Skip to main content

Full text of "Old pewter"

See other formats


OLD  PEWTER 


NEWNES'   LIBRARY   OF 


THE  APPLIED   ARTS 


' 


OLD  PEWTER 

BY  MALCOLM  BELL 


LONDON 

GEORGE  NEWNES  LIMITED 

SOUTHAMPTON  ST.  STRAND  W.C 
NEW  YORK 

CHARLES  SCRIBNERS  SONS 


MK 


FRONTISPIECE 


H.  13" 


COMMUNION  FLAGON,  English.     From  Midhurst  Church, 


CONTENTS 

CHAP.  PAGE 

LIST  OF  ILLUSTRATIONS     .....  vii 

PREFACE  ........  xix 

I.  INTRODUCTORY    .......  i 

II.  WHAT  PEWTER  is       ......  7 

III.  How  PEWTER  WAS  WROUGHT    .         .         .         .24 

IV.  PEWTER  BEFORE  THE  FIFTEENTH  CENTURY          .  44 
V.  PEWTER  IN  THE  FIFTEENTH  CENTURY         .         .  63 

VI.  PEWTER  IN  THE  SIXTEENTH  CENTURY         .         .       72 
VII.  PEWTER  IN  THE  SEVENTEENTH  CENTURY   .         .       93 

VIII.  THE  END  OF  THE  STORY 109 

IX.  COLLECTING  AND  DISPLAYING  PEWTER        .         .     120 

X.  SOME  NOTES  ON  THE  ILLUSTRATIONS  .         .         .147 

A  LIST  OF  USEFUL  BOOKS  OF  REFERENCE     .         .164 

INDEX 165 


LIST  OF   ILLUSTRATIONS 


NOTE. — The  initials  in  this  list,  following  the  numbers  of  the  plates,  denote  the 
ownership  of  the  objects  illustrated  and  are  to  be  read  as  follows  : 

A.G.B. — A.  G.  Bell,  Esq.  M.— Robert  Meldrum,  Esq. 

B.M.— British  Museum  R.M. — Mrs.  Ralston  Mitchell 

B.— Miss  Burrell  H.M.— H.  Murray,  Esq. 

W.B. — Baillie  William  Burrell  F.H.N. — F.  H.  Newbery,  Esq. 

L.c. — Lewis  Clapperton,  Esq.  K.R. — Kennerley  Rumford,  Esq. 

F.  &  s. — Messrs.  Fenton  &  Sons  S.K.M. — South  Kensington  Museum 

G.C. — Glasgow  Corporation  w. — Mrs.  Warren 

B.J.— Mrs.  Borough  Johnson  F.C.Y.— F.  C.  Yardley,  Esq. 

TO  FACE 
PLATE  PAGE 

(S.K.M.)  Communion  Flagon,  English,  from  Midhurst 
Church,  Sussex  Frontispiece 

1.  (B.M.)  Dishes  from  a  hoard  of  pewter  found  by  Rev. 

C.  H.  Engleheart  near  Roman  buildings  at  Apple- 
shaw,  near  Andover,  Hants,  1897  .  .  .  i 

2.  (B.M.)  Dishes  from  a  hoard  of  pewter  found  by  Rev. 

C.  H.  Engleheart  near  Roman  buildings  at  Apple- 
shaw,  near  Andover,  Hants,  1897  .  .  .  2 

3.  (B.M.)  Vessels   and    Dish   from    a    hoard   of  pewter 

found  by  Rev.  C.  H.  Engleheart  near  Roman 
buildings  at  Appleshaw,  near  Andover,  Hants, 
1897  4 

4.  (S.K.M.)  The  Gloucester  Candlestick,     xn  century     .         5 

5.  (M.)  Benitier,  probably  Flemish.     No  marks     .         .         6 

6.  1-2.  (F.  &  s.)  Benetiers,    Flemish,  xvm    century ; 

3-4.  (F.C.Y.)  Benetiers,  Flemish,  xvm  century  .          .          8 

7.  (B.)  Measure,  Pepper-box,  Buttons,  and  Mustard-pot        10 

1.  Measure,  German,  4!"  high. 

2.  Pepper-box,  5^"  high. 


3.  Two  Buttons,  2^"  diam. 

4.  Mustard-pot,  4"  high. 


b  vii 


OLD   PEWTER 

TO  FACE 
PLATE  PAGE 

8.  (w.i.)  Pepper-boxes,  Sugar-castor,  Mustard-pots,  and 

Salt-cellars         .         .         .         .         .         .         .10 

1,  3,  4.  Pepper-boxes,  4^"  to  .«;"  high. 

2.  Sugar-castor,  marked  with  Cupid  holding  trumpet  to  mouth 

with  left  hand,  branch  in  right,  HK,  5§"  high. 
5,  6,  7.  Mustard-pots,  3$"  to  5*^  high. 
8,  9, 10,  ii,  12.  Salt-cellars,  10  ,  marked  with  crowned  X. 
Nos.  3,  7,  9  are  a  set. 

9.  (S.K.M.)  Measure,  Dish,  and  Tankard        .         .         .12 

1.  Measure,  English.    Early  XVIII  century. 

2.  Dish,  English.     Middle  of  XVIII  century. 

3.  Tankard,  English.    XVIII  century. 

10.  (F.C.Y.)  Tankards  and  Tappit  Hen  .         .         .13 

1.  Tankard,  Flemish.     XVII  century. 

2.  Tappit  Hen,  Scotch.    XVIII  century. 

3.  Tankard,  French.    XVIII  century. 

11.  (w.)  Toddy     Ladles,    Soup     Ladles,    Table-spoons, 

Fork,  and  Tea-spoon  .         .         .         .         .14 

1.  Toddy  Ladles,  marked  John  Yates.     VR  crowned. 

2.  Small  fork,  no  mark.    Length  7". 

3.  Soup  ladle,  marked    |  JO  |    |  HN  |    |  YA  |    |TES| 

4.  Soup  ladle,  marked    JASKJ    |BER|    |"RY| 

W.  crown  R. 

BEST  METAL 

WARRANTED 

FOR  USE 

5.  Three  table-spoons,  marked 

\Jo]   |HN|  [YA|  |TES|  VR  crowned. 

6.  Tea-spoon,  marked  Ai 

Crowned  head 
PURE 

12.  (H.M.)  Eight  Spoons 16 

1.  Laton  spoon,  5"  long,  found  in  the  Thames  at  Westminster  ; 

an  engraved  line  runs  along  lower  surface  of  stem.    XIV 
century. 

2.  Bronze  spoon,  6f "  long,  found  in  the  City  of  London.    XIV 

century.     Probably  French. 

3.  Laton  spoon,  6"  long,  French.    XIV  century. 

4.  Laton  spoon,  6f"  long,  found  in  the  Thames  in  London,  stem 

of  diamond  section.     Early  XV  century. 

5.  Laton  spoon,  7"  long,  found  in  London.     Early  XV  century. 

Probably  French. 

6.  Bronze  spoon,  5!"  long,  found  in  London.     XV  century. 

viii 


LIST   OF   ILLUSTRATIONS 

TO    FACE 
PLATE  PAGE 

7.  Bronze  spoon,  6"  long,  stem  of  diamond  section,  found  in 

London.     XV  century. 

8.  Bronze   spoon,  6£"  long,   found  in  London.      XV  century. 

Mark  same  as  No.  2. 
For  marks  on  these  spoons  see  Plate  CIV,  facing  p.  148. 

13.  (H.M.)  Eight  Slip-top  Spoons  ...  .18 

1.  Pewter  slip-top  spoon,  6^"  long,  found  at  Bermondsey,  stem  of 

hexagonal  section.    XVI  century. 

2.  Pewter  slip-top  spoon,  6J"  long,  found  in   London.     XVI 

century. 

3.  Pewter  slip-top  spoon,  6f"  long.    XVI  century. 

4.  Pewter  slip-top  spoon,  6J"  long.     XVII  century. 

5.  Pewter  slip-top  spoon,   6|"  long,  found  in  London.     XVI 

century. 

6.  Pewter  slip- top  spoon,  6"  long,  found  in  York  Road,  West- 

minster.    XVI  century. 

7.  Pewter  slip-top  spoon,  6J"  long,   found  in   London.     XVI 

century. 

8.  Pewter  slip-top  spoon,  6J"  long,  found  in  London.      XVII 

century. 
For  marks  on  these  spoons  see  Plate  CIV ',  facing  p.  148. 

14.  (H.M.)  Eight  Spoons        ......       20 

1.  Pewter  spoon,  6A"  long,  found  in  London.     XVI  century. 

2.  Pewter  spoon,  6±"  long.    XVII  century. 

3.  Laton  spoon,  6$"  long,  found  in  London,  stem  of  hexagonal 

section.     XVII  century. 

4.  Laton  spoon,  with  Apostle  top,  6J"  long.    XVII  century. 

5.  Pewter  spoon,  seal  headed,  hexagonal,  6J"  long,   found  at 

Bermondsey.    XVI  century. 

6.  Pewter  sacramental  spoon,  with  Maidenhead  top,  7^"  long, 

found  in  London.     XVI  century. 

7.  Pewter  spoon,  6J"  long,  found  in  London.    XVII  century. 

8.  Pewter  spoon,  with  Apostle  top,  6f"  long,  found  in  London. 

XVI  century. 
For  marks  on  these  spoons  see  Plate  CIV,  facing  p.  148. 

15.  (H.M.)  Eight  Slip-top  Spoons 21 

1.  Pewter  slip-top  spoon,  6"  long,  found  in  York  Road,  West- 

minster. XVI  century.  Same  mark  as  Plate  XIII.,  Fig.  6. 

2.  Pewter  slip-top  spoon,  6"  long,   found  in   London.     XVI 

century.     Same  mark  as  Plate  XIII.,  Fig.  6. 

3.  The  same  as  Fig.  2. 

4.  Pewter  slip-top  spoon,  6J"  long,  found  in   London.     XVI 

century. 

5.  Pewter  slip-top  spoon,  6£"  long.    XVI  century. 

6.  Pewter  slip-top  spoon,   6^"  long,  found  in  London.     XVI 

century.    Same  mark  as  Fig.  2. 

7.  Pewter  slip-top  spoon,  6|"  long.     XVI  century. 

8.  Pewter  slip-top  spoon,  6|"  long,  found  in  Bermondsey.     XVI 

century. 
For  marks  on  these  spoons  see  Plate  CV,  facing  p.  150. 

ix 


OLD   PEWTER 

TO  FACE 
PLATE  PAGE 

1 6.  (H.M.)  Nine  Slip-top  Spoons 22 

1.  Pewter  slip-top  spoon,  6|"  long,  found  at  Bermondsey.     XVI 

century. 

2.  Pewter  slip-top  spoon,  7^"  long,   found  in   London.     XVI 

century. 

3.  Pewter  slip-top  spoon,  6f  "  long,  found  at  Bermondsey.    XVI 

century. 

4.  Pewter  slip-top  spoon,  6}"  long,  found  at  Bermondsey.    XVI 

century. 

5.  Laton  slip-top  spoon,  6J"  long,  found  at  Bermondsey,  has 

remains  of  tin  plating.    XVII  century. 

6.  Laton  slip-top  spoon,  7"  long,  found  at  Bermondsey.      XVII 

century. 

7.  Laton  slip-top  spoon,  6^"  long,  found  in  London,  has  remains 

of  tin  plating.    XVII  century. 

8.  Pewter  slip-top  spoon,   6£"  long,   found  in  London.     XVI 

century. 

9.  Pewter  slip-top  spoon,  6f"  long,  found  in  London.     XVI 

century. 
For  marks  on  these  spoons  see  Plate  CV,  facing  p.  150. 

17.  (H.M.)  Eight  Seal-top  Spoons .          ....       24 

x.  Laton  seal-top  spoon,  sf"  long,  found   in  London.     XVII 
century. 

2.  Laton  seal-top  spoon,  6i"  long.    XVII  century. 

3.  Laton  seal-top  spoon,  6|"  long.    XVII  century. 

4.  Laton  seal-top  spoon,  6£"  long,  found  in   London.     XVII 

century . 

5.  Laton  seal-top  spoon,  6^"  long,  found  in  London.    XVII 

century. 

6.  Laton  seal-top  spoon,  6^"  long,  found   in  London.     XVII 

century. 

7.  Laton  seal-top  spoon,  6£"  long,  found  in   London.     XVII 

century. 

8.  Laton  seal-top  spoon,  6£"  long,  found   in  London.    XVII 

century. 
For  marks  on  these  spoons  see  Plate  CV,  facing  p.  150. 

1 8.  (H.M.)  Eight  Spoons        .         .         .         .         .         .26 

1.  Laton  slip-top  spoon,  7!"  long,   plated   with  tin,  found  in 

London.    XVII  century. 

2.  Pewter  slip-top  spoon,  7"  long,  found  at  Bermondsey.    XVII 

century. 

3.  Laton  slip-top  spoon,  6£"  long,   found   in  London.     XVII 

century. 

4.  Laton  spoon,  7^"  long,  plated  with  tin.    XVII  century. 

5.  Laton  spoon,  sf"  long,  with  remains  of  tin  plating.    XVII 

century. 

6.  Laton  spoon,  7j"long,  plated  with   tin.     Handle   "  Pied  de 

Biche."     Mark  same  as  Fig.  5. 

7.  Laton  spoon,  7^"  long,  with  remains  of  tin  plating.     Handle 

"  Pied  de  Biche."    XVII  century. 

8.  Laton  rat-tailed  spoon,  7^"  long,  found  in  York  Road,  West- 

minster.   Handle  "  Pied  de  Biche."     XVII  century. 
For  marks  on  these  spoons  see  Plate  CVI,  facing  p.  152. 


LIST  OF   ILLUSTRATIONS 

TO  FACE 
PLATE  PAGE 

19.  (H.M.)  Nine  Rat-tailed  Spoons 28 

1.  Laton  rat-tailed  spoon,  8"  long,  plated  with  tin.   Handle  "Pied 

de  Biche."     XVII  century. 

2.  Pewter  rat-tailed  spoon,  7^"  long,  found  in  Newgate  Street. 

Handle  "  Pied  de  Biche."     XVII  century. 

3.  Laton  rat-tailed  spoon,  7"  long,  found  in  York  Road,  West- 

minster, plated  with  tin.    XVII  century. 

4.  Pewter  rat-tailed  spoon,  7^ "  long,  found  in  Bermondsey.   XVII 

century. 

5.  Pewter  rat-tailed    spoon,   7f"  long,  found  in  Bermondsey. 

XVII  century. 

6.  Pewter  rat-tailed  spoon,  7"  long.    XVII  century. 

7.  Laton  spoon,  5^"  long,  plated  with  pewter,  found  in  London. 

Late  XVII  century. 

8.  Bronze  rat-tailed  spoon,  5f"  long.     XVII  century. 

9.  Rat-tailed   pewter    chocolate  spoon,  4"  long,  found   in  the 

Wandle  at  Wandsworth.    XVII  century. 
For  marks  on  these  spoons  see  Plate  CVI,  facing  p.  152. 

20.  (H.M.)  Eight  Spoons  and  Ladles      ....       29 

1.  Pewter  spoon,   8^"  long,  found  in  London.      Late  XVIII 

century. 

2.  Laton   dog-nose  gravy  spoon,  lof"  long,  plated  with   tin. 

XVII  century. 

3.  Pewter  spoon,  7^  "  long,  found  in  Bermondsey.    Late  XVIII 

century . 

4.  Laton   slip-top  ladle,   8f"    long,  the   handle  of   hexagonal 

section.     XVII  century. 

5.  Laton  slip-top  ladle,  8"  long,  found  in  the  City  of  London,  has 

remains  of  tin  plating.    XVII  century. 

6.  Laton  slip-top  ladle,  7^"    long,  found  in  Suffolk.      XVII 

century. 

7.  Laton  slip-top  ladle,  6J"    long,  found  in  London.     XVII 

century. 

8.  Laton  spoon,  seal-headed,    6|"    long,    has  remains  of  tin 

plating,  found  at  Norwich.    XVII  century. 
For  marks  on  these  spoons  see  Plate  CVI,  facing  p.  152. 

21.  (S.K.M.)  Pyx,  Communion  Flagon,  Chalice  and  Paten         30 

1.  Hexagonal  Pyx,  English.     XIV  century, 

2.  Communion  Flagon,  English.    From  Midhurst  Church,  Sussex. 

1677. 

3.  Chalice  and  Paten.     From  Kriswick. 

22.  (S.K.M.)  Communion  Flagon,  Alms  Dish,  and  Patens        32 

1.  Communion  Flagon,  English.     XVII  century. 

2.  Alms  Dish,  Scotch.     XVIII  century.    Marked  AS.  IK. 

3.  Patens  from  a  Church  in  Yorkshire,  English.     XV  century. 

23.  (w.)  Communion  Jug  and  Two  Chalices  .          .        34 

1.  Chalice,  Italian. 

2.  Earthenware  Communion  Jug  mounted  in  pewter. 

3.  Chalice,  Scotch. 

xi 


OLD   PEWTER 

TO  FACE 
PLATE  PAGE 

24.  (F.C.Y.)  Alms  Dish,  Chalice,  and  Laver     ...       36 

1.  Alms  Dish,  English.     XVIII  century. 

2.  Chalice,  English.     XVIII  century. 

3.  Laver,  Scotch.     Late  XVIII  century. 

25.  (B.)  Church  Flagon,  Scotch,  early  xvui  century  type  37 

26.  (K.R.)  Two  Chalices  and  Church  Flagon,  English      .  38 

27.  (F.  &  s.)  Three  Chalices,  English,  xvm  century          .  40 

28.  (L.C.)  Loving  Cup,  Communion  Cup,  and  Porringer  .  42 

1.  Loving  Cup,  English. 

2.  Communion  Cup,  Scotch. 

3.  Porringer,  French.     Probably  XVII  century. 

29.  (G.C.)  Communion  Cup,  Spice-box,  and  Candlestick  .       44 

1.  Communion  Cup.     1804. 

2.  Spice-box,  French.    XVIII  century. 

3.  Candlestick,  French.     XVIII  century. 

30.  Pocket  Communion  Service,  Sacramental  Cruets,  and 

Measure    ........        45 

1.  (L.C.)  Pocket  Communion  Service  in  wooden  case,  bought  in 

Iceland.    Probably  Danish  or  Scotch. 

2.  (F.  &  s.)  Sacramental  Cruet,  Aqua.     French  or  Flemish. 

3.  (F.  &  s.)  Measure,  temp.  Charles  II.      Found   in  Parliament 

Street,  Westminster. 

4.  (F.  &  s.)  Sacramental  Cruet,  Vinum.     French  or  Flemish. 

31.  (w.)  Sacramental  Cruet  and  Five  Measures        .         .       46 

1.  Cruet,  Vinum. 

2.  Gill  Measure,  modern. 

3.  Half  Mutchkin  Measure. 

4.  Old  Half-glass  Measure. 

5.  Half-gill  Measure. 

6.  Gill  Measure. 

32.  (B.)  Altar  Candlesticks  and  Guild  Cup      ...       48 

1,  3.  Two  Altar  Candlesticks.    XVII  century. 

2.  German  Guild  Cup. 

33.  (F.  &  s.)  Two  Altar  Candlesticks,  Flemish,  xvn  cen- 

tury ...  .50 

34.  (R.M.)  Candelabrum,  Flemish,  xvm  century      .          .        52 

35.  (W.B.)  Alms  Dish,  German,  early  xvm  century.         .       53 

36.  (w.)  Candlesticks 54 

1.  A  pair  of  Pillar  Candlesticks,  Scotch.     XVIII  century. 

2.  Flat  Candlestick,  Scotch.    XIX  century. 

3.  A  pair  of  Pillar  Candlesticks,  Dutch.     XVIII  century. 

xii 


LIST   OF   ILLUSTRATIONS 

TO  FACE 
PLATE  PAGE 

37.  (F.  &  s.)  Oil  Lamps,  Taper  Holder,  and  Candlestick  .       56 

1.  4,  5.  Oil  Lamps,  German.    XVIII  century. 

2.  Taper  Holder,  Flemish. 

3.  Pillar  Candlestick,  French. 

38.  Lamp  Time-keepers         .  .  .58 

1.  (A.G.B.)  Lamp  Time-keeper.    XVII  century. 

2.  (F.C.Y.)  Lamp  Time-keeper.    XVII  century. 

3.  4,  5.  (B.)  Lamp  Time-keepers.     XVII  century. 

39.  (F.C.Y.)  Three  Pillar  Candlesticks     .  .         •       60 

40.  (R.M.)  Pillar  Candlesticks,  Tray,  and  Inkstand  .          .       60 

1.  4.  A  pair  of  Pillar  Candlesticks.     XVIII  century. 

2.  Tray.    XVIII  century. 

3.  Inkstand. 

41.  (w.)  Drinking  Cups         .          .          .          .          .          .61 

1.  Cup,  Scotch.     XVIII  century. 

2.  Loving  Cup.     XVIII  century. 

3.  A  pair  of  Wine  Cups.    XVIII  century. 

42.  (M.)  Three  German  Guild  Cups        .          .         .         .62 

1.  Cup,  on  lid  1721,  three  shields,  right  and  left  a  tankard,  above 

I.W.L.,  below  1713,  centre  one  a  church  with  spire. 

2.  Flagon,  on  front  IMS  1706,  Nuremberg  rose  on  bottom  inside, 

on  the  handle  a  shield,  IMK,  a  wall  and  two  turrets. 

3.  Cup,  inscribed  Johannes  George    Reichel  Johannes   Battzer 

Rellurg,  Anno  1693,  Christope  Stutz. 

43.  (R.M.)  Flagons  and  Beaker      .          .          .          .          .62 

1.  Flagon,  German.    XVIII  century. 

2.  Beaker,  Scotch.    XVIII  century. 

3.  Flagon,  German.     XVIII  century. 


44.  (F.  &  s.)  Tankard,  Measures,  and  Beaker          .         .       64 

1.  Tankard,  Scotch.    XVIII  century. 

2,  3.  Measures,  English.     Early  XVIII  century. 


i.  Tankard,  Scotch.  XVIII  century, 
ly  XVII 
4.  Beaker,  engraved.  Early  XVIII  century.  "Mark,  a  crowned 


45.  (W.B.)  Three   Tankards,     German,    xvn     and    xvm 

centuries  .          .          .          .          .         .          .          .66 

46.  (W.B.)  Four  Tankards,  German,  xvn  century     .          .       68 

47.  (R.M.)  Dishes  and  Tankard 69 

1.  Dish,  English.    XVII  century. 

2.  Tankard,  German.     XVIII  century. 

3.  Deep  Dish,  Scotch.    XVI  century. 

xiii 


OLD   PEWTER 


TO   FACE 
PLATE  PAGE 

48.  (S.K.M.)  Tankards  and  Jug 70 

1.  Peg  Tankard,  Danish.     With  engraved  decoration. 

2.  Jug,  English.    XVIII  century. 

3.  Tankard,  Swedish.     1844. 

49.  (S.K.M.)  Ewer,  Cruet  Stand,  and  Guild  Tankard        .        72 

1.  Ewer  and  Cover.     German. 

2.  Cruet  Stand,  German.     Middle  of  XVIII  century. 

3.  Guild  Tankard,  German.    Dated  1645. 

50.  (F.  &  s.)  Cream-jugs,    Salt-cellar,  Measure,   and  Egg 

Cup  .  .  .  .74 

1.  5.  Two  Cream-jugs. 

2.  Salt-cellar. 

3.  Measure. 

4.  Egg  Cup.    All  English.    XVIII  century. 

51.  (W.B.)  A  Pair  of  Covered  Tankards,  German,   xvm 

century     ........        74 

52.  (W.B.)  Three  German  Guild  Tankards       .         .         .76 

53.  (W.B.)  Three  German  Guild  Tankards,  xvn  to  xvm 

century      ........        77 

54.  (S.K.M.)  Measure   or   Tankard,    German,    late    xvn 

century     ........        78 

55.  (G.C.)  Four  Measures,  Scotch,  xvm  century      .         .        78 

56.  (G.C.)  Five  Measures       .         .         .         .         .         .78 

57.  (S.K.M.)  Wine  Taster  and  Measures  ...       80 

1.  Wine  Taster,  English.    XVII  century.     Dug  up  in  Tottenham 

Court  Road. 

2,  3.  Wine  Measures,  German.    XVIII  century. 

58.  (F.C.Y.)  Bowl,    Water-jug,    Rice    Boiler,    Plate,    and 

Measure    ........       82 

1.  Two-handled  Bowl. 

2.  Water-jug,  English.    XVIII  century. 

3.  Rice  Boiler,  French.    XVIII  century. 

4.  Plate,  English.    XVIII  century.    One  of  a  set  of  six. 

5.  Measure,  English.    XVIII  century. 

59.  (w.)  Three  Tappit  Hens 84 

60.  (R.M.)  Tappit  Hen,  xvm  century      ....  85 

61.  (F.  &s.)  Set  of  eight  French  Measures,  xvm  century  .  86 

62.  (F.  c.  Y.)  Jug,  English,  xvn  century  ...  88 

xiv 


LIST   OF   ILLUSTRATIONS 


TO   FACE 
PLATE  PAGE 

63.  (W.B.)  Flagons 90 

1.  Flagon,  probably  Scotch.     XVIII  century. 

2.  Imitation  Chinese  Flagon,  Dutch.     XVII  century. 

3.  Flagon,  German.     XVII  century. 

64.  (G.C.)  Jugs  and  Flagon    ......       92 

1.  Covered  Jug.    Archangel  mark. 

2.  Flagon. 

3.  Cream-jug. 

65.  (F.C.Y.)  Mustard-pot,  Jug,  and  Measure   ...        94 

1.  Mustard-pot,  English.     XVII  century. 

2.  Hot-water  Jug,  Dutch.     Early  XVIII  century. 

3.  Measure,  English.     Early  XVIII  century. 

66.  (w.)  Jugs       .  94 

1.  Jug,  French.     XVIII  century. 

2.  Jug,  English.     XVIII  century. 

3.  Beer-jug,  English.     XVIII  century. 

67.  (F.C.Y.)  Jugs  and  Coffee-pot 96 

1.  Jug,  Dutch.     Marked  crossed  rose. 

2.  Coffee-pot,  Dutch.     No  marks. 

3.  Jug,  Dutch.    Mark,  a  rose  within  a  circle  of  illegible  lettering. 

68.  (B.J.)  Jug,  George  IV 98 

69.  (F.C.Y.)  Bowls         .......     100 

1.  Barber's  Bowl,  English.     XVIII  century. 

2.  Two-handled  Bowl,  Dutch.     XVIII  century. 

70.  (F.C.Y.)  Three  Porringers         .         .          .          .          .100 

71.  (w.)  Bowls  and  Casket  .          .          .         .         .          .      101 

1.  Two-handled  Bowl,  Scotch. 

2.  Casket,  French.     XVI  century. 

3.  Two-handled  Bowl. 

72.  (F.C.Y.)  Bowls,  Dish,  and  Plate          ....      103 

1.  and  4.  A  Pair  of  Two-handled  bowls. 

2.  Dish,  English.     Dated  1689. 

3.  Plate,  English.     XVIII  century.     One  of  six. 

73.  (F.C.Y.)  Salver,  Jugs,  Dish,  and  Plate        .          .          .104 

1.  Wavy-edged  Salver,  with  feet,  Dutch.    Late  XVIII  century. 

2.  Student's  Beer-jug,  German.     XVII  century. 

3.  Dish,  English.     XVIII  century. 

4.  Water-jug,  English.     Late  XVIII  century. 

5.  Wavy-edged  Plate,  English.    Late  XVIII  century.    One  of  a 

set  of  six. 

C  XV 


OLD   PEWTER 

TO  FACE 
PLATE  PAGE 

74.  (F.C.Y.)  Tankards,  Dish,  Mug,  Measure,  and  Tobacco- 

box  .         .         .         .         .         .         .         .         .106 

1.  Tankard,  English.     Early  XVIII  century. 

2.  Dish,  Scotch,  stamped  with  Initials  A.  D.  V.  and  A.  W.     XVI 

century. 

3.  Mug  with  handle,  English.     Early  XVIII  century. 

4.  Measure,  English.    XVII  century. 

5.  Tankard,  English.    XVIII  century. 

6.  Tobacco-box,  English.    XVIII  century. 

75.  (G.C.)  Soup  Tureen,  Russian.     Archangel  mark          .      108 

76.  (F.C.Y.)  Vegetable  Dish,  Tray,  and  Jugs    .          .         .109 

1.  Vegetable  Dish,  English.     Late  XVIII  century. 

2.  Tray  or  Salver,  Dutch. 

3.  Milk-jug,  English.     Late  XVIII  century. 

4.  Water-jug,  English.    XVIII  century. 

77.  (R.M.)  Ewer  and  Basin    .         .         .         .         .         .in 

78.  (F.C.Y.)  Bowl  and  Jug 112 

1.  Bowl,  Flemish.     XVII  century. 

2.  Jug,  Dutch.     Early  XVIII  century. 

79.  (B.J.)  Mustard-pots,     Salt-cellar,      Pepper-box,      and 

Measure    .  114 

1.  and  3.  Mustard-pots.    XVIII  century. 

2.  Salt-cellar.     XVIII  century. 

4.  Pepper-box.     XVIII  century. 

5.  Measure.    XVIII  century. 

80.  (F.  &  s.)   1-6.  Mustard-pots,  xvii  and  xvni  centuries  .     114 

81.  (F.  &  s.)  i-6.  Pepper-pots,  xvn  and  xvni  centuries  .     114 

82.  (L.C.)  Cream-jug,  Salt-cellars,  Measure,  and  Spoons    .     116 

1.  Cream-jug,  Scotch. 

3.  5,  7.  Three  Salt-cellars,  English. 

9.  Quarter-gill  Measure.    Glasgow  mark. 

2,  4,  6,  8,  10.  Five  Spoons,  Dutch. 

83.  (R.M.)  Measure,  Salt-cellar,  Sugar-castor,  Cruet,  and 

Inkstand  .          .          .          .          .          .         .          .117 

1.  Measure. 

2.  Salt-cellar. 

3.  Sugar-castor. 

4.  Vinegar  Cruet. 

5.  Inkstand. 

84.  (F.C.Y.)  Five  Salt-cellars 117 

xvi 


LIST   OF   ILLUSTRATIONS 

TO   FACE 
PLATE  PAGE 

85.  (F.H.N.)  Sugar-sifters,  Cruets,  and  Spirit  Lamp  .         .118 

i,  2.  Sugar-sifters.     XVIII  century. 

3,  4,  5.  Set  of  Cruets,  Scotch.     XVIII  century. 

6.  Spirit  Lamp. 

86.  (w.)  Cream-jugs,  Herb  Cannisters,  and  Box       .         .118 

1.  Cream-jug,  Scotch. 

2.  Herb  Cannister,  Dutch.     XVII  century. 

3.  Box. 

4.  Herb  Cannister,  Dutch.     Marked  1766. 

5.  Cream-jug. 

87.  (F.  &  s.)  Urns        .......     121 

1.  Dutch  Urn. 

2.  French  Urn.     XVIII  century. 

3.  Dutch  Urn. 

88.  (B.J.)  Tea-pot,  Cream-jug,  and  Coffee-pot  .         .     122 

1.  Tea-pot.    Early  XIX  century. 

2.  Cream-jug.     Early  XIX  century. 

3.  Coffe-pot.     XVIII  century. 

89.  (B.)  Tea-pots  and  Coffee-pot    .         .         .         .         .124 

1.  Tea-pot,  Dutch.    XVIII  century. 

2.  Coffee-pot,  French.    XVIII  century. 

3.  Tea-pot,  Flemish.    XVIII  century. 

90.  (S.K.M.)  Coffee-pot,  Sugar-box,  and  Mustard-pot         .      125 

1.  Coffee-pot,   Louis  XIV.   style,  German.      First  half  XVIII 

century. 

2.  Sugar-box  and  Cover,  Dutch.    Dated  1751. 

3.  Mustard-pot. 

91.  (w.)  Egg  Cups 126 

i,  2,  3.  Egg  Cups.     No  marks. 

4.  5,  6,  7,  8,  9.  Set  of  Six  Egg  Cups,  Scotch.    No  marks. 

92.  (F.C.Y.)  Milk-jug,  Sugar-basin,  and  Tea-pot        .          .      126 

1.  Milk-jug,  English.     Early  XIX  century. 

2.  Sugar-basin,  English.     Early  XIX  century. 

3.  Tea-pot,  English.     Early  XIX  century. 

93.  (K.R.)  Tea-pot  and  Tankard 128 

1.  Tea-pot.     Early  XIX  century. 

2.  Tankard,  English.     XVIII  century. 

94.  (B.J.)  Tea-pots  and  Tobacco-box       .         .         .         .130 

1.  Tea-pot.    Early  XIX  century. 

2.  Tea-pot.     XVIII  century. 

3.  Tobacco-box.    XVIII  century. 

xvii 


OLD   PEWTER 


TO  FACE 
PLATE  PAGE 

95.  (K.R.)  Coffee-pot,  Inkstand,  and  Tobacco-box  .          .132 

1.  Coffee-pot,  Flemish.     XVIII  century 

2.  Inkstand,  French.     XVIII  century. 

3.  Tobacco-box,  French.     XVI II  century. 

96.  (w.)  Snuff-boxes  and  Shoe-buckles  .         .         .  133 

1.  3,  5.  Three  Snuff-boxes. 

2.  4.  A  Pair  of  Shoe-buckles,  the  forks  of  hand-cut  steel. 

97.  (F.C.Y.)  Pepper-pots,  Mustard-pot,  Snuff-box,  and  Egg 

Cup      .  .133 

i,  2.  Two  Pepper-pots. 

3.  Mustard-pot,  English.     Early  XVIII  century. 

4.  Snuff-box,  English.     Middle  XVIII  century. 

5.  Egg  Cup,  with  Bead  Pattern,  English.     XVIII  century. 

98.  (w.)  Ink-pot,  Inkstand,  and  Tobacco-box         .          .133 

1.  Ink-pot,  Scotch. 

2.  Inkstand,  Italian. 

3.  Tobacco-box. 

99.  (F.H.N.)  Flagons  and  Inkstand         .         .         .         .134 

1.  Flagon.    XVII  century. 

2.  Inkstand.     Late  XVIII  century. 

3.  Flagon.    XVII  century. 

100.  (w.)  Oriental  Dagger.     Handle  of  Pewter  and  Bone. 

Top  and  bottom  of  sheath  bound  with  pewter        .      136 

101.  (W.B.)  Food  Bottles  and  Salt-box    .         .         .         .138 

1,  3.  Food  Bottles. 

2.  Salt-box. 

102.  (A.G.B.)  Shaving-pot,     Sugar-sprinklers,     Jug,     and 

Sugar-basin  .          .          .          .          .          .          .140 

1.  Shaving-pot,  English. 

2.  4.  A  Pair  of  Sugar-sprinklers,  Belgian. 

3.  Cider-jug,  Norman. 
5.  Sugar-basin,  Belgian. 

103.  (B.J.)  Mugs,  Flagon,  and  Tankard  .          .          .          .142 

1.  MUST,  William  IV. 

2.  Flagon.     XVIII  century - 

3.  Tankard.     Late  XVIII  century. 

4.  Mug.     Early  XIX  century. 

104.  (H.M.)  Marks  on  Spoons I48 

105.  (H.M.)  Marks  on  Spoons 150 

1 06.  (H.M.)  Marks  on  Spoons I52 

xviii 


PREFACE 

N  compiling  the  present  volume 
I  have  not  attempted  to  disguise 
from  myself  the  fact  that,  how- 
ever keen  my  own  interest  in 
the  subject  may  be,  it  is  destined  to  appeal 
to  the  public  far  more  by  its  numerous  and 
carefully  chosen  illustrations  than  by  anything 
I  may  have  to  say.  I  make  no  pretence  of 
laying  before  the  reader  any  entirely  novel 
discoveries  concerning  pewter.  To  Mr. 
Starkie  Gardner's  paper  in  the  "  Journal  of 
the  Society  of  Arts,"  to  Mr.  Welch's  "  History 
of  the  Pewterers'  Company,"  to  Mr.  Massd's 
"  Pewter  Plate,"  especially  to  the  extremely 
useful  Bibliography  contained  in  it,  and  to 
Mr.  I ngleby  Wood's  "  Scottish  Pewter-ware 
and  Pewterers,"  I  freely  and  gratefully  ac- 
knowledge my  indebtedness  for  the  main 
portion  of  the  facts  herein  set  forth;  and  I 

xix 


OLD  PEWTER 

cannot  too  strongly  recommend  those  whom 
the  present  volume  may  haply  attract  for  the 
first  time  to  a  care  for  pewter  and  its  history 
to  seek  fuller  instruction  in  their  pages.  I 
have  merely  endeavoured  to  gather  here  such 
information  as  may  enable  the  inexperienced 
to  study  the  pictures  with  eyes  not  altogether 
unopened  to  their  meaning,  in  the  hope  that 
a  first  taste  of  this  particular  well  of  know- 
ledge, however  ill  served,  may  tempt  them  to 
fuller  and  deeper  draughts  elsewhere ;  and  if 
I  succeed  in  this  much  I  shall  not  deem  my- 
self to  have  altogether  failed. 

In  conclusion  there  only  remains  to  me  the 
pleasant  task  of  thanking  most  heartily  on  my 
own  behalf,  as  well,  I  trust,  as  on  behalf  of 
any  readers  I  may  have,  those  ladies  and 
gentlemen  who  have  so  generously  allowed 
their  treasures  to  be  photographed  and  re- 
produced, namely  Mrs.  Warren,  Mrs.  Ralston 
Mitchell,  Mrs.  Borough  Johnson,  Miss  Burrell, 
Mr.  Arthur  G.  Bell,  Mr.  William  Burrell,  Mr. 
Lewis  Clapperton,  Messrs.  Fenton  and  Sons, 
Mr.  Robert  Meldrum,  Mr.  H.  Murray,  Mr. 
xx 


PREFACE 

F.  H.  Newbery,  Mr.  Kennerley  Rumford, 
Mr.  Frank  C.  Yardley,  and  the  authorities  of 
the  British  Museum,  the  Victoria  and  Albert 
Museum,  and  the  Glasgow  Corporation 
Museum.  I  have  also  to  express  my  grateful 
thanks  to  Mr.  Yardley  for  cordial  and  invalu- 
able assistance  in  the  preparation  of  this 
work,  particularly  in  the  selection  and  dating 
of  objects  for  reproduction,  and  to  Mr.  Hugh 
P.  Bell  for  the  information  concerning  the 
chemical  analysis  of  pewter. 

MALCOLM  BELL. 


XXI 


ERRATA 

Page  157,  four  lines  from  the  bottom,  read  Plate  XLIX 
Page  159,  four  lines  from  the  bottom,  read  LVII.  i 
Plate  XXXII,  should  read  Altar  Candlesticks  and  German 
Guild  Cup  XVII  Century 


XX11 


OLD  PEWTER 

THE  FIRST  CHAPTER 

INTRODUCTORY 

OOKED  at  rightly  the  rapid 
growth  of  the  quite  modern  taste 
for  collecting  the  many  various 
objects  formerly  fashioned  of 
pewter  must  be  regarded  as 
among  the  healthiest  symptoms  of  the  later 
art  development  in  Europe.  It  is,  in  the 
first  place,  a  genuine  unaffected  taste,  not  a 
mere  fashionable  craze;  nor,  if  one  may  ven- 
ture to  formulate  a  prophecy  concerning 
matters  so  independent  of  any  rational 
foundation  as  the  freaks  of  popular  fancy  are 
apt  to  be,  is  it  ever  likely  to  become  so.  The 
key-note  of  the  fascination  of  pewter  at  its 
best  is  its  simplicity.  It  appeals  to  the  eye  by 
delicate  and  subtle  balances  of  line  and  pro- 
portion and  has  nothing  in  it  of  the  obvious 
and  sensational.  It  does  not,  like  gold  and 
silver  plate,  hint  in  any  way  at  large  intrinsic 

A  I 


OLD  PEWTER 

values,  nor  has  it  their  aggressive  sparkle  and 
glitter  clamouring  decoratively  for  attention. 
Neither  does  it  lend  itself  sympathetically 
to  that  elaborate  manipulation  by  dexterous 
craft  which  attracts  by  its  sheer  elaboration 
and  ingenuity  the  unreasoned  approbation  of 
the  vulgar;  its  high  lights  are  too  subdued,  its 
shadows  too  mellowed,  to  repay  by  brilliancy 
of  effect  for  excessive  labour,  while  the  soft- 
ness of  the  material  is  such  that  where  this 
has  been  bestowed  mistakenly  the  necessarily 
constant  polishing  soon  wears  away  the  sharp- 
ness. It  depends  upon  its  construction,  not 
on  its  added  ornamentation ;  and,  as  too  much 
modern  architecture  makes  evident,  a  refined 
sense  of  proportion,  an  appreciation  of  the 
importance  of  restful  spaces,  are,  for  some 
reason,  yearly  becoming  more  rare  among  us, 
though  for  them  no  superabundance  of  lavish 
decoration,  however  good  in  itself,  can  com- 
pensate. It  has  not,  again,  the  fascinating 
fragility  of  china,  glass,  and  earthenware, 
which  makes  the  joy  of  possessing  them,  to 
quote  Mr.  Gilbert,  "  a  pleasure  that's  almost 
pain."  Lastly,  in  the  history  of  pewter  there 
are  no  great  names  or  conspicuous  schools  for 
the  speculative  dealer  to  boom  into  fictitious 

2 


INTRODUCTORY 

and  precarious  popularity  which  serves  only 
in  the  end  to  unjustifiably  inflate  prices.  The 
works  of  Briot  and  Enderlein  and  some  few 
other  makers  may  perhaps  be  brought  forward 
in  exception  to  this  statement,  but  for  reasons 
more  fully  entered  into  later  these  will  be  ac- 
corded but  brief  notice  in  these  pages.  An 
understanding  of  the  finest  work  done  in 
pewter  implies  an  approbation  of  good  work 
honestly  applied,  of  sound  design  based  on 
actual  needs,  of  unassuming  worth,  not  pre- 
tentious value,  and  these  are  not  qualities 
commonly  held  in  estimation  by  the  bearers 
of  the  longest  purses  in  the  auction-rooms. 

Fifty  years  ago,  or  even  less,  had  any  one 
been  at  the  painsof  giving  the  matter  a  thought, 
the  utter  disappearance  of  all  old  domestic 
pewter,  except  from  a  few  houses  where  the 
former  plenishings  were  preserved  more  out  of 
habit  or  curiosity  than  from  any  real  knowledge 
of  their  beauty,  might  have  seemed  to  be  within 
measurable  distance,  while  far  more  recently 
fine  church  pewter  has  been  melted  down  or 
thoughtlessly  allowed  to  vanish  in  favour  of 
modern  white  metal  or  plated  abominations. 
Who  first  revived  an  interest  in  these  fast 
diminishing  relics  of  the  past  may  already  be 

3 


OLD  PEWTER 

past  discovery,  but  of  late  the  number  of  his 
followers  has  increased  annually,  and  pewter 
is  now  firmly  established  as  one  of  the  subjects 
fully  deserving  a  literature  to  itself. 

Much  has  already  been  elicited  by  patient 
research  and  not  a  little  written,  and  the  time 
is  drawing  near,  if  indeed  it  has  not  arrived, 
when  any  new  volume  on  the  matter  may  be 
asked  to  show  justification  for  its  existence. 

In  the  present  instance  the  number  of  re- 
productions of  carefully  selected  and  well- 
authenticated  typical  examples  must  be  placed 
in  the  first  line  of  defence.  To  the  beginner 
in  the  study  of  any  branch  of  art  production  the 
opportunity  of  examining  specimens  of  gene- 
rally recognised  merit  is  all-essential,  yet  in  few 
is  this  so  difficult  of  attainment  as  in  the  case 
of  pewter.  Indeed,  unless  the  novice  is  fortu- 
nate enough  to  be  acquainted  with  the  owner 
of  a  good  collection,  it  may  be  said  to  be  practi- 
cally impossible.  In  our  museums  attention 
has  been  almost  exclusively  confined  to  articles 
showing  more  or  less  laborious  ornamentation, 
which  are  beside  our  purpose,  while  the  plainer 
articles  of  every-day  use  have  been  well-nigh 
ignored.  It  is  hoped  that  our  illustrations 
will  to  some  extent  remedy  this  defect.  That 
4 


PLATE  IV 


H.  22! 

THE  GLOUCESTER  CANDLESTICK. 
XII  century. 


INTRODUCTORY 

no  object  shall  be  brought  under  the  collector's 
notice  to  which  a  fairly  close  parallel  for  com- 
parison cannot  be  found  in  these  pages  is  too 
much  to  expect,  but  every  effort  has  been 
made  to  render  the  variety  of  forms  and  periods 
as  catholic  and  extensive  as  possible.  With 
regard  to  the  letterpress,  an  attempt,  however 
inadequate,  has  been  made  to  compress  into  a 
volume  of  convenient  size  all  the  more  im- 
portant of  the  facts  that  are  known,  and  these 
have  been  at  times  supplemented  by  earnestly 
weighed,  though  maybe  erroneous,  conjectures. 
We  begin,  in  our  second  chapter,  with  a 
consideration  of  what  pewter  is,  indicating  the 
chief  divergences  in  the  alloys,  and  pointing 
out  various  methods  of  estimating  the  nature 
of  the  one  present  in  any  given  object,  either 
roughly  for  general  use,  or  more  precisely 
by  chemical  analysis  for  the  scientifically 
minded  inquirer.  In  the  next  chapter  the 
different  ways  in  which  the  metal  was  wrought 
into  shape,  and  the  conditions  under  which 
this  was  usually  done,  are  briefly  described 
without  dwelling  with  unnecessary  prolixity 
on  technical  details.  In  the  five  succeeding 
chapters  some  of  the  most  striking  features  in 
the  history  of  pewter  and  pewterers,  as  far  as 

5 


OLD  PEWTER 

known,  are  sketched,  while  the  ninth  chapter 
is  devoted  to  such  suggestions  as  it  is  practi- 
cable to  offer — far  too  few  and  insufficient, 
unfortunately — which  may  serve  to  smooth  to 
some  extent  the  path  of  the  beginner  in  col- 
lecting and  effectively  displaying  his  pewter. 
The  tenth  and  final  chapter  is  devoted  to  a 
brief  indication  of  the  leading  features  in  the 
illustrations.  If  this  work,  despite  its  too 
probable  errors  and  omissions,  should  succeed 
in  arousing  in  but  a  few  hitherto  indifferent 
minds  an  intelligent  interest  in  the  rare 
aesthetic  qualities  of  old  domestic  or  ecclesi- 
astical pewter-ware,  the  author  will  be  amply 
rewarded. 


PLATE  V 


H.  7" 

BENITIER,  Probably  Flemish. 
No  marks. 


THE  SECOND  CHAPTER 

WHAT  PEWTER  IS 


EWTER— the  word  which,  with 
the  pleasing  independence  of  rigid 
orthographic  rules  that  charac- 
terises all  mediaeval  English 
documents,  whether  lay,  legal,  or 
clerical,  appears  under  the  widely  varying  but 
always  recognisable  guises  of  pewtre,  peautre, 
pewtir,  peutre,  peuther,  and  even  pewder  and 
pewdre — is  confidently  asserted  by  Doctor 
Johnson  to  owe  its  derivation  to  the  Dutch 
peauter.  Professor  Skeat,  on  the  other  hand, 
the  weight  of  whose  opinion  cannot  well  be 
over-estimated,  suggests  in  his  Etymological 
Dictionary  as,  at  any  rate,  a  strong  prob- 
ability, that  the  fact  is  the  exact  contrary,  and 
that  the  Dutch  word,  together  with  the  old 
French  peutre,  peautre  or  piautre,  the  Italian 
peltro  and  the  Spanish  peltre,  are  all  borrowed 
from  the  English  word,  which  is  itself  an 
adaptation  of  spelter  ;  nor  is  the  question 
without  a  certain  importance,  since  in  the 
case  of  a  purely  artificial  compound  of  this 
description  a  not  unreasonable  presumption 

7 


OLD  PEWTER 

is  that  the  place  of  its  origin,  at  all  events 
as  far  as  Western  civilisation  is  concerned, 
coincides  with  the  land  which  first  gave  it  a 
name. 

