 
The  last  is  the  fullest  form,  and  must  be  selected.  Westledes 
represents  an  A.S.  west-leodes,  gen.  of  luest-leod,  a  compound 
word.  Leod  means  '  a  man  ' ;  and  ivest-leod  is  '  a  man  from  the 
west.'  (Note  that,  in  Beda's  Eccl.  Hist.,  iv.  1,  the  A.S.  trans- 
lation has  the  gen.  pi.  east-leoda,  i.e.  '  of  men  from  the  east,' 
where  the  Latin  original  has  orientalium.)  The  meaning  is 
'farm  of  the  man  from  the  west.' 


110  THE   PLACE-NAMES    OF   SUFFOLK 

Weston.  Spelt  Weston,  H.R. ;  Westuna,  D.B.,  p.  4.  The 
sense  is  '  west  town  '  or  "'  west  farm.'     It  is  W.  of  Ellough. 

Whitton.  To  the  N.  of  Ipswach.  Spelt  Whitington,  Ipm., 
p.  2.58.  Copinger  cites  several  equivalent  forms,  such  as  Wlii- 
tinton,  Whittington,  Whytingtone,  Whytyngton;  showing  that  the 
name  has  been  contracted.  It  is  the  same  name  as  Whittington 
in  Wore,  spelt  Huitingtun  in  an  A.S.  charter ;  in  Birch,  C.S.,  i. 
497.  The  full  A.S.  form  is  HtuUinga  tun  ;  the  sense  is  '  farm  of 
the  Whitings  '  or  '  of  the  sons  of  White.' 

Winston.  S.  of  Debenham.  Spelt  Wynestou,  H.K.;  Wines- 
tima,  D.B.,  p.  36.  But  Copinger  also  cites  Winerston,  Wynerston ; 
so  that  there  was  once  an  r  before  the  s.  This  represents  the 
A.S.  genitive  Wynheres,  as  occurring  in  Wynheres  stig ;  in 
Birch,  C.S.,  i.  334,  footnote  5.  Wynhere  is  a  known  name.  The 
sense  is  '  Wynhere's  farm.' 

Wiston,  or  WissiNGTON ;  near  Nayland,  Of  course  the 
latter  is  the  older  form.  Spelt  Wysinton,  H.R.  But  this  is 
shortened  from  a  much  more  complex  name,  as  we  learn  from 
iElfflffid's  Will,  in  Birch,  C.S.,iii,602,  where  we  find  Wiswythetun 
mentioned  in  connexion  with  Lavenliam  (about  11  miles  from 
Wiston  in  a  direct  line),  Bildeston  (at  the  same  distance  from  it), 
and  Polstead  (within  four  miles  of  it).  There  is  thus  a  pre- 
sumption that  Wiston  and  Wysinton  (in  H.R.)  are  shortened 
forms  of  Wiswythetun.  Copinger  also  cites  Wisweton  as  a  form 
of  Wiston,  which  clearly  points  to  the  same  form.  If  we  accept 
this  hypothesis,  we  must  enquire  into  its  meaning.  It  is  not 
derived  from  a  personal  name,  but  is  descriptive.  Wis  may  be 
explained  as  short  for  the  A.S.  ivisc,  '  a  meadow,'  a  Avord  dis- 
cussed in  the  Phil.  Soc.  Trans.,  1895-8,  p.  542 ;  and  tuithe  is 
for  withig,  a  withy  or  Avillow.  The  sense  will  then  be  '  farm  of 
the  field- willow.'     For  luisc,  see  Birch,  C.S.,  ii.  412. 

WooLVERSTONE.  Near  the  S.W.  bank  of  the  Orwell.  Spelt 
Wlferstun  (for  Wiilfei^stun),  H.R. ;  Vlverestuna  (with  V  for  U), 
D.B.,  p.  30;  Hulferestuna  (with  H  wrongly  prefixed),  D.B., 
p.  279.  These  represent  the  A.S.  Wulfheres  tun;  and  the  sense 
is  '  Wulfhere's  farm.'     Wulfhere  was  a  very  common  name. 


NAMES   ENDING    IN   -TON,   -TREE,   -WADE  111 

WORLINGTON.  Near  Mildeuhall.  Spelt  Wredelington,  U.K.; 
Wredlingto)i,  Ipm. ;  Wiriiintona,  D.B.,  p.  149.  Copingcr  also 
cites  Wridelyngton,  WretJtelijngton,  WritJielingtoii,  Wredelgnge- 
ton,  &c.  The  prefix  evidently  represents  a  tribal  name,  of  which 
the  gen.  pi.  appears  as  Wredelinga,  Wridelinga,  Wrethelinga,  or 
Writhelinga.  But  all  of  these  are  unknown  forms.  The  sense 
is  '  farm  of  the  Wrethelings  or  Writhelings,'  or  '  of  the  sons  of 
Wrethel  or  Writhel.' 

Wyverstone.  W.  by  N.  from  Mindlesham.  Spelt  Wivers- 
ton,  Ipm.;  Wiiiertestuna  (with  t  for  th),  D.B.,  pp.  57,  oil; 
Wiuerthestuna,  D.B.,  p.  82.  The  prefix  represents  the  A.S. 
Wlferihes,  gen.  of  Wlfertli,  more  correctly  spelt  Wigfrith.  The 
sense  is  *  Wigfrith's  farm.' 

47.  Tree. 

Tree,  in  the  usual  sense,  occurs  in  Pettistree  and  Thed- 
westry. 

Pettistree.  Spelt  Petistre,  Ipm.  Ipm.  has  also  the  form 
Fettes-ho.  The  forms  Peott,  Piott,  and  Piot,  all  given  by 
Sweet,  in  his  Oldest  English  Texts,  p.  536,  may  be  considered 
as  Kentish  variants  of  Pet,  and  so  help  to  establish  that  form. 
Hence  the  sense  is  'Pet's  tr6e.'  See  Wright,  O.E.  Grammar, 
§  93.  The  form  Patta  occurs  in  Birch,  C.S.,  iii.  632,  and  may 
be  related. 

Thedwestry.  This  is  the  name  of  a  hundred ;  and  no 
doubt  the  hundred  met  at  a  particular  tree  that  was  known 
by  this  name.  Spelt  Thedwastre,  Ipm. ;  Thedtvardistre,  H.R. ; 
Theivardestreu,  D.B.,  p.  8;  TJieodivardestreo,  D.B.,  p.  162. 
Copinger  also  gives  Thedwardestre,  Thediuardstree.  The  A.S. 
form  is  Theodweardes-treo ;  O.  Merc,  Theodwardes-treo ;  from 
the  personal  name  Tlieodward.    The  sense  is  '  Theodward's  tree.' 

48.  Wade. 

Wade  represents  the  A.S.  iuced,  a  ford,  shallow  water ; 
cognate  with  L.  uadum.     It  occurs  in  Cattawade. 

Cattawade.  Copinger  gives,  as  old  spellings,  Cataiuade, 
Gattiwade.     Cata  is  precisely  the  O.  Norse  Kata,  gen.  of  Kati, 


112  THE   PLACE-NAMES   OF   SUFFOLK 

a  known  name,  whence  some  known  places  are  derived  (E,ygh). 
The  sense  is  '  Kati's  ford.'  Kati  is  masculine.  Cattawade  is  a 
hamlet  of  Brantham,  near  the  Stour. 

49.    Well. 

Well,  in  the  usual  sense,  occurs  in  Badwell  Ash,  Bardwell, 
Bradwell,  Brightwell,  Bromeswell,  Elmswell,  Eriswell,  Herrings- 
well,  Orwell,  Sizewell,  and  Wordwell ;  eleven  examples. 

Badwell  Ash.  Three  miles  due  N.  of  Elmswell  station, 
and  near  Great  Ashfield.  Copinger  gives,  as  other  names, 
Ashfeld  parva  and  Badewelle  Asfelde.  Badewelle  represents 
the  A.S.  Badan  wella,  i.e.  '  Bada's  well.'  Bada  is  a  known 
name.     Cf.  Bcedewyllan  ;  Birch,  C.S.,  iii.  240  (bottom). 

Bardwell.  N.  of  Ixworth.  Spelt  Berdeivell,  T.N.,  H.R.  ; 
Berdeivella,  D.B.,  p.  221;  Beordewella,  D.B.,  p.  171.  The  forms 
point  directly  to  an  A.S.  form  Beordan  wella,  i.e.  'Beorda's  well.' 
No  instance  of  the  name  Beorda  is  known ;  but  it  may  be 
related  to,  or  an  error  for,  the  known  form  Bearda,  which  occurs 
in  Beardan-lg,  in  the  A.S.  Chronicle,  and  in  Bardan-lg  (the 
0.  Merc,  form)  in  the  Life  of  St  Oswald  by  iElfric. 

Bradwell.  To  the  S.  by  W.  of  Yarmouth.  Spelt  Brade- 
well,  H.R. ;  answering  to  the  A.S.  cet  thdm  hrddan  luellan,  i.e. 
'at  the  broad  well.'  Cf.  Bradfield.  There  are  at  least  six 
Brad  wells. 

Brightwell.  Near  Bucklesham  ;  S.  of  Martlesham.  Also 
known  as  Brightwell-cum-Foxhall.  Spelt  Brihtewella,  D.B., 
p.  211 ;  for  A.S.  cet  thdm  beorhtan  wellan,  lit.  'at  the  bright  (or 
clear)  well.' 

