Post-Medieval Archaeology 44/1 (2010), 1–53

Five centuries of iron working: excavations at

Wednesbury Forge

By PAUL BELFORD

SUMMARY: Archaeological excavations undertaken between 2004 and 2008 at Wednesbury

Forge, Wednesbury, West Midlands encountered extensive remains of timber and masonry
structures and other features. Historical and archaeological evidence revealed a sophisticated
ironworking complex in existence by c. 1600, which was subsequently continually adapted and
redeveloped until the site closed in 2005. Processes included finery and chafery forges, nail-making,
saw-making, gun-making and edge-tool manufacture. Later developments included a wind-powered
grinding mill, internal railway networks, water turbines, rolling mills, housing and workers’
recreational facilities. Archaeological investigations comprised documentary research, excavation,
building recording, oral history and process recording.

INTRODUCTION WEDNESBURY AND ITS INDUSTRIES

Despite the evidently Saxon origins of Wednesbury
This paper describes the results of archaeological
research and investigation undertaken by Iron- as a place name,2 the settlement itself may have
bridge Archaeology at Wednesbury Forge between been preceded by an Iron Age hill fort on what
2001 and 2007. The site was in continuous use as is now Church Hill.3 At Domesday the Manor
an iron forge from the late 16th century to the of Wednesbury was part of Crown lands and
early 21st century. Wednesbury Forge was situated possessed a corn mill.4 The original nucleus of the
in the modern Metropolitan Borough of Sandwell medieval settlement was probably near Church
(historically part of south Staffordshire). It was Hill or within the market place, later spreading to
located approximately 1.5km north-east of the the north-east.5 Industrial activity began early in
medieval town centre of Wednesbury, and 0.5km Wednesbury, with coal and iron being mined in
south-east of Junction 9 of the M6 motorway the area by 1315.6 Geology influenced the early
(NGR SP 0020 9620). The forge was situated in the development of the pottery industry in Wednes-
valley of the River Tame, from which it drew its bury, and potters were first noted in 1422.7 By the
water supply (Fig. 1). The underlying geology is 16th century a substantial ceramic industry had
that of the Coal Measures, specifically the so-called
emerged, which continued to dominate the region
‘Ten Yard Seam’ or ‘Thick Coal’ (in fact, about
a dozen closely overlying seams) of the South until eclipsed by the 18th-century expansion of
Staffordshire Coalfield, which lies very close to the the north Staffordshire potteries.8 The earliest
surface and outcrops in places.1 The upper part of period of pottery production included Cistercian-
this seam was encountered during archaeological type wares and Midlands purple ware, as well
fieldwork, and was cut through by the construction as so-called ‘Wednesbury ware’ (a redware).
of the forge. This is overlain by deposits of boulder Wednesbury pottery products are well known
clay, sands and gravels deposited during the last from contexts across the English Midlands, and
Ice Age. recent archaeological fieldwork in Wednesbury

© Society for Post-Medieval Archaeology 2010 DOI: 10.1179/174581310X12662382629094

44-1-1-PMA 01.indd 1 3/1/2010 7:34:59 PM

 2 PAUL BELFORD

FIG. 1
Wednesbury Forge: site location, showing standing buildings in 2001 and extent of interventions in 2005 (evaluation
trenches in black, numbered) and 2006–7 (area of excavation shown shaded).

has recovered evidence in the form of kiln bases was probably converted to use as a fulling mill by
and waster dumps.9 1423.11 After the Dissolution this site was eventu-
The River Tame provided power to several ally converted to an iron forge and later became
installations in Wednesbury, including the manorial known as Wednesbury Bridge Forge. The heyday
corn mill noted at Domesday. This was demised of this ironworks was in the later 18th century
by William de Heronville (then lord of the manor) when it was occupied by John Wood, who ‘made
to Bordesley Abbey c. 1230, but by c. 1280 the mill a fortune as contractor for the making of Irish
had been sub-let to Sir Thomas Hillary of nearby coinage’.12 By the 1820s it had been re-converted to
Bescot.10 Rivalry between the de Heronville and a corn mill, and it finally closed in 1885.13
Hillary families during the 13th and 14th centuries Other ironworking sites included Sparrow’s
resulted in the construction of a second corn mill Forge, which was a short-lived 18th- and 19th-
in Wednesbury, and the original corn mill site century enterprise; originally horse-powered and

44-1-1-PMA 01.indd 2 3/1/2010 7:35:00 PM

 WEDNESBURY FORGE 3
14
later relocated and converted to water power. Forge’. In 1970 Wednesbury Forge was taken
Iron smelting does not seem to have taken place in over by Spear and Jackson, who continued to
Wednesbury until the late 18th century, although manufacture edge tools there until 2005.
there may have been a furnace in the area as early
as the 1590s.15 THE SITE IN ITS LANDSCAPE SETTING
Wednesbury Forge (the subject of archaeo-
logical fieldwork described here) was a separate The original layout of the site was determined
entity from either of the two corn mills, the fulling by water-power arrangements. This layout was
mill and the Wednesbury Bridge Forge. The detailed fossilized by subsequent developments, and even
history of the Wednesbury Forge site is presented when water power was finally abandoned in the
below; suffice to say here that the earliest docu- early 20th century the form of the landscape
mentary reference is to a ‘riotous and unlawful respected these antecedents. Water power was
assembly’ in 1597, when workers from Wednes- provided by the River Tame, which flows roughly
bury Bridge Forge descended on Wednesbury Forge from west to east through the area. For most of the
armed with shovels and axes.16 In 1606 Wednes- site’s history, a headrace from the Tame diverted
bury Forge was leased by William Comberford water to a pair of pools to the west of the forge.
(then lord of the manor) to Walter Coleman.17 By During the 16th and 17th centuries the northern
the late 1650s the forge was occupied by the Foley pool was predominant; in the 18th, 19th and 20th
partnership, who later sub-let it to various tenants. centuries a southern pool was also in use.18 For
In 1704 the site was let to John Willetts, beginning a period during the 18th century the southern pool
four generations of occupation by that family, who was the principal pool. Throughout the forge’s
manufactured saws and guns. In 1817 the forge existence the main area of forging activities was
was let to Edward Elwell, an edge tool maker, situated to the east of both pools, with water
who bought the site outright in 1831. Even to the supply regulated at the point where they met
present day the site is locally known as ‘Elwell’s (Fig. 2). For most of the 16th to 18th centuries the

FIG. 2
Wednesbury Forge in its landscape setting, based on the 1903 Ordnance Survey map and showing the River
Tame and tributaries.

44-1-1-PMA 01.indd 3 3/1/2010 7:35:02 PM

 4 PAUL BELFORD
forge complex was confined to an area approxi- that summer’s exceptionally wet weather). This
mately 50m (E–W) by 30m (N–S) immediately east concentrated on 16th-, 17th- and 18th-century levels.
of the pools. The outflow (tailraces) of the water- Post-excavation analysis and specialist reporting
power system also proved a resilient arrangement, took place during 2007 and 2008, with the final
the original tailraces being culverted in the 17th ‘grey literature’ report submitted to Sandwell
and 18th centuries and remaining in use well into Metropolitan Borough Council (MBC) Historic
the 20th century. Environment Record in 2009.