The  word,  however,  as  applied  to  the 
material,  can  only  be  regarded  as  generic,  not 
specific.  It  signifies  always  an  alloy  of  two 
or  more  metals  in  which  tin  forms  the  pre- 
ponderating element,  but  the  added  com- 
ponents are  so  varied  in  their  nature  and  so 
diverse  in  their  proportions,  while  the  result- 
ing compounds  are  so  different  in  qualities 
and  appearance,  that  Mr.  Starkie  Gardner, 
attempting  in  an  admirable  paper  read  before 
the  Society  of  Arts  on  May  8,  1894,  to 
decide  to  what  exactly  the  name  pewter 
should  be  applied,  could  only  say  "  the  pro- 
portions are  so  variable  that  it  is  scarcely 
possible  to  exclude  any  in  which  tin  forms  the 
bulk,  where  the  result  is  a  darkish  silvery, 
soft  metal,  fusible  at  a  low  temperature, 
inexpensive,  and  eminently  adapted  to  a 
variety  of  household  and  artistic  purposes  "  ; 
a  definition  which  cannot  be  said  to  err  on  the 
side  of  narrowness.  Yet  any  endeavour  to 
further  restrict  it  leads  straightway  to  con- 
tradiction. A  passage  in  Mr.  Masses  instruc- 
8 


WHAT  PEWTER  IS 

tive  volume  on  Pewter  Plate  (p.  20)  implies 
that  from  his  point  of  view  "  pewter  of  good 
quality  .  .  .  should  contain  no  lead,"  while 
according  to  Mr.  Starkie  Gardner,  strictly 
speaking,  lead  alone  should  be  added,  and  his 
recipe  has  at  least  the  support  of  antiquity, 
for  of  this  nature  was  all  the  old  Roman 
pewter  analysed  by  Mr.  Gowland  and  others. 
In  an  important  appendix  to  a  paper,  printed 
in  Archczologia  in  1898,  describing  the 
remains  of  a  Roman  villa  and  a  number  of 
contemporary  pewter  vessels  unearthed  at 
Appleshaw,  to  which  further  reference  will  be 
made  later  on,  he  discussed  and  tabulated  all 
the  analyses  of  Roman  pewter  that  were  avail- 
able. The  results  show  on  the  surface  an 
extraordinary  divergency,  the  fifteen  samples 
tested  ranging  with  considerable  regularity 
from  a  mixture  of  99.18  of  tin  with  .14  of 
lead,  or  practically  pure  tin,  and  a  probably 
accidental  trace  of  iron,  down  to  a  blend  of 
45.74  of  tin  and  53.34  of  lead  with  traces  of 
iron  and  copper,  but  Mr.  Gowland  in  his 
luminous  comment  makes  it  clear  "that  this 
irregularity  is  more  apparent  than  real."  "  If 
we  now  examine  the  analyses  in  the  table,"  he 
argues,  "  we  will  find  that  the  greater  number 

B  9 


OLD  PEWTER 

of  the  specimens  may  be  placed  in  one  or 
other  of  the  two  following  groups,  A  and  B, 
each  characterised  by  a  special  percentage  of 
tin — Group  A  having  an  average  composition 
of:  tin  71.5,  lead  27.8;  Group  B  with  an 
average  composition  of:  tin  78.2,  lead  21.7 
and  he  further  points  out  that,  as  far  as  the 
small  number  of  analyses  permitted  us  to 
judge,  "  the  first  was  most  generally  employed 
by  the  Romans  during  their  occupation  of 
Britain,"  a  deduction  which  remarkably  agrees 
with  the  analysis  of  some  dating  from  the 
fourth  century  found  at  Aquae  Neriae,  which, 
according  to  Bapst,  contains  tin  and  lead  in 
the  proportions  of  about  7  to  3.  In  conclusion 
Mr.  Gowland  remarks  :  "  It  is  worthy  of  note 
that  the  most  tenacious  alloy  of  tin  and  lead 
closely  approaches  these  Roman  pewters  in 
composition,  a  fact  which  bears  important 
testimony  to  the  knowledge  of  the  properties 
of  metals  possessed  by  the  Romans.  In  the 
preparation  of  these  two  pewters  the  Romans 
seem  to  have  followed  the  practice,  which  still 
survives  in  some  foundries,  of  taking  i  pound 
(libra)  of  the  chief  metal  and  allotting  the 
quantity  of  the  other  metal  to  be  mixed  with 
it  in  the  sub-divisions  of  a  pound  (unciae). 
10 


PLATE  VII 


PLATE  VIII 


For  Descriptions,  sec  back  of  rial e. 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  VII. 

1.  Measure,  German,  4^"  high. 

2.  Pepper  box,  5^"  high. 

3    Two  buttons,  2^"  diam. 
4.  Mustard  pot,  4"  high. 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  VI 1 1. 

1,  3,  4.  Pepper  boxes,  4^"  to  5"  high. 

2.  Sugar  castor,  marked  wiih  Cupid  holding  trumpet  to  mouth 

with  left  hand,  branch  in  right,  HK,  5!"  high. 
5,  6,  7.  Mustard  pots,  3!"  to  5^"  hi^h. 
8,  9,  10,  n,  12.  Salt  cellars,  10",  m irked  with  crowned  X 
Nos.  3,  7,  9  are  a  set. 


WHAT  PEWTER  IS 

Thus  the  pewter  of  Group  A  was  evidently 
made  by  melting  together  i  libra  of  tin  with 
4^  unciae  of  lead,  which,  with  due  allow- 
ance for  the  oxidation  of  part  of  the  tin, 
would  yield  an  alloy  of  that  composition. 
Pewter  B  was  similarly  the  result  of  melting 

1  libra  of  tin  with  3  unciae  of  lead  "  ;  and  he 
finally  calls  attention  to  the  fact  that  of  two 
cakes  of  unwrought  metal  of  Roman   make 
found  in  the  Thames  at  Battersea,  one  agrees 
in  composition  with  Group  A,  the  other  with 
Group  B,  which   circumstance  can  only  be 
regarded   as   convincing   evidence    that    his 
reasoning  is  sound. 

In  later  days,  however,  other  metals  were 
very  generally  employed  in  place  of,  or  in 
addition  to,  the  lead,  as  is  shown  by  other 
analyses  also  carried  out  by  Mr.  Gowland. 
Thus  the  English  "  fine  pewter  "  contains  1 12 
parts  of  tin  to  26  of  copper  and  no  lead  at  all ; 
"  better  pewter"  of  the  first  quality  84  of  tin, 
7  of  antimony,  and  4  of  copper ;  of  third 
quality  56  of  tin,  8  of  lead,  6  of  copper,  and 

2  of  zinc  ;    "  plate   pewter "  of  the   second 
quality  contains  only  tin  and  antimony  in  the 
proportion  of  1 12  to  6  or  7  ;  while  that  of  the 
third  quality  consisted  of  90  parts  of  tin,  7  of 

ii 


OLD  PEWTER 

antimony,  2  of  copper,  and  2  of  bismuth.   The 
use  of  brass  or  copper  was  actually  enf 
for  certain  purposes,  by  the  ordinances  drawn 
up  for    the   guidance  of  pewterers  in  1348, 
wherein  the  "  fine  pewter  "  referred  to 
is  described  as  tin  mixed  with  copper  (or  bras  ) 
"  as  much  as  of  its  own  nature  it  will   tal 
i.e.,  about    I   part  to  4,  and  at  the  same  time 
the  articles  to  be  made  of  this  alloy,  chiefly 
those  that  were  made  square  or  ribbed,  are 
definitely  pres<  jibed,  while,  writing  more  than 
two    hundred    years    latei  .1 

"Description  of  I '.upland  in  Shakespeare's 
Youth,"  when  eulogising  English  pewterers 
and  pewter,  says  :  "J  have  also  been  informed 
that  it  consisteth  of  a  coi  on  which 

hath  ;»>  Ibs.  of  kettle  brass  to  iooolbs.  of  tin, 
whereunto  they  add  3  or  4  Ibs.  of  tin-gloss  (in 
modern  pailanee,  bismuth);  but  as  too  much 
of  this  doth  make  the  stuff  brickie,  so  the 

more  tin  brass  be  Hie  better  is  the  pewter,  and 
more  pioiitable  unto  him  that  doth  buy  and 

purchase  the  same." 

'1  hrproporlionof  lead  1<  ::.,|j  ,,  <|  bythesame 
Ordiu-iiiMs  fa  lh<-  blending  of  what  we  may 
with  Mr.  ( *aidii(  i's  suj)])ojl  bealhjwed  j>erh;ij>s 
to  entitle  true  pewter,  which  was  to  be  used 
12 


- 


:-  a 

:i 

:T- 

:  : 


=   '-c  : 

-  - 

-  ~    -  X 


"     ~ 


-- 

-  s   -        - 
- 


Z   HH 


J 

.Q 

ffi  W 
w   * 


HX 


HX 


WHAT  PEWTER^  IS 

for  making  "pots  rounded,  cruets  rounded, 
candlesticks  and  other  rounded  vessels,"  was 
22  pounds  according  to  one  account,  26  accord- 
ing to  another,  for  each  hundredweight  of  tin, 
though  there  would  seem  to  have  been  some 
laxity  in  carrying  out  this  regulation,  since 
only  two  years  later,  in  1350,  certain  proceed- 
ings at  the  Guildhall  indicate  that  the  then 
customary  blend  was  only  16  pounds  of 
lead  to  112  of  tin.  Small  variations,  in  fact, 
within  reasonable  limits  were  evidently  not 
considered  of  great  importance,  and  every 
centre  of  the  industry  was  a  law  unto  itself. 
The  Montpelier  pewterers,  for  example,  in 
1437  used  a  mixture  of  96  parts  of  tin  to 
4  of  lead  for  dishes  and  porringers,  90  of  tin 
and  10  of  lead  for  ewers  and  salts,  while  the 
Limoges  pewterers  mixed  100  parts  of  tin  with 
only  4  of  lead  ;  the  Nuremberg  pewterers  were 
required  in  1576  to  use  10  pounds  of  tin  to 
every  i  of  lead  ;  finally,  in  France,  during  the 
eighteenth  century  100  parts  of  tin  were  mixed 
either  with  5  parts  of  copper,  or  with  3  of 
copper  and  i  of  bismuth,  or  with  15  parts  of 
lead,  though  at  the  present  day  a  percentage  of 
16.5  of  lead  with  a  narrow  margin  for  error  is 
alone  authorised  as  safe  for  the  storage  of  wine. 

13 


OLD  PEWTER 

The  object  of  these  regulations  in  every 
case  was,  in  the  first  place,  to  •  ensure  suffi- 
cient wearing  power  in  the  article,  and  in  the 
second,  to  protect  the  customer  from  adultera- 
tion and  consequent  deterioration  of  the  alloy 
by  an  excessive  amount  of  lead,  the  cheaper 
component,  and  so  long  as  the  quantity  of  this 
used  was  kept  within  limits,  the  addition 
of  other  materials  would  seem  to  have  been 
left  to  a  considerable  extent  to  the  fancy  of 
the  individual  worker,  who  varied  his  compo- 
sition according  to  the  purpose  to  which  it 
was  to  be  applied,  a  liberty  which  was  probably 
further  secured  to  him  in  early  days  by  the 
imperfection  of  chemical  science,  and  its  in- 
ability to  segregate  minute  percentages  from 
samples  of  small  weight. 

Even  in  these  days  of  prodigiously  im- 
proved methods  the  true  ingredients  of  any 
given  example  cannot  be  regarded  as  ascer- 
tained with  any  very  near  approach  to  accuracy 
byan  analysis  of  single  scrapings  or  other  com- 
paratively infinitesimally  small  proportions 
of  the  whole  bulk.  In  the  course  of  the  dis- 
cussion which  followed  the  reading  of  the 
paper  by  Mr.  Starkie  Gardner,  quoted  above, 
Mr.  Gowland  stated  that  in  analysing  two 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XI. 

1.  Toddy  Ladles,  marked  John  Yates.    VR  crowned. 

2.  Small  fork,  no  mark.     Length  7". 


3.  Soup  ladle,  marked    |jO  I     |  HN       |  YA 


TES 


4.  Soup  ladle,  marked     |  ASK  |     |  BER  |     |~RY 

W.  crown.  R. 

BEST  METAL 

WARRANTED 

FOR  USE 

5.  Three  table  spoons,  marked 

.    |7o~l    |  HN      |  YAJ     I  TES  j    VR  crowned. 

6.  Tea  spoon,  marked  Ai 

Crowned  head 
PURE 


WHAT  PEWTER  IS 

distinct  fragments  of  the  lid  of  a  tea-pot  of 
eighteenth-century  Japanese  pewter  he  found 
that  one  portion  contained  80.48  per  cent,  of 
tin  to  20.02  of  lead,  while  the  other  showed 
77.64  per  cent,  of  tin  and  22.5  of  lead,  though 
there  was  no  reason  to  suppose  that  the 
material  was  not  originally,  when  in  a  state 
of  fusion,  perfectly  homogeneous.  The  cause 
of  this  curious  inconsistency  in  the  results 
was  fully  explained  at  the  same  time.  "  Pro- 
fessor Brown-Austen,"  to  quote  the  Abstract 
of  the  discussion  given  in  the  Journal  of  the 
Society,  "  had  shown  by  means  of  the  thermo- 
electric pyrometer,  that  when  an  alloy  was  in 
the  act  of  cooling,  several  definite  alloys,  in 
which  the  molecules  of  the  metals  were  dif- 
ferently grouped  from  those  of  the  mass,  fell 
out  at  definite  temperatures,  so  that  the  solidi- 
fied metal  did  not  consist  really  of  one  alloy, 
but  was  a  mixture  of  several,  more  or  less 
regularly  diffused  throughout  the  mass,"  the 
margin  of  possible  error  from  this  cause,  ac- 
cording to  Mr.  Gowland,  averaging  about 
2  per  cent,  for  small  vessels  which  would  cool 
quickly  and  fairly  uniformly,  and  as  much  as 
4  per  cent,  in  larger  vessels. 

It  follows  from  this  that  since  no  collector 

15 


OLD  PEWTER 

is  likely  to  sacrifice  an  entire  object  for  the 
purpose  of  ascertaining  its  composition  with 
scientific  precision,  the  quality  of  any  given 
specimen  of  pewter  can  only  be  arrived  at  in  a 
rough  and  ready  fashion.  Excess  of  lead  is  re- 
vealed by  the  weight  and  the  dark  colour  of  the 
surface,  and  the  proportion  may  be  discovered 
approximately,  according  to  Mr.  Massd,  by 
the  degree  to  which  it  is  possible  to  make  a 
mark  on  paper  with  the  metal.  Pure  lead 
leaves  a  dark  mark  like  a  pencil,  pure  tin 
makes  none,  nor  does  an  alloy  which  contains 
more  than  three  parts  of  tin  to  one  of  lead. 
This  combination  just  leaves  a  faint  trace,  and 
the  darker  the  mark  appears  the  more  lead 
and  the  less  tin  has  been  used  in  the  alloy. 
Pure  tin  again,  owing  to  its  peculiar  crystalline 
structure,  gives  out  a  characteristic  sound 
when  scratched  with  a  knife,  and  an  equally 
remarkable  and  unmistakable  crackle  or 
"  cri "  when  bent,  while  lead  gives  no  such 
response.  The  nature  of  the  sound  produced 
may  therefore  serve  as  a  vague  criterion  of 
the  goodness  of  the  pewter,  but  this  test, 
Mr.  Massd  observes,  is  only  trustworthy  to  a 
very  limited  extent,  since  the  addition  of  a 
very  small  amount  of  zinc  to  the  purest  tin 
16 


UJ 

CLX 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XII. 

THE  MAKERS'  MARKS,  WHERE  LEGIBLE,  ARE 
SHOWN  ON  PLATE  CIV.,  FACING  PAGE  148. 

1.  Laton  spoon,  5"  long,  found  in  the  Thames  at  Westminster  ; 

an  engraved  line  runs  along  lower  surface  of  stem.     XIV 
century. 

2.  Bronze  spoon,  6J"  long,  found  in  the  City  of  London.     XIV 

century.     Probably  French. 

3.  Laton  spoon,  6"  long,  French.    XIV  century. 

4.  Laton  spoon,  6^"  long,  found  in  the  Thames  in  London,  stem 

of  diamond  section.     Early  XV  century. 

5.  Laton  spoon,  7"  long,  found  in  London.     Early  XV  century. 

Probably  French. 

6.  Bronze  spoon,  5^"  long,  found  in  London.     XV  century. 

7.  Bronze  spoon,  6"  long,  stem  of  diamond  section,  found  in 

London.     XV  century. 

8.  Bronze  spoon,  6|"  long,    found  in  London.      XV   century. 

Mark  same  as  No.  2. 


WHAT  PEWTER  IS 

at  once  destroys  the  tendency  to  emit  the 
"  crackle."  Lastly,  a  rule  of  thumb  test  em- 
ployed in  former  days  by  French  pewterers 
consisted  in  touching  the  metal  with  a  hot 
iron  and  judging  the  quality  by  the  whiteness 
of  the  scar  which  resulted  on  good  pewter,  or 
the  increasing  depth  of  the  brown  observable 
as  the  stuff  deteriorated. 

The  question,  after  all,  is  for  the  collector 
mainly  an  academic  one  of  little  practical 
importance.  When  the  purchaser  was  buy- 
ing for  domestic  use,  and  in  especial  when  he 
proposed  to  store  in  it  wine,  vinegar,  or  any 
other  liquid  of  which  the  natural  acidity 
might  act  on  any  excess  of  lead,  giving  rise 
to  actively  poisonous  chemical  compounds,  it 
was  a  matter  of  great  moment,  but  such  a 
preponderance  in  no  way  affects  the  artistic 
or  historical  value  of  the  specimen,  and  as 
none  of  the  metals  that  may  be  present  has, 
speaking  broadly,  much  intrinsic  worth,  its 
money  value  is  in  no  way  dependent  upon 
the  character  of  the  material. 

In  case,  however,  any  owner  of  pewter 
should  be  desirous  of  ascertaining  with  some 
degree  of  certainty  the  composition  of  any 
article  in  his  possession,  a  fairly  simple  method 

c  17 


OLD  PEWTER 

of  procedure  is  here  appended.  Scrape  from 
some  inconspicuous  portion  of  the  vessel 
sufficient  material,  taking  due  care  that  the 
instrument  used  is  scrupulously  clean,  as, 
indeed,  must  be  the  case  with  all  the  apparatus 
employed,  and  endeavouring,  if  convenient, 
for  the  reason  given  above,  to  secure  a  number 
of  small  fragments  from  different  parts  of  the 
surface  rather  than  a  larger  scraping  from 
one  only.  If  the  subsequent  operations  are 
conducted  with  great  care  and  a  delicate 
balance  is  used,  from  three  to  four  grams 
weight,  or  about  an  eighth  of  an  ounce, 
should  be  enough,  though  it  is  needless  to 
say  that  every  increase  in  the  portion  experi- 
mented upon  will  diminish  the  liability  to 
error  by  facilitating  the  task  of  weighing  the 
constituents  later  on.  Weigh  the  scraps  of 
metals  so  obtained  carefully  in  an  accurate 
balance,  place  them  in  a  glass  flask  or  other 
convenient  vessel,  and  pour  upon  them  a 
mixture  of  strong  nitric  acid  diluted  with 
half  its  volume  of  water.  The  action  of  the 
acid  on  the  metal  will  be  at  once  observable 
by  a  brisk  effervescence,  and  as  soon  as  this 
appears  to  be  slackening  gently  warm  the 
flask.  When  it  is  finished  a  fine  white 
18 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XIII. 

THE  MAKERS'  MARKS,  WHERE  LEGIBLE,  ARK 
SHOWN   ON   PLATE   CIV.,    FACING    PAGE    148. 

1.  Pewter  slip-top  spoon,  6i"  long,  found  at  Bei  mondsey,  stem  of 

hexagonal  section.     XVI  century. 

2.  Pewter  slip-top  spoon,  6i"   long,   found   in    London.      XVI 

century. 

3.  Pewter  slip-top  spoon,  6J"  long.     XVI  century. 

4.  Pewter  slip-top  spoon,  6A"  long.     XVII  century. 

5.  Pewter  slip-top  spoon,    6J"   long,   found  in   London.      XVI 

century. 

6.  Pewter  slip-top  spoon,  6"  long,  found  in  York  Road,  West- 

minster.    XVI  century. 

7.  Pewter  slip-top  spoon,  6\"  long,    found   in    London.     XVI 

century. 

8.  Pewter  slip-top  spoon,  6^"  long,  found  in  London.      XVII 

century. 


WHAT  PEWTER  IS 

powder  or  precipitate  will  be  visible  sus- 
pended in  the  otherwise  clear  liquid.  That 
is  all  the  tin  present,  now  in  the  form  of 
oxide.  Next  dilute  the  liquid  in  the  flask 
with  about  three  times  its  volume  of  water 
and  heat  it,  keeping  it  meanwhile  in  a  con- 
stant state  of  agitation.  Then  leave  the  pre- 
cipitate to  settle,  and  when  that  is  done  pour 
off  the  liquid  carefully  into  another  vessel, 
taking  care  that  all  the  oxide  remains  behind. 
Preserve  the  liquid  poured  off,  as  that  still 
holds  in  solution  the  remaining  constituents 
of  the  alloy.  To  the  small  residuum  con- 
taining the  precipitate  add  more  nitric  acid, 
diluted  with  about  six  times  its  volume  of 
water,  and  heat  again.  Pass  it  then  through 
a  chemist's  filter-paper  properly  folded  in  a 
glass  funnel,  and  when  the  dilute  acid  has  all 
run  through  pour  water  on  the  powder  on  the 
filter-paper  so  as  to  wash  it  thoroughly.  If 
the  water  on  coming  from  the  filter  is  at  all 
blue  in  colour  copper  is  still  being  dissolved, 
and  the  treatment  with  the  second  orweaker 
dilution  of  nitric  acid  must  be  repeated  until 
on  washing  as  before  no  shade  of  blue  can 
be  detected.  All  the  liquid  that  has  passed 
through  the  filter  should  be  added  to  that 

19 


OLD  PEWTER 

which  was  first  poured  off  and  reserved  for 
further  treatment.  The  filter-paper,  with 
the  powder  adhering  to  it,  must  be  dried 
thoroughly  in  a  small  porcelain  crucible,  and 
when  that  is  done  the  paper  lighted  and  left 
to  burn  to  ashes,  which  will  be  so  infini- 
tesimal in  weight  as  not  to  affect  materially 
the  result.  Heat  the  remnant  over  a  Bunsen 
burner,  or  spirit  lamp,  and  keep  it  for  a  time 
at  a  dull  red  glow  to  drive  off  any  water 
remaining  free  or  in  combination,  and  when 
it  has  cooled  weigh  carefully ;  75  parts  of 
oxide  of  tin  contain  59  parts  of  pure  metallic 
tin,  so  that  if  the  result  be  multiplied  by  59 
and  divided  by  75  you  will  have  the  exact 
weight  of  pure  tin  in  the  alloy  and  can  easily 
calculate  the  percentage. 

The  next  step  is  to  release  the  lead.  Into 
the  liquid  which  has  been  previously  set  aside 
pour  a  small  quantity  of  strong  sulphuric 
acid  and  evaporate  it  down  in  a  glass  or 
porcelain  basin,  taking  care  that  it  does  not 
boil  with  sufficient  violence  to  spurt  over  the 
edge,  until  it  begins  to  give  off  thick  white 
fumes.  This  operation  drives  off  the  nitric 
acid  which  is  no  longer  useful.  Now  dilute 
the  liquid  with  water  and  again  a  cloud  of 
20 


•I 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XIV. 

THE  MAKERS'  MARKS,  WHERE  LEGIBLE,  ARE 
SHOWN  ON  PLATE  CIV.,  FACING  PAGE  148. 

1.  Pewter  spoon,  6A"  long,  found  in  London.     XVI  century. 

2.  Pewter  spoon,  6^"  long.     XVII  century. 

3.  Laton  spoon,  6J"  long,  found  in  London,  stem  of  hexagonal 

section.     XVII  century. 

4.  Laton  spoon,  with  Apostle  top,  6^"  long.    XVII  century. 

5.  Pewter  spoon,  seal  headed,  hevagonal,  6£"  long,    found   at 

Bermondsey.    XVI  century. 

6.  Pewter  sacramental  spoon,  with  Maidenhead  top,  7^"  long, 

found  in  London.     XVI  century  . 

7.  Pewter  spoon,  6£"  long,  found  in  London.     XVII  century. 

8.  Pewter  spoon,  with  Apostle  top,  6J"  long,  found  in  Londcn. 

XVI  century. 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XV. 

THE  MAKERS'  MARKS,  WHERE  LEGIBLE,  ARE 
SHOWN   ON   PLATE   CV.,    FACING   PAGE    150. 

1.  Pewter  slip-top  spoon,  6"  long,  found  in  York  Road,  West- 

minster. XVI  century.  Same  mark  as  Plate  XIII.,  Fig.  6. 

2.  Pewter   slip-top   spoon,  6"   long,    found   in    London.      XVI 

century.     Same  mark  as  Plate  XIII.,  Fig.  6. 

3.  The  same  as  Fig.  2. 

4.  Pewter  slip-top  spoon,  6i"  long,   found  in    London.     XVI 

century. 

5.  Pewter  slip-top  spoon,  6|"  long.     XVI  century. 

6.  Pewter  slip-top   spoon,    6J"  long,  found   in  London.     XVI 

century.     Same  mark  as  Fig.  2. 

7.  Pewter  slip-top  spoon,  6i"  long.     XVI  century. 

8.  Pewter  slip-top  spoon,  6f"  long,  found  in  Bermondsey.     XVI 

century. 


•  J 

o 

g 


WHAT  PEWTER  IS 

fine  white  powder  will  appear.  That  is  the 
lead  in  the  form  of  sulphate  of  lead.  Filter 
and  wash  as  before,  only  adding  a  few  drops 
of  sulphuric  acid  to  the  water  used.  Once 
more  set  aside  the  liquid  that  comes  through 
the  filter,  which  will  still  hold  in  solution  any 
constituents  of  the  alloy  other  than  tin  or 
lead.  Wash  the  powder  in  the  filter  with 
alcohol  or  methylated  spirit,  but  throw  this 
away  after  it  is  done  with  as  it  can  serve  no 
further  purpose ;  then  dry,  burn  off,  heat 
awhile,  and  weigh  as  in  the  case  of  the  tin 
oxide  ;  151  parts  of  sulphate  of  lead  contain 
103  parts  of  lead,  and  a  simple  calculation,  as 
in  the  former  case,  will  give  the  true  weight 
of  the  lead  and  the  percentage. 

If  the  alloy  contains  any  appreciable 
quantity  of  copper  the  liquid  which  has  been 
reserved  from  the  first  washing  will  have 
a  blue  colour.  To  secure  this  copper  is 
easy,  but,  within-doors  at  any  rate,  some- 
what unpleasant.  The  first  step  consists 
in  passing  .through  the  liquid  sulphuretted 
hydrogen  gas,  easily  made  by  pouring  sul- 
phuric acid  on  sulphite  of  iron  in  a  suitable 
vessel,  but  disagreeably  distinguished  by 
possessing  a  powerful  and  highly  offensive 

21 


OLD  PEWTER 

odour  of  rotten  eggs.  A  fine  powder,  black 
in  colour  this  time,  will  soon  appear.  That 
is  the  copper  in  the  form  of  copper  sulphide. 
Repeat  with  this  the  now  familiar  operations 
of  filtering  and  washing.  It  is  then  redis- 
solved  in  dilute  nitric  acid  and  a  solution  of 
caustic  potash  (potassium  hydrate)  is  added. 
The  precipitate  reappears  in  the  form  of 
copper  oxide,  which  may  then  be  filtered 
out,  dried,  burnt  off  and  so  on  previous  to 
weighing  ;  159  parts  of  copper  oxide  con- 
tain 127  parts  of  copper. 

The  final  test  for  zinc  is  easy.  A  solution 
of  sodium  carbonate  added  to  the  remaining 
liquid  will  at  once  precipitate  it  as  zinc  oxide. 
This  should  be  white,  but  will  possibly  be 
more  or  less  tinged  with  reddish  brown,  a 
proof  that  a  slight  accidental  impurity  is 
present  in  the  form  of  iron,  which,  however, 
may  be  ignored.  The  washing,  filtering,  &c., 
are  next  carried  out,  but  the  final  heating  to 
drive  off  the  water  must  be  done  with  great 
care,  as  zinc  evaporates  at  a  comparatively 
low  temperature,  and  floating  off  as  a  gas 
will  be  lost;  81  parts  of  zinc  oxide  contain 
65  parts  of  zinc. 

In  some  kinds  of  pewter,  as  has  been 
22 


CLX 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XVI. 

THE  MAKERS'  MARKS,  WHERE  LEGIBLE,  ARE 
SHOWN  ON  PLATE  CV  ,  FACING  PAGE  150. 

1.  Pewter  slip-top  spoon,  6£"  long,  found  at  Bermondsey.     XVI 

century. 

2.  Pewter  slip-top  spoon,  7^"  long,   found   in   London.     XVI 

century. 

3.  Pewter  slip-top  spoon,  6f"  long,  found  at  Bermondsey.    XVI 

century. 

4.  Pewter  slip-top  spoon,  6i"  long,  found  at  Bermondsey.     XVI 

century. 

5.  Laton  slip-top  spoon,  6£"  long,  found  at  Bermondsey,  has 

remains  of  tin  plating.    XVII  century. 

6.  Laton  slip-top  spoon,  7"  long,  found  at  Bermondsey.      XVII 

century. 

7.  Laton  slip-top  spoon,  6^"  long,  found  in  London,  has  remains 

of  tin  plating.     XVII  century. 

8.  Pewter  slip-top  spoon,   6£"  long,    found  in  London.     XVI 

century. 

9.  Pewter  slip-top  spoon,  6|"  long,  found  in  London.     XVI 

century. 


WHAT  PEWTER  IS 

previously  pointed  out,  antimony  is  employed, 
but  the  separation  of  this  is  a  tiresome  and 
difficult  task,  only  possible  to  a  trained  ana- 
lytical chemist.  Its  presence,  however,  may 
possibly  be  deduced  in  the  process  of  securing 
the  copper  from  the  orange  tinge  which  its 
sulphide,  in  sufficient  quantities,  would  impart 
to  the  black  precipitate. 

As  these  directions  are  intended  solely  for 
those  unaccustomed  to  chemical  operations, 
it  may  not  perhaps  be  impertinent,  in  conclu- 
sion, to  call  attention  to  the  fact  that  the  acids 
used  are  not  only  virulently  poisonous  but 
distinctly  deleterious  in  outward  application 
to  hands,  clothes,  tablecloths,  &c.,  and  should 
consequently  be  handled  with  ample  caution, 
and  either  thrown  away  at  once  or  carefully 
locked  up  when  done  with. 


THE  THIRD  CHAPTER 

HOW  PEWTER  WAS  WROUGHT 


URING  the  Middle  Ages,  when 
the  now  nearly  extinct  craft  of 
the  pewterer  was  in  the  heyday 
of  its  prosperity,  the  actions  of 
its  followers  were  circumscribed, 
their  every  proceeding  regulated,  and  their 
offences  against  the  rules  and  ordinances  from 
time  to  time  established  punished  with  a 
rigour  and  ruthlessness  unequalled  even  by 
the  tyrannous  edictsof  modern  Trades  Unions. 
Unlike  these  last,  however,  the  laws  were 
made  not  solely  for  the  benefit  of  the  so-called 
working  man  but  for  the  general  good.  The 
cynical  doctrine,  caveat  emptor,  found  no 
adherents  in  those  days.  The  rules  of  the 
Company  were  not  indeed  wholly  unconcerned 
with  the  welfare  of  the  craft  and  craftsmen, 
some  of  them  in  fact  display  a  very  sufficient 
amount  of  greed  and  selfishness,  but  they  did 
not  set  on  one  side  as  useless  or  even  harm- 
ful its  good  name  and  honour.  The  modern 
idea  would  seem  to  be  that  no  man,  however 
skilful,  however  diligent,  shall  be  permitted 
24 


DESCRIPTIONS  OF^  OBJECTS  ON  PLATE  XVII. 

THE   MAKKRS'    MAKKS,   WHERE   LEGIBLE,    ARE 
SHOWN    ON    PLATE    CV.,    FACING    PAGE     I  50. 

1.  Laton  seal-top  spoon,  5^"  long,  found   in   London.     XVII 

century. 

2.  Laton  seal-top  spoon,  6i"  long.    XVII  century. 

3.  Laton  seal-top  spoon,  6j"  long.     XVII  century. 

.4.  Laton  seal-top  spoon,  6\"  long,  found  in    London.     XVII 
century . 

5.  Laton  seal-top  spoon,  6^,"  long,  found  in  London.    XVII 

century. 

6.  Laton  seal-top  spoon,  £V  long,  found    in   London.     XVII 

century. 

7.  Laton  seal-top  spoon,  6|"  Icng,  found  in   London.     XVII 

century. 

8.  Laton  seal-top  spoon,  6^"  long,  found    in  London.     XVII 

century. 


HOW  PEWTER  WAS  WROUGHT 

to  turn  out  more  work  or  better  work  than  the 
most  sluggish  and  incompetent  can  produce 
in  the  same  time.  The  old  notion  was  that 
no  man  should  be  allowed  to  foist  upon  an 
innocent  public  work  below  a  certain  definite 
standard  of  merit,  while  any  and  every  man 
should  be  encouraged  to  advance  as  far  beyond 
that  standard  as  in  him  lay.  It  was  recognised 
that  it  was  not  only  to  the  credit  but  to  the 
interest  of  the  craft  as  a  whole  to  deal  honestly 
with  the  customer  and  to  make  sure  that  he 
was  given  exactly  what  he  expected  and  had 
a  right  to,  not  the  omnipresent  "  just  as  good  " 
of  nowadays,  which  in  the  majority  of  cases 
means  infinitely  worse.  How,  when,  and 
where  the  vessel  was  made,  its  weight,  quality, 
and  price,  by  whom,  to  whom,  and  in  what 
places  it  was  sold,  were  all  elaborately  provided 
for,  and  woe  betide  the  knave  or  fool  who 
made  or  dealt  otherwise.  The  first  conviction 
for  illegal  practices  was  followed  by  the  con- 
fiscation of  the  inferior  object  or  material,  and 
the  second  confiscation  was  supplemented  by 
"  punishment  at  the  discretion  of  the  Mayor 
and  Aldermen,"  while  a  third  resulted  in 
expulsion  from  the  craft,  which  in  those 
days  of  closely  guarded  Guilds  must  have 

D  25 


OLD  PEWTER 

meant  in  most  instances  starvation  or  the 
gallows. 

The  Company  did  not,  moreover,  confine 
itself  to  simply  ordering  the  craftsman  as 
trader,  but  interfered  most  minutely  with 
nearly  every  detail  of  his  daily  life.  With 
these  domestic  and  social  restraints,  curious 
and  often  amusing  as  they  are,  we  cannot 
here  concern  ourselves.  The  reader  who 
would  learn  more  may  be  confidently  referred 
to  Mr.  Welch's  exhaustive  "  History  of  the 
Pewterers' Company,"  and  Mr.  Ingleby  Wood's 
no  less  excellent  "Scottish  Pewter  Ware  and 
Pewterers."  We  must  content  ourselves  with 
summarising  the  general  trend  of  the  regula- 
tions governing  the  trade,  without  laying 
stress  upon  the  minor  differences  which 
prevailed  in  different  places. 

Before  the  master-pewterer  could  set  up 
in  business  for  himself  he  had  to  go  through 
a  long  apprenticeship,  as  a  rule  six  years,  with 
an  added  year  during  which  the  work  done  by 
the  then  fully  trained  craftsman  was  supposed 
to  repayhis  master  for  the  cost  he  had  incurred, 
and  no  master  was  allowed  to  be  so  "daring 
as  to  receive  any  workman  of  the  craft  if  he 
have  not  been  an  apprentice."  When  this 
26 


_ 
Q-X 


DESCRIPTIONS  OF  OBJECTS   ON  PLATE  XVIII. 
THE  MAKERS'  MARKS.  WHERE  LEGIBLE,  ARE 

SHOWN  ON  PLATE  CVL,  FACING  PAGE  152. 

1.  Laton  slip-top  spoon,  7!"  long,   plated   with  tin,   found   in 

London.     XVII  century. 

2.  Pewter  slip-top  spoon,  7"  long,  found  at  Bermondsey.     XVII 

century. 

3.  Laton  slip-top  spoon,  6J"  long,   found   in  London.     XVII 

century. 

4.  Laton  spoon,  j\"  long,  plated  with  tin.     XVII  century. 

5.  Laton  spoon,  5^"  long,  with  remains  of  tin  plating.     XVII 

century. 

6.  Laton  spoon,  7 J"  long,  plated  with   tin.     Handle   "  Pied  de 

Biche."     Mark  same  as  Fig.  5. 

7.  Laton  spoon,  7^"  long,  with  remains  of  tin  plating.     Handle 

"  Pied  de  Biche."     XVII  century. 

8.  Laton  rat-tailed  spoon,  7^"  long,  found  in  York  Road,  West- 

minster.    Handle  "  Pied  de  Biche."     XVII  century. 


HOW  PEWTER  WAS  WROUGHT 

time  of  probation  was  out  he  was  expected  to 
produce  under  test  conditions  his  " essay," 
certain  prescribed  vessels  which  differed  in 
nature  at  different  times  and  places,  and  only 
when  these  had  been  submitted  to  and  ap- 
proved by  the  authorities  of  the  craft  was  he 
allowed  to  take  up  his  "freedom,"  register  his 
private  "  touch  "  on  the  proper  plate  at  theCom- 
pany's  Hall,  and  set  up  in  business  for  himself. 
It  was  only  to  the  native-born,  however,  that 
even  these  preliminary  steps  were  open  ;  they 
had  no  altruistic  sentimentality  in  those  days, 
being  fully  determined  to  keep  their  trade  and 
its  secrets  to  themselves  as  far  as  possible, 
and  to  that  end  fining  any  master  employing 
a  foreigner  ten  pounds  besides  confiscating 
any  ware  made  by  him.  The  apprentice 
doubtless  began  with  the  merely  mechanical 
part  of  the  work,  advancing  by  degrees  to 
such  simple  matters  as  the  making  of  spoons, 
a  despised  task  relegated  to  the  young  or  the 
infirm,  and  gradually  progressing  to  more 
important  objects  as  he  gained  skill  in  the  par- 
ticular branch  he  practised.  These  in  England 
were  three — Sadware  men,  Hollow-ware  men 
and  Triflers,  corresponding  fairly  closely  with 
the  French  sub-divisions,  Potters  mditres  de 

27 


OLD  PEWTER 

forge,  Potters  dit  de  rond,  and  Potiers 
menuisiers.  The  Sadware  men,  who  were 
not  rated  very  highly  in  the  craft,  made 
dishes,  trenchers,  chargers,  and  other  more 
or  less  flat  and  open  vessels  of  weight.  The 
derivation  and  significance  of  the  term  are 
doubtful,  and  it  is  perhaps  more  ingenious 
than  allowable  to  refer  it  to  the  original 
meaning  of  "sad,"  as  given  by  Professor 
Skeat,  namely  "  satiated,"  and  to  attribute  its 
use  in  this  connection  to  the  fact  that  the 
material  used  by  the  members  of  this  branch 
was  that  "  fine  pewter"  which,  as  explained  in 
the  last  chapter,  consisted  of  tin  alloyed  with 
as  much  copper  as  "  of  its  own  nature  it  will 
take,"  in  other  words,  satiated,  or,  as  chemists 
would  say,  saturated.  This  conjecture,  far- 
fetched as  it  may  appear,  is  to  some  extent 
supported  by  the  circumstance  that  the  third 
branch,  the  Triflers,  were  indisputably  so 
called  because  they  worked  in  the  mixture  of 
82'or  83  parts  of  tin  to  18  or  17  of  antimony, 
known  as  "trifle."  The  term  Hollow-ware 
men  sufficiently  explains  itself  as  denoting  the 
nature  of  the  object  produced,  not  the  material 
of  which  it  was  fashioned. 

Whichever  of  these  three  branches  of  the 
28 


Q-X 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XIX. 

THE  MAKERS'  MARKS,  WHERE  LEGIBLE,  ARE 
SHOWN  ON  PLATE  CVI.,  FACING  PAGE  152. 

1.  Laton  ratrtailed  spoon,  8"  long,  plated  with  tin.    Handle  ';  Pied 

de  Biche."     XVII  century. 

2.  Pewter  rat-tailed  spoon,  7^"  long,  found  in  Newgate  Street. 

Handle  "  Pied  de  Biche."     XVII  century. 

3.  Laton  rat-tailed  spoon,  7"  long,  found  in  Yuik  Road,  West- 

minster, plated  with  tin.     XVII  century. 

4.  Pewter  rat-tailed  spoon,  7^"  long,  found  in  Bermondsey.    XVII 

century. 

5.  Pewter  rat-tailed    spoon,   j%"  long,  found  in   Bermondsey. 

XVII  century. 

6.  Pewter  rat-tailed  spoon,  7"  long.    XVII  century. 

7.  Laton  spoon,  5^"  long,  plated  with  pewter,  found  in  London. 

Late  XVII  century. 

8.  Bronze  rat-tailed  spoon,  5^"  long.     XVII  century. 

9.  Rat-tailed   pewter    chocolate  spoon,  4"  long,  found    in   the 

Wandle  at  Wandsworth.    XVII  century. 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XX. 

THE  MAKERS'  MARKS,  WHERE  LEGIBLE,  ARE 
SHOWN   ON    PLATE   CVI.,   FACING   PAGE    152. 

1.  Pewter  spoon,   8£"  long,   found   in  London.      Late    XVI II 

century. 

2.  Laton   dog-nose  gravy   spoon,   zof"  long,  plated   with   tin. 

XVII  century. 

3.  Pewter  spoon,  7|"  long,  found  in  Bermondsey.     Late  XVIII 

century. 

4.  Laton    slip-top   ladle,    8J"    long,   the    handle   of    hexagonal 

section.     XVII  century. 

5.  Laton  slip-top  ladie,  8"  long,  found  in  the  City  of  London,  has 

remains  of  tin  plating.     XVII  century. 

6.  Laton  slip-top   ladle,   7^"    long,   found  in   Suffolk.       XVII 

century. 

7.  Laton   slip-top   ladle,   6^"    long,    found   in   London.     XVII 

century. 

8.  Laton  spoon,  seal-headed,    6^"    long,    has   remains   of   tin 

plating,  found  at  Norwich.     XVII  century. 


HOW  PEWTER  WAS  WROUGHT 

trade  he  followed,  the  worker's  first  business 
must  have  been  to  make  his  alloy,  mixing  his 
tin  and  lead  or  copper  in  the  ordained  pro- 
portions, with  a  careful  eye  on  the  "  sharp 
laws  provided  in  that  behalf"  mentioned  by 
Harrison.  At  any  moment  the  searchers 
authorised  by  the  Master  and  Wardens  of  the 
Company  might  swoop  down  upon  him  and 
demand  to  assay  the  work  he  was  engaged 
upon  or  had  already  completed  for  sale,  and 
any  serious  deviation  from  the  standard 
meant  fine  and  confiscation  of  the  ley,  lea,  or 
lay  metal,  as  the  debased  material  was  called. 
That  he  might  not,  however,  be  punished  for 
a  fault  that  was  none  of  his  own,  the  Company 
were  empowered  to  assay  all  tin  on  its  arrival 
in  London,  and  thereby  to  shut  out  "  the  multi- 
tude of  tin  which  was  untrue  and  deceyvable 
brought  to  the  City,  the  defaults  not  being 
perceptible  until  it  comes  to  the  melting." 
Of  this  guaranteed  tin  it  was  customary  to 
cast  a  disk,  keeping  the  mould  so  that  in  any 
case  of  doubt  a  similar  disk  of  the  question- 
able alloy  might  be  cast  in  it,  when,  the 
weights  of  tin  and  lead  or  copper  respectively 
being  known,  the  calculation  of  the  quantity 
of  each  present  was  easily  made. 

29 


OLD  PEWTER 

The  tin  came  from  Cornwall,  the  Warden 
of  the  Company  having  the  legal  right  to 
purchase  at  the  market  price  one-fourth  of  all 
that  came  into  the  market,  retailing  it  at  a 
small  profit  to  the  freemen,  and  it  was  to  the 
superior  quality  of  the  metal  there  obtained 
that  English  pewter  owed  its  high  reputation 
in  foreign  lands.  The  birth  of  this  industry 
is  hidden  in  the  mists  of  remote  antiquity,  for 
as  an  element  in  bronze  the  use  of  the  metal 
goes  back  to  dim  pre-historic  times.  In  the 
Book  of  Numbers,  xxxi.  22,  it  is  included  with 
"  the  gold,  and  the  silver,  the  brass,  the  iron, 
and  the  lead  "  among  the  spoils  "  that  may 
abide  the  fire  "  which  were  ordered  by  Moses 
to  be  made  to  "go  through  the  fire"  for  pur- 
poses of  purification  after  the  victory  over  the 
Midianites.  Isaiah,  i.  25,  mentions  it  meta- 
phorically :  "  I  will  turn  my  hand  upon  thee, 
and  purely  purge  away  thy  dross,  and  take 
away  all  thy  tin  "  ;  and  Ezekiel,  in  the  same 
style,  xxii.  18  and  20,  says,  "  The  house  of 
Israel  has  become  to  me  as  dross  ;  all  they 
are  brass,  and  tin,  and  iron,  and  lead  in  the 
midst  of  the  furnace  "  ;  and,  "  as  they  gather 
silver,  and  brass,  and  iron,  and  lead,  and  tin 
into  the  midst  of  the  furnace,  to  blow  the  fire 
30 


_j  X 

Q.  X 


HOW  PEWTER  WAS  WROUGHT 

upon  it  to  melt  it,  so  will  I  gather  you  in 
mine  anger  and  in  my  fury ;  and  I  will  leave 
you  there  and  melt  you,"  while  in  the  recital 
of  the  various  markets  (xxvii.  12)  contri- 
buting to  the  splendours  of  Tarsus,  which 
precedes  the  prophecy  of  its  downfall,  he  states 
that  "  Tarshish  was  thy  merchant  by  reason 
of  the  multitude  of  all  kinds  of  riches  ;  with 
silver,  iron,  tin,  and  lead  they  traded  in  thy 
fairs."  It  is  not  probable  that  this  tin  which 
was  brought  in  by  the  merchants  of  Tarshish 
came  in  the  ships  of  Phoenician  or  Cartha- 
ginian seamen  from  these  little  islands  in  the 
far-away  northern  seas,  for  Tarshish  has  been 
plausibly  indentified  with  Tartessus,  a  port  in 
the  south  of  Spain,  which  country  is  known 
to  have  produced  tin  at  one  time,  and  Ezekiel 
certainly  wrote  centuries  before  Herodotus, 
who  did  undoubtedly  refer  to  them  as  the 
Cassiterides,  from  the  Greek  word  /cao-o-iVe^o?, 
which  is  made  use  of  by  Homer  also,  and 
other  Greek  writers.  Tin,  "  plumbum  candi- 
dum,"  as  Pliny  calls  it,  in  contradistinction  to 
lead,  "plumbum  nigrum,"  is  frequently  re- 
ferred to  by  Roman  authors,  and  was  very 
extensively  imported  in  their  days  from 
England  to  the  continent  in  the  form  of  small 


OLD  PEWTER 

dice-like  cubes,  being  embarked  at  Ictis,  an 
uncertain  locality  believed  by  various  differing 
authorities  to  be  St.  Michael's  Mount,  Fal- 
mouth,  Weymouth,  and  the  Isle  of  Wight, 
and  delivered  for  distribution  at  Marseilles  in 
Roman  days ;  later  principally  at  Bruges. 
All  the  other  metals  employed  were  also 
home  products,  antimony  in  the  form  of 
sulphide  (stibnite),bismuth,  copper,  zinc  in  the 
form  of  zinc-blende,  and  lead  being  all  found 
in  Cornwall,  though  most  of  the  last,  at  any 
rate,  originally  came  apparently  from  Derby- 
shire, if  we  may  form  any  conclusion  from 
the  fact  that  it  was  known  as  "  Peak  "  to  the 
London  workers. 