Bromeswell.  Near  Melton.  Spelt  Brumesiuelle,  Ipm. 
(which  has  also  such  forms  as  Brumfeld,  Brumlegh) ;  Brumes- 
uelle,  D.B.,  p.  214;  Bromeswella,  D.B.,  p.  25;  Brameswella,  D.B., 
p.  75.  Of  the  three  forms  in  D.B.,  we  must  select  the  first,  as 
agreeing  with  Ipm. ;  whilst  the  second  can  be  accounted  for  by 
the  fact  that  Norman  scribes  frequently  wrote  om  for  um  when 
the  a  is  short.  The  third  form,  with  a,  must  be  wrong,  as  it 
suits  neither  the  old  nor  the  modern  spellings. 


NAMES    ENDING   IN    -WELL  113 

The  name  Brum  occurs  in  Searle ;  and  we  may  conclude 
that  the  A.S.  form  must  have  been  Brumes-wella,  meaning 
'  Brum's  well.'  The  modern  spelling  was  probably  affected  by 
the  influence  of  A.S.  hrom,  '  broom ' ;  i.e.  the  plant  so  called. 
This  was  certainly  the  case  with  Bromsgrove  (Wore),  which 
really  means  '  Brem's  grove,'  as  the  old  spellings  prove ;  see 
Duignan's  Place-Names  of  Worcestershire. 

Elmswell.  Spelt  Elmeswell,  H.R. ;  Elmeswellan,  D.B., 
p.  168.  Here  Elmes  cannot  refer  to  elm  (the  tree),  as  that 
would  have  formed  the  compound  Elmwell.  The  presence  of 
-es  shows  that  Elmes  represents  the  gen.  case  of  a  proper  name, 
which  has  certainly  been  contracted.  It  must  be  short  for 
Elmeres,  which  occurs  in  Birch,  C.S.,  iii.  58,  1.  4  from  bottom, 
and  is  also  spelt  Almeres  in  the  last  line  of  the  page.  Elmeres 
or  Almeres  represents  JElm,eres,  a  later  form  oi  JElfmares ;  and 
the  oldest  form  of  Elmswell  must  have  been  j^lfmd'ves  wella, 
i.e.  '  ^Elfmier's  well'  There  is  an  Elmsall  in  Yorkshire,  which 
similarly  represents  an  A.S.  ^Ifmwres  halh,  i.e.  '  ^Elfmier's 
haugh.'  This  is  Prof  Moorman's  explanation  in  his  Place- 
Names  of  the  West  Riding. 

Eriswell.  Spelt  Erswelle,  T.N. ;  Ersivell,  H.P. ;  Heres- 
wella  (with  H  wrongly  prefixed),  D.B.,  p.  244.  Evidently  a 
contracted  form,  as  shown  by  the  spelling  Everesivell  in  Ipm., 
p.  6.  Everes  represents  the  O.  Merc.  Eferes,  A.S.  Eoferes,  gen. 
of  Ejer,  Eofer,  personal  name,  the  literal  sense  being  '  a  boar ' ; 
and  it  is  cognate  with  L.  aper.  The  meaning  is  '  Efer's  well.' 
Compare  Eversden,  Cambs, ;  Eversley,  Hants. 

Herringswell.  Spelt  Heringeswell,  Ipm. ;  Haringwell, 
T.N. ;  but  older  forms  are  Hernigaiuella,  D.B.,  p.  223 ;  Hern- 
ingawella,  D.B.,  p.  235  ;  Eyrningwella,  D.B.,  p.  156.  As  D.B. 
sometimes  has  e  for  A.S.  y,  these  forms  all. come  from  an  A.S. 
Hyrninga  wella,  i.e.  '  Hyrnings'  well,'  or  '  well  of  the  Hyrniugs ' 
or  '  of  the  sons  of  Hyrn  ' ;  or,  possibly,  '  of  Horn.'  At  any  rate, 
Hyrning  is  a  similar  name  to  Horning;  for  which  see  HORN- 
INGSHEATH. 

G.  A.  S.  Octavo  Series.     No.  XLVI.  8 


114  THE   PLACE-NAMES    OF    SUFFOLK 

Orwell.  The  name  of  a  river ;  but  the  river  was  named 
from  a  well  from  which  it  took  its  rise ;  and  Orwell,  Carabs., 
occurs  as  a  place-name.  The  Cambs.  place-name  represents  an 
A.S.  oran-iuella,  where  oran  is  the  combining  form  of  ora, 
a  border,  brink,  edge,  or  margin  ;  and  the  sense  is  '  well  beside 
the  brink,'  or  'well  beneath  a  brink.'  In  the  A.S.  Chronicle, 
there  is  mention  of  a  river  Arvje  (i.e.  'arrow'),  which  is  sup- 
posed to  be  the  Orwell.  It  may  be  the  same  river  but  it  is 
not  the  same  name. 

SIZEV7ELL.  A  hamlet  in  the  parish  of  Leiston.  It  is  due 
E.  of  Leiston,  and  on  the  coast.  I  know  of  no  old  spelling  ;  but 
Copinger  quotes  Siswell.  Cf.  Siston,  Glouc.  The  A.S.  form 
was  probably  Sisan  luella,  i.e.  '  Sisa's  well.'  The  A.S.  Sisa  is 
implied  in  the  form  Siso,  quoted  in  Searle  from  a  foreign 
source. 

WoRDWELL.  Two  miles  N.W.  of  Ingham  station  (Kelly). 
Spelt  Wridewella,  D.B.,  p.  172 ;  Wride-ioella,  Birch,  C.S.  iii. 
219,  1.  4.  For  A.S.  wrida  luella,  'well  of  the  thicket,'  or  'of 
(the  clump  of)  young  shoots ' ;  where  wrida  is  the  gen.  pi.  of 
wrld.  The  A.S.  wrld  is  the  pro  v.  E.  ride,  'the  quantity  of  wood 
growing  from  one  stump,  a  root-stock  in  coppice.'  '  A  ride  of 
hazle,  &c.,  is  a  whole  plump  of  sprigs  growing  out  of  one  root.' 
Particularly  used  of  the  hazel ;  cf.  A.S.  hcesel-wrid,  '  hazel-ride.' 

50.    WicH,  Wick. 

The  suffix  -wich  or  -luick  represents  the  A.S.  wic,  a  dwelling; 
hence,  a  village.  It  occurs  in  Dunwich,  Hardwick,  Ipswich,  and 
Walberswick. 

Dunwich.  Spelt  Donewic,  T.N.;  H.R.;  Dunemdc,  D.B,, 
p.  62.  A  trisyllabic  form ;  representing  the  A.S.  Dunan  wlc, 
i.e.  '  Duna's  village.'  The  name  Duna  occurs  in  Dunan-heafod 
and  Dunan-hyl  (both  in  Kemble).  It  is  possible  that  the  name 
was  suggested  by  an  older  one.  Beda,  Hist.  Eccl.  ii.  15,  has 
"in  ciuitate  Domnoc";  for  which  the  A.S.  version  has  "on 
Dommocceastre."     Domnoc  is  not  English ;  but  may  be  Celtic. 


NAMES    ENDING    IN    -WICK,    WOM)  115 

In  fact,  we  are  told  that  it  is  so  in  McClure's  British  Place- 
names,  p,  173,  note  1,  where  it  is  said  that  Dumnoc  involves 
a  term  meaning  '  deep,'  with  -oc  as  an  adjectival  termination ; 
i.e.  (as  I  suppose)  the  sense  is  'deepish  ';  and  it  signifies  'a  port 
with  a  deep-water  approach.'  The  base  is  the  Indo-germanic 
*dubnos,  *dumnos,  'deep,'  whence  the  Old  Irish  fn-domain, 
'deep,'  Welsh  dwfn  (fern,  dofn);  see  Stokes-Fick,  Wortschatz 
der  keltischen  Spracheinheit,  p.  153. 

Hardwick.  a  new  parish,  one  mile  W.  b}^  S.  from  Bury 
(Kelly).  Spelt  Herdwice,  R.B.  ;  Hereivic,  H.R.  Herdewic 
answers  to  the  A.S.  Heordewlcum,  dat.  pi. ;  spelt  Heordeivican 
in  Thorpe,  Diplomat.,  p.  594 ;  for  the  A.S.  wtc  was  frequently 
used  in  the  pi.,  as  meaning  '  dwellings.'  The  A.S.  heorde  is  the 
gen.  of  heorde,  a  herd,  a  flock.  The  sense  is  '  herd-dwellings ' 
or  '  herd-village.'  The  place  referred  to  in  Thorpe  is  Hardwick 
in  Northants.  The  derivation  from  the  dat,  pi.  may  explain 
why  the  suffix  remains  as  tuick,  and  did  not  become  luich. 

Ipswich.  Spelt  Gipeswiche,  later  Gippewich,  Ipm.;  Gipeiuic, 
H.R.;  Gypeswich,  Robert  of  Gloucester;  Gipeswiz,  D.B.,  p.  19 
(with  z  =  ts,  for  ch).  A.S.  Gipes  wTc;  A.S.  Chronicle,  an.  993 
(Parker  MS.) ;  where  Gipes  is  the  gen.  of  Gipi  (later  Gipe), 
a  name  not  otherwise  known.  Thus  the  sense  is  'Gipi's  village.' 
The  G  before  i  was  pronounced  as  mod.  E.  y,  and  the  former 
i  was  short ;  so  that  Gipi  was  pronounced  yippy.  The  Norman 
disliked  initial  y  before  i,  and  dropped  it.  Hence  it  took  the 
sound  of  Ippy's  luich,  and  finally  Ipswidi. 

Walberswick.  Spelt  Walberdeswyk,  H.R.  Other  spellings 
in  Copinger  do  not  tell  us  any  more ;  but  the  A.S.  equivalent 
form  is  obvious,  viz.  Wealhheorhtes  wlc ;  O.  Merc.  Walhberhtes 
tulc.  Wealhheorht  occurs  three  times,  and  in  two  instances  is 
also  found  as  Walhert.  The  sense  is  '  Walhberht's  village.'  Or 
-wick  may  have  been  derived  from  the  dat.  pi.  wlcuni;  see 
Hardwick. 