PROJECT BACKGROUND PRESENTATION OF RESULTS

In 2001 Ironbridge Archaeology were commis- Despite the inevitable complexities of a site in
sioned by Crouch Butler Savage Limited (on behalf continuous industrial use for five centuries, it
of their clients Opus Land) to undertake an His- was possible to arrive at a close correspondence
toric Landscape Appraisal of land at Wednesbury between the archaeological sequence and the his-
Forge. This was part of a planning application torical record. Therefore the story of each period
for redevelopment of the land for employment in the site’s development is presented as a self-
purposes; the desk-based assessment was required contained narrative. The documented history
as part of the pre-determination conditions. The enhances the already vivid story told by the archae-
desk-based report identified areas of archaeologi- ology. Archaeology is the act of recovering the past
cal potential, and recommended field evaluation directly from the earth; history — the recovery of
of the site to determine the nature, extent and state the past through the foggy lens of written sources
of preservation of any archaeological remains.19 — is, at Wednesbury, the handmaiden of archaeol-
The development project involved work on several ogy. Description of artefacts and specialist analysis
different landholdings, and due to the complexities is also incorporated into the main body of the nar-
of pre-development negotiation, fieldwork did rative. Inevitably this narrative is an abridgement
not take place until 2005. of the archaeological sequence. Those seeking
The evaluation comprised two stages of trial more detailed analysis of individual elements can
trenching, in April–May and July–August 2005 await the publication of a monograph (currently
(Fig. 1).20 Nine trial trenches were excavated, in preparation), or inspect the relevant ‘grey litera-
focusing on the areas identified as the historic core ture’ reports at the Sandwell MBC Historic Envi-
of the forge (Trenches 1, 2 and 3), surrounding ronment Record, or explore the excavation archive
areas identified as 18th- and 19th-century develop- deposited with Wednesbury Museum.
ment from historical sources (Trenches 4, 5 and
6/6a), and ancillary areas such as managers’ and
PERIOD 1: COMBERFORD’S FORGE (c.
workers’ housing and water-power regulation
1585–1656)
(Trenches 7, 9 and 10). Process recording of forging
operations was also undertaken.
The date of c. 1585 is suggested for the commence-
The results of the field evaluation suggested
ment of operations at Wednesbury Forge. This is
that substantial well-stratified remains of the his-
based on the renewal of an earlier lease in 1606, in
toric core of the forge survived. Since the develop-
which the already-extant forge was described as
ment area included a number of relict watercourses ‘decayed’.21
and made ground of uncertain solidity, develop-
ment was best enabled by the removal and consoli-
dation of such features. Consequently the decision PHASE I: 16TH CENTURY
was taken to mitigate the archaeological impact of The earliest activity on site involved the construc-
the development through preservation by record. tion of two timber-framed wheelpits and their
This took the form of open-area excavation of the associated tailraces. These correspond with the
historic core of the forge, together with isolated 1606 description of an existing finery and chafery,
interventions in outlying areas and a watching powered by water supplied from dammed water-
brief on redevelopment (Fig. 1). Open-area excava- courses with floodgates.22 Wheelpit 1 [2065, 2146]
tion was undertaken in two main phases. The first was located at the northern end of the site, with
took place between March and September 2006, Wheelpit 3 [1869, 2125] situated some 15m to the
and focused on identifying the full extent of the south (Fig. 3). Both wheelpits and associated
historic forge complex and recording 18th-, 19th- tailraces were of similar construction, although
and 20th-century levels. The second phase took Wheelpit 3 and its associated tailrace were heavily
place between January and June 2007, with further truncated by later disturbance, notably the
work in July and September 2007 (interrupted by insertion of a steam engine in Phase XI.

44-1-1-PMA 01.indd 4 3/1/2010 7:35:03 PM

 WEDNESBURY FORGE 5

FIG. 3
Wednesbury Forge: Period 1, Phases I and II. Overall site plan showing 16th- and early 17th-century features.

The northern wheelpit (Wheelpit 1) was the A trench [2066] oriented east–west and
best-preserved example, due to its abandonment approximately 2m wide, had been cut through the
early in the forge’s history. Its construction underlying natural clay. This was then packed with
methods were employed in subsequent wheelpit levelling layers [2018] and [2019]. Two parallel oak
and tailrace structures, so it is described in detail beams, each c. 5m long and 0.15–0.2m square in
here to avoid later repetition. section, were laid 0.5m apart along this cut, and

44-1-1-PMA 01.indd 5 3/1/2010 7:35:03 PM

 6 PAUL BELFORD
were braced with a series of lateral oak timbers and survived to around 1m high. Oak planks were
spaced c. 1m apart. Upright timbers were mortised then nailed to both sides of the resulting box-frame,
into the longitudinal base beams. These uprights forming side-boards for the interior of the wheelpit
averaged approximately 0.15–0.2m square in section and revetting for the exterior (Fig. 4).

FIG. 4
Wednesbury Forge: Wheelpit 1. Top: plan as excavated, showing components and phasing. Bottom: isometric
reconstructed drawings as built and exploded views.

44-1-1-PMA 01.indd 6 3/1/2010 7:35:05 PM

 WEDNESBURY FORGE 7
The rear of the wheelpit had a floor of oak construction of the timberwork, the foundation
planks which curved upwards, secured together trench was then backfilled with redeposited clay
with an interlocking lap joint and pegged together [2067].
using dowels. Curving side-boards were fixed As with Wheelpit 1, construction of Wheelpit
to these base planks with nails (Fig. 5). This con- 3 comprised an oak base frame of longitudinal
struction provided an efficient close fit for a water- members joined with lateral cross-sleepers [1869]
wheel. The resulting wheelpit would have been and [2125] (Fig. 6). Most of the wheelpit was
approximately 0.4m wide. Fragments of straight destroyed by 18th- and 19th-century intrusions,
bucket-boards, on average 0.3m wide with cham- and so it was not possible to determine the original
fered ends, were found beneath the later phase location or size of the water-wheel. The tailrace
of this wheelpit (see below). Wear-marks were had been truncated to the level of the base timbers,
evident on the southern internal side-boards of so uprights and side-boards did not survive.
the wheelpit, and from these the diameter of the
water-wheel was estimated at approximately 4m. PHASE II: EARLY 17TH CENTURY
Assuming that pond levels were consistent between
the 16th and 19th centuries, this water-wheel would In 1597 Thomas and Richard Parkes, together with
have been breast-shot. five labourers, were alleged to have broken into
This method of planked box-frame construc- Wednesbury Forge and expelled William Comber-
tion continued eastwards to form the tailrace, ford and William Whorwood. Whorwood’s iron
although unlike the wheelpit the floor of the mill at Perry Bar was also visited by Parkes; Whor-
tailrace was clay-lined and did not have timber wood retaliated by attacking Parkes’ ironworks at
boarding. Although truncated, fragments of Tail- Perry Bar and Handsworth.23 These actions clearly
race 1 were noted 22m east of the wheelpit. After put the forge out of business for some time, and

FIG. 5
Wednesbury Forge: Wheelpit 1 during excavation. North-facing view of the western end of the structure showing the
wheelpit curvature.