The  alloy  once  duly  compounded  accord- 
ing to  law,  or  bought  ready  mixed — as  was 
evidently  sometimes  the  case  since  it  was  for- 
bidden to  buy  it  by  night,  in  other  words 
clandestinely,  or  from  tilers,  labourers,  boys, 
or  women  !  all  of  whom  were  presumably  ex- 
pected to  have  stolen  it,  or  to  sell  old  pewter 
as  new,  such  when  bought  being  obligatorily 
melted  down  and  recast — its  subsequent  treat- 
ment depended  on  the  article  into  which  it 
was  to  be  converted  and  the  class  of  workmen 
using  it.  For  making  the  largest  dishes  and 
32 


_G 

J3 

O 

CO      £ 

o 

T  if 
"  £  W 


0> 


buO 

w 


o  >> 

I 


O> 

ox 


HOW  PEWTER  WAS  WROUGHT 

chargers  it  was  rolled  into  a  sheet,  and  the 
Sadware  man  was  then  compelled  to  fashion 
it  entirely  by  hammering,  a  process  which 
consolidated  the  metal  and  gave  the  necessary 
rigidity  while  at  the  same  time  producing  the 
desirable    smoothness   of    surface.     Smaller 
plates  and  dishes  and  all  other  articles  were 
cast.     The   moulds   for    this    purpose  were 
usually  made  of  gun-metal  finished  with  pre- 
cise  care,  and   as   these  were   consequently 
expensive  it  was  often  the  custom  for  the 
Company  to  own  them  in  common,  each  mem- 
ber taking  his  turn   in    borrowing  one  or 
another  as  he  required  it,  giving  proper  notice 
to  the  Company  when  he  was  about  to  make 
castings  in  order  that  a  searcher  might  attend 
to  make  sure  that  only  good  metal  was  used. 
Several  lists  of  the  various  forms  of  these 
have  been  preserved,  the  earliest  of  which,  as 
recorded  by  Mr.  Welch,  dating  from  1425,  is 
is  as  follows :  "  i  holow  scharyder,"  i  C  plat- 
molde,   i  C  dysche  molde,  i  C  sawsyrmolde, 
i  medyll  plat  molde,  i  medyll  dysche  molde, 
i  medyll  sawsyr  molde,    i  kyngs  ys  dysche 
molde,  i  holow  dysche  molde,  i  holow  sawsyr 
molde,   i   saly  dysche  molde,   i  saly  sawsyr 
molde,  i  salu  bolle  molde,  i  qware  bolle  molde, 

E  33 


OLD  PEWTER 

i  trechor  molde."  Some  of  these  terms  are 
of  very  doubtful  import,  but  a  comparison  of 
this  list  with  one  of  the  regulation  weights 
for  various  articles  as  laid  down  in  1430, 
making  due  allowance  for  the  eccentricity  of 
spelling  at  the  time,  suggests  that  they  were  a 
mould  for  hollow  chargers,  each  of  which  had 
to  weigh  2j  pounds  (the  larger  ones,  weigh- 
ing respectively  7,  5  and  3^  pounds,  were 
hammered) ;  large,  middle,  and  hollow  plate, 
dish,  and  saucer  moulds,  weighing,  for  the 
first  2^,  2  and  i  J  pounds  each,  for  the  second 
i£,  i£  and  i  pound  each,  and  for  the  third  f , 
TV  and  £  of  a  pound  each  ;  and  a  mould  for 
king's  dishes,  each  of  which  weighed  impounds. 
The  saly  dish,  saucer,  and  bowl  would  seem 
to  correspond  with  the  galley  dishes  and 
saucers  of  the  second  list,  weighing  either  i 
pound  or  f  of  a  pound  each,  which  perhaps 
were  equivalent  to  the  "  flat,  cowped,  and 
squard  saler  "  of  the  York  regulations  of  1419, 
the  "  12  sallite"  dishes  in  an  inventory  of  Sir 
William  Fairfax's  possessions  at  Gilling 
drawn  up  in  1564,  and  Sir  Richard  Poullett's 
"  14  small  sallet  pewter  dishes  "  in  an  inven- 
tory of  1618.  The  meaning  of  a  qware  bowl 
is  hard  to  guess,  for  the  obvious  suggestion, 
34 


CLX 


u  6 


HOW  PEWTER  WAS  WROUGHT 

square,  seems  highly  hazardous,  and  the  sup- 
position that   "trechor"  is   simply  trencher 
misspelt  is  scarcely  less  so.     This  collection 
of  moulds,  however,  was  far  from  complete, 
for  the  regulations  as  to  weight  include  also 
a  middle  charger  weighing  3^  pounds,  a  second 
sized  platter  weighing  2^  pounds,  a  hollow 
dish  weighing  ^  of  a  pound,  a  small  hollow 
dish  weighing!  of  a  pound,  a  Cardinal's  hatte, 
saucers   weighing     15    pounds    the    dozen, 
Florentine  dishes  and  saucers  (greatest  size) 
weighing  13  pounds  the  dozen,  next  size  of 
the  same  weighing  12    pounds,  and  "  small 
holies  "  13  pounds  the  dozen.     The  weights 
and  fashions  seem  also  to  have  been  modified 
as  time  went  on,  for  in  a  further  list  of  York 
moulds  drawn  up  two  hundred  years  later  we 
find  such  new  names  as  "least  dubler,"  "brod 
border  dish,"  "  unmouldishe,"  and  "  banquitin 
dish,"  while  the  smaller  chargers  then  weighed 
4  and  3  pounds,  the   smallest   dish   f   of  a 
pound,  the  largest  platter  3  pounds,  and  the 
smallest  i  pound.     The  mould,  whatever  its 
name  or  shape,  was  first  coated  inside  with 
fine   pumice   powder,    sandarach,  a  resinous 
body  obtained  from  a  small  coniferous  tree, 
Callitris  quadriualvis,  native  in  North-West 

35 


OLD  PEWTER 

Africa,  or  white  of  egg  and  red  ochre,  and 
having  been  securely  closed  if  made  in  two 
pieces,  or  properly  fitted  together  if  in  more, 
was  filled  directly  with  the  molten  metal  and 
left  to  cool.  When  taken  out  the  surface  has 
a  singular  and  somewhat  unpleasing  colour 
and  texture,  and  this  had  to  be  removed, which 
was  effected  in  various  ways.  Sadware  and 
spoons  which  were  cast  in  one  piece  had  to  be 
hammered  and  burnished,  and  any  attempt  to 
save  time  and  labour  at  the  expense  of 
strength  by  the  use  of  a  lathe  was  sternly 
repressed,  while  in  1686  an  ingenious  spoon- 
maker  named  Burton  was  only  grudgingly 
permitted  to  employ  an  "engine"  which, 
probably  by  stamping  them,  turned  out 
spoons  of  good  quality,  on  condition  that  he 
did  not  take  advantage  of  the  lesser  cost  of 
his  process  by  selling  his  goods  under  the 
market  price,  at  that  time  six  shillings  a  gross 
in  the  country  and  four  shillings  in  London. 
The  finishing  of  hollow-ware  was  generally 
a  much  more  elaborate  process.  These  also, 
when  possible,  as  in  the  case  of  porringers 
and  other  more  simple  forms,  were  cast  whole, 
but  the  more  elaborate  pieces  with  curved 
sides,  moulded  rims  and  bases,  handles,  lids 
36 


PLATE  XXV 


H.  i3J" 
CHURCH  FLAGON,  Scotch.     Early  XVIII  century  type. 


HOW  PEWTER  WAS  WROUGHT 

and  other  accessory  parts,  were  cast  in  parts 
which  were  subsequently  soldered  together, 
and  the  completed  article  turned,  scraped, 
and  burnished  on  a  lathe.  Lastly,  the  maker 
was  expected  to  mark  each  vessel  by  stamping 
on  it  the  quality  mark  and  his  own  private 
"touch,"  in  order  that  in  case  of  any  deficiency 
in  weight  or  quality  it  might  be  traced  to 
him.  Failure  to  comply  with  this  sensible 
regulation  was  punishable  by  a  fine  of  forty 
shillings.  A  third  method  of  working  pewter 
was  " spinning,"  by  which  the  metal  was 
pressed  with  a  blunt  steel  tool  into  a  wooden 
mould  spun  on  a  lathe. 

The  "how  "  of  the  manufacture  being  so 
minutely  provided  for,  the  "when"  and  the 
"where"  were  not  neglected.  Working  at 
night  was  forbidden,  on  a  penalty  of  forty 
pence  for  the  first  offence,  eighty  for  the 
second,  and  the  mysterious  and  awesome 
"  discretion  "  of  the  Mayor  and  Aldermen  for 
the  third,  ostensibly  on  the  grounds  that  the 
sight  was  not  so  profitable  by  night  or  so 
certain  as  by  day,  or  on  holy  days,  a  restric- 
tion due  to  the  fact  that  this,  like  all  other 
guilds,  was  in  the  beginning  a  partly  religious 
body.  The  regulations  as  to  the  "  where  "  were 

37 


OLD  PEWTER 

designed,  as  far  as  might  be,  to  keep  the  craft 
in  the  hands  of  the  London  pewterers,  for, 
though  they  had  the  right  of  search  through- 
out England,  and  were  empowered  to  assay, 
rigorously  enough,  we  may  be  sure,  all  pewter 
imported  into  London  from  the  provinces 
(none  was  admitted  from  abroad),  they  sternly 
condemned  as  "  evil-disposed  persons  "  those 
who,  having  learned  their  trade  in  the  City, 
went  "  for  their  singular  lucre  into  strange 
regions  and  countries  "  to  exercise  it,  inciden- 
tally revealing  its  secrets  to  the  "  foreigner." 
They  even  claimed  the  right  to  command  any 
English  pewterers  working  on  the  continent 
to  return  within  three  months  and  perma- 
nently establish  themselves  in  London  on 
pain  of  forfeiting  their  nationality  and  right  to 
"  the  king's  protection,"  while  for  the  same 
selfish  ends  it  was  ordained  that  no  pewterer 
should  work  in  a  shop  open  to  the  street, 
where  any  country  pewterer  passing  by  might 
spy  out  the  process  and  haply  profit  thereby. 
They,  at  any  rate,  were  troubled  by  no  doubts 
as  to  the  superiority  of  a  policy  of  protection 
over  free  trade,  and  "the  open  door"  they 
favoured  was  one  that  only  opened  one  way 
— outwards.  At  the  same  time,  be  it  acknow- 
38 


O 


HOW  PEWTER  WAS  WROUGHT 

ledged,  they  took  good  heed  that  only  honest 
ware  should  issue  from  it,  for  no  ware  was 
allowed  to  be  sent  out  of  the  City  unless  it 
had  been  previously  assayed. 

The  selling  of  the  finished  ware  was  as  fully 
provided  for  as  the  making  of  it.  All  work 
was  to  be  sold  by  the  pewterer  himself  in  his 
own  shop,  unless  by  special  arrangement  with 
the  would-be  purchaser,  or  was  to  be  offered 
publicly  at  recognised  fairs  and  markets  where 
the  searchers  could  inspect  and  test  it  if  they 
thought  necessary,  and  such  pewter  must  be 
new,  the  re-selling  of  old  being  strictly  for- 
bidden. Table  pewter  was  usually  sold  by 
the  "  garnish,"  according  to  Harrison,  con- 
sisting of  12  platters,  12  dishes,  and  12 
saucers,  and  paid  for  by  the  pound,  which 
in  his  time  was  valued  at  sevenpence  or 
eightpence.  Other  articles  were  presumably 
sold  separately,  also  at  so  much  a  pound,  the 
amount  being  calculated  from  the  price  of  tin 
with  a  small  addition  for  the  cost  of  working. 
A  reversal  of  the  rules  as  to  selling  was  made 
in  the  case  of  hiring-out  pewter,  only  old 
pewter  being  admissible  for  the  latter  purpose. 
This  would  seem  to  have  been  a  very  profit- 
able branch  of  the  trade,  whether  the  goods, 

39 


OLD  PEWTER 

known  as  "  feast-vessels,"  were  only  loaned 
temporarily  to  some  one  entertaining  for  the 
nonce  on  an  exceptionally  lavish  scale,  or 
were  let  by  the  year,  as  was  not  infrequently 
the  custom.  An  Earl  of  Northumberland, 
for  example,  during  the  fourteenth  century,  is 
recorded  as  having  been  in  the  habit  of  hiring 
no  less  than  one  hundred  dozen  vessels,  at  a 
charge  of  fourpence  per  dozen  per  annum. 

On  the  ornamentation  of  pewter  there  is 
no  need  to  dwell  at  length.  As  a  broad  rule, 
indeed,  it  may  be  said  that  the  less  extraneous 
decoration  it  has  the  better.  Good  pewter 
should  rely  for  its  decorative  effects  on  its 
structural  incidents.  The  fineness  of  its  lines, 
the  elegance  of  its  curves,  its  Tightness  of 
proportion,  an  obvious  adaptation  to  purpose 
should  be  all-sufficient,  and  any  essential  em- 
bellishments, such  as  the  mouldings  on  rims, 
bases,  or  strengthening  bands,  should  be  as 
simple  as  possible,  and  should  manifestly 
display  their  object  at  a  glance.  The  orna- 
ment should  be  architectural  rather  than 
sculpturesque,  and  this  much  came  fittingly 
within  the  province  of  the  maker.  His  wares 
were  intended  for  use,  and  as  a  consequence 
for  subsequent  fairly  easy  and  wholly  efficient 
40 


HOW  PEWTER  WAS  WROUGHT 

cleansing,  and  such  additions  as  complicated 
coats-of-arms  in  high  relief  in  the  middle  of 
dish  or  platter,  which  were  made  in  later  and 
degenerate  days,  were  opposed  to  both.  Even 
the  attempt  to  tamper  with  the  natural  in- 
tegrity of  the  plain  surfaces  was  rarely  made 
during  the  best  period,  and  still  more  rarely 
successful  ;  and  it  is,  indeed,  at  least  open  to 
question  whether  many,  if  any,  of  the  instances 
that  do  occur  of  chasing,  engraving,  repoussd 
work,  or  other  niggling  supposed  embellish- 
ments were  the  work  of  the  pewterer  himself, 
but  were  not  rather  the  unnecessary  emenda- 
tions of  later  ignorance  and  lack  of  taste.  They 
have  in  most  cases  an  evident  amateurishness 
and  inappropriateness  that  seem  to  stamp 
them  as  often,  if  not  always,  the  evil  inspira- 
tions of  people  who  did  not  rightly  understand 
the  characteristics  of  the  object  or  appreciate 
its  special  charm.  It  would,  perhaps,  be  too 
rash  to  assert  dogmatically  that  the  subtle 
artistic  instinct  which  would  seem  to  have 
been,  almost  universally,  the  heritage  of  even 
the  humblest  craftsman  in  those  days,  assured 
the  pewterer  that  the  texture  and  colour  of 
his  material  had  a  beauty -of  their  own  which 
the  distraction  of  frivolous  detail  would  only 

F  41 


OLD  PEWTER 

detract  from,  but  there  can  be  little  doubt  that 
he  fully  realised  that  his  handiwork  was 
destined  to  constant  and  sometimes  rough 
usage ;  that  this  in  the  hands  of  good  house- 
wives entailed  constant  scouring;  and  that 
in  the  process,  with  a  metal  so  soft  and  easily 
worn  away  as  that  in  which  he  wrought,  any 
decoration,  whether  incised  or  in  relief,  would 
speedily  be  injured  if  not  obliterated.  It  was 
a  period  of  decadence  when  pewter  "  de  belle 
fasson"  and  "  fasson  d'argent"  became  the 
fashion,  and  the  elaborate  salvers,  basins,  and 
ewersof  Briot  and  Enderlein,  cast  in  fragments 
and  laboriously  pieced  together  into  a  useless 
object  speciously  imitating  silver,  were  in- 
evitably the  forerunners  of  the  later  gilding, 
painting,  lacquering,  inlaying,  and  other  gross 
offences  against  the  dignity  of  pewter,  by 
means  of  which  inartistic  workmen,  we  may 
hope  solely  at  the  instigation  of  vulgar  and 
ostentatious  customers,  endeavoured  to  dis- 
guise with  frills  and  fripperies  the  humble 
but  honest  nature  of  their  wares.  It  may 
doubtless  be  maintained  that  fundamentally 
neither  difficulty  of  execution,  durability  of 
the  result  nor  cheapness  of  the  actual  material 
ought  to  influence  the  judgment  in  the  con- 
42 


w 


HOW  PEWTER  WAS  WROUGHT 

sideration  of  a  work  of  art,  but  one  could  not 
but  doubt  the  sanity  of  a  sculptor  who  elected 
to  carve  in  crumblingsandstonewhen  he  might 
with  equal  ease  obtain  enduring  marble,  or 
his  honesty  of  purpose  if  he  painted  it  with 
white  enamel  to  simulate  it.  The  artist  who 
truly  respects  himself  and  his  art  will  not 
sacrifice  durability  to  ease  of  manufacture,  but 
will  strive  to  make  his  work  sound  as  well  as 
pleasing.  It  is  not  because  silver  is  so  much 
more  costly  than  pewter,  but  because  it  is  of 
its  nature  so  much  more  lasting,  that  the  treat- 
ment which  is  right  and  proper  for  the  one  is 
wrong  and  unfit  for  the  other.  To  approach 
in  the  second  the  effects  obtainable  in  the  first 
necessitates  either  unpardonable  weakness  or 
impracticable  weight,  and,  at  the  risk  of 
tediousness,  it  cannot  be  too  often  asserted 
that  pewter  fashioned  in  a  shape  that  clearly 
prohibits  usefulness  is  pewter  misapplied. 


43 


THE  FOURTH  CHAPTER 

SOME   FACTS   ABOUT    PEWTER 

BEFORE  THE  FIFTEENTH 

CENTURY 


S  with  gunpowder,  the  mariner's 
compass,  and  other  useful  inven- 
tions, it  seems  possible  that  the 
Chinese — that  strange  race  who 
advanced  so  far  in  the  path  of 
civilisation  centuries  ago,  and  then,  refusing  a 
step  further,  settled  down  into  hide-bound 
convention — predate  the  Western  nations  in 
the  use  of  pewter  also,  but  the  early  history 
both  there  and  here  is  so  obscure  that  it  is 
impossible  to  speak  with  any  certainty.  In 
the  entire  absence  of  all  early  records  it  is 
permissible  to  believe  that  the  knowledge  of 
pewter  must  be  nearly,  if  not  quite,  as  ancient 
as  that  of  bronze,  for  it  is  scarcely  credible  that 
the  pre-historic  man  who  conceived  the  idea 
of  mixing  tin  and  copper  and  availing  him- 
self of  the  superior  advantages  of  the  blend 
to  either  metal  by  itself  should  have  not 
tested  also  the  desirability  of  combining  the 
brightness,  lightness,  and  rigidity  of  tin  with 
44 


. 


3     O     . 

-II 


I 


w "% 
' 


iga* 

-S   ^  ^  1;  % 

U   O   -   o   O 


BEFORE  FIFTEENTH  CENTURY 

the  greater  toughness  and  malleability  of  lead. 
This,  however,  in  the  present  state  of  our 
knowledge  must  remain  mere  speculation. 
We  first  reach  sure  ground  in  the  days  of  the 
Roman  empire.  Not  only  do  we  meet  with 
probable  references  to  pewter  in  the  works  of 
Latin  writers,  as  in  Plautus,  who  describes  a 
magnificent  feast  served  on  what  is  generally 
considered  to  have  been  pewter,  and  in 
Suetonius,  who  states  that  Vitellius  removed 
the  silver  vessels  from  the  temples,  replacing 
them  with  pewter,  thus  initiating  a  practice 
which  succeeding  generations  followed  more 
than  once,  but  no  inconsiderable  number  of 
pewter  vessels  of  Roman  make  are  still  in 
existence.  Important  finds  have  been  made 
in  our  own  land  at  various  times,  chief  among 
which  ranks  the  remarkable  collection  now  in 
the  British  Museum,  unearthed  by  the  Rev. 
R.  G.  Engleheart,  at  Appleshaw,  Hampshire, 
in  1877,  though  the  discoveries  at  Icklingham, 
Suffolk,  in  1840,  Sutton  in  the  Isle  of  Ely, 
about  1848,  Southward,  Colchester  and  else- 
where were  also  notable.  The  Appleshaw 
district,  which  lies  five  miles  north-west  of 
Andover,  near  the  intersection  of  the  Roman 
roads  which  ran  from  Old  Sarum  to 

45 


OLD  PEWTER 

Silchester  and  from  Winchester  to  Ciren- 
cester,  had  long  been  famous  for  its  remains 
of  Romano-British  constructions,  when  in 
January  1897  Mr.  Engleheart  learned  that  the 
plough  had  revealed  in  the  shape  of  an 
inscribed  stone  indications  of  yet  another. 
The  details  of  the  building,  which  are  fully 
described  in  Archczologia,vQ\.  Ivi.,  do  not  con- 
cern us  here,  but  the  pewter  vessels  found  at 
the  time  are  of  unusual  interest.  "They 
appeared  to  be  designedly  hidden  in  a  pit 
sunk  through  a  cement  floor,  three  feet  below 
the  surface  of  the  field.  The  smaller  vessels 
were  carefully  covered  by  the  larger  dishes." 
They  included,  as  described  by  Mr.  Charles 
K.  Read  in  an  appendix  to  the  paper,  ten 
circular  dishes  from  twenty-two  to  fourteen 
and  three-quarter  inches  in  diameter,  a 
square  dish,  fifteen  and  a  half  inches  in 
diameter,  with  a  semi-circular  projection  in 
the  middle  of  each  side  [Plate  I.  2],  a  chalice- 
shaped  cup  of  which  more  anon,  an  octagonal 
jug,  a  portion  of  a  circular  one  roughly  in- 
scribed VICTRICI,  the  two  last  letters  US 
having  been  broken  off,  fragments  of  three 
cups,  three  bowls  each  with  a  curious  hori- 
zontal flange  round  the  outside,  five  hemi- 
46 


BEFORE  FIFTEENTH  CENTURY 

spherical  bowls  about  two  inches  high  and  from 
six  and  one-eighth  to  four  inches  in  diameter, 
two  saucers,  two  small  plates,  a  portion  of  a 
vase,  a  deep  dish  ten  inches  in  diameter,  and 
an  oval  dish,  originally  nine  inches  in  length, 
of  nearly  pure  tin.  Many  of  them  are  remark- 
able for  their  ornamentation.  Several  of 
the  dishes  [Plate  I.  i  and  3  and  Plate  II.] 
are  decorated  with  geometrical  designs  of 
intricately  interlaced  strapwork  produced  by 
punching  a  wedge-shaped  furrow  afterwards 
filled  in  with  a  black  bituminous  material, 
not  metallic  as  in  the  later  Italian  niello, 
which  when  the  metal  was  bright  must  have 
had  a  highly  decorative  effect.  The  oval 
dish  [Plate  III.  3]  has  a  fish  in  relief 
within  an  oval  border  of  interlaced  work, 
and  one  of  the  saucers  [Plate  III.  i]  is  clearly 
marked  on  the  bottom  with  the  well-known 
Christian  symbol  >P,  in  connection  with  which 
two  facts  Mr.  Engleheart,  in  a  letter  which 
he  has  kindly  permitted  me  to  quote,  offers  a 
most  important  suggestion :  "  Though  I 
dislike  guesses  in  archaeology,"  he  writes, 
"  I  cannot  altogether  dismiss  a  surmise  of 
some  interest  which  has  occurred  to  me.  In 
this  collection  of  vessels  we  have  (i)  a  patera 

47 


OLD  PEWTER 

marked,  as  though  to  distinguish  it,  with  the 
Christian  monogram;  (2)  a  small  dish  stamped 
with  the  fish,  a  well-known  Christian  em- 
blem, the  form  being  almost  exactly  identical 
with  the  one  common  in  the  catacombs ;  (3)  a 
vessel  which  immediately  and  forcibly  suggests 
a  Chalice  [Plate  III.  2].  Is  it  not  at  least  pos- 
sible that  we  have  here  a  very  early  instance  of 
a  '  pewter  communion  service  ? '  If  the  inmates 
of  the  villa  were  Christians,  as  would  seem 
to  be  indicated  by  the  P,  it  is  highly  probable 
that  they  would  earmark  some  of  their  best 
vessels  for  Eucharistic  use.  And  it  is  a 
curious  coincidence  that  a  small  ingot  of 
pewter  in  the  British  Museum  stamped  with 
the  P  is  practically  identical  in  analysis  with 
the  similarly  marked  patera"  on  which  point 
Mr.  Gowland  speaks  even  more  strongly, 
saying  that  "  it  is  of  precisely  the  same  com- 
position." Mr.  Engleheart  goes  on  to  say 
that  "C.  N.  Read  objects  to  the  chalice 
theory  that  the  edge  of  the  bowl  is  reflexed, 
making  it  difficult  to  drink  out  of.  But  the 
vessel  is  surely  a  cup.  He  suggests  it  is  a 
lamp,  but  I  have  never  seen  a  lamp  shaped 
at  all  like  this,  and  I  have  seen  most  of  the 
collections  of  Roman  ware  at  home  and 
48 


PLATE  XXXII 


1     H.  24"  2     H.  23t" 

ALTAR  CANDLESTICKS.     XVII  century. 


3    H. 


BEFORE  FIFTEENTH  CENTURY 

abroad."  When  such  learned  doctors  differ  it 
is  meet  for  the  layman  to  suspend  judgment. 
The  reader  may  form  his  own  conclusions 
from  an  examination  of  the  reproduction 
[Plate  III.] ;  but  with  regard  to  the  objection 
to  the  chalice  theory  founded  on  the  reflexed 
lip  it  may  be  pointed  out  that  in  a  large 
number  of  surviving  chalices  of  ancient  date 
the  lip  will  be  seen  to  be  more  or  less  turned 
outwards  in  the  same  way. 

Mr.  Engleheart's  supposition,  if  it  be  as 
correct  as  it  is  probable,  has  a  considerable 
bearing  on  the  question  of  the  date  of  the 
objects.  In  his  original  remarks  upon  the 
find,  he  argued  on  this  point  as  follows : 
"  Lying  on  the  floor,  below  which  they 
were  buried,  was  a  fragment  of  wall-plaster 
bearing  a  peculiar  pattern  of  red  flower-buds 
on  a  white  ground,  absolutely  identical  with 
plaster  found  in  the  Clanville  villa.  Now 
the  inscribed  stone  found  in  the  latter  proves 
that  the  house  was  inhabited  in  the  year 
234  A.D.,  while  the  coins  cease  with  Decentius, 
351  A.D.  Therefore,  on  the  not  unreasonable 
suppositions  (i)  that  the  plaster,  as  found, 
represents  the  wall-decoration  of  the  houses 
at  the  time  of  their  destruction  or  abandon- 

G  49 


OLD  PEWTER 

ment ;  (2)  that  the  identity  of  design  shows 
a  correspondence  of  dates  ;  (3)  that  the  vessels 
were  concealed  when  the  house  was  abandoned, 
we  may  assign  the  vessels  to  a  period  not  by 
many  years  removed  from  350  A.D."  To 
which  conclusion  he  now  appends  that  as 
"  the  P  would  scarcely  have  appeared  before 
Constantine's  sanction  of  Christianity  in  the 
year  311,"  the  probable  date  cannot  well  be 
earlier  than  that.  As,  moreover,  the  Roman 
legions  left  Britain  in  411  never  to  return,  it 
is  unlikely  that  they  were  made  after  that, 
and  we  may  therefore  feel  assured  that  we 
have  in  them  examples  far  more  ancient  than 
anything  the  Eastern  world  can  show. 

The  oldest  of  these,  indeed,  to  which  an 
approximately  definite  date  can  be  attributed, 
were  some  spoons  seen  by  Mr.  Gowland  in 
the  treasure-house  at  Nara,  in  Japan,  among 
the  dresses  and  decorations  of  the  Court 
which  are  known  to  have  been  deposited 
there  on  the  accession  of  the  Emperor 
Kwammu  in  784  A.D.,  when  Kioto  became 
the  capital,  a  few  years  after  the  first  record 
of  the  use  of  pewter  made  from  tin  found  in 
the  country  instead  of  imported  from  China 
as  heretofore,  which  innovation  took  place 
50 


PLATE  XXXIII 


1     H.  24!"  2     H.  35" 

ALTAR  CANDLESTICKS,  Flemish.     XVII  century. 


BEFORE  FIFTEENTH  CENTURY 

during  the  reign  of  the  Empress  Shokotu, 
which  lasted  from  765  A.D.  to  770  A.D. 

After  the  withdrawal  of  the  Romans,  the 
history  of  England,  as  in  fact  of  all  Western 
Europe,  becomes  one  long  tale  of  ceaseless 
wars.  The  hapless  Britons,  overmastered  by 
the  Picts  and  Scots,  adopted  the  desperate 
remedy  of  calling  to  their  help  the  Angles 
from  the  Jutland  shores,  only  to  find  too  late 
that  in  place  of  a  trusty  house-dog  they  had 
enticed  into  their  fold  a  ravening  wolf.  Year 
after  year,  century  after  century,  with  few  and 
brief  intervals  of  a  respite,  the  rashly  invited 
invaders  tightened  and  widened  their  grasp 
upon  the  land  they  were  called  in  to  defend, 
driving  the  natives  step  by  step  further  and 
further  westward  and  northward,  and  during 
the  protracted  struggle  we  may  feel  sure  that 
any  lingering  traces  of  civilisation  which  had 
survived  the  downfall  of  the  Roman  domi- 
nation were  utterly  swept  away.  How  and 
where  in  Europe  the  memory  of  the  peaceful 
arts,  among  them  that  of  pewter-making, 
was  preserved,  when  and  where  the  seeds, 
so  lying  dormant,  quickened  again  into  life, 
we  do  not  know.  Certain  it  is,  and  that  is 
all  that  is  certain  in  the  matter,  that  between 


OLD  PEWTER 

this  pewter  from  Appleshaw  and  the  next 
objects  of  which  we  have  knowledge  a  gap 
of  some  five  or  six  hundred  years  intervened. 
These,  consisting  of  rings  and  fibulae  of 
Anglo-Saxon  workmanship  attributed  to  the 
ninth  or  tenth  centuries,  are  now  in  the  Guild- 
hall Museum,  but  are  more  interesting  for 
their  antiquity  than  for  their  beauty.  A 
more  important  example  of  early  post- Roman 
European  pewter  is  now  unhappily  only 
known  by  a  drawing,  made  in  1725  from 
an  ornate  chalice,  since  destroyed,  of  very 
uncertain  date  between  the  seventh  and 
eleventh  centuries,  which  was  reproduced 
in  the  Revue  des  Arts  Decoratifs,  published 
in  Paris  in  1883.  The  first  we  hear  of  pewter, 
after  the  Latin  authors  already  mentioned,  is 
in  connection  with  Church  use,  when,  in  1074, 
at  a  Synod  sitting  at  Rouen,  wood  was 
emphatically  forbidden  as  a  material  for 
chalices,  pewter  being  the  only  alternative 
admissible  in  those  cases  where  the  poverty 
of  the  congregation  forbade  the  use  of  the 
more  costly  metals — gold  or  silver.  An 
identical  resolution  was  adopted  for  Eng- 
land by  a  Council  held  at  Winchester  two 
years  later,  but  in  the  next  century,  at  the 
52 


PLATE  XXXIV 


W.  ,8" 


CANDELABRUM,  Flemish.     XVIII  century. 


PLATE  XXXV 


ALMS  DISH,  German      Early  XVIII  century. 


BEFORE  FIFTEENTH  CENTURY 

Council  of  Westminster,  assembled  in  1175, 
under  the  presidency  of  Richard,  Archbishop 
of  Canterbury,  pewter  in  its  turn  was 
degraded  from  its  sacred  offices,  and  the 
bishops  were  commanded  not  "  to  conse- 
crate a  pewter  chalice,"  gold  and  silver 
being  alone  considered  worthy.  A  sense  of 
economy  apparently  limited  this  restriction 
to  vessels  made  for  actual  Eucharistic  use  in 
the  services  of  the  Church,  for  the  chalices 
which  it  was  customary  to  place  in  the  coffins 
of  deceased  churchmen  to  indicate  their  rank, 
were  still,  as  a  rule,  constructed,  roughly 
enough,  of  pewter,  as  were  the  plaques 
sometimes  added,  bearing  the  name  and 
title  of  the  dead  man,  two  of  which  were 
found  at  Mont  St.  Michel,  one  recording 
the  name  of  Robert  de  Torigny,  abbot  from 
1154  to  1 1 86,  and  the  other  of  Martin,  abbot 
from  1 1 86  to  1191.  But  stern  necessity  is 
apt  to  overrule  the  decisions  even  of  Synods 
and  Councils,  and,  as  far  as  England  was 
concerned,  the  regulation  did  not  remain  long 
in  force,  for  in  1194  the  sum  of  10,000  marks 
being  called  for  from  the  nation  at  large  to 
pay  the  ransom  demanded  before  the  release 
of  Richard  Coeur  de  Lion,  the  Church  plate 

53 


OLD  PEWTER 

went  perforce  with  the  rest  of  the  kingdom's 
treasures  into  the  melting-pot,  and  pewter 
was  tacitly  permitted  to  resume  its  erstwhile 
prohibited  place  upon  the  altar. 

Whether  the  white  metal  alloy,  of  which 
the  magnificent  Gloucester  candlestick  [Plate 
IV.],  now  in  the  Victoria  and  Albert  Museum, 
can  in  all  strictness  be  classed  as  pewter,  is  per- 
haps open  to  dispute,  but  it  is  such  a  superb 
example  of  rich  twelfth-century  workmanship 
that  it  is  not  possible  to  pass  it  by  without  brief 
mention ;  nor  need  more  be  accorded  to  the 
grant  in  1201,  by  King  John,  of  charters  to  the 
Stannaries  of  Cornwall  and  Devon,  which  is 
only  incidentally  connected  with  the  subject. 

The  exclusion  of  pewter  from  ecclesiastical 
use,  which  had  so  speedily  fallen  into  abey- 
ance in  England,  was  soon  found  to  be  im- 
practicable in  many  cases  in  France  also,  and 
permission  to  employ  it  had  perforce  to  be 
accorded  to  parishes  which  could  not  afford 
the  nobler  metals  by  the  Council  of  Nimes  in 
1252  and  confirmed  by  that  of  Albi  in  1254. 
The  first  definite  mention  of  pewter  in 
domestic  use  occurs  twenty  years  later  in  a 
record  stating  that  the  meat  for  the  Corona- 
tion banquet  of  King  Edward  I.  of  England 
54 


BEFORE  FIFTEENTH  CENTURY 

(1274)  was  boiled  in  pewter  cauldrons,  and  one 
hundred  dishes,  one  hundred  platters,  and, 
curiously  enough,  more  than  one  hundred 
salt-cellars  formed  part  of  the  three  hundred 
odd  vessels  of  pewter  owned  by  the  same 
king  in  1290.  But  abroad,  at  any  rate,  the 
popularity  of  pewter  must  have  been  well 
established  long  before  that  date,  since,  in  his 
account  of  the  Paris  guilds  in  the  middle 
of  the  thirteenth  century,  Etienne  Boileau 
speaks  of  some  twenty  pewterers  already 
differentiated  into  potters,  nail-makers, 
lorimers,  toy-makers,  and  makers  of  buckles 
and  other  small  goods,  while  the  pewterers 
of  Bruges  were  becoming  noted  about  the 
same  time  for  their  porringers  and  flasks. 

The  continued  increase  of  its  employment 
in  this  way  during  the  succeeding  years  is 
further  indicated  by  a  rough  list  of  Parisian 
craftsmen,  which  shows  that  during  the  eight 
years  between  1292  and  1300  the  number  of 
makers  of  table  vessels  in  wood  had  been 
reduced  by  eighteen  and  their  places  had 
been  filled  by  eight  pewterers,  one  of  whom 
was  a  woman,  "  une  batteresse  detain."  In 
1304  the  pewterers' guild  there  was  so  con- 
firmed in  its  prosperity  that  the  Master  was 

55 


OLD  PEWTER 

thenceforth  ordered  to  pay  a  premium  to  the 
State  on  succeeding  to  office,  unless  his  father 
had  been  Master  before  him.  It  was  doubt- 
less owing  to  this  growing  demand  for  pewter 
in  foreign  lands,  and,  consequently,  for  the 
tin  which  formed  so  essential  an  ingredient 
of  it,  that  in  1305  Edward  I.  confirmed  and 
enlarged  the  Charter  to  theStanners,  relieving 
them,  among  other  provisions,  of  all  duties  ; 
freeing  them  practically  from  all  responsibility 
to  the  general  laws  except  for  capital  crimes  ; 
giving  them  courts,  judges,  and  a  prison  of 
their  own ;  and  bestowing  upon  them  the 
extraordinary  privilege  of  searching  for  and 
securing  tin  and  peat  wherever  they  chose  in 
utter  disregard  of  any  private  ownership  in 
land.  The  decoration  of  pewter,  to  a  certain 
extent,  had  already  begun  early  in  the  four- 
teenth century,  though,  as  has  been  suggested 
in  the  last  chapter,  it  was  not  apparently  as 
yet  undertaken  by  the  pewterers  themselves 
but  by  a  distinct  class  of  craftsmen,  for  Jean 
de  Jeandun,  writing  in  1323,  says  that  there 
were  many  chasers  of  gold,  silver,  pewter, 
and  bronze  on  the  Grand  Pont,  and  as  gold- 
smiths were  not  allowed  to  infringe  upon 
the  pewterers'  province,  nor  the  pewterers 
56 


^'  OX 


O 

T  ffi 


-  s 


ox 


BEFORE  FIFTEENTH  CENTURY 

on  theirs,  it  follows  that  these  impartial 
ornamenters  must  have  been  considered  in- 
dependent of  both. 

Germany  by  this  time  had  developed  a 
sufficiently  important  pewter  industry  of 
her  own  to  necessitate  the  formulating  of 
regulations  at  Augsburg,  in  1324,  for  the  visi- 
tation of  the  workshops  and  investigation 
of  the  material  used  by  properly  qualified 
officials,  armed  with  the  corresponding  powers 
to  punish  offenders,  and  in  the  same  year  the 
name  of  a  pewterer,  Carel,  is  recorded  at 
Nuremberg.  Belgium  was  also  well  to  the 
fore,  for  at  Ath,  in  1328,  the  pewterers'  guild 
was  accorded  the  first  place  in  the  civic  pro- 
cessions, while  a  list  of  pewter  belonging  to 
Clement  of  Hungary,  made  in  the  same  year, 
which  included  one  hundred  and  forty-two 
porringers  and  a  portable  b£nitier,  or  holy- 
water-holder  (see  Plates  V.  and  VI.),  shows 
that  a  recognition  of  the  usefulness  of  pewter 
was  already  widely  spread  in  Europe. 

In  T333  the  town  authorities  of  Poitiers 
found  it  advisable  to  pass  an  edict  against  the 
use  of  inferior  alloys ;  and  the  establishment 
of  the  craft  in  London  is,  for  the  first  time, 
definitely  proved  by  the  record  in  1347  of  the 

H  57 


OLD  PEWTER 

will  of  one  Nicholas  le  Peautrer,  which  is 
noted  in  the  calendar  drawn  up  for  the  Cor- 
poration of  the  City  of  London  by  Dr.  Sharpe. 
The  evidence  is  not,  however,  of  high  im- 
portance, since  the  next  year  witnessed  the 
grant  by  the  Mayor  and  Corporation,  in  re- 
sponse to  the  petition  of  the  London  pewterers, 
of  those  ordinances  regulating  the  conduct 
of  the  trade  and  appointing  a  Master  and 
Wardens  to  enforce  them,  the  chief  effects  of 
which  upon  the  craft  have  been  already  re- 
ferred to  in  the  last  chapter.  An  interesting 
sidelight  is  furthermore  thrown  by  them  upon 
the  extent  of  the  manufacture,  pots,  salers, 
or  salt-cellars,  porringers,  platters,  saucers, 
dishes,  chargers,  square  pitchers,  square  cruets, 
chrismatories,  round  pots,  round  cruets,  and 
candlesticks  being  among  the  articles  speci- 
fied, after  which  array  the  six  quart  pots  pur- 
chased by  John  of  France  in  1351  present  but 
a  humble  appearance. 

Mons,  in  Belgium,  appears  among  the 
centres  of  pewter-making  in  1353,  and  the 
craft  was  so  well  supported  at  Ghent  three 
years  later,  that  its  members  supplied  no  less 
than  thirteen  sergeants  to  the  town  militia. 
The  pewterers'  guild  at  Rouen  emerges  from 

58 


BEFORE  FIFTEENTH  CENTURY 

its  previous  obscurity  in  1369,  when  the  ex- 
change value  of  old  and  new  metal  was  fixed 
by  Royal  Letters.  An  inventory  made  in  1370 
informs  us  that  Henri  de  Poitiers,  Bishop  of 
Troyes,was  the  owner  of  fourteen  dozen  pewter 
porringers,  besides  flagons,  pots,  decanters, 
and  cimaises.  These  last  were  vessels  chiefly 
used  to  contain  the  ceremonial  wine  offered 
to  royal  personages  on  their  arrival  in  the 
neighbourhood  of  a  city  wherein  they  proposed 
to  sojourn  for  awhile.  When  the  contents 
had  been  duly  disposed  of  the  court  atten- 
dants looked  upon  the  vessels  in  which  they 
were  served  as  a  rightful  perquisite,  and  the 
thrifty  burghers,  in  consequence,  who  were 
put  in  any  case  to  quite  sufficient  expense  by 
the  honour  of  the  royal  visit,  economically 
made  them  of  pewter. 

In  1376  the  pewterers  of  Bruges  were  re- 
presented in  the  town  militia  by  eight  ser- 
geants, falling  somewhat  below  the  contingent 
supplied  by  their  brethren  of  Ghent. 

During  the  closing  years  of  this  century 
mentions  of  pewter  in  inventories,  wills,  and 
accounts  become  much  more  frequent.  Thus, 
in  1380,  Michelet  the  Breton,  a  pewterer  of 
Paris,  supplied  six  dozen  dishes  and  twelve 

59 


OLD  PEWTER 

dozen  porringers,  weighing  in  all  474^  marks, 
to  Charles  VI.,  and  in  the  same  year  a  portable 
bdnitier  is  recorded  among  the  possessions 
of  Jean  de  Halomesnil,  one  of  the  canons  of 
Sainte-Chapelle.  Two  years  later  we  find 
the  first  record  of  pewter  candelabra  and 
chandeliers  at  Soignies.  A  reason  for  the 
extreme  rarity  nowadays  of  this  early  pewter 
appears  in  1383,  when  the  before-mentioned 
Michelet  the  Breton  was  paid  24  sols  9 
deniers  of  Paris  for  recasting  twenty-four 
large  pewter  dishes,  and  the  sum  paid  also 
serves  to  show  how  slightly  the  mere  work- 
manship of  such  things  was  rewarded,  since 
the  same  dishes  originally  cost  119  marks. 
Theflabour  expended  in  remaking  them  must 
have  been  exactly  the  same  as  was  exercised 
in  the  first  instance,  and  it  therefore  follows 
that  three  deniers'  worth  of  work  went  to  each 
mark's  weight  of  metal.  A  canon  of  Troyes 
is  found  in  1386  as  the  owner  of  a  good  store 
of  pewter,  though  he  is,  properly  enough,  less 
well  provided  than  his  bishop,  boasting  of 
only  five  dozen  porringers,  with  salvers,  mugs, 
and  cups  ;  and  in  the  same  year  the  city  of 
Amiens  purchased  from  Thibaut  la  Rue, 
"  17  poz  demi-los,"  and  the  city  of  Rouen  a 
60 


PLATE  XXXIX 


1    H.  ?J"     D.  at  base  4V  2     H.  9"     D.  at  base  5J" 

CANDLESTICKS. 


3     H.  8"     D.  at  base  4' 


PLATE  XL 


1     II.  6|" 

CANDLESTICK. 
XVIII  century. 


2    83"  x  7" 

TRAY.     XVII I  century 

3     H.  4" 
INKSTAND. 


4     H.  6f" 

CANDLESTICK. 
XVIII  century. 


BEFORE  FIFTEENTH  CENTURY 

gallon  flagon  which  weighed  twenty-eight 
pounds,  a  moderately  cumbersome  prize  for 
the  official  who  claimed  it  even  after  the 
"  wine  of  honour  "  had  been  finished. 

The  primitive  simplicity  was  by  then  be- 
ginning to  give  place  to  a  more  pretentious 
style,  and  in  1389  the  Archbishop  of  Rheims 
bequeathed  in  his  will,  with  eighteen  dishes 
great  and  small,  forty-eight  porringers,  a 
square  measure,  two  square  quart  pitchers, 
and  other  vessels,  two  round  pitchers  and  two 
measures  of  three  chopins  each,  all  "fasson 
d'argent,"  while  about  the  same  time  one 
Sebaldus  Ruprecht  in  Germany  obtained 
what  must  be  regarded  as  an  equivocal  fame 
for  making  and  fashioning  pewter  which 
could  be  mistaken  for  silver. 

A  woman  pewterer,  one  Isabel  de  Moncel, 
is  first  mentioned  by  name  in  1395  as  work- 
ing at  Paris,  while  in  the  same  city  a  use  of 
pewter  not  contemplated  by  the  maker  comes 
to  light  in  1396,  when  Jean  Leboeuf  was  ac- 
cused of  striking  a  fellow  toper  with  a  wine 
measure,  thus  forestalling  by  more  than  four 
centuries  the  unnamed  individual  who  knocked 
the  late  Mr.  Bardell  "on  the  head  with  a  quart 
pot  in  a  public-house  cellar  "  and  caused 

61 


OLD  PEWTER 

him,  in  the  eloquent  language  of  Serjeant 
Buzfuz,  to  "  glide  almost  imperceptibly  from 
the  world,  to  seek  elsewhere  for  that  repose  and 
peace  which  a  custom-house  can  never  afford.'' 
During  the  fourteenth  century,  then,  we 
find  that  the  use  of  pewter  was  almost  entirely 
confined  to  the  Court,  the  nobility  arid  higher 
Church  dignitaries,  who  would  seem,  however, 
to  have  been  sufficiently  profitable  customers, 
for  the  English  pewterers  throughout  the  last 
thirty  years  of  it  were  ceaselessly  crying  out 
against  the  injury  inflicted  on  them  by  the 
tinkers  and  pedlars,  who  found  it  worth  their 
while  to  go  about  from  house  to  house  and 
town  to  town  in  the  country  recasting  damaged 
pewter.  They  founded  their  demand  for 
redress  on  the  plea  that  these  unlicensed 
workmen  adulterated  the  pewter  with  so  much 
lead  that  the  vessel  afterwards  was  "  not 
worth  the  fourth  part  sold  for,"  and  thereby 
not  only  defrauded  the  owner  but  brought  the 
craft  into  disrepute.  This  we  may  well  be- 
lieve was  perfectly  true,  but  it  is  impossible 
to  avoid  suspecting  at  the  same  time  that  the 
loss  of  business  to  themselves  arising  from 
this  illegal  competition  was  the  chief  basis  of 
their  objection. 
62 


PLATE  XLII 


1     H    2ojt"  2     H.  16" 

GERMAN  GUILD  CUPS. 