51.    Wold. 

Wold  represents  the  O.  Merc,  wald,  A.S.  weald,  a  wood, 
forest ;  just  as  old  is  from  O.  Merc,  aid,  A.S,  eald.     Many  wolds 

8—2 


116  THE   PLACE-NAMES   OF   SUFFOLK 

have    lost   their  trees,  and  are  now  bare.     The  only  Suffolk 
example  is  Southwold. 

SoUTHWOLD.  S]ie\t  Suthwold,  iT^m.,  H.'R.;  Suthwald,'S..'R.; 
Sudwolda,  D.B.,  p.  182  (with  d  for  voiced  th).  From  A.S.  silth, 
south  ;  the  sense  is  obvious. 

52.  Wood. 

Only  in  Hazlewood  ;  two  miles  N.W.  of  Aldeburgh  station 
(Kelly).     The  sense  is  obvious. 

53.  Worth. 

Worth  is  related  to  mod.  E.  worth,  value ;  and  meant 
a  property,  holding,  farm,  an  enclosed  homestead.  The  A.S. 
form  is  worth.  Examples  are  :  Braiseworth,  Chelsworth, 
Dunningworth,  Halesworth,  Hepworth,  Ickworth,  Ixworth, 
Timworth,  Worlingworth. 

Braiseworth,  or  Brayes worth.  Spelt  Breisworth  in 
Kirby.  S.  of  Eye.  Spelt  Bryseworth,  H.R. ;  Breseworth,  Ipm. ; 
Briseworde,  D.B.,  p.  80.  Copinger  also  notes  forms  beginning 
with  Brayss,  Breis,  Bres.  The  use  of  y,  e,  i  after  Br  points  to 
the  A.S.  2/  or  y;  and  suggests  a  form  *Brysan,  gen.  of  *Brysa. 
That  there  was  such  a  name  is  further  supported  by  the  occur- 
rence of  such  place-names  as  Bris-ley,  Bris-ton,  Norf ;  and  still 
more  by  the  forms  Bruse-lowe,  Bris-ingham,  Brise-rvike  in  the 
index  to  Ipm.  I  suggest  the  sense  '  Brysa's  farm.'  Cf.  A.S. 
brysan,  to  bruise. 

Chelsworth.  Near  Bildeston.  Spelt  Ghelesworth,  H.R. ; 
but  Cerleswrda,  D.B.,  p.  176.  The  D.B.  form  answers  to  A.S. 
Ceorles  wyrth;  Birch,  iii.  312;  which  refers  to  this  very  place. 
Ceorles  is  the  gen.  of  A.S.  ceorl,  a  husbandman,  countryman. 
The  sense  is  '  husbandman's  farm ' ;    lit.  '  churl's  farm.' 

Dunningworth.  Tunstall-cum-Dunningworth  ;  near  Alde- 
burgh. Duniworda,  D.B.,  p.  130.  Copinger  gives  the  forms 
Donyngivorthe,  Dimnyngivorthe.  For  A.S.  Dunninga  wyrth ;  i.e. 
'  farm  of  the  Dunnings'  or  'of  the  sons  of  Dunn.'  Both  Dunn 
and  Dunning  are  known  names, 


NAMES   ENDING    IN    -WORTH  117 

Halesworth.  So  spelt;  H.R. ;  T.N. ;  Ipm.  D.B.  has 
Halesuuorda,  p.  25 ;  and  Healesuurda,  p.  87.  Tlic  name 
Halington  (prob.  Hallington,  Lines.)  oeeurs  in  Bireh,  C.S. 
i.  453  (bottom);  with  referenee  to  a  tribe  or  family  of  Halings  ; 
from  a  proper  name  *H8el  or  *HaI,  not  otherwise  known.  The 
sense  may  be  '  Huel's  farm.' 

Hepworth.    Spelt  Hepwrth,  H.R. ;  Hepworda,  D.B.,  p.  170. 

Copinger  also  has  Hepeivorth,  Hipeiuorth.  There  does  not  seem 
to  be  any  reason  why  the  prefix  may  not  be  the  A.S.  lieope, 
M.K  hepe,  a  hip,  i.e.  the  fruit  of  the  dog-rose.  The  sense  may 
be  '  hip-worth,'  or  '  farm  of  wild  roses.'  There  is  a  Hepworth  in 
Yks.,  for  which  D.B.  has  Heppeword.  Cf.  Heope-bricge,  lit. 
'hip-bridge';   Birch,  C.S.,  iii.  567,  1.  4. 

ICKWORTH.  Near  Horningsheath.  Spelt  Ikeiuorth,  Ipm. ; 
kkeivortha,  D.B.,  p.  154,  where  /  is  omitted  by  mistake.  Spelt 
Iccaworth  in  a  late  A.S.  charter,  Kemble,  iv.  222, 1.  2.  For  Iccan- 
worth ;  Iccan  is  the  gen.  of  Icca,  and  occurs  in  Iccan-ora  (ill  spelt 
Iccannore),  in  Birch,  C.S.,  i.  99.     The  sense  is  '  Icca's  farm.' 

IxwoRTH.  N.E.  of  Bury.  Spelt  Ixewortlie,  R.B. ;  Ixeiurth, 
H.R. ;  Ixewr'Q,  D.B.,  p.  174;  Copinger  also  has  Giswortha, 
Gyscewurde.  It  is  the  A.S.  Gyxeiveorth;  Birch,  C.S.,  iii.  219. 
The  prefix  must  have  been,  originally,  a  weak  genitive,  and  the 
vowel  must  have  been  i,  not  y ;  for  G  remains  hard  before  y, 
but  is  lost  before  i  Hence  the  original  form  must  have  been 
Gixan-iuearth ;  and  the  sense  is  '  Gixa's  farm ' ;  where  Gixa 
represents  Gisca.  Gisca  may  stand  for  *Giseca,  a  diminutive 
of  Gisa,  which  is  a  known  name.     Cf.  Gis-wulf. 

TiMWORTH.  N.  of  Bury.  Spelt  Timwrtha,  D.B.,  p.  1G5  ; 
Timeworda,  D.B.,  p.  221.  For  Tlman  lueorth;  and  the  sense 
is  '  Tima's  farm.'     Tima  is  a  known  name. 

WORLING WORTH.  Spelt  Wirlingwo7'th,R.U.;  Wyrlingwortlm, 
D.B,,  p.  175.  But  Copinger  also  has  Weiring lurthe,  Wilrincga- 
wertha,  so  that  the  rl  was  once  Ir.  The  A.S.  form  Wilrincga- 
wyrth  is  in  Thorpe,  Diplomat.,  p.  567.      Wilrincga  is  short  for 


118  THE   PLACE-NAMES   OF   SUFFOLK 

Wilheringa ;  and  the  sense  is  '  farm  of  the  Wilherings,'  or  '  of 
the  family  of  Wilhere.'  Wilhere  is  a  known  name,  Cf 
Wilheringa  (for  Wilheringa  wic) ;    Birch,  C.S.,  ii.  141. 

54.    Yard. 

From  A.S.  geard,  an  enclosure.     It  occurs  in  Bruisyard. 

Bruisyard.  N.W.  of  Saxmundham.  Spelt  Bursyard,  H.R.; 
Bursyerd,  Ipm. ;  Buresiart,  D.B.,  pp.  83,  117.  The  evidence 
seems  to  be  conclusive  as  to  the  fact  that  an  older  form  was 
Burxs-geard.  Biires  is  certainly  a  much  contracted  form,  but, 
fortunately,  it  is  easy  to  restore  it  by  comparison  with  Burslem 
(Staffs.)  and  Buscot  (Berks.).  In  the  former,  Burs-  represents 
the  O.  Merc.  Burgwardes,  A.S.  Burgiueardes,  see  Duignan, 
Place-Names  of  Staffs. ;  and  in  the  latter.  Bus-  is  short  for 
Burs-,  and  represents  the  same  prefix ;  see  my  Place-Names  of 
Berks.  Hence  we  may  explain  Bruisyard  as  representing 
'  Burhward's  yard  '  or  '  enclosure.' 

55.    Some  other  names. 

Having  thus  considered  the  names  which  certainly  seem  to 
be  compounds,  with  known  suffixes,  I  consider  first  Cornard, 
which  is  of  like  formation,  and  lastly  some  names  that  are  of 
different  or  uncertain  formation,  and  cannot  well  be  included 
in  the  foregoing  sets. 

Cornard.     To  the  E.  of  Sudbury  ;  called  Great  Cornard 
there  is  also  Little  Cornard,  not  far  off.     Spelt  Cornerth,  T.N. 
Cornerd,    H.R. ;    Gornerthe,    H.R. ;    Gornerda,    D.B.,    p.    159 
Gornierda,  D.B.,  pp.  12,  223.     The  forms  with  th  must  be  the 
more  original ;  and  of  these  we  may  take  Cornerth  as  the  type. 
The  former  part  of  the  word  is  corn  (A.S.  corn) ;  the  latter  part 
is  probably  not  the  A.S.  eorthe,  mod.  E.  earth,  but  the  rarer 
A.S.    earth,   plough-land,    not    given    in    the    A.S.    Dictionarj^, 
because  it  is  commoner  in  the  '  modified '  form  ierth  or  yrth 
or  irth ;    see  irth  in  Bosworth   and   Toller,  and  earth  (2)  in 
N.E.D.     The  actual  form  earth  (dat.  eai^the)  occurs  in  Birch, 


SOME   OTHER   NAAfES  119 

C.S.,  ii.  195,  ].  22 ;  and  in  the  compound  earth-land  in  Birch, 
C.S.,  i.  502,  1.  6  (with  medwe-land,  i.e.  meadow-land,  in  the 
following  line) ;  and  again  in  the  same,  ii.  40,  1.  4  from  the 
bottom.  The  sense  is  'plough-land  for  corn.'  The  form 
Cornierda  in  D.B.  may  be  compared  with  the  compound 
for-ierth  in  Birch,  C.S.,  ii.  255,  1.  14.  If  this  be  right,  we 
may  add  it  to  the  list  of  compounds ;  for  which  reason  I  place 
it  here. 