44-1-1-PMA 01.indd 7 3/1/2010 7:35:07 PM

 8 PAUL BELFORD
fractured. This suggests a quite serious accident,
such as the unseating of the wheel from its bear-
ings, creating excessive strain on the box-frame
structure and causing it to fail. It is tempting to link
this with sabotage during events in 1597, although
it could have resulted from ordinary wear and tear.
The rebuilding inserted a new base timber [2105]
along the northern side of the wheelpit, into which
new and reused uprights from the original con-
struction were inserted and side-boards re-fixed.
This alteration slightly widened the structure, but
the inner dimensions of the wheelpit itself did not
change (Fig. 4). Dendrochronology on this new
timber gave a date of 1524, although the absence
of sapwood in the sample probably renders this
significantly too early to be a reliable indicator of
felling or construction.
With Wheelpit 1 back in running order, Cole-
man turned his attention to the wider water
supply situation. He built a ‘weere or stank’ on
the northern mill race to Wednesbury Forge.27
Coleman evidently felt that these repairs were
sufficient to permit the rent reduction, and gave
Comberford notice to this effect. However, Comb-
erford was clearly unhappy and brought a Chan-
cery suit against his son-in-law (one wonders
FIG. 6 what his daughter thought about this), in which he
Wednesbury Forge: Tailrace 3 during excavation, alleged that Coleman had failed to repair the forge.
east-facing view (scale 0.3m). In reply, Coleman argued that a lot of this work
had already taken place, and further work was
ongoing.28
in 1606 Comberford leased the forge to his son-in- Comberford was regularly bringing Chancery
law, Walter Coleman of Cannock, in partnership suits against various tenants, and in this case the
with Thomas Chetwynd.24 Coleman and Chetwynd archaeological evidence unequivocally supports
had both been involved in iron forges in the Coleman’s defence. In fact, Coleman had con-
Cannock area; Chetwynd later going on to develop structed two entirely new and extremely sophisti-
the slitting mill at the now-eponymous village and cated water-powered forging installations during
establishing a family concern that had interests the first twelve years of his lease. Both of these
in forges and furnaces elsewhere in Staffordshire were double-channel systems, containing two par-
later in the 17th century.25 allel but offset water-wheels in order to power a
The terms of the 1606 lease required Coleman bellows and adjacent hammers.
to rebuild and repair the forge, and to pay rent
of £20 per annum. This rent could be reduced at WHEELPIT 4
one year’s notice to £5 13s. 4d., providing he only
worked the forge when Comberford could spare Wheelpit 4 and its associated tailrace were built
the water.26 There would have been little incentive approximately 10m south of the then still-extant
for Coleman to make significant improvements Wheelpit 3. Construction was similar to Wheelpit
to the forge, since these would have increased his 1, except that three parallel sets of longitudinal
demand for water and prevented him from paying base timbers were deployed to create a double
reduced rent. wheelpit (Fig. 7). Dendrochronology gave a con-
struction date of c. 1603–14; supported by a yellow
ware jug found in the fill [2158] of the construction
REPAIRS TO WHEELPIT 1 cut. Only the very base of the western end of this
It is perhaps not surprising that Coleman first structure survived (a total length of about 15m),
repaired the existing infrastructure. During the the wheelpit (eastern) end having been removed by
16th century, the mortises along the original the insertion of a brick wheelpit in the 18th century,
northern longitudinal base beam of Wheelpit 1 had and the tailrace truncated by later culverting.

44-1-1-PMA 01.indd 8 3/1/2010 7:35:09 PM

 WEDNESBURY FORGE 9
infrastructure whilst still maintaining production.
Perhaps encouraged by the pending case in Chan-
cery, he turned his attention to the centre of the site.
Adjacent to the old Wheelpit 3 Coleman constructed
another ambitious double-wheel installation which
was evidently intended to replace it.

WHEELPIT 2
Wheelpit 2 (again two wheelpits) was the most
complete surviving example of these early 17th-
century timber installations (Fig. 8). This was 2.3m
wide overall, comprising two parallel wheelpits
and tailraces constructed using three oak base
beams braced by lateral cross-sleepers; the total
surviving length was 29m. Carpenters’ numbering
marks were evident on some of the timbers (Fig. 9).
The northern wheelpit at the western end of this
structure retained its timber base-boards in situ
(Fig. 10). To the north of this were some substan-
tial timbers [2045, 2089]: the relict substructure of
a bellows or anvil installation. Dendrochronology
gave a construction date of between 1616 and
1618 — evidently the work that was ongoing at the
time of the Chancery case.
Ian Tyers’ examination of the Wednesbury
timbers found that they came from trees which
had suffered ‘rapid growth reductions followed
by either steady recovery or permanently slowed
growth’.29 This growth pattern resulted in extreme-
ly dense, gnarled timbers, possibly the result of
lopping of hedgerow or coppice woodland. The
knots and compression wood within these timbers
meant they would have been less likely to split
than straight-grained wood, and so may have been
deliberately selected to allow for the stresses and
strains of machinery. Although it was not possible
to be more than regionally specific about the sup-
ply of timber, the nearby woodlands of Cannock
Chase (where Coleman was born) would have been
of the right character to provide this material.
FIG. 7 A small fragment of water-wheel was found
Wednesbury Forge: Tailrace 4, plan as excavated. in the fill of the northern channel of Tailrace 2
(Fig. 11). This consisted of one section of sole
board, into which were cut three 20mm grooves;
The two channels merged into a single tailrace these held the bucket boards or paddles which
approximately 20m downstream from the wheel- were fixed by nails. This fragment would have been
pit; this fed into an open tailrace which served all part of a wheel that was 4.38m in diameter, which
the water-power installations. A timber platform is consistent with the evidence from Wheelpit 1.
[2156] was located south of Tailrace 4, consisting Moreover, it suggests that the original early 17th-
of two layers of horizontally-laid timbers creating century floor level would have been approximately
a solid platform capable of supporting an anvil. 2.19m above the base of the wheelpit. Given that
With the installation of Wheelpit 4 and its Wheelpit 2 and its associated tailrace was truncat-
tailrace (actually two wheels), Coleman now had ed to a surviving height of not more than 0.3m, it
four operational water-wheels and the beginnings is perhaps not surprising that these massive timber
of a substantial forge operation. The construction structures are the only evidence for the early
of Wheelpit 4 enabled him to consider upgrading history of the site.

44-1-1-PMA 01.indd 9 3/1/2010 7:35:10 PM

 10 PAUL BELFORD

FIG. 8
Wednesbury Forge: Wheelpit and Tailrace 2, plan as excavated and cross-section.

The construction of the two double-wheel was a deliberate filling episode [2127] which con-
installations (Wheelpits 2 and 4) in the southern tained a sherd of blackware dating to the first half
part of the site rendered the earlier Wheelpit 3 of the 17th century.
redundant. The tailrace was filled in two distinct
phases. The first related to the period when the
installation was still in use: a series of laminated PERIOD 2: FOLEY’S FORGE (1656–1704)
silts [1870] containing various ironworking debris,
including finery slags, molten iron slags, smithing William Comberford died in 1625, leaving £20
hearth and hammerscale conglomerates, as well as for the poor of Wednesbury.30 Despite his various
a number of hand-forged nails. The second layer battles with rivals and tenants, and his unabashed

44-1-1-PMA 01.indd 10 3/1/2010 7:35:11 PM

 WEDNESBURY FORGE 11
Trustees, cleared the debts, and three years later
sold the manor (including the forge) to his cousin,
John Shelton of West Bromwich.32 This period can
be divided into two phases.