3     H.  18" 


PLATE  XLIII 


1    H.  10  J"  2     H.6|" 

For  Descriptions  see  back  of  plates. 


3    H.  ng" 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XLII. 

1.  Cup,  on  lid  1721   three  shields,  right  and  left  a  tankard,  above 

I.W.L.,  below  1713,  centre  one  a  church  with  spire. 

2.  Flagon,  on  front  IMS  1706,  Nuremberg  rose  on  bottom  inside, 

on  the  handle  a  shield,  IMK,  a  wall  and  two  turrets. 

3.  Cup,  inscribed  Johannes  George  Reichel   Johannes  Battzer 

Rellurg,  Anno  1693,  Christope  Stutz. 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  XLIII. 

1.  Flagon,  German.     XVIII  century. 

2.  Beaker,  Scotch.    XVIII  century. 

3.  Flagon,  German.    XVIII  century. 


THE  FIFTH  CHAPTER 

SOME  FACTS  ABOUT  PEWTER  IN 
THE  FIFTEENTH  CENTURY 


URING  this  century  the  pew- 
terers  continued  to  enjoy  exalted 
patronage.  In  1401  Isabella  of 
Bavaria  ordered  from  Jehan  de 
Montrousti  nine  dozen  dishes 
and  twenty-three  dozen  porringers  for  her 
own  kitchen  and  thirty  dozen  more  for  the 
Hdtel  St.  Pol.  An  inventory  made  in  the 
course  of  the  next  year  at  Rouen,  quoted  by 
Bapst,  as  far  as  its  exact  meaning  is  recog- 
nisable under  the  originality  of  the  spelling, 
serves  to  show  that  the  tools  used  by  the 
pewterer  in  those  days  were  much  the  same 
as  now,  including  the  lathe  with  its  necessary 
appurtenances,  burnishers,  scrapers,  a  file, 
various  moulds,  punches,  scales,  &c. 

Though  the  manufacture  of  pewter  was  by 
then  almost  universal  in  Northern  Europe,  it 
is  not  until  1406  that  we  discern  any  evidences 
of  its  existence  in  Spain,  but  in  that  year 
the  customary  regulations  and  statutes 
were  drawn  up  at  Barcelona.  In  1407  one 


OLD  PEWTER 

Guillebert  of  Metz  is  noted  as  a  maker  of 
remarkable  works  of  art  in  pewter,  but  no 
known  examples  of  his  handicraft  remain. 
York  would  seem  to  have  captured  at  an 
early  date  the  command  of  the  trade  in  the 
north  of  England,  and  in  1419  the  regulations 
of  her  pewterers  were  codified.  These  were 
on  the  whole  much  the  same  as  those  in  force 
in  London,  though  they  appear  to  have  aimed 
at  an  even  more  rigorous  exclusiveness,  since 
it  was  ordained  that  no  one  was  to  set  up  as 
a  master  in  the  city  who  had  not  served  his 
apprenticeship  within  its  walls. 

Yet  another  royal  order  is  recorded  for 
1422,  when  Charles  VII.  of  France  purchased 
sixty-four  dishes  and  one  hundred  and  fifty- 
eight  porringers  from  Jehan  Goupil  of  Tours, 
but  the  use  of  the  material  had  already,  in 
England  at  least,  descended  far  down  in  the 
social  scale,  for  Robert  Chichely,  Lord  Mayor 
of  London,  in  1423  ordained,  with  a  quaint 
preciseness  as  to  detail,  "that  retailers  of  ale 
should  sell  the  same  in  their  houses  in  pots 
of  pewter  sealed  and  opened,  and  that  who- 
ever carried  ale  to  the  buyer  should  hold  the 
pot  in  one  hand  and  a  cup  in  the  other;  and 
that  all  who  had  pots  unsealed  should  be 


THE  FIFTEENTH  CENTURY 

fined."  No  pronouncement,  unluckily,  is  made 
as  to  what  material  the  cup  should  be  made 
of,  so  it  must  remain  doubtful  whether  we 
have  here  the  earliest  example  of  beer  served 
"  in  its  native  pewter,"  to  quote  Mr.  Robert 
Sawyer.  There  is  almost  as  much  uncertainty 
as  to  the  exact  meaning  of  "  sealed  "  in  this 
connection  as  there  has  been  in  magisterial 
minds  of  late  years  anent  its  exact  signifi- 
cance in  the  Act  relating  to  the  sale  of  intoxi- 
cating liquors  to  infants.  Mr.  Massd  assumes 
that  it  means  stamped  by  the  maker  as  a 
guarantee  of  the  quality  of  the  pewter,  but  it 
would  seem  more  probable  that  the  seal  was 
to  certify  that  the  vessel  held  good  measure, 
a  point  which  certainly  would  more  directly 
concern  the  customer,  as  well  as  the  inn- 
keeper, than  the  nature  of  the  alloy.  The 
spread  of  pewter  among  the  less  exalted  is 
further  illustrated  in  1427  by  the  will  of  John 
Ely,  vicar  of  Ripon,  who  left  half  a  garnish 
and  two  chargers.  The  regulations  as  to  the 
lawful  weights  of  various  articles  made  in 
1430  have  already  been  referred  to,  as  have 
the  two  forms  of  the  alloy  authorised  by  the 
Montpelier  pewterers  in  1437. 

The    London    pewterers  were    evidently 


OLD  PEWTER 

waxing  proud  in  their  growing  prosperity  by 
this  time,  for  in  1438  they  presumed  to  add  to 
their  ordinances  without  consulting  the  Lord 
Mayor,  who  promptly  asserted  his  authority 
and  showed  his  resentment  by  annulling  the 
additions  until  he  had  been  petitioned  with 
due  humility  to  allow  them.  How  great  this 
prosperity  was  is  shown  by  the  fact  that  in 
1444  the  Warden  of  the  Company  thought  it 
advisable  to  ensure  a  sufficient  supply  of  that 
essential  ingredient,  tin,  by  obtaining  the 
right  to  pre-empt  one-quarter  of  all  that  came 
into  the  City.  An  inventory  of  the  same  year 
shows  that  John  Danby  of  Alveston,  a  mere 
commoner,  possessed  "  ix.  pece  led  and  pewd 
vessell,"  worth  two  shillings  and  fourpence, 
while  in  1453  Jacques  Coeur  laid  in  a  supply 
of  pewter  for  his  workpeople.  As  is  so  often 
the  case  with  fashions,  however,  when  the 
common  people  began  to  enjoy  it  the  upper 
classes  thought  it  time  to  eschew  it,  and  in  the 
inventory  of  Sir  John  Fastolfe's  possessions 
in  1459  among  nineteen  thousand  ounces  of 
plate  not  a  grain  of  pewter  is  to  be  found. 
Yet  the  noble  and  royal  visitors  to  subject 
cities  had  still  to  be  content  with  supping 
their  vin  d'honneur  from  humble  pewter,  at 
66 


THE  FIFTEENTH  CENTURY 

any  rate  at  Amiens,  where  cups  for  the  pur- 
pose were  again  purchased  in  1463.  The 
partially  religious  nature  of  the  London  guild 
is  revealed  by  the  mention  of  a  gift  made  in 
1465,  in  which  it  is  described  as  "the 
brotherhood  of  Our  Lady  of  the  Assumption 
of  the  pewterer's  craft,"  and  the  continued 
use  of  their  products  for  church  purposes 
by  a  record  made  in  1467  of  the  existence 
at  "  St.  Stephne's  in  Colman  Streete,"  of 
"3  pair  of  cruets,  22  dishes  for  the  sepulcur, 
2  for  the  pascal,  and  i  on  a  stokke  before 
Synt  John  in  the  church."  In  the  same 
year  the  craftsmen  of  Mons  began  to  mark 
their  wares  with  a  crowned  hammer  and  the 
word  "  fin,"  unless  it  were  spun,  in  which 
case  a  castle  and  the  town  arms  were  used, 
while  pewter  imported  from  England,  a 
common  practice  abroad  at  the  time,  was 
stamped  with  a  crowned  rose.  We  find  in 
1470  that  even  the  rival  Goldsmiths'  Com- 
pany was  investing  in  pewter,  one  pound 
seventeen  shillings  and  sixpence  being  paid 
that  year  "  for  a  garnish  of  2  dozen  pew- 
ter vessels  to  serve  the  company,"  in  spite 
of  the  fact  that  the  pewterers  were  then 
imitating  the  designs  of  the  goldsmiths,  and 

67 


OLD  PEWTER 

in  France  were  buying  from  them  drawings, 
and  patterns  in  pewter  and  clay.     About  the 
same  time  Buschius  of  Hildesheim  undertook 
a  progress  for  the  purpose  of  investigating 
the  state  of  the  Saxon  convents,  and  in  his 
account  pewter  bulks  largely.     The  convent 
at   St.  Cyr  owned   two   hundred   amphorae, 
flagons  and  tankards,  which  does  not  point 
to  any  excessive  austerity ;  that  of  the  Holy 
Cross   of    Erfurth    one    hundred    and    fifty 
amphorae,   seventy   cups,   twelve   jugs,    and 
thirty-three  porringers  ;  the  Cistercians  of  St. 
Martin  had  also  one  hundred  and  fifty  am- 
phorae, flagons  and  porringers  ;  but  the  White 
Ladies  of  Erfurth,  either  owing  to  poverty  or 
a  stricter  rule,  were  satisfied  with  only  forty- 
one  amphorae,  porringers,  and  four  flagons. 
The  year  1473  was  a  notable  one  for  pewterers 
in   England,  for   in    the   course   of  it    King 
Edward  IV.  testified  his  royal  approval   of 
their  labours  by  conferring  on  their  Company 
a   formal  Charter  confirming  the  privileges 
they  enjoyed  and  the  powers   they  had   pre- 
sumably exercised   so   satisfactorily,  and   in 
1478  the  Duke  of  Burgundy  followed  suit  by 
establishing  guilds  in  many  of  the  cities  in 
his  dominions.     A  clear  idea  of  the  immense 
68 


THE  FIFTEENTH  CENTURY 

proportions  the  trade  had  assumed  may  be 
gathered  from  the  fact  that  in  1481  one 
record  of  the  metal  employed  for  organ  pipes 
alone,  a  blend  of  six  parts  of  tin  with  four  of 
lead,  accounts  for  fourteen  thousand  five 
hundred  pounds.  "A  hoole  garnish  of 
peautre  vessel,  two  round  basin  of  peautre  " 
are  found  in  the  will  of  Elizabeth  Lady 
Uvedale  in  1487,  and  pewter  bottles  in  the 
11  Livre  des  Mestiers  "  of  Charles  V.,  while 
from  an  inventory  of  the  goods  of  the 
Pewterers'  Guild  of  London  made  two  years 
later  we  learn  that  their  common  seal  bore 
"the  ymage  of  thassumpcon  of  our  blessyd 
lady  gravy n  theryn  of  silver."  The  fact  that 
the  Company  guaranteed  by  a  stamp  the 
quality  of  pewter  is  confirmed  by  the  entry  in 
1492  of  the  purchase  of  four  new  "  marking 
irons  for  Holoweware  men." 

Scotland,  according  to  Mr.  Ingleby  Wood, 
lagged  far  behind  her  southern  sister  in  the 
pewterer's  craft,  and  this  he  suggests  was  due 
in  part  to  the  greater  general  poverty,  in  part 
to  the  absence  of  native  tin,  which  must  have 
added  very  seriously  to  the  original  cost  01 
the  raw  material,  and  in  part  to  the  facility 
with  which  the  ware  could  be  smuggled  into 


OLD  PEWTER 

the  country,  in  defiance  of  all  statutes,  from 
the  Dutch,  Flemish,  and  French  ports,  with 
which  so  large  a  general  trade  was  carried  on. 
Be  the  causes  what  they  may,  it  is  not  until 
the  very  end  of  the  century,  in  1496,  that  we 
get  the  first  actual  record  of  pewter-making  in 
Scotland,  when  a  second  "  Seal  of  Cause"  or 
Charter  of  Incorporation  was  granted  by  the 
Provost,  Magistrates,  and  Town  Council  of 
Edinburgh  to  the  Hammermen  of  the  City,  in 
whose  ranks,  for  the  first  time,  the  peudrars, 
together  with  the  coppersmiths,  appear,  the 
first  Seal  of  Cause  issued  in  1483  naming 
only  blacksmiths,  goldsmiths,  saddlers, 
cutlers,  and  armourers.  We  learn  from  this 
how  greatly  the  arrangements  in  Scotland 
differed  from  those  in  England,  where  each  of 
these  crafts  had  a  distinct  guild  of  its  own. 
In  Scotland,  on  the  contrary,  all  the  tradesmen 
whose  work  was  in  the  main  executed  with  a 
hammer,  except  the  carpenters,  masons,  and 
usually  plumbers,  were  enrolled  in  one  com- 
prehensive Corporation,  which,  though  differ- 
ing in  its  exact  composition  at  various  times 
and  in  different  places,  generally  extended  to 
glovers,  lorimers,  buckle-makers,  sword- 
cutlers,  gunsmiths,  potters  and  braziers  at 
70 


THE  FIFTEENTH  CENTURY 

Perth,  and  sword-guard  makers  at  Dundee  ; 
skinners,  glaziers,  wrights  and  potters  at 
Aberdeen ;  clock-makers,  bell-makers  and 
plumbers  at  Glasgow;  and  later  on  white- 
ironsmiths,  or  tinsmiths  as  we  should  say 
nowadays,  watch-makers,  carriage-makers, 
bell-hangers,  and  at  St.  Andrews  even  such 
alien  trades  as  dyers,  painters,  and  stationers. 
The  main  features  of  the  development  of 
pewter-ware  during  the  fifteenth  century  were 
the  increasing  range  downwards  in  the  social 
scale  of  its  domestic  use,  the  consequent 
growth  in  numbers  of  the  craftsmen  working 
in  it — Bapst  has  collected  the  names  of  one 
hundred  and  sixteen  workers  during  that 
century  in  France  alone — and  the  beginning 
of  the  custom  among  the  wealthier  classes  of 
having  their  pewter  so  lavishly  adorned  that 
the  cost  became  prohibitive  to  the  lower 
orders,  thus  ensuring  to  themselves  for  a 
while  longer  freedom  from  the  competition 
of  the  vulgar. 


THE  SIXTH  CHAPTER 

SOME  FACTS  ABOUT  PEWTER  IN 
THE  SIXTEENTH  CENTURY 

N  England  at  the  beginning  of 
the  sixteenth  century,  pewter, 
according  to  the  Northumber- 
land household  book,  was  still 
considered  too  expensive  to  be 
common,  but  the  trade  had  nevertheless 
grown  to  be  so  important,  and  the  tempta- 
tions to  fraudulent  practices  on  the  part  of  its 
followers  so  strong  and  apparently  so  fre- 
quently yielded  to,  that  in  1503  an  Act  of 
Parliament  (19  Henry  VII.  c.  6)  was  passed 
to  compel  every  maker  to  mark  his  ware 
with  a  recognisable  touch  of  his  own,  to 
forbid  him  to  sell  outside  his  own  business 
premises  except  in  open  fair  and  market,  and 
to  put  down  the  use  of  false  weights  and 
scales.  The  legal  mind  about  the  same  time 
seems  to  have  realised  that  the  increasing 
independence  of  the  great  City  Companies,  if 
not  dangerous  to  the  Crown,  was  decidedly 
inimical  to  its  own  financial  interests,  and 
that  certain  customary  regulations  adopted 
72 


THE  SIXTEENTH  CENTURY 

by  them  had  the  evil  effect  of  preventing 
many  fat  pickings  from  reaching  the  coffers 
of  the  law.  The  remedy  was  attempted  in 
1504  by  an  Act  providing  that  no  ordinance 
adopted  by  any  one  of  them  should  be  valid 
and  binding  until  it  had  obtained  the  formal 
approval  of  the  Chancellor  or  some  other 
court  official,  which  doubtless  meant  fees, 
and  at  the  same  time  abrogating  the  rule 
which  forbade  brother-members  of  a  guild 
from  going  to  law  with  one  another  in  the 
courts,  but  compelled  them  to  bring  all  dis- 
putes for  arbitration  by  the  wardens  and 
councils,  which  certainly  meant  fees. 

The  "  silver  fashion  "  continued  to  wax  in 
favour  with  the  wealthy  and  the  merely  useful 
to  decline,  if  we  may  judge  by  the  fact  that 
in  1507  the  Duke  of  Burgundy  owned  three 
pitchers  and  three  ewers  of  a  decorative  cha- 
racter, and  only  thirty-two  plates,  the  same 
number  of  porringers  and  a  mustard  for 
practical  purposes.  At  the  same  time  the 
Duke  of  Bpurbonnays  had  three  quart  flagons, 
three  pitchers,  and  three  ewers,  and  in  the 
following  year  the  city  of  Amiens  ordered 
from  Pierre  Hemeron  four  small  ewers  of 
fine  pewter  "  fasson  d'argent."  The  English 

K  73 


OLD  PEWTER 

Parliament  again  took  cognisance  of  the  craft 
in  1512,  in  response  to  a  complaint  that  "  evil- 
disposed  persons"  went  about  the  country 
buying  pewter  and  brass,  which  was  generally 
stolen,  and  afterwards  sold  it  clandestinely  to 
"  strangers  "  who  carried  it  overseas  ;  that  the 
same  persons  were  in  the  habit  of  recasting 
old  vessels,  adulterating  the  material  in  the 
process,  and,  furthermore,  used  beams  and 
scales  so  "  deceivable  and  false  "  "  that  one  of 
them  will  stand  even  with  12  Ib.  weight  at 
the  one  end  against  a  quarter  of  a  Ib.  at  the 
other,"  which  certainly  seems  carrying  fraud 
to  excess ;  and  the  penalty  of  losing  the  beam 
and  paying  a  fine,  of  twenty  shillings,  or,  in 
default,  of  remaining  in  the  stocks  until  the 
next  market  and  elevated  on  the  pillory  as 
long  as  that  lasted,  does  not  sound  unduly 
severe  in  comparison  with  the  fine  of  ten 
pounds  incurred  by  selling  pewter  otherwise 
than  previously  ordained.  The  hospitality 
of  Amiens  would  seem  to  have  been  inex- 
haustible, for  in  1516  it  is  again  found  buying 
thirty-five  small  pewter  mugs  from  Jeanne 
d'Avesne  for  the  purpose  of  presenting  the 
wine  of  loyal  greeting  to  Francois  I.  and 
Louise  of  Savoy.  1518  brings  the  first  inti- 
74 


PLATE  L 


1  2  3     H.  si"  4  5 

CREAM-JUG  MEASURE  CREAM-JUG 


SALT-CELLAR 
English.     XVIII  century. 


EGG-CUP 


PLATE  LI 


A  PAIR  OF  TANKARDS,  German.     XVIII  century. 


THE  SIXTEENTH  CENTURY 

mation  of  the  Incorporation  of  Hammermen 
at  Perth,  but  there  is  good  reason  for  believ- 
ing that  the  book  of  records  bearing  this  date 
had  predecessors  long  since  destroyed,  and  in 
the  same  year  the  town  authorities  of  Edin- 
burgh drew  up  regulations  concerning  the 
stamping  of  the  vessels  used  by  taverners, 
in  order  to  ensure  their  containing  good 
measure,  a  plug  or  "  talpoun  "  on  the  inside 
of  the  neck  indicating  the  height  to  which 
each  was  to  be  filled.  In  1535  we  first  hear  of 
the  Hammermen  of  the  burgh  of  the  Cannon- 
gate,  who  had  an  Incorporation  of  their  own 
distinct  from  that  of  Edinburgh,  and  that  year 
appealed  for  leave  to  build  an  altar  in  the 
abbey  church  and  dedicate  it  to  St.  Eloy, 
always  in  Scotland  the  patron  saint  of  ham- 
mermen. On  the  face  of  it  the  offer  would 
seem  all  to  the  good  of  the  church,  yet  for 
some  reason  it  was  not  accepted  by  the  abbot 
until  eleven  years  later. 

The  English  makers  about  this  time  began 
apparently  to  be  troubled  by  foreign  compe- 
tition. Whether  any  of  the  ware  of  Martin 
Harscher,  a  German  pewterer  who  died  in 
1523  at  the  age  of  eighty-three,  which  was 
proudly  asserted  to  be  superior  in  quality 

75 


OLD  PEWTER 

to  English  pewter,  found  its  way  over  here  is 
doubtful,  but  it  was  considered  desirable,  in 
days  when  Members  of  Parliament  had  no 
Free  Trade  bogey  to  fright  them,  by  an  Act 
passed  in  1538  to  forbid  all  importation 
of  foreign  pewter,  to  deny  to  foreigners  the 
right  of  working  in  England,  and  even  to 
endeavour  to  prevent  Englishmen  from 
working  abroad. 

The  Incorporation  of  Hammermen  at  Glas- 
gow is  first  mentioned  in  1536,  but  pewterers 
are  not  particularly  specified,  probably,  as  Mr. 
Wood  suggests,  because  they  were  too  few  to 
make  it  worth  while,  being  merely  included  in 
the  comprehensive  "  others  within  the  burgh." 
The  dissolution  of  the  monasteries  in  1537, 
or,  rather,  the  inventories  of  their  contents 
which  that  high-handed  proceeding  gave  rise 
to,  might  well  have  been  expected  to  throw 
a  flood  of  light  upon  the  variety  of  domestic 
utensils  in  pewter  employed  at  the  time  ;  but 
as  a  matter  of  fact,  whether  the  Commis- 
sioners did  not  consider  it  as  worthy  of 
notice,  or  whether  the  monks  were  too 
wealthy  to  condescend  to  such  base  material, 
very  little  of  it  appears  in  any  of  the  records, 
and  the  chief  fact  we  learn  is  that  what  we  now 


PLATE  LI  I 


2    H 


3    H.  2ij" 
GERMAN  GUILD  TANKARDS 


THE  SIXTEENTH  CENTURY 

call  porringers  were  known  as  "  counterfettes 
or  podingers." 

The  earliest  ordinances  of  the  Hammermen 
of  St.  Andrews  bear  date  1539,  but  in  this 
case  also  it  is  supposed  that  older  ones  ex- 
isted but  have  been  lost.  In  the  same  year 
King  James  V.  persuaded  a  number  of  foreign 
craftsmen  from  Holland,  France,  and  Spain 
to  settle  in  Scotland,  in  order,  doubtless,  that 
his  subjects  might  learn  from  them,  and  it 
is  at  least  probable  that  some  pewterers  were 
among  them  ;  and  in  the  next  year  the  same 
enlightened  monarch  conferred  upon  the 
Hammermen  of  the  Cannongate  a  ratification 
of  their  "  Seal  of  Cause,"  which  confirmed, 
and  to  some  degree  extended,  their  privileges. 

At  the  same  time  in  England  the  pewterers 
of  York  were  deciding  that  every  pewterer 
was  to  stamp  his  handiwork  with  his  own 
mark,  a  "  counterpayne  "  of  which  was  to  be 
kept  in  the  "common  chambre  "  for  purposes 
of  identification,  in  which  course  they  were 
evidently  following  in  the  footsteps  of  their 
brethren  in  London,  who  already,  according 
to  an  inventory  of  the  date,  had  "  a  table  of 
pewter,  with  every  man's  mark  therein,"  which 
has  unfortunately  long  since  disappeared. 

77 


OLD  PEWTER 

The  various  corporations  obviously  found 
it  easier  to  lay  down  rules  and  regulations 
than  to  enforce  them,  for  again  and  again  we 
find  one  or  another  re-affirming  laws  which 
were  supposed  to  have  been  long  in  force. 
Thus,  in  1543,  the  Hammermen  of  St.  An- 
drews found  it  necessary  to  repeat  an  Act 
governing  the  admission  of  none  but  those 
whose  competence  in  their  particular  craft 
had  been  thoroughly  established,  not  only  to 
the  satisfaction  of  the  selected  officers  belong- 
ing to  it,  but  of  the  Provost  and  Magistrates 
of  the  city,  and  in  the  course  of  the  same 
year  the  Edinburgh  authorities  were  obliged 
to  renew  and  amplify  their  Act  relating  to 
taverners'  measures.  Two  years  later  the 
Edinburgh  Trade  Incorporations  became 
owners  of  the  Chapel  and  Hospital  of  St. 
Mary  Magdalen,  and  the  Charter  of  Dedica- 
tion was  drawn  up  in  accordance  with  the 
wishes  of  one  Janet  Rhynd,  to  whose  generous 
gift  of  two  thousand  pounds  Scots,  or  one 
hundred  and  sixty-six  pounds  sterling,  the 
completion  of  the  building  was  due. 

Meantime  in  France  the  manufacture  of 
what  would  nowadays  be  called  "art"  pewter 
was  progressing  so  rapidly  that  the  gold  and 

78 


H.  6i" 

MEASURE  or  TANK- 
ARD, German.  Late 
XVII  century. 


1     H.  5"  2     H.  6.1"  3     H. 

SCOTCH  MEASURES.     XVIII  century. 


PLATE  LV 


PLATE  LVI 


1     H.4>"  2     H.  3i"  3     H.3l"  4    H.  3J" 


THE  SIXTEENTH  CENTURY 

silversmiths  of  Paris  began  to  take  fright  at 
the  encroachments  of  their  rivals  upon  their 
especial  domains  which  had  clearly  been  initi- 
ated ;  and  in  1545  they  appealed  for  an  enact- 
ment forbidding  the  pewterers  to  work  in  any 
other  metal,  while  they  on  their  part,  in 
order  to  balance  things  and  forestall  a  griev- 
ance on  their  rivals'  side,  undertook  never  to 
work  in  pewter. 

The  first  glimpse  of  pewterers  in  Perth  is 
gained  by  an  odd  side-light  in  1546,  when 
the  hammermen  forbade  the  apprentices, 
among  them  the  "  pewteraris,"  to  play  foot- 
ball on  the  historical  Inch,  whereon  Hal  of 
the  Wynd  fought  his  famous  fight.  1550  is 
famous  in  the  annals  of  purely  decorative 
pewter  as  the  probable  birth-year  of  Francois 
Briot,  in  whose  hands  it  reached  its  highest 
culmination.  Though  over-elaborated  and 
ill  suited  to  the  material,  there  is  no  denying 
to  his  masterpieces  a  wonderful  fertility  of 
invention  and  consummate  finish  of  work- 
manship. His  finest  work  is  undoubtedly 
the  salver  with  a  nude  figure  of  Temperance, 
seated  and  holding  a  cup,  in  the  centre,  alle- 
gorical representations  of  the  elements  sur- 
rounding her,  and  the  sciences  on  the  rim, 

79 


OLD  PEWTER 

the  whole  accompanied  and  embellished  by  a 
variety  of  Renaissance  ornamentation.  Speci- 
mens of  this  may  be  found  both  in  the  British 
Museum  and  at  South  Kensington. 

The  beginnings  of  a  revolution  which  was 
destined  in  after  years  to  cause  the  London 
pewterers  much  vexation  of  spirit,  is  indicated 
in  1552  by  the  rules  promulgated  compelling 
the  makers  of  the  pewter  lids  attached  to 
stoneware  vessels  to  bring  their  whole  weekly 
output  every  "  Satterdaye "  to  the  Hall  in 
order  that,  if  they  were  judged  sufficiently 
well  and  truly  wrought,  the  lids  might  be 
stamped  outside  with  the  mark  of  the  Hall 
as  well  as  with  the  maker's  own  personal 
mark.  Traces  of  the  Reformation  which 
had  devastated  church  and  church  plate 
seventeen  years  before  probably  linger  in 
the  record  on  the  register  of  Waltham  Abbey 
Church  of  the  purchase,  in  1554,  of  a  chris- 
matory  and  pix  in  pewter,  the  former  costing 
three  shillings  and  sixpence,  the  latter  two 
shillings.  The  order  against  buying  by  night 
from  unauthorised  or  suspicious  characters, 
which  has  been  already  referred  to,  was  passed 

in  1555- 

The  high  respect  with  which  pewter  was 
80 


THE  SIXTEENTH  CENTURY 

regarded  in  Scotland  is  shown  by  an  entry 
in  the  town  records  of  Edinburgh,  dated 
June  27,  1559.  In  consequence  of  threatened 
disturbances  the  treasures  of  the  Church  of 
St.  Giles  were  to  be  removed  and  distributed 
for  safety  amongst  various  presumably  trust- 
worthy members  of  the  congregation,  who  did 
not,  however,  apparently  justify  in  all  cases 
the  confidence  so  rashly  reposed  in  them, 
since  even  "  Johne  Charterhous,  elder,  Dean 
of  the  Guild,"  in  spite  of  his  responsible 
position,  never,  as  far  as  we  know,  could  be 
persuaded  to  restore  the  "  twa  candelstyks  of 
tin  "  which  had  been  deemed  worthy  to  adorn 
the  "hie  altar"  until  untoward  circumstances 
flung  them  into  his  too  conscientious  custody. 
Another  of  many  quaint  and  interesting  re- 
cords which  Mr.  Wood  has  unearthed  for  us 
is  found  at  Perth  next  year,  where,  among 
the  duties  prescribed  for  an  apprentice  who 
wished  to  qualify  as  master,  we  discover  the 
obligatory  presentation  of  a  football.  The 
same  year  brought  about  the  downfall  of  the 
Roman  Catholic  Church  in  Scotland  and  the 
establishment  of  Presbyterianism  on  its  ruins, 
and  in  the  process,  what  with  the  vessels 
which  were  carried  abroad  by  the  departing 

L  81 


OLD  PEWTER 

faithful,  those  which  came  into  the  care  of 
such  guardians  as  Johne  Charterhous,  those 
like  the  possessions  of  St.  Giles's  which  were 
sold  in  1561  for  the  public  benefit,  and  those 
that  fell  victims  to  the  zeal  of  the  Reformers, 
the  pre-Reformation  church  plate  without  a 
known  exception  disappeared. 

In  1564  the  question  of  apprentices  was 
finally  settled  for  England  on  a  definite  basis. 
In  accordance  with  a  statute  passed  in  the 
previous  year,  it  was  ordained  that  every 
freeman  of  the  Company  was  entitled  to  take 
one  apprentice,  while  the  Master  and  Wardens 
were  allowed  three,  provided  that  at  the  same 
time  they  employed  not  less  than  two  journey- 
men. They  did  not,  it  would  seem,  approve 
of  the  decorative  debauch  in  progress  among 
their  brethren  across  the  Channel,  supporting 
whole-heartedly  the  rigour  of  the  game,  for 
while  a  Nuremberg  pewterer,  Melchior  Koch, 
who  died  in  1567,  was  gaining  fame  for  a 
process,  the  secret  of  which  he  carried  with 
him  to  the  grave,  of  making  an  alloy  which 
closely  resembled  gold,  the  Englishmen  pro- 
hibited absolutely  the  application  of  gilding 
or  painting  to  the  surface  of  vessels  with  the 
exception,  allowed  by  an  Act  of  1564,  of  small 
82 


THE  SIXTEENTH  CENTURY 

articles  which  had  been  made  solely  to  be 
given  away  as  presents  by  the  maker,  a 
curious  illustration  of  the  minute  details 
with  which  the  makers  of  the  ordinances 
concerned  themselves. 

The  need  of  some  distinguishing  mark  to 
indicate  the  quality  was  first  officially  recog- 
nised in  Scotland  in  1567,  when  an  Act  of 
King  James  VI.  provided  that  "fyne  tyn 
pewder"  was  to  be  marked  with  a  crown  and 
hammer,  and  the  second  quality  with  the 
maker's  name,  with  the  usual  penalties  of 
fine  and  confiscation  if  the  stamps  were 
absent  or  undecipherable.  Before  1570  Dun- 
dee had  followed  Edinburgh  in  her  resolve  to 
ensure  for  her  citizens  good  measure  from  the 
taverners,  and  "ane  iron  stamp  to  mark  ye 
tinn  stoopis  "  was  in  the  charge  of  the  Dean 
of  Guilds  there  that  year. 

In  England  and  on  the  continent  pewter- 
ware  had  by  the  middle  of  the  century  be- 
come almost  an  inevitable  part  of  the  plenishing 
of  every  respectable  family,  so  that  in  1572  a 
simple  draper  of  Paris  could  bequeath  to  his 
heir  ''six  plates,  two  eared  and  two  deep  and 
four  shallow  porringers,  three  large  dishes, 
three  sauce-boats,  a  mustard,  a  salt,  a  couple 

83 


OLD  PEWTER 

of  basins,  a  water-pitcher,  and  a  pint  pot." 
The  pewterers,  in  fact,  were  beginning  to 
boast,  especially  in  Germany,  that  they  could 
and  did  make  everything  in  their  material 
that  the  gold  or  silversmith  could  produce, 
though  they  did  not  at  that  time  confine 
themselves  strictly  to  pewter,  for  the  pewterer 
in  a  "  Treatise  on  Industries,"  by  Harman 
Schoper,  written  in  1573,  is  made  to  declare: 
"  I  make  vases  of  all  sorts  of  molten  metals." 
It  was,  therefore,  scarcely  to  be  wondered  at 
that  in  1579,  some  candlesticks  made  in 
pewter  by  Peter  Schmitt  having  attracted 
too  much  approval,  the  goldsmiths  of  Nurem- 
berg secured  the  adoption  of  a  self-denying 
ordinance,  similar  to  the  earlier  one  agreed 
to  in  Paris,  consenting  to  forego  the  right  of 
themselves  working  in  pewter,  on  condition 
that  the  pewterers  did  not  intrude  upon  their 
speciality. 

The  first  "  Seal  of  Cause"  was  granted  to 
the  Hammermen  of  Aberdeen  in  1579,  but 
from  references  in  the  document  it  is  clear 
that  they  must  have  existed  as  a  properly 
organised  body  before  the  Reformation  of 
1560,  since  it  is  remarked  that  they  still 
owned  and  worshipped  at  an  altar  to  St.  Eloi, 
84 


PLATE 
LX 


H.  iif 

TAPPIT  HEN.    XVIII  century. 


THE  SIXTEENTH  CENTURY 

and,  strangely  enough,  still  lighted  it  with 
candles  made,  as  was  very  general  in  Scotland, 
from  wax  procured  in  the  form  of  fines  from 
offenders  against  their  ordinances.  For  many 
years,  wherever  incorporations  of  the  various 
crafts  existed  in  Scotland  there  had  been 
constant  friction  between  the  deacons  and 
other  officers  elected  by  the  members  and  the 
town  magistrates,  who  represented  for  the 
most  part  the  wealthier  merchants,  and,  jealous 
doubtless  of  the  increasing  power  and  unity 
of  the  craftsmen,  endeavoured  to  interfere 
more  closely  and  frequently  in  their  affairs 
than  the  victims  deemed  desirable,  while  the 
magistrates  on  their  part  were  constantly 
complaining  that  the  deacons  were  always 
striving  to  exceed  their  lawful  powers.  The 
question  was  grappled  with,  and  in  the  end 
set  at  rest  by  an  Act  of  James  VI.,  passed  in 
1581,  which  by  granting  a  Charter  to  the 
crafts  raised  them  to  the  same  level  as  the 
merchants,  and  though  these  last  can  scarcely 
have  been  well  pleased  at  their  practical  defeat, 
peace  seems  to  have  reigned  thereafter.  Dating 
from  1587  the  first  record  book  existing  of  the 
Dundee  Incorporation  was,  as  usual,  evidently 
not  the  earliest,  but  the  history  of  this  body 

85 


OLD  PEWTER 

is  not  of  great  importance  to  us,  since  one 
Master  Gray  is  the  only  "  pewderer  "  belong- 
ing to  it. 

The  same  year  in  England  saw  the  publica- 
tion of  Harrison's  "  Description  of  England  in 
Shakespeare's  time,"  which  gives  many  in- 
teresting pictures  of  the  life  of  the  period,  not 
omitting  references  to  pewter,  which  he  notes 
is  becoming  more  frequent.  "  For  so  common 
were  all  sorts  of  treene  stuffe  in  olde  time, 
that  a  man  should  hardly  find  four  pieces  of 
pewter  (of  which  one  was  peradventure  a  salt) 
in  a  good  farmer's  house.  But  now  a  farmer 
will  think  his  gains  very  small  if  towards  the 
end  of  his  term  he  cannot  have  a  fair  garnish 
of  pewter  on  his  cupboard,  a  bowl  for  wine 
(if  not  a  whole  neast),  and  a  dozen  of  spoons 
to  furnish  up  the  suit."  As  for  the  upper 
classes,  he  says  :  "  Likewise  in  the  houses  of 
knights,  gentlemen,  merchantmen,  and  some 
other  wealthy  citizens,  it  is  not  geson  (un- 
usual) to  behold  generallie  their  great  pro- 
vision of  tapestrie,  Turkic  work,  pewter, 
brass,  and  fine  linen,  and  thereto  costlie  cup- 
boards of  plate  with  five  or  six  hundred  or  a 
thousand  pounds  to  be  deemed  by  estimation," 
a  passage  which  strikingly  recalls  Shakes- 
86 


X 


w 


THE  SIXTEENTH  CENTURY 

peare's  own  description  of  a  well-provided 
house  which  he  put  into  the  mouth  of  Gremio, 
in  the  Taming  of  the  Shrew,  act  ii.,  scene  i. 

"  First,  as  you  know,  my  house  within  the  city 
Is  richly  furnished  with  plate  and  gold  ; 
Basins  and  ewers  to  lave  her  dainty  hands  ; 
My  hangings  all  of  Tyrian  tapestry  ; 
In  ivory  coffers  I  have  stuffed  my  crowns  ; 
In  cypress  chests  my  arras  counterpoints, 
Costly  apparel,  tents  and  canopies, 
Fine  linen,  Turkey  cushions  boss'd  with  pearl, 
Valance  of  Venice  gold  in  needlework, 
Pewter  and  brass  and  all  things  that  belong 
To  house  or  housekeeping." 

What  the  pewter  probably  consisted  of  is 
made  clear  by  an  inventory  made  in  1594  of 
the  contents  of  a  house  at  Gillingham,  occu- 
pied by  Sir  William  Fairfax,  as  reproduced 
in  Archczologia,  vol.  xlviii.  In  the  "wine- 
seller  one  quart  pewter  pott,"  in  the  pantry 
two  basins  and  ewers  of  pewter  valued  at 
fourteen  shillings  and  four  pence,  and  two 
pewter  trays  valued  at  ten  shillings  ;  in  the 
"  kytchine,"  twenty-four  saucers,  twenty-four 
dishes  great  and  small,  twenty-four  platters 
great  and  small,  four  chargers,  and  twelve 
more  dishes.  There  were  also  twelve  saucers, 
twelve  "  sallite  dishes,"  twenty-four  great 


OLD  PEWTER 

dishes,  eighteen  great  platters,  and  "  i  charger 
of  the  greatest  sort,"  which  were  all  new. 
The  value  of  the  whole  collection  being  eigh- 
teen pounds,  six  shillings  and  eightpence. 
Perhaps  the  shortcomings  of  some  of  these 
very  saucers  may  have  led  the  Company  in 
the  following  year  to  decide  that  never  more 
should  saucers  be  wrongfully  finished  by 
turning  on  a  lathe,  but  duly  and  properly 
hammered  "uppon  payne  of  forfayture." 

That  the  French  merchants  did  not  fall 
behind  the  standard  of  luxury  attained  by 
their  English  neighbours  is  shown  by  the 
inventory  of  one  Pierre  de  Capdeville  of 
Bordeaux  in  1591,  which,  in  spite  of  some 
uncertainty  as  to  the  exact  meaning  of  some 
of  the  distinctive  descriptions,  gives  a  good 
idea  of  the  state  of  bourgeois  comfort  at  the 
time.  In  addition  to  a  variety  of  jugs  and 
measures  used  in  the  business  of  selling  wine, 
he  possessed  a  ewer,  two  flagons,  two  ollieres, 
which  may  have  been  cruets  or  oil-jars,  six 
great  dishes  "  du  grande  molle,"  thirteen 
dishes  "  du  deuxieme  molle,"  one  "  du  tiers 
molle,"  eight  "du  petit  molle"  (the  terms 
referring  perhaps  to  the  size  of  the  moulds  in 
which  they  were  cast),  thirty-six  round  plates, 
88 


PLATE  LXII 


JUG,  English.    XVII  century. 


THE  SIXTEENTH  CENTURY 

fourteen  eared  porringers,  seven  sauce-boats, 
two  large  "  gardelles,"  and  one  hundred 
pounds  weight  of  old  pewter,  while  he  was, 
furthermore,  so  far  in  the  fashion  as  to  own  a 
basin  and  ewer  or  cistern  of  ornamental  ware. 
Imperfect  as  are  the  records  of  the  In- 
corporations of  Hammermen  in  most  Scot- 
tish cities,  those  of  Stirling  are  more  than 
usually  so,  only  two  which  in  any  way  con- 
cern us,  covering  the  period  from  1596  to 
1620,  remaining ;  and  though,  like  all  these 
ordinances,  they  contain  much  unconscious 
humour,  and  throw  considerable  light  on  the 
life  of  the  time,  they  do  not  to  any  appreci- 
able extent  advance  our  knowledge  of  the 
state  of  the  pewterer's  craft  in  the  burgh. 
The  most  notable  fact  revealed  by  them  is 
that  between  1599  and  1620  there  were  four 
men  practising  the  trade  there,  a  surprising 
number  considering  the  size  of  the  place, 
surpassing,  as  Mr.  Wood  points  out,  Aber- 
deen, Dundee,  then  the  second  town  in  Scot- 
land, St.  Andrews  and  Perth,  in  the  minute- 
book  of  whose  Incorporation  the  first  name 
tof  a  freeman  "  peutherer  "  does  not  appear 
ill  I597- 
The  confirmation  by  letters  patent  of  the 
M  89 


OLD  PEWTER 

London  Company's  privilege  of  charging  a 
royalty  on  the  smelting  and  *  casting  of  tin, 
obtained  in  1598,  brings  the  history  of  pewter 
in  the  sixteenth  century  to  a  close.  The  most 
remarkable  point  in  it  is  the  still  further 
widening  of  the  social  field  in  which  it  came 
into  common  use  in  England,  though  this 
was  perhaps  not  altogether  without  certain 
incidental  disadvantages  from  the  pewterer's 
point  of  view.  Whether  it  were  cause  and 
effect  due  to  a  resentment  on  the  part  of 
fashion  against  this  increasing  employment 
of  pewter  by  the  commonalty,  or  mere  coinci- 
dence arising  out  of  the  growing  prosperity 
inaugurated  by  the  reign  of  peace  which 
ensued  from  the  overthrow  of  Richard 
on  Bosworth  Field  and  the  accession  to 
the  throne  of  Henry  VII.,  it  would  be 
hard  to  say ;  but,  in  England  at  any  rate, 
even  in  the  opening  years  of  the  century,  we 
begin  to  discover  evidences  of  a  steady  ten- 
dency among  the  higher  classes  towards  the 
uses  of  the  more  precious  metals.  In  1501 
the  banquet  given  in  honour  of  the  politic 
marriage  of  Arthur  Tudor,  the  king's  ill- 
fated  eldest  son  and  heir,  to  the  scarcely  less 
unfortunate  Catherine  of  Arragon,  is  stated 
90 


UH  X 


.    o 
ffi  £* 


THE  SIXTEENTH  CENTURY 

to  have  been  served  on  gold  plate  enriched 
with  pearls  and  other  gems  at  a  total  cost  of 
more  than  twenty  thousand  pounds ;  while 
the  accounts  given  some  years  later  by  Robert 
Amadel  and  Cavendish  of  the  plate  belonging 
to  Cardinal  Wolsey  read  more  like  pages 
from  the  "  Arabian  Nights  "  than  sober  state- 
ments of  fact.  Setting  aside  such  objects  as 
an  image  of  the  Virgin  and  great  candlesticks, 
for  which  pewter  would  have  been  inappro- 
priate, we  find  three  chargers  weighing  nine 
hundred  and  sixty-eight  ounces,  twenty-two 
dishes  weighing  four  hundred  and  fifty-one 
ounces,  and  so  on.  So  prodigious  was  the 
amount,  that  even  at  great  banquets  a  large 
proportion  of  it  was  reserved  solely  for  dis- 
play. "  There  was,"  says  Cavendish,  speak- 
ing of  such  occasions,  "  a  cupboard  as  long  as 
the  chamber  was  in  breadth,  with  six  deskes 
in  height,  garnished  with  guilt  plate,  and  the 
nethermost  deske  was  garnyshed  all  with  gold 
plate,  having  with  lights  one  pair  of  candle- 
sticks with  silver  and  guilt,  being  curiously 
wrought,  which  cost  three  hundred  marks. 
This  cupboard  was  barred  round  about  that 
no  man  might  come  nigh  it,  for  there  was 
none  of  all  this  plate  touched — there  was 


OLD  PEWTER 

sufficient  besides."  As  this  example  was 
certainly  followed,  though  less  lavishly  per- 
haps by  smaller  men,  the  great  Cardinal  was 
clearly  no  good  friend  to  the  pewterers  ;  and 
they,  at  least,  can  have  felt  no  great  reason  to 
lament  his  fall. 

Other  noticeable  features  in  this  century 
are  the  continued  growth  of  the  craft  in 
Scotland,  the  rapid  development  of  the 
manufacture  of  highly  decorated  ware  in 
France  and  Germany,  and  the  regrettable 
and  wholly  irreparable  destruction  of  church 
plate  during  the  Reformation  in  England  and 
Scotland. 


92 


o 


O 


o 


Q 
W 
tti 
W  . 

>•£ 
O  S 
O  S 


THE  SEVENTH  CHAPTER 

SOME   FACTS   ABOUT   PEWTER   IN 
THE   SEVENTEENTH   CENTURY 

ROUBLES  began  to  thicken 
round  the  London  pewterers 
during  this  century.  As  early 
as  1 60 1  the  provincial  compe- 
tition was  making  itself  so 
markedly  felt  that  they  resolved  to  insist 
that  thenceforth  no  craftsman  under  their 
control  should  work  publicly  in  an  open  shop, 
thereby  giving  "  occasion  that  pewterers  of  the 
country  and  others  shall  come  to  great  lyght 
of  farther  knowleg  to  the  great  hindraunce 
of  the  Company,"  and  so  serious  did  they 
deem  the  danger  to  be  that  a  first  failure  to 
comply  with  the  requirements  of  the  ordinance 
was  punishable  by  a  fine  of  thirteen  shillings 
and  fourpence,  a  second  by  one  of  twenty 
shillings,  and  the  third  by  expulsion,  so  that 
"  no  brother  of  the  Company  shall  buy  and 
sell  with  them,"  an  ancient  precedent  for 
boycotting. 

1602  is  notable  as  the  date  inscribed  on 
the  most  interesting  surviving  specimen  of 

93 


OLD  PEWTER 

Scottish  pewter,  the  Pirley-Pig  of  Dundee, 
which  is  one  of  the  very  few  examples,  show- 
ing added  exterior  ornament.  It  is  a  flattened 
sphere  six  inches  in  diameter  and  three  in 
height,  with  a  narrow  opening  in  one  side 
for  the  insertion  of  coins,  and  was  used  for 
many  years  to  hold  the  fines  for  non-attend- 
ance paid  by  members  of  the  Council.  For 
a  time  it  disappeared,  but  by  a  most  fortunate 
chance  was  rediscovered  in  1839  in  a  heap  of 
old  metal  ear-marked  for  the  foundry,  and 
now  once  more  forms  one  of  the  treasures  of 
the  Town  Hall. 