Barrow.  A  well-known  word,  meaning  a  funeral  mound 
or  tumulus,  or  sometimes  simply  a  hill ;  from  the  M.E,  berwe, 
herewe,  A.S.  heorge,  dat.  of  heorh  (O.  Merc.  herJi),  a  hill, 
a  barrow. 

Beccles.  Spelt  Beccles,  H.R. ;  Becles,  T.N. ;  D.B.,  pp.  6, 
178.  We  may  also  compare  Beclinge,  Ipm. ;  Becclinga,  D.B., 
p.  116  ;  where  Beccling  is  a  patronymic,  formed  from  the  name 
Beccel.  Beccles  is  the  gen.  of  Beccel ;  and  stands  alone  as 
indicating  the  name  of  the  possessor  of  the  original  settlement ; 
just  as  we  might  call  a  farm  Smith's,  meaning  Smith's  farm. 
Hence  the  sense  is  '  Beccel's,'  meaning  a  settlement  of  Beccel. 
In  the  A.S.  Life  of  St  Guthlac,  ch.  vii.,  there  is  mention  of 
'  Beccel  the  priest.' 

Beck  Row.  A  hamlet  of  Mildenhall.  The  prov.  E.  heck 
means  '  a  small  stream ' ;  from  the  O.  Norse  bekker,  a  brook. 
"  Towards  the  Fens  [near  Mildenhall]  are  several  large  Streets 
as  big  as  ordinary  Towns,  called  by  the  Inhabitants,  Rows ;  as 
West  Row,  Beck  Row,  and  Holywell  Row";  Kirby. 

BOULGE.  Spelt  Bulge,  Ipm. ;  Bidges,  D.B.,  pp.  77,  134. 
The  name  must  have  been  quite  recent,  at  the  time  of  the 
compilation  of  D.B.  in  1086.  It  is  obviously  not  of  English, 
but  of  Norman  origin  ;  and  D.B.  preserves  the  correct  form. 
Bulges  is  an  Old  Norman  plural,  answering  to  a  later  0.  French 
houges,  given  by  Godefroy  s.v.  Bouge,  s.m.  'terrain  inculte  et 
couvert  de  petites  brandes.'  It  therefore  signifies  'lands  not 
yet  cultivated,  but  covered  with  heather.' 


120  THE   PLACE-NAMES   OF   SUFFOLK 

Brome,  or  Broome.  Pronounced  as  E.  broom.  S.  by  W.  of 
Diss.  Spelt  Brom,  D.B.,  pp.  59,  117.  The  njod.  E.  broom;  as 
a  plaut-name.     Broom  was  there  abundant. 

BuRES  St  Mary.  On  the  Stour ;  W.  of  Nayland.  A  bridge 
across  the  river  leads  to  Bures  in  Essex,  Spelt  Bures,  H.R., 
T.N.,  Ipm. ;  Bare,  D.B.,  p.  223.  The  A.S.  bur,  a  bower,  cottage, 
was  masc. ;  with  the  pi.  buras.  But  this  would  have  given 
a  mod.  E.  Bowers.  The  preservation  of  a  long  u  only  takes 
place  in  words  of  French  origin.  The  right  explanation  seems 
to  be  that  it  is  an  Old  Norman  bures,  pi.  of  bure,  which  was  not 
a  word  of  Latin  origin,  but  merely  borrowed  from  the  O.H.G. 
and  A.S.  bur.  The  forms  bur,  bure,  are  given  as  modern  Norman 
words  in  Moisy's  Diet,  of  the  Norman  dialect ;  and  he  quotes 
the  Lat.  form  bui-us  from  a  Caen  chartulary;  adding  that 
Norman  also  possesses  the  dimin.  buron,  a  hut.  This  only 
affects  the  phonology,  not  the  sense.  We  may  explain  Bures 
to  mean  '  bowers,'  i.e.  a  collection  of  cottages  or  huts.  In  the 
A.S.  Chron.,  an.  1094,  we  find  mention  of  a  'castel  set  Bures,' 
i.e.  a  castle  at  Bures  in  Normandy,  in  the  department  of  Seine 
Inferieure.     This  is  a  proof  that  Bures  is  Norman. 

Capel.  S.E.  of  Hadleigh.  Kelly  names  Capel  St  Mary 
and  Capel  St  Andrew.  Spelt  Capele,  H.E. ;  Capeles,  D.B., 
p.  25.  It  is  interesting  to  see  that  D.B.  uses  the  plural.  The 
word  is  Norman  ;  capeles  is  the  pi.  of  capele,  a  chapel.  The 
F.  chapelle  is  capele  in  Old  Norman ;  the  latter  form  occurs 
in  La  Chanson  de  Roland,  1.  52.  Tlie  Welsh  form  is  capel, 
not  capele. 

Clare.  Spelt  Clare,  H.R. ;  Ipm  ;  Claram  (Lat.  ace),  D.B., 
p.  218,  From  the  A.S.  Clare,  a  personal  name;  a  witness  bearing 
this  name  signs  an  A.S.  charter  dated  A.D.  949 ;  Birch,  C.S.,  iii. 
38.  It  can  hardly  have  been  a  true  A.S.  name ;  it  was  very 
likely  borrowed  from  L.  cldrus,  illustrious.  Earls  of  Clare  took 
their  name  from  this  place, 

COLNEIS.  The  name  of  a  hundred,  which  comprised  the 
land  lying  between  the  rivers  Orwell  and  Deben.  Spelt 
Colneise,   Colneyse,   II,R. ;    Colneyse,   Ipm.     D.B.    has    Colenese, 


SOME    OTHKR    NAMES  121 

p.  118;  Colenesse,  p.  2*3.  This  is  a  name  of  extraordinary 
difficulty ;  and  I  can  only  guess  at  it.  The  forms  -nese,  -nesse 
may  represent  the  dat.  of  the  A.S.  ness,  a  promontory,  a  head- 
land ;  while  -neise,  -neyse  may  be  the  O.  Norse  nesi,  dat.  of  the 
O.N.  nes,  with  the  same  sense.  The  hundred  of  Colness  does, 
in  fact,  consist  of  one  long  headland,  which  narrows  down  to 
Langer  Point.  The  name,  accordingly,  is  '  Col-ness  ' ;  where  the 
sense  of  Col-  is  not  known.  But  both  Cole  and  Colne  are 
river-names,  and  either  will  suit.  I  suggest  that  one  or  other 
of  these  names  was  the  old  name  of  the  Deben  ;  it  has  been 
shown  above  that  Deben  is  a  new  name,  due  to  the  place-name 
Debenham. 

Combs.  S.  of  Stowmarket.  Spelt  Combes,  Ipm. ;  Gambes, 
T.N. ;  Camhas,  D.B.,  p.  21.  Cambas  is  precisely  the  A.S. 
cambas,  pi.  of  camb,  a  comb,  a  crest,  a  top.  The  sense  is 
'  crests '  ;  with  reference,  as  I  suppose,  to  hill-tops.  The 
spelling  with  a  shows  that  it  is  quite  a  different  word  from 
combe,  a  valley,  so  common  in  place-names. 

CoPDOCK.  S.W.  of  Ipswich.  Spelt  Gopedok,  H.R. ;  Goppe- 
doc,  H.R. ;  Goppedock,  Ipm. ;  Goppedhak,  T.N.  Obviously  for 
A.S.  copped  dc,  a  '  copped '  or  pollarded  oak.  The  sense  is 
therefore  'pollard-oak.'  In  Birch,  C.S.,  ii.  241,  we  have  the 
same  expression,  but  in  the  accusative  case;  'on  tha  coppedan 
ac,'  i.e.  to  the  pollard  oak. 

Cove.  There  is  a  North  Cove  and  a  South  Cove,  near 
Covehithe,  which  is  between  Lowestoft  and  Southwold.  Cove 
is  from  the  A.S.  cofa,  a  cove,  a  cave,  a  place  of  shelter.  The 
spelling  Coua  occurs  in  D.B.,  p.  25,  66. 