PHASE III: MID- TO LATE 17TH CENTURY

After acquiring Wednesbury Forge, John Shelton
promptly leased it to Thomas Foley of the well-
known family of Midlands ironmasters. In 1667
the site was transferred to Thomas’ son Philip
Foley, presumably part of the elder Foley’s settle-
ment at this time to Philip of all his Midlands
iron-making concerns.33 Wednesbury was a closely
integrated part of Foley’s Stour valley group of
FIG. 9 ironworks, pig iron coming from the blast furnace
Wednesbury Forge: carpenters’ numbering marks on at Hales and the wrought iron bar sent down to
one of the base beams of Tailrace 2. the slitting mill at Bustleholme.34 During the first
few years of this period, at least until 1672, the
forge was managed on Foley’s behalf by William
Spencer. Output of wrought iron during those
six years ranged from 92 to 141 tons per year; the
snapshot provided by the year-end account taken
on 27 March 1669, for example, reveals over 68
tons of cast iron and nearly 5 tons of wrought iron
on the premises, together worth nearly £470, as
well as 75 loads of charcoal worth over £78.35
This scale of operations was made possible by
various improvements and additions to the site.
Archaeologically, since the 17th-century ground
levels were removed by later developments, the
above-ground details of these improvements are
not entirely clear. However, the evidence of the
water management system, together with residual
finds from later phases, suggests a period of expan-
sion and improvements in productivity during the
third quarter of the 17th century (Fig. 12).
The principal work involved culverting the
formerly open tailraces leading from double-
wheelpits 2 and 4. This involved partial removal
of timber structures built by Coleman, and their
replacement by brick arched culverts leading to the
open watercourse. In both Tailrace 2 and Tailrace
4 existing base beams were used as foundations for
the construction of new brickwork, and the new
culverts exactly followed their predecessors’ line.
FIG. 10 In both cases the northern channel was kept in use
Wednesbury Forge: Wheelpit 2 looking west, showing whilst the southern channel was constructed, thus
timber base-boards (scale 2m). enabling the water-power system to remain in use.
In Tailrace 2 the brick culvert [1630, 1685,
1868, 2116] incorporated a square extension to the
north, possibly a sluice or regulator. Abutting the
Catholicism, he had built a successful political and south side of Tailrace 2 was a brick-built circular
business career and had sired numerous children feature, 2.25m in diameter [1855], possibly a cistern
with his two wives. The manor passed to his son for quenching; this had an outlet into the southern
William and then to his grandson, John Comber- wall of the new culvert. The fill [1860] consisted
ford, in 1653.31 John settled the estate as a whole on of very heavily cemented dark silt with slag and

44-1-1-PMA 01.indd 11 3/1/2010 7:35:12 PM

 12 PAUL BELFORD

FIG. 11
Wednesbury Forge: water-wheel fragments from Tailrace 2 (sole board, left) and Tailrace 1 (bucket fragments, right).

ash inclusions. Parts of the earlier timber structure silty clay [2069], indicating that the water supply
(including the side-boards) were retained and built had ceased altogether. The 1669 accounts refer to
around; to the east this section of culvert stopped ‘one pair of old finery bellows’, and these may
short of the main open watercourse. Finds from relate to an installation associated with Wheelpit 1.
deposits in the now-redundant northern channel
of Tailrace 2 [2061, 2135], principally coarse earth- PHASE IV: LATE 17TH CENTURY
enwares, suggest that it had gone out of use by the
final quarter of the 17th century. The culverting of The culverting of the tailraces was only the first
Tailrace 4 to the south was achieved in much the stage in upgrading the forge. In 1676 the Foleys
same way, although here the walls of the culvert sub-let the site to Humfrey Jennens, who contin-
[1173, 1174, 1182, 1183, 1566] were built on the ued to produce bar iron for the Foley concern.
outer base-frame timbers of the tailrace, thus In 1677 Foley was granted the lease of ‘the leat to
enclosing both channels (Fig. 13). Unlike its north- the forge . . . called Wednesbury Forge’,37 which
ern counterpart, culverting here extended to the evidently provided security of water supply and
edge of the main open watercourse. The new north gave him the confidence to embark on a series of
and south culverts were linked by a further brick further improvements to the water system. Repairs
culvert oriented north–south [1856], incorporating were recorded in 1678, and a further agreement
the cistern [1855]. was struck between Philip Foley and Humfrey
All of these new culverts and associated Jennens the following year.38
features were built of handmade clamp-fired bricks
bonded with sandy lime mortar. The base of these BRICK WHEELPITS
culverts was formed by compacting a mixture of
forging waste, hammerscale and slag [2157] onto The historical evidence supports the archaeologi-
the underlying clay, creating an impermeable cal record, since it was at around this time that
base. Philip Foley began to make a serious investment
Thus Philip Foley, with his manager William in the wheelpits themselves. Both wheelpits associ-
Spencer, had created an efficient finery and chafery ated with the newly-culverted Tailraces 2 and 4
forge in an enlarged working area between two were reconstructed in brick, presumably the 1678
existing double-wheel installations. Equipment at repairs. In the case of the southern wheelpit (feed-
the forge in 1669 included two pairs of finery and ing into Tailrace 4), the existing timber structures
chafery bellows, two grindstones, hammer beams were simply replaced by one in brick, on the same
and a spare anvil.36 It was probably during this east–west orientation. Two severely-truncated
period that the old timber-framed Wheelpit 1 at the walls [1641] at the very base of the later wheelpit
northern end of the site was abandoned. A period survived from the 17th-century rebuild; these
of intermittent water flow was suggested by the were curved, suggesting that the rebuilt wheelpit
accumulation of fine silts with inclusions of decom- may have been significantly wider than the pair
posed vegetation [2094]; this was overlain by clean of wheelpits which it replaced.

44-1-1-PMA 01.indd 12 3/1/2010 7:35:14 PM

 WEDNESBURY FORGE 13

FIG. 12
Wednesbury Forge: Period 2, Phases III and IV. Overall site plan showing later 17th- century features.

More dramatic was the redevelopment of the was made to reorientate the wheelpit through
water-power installation to the north. The timber 90°, and relocate it to the west. This new northern
Tailrace 2 had already been replaced by brick wheelpit could have been over 2.5m wide (ie. east–
culverts, but the replacement of the water-wheels west); however its original extent was not clear due
was not as straightforward. Instead, the decision to later alterations — the extant wheelpit which

44-1-1-PMA 01.indd 13 3/1/2010 7:35:15 PM

 14 PAUL BELFORD

FIG. 13
Wednesbury Forge: culverting the tailraces during Phase III. Top: finds from deposits associated with the infilling of
Tailrace 3 and the culverting of Tailrace 4. Bottom: west-facing view of Tailrace 2 during excavation showing how the
brick culverts rested directly on their timber antecedents (scale 2m).

replaced it measured 2.8 × 5.5m. This later wheel- curved to accommodate the water-wheel, just as its
pit incorporated a 1.5m-high section of original timber antecedent (Wheelpit 1) had done.
17th-century brickwork in its southern elevation The reason for the reorientation of this fea-
[1730]. This was constructed of handmade clamp- ture is not immediately clear. Possibly this was a
fired bricks bonded in lime mortar, and had been way in which the new wheelpit could be constructed