To  the  collector  1603  is  not  without  im- 
portance, as  in  that  year  by  a  Canon  of  the 
Reformed  Church  of  England  it  was  required 
that  the  wine  should  be  brought  "to  the 
Communion-table  in  a  clean  and  sweet  stand- 
ing pot  or  stoup  of  pewter — if  not  of  purer 
metal/'  and  it  is,  therefore,  safe  to  assume 
that  no  pewter  church  flagons  are  of  earlier 
date.  That  this  permission  to  use  pewter 
was  promptly  taken  advantage  of  is  shown  by 
an  entry  at  Strood  near  Rochester  in  1607  of 
11  the  purchase  from  Robert  Ewer  (for  95.  6d.) 
Two  pewter  pots  to  serve  the  wine  at  the 
Communion " ;  while  one  still  existing  at 
94 


g 

hJ    O 


tuO 

H 

w 

"•n   O 

53  °r  >> 


THE  SEVENTEENTH  CENTURY 

Werrington  in  Northamptonshire  bears  an 
inscription,  "  Ex  dono  Edmundi  Pennye  et 
Franciscae  uxoris  ejus  ad  usum  Capellae  de 
Werrington  1609." 

An  example  of  seemingly  calm  appropria- 
tion of  another  man's  ideas,  bearing  the  date 
1 6 1 1,  is  to  be  seen  at  the  Victoria  and  Albert 
Museum  in  the  form  of  a  salver  of  Nurem- 
berg pewter  which  is  practically  a  reproduc- 
tion of  Briot's  Temperantia  salver,  though 
it  is  unblushingly  signed,  SCULPEBAT 
GASPARD  ENDERLEIN,  and  stamped 
with  his  initials  G.  E. 

About  this  time  we  begin  to  perceive  an 
augmenting  activity  in  the  craft  in  England, 
as  indicated  by  the  lengthening  catalogue  of 
articles  turned  out.  Among  the  new  names 
extracted  by  Mr.  Welch  from  a  list  made  by 
the  Company  in  1612,  are  "  chapnets,"  small 
vessels  for  church  use,  weighing,  according 
to  size,  one  pound  and  a  half  or  one  pound 
the  half  dozen ;  great  beakers,  both  wrought 
and  plain,  middle,  small,  and  children's 
beakers,  large  and  small  beer  bowls,  large 
wrought  cups,  two  sizes  of  French  cups, 
"  spowt  potts  "  of  various  capacities,  hawkes- 
bills  and  ravensbills,  which  were  ewers  of 

95 


OLD  PEWTER 

different  sorts,  named  perhaps  from  the  beak- 
like  shape  of  the  spout ;  thurndells  and  half 
thurndells,  according  to  Mr.  Mass£,  on  the 
authority  of  Mr.  St.  John  Hope,  a  Wiltshire 
name  for  vessels  containing  three  pints  ;  and 
a  number  of  measures,  "new  quarts,  new 
great,  small  and  halfe  potts,  hooped  thurn- 
dells, Winchester  quarts  and  pints,  with  or 
without  lidds,  long  hooped  Winchester  pyntes 
and  Jeayes  danske  potts,"  which  last  are  a 
mystery.  The  rigorous  distinctions  between 
the  various  branches  of  craftsmen  seem  to 
have  been  somewhat  relaxed  about  the  same 
period,  for  the  Triflers  were  including  not  a 
little  hollow-ware  in  their  output. 

The  curious  independence  of  the  Scottish 
towns  in  those  days  of  bad  roads  and  little 
travel  is  suggested  by  the  fact  that  only  in 
1614  did  the  Incorporation  of  Dundee  ordain 
the  hall-marking  by  the  Dean  of  Guild  and 
Town  Baillie  of  tavern  measures,  nearly  a 
century  after  Edinburgh  had  adopted  the 
same  sensible  course,  although  as  early  as 
1563,  and  again  in  1568  and  1613,  stern 
punishment  had  been  provided  for  the  use 
of  false  measures. 

Still  new  uses  for  pewter  come  to  light  in 

96 


£  o 


a  ^ 


THE  SEVENTEENTH  CENTURY 

a  list  of  " trifles"  in  1614,  such  as  basins, 
bowls,  pastie  plates,  pye  coffins,  limbecks, 
cefters,  and  still-heads,  so  that  it  is  small 
wonder  that  in  my  lord  Northampton's 
kitchens  that  year  the  weight  of  the  pewter 
amounted  to  three  hundred  pounds.  Thirteen 
candlesticks  of  pewter  form  a  noteworthy 
item  in  the  inventory  of  Sir  Thomas  Hoskyns 
of  Oxted,  dated  1615;  the  rest,  however,  con- 
sisting of  eight  dozen  dishes  of  all  sorts,  four 
dozen  saucers,  and  four  flagons,  is  of  the 
usual  type  of  kitchen  furniture. 

A  second  wave  of  revolution  swept  over  the 
Church  of  Scotland  in  1617,  when  James  I. 
(VI.  of  Scotland)  forced  upon  his  old  kingdom 
the  form  of  worship  prevailing  in  his  new  one, 
under  the  name  of  Episcopalianism,  and  the 
change  of  ritual  unquestionably  contributed  to 
the  destruction  of  much  of  any  plate  that  had 
survived  from  1567.  It  also  brought  into 
ecclesiastical  use,  apparently  for  the  first 
time,  a  new  vessel  in  the  "  laver,"  a  form 
of  flagon,  usually  with  a  narrow  spout,  which 
was  used  either  for  pouring  the  water  at 
baptism  on  to  the  face  of  the  child  or  simply 
to  convey  the  water  to  the  font.  No  pewterer 
is  especially  mentioned  as  a  member  of  the 

N  97 


OLD  PEWTER 

Incorporation  of  St.  Andrews  until  1619, 
and  Mr.  Wood  conjectures  that  the  supplies 
of  ware  which  must  have  been  demanded  by 
the  castle,  the  university,  and  the  cathedral, 
were  obtained  from  the  gypsies  and  other 
wandering  unfreemen,  who  are  known  to  have 
driven  a  thriving  trade  in  Scotland  just  as 
did  the  "deceivable  hawkers"  in  England, 
for  whose  suppression  the  London  Company 
were  once  more  clamouring  in  1621  as  they 
had  done  ineffectually  so  many  times  before, 
A  further  element  of  tribulation  for  it  appears 
about  this  time  under  the  guise  of  "  the 
Crooked  Lane  men "  whose  ware  was  as- 
serted by  it  in  1634  to  be  a  "counterfeiting 
of  the  reall  commodity  of  Tynn."  What  this 
ware  was  actually  made  of  cannot  be  stated 
with  absolute  certainty.  That  it  was  not 
regarded  as  fraudulent,  except  by  the  pew- 
terers,  who  can  scarcely  be  considered  as  in- 
dependent or  disinterested  judges,  is  clear, 
since  their  hostile  attempts  were  unsuccessful ; 
and,  as  workers  in  tin-plate,  under  the  name 
of  white-ironsmiths,  were  accepted  as  journey- 
men by  the  Corporation  of  Aberdeen  as  early 
as  1649,  and  as  full  freemen  of  that  of  Glas- 
gow in  1652,  it  is  scarcely  a  rash  conclusion 
98 


PLATE  LXVIII 


. 


: 


H.'io" 


JUG,  George  IV 


THE  SEVENTEENTH  CENTURY 

that  in  these  obnoxious  Crooked  Lane  men 
we  have  the  first  of  those  tinsmiths  whose 
handiwork  in  years  to  come  was  fated  to 
play  no  small  part  in  the  overthrow  of  the 
pewterer's  craft. 

It  is  rather  sad  in  the  face  of  this  aggres- 
sive eagerness  on  the  side  of  the  pewterers  to 
protect  "  his  Matits  subjects  "  from  "  deceipt  or 
wrong  "  to  find  that  they  were  themselves  en- 
gaged at  the  very  time  in  marking  their  wares 
in  a  manner  so  closely  resembling  the  four 
hall-marks  which  the  Goldsmiths'  Company 
stamped  on  real  plate,  that  in  1635  the  latter 
appealed  for  redress  to  the  Privy  Council,  and 
that  in  response  the  Court  of  Aldermen 
ordained  that  the  pewterer  should  use  only 
one  stamp,  though  he  might  add  the  arms  or 
sign  of  the  purchaser  on  his  desiring  it. 

Once  more,  in  1638,  there  came  an  up- 
heaval in  the  Church  of  Scotland,  when  the 
people,  who  had  always  more  or  less  hated 
and  actively  resented  it,  revolted  against 
Episcopalianism,  and  pledged  themselves  to 
the  New  Covenant,  but  the  result  of  this  was 
probably,  at  any  rate  at  first,  an  increase  in 
the  use  of  pewter  for  church  vessels,  the  more 
precious  metals  being  absorbed  in  providing 

99 


OLD  PEWTER 

means  for  carrying  on  the  conflict  which 
followed.  1640  is  the  earliest  date  on  the 
first  of  the  touch-plates  still  preserved  at 
Pewterers'  Hall,  forming  with  a  windmill  the 
touch  of  one  N.M.,  whose  initials  are  enclosed 
in  an  octagon  ;  while  1641,  according  to  the 
Rev.  J.  E.  Nightingale's  "  Church  Plate  of 
Dorset,"  is  the  oldest  date  on  pewter  found  in 
that  county,  viz.,  two  flagons  inscribed,  "  Ex 
dono  Henrie  Arnoldi  Ilsingtoniensis." 

The  same  year  witnessed  the  passing  of 
an  Act  by  the  Parliament  of  Scotland  com- 
manding every  pewterer  to  mark  his  ware 
with  the  thistle,  the  Deacon's  mark,  and  his 
own  name,  and  ordaining  that  none  but  the 
finest  pewter,  such  as  in  England  bore  the 
mark  of  the  rose,  was  to  be  employed.  It  is 
probably  owing  to  the  regrettable  carelessness 
displayed  in  preserving  the  records  of  the 
Hammermen  of  Glasgow  that  we  find  no 
mention  of  a  pewterer  in  such  as  remain 
before  1648,  for,  though  the  now  large  and 
prosperous  city  was  at  that  time  but  a  small 
place,  it  had  its  cathedral,  and  it  seems 
scarcely  credible  that  there  should  have  been 
no  resident  craftsmen  to  meet  the  demand 
that  must  have  existed  for  the  making  and 
100 


w 


tu 
a 

w 


o 


THE  SEVENTEENTH  CENTURY 

repairing  of  pewter  ware.  An  unusual  but 
significant  employment  for  a  member  of  the 
craft  is  revealed  by  an  Act  of  Parliament 
passed  at  Edinburgh  in  1649  to  authorise  the 
payment  of  a  sum  of  four  hundred  and  forty- 
six  pounds  odd  sterling  (five  thousand  three 
hundred  and  fifty-three  Scots)  due  to  one 
James  Monteith  from  the  Government  for 
making  musket  and  pistol  balls,  presumably 
for  the  Parliamentary  forces  in  Scotland  and 
England,  two  years  before,  an  account  which 
was  duly  discharged,  thus  saving  the  creditor 
from  carrying  out  his  rather  Gilbertian  threat 
of  deserting  his  wife  and  family.  Nor  can 
James  Monteith  have  been  the  only  one 
among  the  pewterers  to  profit  handsomely  by 
the  internal  dissensions,  for  to  his  English 
contemporaries  also  it  must  have  come  as  a 
not  wholly  disagreeable  illustration  of  the 
old  proverb  anent  an  ill  wind.  As  the  long- 
drawn-out  contest  between  King  and  Parlia- 
ment dragged  on  its  weary  length,  nobleman 
after  nobleman  and  gentleman  after  gentle- 
man sent  his  rich  stores  of  gold  and  silver 
plate  to  be  melted  down  to  meet  the  endless 
drain  upon  the  war-chest,  and,  since  men  must 
eat  even  in  the  intervals  of  victory  and  defeat, 

101 


OLD  PEWTER 

the  vanished  treasures  were  doubtless  replaced 
by  humble  pewter.  Even  royalty  itself,  as  we 
know,  was  fain  at  last  to  be  content  with  such, 
for  in  the  possession  of  Mrs.  F.  W.  Barry 
there  is  a  large  and  handsome  rose-water  dish, 
which  was  one  of  a  set  of  six  provided  for  the 
King's  use  when  he  lay  at  York.  The  opposing 
party  in  their  turn  also  made  much  work  for  the 
pewterers'  lathes  and  hammers,  more  especially 
when  they  flung  out  from  the  churches  the 
fonts  as  savouring  particularly  of  papistry, 
necessitating  the  use  of  pewter  bowls  and 
basins  for  baptismal  purposes. 

In  France  during  this  half-century  the  use 
of  pewter — except  the  ornamental  kind — by 
the  nobility  had  steadily  declined,  and  not  even 
the  permission  given  in  1650  to  gild,  silver, 
or  lacquer  it,  forbidden  till  then  save  for 
church  use,  was  successful  in  inducing  them 
to  return  to  it  to  any  great  extent,  although 
by  royal  command,  in  order  to  defray  the 
expenses  of  the  endless  wars,  their  plate  was 
sent  to  the  Mint,  the  King  leading  the  way 
with  "tables,  candelabra,  large  seats  of  silver," 
so  "enriched  with  figures,  bas-reliefs,  and 
chasings  "  that,  though  they  had  cost  him  ten 
million  francs,  the  metal  when  melted  down 
1 02 


*  t 
•8  : 


W 


THE  SEVENTEENTH  CENTURY 

was  worth  no  more  than  three.  So  despised, 
indeed,  had  pewter  become,  that  the  Grande 
Mademoiselle  when  visiting  Nanteuil  six 
years  later  thought  it  needful  to  her  dignity 
and  delicate  susceptibilities,  while  commend- 
ing her  supper,  to  deprecate  the  pewter  it  was 
served  on. 

The  punishment  recorded  in  1652  of 
William  Abernethie  of  Edinburgh,  for  using 
bad  metal,  is  remarkable  as  the  only  instance 
of  such  a  necessity  arising,  a  praiseworthy 
proof  of  the  faithfulness  with  which  the 
Incorporation  and  craftsmen  fulfilled  their 
obligations  to  the  public. 

It  may,  perhaps,  be  doubted  whether  the 
London  Company  was  so  unselfishly  con- 
siderate for  the  consumer  when  in  the  same 
year  it  forbade  Thomas  Allen  to  lend  his 
professional  assistance  to  a  Major  Purling, 
who  had  invented  a  new  material  which  he 
called  Silvorum,  and  in  the  following  year 
fined  Lawrence  Dyer  for  working  in  it,  though 
the  last,  be  it  acknowledged,  did  aggravate  his 
offence  by  selling  it  unmarked.  One  would 
like  to  know  whether  he  paid  any  or  all  of  his 
fine  in  the  farthings  stamped  "£  of  an  ounce 
of  fine  pewter,"  which  Cromwell  found  it 

103 


OLD  PEWTER 

necessary  to  issue  that  year  in  imitation  of 
the  tradesmen's  tokens,  which  had  for  many 
years  been  fashioned  in  that  material.  Cer- 
tainly it  was  on  no  sentimental,  altruistic 
grounds  that  in  1658,  when  they  were  debarred 
from  introducing  their  ware  into  Bordeaux, 
doubtless  with  its  large  wine  trade  a  paying 
customer,  the  London  pewterers  grumbled, 
quite  regardless  of  the  fact  that  their  own  laws 
against  the  importation  of  foreign  ware  were 
of  the  strictest. 

With  the  return  of  Charles  II.  to  the  throne 
in  1660  the  hated  yoke  of  Episcopalianism 
was  again  laid  on  the  shoulders  of  the  Church 
of  Scotland,  and  doubtless  a  new  era  of  melt- 
ing down  and  refashioning  church  plate, 
pewter  and  otherwise,  ensued ;  but  the  interests 
of  the  pewterers  were  not  ignored,  for  in  1661 
an  Act  was  passed  absolutely  prohibiting  the 
exportation  of  old  pewter,  a  precaution  by  no 
means  unnecessary  if  we  may  deduce  the  rarity 
of  the  metal  there  from  the  fact  that  it  was 
estimated  at  nearly  three  times  its  value  on 
the  south  side  of  the  border.  This  high  price, 
in  fact,  would  seem  to  have  had  a  damaging 
effect  upon  the  business  of  the  craftsmen,  and 
to  have  driven  those  of  Edinburgh  into  under- 
104 


§^  S  rt 

O  ^°  o 
^.  Q  w  ^  2 
T  w    ^n  x' 
«^W>0;s 


<£  >> 

.     1      rrt      1-     ' 


IiJ  rt 


^4  | 


QX 


T  ZD 


Qx:  » 

-   W'S  1 
's  o  *  J 

Q'  §  ^  "w 


THE  SEVENTEENTH  CENTURY 

taking  jobs  outside  their  regular  profession, 
for  an  Act  was  called  for  in  1663  to  decide 
that  plumbers  belonged  to  a  distinct  trade, 
and  that  pewterers  had  no  right  to  work  in 
lead. 

Cisterns  and  toys  next  appear  among  the 
articles  fashioned  of  pewter  in  England,  Mr. 
Samuel  Pepys  having  purchased  one  of  the 
former  in  1667,  while  Francis  Lea  was  fined 
ten  shillings  in  1668  for  making  the  latter  of 
inferior  quality,  for  even  to  such  small  deer 
did  the  powerful  Company  deign  to  attend  in 
the  intervals  of  such  weightier  business  as 
joining  the  Girdlers  in  1669  in  opposing  the 
endeavour  of  the  poor  Crooked  Lane  men  to 
get  a  charter  for  themselves,  or,  as  in  1671, 
making  puzzling  and  apparently  contradictory 
resolutions  as  to  the  lawful  use  of  the  rose 
and  crown  stamp  and  the  addition  or  omission 
of  the  name  of  the  maker  in  full. 

Pewter  spoons  and  forks  are  found  in  a 
French  inventory  of  1672. 

A  series  of  ordinances  issued  by  the  Aber- 
deen Incorporation  are  interesting  as  showing 
with  what  intimate  details  of  the  daily  life  of 
its  members,  quite  apart  from  their  work,  it 
was  suffered  to  interfere,  more  especially  in  the 

o  105 


OLD  PEWTER 

case  of  apprentices,  who,  if  the  rules  were 
conscientiously  enforced,  must  have  had  a 
dull  and  rather  difficult  time.  The  quarrel 
between  the  pewtereis  and  plumbers  of 
Edinburgh  came  to  a  head  next  year,  and 
resulted  in  1679  in  an  action  brought  by  the 
former  in  which  they  endeavoured  to  establish 
their  right  in  the  teeth  of  the  Act  of  1663  to 
work  in  lead,  and  some  of  the  pleadings  show 
the  smallness  of  the  trade,  owing  to  which 
the  craftsmen  were  unable  to  exist  by  it 
alone.  Yet  people  of  fair  social  standing  did 
not  hesitate  to  embark  in  it,  a  nephew  of  the 
Laird  of  Wallieford  being  apprenticed  in 
1687.  The  collapse  of  Episcopalianism  and 
revival  of  Presbyterianism  on  the  accession 
of  William  of  Orange  in  1688  led  once  again 
to  alterations  in  church  plate,  though  on  this 
occasion  much  more  of  it  was  preserved  in- 
tact, and  some  of  it  still  survives. 

In  the  course  of  the  same  year  the  pew- 
terers  of  London  made  it  clear  that,  however 
cordially  they  might  welcome  a  king  from 
foreign  parts,  they  were  not  prepared  to 
extend  their  toleration  to  rival  workmen,  for 
one  Mark  Henry  Chabrolles,  a  Frenchman, 
was  bluntly  forbidden  to  open  a  "shopp." 
106 


CL_I 


DESCRIPTIONS  OF  OBJECTS  ON  PLATE  LXXIV. 

1.  Tankard,  English,  4"  high.     Early  XVIII  century. 

2.  Dish,  Scotch,  stamped  with  initials  ADV  and  AW.  dinm.  i^ 

XVI  century. 

3.  Mug  with  handle.  English,  si"  high.     Early  XVII I  century. 

4.  Measure,  English,  2l"  high,  2|"  diam.     XVII  century. 

5.  Tankard,  English,  4"  high.     XVIII  century. 

6.  Tobacco-box,  English,  5^"  high,  4"  diam.     XVIII  century. 


THE  SEVENTEENTH  CENTURY 

Learning  later  that  he  was  a  Protestant 
refugee,  they  consented  to  his  working 
"some  time  longer,"  but  in  1692,  repenting 
of  its  leniency,  the  Court  finally  ordered  him 
to  abandon  the  trade  of  pewterer  by  the 
24th  of  August,  as  they  recognised,  some- 
what tardily,  that  his  practising  it  was  illegal. 
At  the  same  time  it  was  tinkering  with  the 
rules  as  to  stamping  wares  with  a  persistency 
and  contrariety  that  would  have  rejoiced  the 
heart  of  a  War  Office  permanent  official.  In 
1690  complaint  was  made  that  one  Samuel 
Hancock  was  brazenly  stamping  his  name  in 
full  upon  his  plates,  yet  in  1692  the  Court 
ordered  or  allowed  this  to  be  done,  only  to 
renew  its  prohibition  six  months  later,  which 
was  again  withdrawn  in  1697. 

A  significant  entry  appears  in  the  records 
of  Aberdeen  for  1694  admitting  white-iron- 
smiths  to  the  freedom  of  the  Incorporation, 
by  which,  as  far  as  the  pewterers  were  con- 
cerned, it  was  opening  the  gate  to  an 
insidious  foe.  Indeed,  everywhere  at  the 
end  of  the  seventeenth  century  signs  of  the 
approaching  end  of  the  prosperous  days  of 
the  craft  were  becoming  plainly  visible.  In 
Scotland  the  white-ironsmiths,  in  England 

107 


OLD  PEWTER 

the  makers  of  glass  and  earthenware,  were 
steadily  absorbing  more  and  more  of  the 
trade ;  so  much  so  that  the  Pewterers'  Com- 
pany attempted,  though  in  vain,  to  procure 
the  passing  of  an  Act  of  Parliament  enforcing 
the  sale  on  draught  of  beer  and  other  intoxi- 
cating liquors  in  pewter  vessels  only,  declar- 
ing with  sublime  effrontery  that  no  others 
held  full  measure.  It  may  even  be  doubted 
whether  the  long  wars  with  France  which 
followed  the  triumph  of  William  of  Orange 
and  the  final  overthrow  of  the  Stuarts  gave 
much  stimulus  to  the  pewterers'  trade, 
though  William,  after  having  been  driven  to 
the  same  expedients  as  his  rival  LouisXIV. — 
melting  down  the  larger  portion  of  the  pro- 
fusion of  silver  accumulated  by  Charles  II., 
and  calling  on  his  subjects  for  similar  sacri- 
fices— was  finally  reduced  to  striking  part  of 
his  coinage  in  pewter. 


1 08 


PLATE  LXXV 


H.   12" 


SOUP  TUREEN,  Russian.     Archangel  mark, 


1 

w 

*  O  £ 

X  D  3 

*sj 

II 


W 


as  s 


THE  EIGHTH  CHAPTER 

THE  END  OF  THE  STORY 

HE  history  of  pewter  during  the 
eighteenth  and  nineteeth  cen- 
turies is  an  unrelieved  record  of 
steady  continuous  decline,  of 
loss  by  the  Companies  and 
Incorporations  of  their  power  to  control  the 
trade,  of  loss  finally  by  the  craftsmen  of  the 
trade  itself.  At  the  beginning  of  the  period 
the  London  Company  was  still  struggling 
with  the  vexed  question  of  the  marking  of 
pewter.  Regulation  after  regulation  had 
ordained  that  each  maker  should  possess  a 
mark  peculiar  to  himself,  that  this  mark 
should  be  registered  on  a  plate  kept  for  the 
purpose  at  the  Company's  Hall,  and  that  every 
master  should  stamp  this  touch  on  every  piece 
of  pewter  that  issued  from  his  shop.  Yet  in 
1702  it  was  still  necessary  for  the  Company 
to  obtain  a  charter  empowering  them  to  make 
more  regulations,  or  rather  to  reiterate  those 
that  had  been  laid  down  so  often  before.  The 
attempt  to  prevent  the  importation  of  foreign 
pewter  would  appear  to  have  been  no  more 

109 


OLD  PEWTER 

successful,  in  spite  of  the  fact,  as  proved  by  a 
number  of  tests  undertaken  by  the  Company 
in  1709,  that  it  was  immensely  inferior  in 
quality  to  the  native  material. 

In  Scotland  at  the  same  period  the  in- 
sidious white-ironsmiths  were  gradually  ex- 
tending their  boundaries.  One  is  mentioned 
at  Edinburgh  in  1713;  another,  the  first,  was 
admitted  as  a  freeman  of  the  Dundee  Incor- 
poration in  1715  ;  while — a  still  more  signifi- 
cant fact — in  1720  the  last  pewterer,  one 
Patrick  Sampson  of  Dundee,  joined  the  St. 
Andrews  Incorporation. 

The  grip  of  the  London  Company  weak- 
ened yearly.  Few  rules  had  been  more 
persistently  enforced  than  that  forbidding 
anything  approaching  to  self-advertisement, 
yet  in  1727  one  Samuel  Smith  had  no  hesita- 
tion in  asserting  publicly  by  means  of  the 
stamp  upon  his  wares  that  these  were  "good 
mettle  made  in  London."  As  far  as  the 
country  was  concerned  matters  were  even 
worse.  The  Company,  which  had  owned  and 
regularly  exercised  the  right  of  search  for 
inferior  ware  throughout  the  length  and 
breadth  of  England,  though  assured  in  1729 
that  such  was  being  offered  for  sale  in  large 
no 


PLATE  LXXVII 


Ewer,  H.  8|"     Basin,  14"  x  io|" 

EWER  AND  BASIN. 


THE  END  OF  THE  STORY 

quantities  at  Bristol,  no  longer  dared  to  put 
their  prerogatives  into  practice,  and  by  a 
timorous  inaction  surrendered  their  whole 
position  and  abandoned  their  justification  for 
continued  existence.  To  issue  edicts  to  which, 
on  your  own  confession,  you  have  no  means 
of  compelling  obedience,  is  to  wilfully  court 
humiliation.  The  weakness  of  the  authorities 
was  furthermore  reflected  in  the  trade  itself. 
The  makers  lost  their  originality  and  in- 
dividuality, and  more  and  more  as  time  went 
on  we  find  them  content  merely  to  re-echo  the 
work  of  the  gold-  and  silver-smiths.  This  is 
especially  observable  in  their  table  ware,  as, 
for  example,  in  the  dish  [Plate  IX.  2],  the 
mustard  pots  [Plates  VII.  4,  and  VI 1 1. 6  and  7], 
pepper  pots  [Plate  VIII.  2  and  3],  and  salts 
[Plate  VIII.  9  and  12],  which,  admirable  as 
they  are  in  design,  are  only  slavish  copies 
of  the  silver  of  the  time. 

In  1729  the  last  pewterer  was  enrolled 
among  the  Hammermen  of  the  Cannongate, 
and,  moreover,  the  Patrick  Sampson  who,  as 
mentioned  above,  had  migrated  from  Dundee 
to  St.  Andrews  nine  years  previously,  so 
evidently  had  found  business  in  that  city 
unsatisfactory  that  he  shook  its  dust  from  his 

in 


OLD  PEWTER 

feet,  and  returning  to  his  former  place  of 
residence  sought  and  obtained  admission  to 
the  Incorporation  there,  while  the  craftsmen 
of  St.  Andrews  set  themselves  to  the  task  of 
reducing  the  overcrowding  of  the  Incorpor- 
ation which  had  resulted  from  the  ill-con- 
sidered diminution  of  the  entrance  fees  some 
years  earlier,  by  enacting  that  no  one  who  was 
not  either  the  son  of  a  freeman  or  married  to 
the  daughter  of  one  should  thenceforth  be 
entitled  to  belong  to  it.  How  far  this  served 
to  improve  matters  for  the  time  we  do  not 
know,  but  in  endeavouring  to  bolster  up  an 
out-of-date  monopoly  they  were  foredoomed 
to  failure  in  the  end.  The  spirit  of  the  age, 
with  its  growing  sense  of  personal  freedom, 
was  against  all  such  galling  restrictions  on 
the  liberty  of  the  individual,  however  well 
directed  to  the  public  benefit  they  may  have 
been,  and  from  an  order  issued  in  1732  to  its 
members  by  the  Incorporation  of  Perth  to  spy 
out  what  burgesses  of  the  town  were  illegally 
trafficking  with  the  unfreemen  it  is  clear  that 
the  consumer  from  without  was  aiding  and 
abetting  the  revolt  from  within.  One  by  one 
the  Incorporations  had  to  retreat  from  positions 
formerly  held  in  strength.  Thus  in  1733 
112 


£p          Q 

<N    Q 


THE  END  OF  THE  STORY 

Patrick  Campbell  was  admitted  as  freeman  of 
both  "  the  Coppersmith  and  the  Pewter  Arts,'1 
though  only  nine  years  earlier  Ninian  Grey, 
although  Deacon  of  the  Incorporation,  was 
brought  to  book  for  working  at  the  two. 
The  same  year  the  pewterers  of  Edinburgh, 
beginning,  maybe,  to  realise  their  danger, 
endeavoured  to  assert  their  authority  over 
the  white-ironsmiths  by  denying  them  the 
right  to  alter  without  permission  their  "  essay," 
as  the  test-pieces  which  a  man  had  to  make 
before  admission  to  the  freedom  were  called, 
but  they  had  to  give  way,  a  result  which 
might  act,  but  did  not,  as  a  warning  to  their 
neighbours  of  the  Cannongate,  who,  all  un- 
suspecting, gave  entrance  to  their  first  white- 
ironsmith  that  very  year. 

The  London  Company  must  have  been 
equally  ready  to  forego  its  lawful  powers 
under  continued  pressure,  or  John  Jupe  would 
never  have  dared  in  1736,  not  only  to  label 
his  metal  "  Superfine,"  but  to  crown  his 
defiance  with  the  additional  insult  "  French  "  1 

By  1739  the  white-ironsmiths  of  Edin- 
burgh had  waxed  so  fat  in  their  own  estima- 
tion that  they  began  to  kick,  urging  complaints 
against  the  pewterers  and  demanding  the 

p  113 


OLD  PEWTER 

right  to  establish  a  definite  and  distinct 
craft  of  their  own,  but  that  time  they  went 
too  far,  and  the  pewterers  firmly  and  suc- 
cessfully rejected  the  appeal.  At  Perth, 
in  the  course  of  the  same  year,  a  tinsmith, 
the  first  to  be  found  in  the  records,  actually 
appealed  to  the  Incorporation  for  charitable 
relief,  a  disaster  to  one  of  the  rival  craft 
which  should  have  afforded  unbounded  satis- 
faction to  the  pewterer  members,  and  not 
the  less  so  because  George  Brown,  the  bene- 
ficiary, had  formerly  been  a  pewterer  and  an 
officer  of  the  Incorporation  withal,  and  had 
deserted  them  for  the  enemy,  evidently  with 
very  poor  results  as  far  as  he  was  concerned. 
The  year  1745,  so  fraught  with  bitter  memories 
for  Scotland,  brought  about  a  further  reduc- 
tion in  the  amount  of  any  early  plate  yet 
remaining  to  the  congregations  of  the  dis- 
established, but  not  deserted,  Episcopalian 
Church,  whose  members,  naturally  taking  the 
side  of  the  Pretender  in  the  ill-fated  attempt 
at  rebellion,  were  ruthlessly  harried  by  the 
victorious  Duke  of  Cumberland,  their  plate 
being  carried  off  and  their  chapels  burned, 
though  this  in  the  upshot  proved  of  some 
small  benefit  to  the  pewterers.  For  the 
114 


PLATE  LXXIX 


1  2345 

H.  3-J"  D.  2!"  H.  2J"  D.  2|"          H.  2"  D.  2J"     H.  4|"  Base  2"        H.  2J"  D.  3" 

MUSTARD-POT.      SALT-CELLAR.      MUSTARD-     PEPPER-BOX.     MEASURE. 

XVIII  century.  XVIII  century.  POT.    Late      XVIII  century.       XVIII  century. 

XVIII  cen- 
tury. 


PLATE  LXXX 


3  45 

From  4"  to  6"  high 

MUSTARD-POTS.    XVII  and  XVIII  centuries. 


PLATE  LXXXI 


Bi 


1  234 

From  4!"  to  6"  high 

PEPPER-POTS.     XVII  and  XVIII  centuries. 


THE  END  OF  THE  STORY 

congregations,  impoverished  and  oppressed 
could  only  afford  the  more  modest  ware 
when  replenishing  their  ravaged  stores,  and 
as  a  consequence  most  of  the  Episcopalian 
pewter  plate  still  in  existence  dates  from  the 
nearer  side  of  "  the  Forty-five."  This  demand, 
however,  was  far  too  small  to  provide  any 
serious  check  on  the  downward  path  of  the 
doomed  craft ;  in  the  very  next  year  the 
Dundee  Incorporation  enrolled  the  last  on  its 
books,  tinsmiths  thenceforth  occupying  their 
places,  while  after  1747  no  Perth  hammer- 
man can  be  shown  to  have  made  a  living  out 
of  pewter  alone,  combining  the  craft  with  work 
in  copper  or  some  other  trade,  a  practice,  as  has 
been  said,  at  one  time  utterly  prohibited. 

In  London  at  the  same  time  the  Company, 
with  an  inability  to  recognise  or  an  unwilling- 
ness to  admit  the  hollowness  of  their  preten- 
sions which  is  almost  pathetic,  were  still 
perseveringly  legislating  on  the  subject  of 
touches,  ordaining  that  all  objects  large 
enough  should  bear  the  Christian  name  and 
surname  of  maker  or  vendor,  and  smaller 
articles  his  registered  mark,  though  their 
edicts  were  by  then  so  openly  disregarded 
that  one  Edward  Box  two  years  before,  with 

"5 


OLD  PEWTER 

apparently  complete  impunity,  advertised  by 
means  of  the  stamp  upon  his  ware  that,  in  his 
opinion,  there  was  "  No  better  in  London." 

For  the  Episcopalians  of  Scotland  we  may 
well  feel  compassion  in  regard  to  the  destruc- 
tion of  their  ancient  plate,  since  it  was  sacri- 
ficed nobly  in  a  gallant  if  mistaken  effort  to 
uphold  a  falling  cause,  but  for  the  Presby- 
terians, in  their  almost  conspicuous  lack  of 
early  vessels,  we  have  nothing  but  contempt. 
Without  compulsion,  of  their  own  foolish  free 
will,  they  began  to  consider  such  precious 
relics  as  old-fashioned,  and,  from  about  the 
year  1750  onwards,  remodelling  silver  plate 
and  destroying  pewter  altogether,  they  wrought 
their  own  devastation.  Small  wonder  that 
the  pewterers  were  dying  out.  The  last  was 
enrolled  at  Aberdeen  in  1765,  and  at  Perth  in 
1771,  while  at  Glasgow  by  1776  the  white- 
ironsmiths  were  undisguisedly  making  num- 
bers of  articles  which  should  have  pertained 
properly  to  the  pewterer.  Nor  do  matters  seem 
to  have  been  any  better  abroad,  for  at  Paris 
in  1776  the  hitherto  independent  pewterers 
were  joined  in  somewhat  uncongenial  union 
with  the  coppersmiths  and  scale-makers  to 
form  a  single  company.  It  was  probably 
116 


PLATE  LXXXIII 


1  D.  4>' 

MEASURE.  SALT. 


3    H.  5g"  4  5 

SUGAR-        VINEGAR  INKSTAND. 

CASTOR.       CRUET. 


PLATE  LXXXIV 


1    H.  if"  2    H.  ij" 

SALTS. 


3     H.  2"  4     H.  2 '          5     H. 


THE  END  OF  THE  STORY 

owing  rather  to  the  badness  of  trade  in 
general  than  to  any  successful  competition 
on  the  part  of  the  pewterers  that  no  white- 
ironsmith  appears  in  the  records  of  St. 
Andrews  until  1787,  for  by  1794  the  admis- 
sions of  tinsmiths  as  compared  with  pewterers 
were  as  six  to  one,  while  those  at  Edinburgh 
in  1795  were  strong  enough  to  so  far  turn  the 
tables  as  to  secure  the  condemnation  of  a 
pewterer  for  undertaking  their  work.  Well 
might  the  offender,  James  Wright,  exclaim  in 
bitterness,  "  Ichabod,  Ichabod — thy  glory  has 
departed." 

It  would  be  a  useless  and  ungrateful  task 
to  follow  the  practical  extinction  of  the  craft 
to  the  bitter  end  in  the  nineteenth  century. 
The  ever-increasing  cheapness  of  glass  and 
chinaware  ousted  pewter  from  table  use,  the 
discovery  of  the  possibility  of  coating  steel 
with  tin,  forming  what  is  now  known  as  block 
tin,  introduced  a  material  both  stronger  and 
lighter  for  hollow-ware,  the  invention  of 
Britannia  metal — a  form  indeed  of  pewter — 
German  silver,  nickel  plate,  and  other  com- 
positions, either  more  silver-like  in  themselves 
or  more  easily  susceptible  to  plating  with 
silver,  banished  it  for  the  most  part  from 

117 


OLD  PEWTER 

taverns  and  public  eating-houses,  and  the 
isolation  of  zinc  as  a  distinct  metal  and  its 
employment  in  combination  in  the  form  of 
galvanised  iron  gave  a  result  more  convenient 
for  many  and  varied  domestic  and  commercial 
purposes.  To  complete  the  downfall,  the 
pewterers  themselves  made  no  effort  to  meet 
the  altering  conditions.  With  that  fine  British 
conservative  obstinacy  which  is  by  no  means 
extinct  as  yet  among  us,  they  declined  to  adapt 
their  ancient  methods  to  new  demands.  They 
adhered  resolutely  to  the  good  old  principle, 
which  has  exercised  such  unmistakable  effects 
upon  our  foreign  trade,  that  it  is  the  business 
of  the  customer  to  want  what  the  manufacturer 
chooses  to  make,  in  no  way  whatever  that  of 
the  manufacturer  to  make  what  the  customer 
wants,  and,  clinging  with  limpet-like  consis- 
tency to  the  rock  of  trade  traditions,  they  let 
the  tide  of  public  requirements  ebb  away  from 
them,  never  to  return.  The  last  touch  on  the 
plates  at  Pewterers'  Hall  is  dated  1824,  and 
though  the  Company  still  survives,  and  is  by 
no  means  the  smallest  or  poorest  among  its 
fellows,  its  official  purpose  is  non-existent. 
A  Mr.  James  Moyes  kept  a  pewterer's  shop  in 
the  West  Bow,  Edinburgh,  until  some  time 
118 


PLATE  LXXXV 


L.  6'  2  3     H.  5"         4     II.?,;.  5     H.  5f  6     H.  aj 

SUGAR-SIFTERS.     SET  OF  CRUETS,  Scotch.     SPIRIT-LAMP. 
XVIII  century.  XVIII  century. 


PLATE  LXXXVI 


1     H.  4" 

2     H.  5S" 

3     H    i}8" 

4     H.  5|" 

CRE^M-IUG 

HERB-CAN- 

15OX.   No 

HERB-CAN- 

Scotch.     iNo 

NISTER, 

mark. 

NISTER,  Dutch. 

mark. 

Dutch.     XVII 

Marked  1766,  a 

century.    Mark 

Cupid  in  profile 

a  standing 

blowing  a  horn. 

figure. 

P.S.B. 

5    H   4J" 

CREAM-JUG, 

No  nicirk. 


THE  END  OF  THE  STORY 

between  1870  and  1880,  though  it  is  said  that 
for  some  years  previously  he  had  ceased  the 
manufacture.  The  Incorporations  of  Ham- 
mermen at  Aberdeen,  Dundee,  Perth,  and 
Stirling  still  hold  meetings  and  make  a  poor 
pretence  of  carrying  on  their  business,  but  as 
they  possess  no  executive  powers  the  prac- 
tical effect  is  insignificant ;  indeed,  as  far  as 
pewter  is  concerned,  is  nil.  Some  little 
pewter  is  still  made  in  London  on  the  old 
lines,  and  there  are  even  now  eating-houses 
where,  with  a  pleasant  if  somewhat  artificial 
conservatism,  the  guest  is  served  on  pewter, 
but  the  recent  attempts  to  revive  the  craft  by 
bending  it  to  the  too  often  ill-directed  pur- 
poses of  the  so-called  new  art  have  not  as 
yet  produced  much  calling  for  extended 
notice. 


119 


THE  NINTH  CHAPTER 

COLLECTING  AND  DISPLAYING 
PEWTER 

N  the  matter  of  a  scientific  know- 
ledge of  his  chosen  subject 
founded  on  accurate  and  organ- 
ised information  the  collector  of 
pewter  is,  and  most  indisputably 
will  always  remain,  sadly  at  a  disadvantage 
in  comparison  with  the  collector  of  gold  and 
silver  plate.  In  theory,  had  the  regulations 
of  the  Companies  and  Incorporations  been 
rigorously  enforced  or  loyally  adhered  to, 
and  had  the  records  been  faithfully  kept  and 
carefully  preserved,  there  should  have  been 
no  more  difficulty  in  establishing  the  identity 
of  the  maker  of  any  specimen  and  ascertain- 
ing its  approximate  date  in  the  one  case  than 
in  the  other.  Every  article  would  have  borne 
the  touch  of  the  craftsman.  A  comparison  of 
this  with  the  touch-plate  and  a  consultation 
of  the  register  of  touches  and  of  the  list  of 
freemen  would  have  been  all  that  was  neces- 
sary for  full  enlightenment.  Unfortunately 
this  ideal  state  of  things  is  very  far  from 

120 


U 


a  o 


COLLECTING  AND  DISPLAYING 

prevailing.  In  practice  it  may  be  said  that 
it  is  the  exception  rather  than  the  rule  when 
this  is  possible.  Indeed,  it  would  probably 
be  no  exaggeration  to  assert  that  nine  out  of 
ten  of  the  existing  specimens  of  pewter  bear 
no  mark  at  all,  and  that  of  those  which  are 
marked  ninety-nine  out  of  every  hundred 
now  convey  no  intelligible  meaning.  There 
are  indeed,  as  has  been  previously  observed, 
five  plates  of  touches  at  Pewterers'  Hall,  and 
Mr.  Massd  has  patiently  described  every 
touch  upon  them,  but  the  registers  which 
should  have  preserved  the  names  of  their 
owners  have  disappeared,  and  the  result  in 
the  present  state  of  our  knowledge  is  only 
useful  to  a  limited  extent.  The  number  of 
these  touches  is  nearly  twelve  hundred,  but 
of  these  only  forty-one  give  the  name  and 
date.  In  the  case  of  two  hundred  and  fifty 
the  name  alone  is  recorded,  but,  as  the  same 
is  included  in  the  list  of  freemen  also  repro- 
duced by  Mr.  Massd,  we  may,  in  the  majority 
of  them,  arrive  with  fairly  assured  certainty 
at  the  period  at  which  the  vessel  so  marked 
was  made,  while  in  three  instances  the  date  is 
given  and  the  name  may  be  conjectured.  In 
no  less  than  three  hundred  and  six  instances, 

Q  121 


OLD  PEWTER 

on  the  other  hand,  the  name  on  the  touch  is 
not  to  be  found  in  the  list  of  freemen  ;  in  one 
hundred  and  fifty-six  a  date,  with  initials  or  a 
device,  is  all  the  information  afforded,  and  in 
all  the  remaining  cases  there  is  neither  name 
nor  date  to  guide  us.  The  authorities,  in 
fact,  as  long  as  their  power  was  paramount, 
did  all  they  could  to  baffle  the  collector  of  the 
future,  unconsciously,  of  course,  for  it  can 
never  have  occurred  to  craftsmen,  working 
their  best,  indeed,  but  with  no  deliberate 
artistic  intent,  that  posterity  would  take  any 
interest  in  their  productions,  but  none  the 
less  effectively,  since  these  dateless  initials 
and  nameless  devices  may  be  traced  directly 
to  their  horror  of  self-advertisement  and 
their  consequent  prohibition  of  a  touch 
bearing  name  or  address,  even  so  wide  a 
one  as  "  London "  having  been  for  long 
forbidden. 

Had  the  touches  only  been  put  upon  the 
plates  in  orderly  succession  we  should  have 
been  less  at  a  loss,  for  it  would  at  all  events 
have  been  possible  to  assign  a  date  to  a  touch 
lacking  one  somewhere  between  the  nearest 
previous  and  succeeding  ones  which  were 
dated ;  but  there  is  no  doubt  whatever  that 
122 


COLLECTING  AND  DISPLAYING 

no  system  was  followed,  but  that  the  striker 
chose  his  own  place  haphazard,  left  his  mark 
where  he  would,  and  actually,  to  a  certain 
extent,  made  choice  of  what  plate  he  would 
strike  it  on.  In  no  other  way  can  we  account 
for  the  fact  that  the  earliest  remaining  plate 
bears  such  dates  as  1679,  1680,  1687,  &c., 
though  the  second  shows  1676,  and  the 
fourth  plate  was  stamped  in  1704,  1705,  and 
so  on,  while  the  third  contains  one  as  late  as 
1786,  and  many  others  of  the  later  half  of 
that  century.  Nor  is  this  the  last  of  the 
sources  of  possible  confusion.  Makers  some- 
times marked  their  wares  with  a  touch  differ- 
ing from  the  one  on  the  touch-plate,  and  this 
not  only  in  those  cases  where,  as  a  penalty  for 
infringing  some  regulation,  or  for  other  good 
reason,  one  was  ordered  to  change  his  touch, 
duly  recording  the  new  one.  Thus  Tim  Fly, 
who  was  Warden  in  1737  and  Master  in 
1739,  and  should  therefore  have  been  staunch 
in  upholding  the  rules  of  the  Company, 
registered  his  name  and  a  punning  device  of 
a  fly  as  his  distinguishing  mark,  yet  a  speci- 
men in  the  collection  of  Mr.  Hugh  Bryan 
bears  his  name  and  a  cock  and  crown,  which 
is  not  to  be  found  on  the  touch-plates, 

123 


OLD  PEWTER 

together  with  the  forbidden  four  small 
stamps  simulating  the  hall-marks  on  gold 
or  silver  plate.  Finally,  to  crown  our  un- 
certainty, some  makers,  either  illegally  and 
probably  with  fraudulent  intent,  or  with  due 
permission  given  on  sufficient  grounds, 
used  the  touch  registered  by  another  man, 
while  some  provincial  pewterers,  as,  for 
instance,  one  Stephen  Maxwell,  of  Glasgow, 
went  so  far  as  to  stamp  "  London"  on 
their  home-made  goods,  doubtless  to  enhance 
the  price.  These  Edinburgh  touches,  as  far 
as  they  go,  are  much  more  instructive  than 
those  of  London,  as  a  large  majority  of  the 
one  hundred  and  twenty-nine  described 
by  Mr.  Wood  are  dated  between  1600  and 
1764,  and  in  a  great  number  of  cases  it  has 
been  possible,  with  more  or  less  certainty,  to 
connect  the  initials  on  them  with  a  contem- 
porary name  in  the  existing  lists  of  freemen 
and  apprentices.  Unfortunately  only  two  of 
these  touch-plates  or  counterpanes  survive, 
and  they,  it  would  seem,  by  a  happy  and 
mysterious  accident.  Whether  the  Incor- 
porations of  the  other  Scottish  cities  kept 
their  own  touch-plates,  as  York,  in  England, 
did,  is  not  known.  Probably  they  did,  as 
124 


fe    3 

h-i 


HX 


COLLECTING  AND  DISPLAYING 

they  certainly  made  regulations  as  to  touches, 
but  in  no  instance  has  any  been  preserved. 
With  regard  to  private  touches,  then,  the 
owner  of  any  vessel  bearing  one  would  be 
well  advised  to  search  through  the  lists  given 
by  Mr.  Massd  and  Mr.  Wood  on  the  chance 
of  identifying  it,  though  the  likelihood  of  his 
doing  so,  or  of  obtaining  any  great  accession 
of  knowledge  if  he  does  so,  cannot  be  said  to 
be  very  great.  Speaking  broadly,  and  with 
due  regard  to  exceptions,  we  may  infer  from 
Mr.  Masses  list  that  up  to  the  end  of  the 
seventeenth  century  the  proportion  of  stamps 
bearing  names  to  those  bearing  initials  and 
devices  was  small,  that  names  in  full 
increase  constantly  in  frequency  during  the 
eighteenth  century,  and  that  by  the  end  of  it 
the  use  of  initials  alone  had  almost  died  out. 
In  respect  to  the  general  marks  which 
every  pewterer  was  supposed  to  use,  matters 
are  not  in  a  much  more  satisfactory  con- 
dition. The  rose  or  the  x,  crowned  or 
uncrowned,  were  English  guarantees  of 
quality  at  an  early  date,  but  the  former  was 
sometimes  used  under  pretence  of  being  a 
private  touch  on  inferior  ware,  and  in  the 
eighteenth  century  it  was  adopted  without  any 

125 


OLD  PEWTER 

such  meaning  as  it  had  in  the  south  by  many 
pewterers  in  Scotland.     A  similar  mark  also 
occurs  on  pewter  of  the  second  quality  made  at 
Lidge,  and  on  ware  from  Ghent,  Nuremberg, 
and  many  other  places  in  Holland,  Flanders, 
and  Germany.     The  x,  whether  crowned  or 
not,  denoting  extraordinary  ware,  and    four 
small  marks  imitating  those  on  silver  plate, 
were  frequent  in  London  from  at  least   the 
beginning  of  the  seventeenth  century  in  spite 
of  the  energetic  protests  of  the  Goldsmiths, 
but   these   also   were  adopted    in    Scotland 
during   the    eighteenth    century,   while    the 
thistle,  which  after  1641  was  an  evidence  of 
good  material   in   Scotland,  was   sometimes 
used  by  English  makers,  possibly  Scotsmen 
naturalised,    in    their    private     touches.     A 
crowned  hammer  was  a  mark  used  at  Mons 
as  early  as  1467,  sometimes  accompanied  by 
the  word  "fin,"  and  the  same  mark,  without 
the  word,  of  course,  was  used  as  a  sign  of 
first  quality  in  Scotland  during  the  sixteenth 
century.     The  fleur-de-lys  was  the  mark  of 
third  quality  at  Lidge,  but  was  also   often 
used  in  France,  and   the   same  was,  by  an 
order  made  in   1548,  to  be  stamped  on  the 
lids   attached   in    London   to    stoneware    or 
126 


PLATE  XCI 


fT     C 
1     1 


J 


I 

;f 
.* 


J 


1  2  3 

EGG-CUPS,  no  marks. 