Ellough.  S.  by  W.  of  Beccles.  Kelly  calls  it  Ellough,  or 
Willingham  All  Saints,  or  Willoiigh.  The  last  name  seems 
to  be  a  jumble  of  the  other  two.  H.R.  has  Elr/  villa ;  D.B., 
p.  6,  has  in  Elga  et  in  WillingaJtam.  Copinger  gives  also  the 
forms  Elloiue  and  Helw.  This  difficult  name  can  be  fully 
explained  by  comparison  with  a  name  in  Yorkshire.  In  York- 
shire Place-Names,  by  J.  Horsfall  Turner,  p.  114,  there  is  a  note 
that  a  place   now  called   Hellaby  is  spelt  in   D.B.  as   Elgebi 

8—5 


122  THE    PLACE-NAMES    OF    SUFFOLK 

(twice).  The  suffix  -by  shows  that  the  prefix  Elge-  is  Norse, 
and  the  fact  that  the  modern  name  begins  with  H  shows  that 
this  is  a  D.B.  spelling  of  Helge-.  Rygh  shows  that  the  O.  Norse 
personal  name  Helgi  is  extremely  common,  and  that  a  large 
number  of  place-names  beginning  with  Helge-  (including 
Helgeby)  are  derived  from  it.  Moreover,  Helge  would  regularly 
become  Helwe  in  Mid.  Eng.  (cf.  the  form  Heliu  above),  and 
would  then  necessarily  become  Hellow ;  cf.  mod.  Yi.  fellow  from 
O.  'Norse  felagi.  The  dropping  of  initial  h  in  Norman  is  usual, 
because  it  was  not  pronounced ;  and  in  this  instance  it  has 
affected  even  the  English  form  ;  so  that  Hellow  became  Ellow, 
of  which  Ellough  is  a  mere  variant.  This  loss  of  h  is  aptly 
illustrated  by  the  occurrence  in  Ipm.  of  a  Lines,  place-name 
that  is  spelt  both  as  EUowe  and  as  Hellowe ;  without  any  modern 
equivalent.  Hence  the  D.B.  form  Elga  is  exactly  the  O.N, 
Helga,  gen.  of  Helgi ;  and  the  sense  is  '  Helgi's,'  i.e.  '  Helgi's 
settlement.' 

Eyke.  N.E.  of  Woodbridge.  Spelt -%A;e,  Ipm.  From  O.N, 
eik,  an  oak  ;  gen.  eikar,  eikr,  dat.  eik.  The  sense  is  either  '  oak,' 
or  'at  the  oak,'  in  the  dative.  The  diphthong  ey,  representing 
Mid.  Eng.  ei,  is  characteristic  of  Norse. 

Groton.  W.  of  Hadleigh.  Spelt  Groten,  H.R. ;  Grotene, 
Ipm. ;  Grotma,  D.B.,  pp.  13,  158.  It  is  obvious  that  the  suffix 
was  not  originally  -ton,  and  that  it  must  be  otherwise  explained. 
Note  that  the  modern  form  should  be  Groten,  as  it  is  sometimes 
written.  The  nearest  A.S.  form  is  grot-an,  nom,  pi,  of  grata, 
prob.  '  a  particle  of  grit,'  found  in  mere-grota,  a  sea-pebble,  a 
pearl  (see  A.S.  Diet.).  The  sense  would  then  be  '  sands '  or 
'  gritty  plains.'  Cf.  prov.  E.  greet,  grit,  gravel,  also  found  in  the 
forms  grote,  grute,  grut ;  and  prov.  E.  gritten,  adj.  sandy.  The 
form  grot-en  might  be  adjectival. 

HoxNE,  The  name  of  a  hundred ;  the  modern  village  is 
near  the  Waveney,  almost  due  N.  of  Debenham.  Spelt  Hoxene, 
H.R. ;  Hoxana,  D.B.,  p.  197.  The  form  exactly  answers  to  A.S. 
Hoxena,  gen.  pi.  of  a  nom.  pi.  Hoxan,  which  might  very  well 
represent  the  name  of  a  small  tribe  of  settlers,  just  as  we  find 
mention  of  the  Wixan  (see  the  A.S.  Dictionary),  and  of.  the 


SOME    OTHER    NAMES  123 

celebrated  tribe  of  Seaxan.  We  may  therefore  explain  the 
name  as  meaning  *  settlement  of  the  Hoxan.'  I  owe  this 
suggestion  (which  is  to  me  convincing)  to  Mr  A.  Anseombe, 
whom  I  consulted  in  this  instance.  The  modern  pronunciation, 
as  Hoxen,  results  from  the  loss  of  the  inflectional  -e. 

Iken.  On  the  S.  bank  of  the  river  Aide,  ai\d  almost  due 
W.  of  Aldeburgh.  Spelt  Ikene,  H.R.;  T.N. ;  Ykene,  R.B.,  T.N. 
Here  the  form  suggests  an  A.S.  gen.  pi.  Iccena ;  allied  to  the 
proper  name  Icca  which  occurs  in  ICKWORTH.  The  nom.  pi. 
would  be  Iccan ;  and  the  sense  would  be  '  a  settlement  of  the 
Iccan,'  or  '  of  the  followers  of  Icca.'  Not  from  lea,  with  one 
c,  as  this  would  certainly  give  Iche.  Cf.  A.S.  Icena,  the  river 
Itchen. 

Landguard.  "  Landguard  Fort  stands  on  the  extreme 
Western  point  of  this  parish " ;  Kelly,  s.v,  Felixstow,  The 
present  name  is  an  ingenious  adaptation,  as  if  it  were  a  '  land- 
guard,'  or  a  fort  to  guard  the  land,  which  is  not  a  distinctive 
feature  in  forts ;  they  all  do  the  same.  In  Philips'  map  it  is 
Landger  Point,  where  the  latter  syllable  is  -ge7\  In  the 
Beauties  of  England,  1813,  xiv.  235,  it  is  Langnard,  with  the 
former  syllable  as  La7i- ;  and  at  p.  273  of  the  same  we  read 
that  "  here  was  a  ridge,  two  miles  along  the  sea,  called  Langer- 
ston,  dangerous  to  ships";  so  that  in  1813  the  name  was  really 
Langer  or  Langar,  which  may  be  compared  with  Langar  in 
Notts.  So  also  in  Kirby,  p.  91 : — "  Langer-Fort,  and  not  Land- 
guard  Fort,  as  it  is  corruptly  and  vulgarly  called."  There  is 
still  a  Langer  Common  (misspelt  Landguard  on  Ordnance  map) 
in  Felixstowe  parish.  The  etymology  is  easy,  viz.  from  A.S. 
lang  gdra,  i.e.  '  long  gore,'  which  precisely  describes  it.  A  gore 
is  a  promontory,  or  a  triangular  piece  of  a  land  with  a  pointed 
end  ;  from  A.S.  gar,  a  spear,  point. 

LoES.  The  name  of  a  hundred.  Spelt  Lose,  H.R.  ;  T.N. ; 
Losa,  H.R. ;  D.B.,  pp.  11,  215.  It  perhaps  represents  A.S. 
Hlossan,  gen.  of  Hlossa,  a  personal  name.  We  find  it  in 
Hlossan-ham ;  in  Birch,  C.S.,  i.  207.  If  this  be  right,  the 
sense  refers  to  a  settlement  '  of  Hlossa.'  Cf.  Hlos-hrycg,  Hlos- 
wudu ;    both  in  Kemble's  Index, 


124  THE    PLACE-NAMES   OF   SUFFOLK 

LoUND.  N.  by  W,  from  Lowestoft.  Spelt  Lunda,  D.B., 
p.  6.  Ipm.  has  Lund,  Lound  (both  in  Notts.).  Not  English, 
but  a  well-known  Norse  word ;  from  O.  Norse  lundr,  a  grove ; 
cf.  Lund  in  Sweden.     The  sense  is  '  grove.' 

Mellls  (see  p.  104).  Spelt  Melles,  R.B. ;  Ipm. ;  D.B.,  p.  181. 
It  is  explained  by  comparing  it  with  the  prov.  E.  meal,  "  a 
sandbank  or  sand-hill,  frequent  in  proper  names:  gen.  in  the 
plural  "  ;  E.D.D.  From  Norvv.  mel,  a  sandbank  along  a  lake  or 
river-course ;  O.  Norse  melr,  a  sandbank  overgrown  with  bent- 
grass,  or  gen.  a  sandbank  whether  overgrown  or  bare ;  frequent 
in  Icel.  local  names. 

Onehouse.  Spelt  Onhus,  H.R.;  Anhus,  D.B.,  p.  160;  Anehus, 
D.B.,  p.  311  ;  Annhus  (not  Anuhus),  D.B.,  p.  121.  It  is  the  A.S. 
dti  h'lis,  lit.  '  one  house.'  It  is  now  a  hamlet  of  scattered  houses, 
but  there  is  still  a  Onehouse  Hall.  In  the  Beauties  of  England, 
1813,  xiv.  210,  we  are  told  that  "on  the  site  of  the  old  hall... 
a  farmhouse  has  been  built."  Probably  that  old  hall,  or  a  much 
older  house  on  the  same  site,  was  the  original  One  House. 

Rede,  or  Reed.  S.S.W.  of  Bury.  Spelt  Rede,  Ipm, ;  Reoda, 
D.B.,  p.  155  ;  Reda,  D.B.,  pp.  202,  222.  For  A.S.  Readan ;  as  in 
Readan-clif,  -cumb,  -die,  -flod,  -ford,  &c. ;  all  in  Kemble's  Index. 
Readan  is  the  gen.  case  of  Read, '  the  Red,'  still  common  as  Reade, 
Read,  Reid,  &c.     The  sense  is  '(settlement)  of  the  Red  one.' 

Rishangles.  S.  by  E.  from  Eye.  Spelt  Rishanggeles,  Ipm.; 
but  Risangra,  D.B.,  p.  85.  The  prefix  is  A.S.  rise,  a  rush.  The 
suffix  seems  to  be  the  A.S.  hangra,  of  which  the  true  sense  is 
'a  hanging  wood  on  a  hill-side';  see  the  Crawfurd  Charters, 
p.  134.     The  sense  is  'rushy  slopes,  with  trees  upon  them.' 

Snape.  N.  of  the  Aide  above  Aldeburgh.  Spelt  Snape 
H.R. ;  Snapes,  D.B.,  p.  71.  Cf.  A.S.  sncep,  as  in  the  following  : 
lit  o3  mearc  andlang  diin  and  sncep,  '  out  as  far  as  the  boundary 
along  the  down,  &c.' ;  the  sense  of  sna^p  being  here  unknown  ; 
see  Birch,  C.S.,  iii.  362.  But  the  E.D.D.  has  snape,  '  a  spring, 
a  moist,  boggy  place  in  a  field ' ;  known  in  Dors.,  Som.,  and 
Devon.  This  may  be  the  right  explanation  here  as  Snape  is  in 
a  low  situation. 