44-1-1-PMA 01.indd 14 3/1/2010 7:35:16 PM

 WEDNESBURY FORGE 15
whilst keeping the old wheelpits in use. However, it The 1669 accounts showed ‘10 pair bedsteads’ in
may also reflect the introduction of new processes ‘workmen’s houses’ on the site, and at least that
to the forge, such as a rolling mill. Certainly the number would have been employed at the works.40
forge was being used for ‘plating’ by the first de- These included Nicholas Record, the hammerman
cade of the 18th century.39 between 1672 and 1685, William Whiston, the finer
in 1679, and labourers including Ishmael Bomford,
NEW BUILDINGS AND OLD RUBBISH Richard Seamore and Roger Brisburn who appear
in various records during the 1680s.41 No doubt
A substantial timber building was also erected they drank from blackware mugs, sherds of which
on the site during this period, presumably to house were found in the silt of the tailraces; perhaps the
the new and expanded machinery of the forge. handmade brass thimble came from one of their
Parts of this structure were reused during Phase households.
VIII as a boring mill base. These timbers included When Humphrey Jennens died in 1690, his
reused sill beams [1473] and [1476] and joists [1487] son John took over the lease.42 This would seem to
and [2112]. This or another building incorporated be a period of decline, for the Foley accounts show
a small brick-built cellar [1682], also constructed that in 1692 only 32 tons of pig iron were purchased
in the third quarter of the 17th century, and incor- from Hales furnace.43 The extent to which John
porating into its floor [1449] in situ timbers from Jennens was directly involved with the forge is
the long-infilled former Tailrace 3. The cellar was uncertain; he employed Joshua Spurr and Thomas
reached by a curved brick staircase which showed Lowe at the site between 1693 and 1698.44 Jennens
considerable wear [1461]. seems to have spent as much time filling in the
As noted above, the old Wheelpit 1 was old Tailrace 1 as he did producing iron. This fea-
already redundant, and began to be filled during ture was deliberately filled in with clay, sand and
this period with waste and rubbish. The lower other debris, pottery from these layers [2070, 2071,
fill deposits [2091, 2147] of this feature contained 2148, 2149, 2150] firmly pointing to complete
charcoal and forging waste as well as general rub- abandonment by the later 17th century.
bish, probably representing clearance of buildings The Jennens’ initial 21-year lease would have
on site and a consequence of the new building pro- expired in 1697, although the agreement of 1679
gramme. Detritus included a 16th-century pewter may have extended this to 1700. By this time,
spoon and a handmade brass thimble (Fig. 14). however, the Foley empire had been reorganized,

FIG. 14
Wednesbury Forge: finds from the Phase IV infilling of Tailrace 1.

44-1-1-PMA 01.indd 15 3/1/2010 7:35:18 PM

 16 PAUL BELFORD
and Wednesbury Forge formed part of the group ‘MR WILLITS’ FOLLY’ (THE WINDMILL)
known as the ‘Ironworks in Partnership’. This Probably one of the most unlikely of the many
partnership was formed in 1692 and included four extraordinary metallurgical experiments seen in
furnaces, four slitting mills, thirteen forges and a
the West Midlands during the 18th century took
warehouse. These were mostly along the Stour,
place at Wednesbury. Already operating a finery
but also included sites in the Forest of Dean and
and chafery, a rolling mill and various grinding
others, such as Wednesbury, on the Tame.45 In the
wheels John Willetts, like his contemporaries,
first few years of the 18th century the ‘Ironworks in
Partnership’ had given up their Midlands works suffered periodic water shortages. He therefore
in order to concentrate more on Sussex and the chose to develop what was probably the only wind-
Forest of Dean. powered grinding and boring mill in the history of
ferrous metalworking. This structure was located
in the space between the sluices serving the two
PERIOD 3: WILLETTS’ FORGE (1704–1817) wheelpits — a logical place for a power source,
but an extremely difficult position from which to
The Foley partnership quickly divested themselves operate a windmill. As the Swedish industrialist
of Wednesbury Forge (and their other River Angerstein noted in the 1750s: ‘Willits [sic] was
Tame sites), and by January 1704 the site was in willing to spend a considerable amount of money
the occupation of John Willetts. Roger Colley and to learn of a way to take in the sails without stop-
Joseph Beeche were forgemen in 1707 and 1708 ping the mill, because the stopping is very danger-
respectively.46 At that time Richard Shelton still ous in view of the position of the mill just above
owned the site. An agreement between Shelton the water . . . [the mill was] . . . generally called
and Richard Parkes in 1708 shows John Willetts “Mr. Willits Folly” by people living in the
working the ‘iron mill commonly called Wednes- district’.49
bury Forge . . . now used for plating’.47 This was Only the northern half of the windmill build-
possibly a rolling mill. Saw-making was the main ing survived. The tapering walls [1578] stood to a
activity on site during the first half of the 18th height of around 2.15m, and were partly founded
century (Fig. 15).48 on redeposited natural clay. The extant structure
was sub-circular in plan, with an angled south-
PHASE V: EARLY 18TH CENTURY eastern corner (Fig. 16 and Fig. 29). The external
diameter was 7.98m. Although later strengthening
Willetts culverted more of the watercourse system, had added some thickness to the wall, the original
perhaps in order to accommodate the new plating structure appeared to be five bricks thick; these
facility. The main open watercourse was impres- were bonded in a mixture of conventional and
sively culverted in brick; the resulting structure hydraulic lime mortar, and the whole exterior of
[1343, 1364, 1760, 1798, 1850] was slightly curved the wall was rendered since it formed part of the
in plan, and ran across the entire site in a broadly
sluice system.
north-easterly orientation. It was constructed of
Two elliptical-arched doorways provided
new and reused bricks, some of the latter partly
access to the north and south sides of the building.
vitrified and presumably from earlier forge struc-
The southern door was accessed from the outside
tures. The culvert consisted of a single curved
by a set of steps [1693] which worked their way
arch, varying in width but generally between 2.5
and 3.4m wide. The maximum height of the culvert around the external wall; these were made of
was estimated around 2.8m. Until the abandon- brick and edged with hard grey sandstone blocks.
ment of water power in the mid-20th century The remains of a third doorway survived on the
(Phase XIII) this 67m-long tunnel continued to truncated north-eastern end of the building.
drain the water away from the wheelpits and back Perhaps inevitably in view of its location, the
to the River Tame. At the time of excavation it windmill’s construction prompted some rebuilding
contained standing water, although the outlet to of the sluices leading to the southern wheelpit.
the river had been diverted. Elements of this work survived in the form of the
The eastern end of Tailrace 2 was subsequent- south-western wall of the watercourse [1581] and
ly culverted [1429, 1430, 2117] so as to connect with a brick floor between it and the windmill wall.
the main outflow. Again, the new culverts were The watercourse wall [1581] was 0.35m thick and
built directly on the line of the timber tailrace. This survived to approximately 7m in length; it curved
1.2m-wide structure had vertical side-walls and a slightly from the north-west to the south-east in
shallow arched roof; it was consistently 1.2m wide order to direct the water into the southern wheelpit
until its junction with the main culvert. constructed in Phase IV (see above). This brick

44-1-1-PMA 01.indd 16 3/1/2010 7:35:19 PM

 WEDNESBURY FORGE 17

FIG. 15
Wednesbury Forge: Period 3, Phases V and VI. Overall site plan showing early 18th-century features.