4  5  6        H.  2l"        7 

SET  OF  SIX  EGG-CUPS,  Scotch.     No  marks. 


PLATE  XCII 


1     H.  31" 


3     H. 


2     H.  2"  D.  3" 

MILK-JUG,  English.    SUGAR-BASIN,       TFAPOT,  English.     Early 
Early  XIX  century.        English.      Early         XIX  century 
XIX  century. 


COLLECTING  AND  DISPLAYING 

earthenware  pots,  on  the  outside  to  begin 
with,  though  in  1559  it  was,  for  some  obscure 
reason,  transferred  to  the  inside.  Not  even 
"  London"  or  "  Made  in  London"  can  be 
accepted  as  absolute  evidence  of  provenance, 
as  has  been  remarked  before.  The  castle 
of  Edinburgh,  though  by  no  means  in- 
variably present,  seems  a  fairly  certain  mark 
of  nationality,  but  may  possibly  be  confused 
with  the  castle  used  at  Mons,  and  the 
double  eagle  and  half  eagle  between  the  flails 
of  Nuremberg  appear  to  have  been  almost 
the  only  marks  that  were  not  duplicated 
elsewhere. 

When  pewter  bearing  a  mark  affords 
such  shifting  and  uncertain  footing  to  the 
inquirer,  it  is  small  wonder  that  it  should 
present  a  very  quagmire  of  doubt  and 
hesitation  when  it  is  entirely  without  mark  of 
any  kind,  and  this  is  only  too  commonly  the 
case.  In  the  excellent  illustrated  "Cata- 
logue of  the  Pewter  Exhibition,"  organised 
by  Mr.  Massd  at  Clifford's  Inn  Hall  in 
1904,  the  words  "  Maker's  mark  none"  or  an 
equally  expressive  silence  occur  again  and 
again.  Out  of  fifty-nine  vessels  still  existing 
in  twenty-nine  churches  in  the  Diocese  of 

127 


OLD  PEWTER 

Llandaff,  as  recorded  in  Mr.  J.  E.  Halliday's 
published  "  Terrier,"  only  fifteen  are  marked, 
and  the  list  of  evidences  to  the  same  effect 
might  be  almost  indefinitely  extended,  but 
every  collector  of  pewter  knows  how  greatly 
unmarked  specimens  preponderate.  The 
reasons]  are  not  far  to  seek,  and,  though  in 
the  main  conjectural,  are  not  without  con- 
siderable antecedent  probability.  In  the  first 
place,  though  the  gold- and  silver-smiths  were 
compelled  to  carry  their  finished  wares  to  the 
Company's  Hall,  where  they  were  assayed 
and  stamped  by  the  proper  officials,  while 
any  evasion,  owing  to  the  intrinsic  value  of 
the  material,  would  certainly  be  pointed  out 
by  the  purchaser  and  promptly  and  severely 
punished,  the  pewterer  was,  in  most  cases, 
allowed  to  mark  his  wares  in  his  own  shop, 
subject  always  to  surprise  visits  by  the 
Company's  searcher,  and  as  the  customer 
would  doubtless  be  less  particular  as  to  the 
exact  quality  of  this  cheaper  metal,  avoidance 
of  the  obligatory  marks  must  have  been 
much  easier.  In  the  second  place,  it  is 
certain  that  the  use  of  pewter  would  have 
lingered  on  in  the  remoter  country  districts 
for  many  years  after  it  had  gone  out  of 
128 


be 

C  . 

w 


~  w 


So 
a 
W 


COLLECTING  AND  DISPLAYING 

fashion  in  London  and  other  large  towns, 
while  the  London  Company  lost  control  over 
the  provincial  makers  long  before  their 
authority  ceased  to  be  effective  in  the  City. 
Furthermore,  even  when  their  powers 
theoretically  extended  over  the  whole  of 
England  we  know  that  tinkers,  gypsies,  and 
other  travelling  pewterers  went  about  the 
country  melting  down  and  re-casting  worn- 
out  pewter,  in  spite  of  frequently  renewed 
edicts  against  them,  so  that  in  the  course  of  a 
century  or  two  very  little,  if  any,  of  the  ware 
can  have  retained  its  original  form  and 
markings.  Now  it  is  highly  unlikely  that 
these  unlicensed  workers  whom  the  Com- 
pany was  so  ceaselessly,  though  vainly,  en- 
deavouring to  suppress  would  have  stamped 
any  mark  which  might  lead  to  their  identifi- 
cation on  their  wares,  nor  would  the 
employer  be  keen  in  demanding  such,  and  as 
a  very  large  proportion  of  the  old  pewter, 
which,  now  that  the  demand  for  it  is  grow- 
ing, is  finding  its  way  into  the  market,  comes 
from  these  rural  districts,  the  general  lack  of 
marks  is  in  no  way  surprising. 

This  conclusion,  if  held  valid,  compels  us 
to  the  recognition  of  yet  another  stumbling- 

R  129 


OLD  PEWTER 

block  in  the  path  of  a  collector  anxious  to 
date  his  pewter,  for  these  pedlars  working,  as 
they  did,  surreptitiously  in  constant  fear  of 
detection,  and  in  a  necessarily  small  way, 
would  not  have  been  able,  even  if  they  could 
have  afforded,  to  procure  a  constant  supply  of 
new  moulds,  nor  could  they  well  have  carried 
about  with  them  an  extensive  variety  of  such 
bulky  and  ponderous  objects  without  attract- 
ing undesirable  attention  from  the  authorities 
of  any  town  they  passed  through,  and  it 
follows  that  their  productions,  in  most,  if  not 
in  all  cases,  must  have  been  antiquated  in 
style  to  an  unascertainable  extent.  In  the 
absence  of  marks,  however,  style  is  practically 
the  only  indication  of  time  and  place  of  manu- 
facture, and  when  it  is  probable,  considering 
the  durability  of  the  massive  metal  moulds, 
that  in  remote  country  neighbourhoods  vessels 
were  being  turned  out  in  shapes  a  century  or 
more  perhaps  behind  the  prevailing  taste  in 
London,  an  element  of  uncertainty  to  a  quite 
incalculable  amount  is  introduced  into  all  such 
deductions.  Nevertheless  uncertainty  is  pre- 
ferable to  absolute  ignorance,  and,  keeping 
well  in  mind  this  possible  source  of  error,  we 
must  continue  to  judge  the  period  of  a  vessel 
130 


COLLECTING  AND  DISPLAYING 

from  its  form.  There  is  much  yet  to  be  done 
in  accumulating  and  codifying  the  necessary 
materials  even  for  this,  and  only  vague  and 
general  indications  can  as  yet  be  attempted, 
taking  the  form  rather  of  rough  and  detached 
notes  than  of  any  connected  story  of  develop- 
ment. No  hard-and-fast  dogmatic  proposi- 
tions can  be  accepted.  Speaking  only 
generally,  then,  and  taking  as  understood  the 
"  possibly's  "  and  "  perhaps's  "  with  which 
every  paragraph  should  bristle  if  we  would 
keep  strictly  within  the  four  corners  of  allow- 
able hypothesis,  we  may  venture  to  assume 
that  the  greater  the  simplicity  the  greater  the 
age.  Straight  or  slightly  waved  lines  pre- 
ceded swelling  curves,  flat  unadorned  lids 
came  before  domed  tops  with  knobs  or  crests, 
few  and  simple  mouldings  were  the  forerunners 
of  many  and  elaborate  ones.  This  decorative 
evolution  as  displayed  in  the  thumb-pieces  of 
tavern-measures  made  in  Scotland  is  effec- 
tively described  and  illustrated  by  Mr.  Wood 
in  the  volume  so  often  quoted  before,  and  at 
the  same  time  he  lays  down  various  broad 
rules  of  the  greatest  importance  to  the  collector. 
For  example  :  that  a  vessel  with  the  lower 
part  of  the  handle  attached  directly  to  the 


OLD  PEWTER 

body  is  older  than  one  in  which  it  is  connected 
with  it  by  a  short  intervening  peg  or  strut ; 
that  the  Scottish  "quaigh"  differs  from  the 
porringer,  so  often  incorrectly  called  a 
bleeding-vessel  or  barber's  bowl,  by  the  fact 
that  the  first  was  round-bottomed,  the  second 
flat,  and  that  the  handles  or  "  lugs  "  were 
always  plain  and  solid,  not  pierced  or  otherwise 
ornamented ;  that  the  older  a  plate  is  the 
broader  is  the  rim  and  the  thicker  the  metal 
of  which  it  is  made,  though  this  refers  chiefly, 
if  not  wholly,  to  large  plates ;  that,  owing  to 
the  national  partiality  for  broth,  porridge,  and 
other  more  or  less  fluid  foods,  their  plates  were 
generally  deeper  than  those  used  in  England  ; 
and  that  up  to  1707,  and  less  regularly  up 
to  1835,  the  Scots  measures  differed  con- 
spicuously from  the  English,  the  former  being 
the  Scots  gill,  which  was  eight-tenths  of  the 
English  gill,  the  mutchkin,  which  held  three 
English  gills,  the  chopin,  containing  one  pint 
two  gills  English  measure,  the  Scots  pint, 
familiarly  known  as  a  "Tappit-hen  "  [Plate 
X.  2],  a  peculiarly  characteristic  type  of  vessel 
which  equalled  three  of  the  degenerate  English 
pints,  and  the  noble  Scots  gallon,  which 
absorbed  three  petty  English  ones ;  while  in 
132 


\ 


PLATE  XCVI 


i   L.  2Ty 

SNUFF-BOX.  No 
marks.  XVIII  cen- 
tuiy 


PLATE  XCVI  I 


2    L.  2J"  4     W.  ij" 

A  PAIR  OF  SHOE  BUCKLES.     The 
forks  of  hand-cut  steel.     No  marks. 

3    L   3i" 
SNUFF-BOX. 

No  marks. 


SNUFF-BOX, 
Scotch.     No 
marks. 


1     H.  5"  2     H.  3i"  3     H.  3i"  4    L.  3i" 

PEPPER-POTS,  English.         MUSTARD-POT,  Engl^h.  SNUFF-BOX 

XVIII  centuiy.     Earlv  Early  XVIII  century.  English.   Middle 

XIX  century.  XVIII  century. 


5     H.  aj" 

EGG-CUP, 
with  bead 
pattern, 
English. 
XVI 1 1  cen- 
tury. 


PLATE  XCVI  1 1 


INK-POT,  Scotch.     No  INKSTAND,  Italian.    With  ink-    TOBACCO-BOX, 

mark.  holder,  sand-box,  holes  for  two          marks, 

quills,  and  drawer  for  wafers. 


No 


COLLECTING  AND  DISPLAYING 

his  two  chapters  on  "  Church  Vessels  before 
and  after  the  Reformation"  he  provides  a 
wealth  of  information  as  to  the  various  shapes 
which  these  assumed  at  different  periods,  and 
the  distinctive  features  of  those  appertaining 
to  the  Presbyterian  and  Episcopalian  forms 
of  worship  respectively.  In  considering  style 
and  estimating  from  it  the  probable  date  of 
pewter  ware,  a  knowledge  of  the  contemporary 
work  of  the  gold-  and  silver-smiths  will  be 
found  of  the  greatest  assistance.  It  is  cer- 
tain that  the  pewterers  were  accustomed  to 
follow  with  considerable  closeness  the  designs 
in  favour  with  them,  doubtless  at  the  insti- 
gation of  their  customers,  who  desired  that 
their  table  furnishing,  though  cheap,  should 
be  in  the  mode.  More  especially  would  this 
seem  to  have  been  the  case  with  Church  plate, 
and  naturally  enough  only  those  parishes 
which  were  too  poor  to  afford  better  would 
content  themselves  with  the  baser  and  doubt- 
fully canonical  material,  and  they  would  have 
been  more  than  human  had  they  not  done 
their  best  to  conceal  from  stranger  eyes  the 
narrowness  of  their  circumstances.  To  this 
desire  to  keep  up  appearance,  quite  as  much  as 
to  a  sense  of  the  sacredness  of  the  vessels,  may 

133 


OLD  PEWTER 

perhaps  be  ascribed  the  precise  rules  as  to 
"scouring"  it  which  were  in  force,  one  of 
which,  quoted  from  Boissonet  by  Bapst, 
some  collector  may  be  inclined  to  try.  "  It 
must  be  washed  every  three  months  in  hot 
soap-suds,  and  be  rubbed  with  oats  or  other 
husk-bearing  grain,  or  with  broken  egg-shells; 
then  washed  in  clean  water,  dried,  and  wiped 
with  a  clean  cloth." 

Thanks  to  the  inclusion  in  the  hall-marks 
on  gold  or  silver  plate  of  a  letter  definitely 
denoting  the  year  of  manufacture,  it  is,  as  a 
rule,  easy  to  fix  the  date  of  these,  and  it  will 
be  a  fairly  safe  presumption  that  pewter  in 
the  same  fashion  is  of  somewhat  later  make. 
Coats-of-arms  engraved  upon  it  may  also  at 
times  afford  indications  of  the  date,  though 
these  must  always  be  regarded  with  some 
suspicion.  Even  if  genuine  there  can  be  no 
assurance  that  they  have  not  been  added 
later,  while  there  is  only  too  good  reason  to 
suspect  that  the  forging  of  them  is  becoming 
a  common  practice  among  some  unscrupulous 
dealers. 

To  the  young  collector  anxious  for  certain 
guiding  in  the  subject  he  has  selected  it  is  to 
be  feared  the  foregoing  remarks  will  prove 
134 


COLLECTING  AND  DISPLAYING 

not  a  little  disappointing,  but  to  such  a  one 
it  may  be  suggested  that  it  is  to  the  gradual 
increase  of  knowledge  accruing  from  it  that 
one  of  the  highest  pleasures  of  forming  a 
collection  is  due.  He  who  buys  wholesale, 
relying  solely  on  the  dictum  of  the  dealer  or 
the  authority  of  the  expert,  without  attempting 
to  arrive  at  any  opinion  founded  on  his  own 
judgment,  misses  one  of  its  chiefest  joys.  It 
is  better  to  make  a  few  mistakes,  provided 
that  these  are  not  too  costly,  and  to  profit  by 
them,  than  to  depend  in  ignominious  ignorance 
upon  the  real  or  assumed  experience  of 
another,  however  inglorious  a  security  this 
may  by  chance  result  in.  If  he  does  not 
appreciate  the  simple  beauties  of  good  pewter 
no  one  is  likely  to  be  tempted  to  possess  it, 
and  if  an  object  displays  this  unaffected 
beauty  to  its  owner's  eyes  it  is  not  a  matter  of 
very  great  importance  to  what  century  exactly 
it  may  belong,  always  presuming  that  its 
alleged  date  has  not  been  made  the  excuse  for 
extorting  a  purely  fancy  price. 

In  conclusion  a  few  lines  may  be  devoted 
to  indicating  briefly  some  of  the  articles  in 
pewter  which  the  collector  may  hope  to  find, 
though  it  would  be  tedious  to  attempt  a 

135 


OLD  PEWTER 

complete  catalogue  of  the  uses  to  which  it 
was  applied.  Plates  and  dishes,  flagons  and 
tankards,  measures  and  drinking  vessels  of 
diverse  forms  and  sizes  are  comparatively 
common,  as  are  candlesticks,  inkstands, 
porringers,  bowls  and  basins  of  various  sorts  ; 
while  among  the  less  frequent  objects  are 
articles  of  church  plate,  lavers,  chalices,  patens, 
baptismal  bowls,  fonts,  and  so  on,  guild  cups 
from  foreign  cities,  salt-cellars,  pepper  and 
mustard  pots,  tobacco  and  snuff  boxes,  candle 
boxes,  tea-caddies,  lamps,  some  with  glass 
reservoirs  and  a  divided  band  indicating  the 
hour's  consumption  of  oil  for  night  use,  spoons 
and  forks,  Scotch  communion  tokens,  beggars', 
porters'  and  other  badges,  tea  and  coffee  pots, 
punch  ladles,  colanders,  powder-puff  boxes 
and  money-boxes,  eggstands,  haggis-dishes, 
and  quaighs.  The  seeker  after  pewter  has  no 
need  to  fear  monotony  in  his  pursuit. 

When  the  collection  has  been  made  or  has, 
at  any  rate,  attained  a  sufficient  importance 
in  quantity  and  quality,  the  question  of  its 
disposition  so  as  to  display  its  decorative 
powers  and  aesthetic  features  in  the  most 
complete  and  characteristic  manner  becomes 
one  for  serious  consideration.  Beyond  all 
136 


o  s 


COLLECTING  AND  DISPLAYING 

question  the  happiest  results  are  secured  in 
those  few  but  fortunate  cases  where  in  old 
country  houses  the  vessels  made  for  and  used 
by  departed  ancestors  remain  in  their  original 
habitation,  either  still  spread  out  on  the  very 
shelves  and  dressers  primarily  designed  to 
receive  them  or  removed  with  loving  care  into 
equally  congenial  but  more  generally  acces- 
sible surroundings.  This,  however,  may  be 
rapidly  passed  over  as  an  usually  unattainable 
counsel  of  perfection.  The  principle,  never- 
theless, is  one  that  should  be  borne  carefully 
in  mind  when  the  subject  of  arrangement 
calls  for  attention.  It  must  never  for  one 
moment  be  forgotten  that,  leaving  out  of 
account  the  highly  ornamented  but  artistic- 
ally wrong  works  of  Briot,  Enderlein,  and  the 
like,  the  prime  intention  of  pewter  was 
practical,  every-day  usefulness,  that  the 
beauties  it  possesses  are  incidental  rather 
than  intentional,  arising  from  the  perfection 
of  its  construction  and  adaptation  to  purpose, 
and  that  consequently  any  setting  that  tends 
in  any  way  to  weaken  or  destroy  the  clear 
impression  of  its  maker's  meaning  will  be 
ipso  facto  a  bad  one.  In  short,  though 
devoting  it  in  reality  solely  to  purposes  of 

s  137 


OLD  PEWTER 

ornament,  we  should   endeavour  to  retain  a 
suggestive  reminiscence  of  possible  use. 

One  cannot,  of  course,  dogmatically  pre- 
dicate the  hidden  influences  working  on 
another's  mind,  but  one  may  tentatively 
assume  that  like  effects,  as  a  rule,  spring 
from  like  causes,  and  deduce  from  that  pos- 
tulate a  belief  that  few,  if  any,  true  lovers  of 
pewter  are  blind  altogether  to  the  large  part 
which  associations  play  in  the  fostering  of 
their  devotion.  It  is  out  of  its  very  home- 
liness that  pewter  weaves  one  of  its  most 
potent  spells.  From  the  world  at  large 
crowned  heads  and  belted  earls  are  too  far 
removed  to  greatly  stir  the  imagination.  The 
gorgeous  gold  and  silver  of  royal  feasts  and 
noble  revels,  flashing  beneath  the  electric 
lamps,  dazzle  more  than  they  delight.  The 
modest,  moonlight  sheen  of  pewter,  on  the 
other  hand,  blinking  demurely  on  the  oak 
dresser  in  the  dancing  firelight  of  a  winter's 
night,  speaks  to  us  of  ourselves.  We  may 
trace  it  back  in  fancy  to  the  shelves  of  some 
stout  and  prosperous  burgher,  long  since 
crumbled  into  dust  beneath  some  silent  city 
church,  to  the  stone-paved  house-place  of 
some  far-off  farm-house,  at  the  highest  to 

138 


COLLECTING  AND  DISPLAYING 

some  comfortable  country  manor  inhabited 
by  well-to-do  gentry,  for  we  prefer  to  keep 
aloof  from  the  clattering  scullions  in  the 
kitchens  of  the  great.  Or  giving  looser  reins 
to  the  Pegasus  of  our  musings  as  the  glow- 
ing coals  fall  together  and  the  flames  leap 
up  anew,  we  may  more  definitely  conceive 
our  mind-pictures,  and  bring  into  them  well- 
known  figures  from  the  distant  past.  That 
measure  may  have  stood  upon  the  board 
between  Shakespeare  and  Ben  Jonson  at  the 
Mermaid  Tavern  ;  from  that  tankard  Mar- 
lowe may  have  supped  his  last  draught  before 
the  footman's  sword  brought  to  nothing  his 
mighty  if  erratic  genius  ;  Dicky  Steele  may 
have  taken  that  candlestick  in  his  shaky  hand 
to  re-light  his  oft-extinguished  pipe  ;  and  why 
should  not  Robby  Burns  have  filled  a  mug 
from  that  "  Tappit  hen  "  ? 

But  to  feel  certainly  that  these  agreeable 
pictures  are  provable  facts,  we  must  have 
our  pewter  in  congenial  surroundings.  It 
cannot  be  denied,  indeed,  that  to  an  owner 
who  fully  realises  its  limitations  as  well  as 
its  capabilities,  pewter  is  to  no  small  extent 
exacting  in  its  demands.  It  possesses  so 
marked  an  individuality,  so  large  a  measure 

139 


OLD  PEWTER 

of  dignity  and  solidity,  that  it  will  combine 
harmoniously  with  few  of  the  ordinary  em- 
bellishments of  the  average  house.  No  one, 
it  may  be  taken  for  granted,  would  dream  of 
enshrining  it  in  the  fanciful  glass  cases  or 
on  the  delicate  velvet-covered  shelves  and 
niches  of  a  drawing-room,  or  of  forcing  it 
into  incongruous  propinquity  to  the  dainty 
subtleties  of  Dresden  china  or  Venetian  glass. 
Such  a  proceeding  would  be  too  flagrantly 
absurd  ;  yet,  setting  aside  such  extremes  of 
bad  judgment,  it  is  not  difficult,  with  the 
best  intentions,  to  fall  into  error  nearly  as 
serious ;  and  complete  success  will,  in  most 
cases,  be  found  in  keeping  it  apart  in,  as  far 
as  possible,  conditions  approximating  in  their 
simplicity  to  what  may  not  inaptly  be  termed 
natural.  We  cannot  all  command  for  it  stone 
floors,  wood  panelling  and  antique  furniture  ; 
nor  ought  we  to  expect,  even  if  we  could, 
any  one  but  an  enthusiast  so  far  to  sacrifice 
comfort  to  aesthetic  appropriateness ;  but  we 
may  at  least  spare  it  the  indignity  of  an 
anachronistic  connection  with  gilded  gim- 
cracks  and  modern  wall-papers.  The  charm 
of  well-wrought  and  well-designed  pewter  lies, 
in  the  main,  in  form  and  colour,  and  the 
140 


COLLECTING  AND  DISPLAYING 

fullest  expression  of  these  should  be  the  sole 
aim.  Isolation,  and  the  carefully  considered 
but  not  too  obviously  intentional  combina- 
tion of  sufficiently  numerous  examples,  will 
most  surely  secure  the  desired  result.  Plates, 
dishes,  and  other  shallow  objects  may  be 
arranged  effectively  and  justifiably  on  shelves 
or  ledges  of  proportionate  apparent  strength 
in  passages  or  ante-rooms,  the  wood  being 
either  naturally  dark  or  stained,  and  the  back- 
ground of  some  uniform  light  tint,  undis- 
turbed by  fretful  patterns,  in  harmony  with 
the  general  tone  of  the  material.  In  the 
selection  of  this  there  is  ample  scope  for 
personal  preferences,  though  it  will  be  found 
that  plain,  unpretentious  whitewash  is  hard 
to  beat.  Never,  under  any  circumstances, 
should  a  specimen  be  suspended  in  any  way. 
A  porcelain  plate  seemingly  adherent  to  the 
wall  is  ridiculous  enough ;  a  pewter  plate 
under  such  inane  conditions  is  an  eyesore. 
More  bulky  objects,  jugs,  tankards,  and  so 
on,  backed  by  plates  and  dishes  in  goodly 
array,  will  find  a  fitting  home  in  the  dining- 
room,  and  the  trouble  and  expense  will  be 
abundantly  repaid  by  the  enhanced  beauty  of 
the  exhibit  if  an  old  oak  or  walnut  dresser 

141 


OLD  PEWTER 

is  secured  for  their  reception,  and  smaller 
hollow-ware,  mugs,  porringers,  &c.,  are  hung 
on  hooks  along  the  front  edges  of  the  shelves. 
These,  with  candlesticks,  snuff-boxes,  and  the 
endless  variety  of  smaller  articles,  may  also 
be  well  disposed  upon  the  mantelpiece  and  on 
shelves  over  it ;  always,  however,  taking  good 
care  that  none  are  placed  too  high  above  the 
level  of  the  eye,  that  the  examples  are  neither 
so  widely  dispersed  as  to  suggest  specimens 
in  a  museum  nor  so  crowded  together  as  to 
interfere  with  the  spectator's  full  enjoyment 
of  the  lines  and  curves  of  each  individual 
object ;  and  lastly,  that  the  illumination  shall 
be  as  full  as  can  be  managed,  for  the  colour  of 
pewter  is  of  a  delicate  nature,  and  in  a  dim 
light  it  is  apt  to  take  on  a  dull  and  leaden 
hue. 

The  final  steps  before  installing  the  col- 
lection in  its  destined  place  are  the  cleaning 
and  the  repairing,  as  far  as  is  desirable,  the 
effects  of  time  and  ill-usage.  As  to  the 
extent  to  which  the  former  process  may 
legitimately  be  carried  there  may  be  no  little 
difference  of  opinion.  Undoubtedly,  when 
the  objects  first  came  fresh  and  fair  from  the 
maker's  workshop  they  were  as  bright  as 
142 


COLLECTING  AND  DISPLAYING 

silverware,  and  it  cannot  be  denied  that  a 
well-chosen  display,  cleaned  and  polished  to 
the  highest  point  of  brilliancy,  is  an  attractive 
sight.  Yet,  on  the  other  hand,  one  cannot 
altogether  unreluctantly  destroy  the  results 
of  the  mellowing  hand  of  time,  and  the 
problem  would  seem,  on  the  whole,  one  of 
those  that  is  best  left  to  individual  predilec- 
tion for  solution.  It  may,  however,  be 
remarked  that,  according  to  Mr.  Gowland, 
the  Japanese,  whose  supreme  attainments  in 
subtleties  of  taste  Western  nations  have  long 
since  learned  to  acknowledge,  never  attempt 
to  clean  or  polish  pewter  after  it  has  left  the 
fashioner's  hands,  confining  themselves  to  an 
occasional  rubbing  with  a  piece  of  silk  or 
cotton.  The  result  of  this  is  that  the  various 
slightly  differing  alloys  which,  as  has  already 
been  pointed  out,  develop  during  the  process 
of  cooling,  acquire  somewhat  different  tones 
in  the  course  of  the  slow  oxidisation  which 
they  undergo  when  exposed  to  the  air,  and 
the  consequent  mottling  of  a  light  and  darker 
patina,  when  well  marked  and  evenly  dis- 
tributed, is  a  highly  valued  and  much 
appreciated  feature  of  the  ware. 

With  regard  to  repairs,  unless  these  are 

H3 


OLD  PEWTER 

absolutely  necessary  it  is  better  as  a  general 
rule  to  refrain  from  the  attempt,  but  if  such 
must  be  undertaken  the  judicious  lover  of 
his  pewter  will  be  at  the  slight  pains  of 
learning  to  execute  them  himself.  The  use 
of  a  blow-pipe  and  soldering-iron  for  patch- 
ing up  holes  and  rents,  though  calling  for 
differences  of  manipulation  under  differing 
circumstances,  and  not  easy  therefore  to 
make  clear  in  words,  is  not  in  actual  practice 
difficult  to  acquire.  A  few  experiments  on 
objects  of  no  importance  will  soon  lead  to 
dexterity,  and  the  owner  of  some  neatly 
mended  treasure  will  feel  himself  well  re- 
warded should  he  subsequently  encounter  in 
some  rival  collection  the  ghastly  results  of 
confiding  such  a  delicate  operation  to  the 
clumsy  mercies  of  the  average  British  work- 
man. For  the  dents,  which  few  hollow 
vessels  altogether  escape  in  the  chances  and 
changes  of  life,  pressure  applied  artfully,  and 
above  all  very  gently  and  gradually,  inside 
with  the  round  knob  of  a  poker,  the  head  of 
a  curtain-pole,  or  even  a  wooden  spoon  will 
usually  be  found  effectual,  but  in  more 
obstinate  cases  it  may  be  needful,  after 
placing  a  hard  object  of  the  right  shape 
144 


COLLECTING  AND  DISPLAYING 

inside,  to  beat  very  lightly  with  a  curved- 
faced  hammer  outside  all  round  the  circum- 
ference of  the  bulge,  which,  if  done  with 
caution  will  eventually  restore  it  to  its 
former  level.  Slight  scratches  are  easily  re- 
moved by  rubbing  with  fine  emery-powder, 
but  no  attempt  should  be  made  to  eradicate 
the  deeper  knife-cuts  which  will  generally 
be  found  on  plates  and  dishes  that  have  been 
much  used.  It  is  not,  after  all,  advisable  to 
endeavour  to  render  old  pewter  indistinguish- 
able from  new. 

In  the  case  of  the  irregular  black  blotches 
and  stains  caused  by  oxidisation,  which  in- 
variably occur  upon  examples — often  the 
richest  prizes — which  have  been  brought  to 
light,  forgotten  and  long-neglected,  from  the 
obscure  recesses  of  a  broker's  shop  or  country 
cottage,  a  long  soaking  in  paraffin  is  the 
surest  and  safest  remedy,  for,  though  it  may 
be  done  more  rapidly  by  a  careful  treatment 
with  acids,  the  process  is  not  without  its 
inconveniences,  not  to  speak  of  dangers  to 
hands  and  garments. 

Finally,  when  all  that  is  needed  has  been 
done,  and  supposing  there  is  no  intention 
of  using  the  ware,  as  will  most  often  be  the 

T  145 


OLD  PEWTER 

case,  a  very  thin  coating  of  oil,  or  still  better, 
vaseline,  applied  to  the  entire  surface  of  each 
specimen  will  form  an  admirable  protection 
against  the  effects  of  the  atmosphere,  and 
will  greatly  reduce  the  frequency  of  necessary 
cleanings. 


146 


THE  TENTH  CHAPTER 

SOME  NOTES  ON  THE 
ILLUSTRATIONS 

FORKS  AND  SPOONS 

HESE  are  the  forms  in  which 
we  most  seldom  find  pewter  at 
the  present  day.  Indeed  the 
first  are  the  very  rarest  of  all 
objects.  When  gathering  speci- 
mens for  the  Exhibition  at  Clifford's  Inn,  Mr. 
Massd  was  only  able  to  secure  a  single  fork, 
and  an  inspection  of  many  collections  made 
for  the  purpose  of  selecting  the  materials  for 
the  illustration  of  this  volume  only  revealed 
one  further  example.  Nor  is  this  scarcity  in 
any  way  surprising.  It  is  not  probable  that 
forks  were  made  in  any  great  quantity  in  so 
unsuitable  a  metal.  Strength  and  sharpness 
are  the  fundamental  qualities  required  in  the 
prongs  of  a  fork,  and  neither  could  be  assured 
permanently  in  so  soft  a  medium.  The  points 
would  very  soon  get  blunt,  and  the  neces- 
sary slenderness  of  the  various  parts  would 
quickly  lead,  under  even  fair  wear  and  tear,  to 

H7 


OLD  PEWTER 

breakage.  The  one  we  reproduce  in  the  middle 
of  Plate  XI.  2  is  small  and  exquisitely  finished 
with  well-designed  ornament,  and  may  per- 
haps have  once  formed  part  of  a  dessert  service, 
to  which  use  alone  pewter  might  be  applic- 
able, but  it  seems  more  probable  that  it  is  in 
fact  one  of  those  trial  pieces  which  the 
goldsmiths  and  silversmiths  are  known  to 
have  made  in  pewter. 

Spoons,  though  decidedly  less  exceptional, 
are  very  far  from  common.  The  thinness  of 
the  stem  and  the  inevitable  point  of  weakness 
at  the  junction  of  this  with  the  bowl  rendered 
them  very  liable  to  damage  under  rough 
usage,  while  the  small  quantity  of  metal 
contained  in  each  would  reduce  the  probability 
of  any  attempt  at  repairs,  even  were  this 
practicable.  The  few  early  examples,  conse- 
quently, which  we  are  able  to  present,  dating 
from  the  fourteenth  and  fifteenth  centuries, 
are  made  of  bronze  or  laton,  a  form  of  brass, 
and,  though  not  properly  coming  within  our 
subject,  are  given  as  types  of  the  shapes  which 
were  no  doubt  adopted  also  by  the  pewterers. 
Of  fourteenth-century  make  are  Nos,  i  and  2 
[Plate  XII.]  in  laton  and  3  in  bronze ;  and  of 
the  fifteenth,  4  and  5  in  laton  and  6,  7  and  8 
148 


NOTES  ON  THE  ILLUSTRATIONS 

in  bronze.  There  are,  on  the  other  hand,  no 
bronze  or  laton  spoons  of  sixteenth-century 
make  in  this  collection,  all  the  twenty-four 
of  that  period  being  pewter.  The  bowl  in 
each  is  of  the  broadly  oval  shape,  charac- 
teristic also  of  earlier  examples,  which  differs 
so  conspicuously  from  our  present  spoons  in 
being  broadest  at  the  end  and  narrowing  up- 
wards towards  the  handle — obovate,  to  employ 
a  botanical  term — instead  of  tapering  down 
from  below  the  handle  towards  the  point. 
The  handles,  however,  present  a  considerable 
variety.  Most  of  them  belong  to  what  is 
known  as  the  "  Slip-top  "  pattern,  in  which  the 
stem,  which  has  straight  parallel  sides  and  is 
as  a  rule  only  slightly  flattened,  is  abruptly 
and  obliquely  truncated,  presenting  an  acute 
angle  at  one  edge  and  an  obtuse  one  at  the 
other,  as  if  it  had  been  cut  off  by  a  slanting 
blow  from  a  chisel.  Such  are  Nos.  i,  2,  3,  5, 
6  and  7  in  Plate  XIIL,  all  in  Plate  XV., 
and  Nos.  i  to  4  and  6  and  9  in  Plate  XVI. 
Of  these,  No.  6  Plate  XIIL,  and  Nos.  i,  2,  3 
and  6  in  Plate  XV.  all  have  the  same  maker's 
mark  with  pellets  above  and  below  in  a 
dotted  circle,  and  the  owner's  initial,  A,  and 
were  all  obtained  from  the  same  excavation 

149 


OLD  PEWTER 

in  York  Road,  Westminster.  The  remain- 
ing four  are  all  different;  one,  No.  6  in  Plate 
XIV.,  is  a  very  curious  Sacramental  spoon 
with  a  half-length  figure  on  a  rough  capital  at 
the  top  of  the  stem,  from  which  it  and  others 
of  the  same  form  obtain  the  name  of  "Maiden- 
head "  spoons  ;  a  second,  No.  5  in  the  same 
plate,  has  a  roughly  moulded  hexagonal  head 
with  a  flat  top,  hence  called  "Seal-headed"  ; 
the  third,  No.  8,  also  in  Plate  XIV.,  is  of  the 
familiar  "Apostle  "  design  ;  while  the  fourth, 
No.  i,  Plate  XIV.  again,  has  an  acorn  head. 
The  laton  spoons  again  preponderate  among 
those  of  the  seventeenth  century,  there  being 
no  less  than  twenty-eight,  of  which  only  those 
presenting  some  marked  peculiarity  or  forms 
not  found  in  the  pewter  specimens  need  be 
mentioned.  No.  4  Plate  XIV.  is  a  laton 
"Apostle"  spoon,  No.  8  Plate  XVII.,  No.  3 
Plate  XIV.,  Nos.  i,  2,  3,  4,  5,  6  and  7,  all  in 
Plate  XVII.,  while  No.  8  in  Plate  XX.  has  a 
sort  of  shallow  urn-shaped  top  which  is  quite 
peculiar.  An  equally  unusual  type  is  No.  7 
Plate  XX.,  which  has  a  very  deep  circular 
bowl  and  a  short  flattened  handle  ;  while  in  the 
large  gravy  spoon,  No.  2,  we  have  in  both 
the  bowl  and  the  so-called  "  dog-nose  "  handle 
150 


o  !  co 


NOTES  ON  THE  ILLUSTRATIONS 

a  distinct  tendency  towards  more  modern 
shapes,  and  in  No.  4,  both  in  the  same  plate, 
a  curious  pierced  ladle.  Of  the  pewter  spoons, 
No.  2  Plate  XIV.,  with  the  seated  lion  on  the 
handle,  is  noteworthy.  No.  2  Plate  XVIII. 
has  a  decidedly  broadened  and  flattened  slip- 
top  handle  of  the  type  known  as  "  Puritan," 
of  which  other  examples  in  laton  are  Nos. 
i,  3,  4  and  5,  all  in  the  same  plate,  the 
notches  on  the  head  of  the  last  perhaps  indi- 
cating the  first  beginnings  of  the  three-lobed 
broadened  top  known  as  "  Pied-de-biche,"  a 
further  approach  to  which  is  seen  in  No.  6 
in  the  same  plate.  The  subsequent  develop- 
ments of  this  idea  were  very  varied,  and  are 
well  illustrated  in  Plates  XVIII.  and  XIX. 
Thus  in  No.  7  Plate  XIX.  a  slight  deepening 
of  the  notches  gives  three  regularly  rounded 
lobes,  a  more  marked  deepening  in  No.  9 
Plate  XIX.  (a  highly  ornamented  pewter 
spoon) and  No.  8  Plate  XVIII.  gives  a  broad 
round  central  lobe  with  a  claw-shaped  projec- 
tion on  each  side.  In  No.  3  Plate  XIX.  these 
are  cut  off,  leaving  a  single  lobe.  In  No.  4  Plate 
XIX.  the  notches  are  shallow,  forminga pear- 
shaped  expansion,  and  in  Nos.  2,  5,  and  6  Plate 
XIX.  they  are  variously  modified.  Nos.  i 


OLD  PEWTER 

and  3  Plate  XX.  are  eighteenth-century  work, 
evidently  founded  on  the  contemporary  silver 
work,  and  differ  little  from  those  made  at  the 
present  day.  The  spoons  in  Plate  XI.  are  of 
the  nineteenth-century  make,  the  soup  ladle 
on  the  right  bearing  the  initials  W.  R.  flank- 
ing a  crown  and  the  name  ASH  BER  RY 
in  three  separate  stamps,  while  that  on  the 
left,  the  three  table-spoons  and  the  ten  toddy 
ladles,  are  marked  JO  HN  YA  TES  simi- 
larly divided  into  four,  and  a  crowned  V.R. 

CHURCH  VESSELS 

The  earliest  examples  of  these,  and  indeed 
the  most  ancient  objects  here  reproduced,  are 
the  three  Roman  vessels  from  Appleshaw  in 
Plate  III.,  but  as  these  have  been  fully  dis- 
cussed in  an  earlier  chapter  no  further  refer- 
ence to  them  need  be  made  here.  The  next  in 
point  of  age  is  the  hexagonal  pyx  on  the  left 
of  Plate  XXL,  with  the  Annunciation  and 
the  arms  of  England  and  France  on  the  lid, 
which  dates  from  the  fourteenth  century, 
while  the  two  patens  from  a  church  in  York- 
shire on  the  right  of  Plate  XXII.  3,  are  of 
fifteenth-century  make.  To  the  seventeenth 
belong  the  fine  English  Communion  flagon 
152 


NOTES  ON  THE  ILLUSTRATIONS 

on  the  left  of  the  same  plate  (i),  that  in  the 
centre  of  Plate  XXI.  2  which  bears  the  date 
1677  and  came  from  Midhurst  in  Sussex, 
and  probably  the  Dutch  earthenware  flagon 
mounted  in  pewter  in  the  centre  of  Plate 

XXIII.  2.     The  two  other  flagons  in  Plates 

XXIV.  3   and   XXV.,  one  with,  the   other 
without,  a  spout,  though  belonging   to  the 
earlier  flat-lidded  type,  are  eighteenth-century 
work,  and  to  the  same  period  may  be  attri- 
buted the  various  forms  of  chalice,  those  in 
Plates  XXIV.  2,  XXVI.  i  and  3,  and  XXVII. 
being    English,    that    on   the   left    of   Plate 
XXIII.  3,  and  those  in  Plates  XXVIII.  2  and 
XXIX.  i,  Scotch,  the  latter  dated  1804,  while 
that  on  the  right  of  Plate  XXIII.  i  is  Italian. 
An  uncommon  object  is  the  minute  pocket 
Communion  set  in  Plate  XXX.  i,  the  chalice  in 
which  is  only  two  inches  high,  with  its  roughly 
carved  wooden  case,  which  comes  from  Ice- 
land.     Rare  also  are  the  two  Communion 
cruets,  or  chapnets,  in  Plate  XXX.  2,  4,  of 
French  or  Flemish  workmanship,  their  re- 
spective uses  being  indicated   by  the   letter 
which  forms  the  handle  to  the  lid,  A  on  the 
left   for  Aqua,  V.  on  the  right  for  Vinum, 
of  which  last  a  second   and    more  graceful 

u  153 


OLD  PEWTER 

specimen  stands  on  the  extreme  left  of  Plate 

XXXI.  i.     The  portable  bdnitiers,  so  often 
mentioned  in  mediaeval  wills  and  inventories, 
are  well  represented,  that  in  Plate  V.  being 
adapted  for  suspending  only,   2,  3  and  4  in 

'  Plate  VI.  for  either  hanging  or  standing, 
while  i  in  the  same  plate  could  only  be  used 
on  a  table  or  altar.  They  are  all  of  Flemish 
make,  as  are  the  church  candlesticks  in  Plates 

XXXII.  i    and   3,  and  XXXIII.,  and  the 
candelabrum  in  Plate  XXXIV. 

DOMESTIC  CANDLESTICKS  AND  LAMPS 

Candlesticks,  being  easily  made  strong 
and  not  necessarily  subject  to  much  handling, 
are  comparatively  common,  and  do  not  as  a 
rule  present  any  marked  differences  from 
those  of  the  same  dates  made  in  brass  or 
silver.  Those  in  the  centre  and  on  the  left 
of  Plate  XXXVI.  i  and  2  are  Scotch,  the  pair 
on  the  right  of  the  same  plate  (3)  Dutch.  The 
larger  one  in  Plate  XXXVII.  3  is  French, 
and  the  little  taper-holder  on  the  left  of  it  (2) 
Flemish.  Lamps  are  more  rarely  found  in 
good  condition,  and  the  three  of  German 
eighteenth-century  make  in  Plate  XXXVII. 
i,  4  and  5  are  exceptionally  interesting, 
154 


NOTES  ON  THE  ILLUSTRATIONS 

though  still  more  so  are  those  with  glass 
reservoirs  in  Plate  XXXVIII.  In  these  the 
metal  band  which  supports  the  oil-holder  is 
divided  into  numbered  spaces,  by  means  of 
which,  as  the  lamp  burned  and  the  oil  sank 
down,  the  passage  of  the  night  hours  was 
roughly  indicated.  That  this  would  be  fairly 
accurate  in  the  flat  and  cylindrical  forms  (i, 
3  and  5)  is  easy  to  believe,  but  the  later  hours 
in  the  two  with  pear-shaped  reservoirs  (2  and 
4)  must  have  appeared  to  have  sped  much 
more  rapidly  than  the  earlier  ones,  since,  at 
any  rate,  no  alteration  has  been  made  in  the 
distances  apart  of  the  divisions  to  compen- 
sate for  the  reduced  bulk  of  oil  contained 
between  each  two. 

CUPS  AND  TANKARDS 

The  line  of  demarkation  between  these 
two  and  between  the  second  and  flagons  is 
not  always  an  easy  one  to  draw,  but,  speaking 
broadly,  the  first  may  be  defined  as  drinking- 
vessels  with  two  handles  or  none,  the  second 
as  one-handled  vessels  intended  to  be  drunk 
out  of,  and  the  last  as  similar  vessels  destined 
to  hold  liquors  which  were  poured  into  others 
for  the  consumer.  Cups  may  be  conveniently 

155 


OLD  PEWTER 

divided  into  two  classes,  those  with  stems, 
and  those  without.  The  former,  with  the 
exception  of  a  certain  type  to  be  considered 
hereafter,  are  rarely  met  with  in  pewter,  and 
when  they  do  occur  are  generally  those  direct 
imitations  of  silver-work  which  have  perhaps 
been  more  than  sufficiently  deprecated  already, 
and  there  is  consequently  no  necessity  for 
pointing  out  the  constructional  defects  in  the 
four  cups,  light  and  graceful  as  they  are 
in  design,  represented  in  Plate  XLI.  The 
stemmed  cups,  mostly  made  in  Germany  for 
the  use  of  various  trade  guilds,  are  less 
uncommon,  and  examples  are  to  be  found 
in  most  large  collections.  Like  much  other 
German  work,  they  are  apt  to  show  a  tendency 
to  exaggeration  of  outline  and  over-redun- 
dancy of  ornament,  but  the  simpler  forms, 
such  as  those  shown  in  Plates  XXXII.  2  and 
XLI  I.  i  and  3,  are  not  without  a  certain 
dignity.  A  beautiful  example  of  the  stemless 
type  is  the  little  English  cup  on  the  left  of 
Plate XXVIII.  i.  The  "beaker"  or  "tumbler" 
shape,  illustrated  in  Plates  XLIII.  2  and 
XLIV.  4,  is  far  commoner,  these  having  been 
largely  imported  from  the  Netherlands  into 
Scotland,  where,  after  the  Reformation,  they 

156 


NOTES  ON  THE  ILLUSTRATIONS 

seem  to  have  been  generally  used  as  Com- 
munion cups. 