SOME    OTHER    NAMES  125 

Stoven.  N.W.  of  Southwold.  Spelt  Stoiine,  D.B.,  p.  106, 
with  w  for  v;  Stouone,  D.B.,  p.  251,  with  u  for  v.  In  the  Cursor 
Mundi,  8036,  is  the  line  : — "  Thai  three  stod  on  a  stouen,"  they 
three  stood  on  a  stovin;  where  other  MSS.  have  stalke,  a  stalk, 
or  stocke,  a  stock.  And  the  E.D.D.  explains  stovin,  a  stump  or 
stake,  the  part  of  a  hawthorn  left  in  a  hedge  after  '  splashing ' 
it ;  Leicestershire.  A.S.  stofn,  a  stem,  tree-stump ;  Icel.  stofn, 
a  stump  of  a  cut  tree.  This  is  a  good  example  of  the  frequently 
trivial  origin  of  a  place-name.  It  merely  means  '  stump  of  a  cut 
tree.' 

Thwaite.  Near  the  river  Dove ;  S.  by  W.  from  Eye.  A 
well-known  word  in  the  North ;  from  the  O,  Norse  thiueit,  a 
clearing  in  woods.  So  that  the  sense  is  '  a  clearing.'  It  is 
chiefly  remarkable  for  its  occurrence  so  far  to  the  South.  There 
are  two  more  Thwaites  in  Norfolk. 

Weybread.  S.W.  of  Bungay.  Spelt  Weybred,  H.R. ;  Weij- 
bredd,  Ipm. ;  Weibrada,  D.B.,  pp.  11,  98.  A.S.  wegbrwde,  lit. 
*  way-breadth,'  i.e.  the  broad  plant  by  the  wayside ;  a  name  for 
the  common  plantain,  from  its  flat  growth.  It  merely  means 
'  plantain.' 

WooLPiT.  Between  Bury  and  Stowmarket.  Spelt  Wulpet, 
H.R.;  Wolpet,  Wulpet,  Ipm.;  Wlpet,  H.R.;  Wlfpeta,  D.B.,  p.  164. 
It  answers  to  A.S.  Widfpyt ;  a  wolf-pit ;  a  pit  in  which  to  catch 
wolves.  The  dat.  pi.  widfpyttun  is  in  Birch,  C.S.,  iii.  184.  The 
sense  is  therefore  '  wolf-pit.'  We  should  particularly  notice  the 
dialectal  (Suffolk)  pet,  in  place  of  the  A.S.  pyt,  E.  pit.  The 
same  form,  pet,  occurs  in  Old  Frisian ;  and  Widpet  may  have 
been  due  to  Frisian  influence. 


56.    Concluding  Remarks. 

Owing  to  the  large  number  of  place-names,  this  investiga- 
tion has  necessarily  taken  up  much  space ;  but  room  must  be 
found  for  a  brief  statement  of  general  results. 

The  traces  of  Celtic  are  extremely  slight,  even  among  the 
river-names.     The  Kennet  and  the  Ouse  are  of  Celtic  origin; 


126  THE    PLACE-NAMES    OF   SUFFOLK 

but  the  Butley  river  was  named  from  Butley ;  the  Deben  from 
Debenham ;  the  Breton,  afterwards  shortened  to  Bret,  from 
Brettenham;  the  Gipping,  from  Gipping ;  the  Thet,  from 
Thetford ;  the  Box,  from  Boxford ;  the  Yox,  from  Yoxford ; 
and  the  Aide,  from  Aldeburgh.  The  Blythe,  the  Orwell,  and 
the  Waveney  are  clearly  English  in  form.  The  Lark  seems  to 
have  been  made  out  of  Lackford,  and  the  Linnet  is  its  playfully 
named  companion.  The  Welshman,  or  '  foreigner,'  is  alluded  to 
in  Walpole,  Walton,  and  Walsham. 

There  were  certainly  Frisians  settled  in  Suffolk,  viz.  at 
Freston,  Friston,  and  Fressingfield.  The  Frisian  e  (for  AS.  a) 
is  apparent  in  Bredfield ;  it  is  even  likely  that  the  Frisian  e  (for 
A.S.  y)  is  seen  in  Gedding,  Hertest,  Kedington,  Nedging,  and 
Woolpet  (older  form  of  Woolpit). 

The  traces  of  Danes  and  Norsemen  are  not  very  numerous, 
but  quite  clear  and  decided.  There  are  four  names  ending  in 
-hy — Ashby,  Barnby,  Risby  and  Wilby ;  Baylham  probably 
contains  the  0.  Icelandic  hceli,  '  a  farm.'  Ingham  contains  the 
O.N.  eng,  a  meadow ;  and  Kirkley  and  Kirton  show  the  Norse 
form  of  'church.'  Bungay,  Eyke,  Lound,  Thwaite  are  all 
Norse;  and  Norse  names  or  prefixes  occur  in  Blundeston, 
Cratfield,  Drinkstone,  Flixton,  Gisleham,  Grundisborough, 
Gunton,  Hasketon,  Kesgrave,  Kettlebaston,  Kettleburgh,  Lax- 
field,  Ringsfield,  Ringshall,  Risbridge,  Thrandeston,  Ubbeston, 
and  Uggeston.     There  is  a  Norse  suffix  in  Lowes-toft. 

Finally,  there  are  even  traces  of  Norman ;  as  in  Boulge, 
Bures,  and  Capel ;   and  the  -le-  in  Walsliam-le- Willows. 

As  to  the  names  that  are  purely  English,  they  show  decided 
traces  of  belonging  to  the  Mercian  or  Midland  dialect,  as  dis- 
tinguished from  the  Anglo-Saxon  or  Southern;  which  is  a 
matter  of  no  small  importance.  It  shows  that  we  may  fairly 
include  Suffolk  amongst  the  rather  limited  number  of  counties 
that  have  helped  to  build  up  that  East-Midland  dialect  which 
was  destined  to  supersede  all  others  and  to  become  the  speech 
of  the  empire. 


INDEX 


-acre,  4 
Acton,  92 

Akenhara,  48 
Aldborough,  7 
Alder  ton,  93 
Aldham,  48 
Aldriugliam,  48 
Alpheton,  93 
Ampton,  93 
Ashbocliing,  71 
Ashby,  13 
Ashfield,  24 
Aspall,  42 
Assiugton,  93 
Athelington,  93 

Babergh  Hundred,  5 

-bach,  4 

Bacton,  93 

Badingham,  or  Baddingham,  48 

Badley,  77 

Badwell  Ash,  112 

Ballingdon,  18 

Bardwell,  112 

Barham,  48 

Barking,  71 

Barnaby  or  Barnby,  13 

Barnham,  48 

Baruingham,  49 

Barrow,  119 

Barsham,  49 

Barton,  94 

Battisford,  31 

Bawdsey,  21 

Baylham,  49 

Bealings,  71 

Beccles,  119 

-beck,  5 


Beck  Row,  119 
Bedtield,  24 
Beduiglield,  25 
Belstead,  86 
Belton,  94 
Benacre,  4 
Benhall,  43 
Bentley,  77 
-bergh,  5 
Bergholt,  70 
Beyton,  94 
Bildeston,  94 
Blackbourn  Hundred,   10 
Blakenham,  49 
Blaxball,  43 
Blundeston,  94 
Blythburgh,  7 
Blythford  or  Blyford,  31 
Blything  Hundred,   72 
-borough,  6 
Bosmere  Hundred,  82 
Botesdale,  15 
Boulge,  119 
-bourn,  10 
Boxford,  32 
Boxstead,  87 
Boyton,  94 
Bradfield,  25 
Bradley,  77 
Bradwell,  112 
Braisworth,  116 
Bramfield,  25 
Bramford,  32 
Brampton,  94 
Braudeston,  95 
Brandon,  18 
Brantham,  49 
Bredfield,  25 


128 


THE    PLACE-NAMES   OF   SUFFOLK 


Brettenbam,  49 

Bricett,  Si 

-bridge,  10 

Brigbtwell,  112 

Brockford,  32 

Brockley,  77 

Brome,   120 

Bromeswell,  112 

-brook,  11 

Browstou,  95 

Bruisyard,  118 

Bruudisb,  20 

Bucklesbam,  49 

Bulcamp,  14 

Bungay,  21 

Bures  St  Mary,  120 

Burgate,  40 

Burgb     (in    Colneis     Hundred,     near 

Felixstowe),  6 
Burgh  (Castle)  on  the  Waveuey,  6 
Burgh  (3  miles  N.  W.  of  Woodbridge),  6 
Burstall,  86 
-bury,  6 

Bury  (St  Edmund's),  8 
Butley,  78 
Buxhall,  43 
-by,  12 

-camp,  14 
Campsey  Ashe,  22 
Capel,   120 

Carlford  Hundred,  32 
Carlton,  95 
Cattawade,  111 
Cavendish,  20 
Cavenham,  50 
Charsfield,  25 
Chattisham,  50 
Chedburgh,  9 
Chediston,  89 
Chelmondiston,  95 
Chelsworth,  116 
Chevington,  96 
Chillesford,  32 
Chilton,  96 
Clare,  120 
-clay,  14 