wall was up to 1.95m in height, and stepped out in BUILDING A (WILLETTS’ HOUSE)
a series of five steps to form a curved weir. A floor In addition to his ‘folly’, John Willetts also built a
of bricks on edge [1692] was laid between this wall house (Building A), north-east of the main forging
and the windmill. complex (Fig. 17). At its core was a roughly square

44-1-1-PMA 01.indd 17 3/1/2010 7:35:19 PM

 18 PAUL BELFORD

FIG. 16
Wednesbury Forge: Windmill. Plan (left) and internal elevation from A–B.

brick building, 9.7x8.8m in plan. The surviving part of the cellar had a brick floor [1520] which had
foundations suggest that the house was two rooms been built on a series of levelling layers made up of
deep (east–west) and three bays wide (north–south). forging waste and other rubbish, one of which
The main building was subdivided by a brick wall [1738] contained 17th-century ceramics, including
[1548] separating the northern third from the a Wednesbury ware jug handle. A doorway in the
southern two-thirds. Fireplaces were located in the dividing wall led to the eastern room of the cellar,
north-eastern corner of the northern room [1524] the floor of which was compacted earth [1542].
and the north-western corner of the southern room This end of the cellar was also accessed externally
[1552]. Further subdivision of the main floors of from the east via a set of brick and wooden steps
the house above was indicated by a substantial [1533].
sandstone column base [1554], halfway between Domestic water was supplied by a 1m diame-
the north and south walls of the southern room. ter brick-lined well [1464], located approximately
The column base and buttresses to the central wall 4m west of the house (in the back yard). This
were slightly west of the centre line of the building; survived to a (truncated) depth of 3.48m. Child
assuming that the larger rooms would have been and adult leather shoes were recovered from the
the principal reception rooms at the front of the waterlogged primary fill [2110] at the base of the
building, the house appears to have faced east (i.e. well (Fig. 17).
looking away from the forge). There is some evidence that Willetts applied
A further range existed to the north, built his eccentric ingenuity to domestic sewage arrange-
at the same time as the main house but evidently ments as well as forging operations. On the eastern
a service range. Archaeologically this comprised side of the house was a feature possibly originally
a cellar [1432], the base of which was below the associated with the disposal of waste or foul water.
level of the main core of the building. The cellar This comprised a large brick-built circular tank
was 8.7m long (east–west) and 2.5m wide, and was [1820], 4.1m in diameter, with a neatly-laid floor
divided into two halves. The western half was of reused, handmade roof tiles [1849]. A narrow
approximately 5m long, and was accessed inter- brick-built drainage culvert [1836] sloped down
nally by a brick-built spiral staircase in the north- from the south-eastern side of this cistern into
eastern corner of the main building [1517]. This the nearby primary culvert [1364]. A brick floor

44-1-1-PMA 01.indd 18 3/1/2010 7:35:21 PM

 WEDNESBURY FORGE 19

FIG. 17
Wednesbury Forge: Building A (Willett’s House). Overall plan showing Phase V, VI and IX features (top), with finds
from early occupation phases. All finds drawings to same scale.

44-1-1-PMA 01.indd 19 3/1/2010 7:35:22 PM

 20 PAUL BELFORD
surface [1817] linked this to a rectangular brick in operation 70 years later in a ‘gun factory and
drain [1508] with an iron grate adjacent to the saw-blade factory’ at Wednesbury, in which ‘iron
house, and presumably collecting water from filings and small pieces of iron . . . were put into
it. The primary fill of this drain [1509] included crucibles that were placed in a reverberatory fur-
various tin-glazed and black-glazed earthenwares nace . . . the resulting lump of iron was removed
(Fig. 17). from the pot and forged’.53
Buildings C and D, at the far eastern end of
BUILDING B the site, may have been involved in these processes
(Fig. 18). Building C was a roughly square brick
A series of relatively short-lived buildings were structure [1741] 6.4x7.2m in plan and oriented
constructed at the far eastern end of the site, north–south, subdivided into four rooms. The
directly on top of the newly-built main culvert. western rooms each contained a small cellar [1757,
These structures were evident as a series of heavily 1758], with steps down from the former floor levels.
truncated quarry-tiled surfaces [1767], overlain in In the southern cellar there were two sets of steps
Phase VI by later industrial buildings. These are [1755, 1759] giving access from east and west; both
likely to have been domestic buildings, perhaps cellars were paved with bricks and quarry tiles.
workers’ or servants’ houses. The eastern part of the building contained ash pits
for two back-to-back forging hearths [1748]. Finds
PHASE VI: EARLY TO MID-18TH included black-glazed earthenware and Midlands
CENTURY purple ware, suggesting a date in the first half of
the 18th century. Building C could be interpreted
The enterprising John Willetts was succeeded in as a series of small reheating furnaces and adjacent
1722 by his namesake son, who continued at the forging hearths.
forge until his death in 1753. The younger John Immediately east of Building C were the foun-
continued principally in the manufacture of saws; dations of another building, of which the central
he was listed as a saw-maker in 1726,50 for example, north–south wall [1795] survived to a length of
although he was also involved in gun-making. 4.3m. This structure (Building D) contained a pair
Thomas Kester and Jonah Davis are recorded as of rectangular forging hearth bases which mirrored
working at the forge in the 1740s.51 those in Building C. Both of the hearth bases [1792]
No significant alterations were made to the and [1793] were filled with a compact ashy-silt
main complex during this period, most new build- with charcoal inclusions [1796] and [1797], which
ing being on the fringes of the site. This included also contained clay pipe stems. To the west was a
the replacement of the short-lived domestic dwell- mortar floor surface [1794].
ings (Building B) with a series of hand forges.
At the same time Willetts also constructed a series
of boundary walls enclosing the eastern end of DOMESTIC IMPROVEMENTS
the site. The northern wall [1740, 1777], of which a Willetts also extended the house built by his
15.2m length survived, was of conventional brick father (Fig. 17) by adding a new southern wing.
construction. In contrast, the southern wall [1379, This survived as a substantial brick-built cellar
1384, 1754], also of brick, was founded on a series measuring 7.7x2.7m [1424, 1427, 1432] and ori-
of arches. These arches rested on brick piers; this ented north–south, partly overlying the culverted
system was devised to overcome boggy ground in northern tailrace. The eastern elevation incorpo-
this area, presumably a relic of the formerly open rated an opening at below-ground level to take
watercourse. At the northern-eastern end of this it over the culvert, and the area beneath the wall
21.5m stretch of wall was a gatehouse: a small and around culvert was packed with material which
square structure 1.64x1.71m in plan with a brick included 17th- and early 18th-century pottery. This
floor [1753]. end of the building subsequently suffered serious
subsidence, particularly in the cracked and bowing
south-west corner. The cellar had a brick floor
HAND-FORGE BUILDINGS (BUILDINGS C
[1422], and was later subdivided into two rooms by
AND D)
a thin brick partition [1428].
The production of small batches of high-grade iron Domestic rubbish from this period was
for special purposes seems to have been widespread dumped in a large pit to the east of the house,
in the Black Country. Robert Plot described in and to the south of Building C. This pit [1810]
1686 how ‘the best iron of all’ was made from the was up to 3m wide and contained a quantity of
‘filings and pareings of the locksmiths’.52 Reinhold broken glass, animal bone, clay pipes, shells, brick
Rücker Angerstein described a similar process rubble and a ceramic assemblage consisting of