Tankards,  again,  may  be  considered  as 
falling  into  two  categories,  those  with  lids 
and  those  without.  The  most  usual  form 
of  the  first  is  that  with  straight  sides  more 
or  less  relieved  by  hoops  and  bands,  as  in 
the  one  on  the  right  of  Plate  IX.  3,  the  four 
in  Plate  XLVL,  and  one  in  the  centre  of 
Plate  XLVII.  2,  or  slightly  conical  or  curved, 
as  in  Plates  XLIII.  i  and  3,  and  XLVII  I.  3, 
but  from  these  simple  lines  we  find  every 
graduation  to  the  full,  bold  curves  of  the 
"balustrade"  type,  illustrated  in  the  middle 
of  Plate  L.  3,  or  such  fantastic  designs  as 
the  German  guild  tankard,  dated  1645,  on 
the  right  of  Plate  XL.  i.  Simplicity,  as  a 
rule,  marks  the  lidless  tankard,  characteristic 
examples  of  which  are  shown  in  Plates  XLI V. 
i,  XCIII.  2,  and  CIII.  3. 


157 


OLD  PEWTER 

MEASURES 

The  segregation  of  these  is  often,  it  must 
be  owned,  somewhat  arbitrary,  since  they 
graduate  indefinitely  into  tankards  on  the 
one  hand  and  flagons  on  the  other,  for 
measures  such  as  the  Scotch  half-mutchkin, 
gill,  half-gill,  and  half-glass  [Plates  XXXI. 
3,  2,  6,  5,  and  4,  LV.  and  LVL],  and  the 
English  one  which  stands  second  from  the  right 
hand  in  Plate  XLIV.3,were  undoubtedly  used 
as  drinking-vessels,  while  the  flagon-shaped 
one  on  the  left  of  it  (2),  and  those  in  Plates 
IX.  i,  LVII.  2  and  3,  and  LVIII.  5,  are  equally 
well  adapted  for  filling  round.  Most  inter- 
esting among  these  last  are  the  Tappit  hens 
previously  mentioned,  examples  of  which  are 
given  in  Plates  X.  2,  LIX.  and  LX.,  the  last 
in  especial  being  of  unusual  importance.  Mr. 
Ingleby  Wood  could  only  rely  upon  tradition 
for  the  fact  that  hot  drinks  were  brought 
out  in  these  vessels  to  the  passengers  upon 
coaches  at  the  changing-places,  and  served 
round  in  a  small  pewter  cup,  but  we  have 
direct  proof  in  the  existence  still  in  its  place 
of  the  cup  fitting  into  the  throat  of  the 
vessel. 

158 


NOTES  ON  THE  ILLUSTRATIONS 

FLAGONS  AND  JUGS 

There  is  probably  more  diversity  of  shape 
and  size  in  these  than  in  any  other  class  of 
vessels,  and  many  of  them  are  remarkable 
for  their  unforced  beauty  of  line  and  propor- 
tion. It  would  be  hard,  for  instance,  to 
improve  upon  those  reproduced  in  Plates 
LVIII.  2  and  LXIL,  both  of  which  are 
English,  the  former  eighteenth,  the  latter 
seventeenth  century.  The  German  taste 
was,  as  usual,  more  ornate,  but  there  is  much 
dignity  in  the  large  guild  flagons  in  Plates 
LI.,  LIII.,  LXIII.  3,  and  XLII.  2. 

PORRINGERS 

These,  which  are  frequently,  and  generally 
erroneously,  called  bleeding-bowls  or  barber's 
bowls,  were  very  commonly  in  use  in  the 
Middle  Ages,  and  great  numbers  of  them 
still  exist.  The  ornamentation  of  the  bowl 
when  present  was  generally  confined  to  a 
Tudor  rose  in  relief  on  the  bottom,  as  in 
Plates  LXIX.  2  and  LVII.  2,  but  a  consider- 
able freedom  of  treatment  was  allowed  in  the 
handles.  In  some  cases  these  were  quite 
plain,  or  with  a  slightly  curved  outline  [Plate 

159 


OLD  PEWTER 

LXX].  In  the  one  on  the  left  of  Plate  LXIX.  i 
the  border  is  deeply  crenated,  in  that  on  the 
right  of  the  same  plate  and  in  those  in  Plates 
LVII.  i  and  LXXI.  i  and  3  the  ears  are 
elegantly  and  elaborately  pierced,  while  in 
Plate  XXVIII.  3  we  have  a  rare  instance  of 
solid  ears  with  a  pattern  in  relief. 


PLATES  AND  DISHES 

Beyond  doubt  these  are  the  commonest 
objects  which  will  come  under  the  collector's 
notice,  ranging  in  point  of  date  from  the 
Appleshaw  Roman  pewter  [Plates  I.  and  II.] 
up  to  the  present  day,  and  in  size  from  a  few 
inches  in  diameter  to  two  feet  or  more  [Plate 
LXXII.  2  and  3].  The  rims  of  the  plates 
were  originally  quite  plain,  or  strengthened 
by  a  plain  band  below.  This  in  later  days 
was  placed  above,  and  frequently  adorned 
with  mouldings,  as  in  Plates  XXII.  2, 
LVIII.  4,  LXXII.  2,  and  LXXIII.  3,  while 
in  the  eighteenth  century  the  rim  was  often 
broken  into  curves  and  the  border  variously 
beaded  or  moulded  in  imitation  of  silver 
work  [Plates  IX.  2,  and  LXXIII.  i  and  5]. 
The  surface  also  presented  temptations  to 
1 60 


NOTES  ON  THE  ILLUSTRATIONS 

the  would-be  decorator,  as  shown  in  Plates 
XXXV.  and  XLVII.  i. 

TABLE  PLATE 

Practically  every  article  used  in  the 
serving  or  consuming  of  food  has  been  at 
one  time  or  another  manufactured  in  pewter. 
We  have  soup-tureens  [Plate  LXXV.],  vege- 
table dishes  [Plates  LXXII.  i  and  4,  and 
LXXVI.  i],  ewers  and  basins  for  washing  the 
hands  [Plate  LXXVIL1,  and  bowls  for 
washing  the  dishes  [Plate  LXXVIII.].  Salt 
cellars,  pepper  and  mustard  pots  appear  in  a 
variety  of  forms  too  large  to  specify  in  detail 

([Plates  LXXX.  to  LXXXIV.  and  others],  but 
note  must  be  made  of  a  curious  kind  of 
vinegar  sprinkler  of  Scottish  origin  which 
exactly  resembles  in  its  construction  the 
small  spirit  lamps  often  handed  to-day  to 
after-dinner  smokers  [Plates  LXXXIII.  4  and 
LXXXV.  6] .  When  tea  came  into  general  use 
pewter  was  devoted  to  its  service.  The  herb 
was  kept  in  pewter  bottles  [Plate  LXXXVI. 
2  and  4],  and  urns  [Plate  LXXXVIL],  tea 
and  coffee  pots,  milk  jugs,  and  sugar  bowls 
[Plates  LXXXVIII.  &c.]  replaced  the  earlier 
beer  jug  and  tankard  at  the  breakfast  table. 

x  161 


OLD  PEWTER 

Even  the  matutinal  egg  appeared  in  a  pewter 
cup  [Plate  XCI.],  and  the  whole  was  some- 
times brought  in  on  a  pewter  tray  [Plate 
LXXVI.  2]. 

MISCELLANEOUS  OBJECTS 

The  number  and  variety  of  these  is  almost 
infinite.  Boxes  to  hold  tobacco  [Plates  LXX I V. 
6,  XCIV.  3,  and  XCV.  3]  and  snuff  [Plates 
XCVI.  i  and  3,  and  XCVII.  4]  for  the  master, 
spices  [Plate  XXIX.  2],  powder  or  patches 
[Plate  LXXXVI.  3],  sugar  [Plate  XC.  2]  or 
other  trifles  for  the  mistress  [Plates  XL.  3,  and 
XC.  2],  inkpots  of  diverse  sorts  [Plates  XCV.  2, 
XCVIII.  i  and  2,  and  XCIX.  2],  shoe  buckles 
[Plate  XCVI.  2  and  4,  and  buttons  [Plate  VII. 
3]  were  only  a  tithe  of  the  fancy  articles  turned 
out  by  the  trifler.  In  the  East  pewter  was 
even  allowed  the  honour  of  adorning  weapons 
[Plate  C.]. 

A  puzzle  which  few  people  succeed  in 
solving  unaided  is  presented  by  the  object 
in  the  centre  of  Plate  LVIII.  3.  It  is  a 
round  vessel  pierced  with  holes  and  closed  by 
a  screw  lid  with  a  twisted  handle,  and  was, 
and  doubtless  still  is,  used  by  the  French 
peasant  women  to  keep  the  rice  apart  from 
162 


NOTES  ON  THE  ILLUSTRATIONS 

the  meat  and  broth  with  which  it  was 
boiled.  Another  puzzle  which  still  awaits 
solution  is  the  exact  purpose  to  which  the 
screw-top  bottles  in  Plate  CI.  i  and  3  were 
devoted.  They  are  generally  six-sided, 
though  the  one  on  the  left  of  our  plate  is 
flattened,  and  in  the  catalogue  of  the  Clifford's 
Inn  Exhibition  are  called  food-bottles  or  food 
carriers,  but  this  title  seems  to  be  merely 
speculative,  and  there  is  yet  opportunity  for 
some  pewter-lover  to  tell  us  what  they  really 
were. 


163 


USEFUL  BOOKS  OF  REFERENCE 

History  of  the  "Worshipful  Company  of  Pewterers  of 
the  City  of  London.  Based  upon  their  own  Records. 
By  CHARLES  WELCH,  F.S.A.  Two  vols.,  410.  (1902.) 

Pewter  Plate.  A  Historical  and  Descriptive  Handbook.  By 
H.  J.  L.  MASSE,  M.A.  (1904.) 

Scottish  Pewter- Ware  and  Pewterers.  By  L.  INGLEBY 
WOOD.  (1905.) 

Les  Metaux  dans  1'antiquite  et  au  moyen  age.  L'Etain. 
Par  GERMAIN  BAPST.  Avec  n  planches,  &c.  Paris.  (1884.) 

Old  Scottish  Communion  Plate.  By  THOMAS  BURNS, 
F.R.S.E.  With  a  Preface  by  the  Right  Rev.  J.  MACGREGOR 
.  .  .  and  Chronological  Tables  of  Scottish  Hall-marks,  pre- 
pared by  A.  J.  S.  BROOK.  Illustrated.  (1892.) 

Dictionnaire  de  I'Ameublement  et  de  la  Decoration 
depuis  le  xiiie  siecle  jusqu'a  nos  jours.  Par  HENRI 
HAVARD.  Four  vols.,  410.  Paris.  (1887-90.) 

Histoire  du  Mobilier :  recherches  et  notes  sur  les  objets 
d'art  qui  peuvent  composer  rameublement  et  les 
collections  de  1'homme  du  monde  et  du  curieux. 
Par  ALBERT  JACQUEMART.  8vo.  Paris.  (1876.) 

Fran£ois  Briot,  Caspar  Enderlein  und  das  Edelzinn. 
VON  HANS  DEMIANI.  4to.  Leipzig.  (1897.) 

The  Perth  Hammermen  Book.  With  an  Introductory 
Sketch  by  COLIN  A.  HUNT  of  Perth.  4to.  (1889.) 

The  Church  Plate  of  the  County  of  Northampton.  By 
C.  A.  Markham.  Illustrated.  8vo.  (1894.) 

The  Church  Plate  of  the  County  of  Dorset.  With 
Extracts  from  the  Returns  of  Church  Goods  by  the 
Dorset  Commissioners  of  Edward  VI.  By  the  Rev.  JAMES  E. 
NIGHTINGALE.  (1889.) 

The  Church  Plate  of  the  County  of  Wilts.  Including 
that  part  of  the  County  now  in  the  Diocese  of  Gloucester 
and  Bristol.  From  Returns  made  by  the  Rev.  J.  E.  NIGHT- 
INGALE and  the  Rev.  E.  H.  GODDARD.  (1891.) 

The  Church  Plate  of  the  County  of  Norfolk.  By  the 
Rev.  J.  E.  NIGHTINGALE. 

An  Inventory  of  the  Church  Plate  of  Leicestershire, 
with  some  Account  of  the  Donors.  With  Plates.  By 
ANDREW  TROLLOPE.  (1890.) 

Elizabethan  England.  From  a  Description  of  England  by 
William  Harrison  in  "  Hollinshed's  Chronicles."  Edited  by 
LOTHROP  WITHINGTON.  With  an  Introduction  by  F.  J. 
FURNIVALL.  "Camelot  Classics."  (1886.) 


INDEX 


ABERNBTHIE,  WILLIAM,  of  Edin- 
burgh,  fined,    1652,   for  using 
bad  metal,  103 
"Acorn -head"     pattern.       See 

Spoons 
Allen,     Thomas,     forbidden     to 

assist  a  new  invention,  103 
Alloy : 

Proportions  variable,  8  ;  pro- 
portions in  Roman  and 
English  pewter,  8-11  ;  pro- 
portions legalised,  11-14; 
proportions  in  1350,  13  ; 
object  of  regulations,  14; 
inconsistencies  in  analysis, 
1 5  ;  discussed  at  the  Society 
of  Arts,  ib.  ;  proportions  not 
of  practical  importance  to 
the  collector,  17  ;  simple 
method  of  analysis,  18-23  > 
precautions  taken  in  assay- 
ing the  metal  of  the  worker, 
29  ;  purchase  by  night  for- 
bidden, 32  ;  alloy  resem- 
bling gold  made  by  a  Nu- 
remberg craftsman,  82 
Alms  Dish.  See  Church  Vessels, 

&c. 

Amadel,  Robert,  reference,  91 
Altar  Candlesticks.     See  Church 

Vessels,  &c. 

Anglo-Saxon  work,  specimens  in 
the  Guildhall  Museum,  Lon- 
don, 52 

Antimony,     its     proportions     in 
some    kinds    of    pewter,     1 1  ; 
separation    by    analysis    only 
possible  to  a  trained  chemist, 
23  ;   references,  12,  28,  32 
Apostle  spoons.     See  Spoons 
Appleshaw,  discovery  of  Roman 
specimens     at,     45,    52,    160. 
See  plates  i,  n,  in 


Apprenticeship,  the  "  essay " 
and  subsequent  registration  of 
private  "  touch,"  26,  27 

Aquae  Neriae,  analysis  of  fourth- 
century  specimens  found  at, 
10 

"  Archaeologia,"  inventory  of  the 
contents  of  a  house  at  Gilling- 
ham,  quoted,  87  ;  references, 
9,46 

Arnold,  Henrici,  of  Ilsington,  a 
donor  of  Church  pewter,  1641, 
100 

Arthur  Tudor,  reference,  90 

"  Arts, Journal  of  the  Society  of," 
discussion  on  alloys,  quoted,  8, 

15 
Ashberry,  his  mark,  plate  xi,  iv  ; 

14,  52 

Ath,  reference,  57 
Augsburg,  reference,  57 
Avesne,    Jeanne    d',    a    French 

craftsman,  1516,  74 

BAPST,  M.  Germain,  on  Roman 
alloy,  10  ;  references,  63,  71, 

J34 

Barber's  Bowl.     See  Bowls 

Barry,  Mrs.  F.  W.,  rose-water 
dish  in  her  possession  pro- 
vided for  Charles  I.,  102 

Battersea,  composition  of  Roman 
metal  found  at,  n 

Beaker.     See  Bowls 

Beer-Jugs  : 

English  ;    eighteenth  century ; 

plate  LXVI,  iii,  94 
Students'  Beer-Jug  ;  German  ; 
seventeenth   century  ;    plate 
LXXIII,  ii,  104 

Belgian  Pewter  :  Bruges  crafts- 
men noted  for  porringers  and 
flasks,  55  ;  the  guild  at  Ath  in 


165 


OLD  PEWTER 


Belgian  Pewter — cont. 

1328,  57  ;  Mons  a  centre  in 
I353»  58  ;  craft  well  supported 
at  Ghent,  ib.  ;  Bruges  crafts- 
men support  the  town  militia, 
J376,  59  ;  Mons  marks  in 
1467,  67 
Benitier.  See  Church  Vessels, 

&c. 

Bermondsey,    spoons    found    at, 
See  plates    xin,    i,    18  ;     xiv, 
v,  20  ;    xv,  viii,  21  ;  xvi,  i,  iii, 
iv,    v,    vi,    22 ;  xvin,    ii,    26  ; 
xix,  iv,  v,  28  ;   xx,  iii,  29 
Boileau,  Etienne,  his  account  of 
the  Paris  guilds  in  thirteenth 
century,  alluded  to,  55 
Boissonet,  1'Abbe,  his  method  of 
"  scouring  "     quoted     by     M. 
Bapst,  reference,  1 34 
Bourbon,     the     Duke     de,     his 

pewter  vessels,  1507,  73 
Bowls : 

Barber's  Bowl,  English  ;  eigh- 
teenth century  ;  plate  LXIX, 
i  ;  100,  160 

Beaker,  engraved  ;  early  eigh- 
teenth century  ;  mark,  a 
crowned  rose  ;  plate  XLIV, 
iv;  64,  156 

Beaker,    Scotch  ;     eighteenth 
century  ;      plate    XLIII,     ii  ; 
62,  156 
Bowls  ;     plate    LXXII,    i,    iv  ; 

103,  161 

Bowl  ;  marks,  crowned  rose 
and  W.R.W.  ;  plate  LXXI, 
iii  ;  101,  160 

Bowl,  Flemish  ;  seventeenth 
century  ;  plate  LXXVIII,  i  ; 
112,  161 

Bowl,   Scotch  ;    mark  on  one 
lug     indecipherable  ;      plate 
LXXI,  i  ;    101,  160 
Two-handed  bowl  ;   plate  LVIII, 

i  ;    82 

Dutch  ;  eighteenth  cen- 
tury ;  plate  LXIX,  ii  ;  100, 
159 

1 66 


Box,  Edward,  London,  disre- 
gards advertising  rule,  115 

Box  ;  no  mark  ;  plate  LXXXVI, 
iii  ;  118,  162 

Briot,  Fran9ois,  references,  3,  42, 

79,  95>  137 

Britannia  metal,  &c.,  aids  the 
downfall  of  pewter,  117 

British  Museum,  Decorated  speci- 
mens at,  80  ;  references,  45,  48; 
see  also  plates  i,  n,  in 

Brown-Austen,  Professor,  refe- 
rence, 15 

Brown,  George,  Perth  tinsmith, 
applies  for  relief,  114 

Bruges,  references,  32,  55,  59 

Bryan,  Mr.  H.,  specimen  in  his 
possession  made  by  the  Master 
of  the  Company,  123 

Burgundy,  Duke  of,  establishes 
Guilds,  1478,  68  ;  his  pewter 
more  decorative,  1507,  73 

Burns,  Robert,  reference,  139 

Burton,  spoon-maker,  1686,  un- 
willingly allowed  to  use  a 
lathe,  36 

Buschius  of  Hildesheim,  his 
record  of  Saxon  pewter,  1470, 68 

Buttons,  Two  ;  plate  vn,  iii  ; 
10,  162 

CAMPBELL,    PATRICK,    reference, 

"3 

Candelabrum,  Flemish  ;  eigh- 
teenth century  ;  plate  xxxiv  ; 

52,  154 
Candlesticks  : 

Candlestick,  French  ;  eigh- 
teenth century  ;  plate 
xxix,  iii;  44,  154 

French  ;     plate    xxxvu. 

iii;    56,  154 

Candlesticks  ;  plates  xxxix, 
i,  ii,  iii ;  60,  154 

eighteenth  century ;  plates 

XL,  i,  iv  ;    60,   1 54 
Dutch;    eighteenth    cen- 
tury ;    plate  xxxvi,  iii  ; 
54,  154 


INDEX 


Candlesticks — cont. 

Scotch  ;   eighteenth  cen- 

tury ;    plate    xxxvi,  i  ; 

54,   154 

Scotch  ;    nineteenth  cen- 

tury ;    no  marks  ;   plate 
xxxvi,  ii;    54,  154 
Taper-Holder,   Flemish  ;  plate 

xxxvn,  ii;    56,  154 
Gloucester    Candlestick,    The, 
twelfth  century  ;    plate  iv  ; 

5,  54 

Canterbury,  Richard,  Archbishop 
of,  reference,  53 

Capdeville,  Pierre  de,  of  Bor- 
deaux, extracts  from  inven- 
tory, 1591,  88 

"Cardinal's  hatte  "  mould,  35 

Carel,    a    Nuremberg    pewterer, 

1324,  57 

Casket,  French ;  sixteenth  cen- 
tury ;  plate  LXXI,  ii  ;  101 

Cassiterides,  the  "  Tin  Islands  " 
mentioned  by  Herodotus,  31 

Catherine  of  Arragon,  reference, 
90 

Cavendish,  his  account  of  Wol- 
sey's  plate,  91 

Chabrolles,  Mark  Henry,  a 
Frenchman,  refused  permis- 
sion to  trade  in  London.  1688, 
106 

Chalice.     See  Church  Vessels 

"  Chapnets."  See  Church  Ves- 
sels," &c. 

Charles  II.,  references,  104,  108 

Charles  V.,  reference,  69 

Charles  VI.,  reference,  60 

Charles  VII.  of  France,  his  ex- 
tensive purchase  in  1422,  64 

Charterhous,  John,  refuses  to 
restore  two  candlesticks  at 
Edinburgh,  1559,  81,  82 

Chichely,  Robert,  Lord  Mayor  of 
London,  ordains  the  use  of 
pewter  by  ale-dealers,  1423, 
64 

Chinese,  conjecture  as  to  the  in- 
vention of  pewter  by  the,  44 


Chopin,  a  Scottish  measure,  61, 

132 
Church    Flagon.        See    Church 

Vessels,  &c. 
Church  Vessels,  &c.  : 

Alms  Dish ;  Scotch  ;  eigh- 
teenth century  ;  mar- 
ked AS.  IK;  plate 
xxii,  ii  ;  32,  160 

English  ;  eighteenth  cen- 
tury ;  plate  xxiv,  i  ;  36 

German  ;  early  eigh- 
teenth century  ;  plate 
xxxv,  53,  161 

Altar  Candlesticks  ;  eigh- 
teenth century  ;  plate 
xxxii,  i,  iii  ;  48,  154 

Flemish ;      seventeenth 

century ;    plate   xxxin, 
i,  ii  ;   50,  154 

Benitier,  probably  Flemish ; 
no  marks  ;  plate  v  ;  6,  57, 

154 

Benitiers,  Flemish  ;  eigh- 
teenth century  ;  plate  vi, 
i,  ii,  iii,  iv;  8,  57,  154 

Chalice  and  Paten,  from  Kris- 
wick  ;  plate  xxi,  iii  ;  30 

Chalice,  Italian  ;  no  marks  ; 
plate  xxm,  i  ;  34,  153 

Scotch  ;  no  marks  ;  plate 

xxin,  iii ;  34,  153 

English ;  eighteenth  cen- 
tury ;  plate  xxiv,  ii  ;   36,  153 

Chalices  and  Church  Flagon  ; 
English  ;  plate  xxvi,  i,  ii, 

iii  ;    38,   153 

Chalices,  •  English  ;  eigh- 
teenth century  ;  plate 
xxvn,  i,  ii,  iii;  40,  153 

"  Chapnets,"  small  church  ves- 
sels, 95,  153 

Church  Flagon,  Scotch  ;  early 
eighteenth  century  type  ; 
plate  xxv  ;  37,  153 

Communion  Cup  ;  Scotch  ; 
plate  xxvin,  ii  ;  42 

1804  ;   plate  xxix,  i  ;  44, 

153 

167 


OLD  PEWTER 


Church  Vessels,  &c. — cont. 

Communion  Flagon  ;  English  ; 
from  Midhurst   Church, 
Sussex  ;      frontispiece  ; 
plate  xxi,  ii  ;    30,  153 
Flagon  ;  English  ;  seven- 
teenth   century  ;     plate 
xxii,  i  ;    32,  153 
Jug,  earthenware,  moun- 
ted    in     pewter  ;        mark : 
angel       standing       holding 
crossed  branch  ;    letters  in- 
distinct ;      plate    xxm,    ii  ; 

34, 153 

Laver,  Scotch  ;  late  eighteenth 
century  ;  plate  xxiv,  iii  ;  36 

J53 

Patens,  from  a  Church  in 
Yorkshire,  England  ;  fif- 
teenth century  ;  plate  xxii, 
iii;  32,  152 

Pocket  Communion  Service,  in 
wooden  case,  bought  in 
Iceland,  probably  Danish 
or  Scotch ;  plate  xxx,  i  ; 

45,  153 

Pyx,      hexagonal  ;      English  ; 
fourteenth    century ;     plate 
xxi,  i  ;    30,  152 
Sacramental     Cruet ;     Aqua ; 
French      or      Flemish  ; 
plate  xxx,  ii;  45,  153 

Cruet  ;    Vinum  ;    French 

or  Flemish  ;   plate  xxx, 

iv;  45»  !53 

Spoon,    with    "  Maiden- 

head "  top,  found  in 
London  ;  sixteenth  cen- 
tury ;  plate  xiv,  vi ;  20, 
150 

Church,  utilisation  of  pewter  for 
the  ;  early  records,  52  ;  its 
use  forbidden,  1175,  53;  its 
exclusion  found  impracticable 
in  France,  54  ;  vessels  in  use 
at  St.  Stephen's,  Coleman 
Street,  London,  1467,  67  ; 
vessels  at  the  Cistercian  Con- 
vent of  St.  Martin,  68  ;  at 

1 68 


Church — cont. 

the  Convent  of  St.  Cyr,  ib. 
very  little  recorded  at  the  dis 
solution  of  the  monasteries, 
76  ;  Chrismatory  and  pyx 
bought,  1554,  by  Waltham 
Abbey,  80  ;  the  plate  of  St. 
Giles,  Edinburgh,  removed  for 
safety,  1559,  81  ;  disappear- 
ance of  much  pre-Reformation 
plate  in  Scotland  in  1559,  82  ; 
destruction  of  church  plate 
during  the  sixteenth  century, 
92  ;  purchase  at  Strood,  1607, 
94  ;  safe  to  assume  that  no 
pewter  Church  flagons  are 
earlier  than  1603,  94  ;  in- 
scription on  vessel  at  Werring- 
ton,  95  ;  effects  of  the  revolu- 
tion of  1617  in  the  Church  of 
Scotland,  97  ;  first  appear- 
ance of  the  "  laver,"  ib.  ;  ear- 
liest examples,  152  ;  English, 
Scotch,  Italian,  Icelandic,  and 
French  or  Flemish  specimens. 

153 
Cider  Jug,  Norman  ;  plate  en,  iii  ; 

140 

Clifford's   Inn   Exhibition,    refe- 
rences, 127,  147,  163 
Cceur,   Jacques,  supplies  pewter 

for  his  work-people,  66 
Coffee  Pots  : 

Eighteenth     century  ;      plate 

LXXXVIII,  iii;    122,  161 
Dutch  ;      no     marks  ;      plate 

LXVII,  ii  ;    96 
Flemish  ;    eighteenth  century  ; 

plate  xcv,  i  ;    132 
French  ;    eighteenth  century  ; 

plate  LXXXIX,  ii  ;    124 
Louis    XIV.    style ;     German  ; 
first  half  of  eighteenth  cen- 
tury ;    plate  xc,  i  ;    125 
Collecting :     not    a    fashionable 
craze,    i  ;    its    attractions    as 
compared  with  other  objects, 
2  ;     taste   increasing,   4  ;    ex- 
amination of  good  specimens 


INDEX 


Collecting — cont. 

essential,  ib.  ;  aim  and  scope 
of  the  book,  5  ;  proportions  of 
of  alloy  not  important,  17  ; 
pewter  Church  flagons  before 
1603  unlikely,  94  ;  collecting 
and  displaying,  120  ;  diffi- 
culties caused  by  imperfect 
records  of  the  companies,  ib.  ; 
five  plates  of  "  touches  "  at 
Pewterers'  Hall,  121  ;  paucity 
of  marks  on  existing  speci- 
mens, ib.  ;  confusing  effects  of 
early  rules,  122  ;  Edinburgh 
"  touches  "  more  instructive 
than  those  of  London,  124  ; 
advice  to  owners  of  vessels 
with  private  "touches,"  125  ; 
marks  generally  misleading 
and  confusing,  125  ;  Nurem- 
berg almost  the  only  marks 
not  duplicated  elsewhere,  127  ; 
reasons  for  the  absence  of 
marks,  128  ;  effects  of  the  use 
of  old  moulds  by  travelling 
pewterers,  130  ;  style,  practic- 
ally, the  only  indication  of 
time  and  place  of  manufac- 
ture, ib.  ;  differences  between 
Scotch  and  English  measures, 
132  ;  knowledge  of  contem- 
porary work  of  gold-  and 
silver-smiths  valuable,  133  ; 
Boissonet's  rules  for  "  scour- 
ing," 134;  coats  of  arms 
doubtful  indications  of  date, 
134  ;  experience  and  genuine 
interest  in  the  subject  neces- 
tary,  135  ;  some  of  the  articles 
collectors  may  hope  to  find, 
136  ;  unadvisable  to  render 
old  pewter  indistinguishable 
from  new,  145 

Communion      Cups,     &c.        See 
Church  Vessels,  &c. 

Cornwall,  its  metals,  30,  32 

Cream  Jugs  : 

Plate     LXIV,     iii,     92  ;      plate 
LXXXVI,  v,   118 


Cream  Jugs — cont. 

English  ;    eighteenth  century  ; 

plate  L,  i,  v  ;    74 
Early      nineteenth      century  ; 
plate     LXXXVIII,      ii  ;      122, 
161 

Scotch  ;   plate  LXXXII,  i  ;    116 
Scotch ;       no      mark  ;      plate 

LXXXVI,  i  ;    118 
Cromwell,      Oliver,       reference, 

103 

"  Crooked  Lane  men  "  and  their 
ware  in  1634,  98  ;  references, 

99»  105 
"  Crooked  Lane  men  "  and  their 

ware  in  1634,  98 
Cruets  : 

Cruet,    Vinum  ;     no    marks  ; 

plate  xxxi,  i;    46,  154 
Cruet-Stand  ;     German  ;    Mid- 
dle  of  eighteenth  century  ; 
plate  XLIX,  ii;    72 
Cruets,  Set  of  ;   Scotch  ;   eigh- 
teenth century ;  plate  LXXXV, 
iii,  iv,   v  ;    118 
Cumberland,  Duke  of,  reference, 

114 
Cups.     See  Tankards 

DAGGER,  Oriental  ;  pewter  and 
bone  ;  sheath  bound  with 
pewter;  plate  c,  136,  162 

Danby,  John,  of  Alveston,  his 
pewter  of  1444,  66 

Decoration :  characteristics  of 
good  specimens,  40  ;  supposed 
embellishments  probably  not 
the  work  of  the  maker,  41  ; 
chasers  alluded  to  by  Jean  de 
Jeandun,  1323,  56;  Francois 
Briot  famous  in  sixteenth  cen- 
tury, 79  ;  gilding  and  painting 
prohibited,  1564,  82  ;  develop- 
ment of  French  and  German 
decorated  ware  during  the 
sixteenth  century,  92 

Dishes  : 

English ;  seventeenth  century  ; 
plate  XLVII,  i ;  68,  161 


169 


OLD  PEWTER 


Dishes— cont. 

English;  eighteenth  century  ( ?); 
date  1689  ;  plate  LXXII, 
ii;  103,  160 

middle     of      eighteenth 

century  ;    plate   ix,   ii  ; 
12,  in,  160 

eighteenth  century;  plate 

LXXIII,  iii  ;  104,  1 60 

Scotch  ;    sixteenth    cen- 

tury; plate  LXXIV,  ii;  106 

sixteenth  century  ;  plate 

XLVII,  iii  ;   68 

Roman  ;  found  at  Appleshaw  ; 
plates  I,  II,  in  ;  I,  2.  4,  47, 
48,  49,  160 

Displaying :  original  surround- 
ings the  best,  137;  its  utility 
to  be  considered  in  the  choice 
of  surroundings,  ib.  ;  its  home- 
liness appeals  to  the  sympa- 
thetic and  imaginative,  138  ; 
its  limitations  and  individu- 
ality* I39  J  incongruous  proxi- 
mity to  choicer  objects  to  be 
avoided,  140  ;  suggestions  for 
housing  and  placing,  141  ; 
cleaning  and  repairing,  142  ; 
evidences  of  age  to  be  pre- 
served, 145 

"Dog-nose"  handle.     See  Spoons 
Dyer,  Lawrence,  fined  for  work- 
ing a  new  invention,  103 

EDWARD  I.,  pewter  used  at  his 

coronation,     54  ;      confirms 

charter  to  the  Stanners,  56 
Edward  IV.  confers  a  charter  on 

the  Guild,  1473,  68 
Egg-Cups  : 

English  ;  eighteenth  century  ; 
plate  L,  iv  ;  74 

with  bead  pattern  ;  eigh- 
teenth century  ;  plate  xcvn, 
v  ;  133 

Scotch  ;  set  of  six  ;  no  marks  ; 
plate  xci,  iv-ix  ;  126,  162 

No  marks  ;  plate  xci,  i,  ii,  iii  ; 
126,  162 

170 


Ely,  John,  Vicar  of  Ripon, 
bequest,  1427,  65 

Enderlein,    Gaspard,    references, 
3,42,  95,  137 

Engleheart,    Rev.  C.  H.,  disl 
found  at  Appleshaw  by,  plates 
i,  n,  in,   i,  2,  4;    his  deduc 
tions    concerning    the    marks 
on,    47,    48  ;     references,    45, 
46,  49 

English  Pewter  :    moulds  owned 
in  common,   33  ;    foreign  ii 
portations      prohibited,      38  ; 
some  causes  of  decadence,  42 ; 
its    domestic    use    first    men- 
tioned in  1274,  54;    the  craf 
established  in  London  in  1347, 
57  ;    master  and  wardens  ap- 
pointed to  enforce  ordinances, 
1348,    58  ;    during   fourteenth 
century    used    principally    by 
the      Court,      nobility,      and 
Church   dignitaries,   62  ;    out- 
cry in  the  fourteenth  centui 
against  pedlars  and  unlic 
workmen,  62  ;    York,  in  14 1< 
the  principal  trading  place  fc 
the  North  of  England,  64  ;  i1 
use  ordained,  in  1423,  by  re- 
tailers of  ale,  64  ;    bequest  by 
John    Ely,    Vicar    of    Ripon, 
65  ;     London   pewterers   pros- 
perous   in    1438,    66  ;    pewter 
possessed  by  John  Danby  in 
1444,    66 ;     its    use    by    tl 
nobility  declining  in  1459,  66 
partly  religious  nature  of  tl 
London  Guild,   67  ;    imports 
tions   from   England   stamped 
by  the  importers  in  1467,  with 
a   crowned   rose,   67  ;    pewter 
bought  in  1470,  by  the  Gold- 
smiths'   Company,    ib.  ;    Ed- 
ward IV.  confers  a  charter  on 
the  Guild,  68  ;  Lady  Uvedale's 
bequest  in    1487,   69  ;    stami 
and    seal    of    London    Guil 
1492,  69  ;  development  during 
fifteenth     century    tends     to- 


INDEX 


English  Pewter — cont. 
wards  articles  of  domestic  use, 
71  ;    at  beginning  of  sixteenth 
century  too   expensive   to   be 
common,  72  ;    in   1503  marks 
and    "  touches "    made    com- 
pulsory, ib.  ;   Guilds,  in  1504, 
restricted  in  their  powers  with 
regard  to  disputes,  73  ;    craft 
complains  of  "  strangers  "  and 
false      weights,       1512,      74; 
troubled   by  foreign  competi- 
tion,    1535,     75  ;      protective 
measures  passed  in  1538,  76; 
special    marks     ordained     for 
York    craftsmen,     1540,     77  ; 
stamping  pewter  lids  for  stone- 
ware vessels  compulsory,  80  ; 
buying  at  night  by  pewterers 
prohibited,  32,  80  ;  number  of 
apprentices  limited,  82  ;    pew- 
ter in  great  demand  in  middle 
of     sixteenth     century,      83  ; 
London    Company    privileged 
to  charge  royalties  on  smelting 
and  casting  of  tin,  90  ;   pewter 
falls  into  disuse  by  the  higher 
classes,    90  ;     provincial   com- 
petition,   93  ;     increased    ac- 
tivity  in    1612,    95  ;     distinc- 
tions between  various  branches 
of     craftsmen     relaxed,     96  ; 
new  uses  in   1614,  97  ;    Lord 
Northampton's       pewter      in 
1614,  ib.  ;    the  pewter  of  Sir 
Thomas     Hoskyns     in     1615, 
ib.  ;   clamour  for  the  suppres- 
sion of  "  deceivable  hawkers  " 
in    1621,   98  ;    pewterers   imi- 
tate the   marks   of  the  Gold- 
smiths'   Company,    1635,    99  ; 
1641     the     earliest     date     of 
Dorsetshire  pewter,  100  ;    res- 
trictive methods  of  the  London 
Company,    103  ;     pewter    cis- 
terns and  toys  appear  in  1667, 
105  ;    penalty  for  making  in- 
ferior toys,  ib.  ;    exclusion  of 
a  French  craftsman,   106;  in- 


English  Pewter — cont. 

consistent  attitude  of  the 
London  Company  in  1690, 
107  ;  decline  of  the  craft 
evident  at  end  of  the  seven- 
teenth century,  ib.  ;  craft  de- 
clines during  eighteenth  and 
nineteenth  centuries,  109 ; 
rule  against  self-advertise- 
ment falls  into  disuse,  1727, 
no  ;  London  Company  power- 
less, ib.,  113,  115  ;  practical 
extinction  of  the  craft  in  the 
nineteenth  century,  117,  118  ; 
last  "  touch "  registered  at 
Pewterers'  Hall,  1824,  ib.  ; 
efforts  to  revive  the  craft  in- 
appreciable, 119 

Ewer,  Robert,  sells  pewter,  1607, 

94 

Ewer  and  Basin  ;    plate  LXXVII, 
in,   161 

and    Cover  ;     German  ; 

plate  XLIX  ;    i.  72 

FAIRFAX,  Sir  William,  his  "  sal- 
lite  "  dishes,  34  ;  inventory 
of  the  contents  of  his  house  at 
Gillingham,  quoted,  87 

Falstolfe,  Sir  John,  reference,  66 ; 

"  Fasson  d'argent,"  a  sign  of 
decadence,  42  ;  vessels  made 
of,  6 1  ;  reference,  73 

"  Feast  vessels,"  the  loan  of,  40 

FitzHenry,  J.  H.,  Esq.,  cruet- 
stand  and  sugar-box  given  by  ; 
plates  XLIX,  ii,  xc,  ii  ;  72,  125 

Flagons  and  Jugs  : 

Flagon  ;   plate  LXIV,  ii  ;   92 

seventeenth       century ; 

plate  xcix,  i ;    1 34 

seventeenth      century ; 

plate  xcix,  iii ;    1 34 

Imitation  Chinese  ; 

Dutch ;  seventeenth  cen- 
tury ;  plate  LXIII,  ii ;  90 

German  ;      seventeenth 

century  ;     plate    LXIII  ; 
iii.,  90,  159 

171 


OLD  PEWTER 


Flagons  and  Jugs  —  cont. 

Flagon  ;  eighteenth  century  ; 
plate  cm,  ii  ;  142 

-  1706,  Nuremberg  mark  ; 

plate  XLII,  ii  ;   62,  159 

-  German  ;          eighteenth 

century  ;      plate    XLIII, 
i  ;    62,   157 

-  German  ;          eighteenth 

century  ;      plate    XLIII, 
iii  ;    62,  157 

-  Probably  Scotch  ;    eigh- 

teenth    century  ;     plate 
LXIII,  i  ;    90 

Jug  ;  Dutch  ;  early  eigh- 
teenth century  ;  plate 
LXXVIII,  ii  ;  112 

-  Covered  ;       Archangel 
mark  ;     plate    LXIV,    i  ; 


92 
Dutch 


mark,    crossed 
plate    LXVII,    i  ; 


rose 
96 

-  Dutch  ;    majk,  a    rose  ; 
plate  LXVII,*  iii;    96 

-  English  ;      seventeenth 
century  ;  plate  LXII,  88, 

159 

-  English  ;  eighteenth  cen- 

tury ;    plate  XLVIII,  ii  ; 
70 

-  English  ;  eighteenth  cen- 

tury ;     plate    LXVI,    ii  ; 

94 

-  French  ;   eighteenth  cen- 

tury ;   plate  LXVI,  i  ;  94 

-  George       IV.  ;         plate 
LXVIII  ;    98 

Fly,  Tim,  his  punning  device,  123 
Food  Bottle  ;  square  ;  plate  ci,  i  ; 

138,  163 
-  Bottle  ;     sexagonal  ;    plate 

ci,  iii  ;     138,    163 
Forks  :    pewter  forks  mentioned 

in     a     French     inventory     of 

1672,  105  ;    their   rarity,  147  ; 

Fork,   small,  no  mark  ;    plate 

xi,  ii  ;    14 
Fran9ois  I.,  reference,  74 

172 


French  Pewter :  proportions  of 
alloy  during  the  eighteenth 
century,  13  ;  mentioned  in 
Boileau's  account  of  the  Paris 
Guilds  in  the  thirteenth  cen- 
tury, 55  ;  mentioned  in  a 
further  h'st  of  Parisian  crafts- 
men, ib.  ;  Guild  prosperous 
in  1304,  ib.  ;  edict  against 
inferior  alloys  passed  at  Poi- 
tiers, 1333,  57  ;  Guild  at 
Rouen,  1369,  58  ;  vessels 
owned  by  Henri  de  Poitiers, 
Bishop  of  Troyes,  59  ;  vessels 
considered  as  perquisites  by 
court  attendants,  59  ;  Miche- 
let  the  Breton,  a  craftsman  of 
1380,  59  ;  record  in  1380  of  a 
Benitier  in  the  possession  of 
Jean  de  Halomesnil,  60  ;  store 
possessed  by  a  canon  of 
Troyes,  1386,  60  ;  the  reward 
for  workmanship  very  slight, 
ib.  ;  the  city  of  Amiens  buys 
pewter  from  Thibaut  la  Rue, 
ib.  ;  the  city  of  Rouen  pur- 
chases a  gallon  flagon,  61  ; 
bequest  of  the  Archbishop  of 
Rheims,  1389,  61  ;  Isabel  de 
Moncel,  a  Parisian  crafts- 
woman  of  1395,  ib.  ;  mis- 
applied use  of  a  wine  measure 
by  one  Jean  Le  Bceuf,  1396, 
ib.  ;  tools  used  at  Rouen, 
1402,  63  ;  Isabella  of  Bava- 
ria's order  for  the  Hotel  St. 
Pol,  1401,  63  ;  Guillebert  of 
Metz  a  noted  maker,  1407, 
64  ;  extensive  purchase  by 
Charles  VII.  of  France  in 
1422,  64  ;  Jacques  Coeur  in 
1453  supplies  pewter  for  his 
workpeople,  66  ;  Amiens  buys 
pewter  cups  for  noble  visitors, 
1463,  67  ;  Duke  of  Burgundy 
in  1478  establishes  Guilds, 
68  ;  pewter  bottles  in  the 
"  Livre  des  Mestiers "  of 
Charles  V.,  69 ;  increase  of 


INDEX 


French  Pewter — cont. 

craftsmen  during  the  fifteenth 
century,  71  ;  the  "silver 
fashion  "  increases  in  favour, 
I5°7>  73  5  Duke  of  Burgundy's 
vessels,  in  1507,  incline  to 
a  decorative  character,  73  ; 
Duke  of  Bourbon's  vessels  in 
1507,  ib.  ;  Amiens  buys  vessel 
in  1507  from  Pierre  Hemeron, 
ib.  ;  in  1516  from  Jeanne 
d'Avesne,  74  ;  Paris  silver- 
smiths take  action  against 
the  pewterers,  1545,  79  ; 
Briot  famous  as  a  decorator  of 
pewter,  sixteenth  century, 
79  ;  Paris  draper's  considera- 
ble bequest,  1572,  83  ;  extracts 
from  inventory  of  Pierre  de 
Capdeville  of  Bordeaux,  1591, 
88  ;  development  of  decorated 
ware  during  sixteenth  cen- 
tury, 92  ;  decline,  except  the 
ornamental  kind,  during  first 
half  of  seventeenth  century, 
1 02  ;  pewter  spoons  and  forks 
mentioned  in  French  inventory 
1672,  105 

Galbraith,  Glasgow,  his  mark ; 
plate  XLIV,  i  ;  64,  157 

Gardner,  Mr.  Starkie,  quoted,  8, 
9,  12,  14 

German  Pewter :  development 
of  the  industry,  1324,  57  ; 
Sebaldus  Ruprecht  famous  for 
pewter  which  could  be  mis- 
taken for  silver,  61  ;  Buschius 
of  Hildesheim  records  pewter 
in  Saxon  convents,  68  ;  pew- 
ter in  the  convents  at  Erfurth, 
ib.  ;  Martin  Harscher,  a  noted 
German  craftsman,  1523,  75  ; 
Melchior  Koch,  a  Nuremberg 
craftsman,  makes  an  alloy  re- 
sembling gold,  1564,  82  ;  pew- 
terers claim  to  make  every 
article  produced  by  gold-  or 


silversmiths,   1573,  84  ; 
Schmitt's      candlesticks 


Peter 
ob- 


German  Pewter — cont. 

jected    to    by    goldsmiths    of 
Nuremberg,  84  ;   development 
of  decorated  ware  in  the  six- 
teenth century,  92  ;   salver  of 
Nuremberg  a  reproduction  of 
Briot's  Temperantia  salver,  95 
Ghent,  references,  58,  59,  126 
Gilbert,  W.  S.,  references,  2,  101 
Gill  Measure.     See  Measures 
Gloucester       Candlestick.       See 

Candlesticks 

Goldsmiths'  Company  purchase 
pewter,  1470,  67  ;  their  marks 
imitated  in  1635,  99 
Goupil,  Jehan,  of  Tours,  sells 
pewter  to  Charles  VII.  of 
France,  64 

Gowland,  Mr.,  on  the  treatment 
of  pewter  by  the  Japanese, 
143  ;  references,  9,  10,  n,  14, 
15,  48,  50 

Gray,  Master,  of  Dundee  Corpo- 
ration, 85 
Grey,     Ninian,    works    at    two 

trades,  113 

Guild  Cups.     See  Tankards 
Guildhall,  London,  reference,  13 
Guildhall      Museum,       London, 

Anglo-Saxon  specimens  at,  52 
Guillebert    of    Metz,     a    noted 
French  craftsman,  64 

HALF -GiLL Measure.  SeeMeasures 

Half-Glass  Measure.  SeeMeasures 

Half    Mutchkin.     See    Measures 

Halliday,  J.E.,  his  "Terrier,"  or 

inventory    of    the    Diocese    of 

Llandaff,  reference,  128 

Halomesnil,    Jean    de,    benitier 

possessed  by,  60 

Hammermen :  called  "  Sad- 
ware  "  men,  27  ;  articles  made 
by,  28  ;  their  work  fashioned 
by  hammering,  33  ;  incor- 
porated at  Perth,  75  ;  at 
Cannongate,  ib.  ;  patron  saint 
St.  Eloi,  ib.  ;  incorporated  at 
Glasgow,  76  ;  earliest  ordi- 

173 


OLD  PEWTER 


Hammermen — con/, 

nances  at  St.  Andrew's,  77  ; 
James  V.  ratifies  "Seal  of 
Cause "  of  the  Cannongate, 
1540,  ib.  ;  apprentices  of 
Perth  forbidden  to  play  on 
the  Inch,  79  ;  first  "  Seal  of 
Cause  "  granted  to  Aberdeen, 
1579,  84;  Stirling  records 
imperfect,  89  ;  pewterers  not 
mentioned  in  Glasgow  records 
before  1648,  100 ;  last  Can- 
nongate enrolment,  1 1 1  ;  craft 
combines  with  other  trades 
at  Perth,  1747,  115  ;  meetings 
still  held  at  Aberdeen,  Dundee, 
Perth,  and  Stirling,  119 

Hancock,  Samuel,  London  Com- 
pany's attitude  towards,  1690, 
107 

Harrison's  "  Description  of  Eng- 
land," quoted,  12,  29,  39,  86 

Harscher,  Martin,  a  famous  Ger- 
man craftsman,  reference,  75 

Hemeron,  Pierre,  a  French  crafts- 
man, 1507,  73 

Henry  VII.,  references,  72,  90 

Herb-Cannisters  : 

Dutch  ;  seventeenth  century  ; 
mark,  a  standing  figure  ; 
plate  LXXXVI,  ii  ;  118,  161 
Dutch  ;  marked  1766,  a  Cupid 
in  profile  blowing  a  horn  ; 
PSB  ;  plate  LXXXVI.  iv  ; 
118,  161 

Herodotus,  reference,  31 

Hiring  out  articles  a  profitable 
branch  of  the  trade,  39 

Hollow-ware,  finishing  process,  36 

Hollow- ware  men,  references,  27, 
69 

Hope,  Mr.  St.  John,  reference,  96 

Home,  John,  of  Snow  Hill,  his 
mark  on  early  English  Mea- 
sure ;  plate  ix,  i  ;  12,  158 

Hoskyns,  Thomas,  of  Oxted,  his 
pewter  in  1615,  97 

Hungary,  Clement  of,  list  of  his 
pewter  vessels,  alluded  to,  57 

174 


ICELAND,  Pocket  Communion 
Service  bought  in  ;  plate  xxx, 

i;  45 
Ictis,  named  by  old  historians, 

supposed    to    be    either    St. 