Claydon  Hundred,   18 
Clopton,  96 
Cockfield,  26 


Coddeuham,  50 
Colneis  Hundred,  120 
Combs,  121 
Coney  Weston,  96 
Cookley,  78 
Copdock,   121 
Cornard,  118 
Gorton,  97 
Cosford  Hundred,  32 
Cotton,  97 
Cove,  121 
Covehithe,  68 
Cowlinge,  72 
Cransford,  33 
Cratfield,   26 
Greeting,   72 
Cretiugham,  50 
Crowfield,  26 
Gulford,  33 
Gulpho,  68 

-dale,  15 
Dalham,  50 
Dalliughoo,  69 
Darmsden,  19 
Darsham,  50 
Debach,  4 
Debeuham,  51 
-den,  15 
-dene,   15 
Bennington,  97 
Denham,  51 
Denstone,  98 
Depden,  15 
-don,  18 
-down,   18 

(Santon)  Downham,  51 
Drinkstone,   98 
Dunningworth,  116 
Dunwich,  114 

Easton,  98 
-edish,  20 
Edwardstone,  98 
Eleigh,  78 
Ellough,  121 
Elmham,  52 
Elmsett,  85 
Elmswell,  113 
Elveden  or  Elden,  16 


INDEX 


129 


Eriswell,  113 
Erwarton,  98 
Euston,  98 
Exning,  72 
-ey,  21 
Eye,  21 
Eyke,  122 

Fakenham,  52 
Falkenham,  52 
Farnham,  52 
Felixstowe,  90 
Felsham,  52 
-field,  24 
Finborough,  5 
Finningham,  53 
-fleet,  30 
Flemptou,  99 
Flixton,  99 
Flowton,  99 
-ford,  31 
Fornham,   53 
Foxhall,  43 
Framlingham,  53 
Framsden,  16 
Freckenham,  53 
Fressingfield,  26 
Freston,  99 
Friston,  99 
Fritton,  100 
Frostenden,  16 

-gate,  39 
Gazeley,  78 
Gedding,  73 
Gedgrave,  40 
Gippiug,  73 
Gisleham,   53 
Gislingham,  54 
Glemham,  54 
Glemsford,   33 
Gorleston,  100 
Gosbeek,  5 
-grave,  40 
Groton,  122 
Grundisburgh,  7 
Gunton,  100 

Hacheston,  100 
Hadleigh,  78 


-hale,  42 

Halesworth,  117 

-hall,  42 

-ham,  47 

Hard  wick,  115 

Hargrave,  41 

Harkstead,  87 

Harleston,  100 

Hartest,  70 

Hartismere  Hundred,  82 

Hasketon,  100 

-haugh,  65 

Haughley,  79 

Haverill,  67 

Hawkedon,  19 

Hawstead,  87 

Hazlewood,  116 

-heath,  66 

Helmingham,  54 

Helmley  or  Hemley,  79 

Hemingstone,  101 

Hengrave,  41 

Henham,  54 

Henley,  79 

Henstead,  87 

Hep  worth,  117 

Herringfleet,  30 

Herringswell,  113 

Hessett  or  Hedgsett,  85 

Heveniugham,  54 

Higham,  54 

-hill,  67 

Hinderclay,  14 

Hintlesham,  55 

Hinton,  101 

Hitcham,  55 

-hithe,  67 

-hoe,  68 

Holbrook,  11 

HoUesley,   79 

-holt,  70 

Holton,   101 

Homersfield,  26 

Honington,  101 

Hoo,  68 

Hopton,  102 

Horham,  55 

Horningsheath  or  Horringer,  66 

Hoxne,  122 

Hundon,  16 


130 


THE   PLACE-NAMES   OF   SUFFOLK 


Hunston,  102 
Huntingfield,  27 
•hurst,  70 

Icklingham,  55 
Ickworth,   117 
Iken,  123 
Ilketshall,  43 
■ing,  71 
Ingham,  55 
Ipswich,  115 
Ixworth,  117 

Kediugton,  102 
Kelsale,  44 
Kentford,  33 
Kenton,  102 
Kersey,  22 
Kesgrave,  41 
Kessingland,  75 
Kettlebaston,  103 
Kettleburgh,  9 
Khkley,  80 

Kirton  or  Kirkton,  103 
Knettishall,  44 
Knodishall,  44 

Lackford,  33 
Lakeuheath,  66 
-land,  75 
Landguard,  123 
Langham,  56 
Lavenham,  56 
Lawshall,  45 
Laxfield,  27 
Layham,  56 
Leavenheath,  67 
Leiston,  103 
Letheringham,  56 
Levington,  103 
-ley,  77 
Lidgate,  40 
Lindsey,  23 
Linstead,  88 
Livermere,  83 
Loes  Hundred,  123 
Lothingland  Hundred,  76 
Lound,  124 
-low,  81 
Lowestoft,  92 


Marlesford,  34 
Martlesham,  57 
-meadow,  82 
Melford,  34 
Mellis,  124 
Melton,  103 
Mendham,  57 
Mendlesham,  57 
-mere,  82 
Metfield,  27 
Mettingham,  57 
Mickfield,  27 
Middleton,   104 
Milden,  74 
Mildenhall,   45 
Monewden,  16 
Moulton,  104 
Mutford,  35 

Nacton,  104 
Nay  land,  76 
Nedging,  74 
Needham  (Market),  58 
Nettlestead,  88 
Newbourn,  10 
Newton,  104 
Norton,  105 
Nowton,  105 

Oakley,  80 
Occold,  70 

Ofton  or  Offton,  105 
Onehouse,  124 
Ortord,  35 
Orwell,  114 
Otley,  80 
Oulton,  105 
Owsden,  17 

Pakefield,  28 
Pakenham,  58 
Palgrave,  41 
Parham,  58 
Peasenhall,  45 
Pettaugh,  65 
Pettistree,   111 
Playford,  35 

Plomesgate  Hundred,  40 
Polstead,  88 
-pool,  83 


INDEX 


131 


Poslingford,  35 
Prestou,  105 

Earasholt,  70 
Eattlesden,  17 
Eaydon  or  Eeydon,  20 
Eede,   121 
Eedgrave,  42 
Eedishara,  58 
Eedlinglield,  28 
Eendham,  58 
Eendlesham,  59 
Eickinghall,  46 
Eingsfield,  28 
Eingshall,  46 
Eisbridge  Hundred,   10 
Eisby,  13 
Eishangles,  124 
Eougharu,  59 
Eumburgh,  7 
Eushbrooke,  11 
Eushmere,  83 

Samford  Hundred,  36 
Sapiston,  105 
Saxham,  59 
Saxmundham,  59 
Saxstead,  88 
Semer,  88 
-set,  84 

Shadingfield,  28 
Shelland,  76 
Shelley,  80 
Sbimpling,  74 
Shipmeadow,  82 
Shotley,  80 
Shottisham,  59 
Sibton,  105 
Sizewell,  114 
Snape,  124 
Soham,  60 
Somerleyton,  106 
Somersham,  60 
Sotherton,  106 
Sotterley,  80 
Southolt,  70 
Southwold,  116 
Spexhall,  46 
Sproughton,  106 
-stall,  86 


StanningfieUl,  29 
Stansfield,  29 
Stanstead,  88 
Stanton,  106 
-stead,  86 
Sternfield,  29 
-stoke,  89 
Stoke-by-Clare,  89 
Stoke-by-Nayland,  89 
Stoke  Ash,  89 
-stone,  89 
Stonham,  60 
Steven,  125 
-stow,  89 

Stow  Hundred,  89 
(West)  Stow,  89 
Stowmarket,  90 
Stradbroke,  11 
Stradishall,  46 
Stratford,  36 
Stuston,  106 
Stutton,  106 
Sudbourn,   10 
Sudbury,  9 
Sutton,  107 
Sweffling,  75 
Swillaud,  77 
Syleham,  61 

Tannington,  107 
Tattingstone,  107 
Theberton,  107 
Thedwestry  Hundred,  111 
Thelnetham,  61 
Thetford,  36 
Thingoe  Hundred,  69 
Thorington,  107 
Thorndon,  20 
Thornham,  61 
-thorpe,  91 
Thorpe,  91 

Thorpe-by-Ixworth,  91 
Thorpe-Morieux,  91 
Thrandeston,  108 
Thredling,  75 
Thurlow,  81 
Thurlston,  108 
Thurston,  108 
Thwaite,   125 
Tim  worth,  117 


132 


THE    PLACE-NAMES   OF   SUFFOLK 


-toft,  92 
-ton,  92 
Tostoek,  89 
-tree,   111 
Trimley,  81 
Troston,  108 
Tuddenbam,  61 
Timstall,  86 

Ubbeston,  109 
Ufford,  37 
Uggesball,  47 

-wade,  111 

Walberswick,  115 

Waldingfield,  29 

Waldringfield,  29 

Walpole,  84 

Walsbam,  62 

Walton,  109 

Wangford  (in  Blytbing  Hundred,  N.W. 

of  Southwold),  37 
Wangford  (in  Lackford  Hundred,  S.W. 

of  Brandon),  37 
Wangford  Hundred,  38 
Wantisden,  18 
Wasbbrook,  12 
Wattisfield,  29 
Wattisham,  62 
-well,  112 
Wenham,  62 
Wenhaston,  109 
Westerfield,  30 
Westhall,  47 
Westborpe,  91 
Westleton,  109 
Westley,  81 
Weston,  110 
(Market)  Weston,  103 
Wetberden,  18 
Wetberingsett,  85 


Weybread,   125 
Wbatfield,  30 
Wheluetham,  63 
Whepstead,  88 
Wberstead,  88 
Wbitton,  110 
-wicb,  114 
-wick,  114 

Wickbambrook,  12,  64 
Wiekham  Market,  63 
Wickbam  Skeitb,  64 
Wilby,  13 

Wilford  Hundred,  38 
Willingbam,  64 
Willisbam,  64 
Wingfield,  30 
Winston,  110 
Wissett,  85 

Wiston  or  Wissiugton,  110 
Witbersdale,  15 
Withersfield,  30 
Witnesbam,  64 
Wixoe  or  Wbixoe,  69 
-wold,  115 
-wood,  116 
Woodbridge,  11 
Woolpit,   125 
Woolverstone,  110 
Wordwell,  114 
Worliogbam,  65 
Worlington,  111 
Worlingwortb,  117 
-worth,  116 
Wortham,  65 
Wratting,  75 
Wrentham,  65 
Wyverstone,  111 

-yard,  118 
Yaxley,  81 
Yoxford,  39 


cambbidge:    printed  by  john  clay,  m.a.  at  the  univeksity  peess 


CAMBRIDGE   ANTIQUARIAN   SOCIETY. 