44-1-1-PMA 01.indd 20 3/1/2010 7:35:23 PM

 WEDNESBURY FORGE 21

FIG. 18
Wednesbury Forge: Buildings C, D and E, overhead view looking north-west. Building C is bottom right (light brick
square structure and floor surface just to the right of the shadow). Building D is represented by the two square hearths
and associated walls and floors to the west of Building C and crossed by the shadow. The remaining walls are Building
E (Phase IX) (scale 2m).

black-glazed earthenware storage jars and salt- thus continued his father’s excursions into the
glazed wares all datable to the 18th century. It was increasingly lucrative gun-making trade. Wednes-
also during this period that the possible former bury Forge was one of three gun-making sites
sewage system was abandoned, although the cis- being used in the 1750s by Willetts, who was sup-
tern continued in use for industrial water storage. plying guns to the Board of Ordnance.57 Iron was
Local physician Dr Richard Wilkes visited the purchased from various sources throughout the
Willett family in November 1736 for the treatment period of the Willetts’ occupation, including local
of colic;54 it can be speculated that the presence furnaces such as Hales and Aston as well as more
of a poorly-drained tank full of sewage may have remote suppliers based in Bristol and Hull. The site
had adverse effects on the health of the family. included a plating forge making skelps,58 a boring
mill and a mill to grind off the outside of the gun
PHASE VII: MID- TO LATE 18TH CENTURY barrels.59 Gun-making required substantial boring,
grinding and forging facilities, as well as hearths
John Willetts’ son Benjamin survived the various and furnaces (Fig. 19).
health hazards encountered on the forge site, and
took over the works on his father’s death in 1753.55
The saw-makers of Sheffield had developed a REBUILDING THE NORTHERN WHEELPIT
strong market position since the development of The windmill experiment having proved unsuc-
crucible steel,56 and their Midlands counterparts cessful, attention was paid to improving the effi-
began to turn to other sources of revenue. Willetts ciency of the water-power installation. This work

44-1-1-PMA 01.indd 21 3/1/2010 7:35:23 PM

 22 PAUL BELFORD

FIG. 19
Wednesbury Forge: Period 3, Phases VII and VIII. Overall site plan showing mid- to late 18th-century features.

evidently resulted in some disruption to the neigh- The rebuilding of the wheelpit involved the
bourhood; the owner of a nearby coal mine brought construction of new brick walls to the northern
a case in 1756 claiming that his workings had [2031], eastern [1732] and western elevations
been flooded by the temporary ‘diversion of a (Fig. 20). The western elevation included a semi-
watercourse at the forge pool’.60 circular 5m length of brick wall in English Garden

44-1-1-PMA 01.indd 22 3/1/2010 7:35:25 PM

 WEDNESBURY FORGE 23

FIG. 20
Wednesbury Forge: northern wheelpit, eastern elevation showing Phase VII brickwork and Phase VIII iron framing
and bearing box (scale rod in 0.5m divisions).

Wall bond, which also acted as part of the forge the now long-abandoned Tailrace 1 and passing
pool retaining wall. Associated with this were two to the north of Willetts’ house. It fed back into
north–south brick walls [2026] and [2034], which, the main culverted tailrace outside the site bound-
together with the second part of the western eleva- ary. This new culvert rendered the old east–west
tion (another north–south wall [1591] to the east) tailrace (Tailrace 2) redundant.
formed part of the water regulation mechanism.
These two western walls were joined by a brick
NEW GRINDING AND BORING MILLS
floor surface [1729, 2064], which extended to the
northern side of the windmill building. The eastern The reconstruction of the northern wheelpit freed
elevation [1732] was built of the same materials; up a large area of the forge, and this was rebuilt
neither of these side walls directly abutted the as a grinding and boring mill, incorporating the
existing southern wall [1730], which had been earlier structures that had formed part of the ‘folly’
built during Phase IV and was retained), but were (Fig. 21). This new mill measured 20m north–south
separated from it by a thin lens of compacted and 10m wide, and was divided into three sections:
demolition material and rubble. grinding mills to the north and south and a boring
The new wheelpit was served by an entirely mill in the central section. Power was transmitted
new culverted tailrace, which exited from the to the grinding mills along a narrow brick channel
north-eastern corner of the wheelpit [2031]. This running north–south immediately to the east of
brick culvert [2012, 2038] comprised an arched the eastern elevation of the wheelpit.
roof on vertical side-walls and was 1.30m in The southern grinding complex reused the
section; it survived to a length of approximately former windmill [1578]. The original doors to the
32m. It ran in a north-easterly direction, truncating mill were blocked and the interior was squared off

44-1-1-PMA 01.indd 23 3/1/2010 7:35:27 PM

 24 PAUL BELFORD

FIG. 21
Wednesbury Forge: grinding and boring mill, showing Phase VII and VIII features and Wilkinson token found under
Phase VIII boring mill timber [1476].

by the insertion of a wall along the inside of the with oxidized wheelswarf [2027]. Again the bases
western elevation. Three new brick-built grinding of the pits were timber, and the northern end of this
wheel pits were located inside the building [1696, structure overlay the newly-built northern culvert.
1704, 1705], cooled and lubricated by water, Further extensions were made to the grinding and
which was led away from the grinding mill down boring mill in subsequent phases. No grindstones
a sloping central channel [1695] to a small iron were found in situ, but a considerable number were
chamber [1629] which fed into a small drain. There encountered in late 18th- and 19th-century con-
were slight variations in the size of the wheelpits, texts all over the site, where they had been reused
with internal measurements between 1.8m and in a variety of structures including buildings,
2.1m long and 0.35m to 0.55m wide. They extended hammer bases and a railway track. Grinding was
up to 1m below the ground surface; the base of an extremely dangerous occupation, due largely to
each pit was formed from timber seated on the the inhalation of dust, but also the explosion of
natural clay. These grinding pits were partly filled grindstones. So in 1767 Joseph Stevens was killed
with an accumulation of compacted wheelswarf at Wednesbury ‘by the breaking of a Stone on
[1716],61 thicker at each end of the pit due to the which he was grinding tools’.63
action of the wheel; indeed the whole interior of
this building was covered with a thick lamination
of oxidized wheelswarf adhering to the floor and HEARTHS AND FLUES
walls. This archaeological evidence supports an To the east of the new northern grinding mill
1831 document referring to a ‘grinding mill which were a pair of forging hearths and associated
had formerly been a windmill’.62 underground flues (shown on Fig. 19). The eastern
The boring mill and the northern grinding forging hearth [1505, 2010] was rectangular in
mill complex was a new build [2020], which origi- plan and oriented north–south; it measured 3.5x
nally included three grinding wheel pits. These pits 1.64m and was built of firebricks. Vitrification
[1493], [1496] and [1498] were built in the same and heating were evident throughout, the thermal
manner as those in the former windmill, albeit to stresses requiring the later addition of buttresses
a more consistent size, and were all partly filled [1502] to the southern corners and some internal