Michael's    Mount,    Falmouth, 

Weymouth,    or    the    Isle    of 

Wight,  32 
Ink-Stands,  &c.  : 

Ink-pot,    Scotch  ;    no   mark  ; 
plate  xcvin,  i  ;    133,  162 

Inkstand  ;  plate  XL,  iii ;  60,  162 

plate  LXXXIII,  v  ;    117 

late  eighteenth  century  ; 

plate  xcix,  ii  ;    134,  162 
French  ;  eighteenth  cen- 
tury ;     plate    xcv,    ii  ; 

132,  162 

Italian  ;  plate  xcvin,  ii  ; 

133,  162 

Isabella  of  Bavaria,  her  pur- 
chase for  the  Hotel  St.  Pol  in 
1401,  63 

Isle  of  Wight,  reference,  32 

JAMES    V.    persuades    foreign 
craftsmen   to   settle   in    Scot- 
land, 1539,  77 

James  VI.  grants  charter  to  the 
crafts,  1581,  85  ;  references, 
97»  83 

Japan,  spoons  seen  by  Mr.  Gow- 
land  at  Nara,  50 

Japanese   treatment  of   pewter. 

I5»  143 
Jeandun,  Jean  de,  his  reference 

to  French  decorators,  56 
John  II.  of  France,  his  pewter, 

I35i»  58 

John,  King,  reference,  54 
ohnson,  Dr.,  reference,  7 
onson,  Ben,  reference,  139 
ugs.     See  Flagons 
upe,  John,  his  defiant  attitude 

towards  the  London  Company, 

1736,  113 

KOCH,  Melchior,  his  alloy  resem- 
bling gold,  82 


INDEX 


Kriswick,     chalice     and     paten 

from,  plate  xxi,  iii,  30 
Kwammu,    Emperor    of    Japan, 

reference,  50 

LADLES : 

Laton  slip-top  Ladle,  handle 
of  hexagonal  section 
seventeenth  century  ; 
plate  xx,  iv  ;  29,  1 5 1 

slip-top  Ladle,  found  in 

the  City  of  London  ; 
has  remains  of  tin  pla- 
ting ;  seventeenth  cen- 
tury ;  plate  xx,  v  ;  29 

slip-top  Ladle,  found  in 

Suffolk  ;       seventeenth 

century  ;   plate  xx,  vi  ; 

29 
slip-top  Ladle,  found  in 

London ;       seventeenth 

century  ;  plate  xx.  vii  ; 

29,  150 
Soup     Ladle,     marked     John 

Yates  ;  plate  xi,  iii  ;   14, 

IS2 

marked   Askberry,   &c.  ; 

plate  xi,  iv  ;    14,  152 
Toddy   Ladles,   marked    John 
:,    Yates,  &c.  ;  plate  xi,  i  ;    14, 
1     152 

"  La      Grande      Mademoiselle  " 
(the  Duchesse  de  Montpensier), 

T    I03 
Lamps  : 

Rarely  found  in  good  con- 
dition, 154 

Lamp  Time-Keepers  ;  seven- 
teenth century;  plate  xxxviu 

i-v;    58,  155 

Oil    Lamp  ;     German  ;     eigh- 
teenth   century  ;     plate 
xxxvii,  i  ;    56,  154 
Lamps  ;      German  ;     eigh- 
teenth century;  plate  xxxvii 
iv,  v  ;    56,  154 
Spirit-Lamp  ;  plate  LXXXV,  vi; 

118,  161 
La  Rue,  French  craftsman,  60  ; 


"  Laver,"  the,  its  first  appear- 
ance, 97.  See  Church  Vessels, 
&c. 

Le  Boeuf,  Jean,  in  1396  fore- 
stalls an  incident  recorded  in 
"  Pickwick,"  6 1 

Le  Peautrer,  Nicholas,  estab- 
lishment of  the  craft  in  London 
proved  by  his  will,  1347,  58 

Lea,  Francis,  fined,  1668,  for 
making  inferior  toys,  105 

Limoges,  composition  of  pewter 
at,  13 

"  Livres    des    Mestiers  "    of 
Charles     V.     pewter     bottles 
mentioned  in,  69 

London,  spoons  found  in,  plate 
xu,  ii,  iv-viii,  16  ;  plate  xin,  ii, 
v,  vii,  viii,  18  ;  plate  xiv,  i,  iii, 
vii,  viii,  20  ;  plate  xv,  ii,  iv,  vi, 

21  ;    plate  xvi,  ii,  vii,  viii,  ix, 

22  ;    plate  xyn,  i,  iv-viii,  24  ; 
plate  xvni,  i,  iii,  26  ;  plate  xix, 
vii,  28  ;  plate  xx,  i,  vii,  29 

Louise  of  Savoy,  reference,  74 
Louis   XIV.   sends  his  plate  to 

the  mint,  102,  108 
Loving  Cup.     See  Tankards 

MACHINES,  their  use  discouraged 
by  early  craftsmen,  36 

Marks  and  Marking  :  vessels  to 
be  stamped  with  quality  mark 
and  private  "  touch,"  37  ; 
Mons  marks  in  1467,  67  ; 
marking  made  compulsory  by 
Act  of  Parliament,  1503,  72  ; 
enacted  in  Scotland,  1567, 
83  ;  hall-marking  of  tavern 
measures  ordained  by  Dundee 
guild,  1614,  96  ;  compulsory 
in  Scotland,  1641,  100  ;  plates 
of  "  touches  "  at  Pewterers' 
Hall,  121  ;  few  marks  on 
existing  specimens,  121  ;  con- 
fusion caused  by  the  au- 
thorities, 122  ;  Edinburgh 
"touches,"  124;  advice  to 
owners  of  specimens  with 


OLD  PEWTER 


Marks  and  Marking — cont. 

private  "  touches,"  125  ;  marks 
in  general  misleading  and  con- 
fusing, 125,  127  ;  Nuremberg 
marks  the  most  reliable,  127  ; 
reasons  for  the  absence  of 
marks,  128  ;  marks,  where 
legible,  of  spoons  on  plates 
xn-xx.  See  special  plates  civ, 
cv,  cvi,  148,  150,  152 

Marlowe,  Christopher,  reference, 
138 

Martin,  abbot,  1186-1191,  early 
pewter  plaque  recording  name 
and  title,  53 

Masse,  Mr.,  "  Catalogue  of  the 
Pewter  Exhibition  "  organised 
by  him  at  Clifford's  Inn  Hall, 
references,  127,  147  ;  alluded 
to,  9,  16,  65,  96,  121,  125 

Maxwell,  Stephen,  of  Glasgow 
uses  London  stamp,  124 

Measures  : 

Comparison    of    Scotch    and 

English,  132 

Measure;  plate  LXXXIII  ;    117 

Measures ;  plate LVI,  i-v ;  78, 1 58 

Measure,    temp.    Charles    II. 

found     in     Parliament 

Street ;  plate  xxx,  iii ;  45 

English  ;        seventeenth 

century  ;  plate  LXXIV, 
iv.  ;  106 

English;  early  eighteenth 

century  ;  plate  LXV,  iii  ; 

94 

English  ;       early  «•  eigh- 

teenth century  ;  mark, 
"  John  Home,"  &c.  ; 
plate  ix,  i  ;  12,  158 

English  ;       early      eigh- 

teenth century  ;  plate 
XLIV,  ii  ;  64,  158 

English  ;       early      eigh- 

teenth century  ;  plate 
XLIV,  iii  ;  64,  158 

English  ;  eighteenth 

century  ;  plate  L,  iii  ; 
74,  157 


Measures — cont. 

Measures  English  ;  eighteenth 
century  ;  plate  LVIII, 
v ;  82,  158 

eighteenth  century;  plate 

LXXix,  v  ;    114 

Measures,  French,  A  set  of, 
plate  LXI,  86 

Scotch  ;   eighteenth  cen- 
tury ;  plate  LV,  i-iv  ;  78,  158 

Measure,  German  ;  plate  vn,  i ; 
10 

or      Tankard,      German 

late     seventeenth     century 
plate  LIV  ;   78 

Gill  Measure  ;  no  marks  ; 
plate  xxxi,  vi  ;  46,  158 

Measure,   Modern,   mark 

VR  Crowned  ;    plate  xxxi, 
ii;    46,   158 

Half  Gill  Measure  ;  marked 
like  Mutchkin  but  with 
two  oval  stamps  instead 
of  Glas.  and  shield  ;  plate 
xxxi,  v;  46,  158 
Quarter  Gill  Measure  ;  Glas- 
gow mark  ;  plate  LXXXII,  ix ; 
116 

Half  Mutchkin  Measure  ; 
marked  Imperial  crown 
Standard,  &c.  ;  plate  xxxi, 
iii;  46,  158 

Old  Half  Glass  Measure  ;    no 
marks ;  plate  xxxi,  iv ;  46, 158 
Tappit  Hen.  ;    eighteenth  cen- 
tury ;  plate  LX  ;  85,  158 
Scotch  ;        eigh- 
teenth   century  ;     plate 
x,  ii ;    13,  132,  158 

Hens  ;  plate  LIX,  i,  ii,  iii ; 

84,  158 

Wine       Measures,       German  ; 
eighteenth    century  ;     plate 
LVII,  ii,  iii ;   80,  158 
Michelet   the   Breton,   a   French 

craftsman  of  1380,  59,  60 
Midhurst,     Communion     Flagon 
from,  frontispiece  ;    plate  xxi, 
ii  ;    30.  J53 


176 


INDEX 


Milk  Jugs  : 

English  ;  late  eighteenth  cen- 
tury ;  plate  LXXVI,  iii  ; 
109 

early  nineteenth  century  ; 

plate  xcn,  i  ;    126 

Miscellaneous  objects,    162 

Moncel,  Isabel  de,  a  French 
craftswoman,  1395,  61 

Mons,  a  Belgian  centre  of  the 
craft,  58,  67 

Mont  St.  Michel,  early  pewter 
plaques  found  in  coffins  at,  53 

Monteith,  James,  a  craftsman, 
1649,  his  claim  for  making 
bullets,  101 

Montpelier,  composition  of  pew- 
ter there  in  1437,  13  ;  refe- 
rence, 65 

Montrousti,  Jehan  de,  supplies 
pewter  to  Isabella  of  Bavaria, 
63 

Moulds :  list  made  by  Mr. 
Welch,  33  ;  new  names  ap- 
pearing in  a  further  York  list, 

Moyes,  Mr.  James  one  of  the 
last  Edinburgh  pewterers,  118 

Mugs: 

Mug,  early  nineteenth  cen- 
tury ;  plate  cm,  iv  ;  142 

William  IV.  ;    plate  cm, 

i  ;     142 

with     handle,     English ; 

early     eighteenth      cen- 
tury ;  plate  LXXIV,  iii ;  106 

Mustard  Pots  : 

Mustard  Pot ;  plate  vii,  iv  ; 
10,  in 

English ;      seventeenth 

century  ;    plate  LXV,  i  ; 

English  ;      early     eigh- 
teenth   century  ;     plate 
xcvii,  iii  ;     133 

eighteenth  century  ;  plate 

LXXIX,  i  ;     114 

late  eighteenth  century  ; 

plate  LXXIX,  iii  ;    114 


Mustard  Pots— cont. 

Mustard  Pot;  marked  with  a 
figure  floating  in  the  air ; 
plate  xc,  iii  ;  125 

Mustard  Pots  ;  plate  vm,  v, 
vi,  vii  ;  10,  in 

seventeenth  and  eigh- 
teenth centuries;  plate 
LXXX,  i-vi  ;  114,  161 

NEWGATE  Street,  Rat-tailed 
spoon  found  at  ;  plate  xix,  ii  ; 
28 

Nightingale,  Rev.  J.  E.,  "  Church 
Plate  of  Norfolk,"  reference, 
100 

Northampton,  Lord,  his  pewter 
in  1614,  97 

Northumberland,  an  Earl  of, 
in  the  habit  of  hiring  pewter 
vessels,  40 

Norwich,  Seal-headed  spoon 
found  at,  plate  xx,  viii,  29 

Nuremberg,  composition  of  pew- 
ter at,  1576,  13  ;  marks  the 
most  reliable,  127  ;  refer- 
ences, 57,  126,  127 

OLD    Half   Glass   Measure.     See 

Measures 
Oriental  Dagger.     See  Dagger 

PARLIAMENT  Street,  Whitehall, 
Measure  found  in  ;  plate  xxx, 

iii;  45 

Paten.     See  Church  Vessels,  &c. 
Peg  Tankard.     See  Tankards 
Penny,  Edmund,  gives  a  Church 

vessel,  1609,  95 
Pepper-Boxes  and  Pots  : 

Pepper-box  ;   plate  vii,  ii  ;    10 

eighteenth  century  ;  plate 

LXXIX,  iv ;    114 
Pepper-boxes  ;  plate  vm,  i,  iii. 

iv  ;    10,  in 

Pepper-Pot,  English  ;  eigh- 
teenth century  ;  plate 
xcvii,  i  ;  133 


177 


OLD  PEWTER 


Pepper-Boxes  and  Pots — cont. 
Pepper -Pot ;     English  ;     early 
nineteenth        century ; 
plate  xcvn,  ii  ;    133 

seventeenth    and    eigh- 
teenth century ;  plate  LXXXI, 
i-vi  ;    114,    161 

Pepys,  Samuel,  buys  a  pewter 
cistern,  1667,  105 

Pewter  :  its  disuse  for  domestic 
and  other  purposes,  3  ;  deri- 
vation of  the  term,  7  ;  com- 
position of  English  "  fine  pew- 
ter," ii  ;  its  composition 
according  to  Harrison  the 
historian,  12  ;  rough  and 
ready  tests  for  ascertaining  its 
quality,  16  ;  tests  used  by 
French  pewterers,  17  ;  how 
it  was  wrought,  24  ;  all  the 
metals  employed  home  pro- 
ducts, 32 

Pewterers'  Company  oppose 
"  Crooked-lane  men,"  105 

Pewterers'  Hall,  1640  the  earliest 
date  on  "  touch  "  plate  pre- 
served there,  100  ;  the  last 
"  touch  "  registered  there  in 
1824,  118 

"  Pied  de  Biche "  specimens. 
See  Spoons 

Pirley  Pig  of  Dundee  :  date  ; 
description  ;  use  ;  loss,  and 
recovery  in  1839,  94 

Pitt  and  Davley,  their  mark  ; 
plate  ix,  iii  ;  12 

Plates  : 

Plate,  one  of  six  ;  English  ; 
eighteenth  century ; 
plate  Lxxn,iii;  103,  160 

one  of  a  set  of  six,  English 

eighteenth     century  ; 
stamped     with    initials 
ICB    and    IBB ;     plate 
LVIII,  iv  ;   82,  160 

Wavy-Edged,    English ; 

late  eighteenth  century  ; 
one  of  a  set  of  six  ;  plate 
LXXIII,  v  ;    104,  1 60 

178 


Plautus,  reference  to  pewter,  45 
Pliny,  his  reference  to  tin,  31 
Plumbum  candidum,  the  ancient 

term  for  tin,  31 
Plumbum    nigrum,    the    ancient 

term  for  lead,  31 
Pocket  Communion  Service.     See 

Church  Vessels,  &c. 
Poitiers,    Henry    de,    Bishop    of 

Troyes,  his  pewter,  59 
Porringers  : 

Porringer,  French  ;  probably 
seventeenth  century  ;  plate 
xxvin,  iii  ;  42,  1 60 

Porringers  ;  plate  LXX,  i,  ii,  iii  ; 

100,  160 
Poullett,  Sir  Richard,  his  "  sallet 

pewter  dishes  "  of  1618,  34 
Protective  safe-guards  of  the  old 

craftsmen,  38 
Purling,     Major,     his    invention 

called  Silvorum,  103 
"  Puritan  "  handle.     See  Spoons 
Pyx.     See  Church  Vessels,  &c. 


IUAIGH,  132,  136 
>uarter-gill  Measure, 
sures 


See  Mea- 


"  RAT-TAIL  "  pattern.  See 
Spoons 

Read,  C.  K.,  references,  46,  48 

Regulations  of  the  early  crafts- 
men designed  to  protect  the 
purchaser  and  to  insure  good 
workmanship,  24,  25  ;  the 
old  regulations  summarised, 
26  ;  days  and  hours  of  manu- 
facture, 37  ;  lids  of  stone- 
ware vessels,  80  ;  apprentices, 
1564,  82  ;  sale  of  goods,  39 

"  Revue  des  Arts  Decoratifs," 
early  post-Roman  Chalice,  now 
destroyed,  reproduced  in,  52 

Rheims,  Archbishop  of,  his  be- 
quest, 1389,  6 1 

Rhynd,  Janet,  her  generous  gift 
to  the  Edinburgh  Trade  In- 
corporation, 78 


INDEX 


Rice-Boiler  ;  French  ;  eigh- 
teenth century  ;  plate  LVIII, 
iii;  82,  162 

Richard  Cceur  de  Lion,  Church 
plate  melted  to  procure  his 
ransom,  53 

Richard  III.,  King,  reference,  90 
Roman  pewter,  divergent  results 
of  analysis,  9  ;   the  discoveries 
at      Appleshaw,     Icklingham, 
Sutton,    and    Colchester,    45  ; 
conjectures    as    to    date,    49  ; 
effects    of    the    Roman    with- 
drawal on  the  industry,  5 1 
Rouen,  references,  58,  60,  63 
Ruprecht,     Sebaldus,       German 
craftsman,   1389,  61 

SACRAMENTAL  Cruets.  See  Church 
Vessels,  &c. 

Sadware,  how  fashioned,  36 

Sadware  men.     See  Hammermen 

St.  Cyr,  pewter  at  the  convent 
of,  68 

St.  Eloi,  the  patron  saint  of 
Scotch  Hammermen,  75,  84 

St.  Martin,  pewter  at  the  Cis- 
tercian convent  of,  68 

St.    Michael's   Mount,   reference, 

32 

St.  Stephen's,  Coleman  Street, 
London,  pewter  vessels  in  use, 
1467,  at,  67 

Salt-Cellars  and  Boxes  : 
Salt-Box  ;    plate  ci,  ii  ;    138 
Salt-Cellars  ;      plate     LXXXIII, 
ii  ;    117,  161 

plate  LXXXIV,  i-v  ;    117, 

161 

marked    with    crowned 

X  ;    plate  vm,  viii-xii  ; 

10,    III 

Salt-Cellar,  English  ;  eigh- 
teenth century  ;  plate 
L,  ii  ;  74 

eighteenth  century ;  plate 

LXXIX,  ii  ;  1 14 

Cromwellian  ;  plate 

LXXXII,  iii  ;  116,  161 


Salt-Cellars— cont. 

Salt-Cellar,      English  ;      plate 
LXXXII,  v  ;    116,  161 

English ;  Georgian  ;  plate 

LXXXII,  vii  ;    116,  161 

Salver,  Wavy-Edged,  with  feet, 
Dutch  ;  late  eighteenth  cen- 
tury ;  plate  LXXIII,  i,  104, 
1 60  ;  salver  of  Nuremberg, 
1611,  a  reproduction  of  Briot's 
Temperantia  salver,  95 

Sampson,  Patrick,  references, 
no,  in 

Sandarach,  or  Callitris  quadri- 
valsis,  used  for  coating  moulds, 

Schmitt,  Peter,  a  German  crafts- 
man, 84 

Schoper,  Harman,  "  Treatise  on 
Industries,"  quoted,  84 

Scotch  Pewter:  Scotland's  ad- 
vance in  the  craft  retarded 
by  the  absence  of  native  tin, 

69  ;      manufacture     first     re- 
corded   in    Edinburgh     1496, 

70  ;     arrangements   of   Guilds 
differ  from  those  in  England, 
ib.  ;  first  recorded   incorpora- 
tion of  Hammermen  at  Perth, 
I5I^»     75  ;      at     Cannongate, 
1535,  ib.  ;  St.  Eloi  the  patron 
saint,   75  ;    first  recorded  in- 
corporation of  Glasgow  Ham- 
mermen,   76  ;     earliest    ordi- 
nances of  St.  Andrews  Ham- 
mermen,  77  ;    James  V.  per- 
suades  foreign    craftsmen    to 
settle  in  Scotland,  and  ratifies 
the   "  Seal  of  Cause  "  of  the 
Cannongate  Hammermen,  ib.  ; 
measures   reimposed,  1543,  to- 
secure    competent    craftsmen  „ 
78  ;    the  Edinburgh  Corpora- 
tion   become    owners    of    the 
Chapel  of  St.  Mary  Magdalen, 

78  ;     Perth    apprentices    for- 
bidden to  play  on  the  Inch, 

79  ;      Church     plate     of     St. 
Giles',  Edinburgh,  removed  for 

179 


OLD  PEWTER 


Scotch  Pewter — cont. 

safety,  81  ;  Perth  apprentices 
wishing  to  become  masters 
present  a  football  to  the  Guild, 

8 1  ;     disappearance    of    much 
pre- Reformation  Church  plate, 

82  ;   marks  enacted,  1567,  83  ; 
Dundee  stoups  ordered  to  be 
officially    marked,    ib.  ;     first 
"  Seal  of  Cause  "   granted  to 
Aberdeen    Hammermen,     84  ; 
James     VI.     grants     charter, 
1581,    raising    crafts    to    the 
level  of  merchants,   85  ;    first 
record  of   the   Dundee   Incor- 
poration, ib.  ;    records  of  the 
Stirling    Hammermen,     1596- 
1620,   more   than   usually  im- 
perfect, 89  ;  continued  growth 
of  the  craft  in  Scotland  during 
the     sixteenth     century,     92  ; 
date,  description,  and  use  of 
the    Pirley    Pig,    94  ;    its   loss 
and      recovery,      ib.  ;       Hall- 
marking   ordained,     1614,    by 
Dundee     Guild,     96  ;      effects 
of  the  Revolution  of   1617  in 
the   Church  of   Scotland,   97  ; 
first  appearance  of  the  "  laver," 
ib.  ;      no     pewterer      in      St. 
Andrew's    Guild    until     1619, 

98  ;     effects    of    the    Church 
troubles  in  1638  on  the  craft, 

99  ;  marks  made  compulsory, 
1641,      100  ;      pewterers     not 
mentioned  in  Glasgow  records 
before  1648,  100  ;   James  Mon- 
teith's      claim      for      making 
bullets,  101  ;   penalty  in  Edin- 
burgh   for    using    bad    metal, 
103  ;  effects  of  the  Restoration 
on  the  craft  in  Scotland,  104 ; 
Act  to  prevent  the  craft  from 
encroaching    upon    the    trade 
of  plumbers,  105  ;    interesting 
Ordinances  issued  by  the  Aber- 
deen Guild,   105  ;    quarrel  be- 
tween pewterers  and  plumbers 
of  Edinburgh,  106  ;    effects  of 

1 80 


Scotch  Pewter — cont. 

the  accession  of  William  III. 
on  Church  plate,  106  ;  white 
iron-smiths  admitted  to  the 
freedom  of  Aberdeen  Guild, 
107  ;  decline  of  the  craft  and 
threatened  absorption  by 
white  iron-smiths,  no;  en- 
rolment of  the  last  pewterer 
at  the  Cannongate,  in  ; 
craftsmen  of  St.  Andrews  and 
Perth  fail  to  preserve  their 
monopoly,  112,  113;  efforts 
of  Edinburgh  white  iron- 
smiths  to  found  a  distinct 
craft,  114;  Dundee  corpora- 
tion enrols  the  last  pewterer, 
115;  Perth  hammermen  com- 
bine with  other  trades,  ib.  ; 
effects  of  the  Presbyterians  on 
the  craft,  116  ;  last  pewterers 
enrolled  at  Aberdeen  and 
Perth,  ib.  ;  white  iron-smiths 
openly  make  pewter  articles, 
ib.  ;  St.  Andrews  pewterers 
outnumbered  by  tin-smiths, 
117;  Aberdeen,  Dundee,  Perth, 
and  Stirling  Hammermen  still 
hold  meetings  but  have  no 
executive  power,  119  »_ 

"  Seal-headed  "  pattern.  See 
Spoons 

Shakespeare,  pewter  mentioned 
by,  87  ;  reference,  139 

Sharpe,  Dr.,  his  Calendar  drawn 
up  for  the  Corporation  of  the 
City  of  London,  58 

Shaving- Pot  ;  English  ;  plate  en, 
i ;  140 

Shoe  Buckles,  A  pair  of  ;  no 
marks  ;  plate  xcvi,  ii,  iv  ;  133, 
162 

Shokotu,  Empress  of  Japan, 
reference.  51 

Silvorum,  Purling's  new  inven- 
tion, 103 

Skeat,  Professor,  references,  7, 
28 

"  Slip-top  "  pattern.  ~  See  Spoons 


INDEX 


Smith,  Samuel,  London,  breaks 

advertising  rule,  116 
Snuff-Boxes  : 

Snuff-Box  ;    no  marks  ;    plate 
xcvi,  v  ;    133 

no    marks  ;     eighteenth 

century  ;  plate  xcvi,  i  ; 
133,  162 

English ;     middle    eigh- 

teenth century  ;  plate 
xcvn,  iv;  133,  162 

Scotch  ;  no  marks  ;  plate 

xcvi,  iii  ;   133,  162 
Soignies,  reference,  60 
Soup  Ladle.     See  Ladles 
Soup    Tureen,    Russian  ;     Arch- 
angel mark  ;  plate  LXXV  ;  108, 
161 

Spanish  Pewter,  statutes  and 
regulations  drawn  up,  1406, 
at  Barcelona,  63 

Spice-Box,   French  ;    eighteenth 
century  ;    plate  xxix,  ii  ;    44, 
162 
Spinning,  a  method  of  working 

pewter,  37 
Spoons : 

Bronze,  found  in  the  City  of 
London;  fourteenth  cen- 
tury ;  probably  French : 
plate  xn,  ii  ;  16,  148 
found  in  London  ;  fif- 
teenth century  ;  plate 
xii,  vi;  16,  148 

stem  of  diamond  section, 

found  in  London  ;  fif- 
teenth century  ;  plate 
xn,  vii ;  1 6,  148 

found   in    London  ;     fif- 

teenth century  (mark, 
same  as  No.  2)  ;  plate 
xn,  viii ;  16,  148 

rat-tailed  ;     seventeenth 

century  ;  plate  xix,  viii  ; 

28,    151 
Dutch  ;     plate   LXXXII,   ii,   iv, 

vi,  viii,  x  ;    1 16 
Laton,   found   in  the  Thames 

at  Westminster  ;    four- 


Spoons — cont. 

teenth  century  ;  plate 
xn,  i ;  16,  148 

Laton  French  ;  fourteenth  cen- 
tury ;  plate  xn,  iii  ;    16 

found  in  the  Thames  in 

London,  stem  of  dia- 
mond section  ;  early 
fifteenth  century  ;  plate 
xn,  iv;  16,  148 

found  in  London  ;    early 

fifteenth  century  ;  pro- 
bably French  ;  plate 
xii,  v;  16,  148 

plated     with     pewter  ; 

found  in  London  ;  late 
seventeenth  century  ; 
plate  xix,  vii;  28,  151 

plated  with  tin  ;    seven- 

teenth century  ;  plate 
xvin,  iv ;  26,  151 

with     remains     of     tin 

plating  ;       seventeenth 
century  ;     plate    xviu, 
v;    26,   151 

plated  with  tin  ;    handle 

"  Pied  de  Biche  "  (mark 
same  as  No.  5 ) ;  plate 
xvin,  vi ;  26,  151 

with     remains     of     tin 

plating  ;    handle  "  Pied 
de      Biche  "  ;       seven- 
teenth   century  ;     plate 
xvni,  vii  ;    26,  151 

rat-tailed  ;  plated  with 

tin  ;  handle  "  Pied  de 
Biche  "  ;  seventeenth 
century  ;  plate  xix,  i  ; 
28,  151 

rat- tailed,  found  in  York 

Road,  Westminster  ; 
plated  with  tin  ;  seven- 
teenth century  ;  plate 
xix,  iii;  28,  151 

rat-tailed,  found  in  York 

Road,  Westminster  ; 
handle  "Pied  de  Biche  "  ; 
seventeenth  century  ; 
plate  xvni,  viii; 26,  151 

181 


OLD  PEWTER 


Spoons — cont. 

Laton,  seal-top,  found  in  Lon- 
don ;  seventeenth-cen- 
tury ;  plate  xvn  i  ; 
24,  150 

seal-top ;  seventeenth 

century  ;  plate  xvn, 
ii;  24,  150 

seal-top ;      seventeenth 

century  ;      plate     xvu, 
iii  ;    24,  150 

seal-top,  found  in  Lon- 

don ;  seventeenth  cen- 
tury ;  plate  xvu,  iv  ; 
24,  150 

seal-top,  found  in  Lon- 

don ;  seventeenth  cen- 
tury ;  plate  xvu,  v  ; 
24,  150 

seal-top,  found  in  Lon- 

don ;  seventeenth  cen- 
tury ;  plate  xvu,  vi  ; 
24,  150 

seal-top,  found  in  Lon- 

don ;  seventeenth  cen- 
tury ;  plate  xvu,  vii ; 
24,  150 

seal-top,  found  in  Lon- 

don ;  seventeenth  cen- 
tury ;  plate  xvu,  viii  ; 
24,  150 

seal-top,  with  remains  of 

tin-plating,  found  at 
Norwich  ;  seventeenth 
century  ;  plate  xx,  viii  ; 
29,  150 

slip-top,  plated  with  tin  ; 

found  in  London  ; 
seventeenth  century  ; 
plate  xvui,  i  ;  26,  151 

slip-top,   found   at  Ber- 

mondsey  ;  has  re- 
mains of  tin  plating  ; 
seventeenth  century  ; 
plate  xvi,  v  ;  22 

slip-top,    found   at   Ber- 

mondsey  ;  seventeenth 
century  ;  plate  xvi,  vi  ; 

22,   149 

182 


Spoons — cont. 

Laton,  slip-top,  found  in  Lon- 
don ;  has  remains  of 
tin  plating  ;  seven- 
teenth century  ;  plate 
xvi,  vii ;  22 

slip-top,  found  in    Lon- 

don ;  seventeenth  cen- 
tury ;  plate  xvui,  iiij; 
26,  151 

spoon,  found  in  London  ; 

stem  of  hexagonal  sec- 
tion ;  seventeenth  cen- 
tury ;  plate  xiv,  mf; 
20,  150 

spoon,  with  Apostle  topr; 

seventeenth  century^; 
plate  xiv,  iv  ;  20,  150" 

dog-nose  gravy  spoon, 

plated  with  tin  ;  seven- 
teenth century  ;  plate 
xx,  ii;  29,  150 

Pewter,  found  in  London  ; 
sixteenth  century  ;  plate 
xiv,  i  ;  20,  150 

seventeenth       century  ; 

plate  xiv,  ii ;    20,  1 5 1 

found  in  London  ;  seven- 
teenth century  ;  plate 
xiv,  vii  ;  20 

found  in  London  ;    late 

eighteenth  century  ; 
plate  xx,  i;  29,  152 

found     in    Bermondsey, 

late  eighteenth  century ; 
plate  xx,  iii,  29,  152 

with  Apostle  top,  found 

in  London  ;  sixteenth 
century  ;  plate  xiv, 
viii  ;  20,  150 

rat-tailed,  found  in  New- 
gate Street ;  handle 
"  Pied  de  Biche  "  ; 
seventeenth  century  ; 
plate  xix,  ii;  28,  151 

rat-tailed,  found  in  Ber- 

mondsey ;  seventeenth 
century  ;  plate  xix,  iv  ; 
28,  151 


INDEX 


Spoons — cont. 

Pewter, rat- tailed, found  in  Ber- 
mondsey  ;  seventeenth 
century  ;  plate  xix,  v  ; 
28,  151 

rat-tailed,  seventeenth 

century  ;  plate  xix,  vi  ; 
28,  151 

rat  -  tailed  chocolate 

spoon,  found  in  the 
Wandle,  at  Wands- 
worth  ;  seventeenth 
century  ;  plate  xix,  ix  ; 
28,  151 

seal -headed,    hexagonal, 

found  at  Bermondsey  ; 
sixteenth  century ;  plate 
xiv,  v;  20,  150 

slip-top,  found  at  Ber- 
mondsey, stem  of  hexa- 
gonal section  ;  fifteenth 
century  ;  plate  xin,  i  ; 
1 8,  149 

slip-top,  found  in  Lon- 
don ;  sixteenth  cen- 
tury ;  plate  xin,  ii ; 
18,  149 

slip-top,  sixteenth  cen- 
tury ;  plate  xin,  iii  ; 
i 8,  149 

slip-top  ;  seventeenth 

century  ;  plate  xin,  iv  ; 
18 

slip-top,  found  in  Lon- 
don ;  sixteenth  cen- 
tury ;  plate  xin,  v  ; 
18,  149 

slip-top,  found  in  York 

Road,  Westminster  ; 
sixteenth  century  ;  plate 
xin,  vi  ;  1 8,  149 

slip-top,  found  in  Lon- 
don ;  sixteenth  cen- 
tury ;  plate  xin,  vii  ; 
18,  149 

slip-top,   found  in   Lon- 

don ;  seventeenth  cen- 
tury ;  plate  xin,  viii  ; 
18 


Spoons — (cont,) : 

Pewter,  slip-top,  found  in  York 
Road,  Westminster ;  six- 
teenth century ;  plate 
xv,  i  (same  mark  as  No. 
6)  ;  plate  xin  ;  21,  149 

slip-top,  found  in   Lon- 
don ;      sixteenth     cen- 
tury    (same    mark    as 
No.  6,  plate  xin)  ;  plate 
xv,  ii  ;    21,  149 

slip-top,  the  same  as  No. 

2  ;  plate  xv,  iii  ;  21, 
149 

slip-top,  found  in  Lon- 
don ;  sixteenth  cen- 
tury ;  plate  xv,  iv  ;  21, 
149 

slip-top,    sixteenth   cen- 

tury ;  plate  xv,  v  ;  21, 
149 

slip-top,   found  in  Lon- 

don ;  sixteenth  cen- 
tury (same  mark  as 
No.  2)  ;  plate  xv,  vi  ; 

21,  149 

slip-top  ;  sixteenth  cen- 
tury ;  plate  xv,  vii  ;  21, 
149 

found   in    Bermondsey  ; 

sixteenth  century  ;  plate 
xv,  viii ;  21,  149 

slip-top,   found  in  Lon- 

don ;  sixteenth  cen- 
tury ;  plate  xvi,  ix  ; 

22,  149 

slip-top,   found  in  Lon- 

don ;  sixteenth  cen- 
tury ;  plate  xvi,  viii  ; 
22 

slip-top,  found   at   Ber- 

mondsey ;  sixteenth 
century  ;  plate  xvi,  i  ; 
22,  149 

slip-top,  found  in  Lon- 
don ;      sixteenth     cen- 
tury ;  plate  xvi,  ii  ;   22, 
149 

slip-top,  found   at   Ber- 


183 


OLD  PEWTER 


Spoons — COM/. 

mondsey  ;  sixteenth 
century  ;  plate  xvi,  iii  ; 
22,  149 

Pewter,  slip-top,  found  at  Ber- 
mondsey  ;  sixteenth 
century  ;  plate  xvi,  iv  ; 
22,  149 

slip-top,  found  at  Ber- 

mondsey  ;    seventeenth 
century  ;     plate    xvm, 
ii  ;    26,  1 5 1 

Sacramental        Spoon.         See 

Church  Vessels,  &c. 
Table-Spoons,    three,   marked 
John     Yates,     &c.  ;      plate 
XI,  v  ;    14,    152 
Tea  Spoon,  marked  Ai   &c.  ; 

plate  xi,  vi;    14,  152 
Stannaries,  Edward  I.   confirms 

Charter  to,   56 

Steele,  Richard,  reference,  139 
Students'    Beer   Jug.     See   Beer 

Jugs 

Suetonius,   his  reference  to  the 
substitution     of     pewter     for 
silver  vessels,  45 
Sugar-Basins,    Castors,    &c.  : 
Sugar- Basin  ;    English  ;    early 
nineteenth        century  ; 
plate  xcn,  ii  ;    126 

Belgian  ;     plate    en,    v  ; 

140 

Sugar-Box  and  Cover,  Dutch  ; 
dated  1751;  plate  xc,  ii  ; 
125,  162 

Sugar-Castor  ;  plate  LXXXIII, 
iii;  117 

marked  with  Cupid,  &c.  ; 

plate  viu,  ii  ;    10 
Sugar-Sifters  ;   eighteenth  cen- 
tury ;     plate    LXXXV,    i,     ii ; 
118 

Sugar      Sprinkler,       Belgian  ; 
plate  en,  ii ;    140 
—  Sprinkler,  Belgian  ;   plate 
cii,  iv  ;    140 


TABLE  Plate,  161 

Table  Spoons.     See  Spoons 

Tankards  and  Cups  : 

Cup,  on  lid  1721,  &c.  ;  plate 
XLII,  i;  62,  156 

inscribed  Johannes 

George     Reichel,     &c., 
1693  ;     plate    XLII,    iii  ; 
62,  156 

German     Guild,    seven- 

teenth century  ;  plate 
xxxn,  ii  ;  48,  156 

Scotch  ;  eighteenth^cen- 

tury  ;  plate  XLI,  i  ;  61, 
156 

Loving  Cup  ;  English  ;  plate 
xxvm,  i;  42,  153, 
156 

eighteenth        century  ; 

plate  XLI,  ii  ;    61,  156 

Peg   Tankard,    Danish  ;    with 
engraved      decoration      in  - 
scribed      Kleinreide,      &c.  ; 
plate  XLVIII,  i  ;    70 
Tankard  ;       late      eighteenth 
century  ;   plate  cm,  iii  ; 
142,  157 

engraved  with  IAM  be- 

neath an  anchor,  &c.  ; 
English ;  eighteenth 
century  ;  plate  ix,  iii  ; 
12,  157 

English  ;  early  eighteenth 

century  ;     plate   LXXIV, 


i;    106 
English  ; 

century 

v;    106 
English  ; 

century  ; 


eighteenth " 
plate   LXXIV, 


eighteenth 
plate    xcin, 
11 ;     128,    157 
Flemish  ;        seventeenth 
century  ;     plate    x,     i ; 

13 

French  ;  eighteenth  cen- 
tury ;  plate  x,  iii  ;  13 

German  Guild  ;  dated 
1645  ;  plate  XLIX,  iii  ; 
72,  157 


184 


INDEX 


Tankards  and  Cups — cont. 

Tankard ;  German  ;  eighteenth 
century  ;  plate  XLVII, 
ii  ;  68,  157 

Scotch  ;  eighteenth  cen- 
tury ;  marked  "  Gal- 
braith  Glasgow  "  ;  plate 
XLIV,  i;  64,  157 

Cup,  Swedish,  1844  ;  marks, 
Arms  of  Sweden,  &c.  ; 
plate  XLVIII,  iii  ;  70, 

157 

Tankards,  German  Guild  ; 
plate  LII,  i,  ii,  iii  ;  76 

German  Guild  ;  seven- 
teenth to  eighteenth 
century  ;  plate  LIII,  i, 
ii,  iii;  77,  159 

German ;  seventeenth 

century  ;  plate  XLVI, 
i-iv  ;  68 

German  ;  seventeenth 

and  eighteenth  cen- 
turies ;  plate  XLV,  i,  ii, 
iii  ;  68 

German,  A  Pair  of, 

eighteenth  century ;  plate 

LI;    74,   159 

Wine  Cups  ;    eighteenth  cen- 
tury ;      no     marks  ;      plate 
XLI,  iii,  iv;    61,  156 
Taper-Holder.     See  Candlesticks 
Tappit-Hen.     See  Measures 
Tea-Pots  : 

Tea-Pot  ;  eighteenth  century  ; 
plate  xciv,  ii  ;  130 

early  nineteenth  century  ; 

plate  LXXXVIII,  i  ;  122, 
161 

early  nineteenth  century  ; 

plate  xciv,  i  ;    130 
Dutch  ;    eighteenth  cen- 
tury ;    plate  LXXXIX,  i  ; 
124 

English  ;    plate  xcm,  i  ; 

128 

.  English ;  early  nine- 
teenth century ;  plate 
xcn,  iii  ;  126 


Tea-Pots— -cont. 

Tea-Pot,  Flemish  ;  eighteenth 
century  ;   plate  LXXXIX, 
iii  ;    124 
Temperance    salver    by    Briot, 

references,  79,  95 
Tests  for  ascertaining  the  quality 

of  pewter,  17,  16 
Tin :  reputation  of  English 
pewter  due  to  its  superior 
quality,  30 ;  early  Biblical 
and  Classical  references  to, 
30,  31  ;  craft  in  Scotland 
retarded  by  its  absence,  69  ; 
mentioned  by  Shakespeare, 
87  ;  London  Company  privi- 
leged to  charge  royalty  on, 
1598,90 
Tobacco-Boxes  : 

Tobacco-Box ;  no  marks ;  plate 
xcvui,  iii ;  133 

eighteenth  century  ;  plate 

xciv,  iii  ;  130,  162 
English  ;  eighteenth  cen- 
tury ;    plate  LXXIV,  vi  ; 
1 06,  162 

French  ;  eighteenth  cen- 
tury ;  plate  xcv,iii  ;  132, 
162 

Toddy-Ladles.     See  Ladles 
Torigny,  Robert  de  (1154-1186), 
early  pewter  plaque  recording 
his  name  and  title,  53 
Tottenham    Court    Road,    Wine 
Taster  dug  up  in,  plate  LVII,  80 
Trays  : 

Tray  ;  eighteenth  century  ; 
plate  XL,  ii  ;  60 

or  Salver,  Dutch  ;    plate 

LXXVI,  ii ;    109,  162 
Triflers,    workers    in    the    metal 
known    as    "  trifle,"    27,    28  ; 
in  1612  included  hollow-ware 
in  their  output,  96 
Troyes,  references,  59,  60 

URNS  : 

Urn,  Dutch;  plate  LXXXVII. 
i  ;  121,  161 


2A 


OLD  PEWTER 


Urns — cont. 

Urn,   Dutch ;    plate   LXXXVII, 
iii  ;    121,   161 

French  ;    plate  LXXXVII, 

ii ;    121,  161 

Uvedale,  Lady  Elizabeth,  her 
bequest  of  pewter,  1487,  69 

VEGETABLE  DISH  ;  English  ;  late 
eighteenth  century ;  plate 
LXXVI,  i  ;  109,  161 

Vessels  and  Dish,  Roman,  found 
by  Rev.  C.  H.  Engleheart ; 
plate  in,  i,  ii,  iii  ;  4,  47,  48,  49 

Victoria  and  Albert  Museum, 
Gloucester  Candlestick  at, 
plate  iv,  5,  54  ;  specimens  of 
decorated  pewter  at,  80  ;  Nu- 
remberg Salver  there,  a  re- 
production of  Briot's  Tem- 
perantia  Salver,  95.  See  also 
frontispiece  and  plates  ix,  xxi, 

XXII,  XLVIII,  XLIX,  LIV,  LVII,  XC 

Vinegar  Cruet ;  plate  LXXXIII, 
iv;  117,  161 

Waltham  Abbey  Church,  pewter 

purchased  for,  80 
WANDSWORTH,  rat-tailed  pewter 
chocolate     spoon     found     at, 
plate  xix,  ix;  28 
Water-Jugs  : 

Hot- Water  Jug  ;  Dutch  ;  early 
eighteenth  century ;  plate 
LXV,  ii;  94 

Water-Jug ;  English ;  eigh- 
teenth century  ;  plate 
LVIII,  ii;  82,  159 
English  ;  eighteenth  cen- 
tury ;  plate  LXXVI,  iv  ; 
109 

English  ;   late  eighteenth 

century  ;     plate  LXXIII, 
iv  ;    104 


Welch,  Mr.,  "History  of  the 
Pewterers'  Company,"  26  ;  list 
of  moulds,  33  ;  reference, 

95 

Westminster,  spoons  found  at : 
plates  xn,  i,  16  ;  xni,  vi,  18  ; 
xv,  i,  21  ;  xvin,  viii,  26  ; 
xix,  iii ;  28 

Weymouth,  reference,  32 

White  iron-smiths,  their  absorp- 
tion of  the  craft,  107,  no,  112, 
113,  116,  117 

William  III.  compelled  to  strike 
part  of  his  coinage  in  pewter, 
1 08  ;  reference,  106 

Wine  Measure.     See  Measures 

Wine  Taster,  English ;  seven- 
teenth century  ;  dug  up  in 
Tottenham  Court  Road  ;  plate 
LVII,  i  ;  80,  1 60 

Wolsey,  Cardinal,  plate  belonging 
to,  91 

Woman,  A,  belongs  to  the  craft 
in  the  thirteenth  century, 

Wood,  Mr.  Ingleby,  "Scottish 
Pewter-Ware  and  Pewterers," 
reference,  26  ;  "  Church  Ves- 
sels before  and  after  the  Re- 
formation," reference,  133  ; 
references,  69,  76,  81,  89,  98, 
124,  125,  131,  158 

Wright,  James,  of  St.  Andrews, 
reference,  117 


YATES,     JOHN,     his     mark     on 

Spoons  and  Ladles,  plate  xi, 

iii,  v  ;  14,  152 
York,  the  principal  trading  place 

for  the  North  of  England  in 

1419,  64,  77 
Yorkshire,  Patens  from  a  Church 

in,  plate  xxn,  iii;  32 


Printed  by  BALLANTYNE  &*  Co.  LIMITED 
Tavistock  Street,  London 


o 


PLEASE  DO  NOT  REMOVE 
CARDS  OR  SLIPS  FROM  THIS  POCKET 

UNIVERSITY  OF  TORONTO  LIBRARY 


NK 

84  OM 
BM5 
C.I 
ROBA