REGENT   PUBLICATIONS. 

[A  Complete  Catalogue  can  he  had  on  applicatio7i.] 

Proceedings,  1911-12.     Michaelmas  Term.     With  Communica- 
tions, No.  LXI.     pp.  1—59.     Plates  I— IV.     3s.  Qd  net. 

Allen,  F.  J.,  M.D.,  Church  Spires  of  Cambridgeshire.  Brindley,  H.  H., 
M.A.,  Notes  on  Mediaeval  Shi^Ds  (n.  p.).  Crewdson,  Bev.  Canon,  M.A., 
Psychical  Phenomena  in  Ancient  Mythology  (n.  p.).  Johns,  Mrs,  Early 
developments  in  Egyptian  civilization  (n.  p.).  Petrie,  Professor  W.  M. 
Flinders,  D.C.L.,  Eoman  Portraits  from  Egypt  (n.  p.).  Eushe,  Eev.  J.  P., 
Origin  of  S.  Mary's  Gild,  Cambridge,  and  its  connection  with  Corpus 
Christi  College.  Stokes,  Eev.  Dr,  F.S.A.,  Old  Trees  of  Cambridge  (n.  p.). 
Report  for  year  1910-11. 

Proceedings  Lent  and  Easter  Terms  1911-12.     7s.  6o?.  net. 

Lent  Term.     With  Communications,  No.  LXIL     pp.  60 — 117. 
Plates  V— X. 

Forster,  E.  H.,  M.A.,  Excavations  at  Corstopitum  (n.p.).  James,  M.  R., 
Litt.D.,  F.B.A.,  Earliest  Inventory  of  Corpus  Christi  College.  Johns, 
Eev.  C.  H.  W. ,  Litt.D.,  Debt  of  Europe  to  the  Ancient  East  (n.  p.). 
Moir,  J.  Eeid,  Palaeolithic  Workshop  of  mid-palaeolithic  age  at  Ipswich 
(n.  p.).  Myers,  C.  S.,  M.D.,  Sc.D.,  Primitive  Music  (n.  p.).  Seward, 
Professor  A.  C,  F.E.S.,  Churches  of  Gothland.  Walker,  Rev.  F.  G.,  M.A., 
Eoman  Pottery  Kilns  at  Horningsea  (to  be  printed  later).  Wyatt,  A.  J., 
M.A.,  Anglo-Saxon  Eiddles  (n.  p.). 

Easter  Term.     With  Communications,  No.  LXIII.     pp.  118 — 
200.     Plates  XI — XIII  and  other  illustrations. 

Brindley,  H.  H.,  M.A.,  Fishing  boats  in  a  window  of  1557  in  Auppegard 
church,  Normandy.  Duckworth,  W.  L.  H.,  M.D.,  Sc.D.,  Eeport  on 
Human  Bones  from  Eoman  and  Saxon  Site  in  Grange  Eoad,  Cambridge. 
Duckworth,  W.  L.  H.,  M.D.,  Sc.D.,  Eeport  on  some  Human  remains  from 
Hyning,  Westmorland.  Gaselee,  S.,  M.A.,  Relic  of  Samuel  Pepys.  Palmer, 
W.  M.,  M.D.,  College  Dons,  County  Clergy  and  University  Coachmen. 
Skeat,  Professor,  Litt.D.,  F.B.A.,  Place-Names  of  Suffolk  (this  paper  is 
printed  in  the  Society's  8vo.  Publications).  Smith,  Eev.  F.,  Comparative 
morphology  of  Scottish  and  Irish  palaeolithic  relics  (n.  p.).  Walker, 
Eev.  F.  G.,  M.A.,  Eoman  and  Saxon  remains  from  Grange  Eoad, 
Cambridge.  Walker,  Rev.  F.  G.,  M.A.,  Palaeolithic  Flint  Implements 
from  Cambridgeshire.    Index  to  Vol.  XVI. 

n.  p.  means  that  the  Communication  has  not  been  printed  in  full. 


CAMBRIDGE   ANTIQUARIAN   SOCIETY. 

RECENT    PUBLICATIONS. 

QUARTO    (NEW)   SERIES. 

I.  The  Dual  Origin  of  the  Town  of  Cambridge.  By  Arthur 
Gray,  M.A.     pp.  32  +  2  plates.     1908.     3s.  %d. 

II.  King's  Hostel,  Trinity  College  Cambridge.  By  W".  D. 
Oaroe,  M.A.,  F.S.A.  pp.  38+  13  plates  and  26  illustrations.  1909. 
10s.  M. 

OCTAVO   SERIES. 

XXXV I.  Cambridgeshire  Place-names.  By  the  Rev.  W.  W. 
Skeat,  Litt.D.     pp.  82.     1911.     5s.  net.     (Second  Edition.) 

XXXVII.  Calendar  of  the  Feet  of  Fines  of  Hutingdon- 
SHiRE.  Edited  by  G.  J.  Turner,  M.A.  pp.  clxiv  +  300.  1913. 
10s.  net. 

XXXVIII.  The  Verses  formerly  inscribed  on  the  Twelve 
Windows  in  the  Choir  of  Canterbury  Cathedral.  Edited  by 
M.  R.  James,  Litt.D.     pp.  42.     1901.     2s.  net. 

XXXIX.  Cambridge  Gild  Records.  Edited  by  Mary  Bate- 
son,  with  a  preface  by  the  Rev,  W.  Cunningham,  D.D.  pp.  176. 
1903.     7s.  6d  net. 

XL.  Annals  of  Gonville  and  Caius  College,  by  John 
Caius.     Edited  by  J.  Venn,  Sc.D.    pp.  xiii  +  431.    1904.    15s.  net. 

XLI.  The  Chaplains  and  the  Chapel  op  the  University 
of  Cambridge  (1256—1568).  By  the  Rev.  H.  P.  Stokes,  LL.D. 
pp.  viii  +  100  +  5  2)lates.     1906.     5s.  net. 

XLII.  The  Place-names  of  Bedfordshire.  By  the  Rev. 
Professor  W.  W.  Skeat,  Litt.D.  5s.  net.  For  Place-names 
of  Cambridgeshire  see  No.  XXXVI  and  Place-names  op 
Huntingdonshire  Proceedings  No.  XLIV. 

XLIII.  The  Riot  at  the  Great  Gate  of  Trinity  College, 
February,  1610—11.  By  J.  W.  Clark,  M.A.,  F.S.A.,  Registrary 
of  the  University,      pp.  xxvi  +  42  +  1  p^aie.      1906.     2s.  6d.  net. 

XLIV.  Outside  the  Trumpington  Gates  before  Peterhouse 
was  founded.  By  the  Rev.  H.  P.  Stokes,  LL.D.,  F.S.A.  pp.  88 
+  2  p>lates.     1  908.     5s.  net. 

XLV.  The  Esquire  Bedells  of  the  University  of  Cam- 
bridge (1250—1910).  By  the  Rev.  H.  P.  Stokes,  LL.D.,  F.S.A. 
pp.  xii  + 140+ 8  ^Zaies.  1911.  (Published  by  private  donation  for 
members  only.) 

EXTRA   PUBLICATIONS. 
Not  gratuitous  to  members. 

LuARD  Memorial  Series  :  Records  of  the  University.  To 
be  completed  in  about  five  volumes,  8vo.  Subscribers,  21s.  Members 
of  the  Society,  15s. 

Vol.  I.  "Grace  Book  A,"  containing  the  Proctors'  Accounts 
AND  other  Records  of  the  University  of  Cambridge  for 
the  years  1454 — 1488.  Edited  by  Stanley  M.  Leathes,  M.A. 
pp.  xliv  +  276.     1897. 

Vol.  II.  "Grace  Book  B,"  Part  I,  1488—1511.  Edited  by 
Mary  Bateson.     pp.  xxvii  +  309.     1903. 

Vol.  III.  "Grace  Book  B,"  Part  II.  Edited  by  Mary 
Bateson.     1905. 


RETURN  TO  the  circulation  desk  of  any 
University  of  California  Library 

or  to  the 

NORTHERN  REGIONAL  LIBRARY  FACILITY 
BIdg.  400,  Richmond  Field  Station 
University  of  California 
Richmond,  CA  94804-4698 

ALL  BOOKS  MAY  BE  RECALLED  AFTER  7  DAYS 

•  2-month  loans  may  be  renewed  by  calling 
(510)642-6753 

•  1-year  loans  may  be  recharged  by  bringing 
books  to  NRLF 

•  Renewals  and  recharges  may  be  made 
4  days  prior  to  due  date 

DUE  AS  STAMPED  BELOW 


NOV  1  3  ZOOJ 


DD20  15M  4-02 


U.  C    BERKFLFY  I  IBRARIES 


CDS134T7flT 


'3;3l2f6^ 


UNIVERSITY  OF  CAUFORNIA  LIBRARY