44-1-1-PMA 01.indd 24 3/1/2010 7:35:29 PM

 WEDNESBURY FORGE 25
reconstruction. The western hearth was more road to Wednesbury. There was a drainage chan-
severely truncated, and only the ashpit [2014] nel running along its eastern elevation. This was
survived. This was constructed in a similar fashion, built of a single course of bricks and was roofed
and had been modified by the insertion of a divid- using alternate chamfered bricks to create a water-
ing wall [2018]. Both ashpits contained primary tight construction. Part of the drainage channel
deposits of charcoal and ash [2016] and [2017]. had collapsed and been allowed to silt up naturally.
The hearths were connected to a further small- Within this fill [2005] were ceramics and gun flints
er hearth to the north-east [2010] by a system of of mid- to late 18th-century date.
underground brick flues [1073], [2009], [2011] and Further flints were also found in the fill of
[2012], which extended for approximately 25m Willetts’ domestic cistern [1825], and in some of the
north-west to south-east. This early flue system layers underlying the short-lived western extension
had been truncated by later developments and to the south cellars. Taken together, this assem-
only survived in fragments, including a substantial blage of gun flints (Fig. 22) shows how sophisti-
section only evident as a backfilled robbing trench. cated the Willetts’ gun-making enterprise had
Surviving sections indicated a single arch 1m in become. Although the historical record suggests
section, filled with successive layers of burnt ash, that they were only making barrels, the archaeol-
charcoal and silt deposits [2123]. No trace survived
ogy shows that the firm was involved in the whole
of the buildings which had housed these features.
process during the second half of the 18th century.
By this time such flints were partly machine-made
GUN-FINISHING WORKSHOP (BUILDING (Rees’ Cyclopaedia shows ‘tools for cutting gun
H) flints’),64 and, although they may have been made
Building H was a substantial brick building on site, the absence of unfinished flints, cores or
situated to the north of the northern grinding mill flakes suggests that they were imported as finished
and its adjacent hearths and flues. This fragmen- items. Four types were identified, from which it
tary rectangular structure [2001, 2008], oriented can be inferred that different weapons such as mus-
north–south, was the most northerly set of build- kets, pistols and perhaps latterly rifles, were being
ings during Phase VII, and was nearest to the made.

FIG. 22
Wednesbury Forge: gun flints. A representative selection of the four types, featuring flints from contexts
[1755] and [1825].

44-1-1-PMA 01.indd 25 3/1/2010 7:35:30 PM

 26 PAUL BELFORD
FURTHER CHANGES TO THE CULVERTS columns. Integral to the casting of the horizontal
It was during this later 18th-century phase that beams was a pair of lugs to locate the bearing
further brick culverts were constructed in the assembly. The water-wheel was also of cast iron,
southern part of the site. The precise function of and a section of the sole board was recovered
these is not clear, since their origin lay outside the from the base of the wheelpit during excavation.
excavation area. However, they appear to have This fragment was 30mm thick and 1.44m long; it
supplied water for grinding, boring and rolling contained ribs to house the wooden boards and
processes in the southern part of the site rather represented a one-twenty-fourth section of the full
than acting as tailraces for water-power installa- wheel, which would have been cast in sections to
tions. These culverts [1342, 1353, 1567, 1568, 1569] enable easy transportation to site and assembly.
were brick-arched and varied from 1m to 2.5m in The wheel had gouged an impressive scrape mark
width. They all ran from south-west to north-east in the northern elevation, wearing away at the
and linked with the primary culvert built in Phase iron frame as well as the brickwork. The diameter
V. Meanwhile, the now-abandoned 17th-century of this breast-shot wheel was 5.5m.
culverts were all filled during this period. These The remainder of the wheelpit was of relatively
deposits yielded mid- to late 18th-century ceramics, conventional construction, the eastern elevation
crucible fragments and other artefacts, including comprising a brick arch [1660] leading to the earlier
leather. culvert [1182, 1566]. The western elevation had
been almost entirely removed during the later
insertion of a turbine.
PHASE VIII: LATE 18TH AND EARLY 19TH A similar iron-framed structure was inserted
CENTURIES into the northern wheelpit at around the same time
The presence of these new southern culverts reflects (shown in Fig. 20). Here the western elevation was
investment in the southern pool. The pool was partly rebuilt to allow the insertion of a cast-iron
probably enlarged after the death of Benjamin horizontal beam [1718] into the brickwork. In
Willetts in 1786, when the forge passed to his the eastern elevation the equivalent horizontal
son, another Benjamin. He operated the forge in a member was supported on a freestanding iron
partnership including his mother, the concern frame within the wheelpit itself [1731]. This
being known as Short, Willetts & Co.65 Unfortu- arrangement meant that it was not necessary to
nately, the younger Benjamin’s tenure was short- rebuild the western wall of the grinding and
lived; he died in 1794 and the forge was operated boring mill. Here again both horizontal beams
by his widow in another partnership with a Mr contained integrally-cast lugs to locate the bearing
Holden. This sequence of events gave rise to the mechanism.
name ‘Mrs Willetts’ Pool’ which was used for
the southern pool from the late 18th century TIMBER BORING MILL BASE
onwards.66
The improvements to the water-power system
prompted further alterations to the boring mill
IRON-FRAMED WHEELPITS complex. The demolition of the substantial Phase
The improvements to the southern pool increased IV timber building provided the raw materials
the supply of water to the forge, enabling modern- for the base for a new boring mill. The installation
ization of the water-power system. The first of this facility necessitated the extension of the
development was the rebuilding of the southern boring mill to the south and east, and the timbers
wheelpit, replacing the century-old brick structure were in fact incorporated into the foundations of
constructed in Phase IV. Taking advantage of this building.
the latest developments in ferrous technology,67 The new structure contained a series of
one of the Mrs Willetts then in control of the forge substantial parallel timbers which overlay the cul-
decided to install an iron-framed wheelpit. verted (and now redundant) Tailrace 2 (Fig. 24).
The iron framing consisted of vertical cast- These timbers [1473, 1476, 1488] were over 5m long
iron columns, bedded into a brick base [1641] and 0.5 × 0.3m in section; they were laid horizon-
which incorporated elements of the earlier struc- tally in three parallel rows oriented east–west and
ture (Fig. 23). These columns [1652] were 0.22m joined by a series of north–south timbers [1472]
square in section and stood 2.5m high; they were and [1474]. A cast-iron beam [2144] was bolted
situated 1.35m apart and the space between to the eastern end of the structure; vertical iron
was infilled in brick. The top of each side of the bars were bolted to this beam, corresponding
wheelpit was formed by a cast-iron horizontal with similar iron bars bolted to the north–south
beam [1638, 1639], socketed into the upright timbers. A dark oily silt [1507] filled the area

44-1-1-PMA 01.indd 26 3/1/2010 7:35:31 PM

 WEDNESBURY FORGE 27

FIG. 23
Wednesbury Forge: the iron-framed southern wheelpit constructed in Phase VIII. Top: photograph showing the
north-eastern end of the wheelpit and the water-wheel scar (scale 2m). Bottom: south-facing elevation.

between the timbers. Taken together, the form, ori- token of that date was found in the upper horizon
entation and nature of this substructure suggests of the levelling layer [2082], beneath timber [1476].
that it supported a boring mill powered by the The token had been modified by its owner, who
northern water-wheel. The installation of this had punched a hole in it, presumably to create a
machinery took place after 1788, since a Wilkinson pendant or necklace. Given John Wilkinson’s

44-1-1-PMA 01.indd 27 3/1/2010 7:35:31 PM

 28 PAUL BELFORD

FIG. 24
Wednesbury Forge: timber-framed boring mill base during excavation (scale 2m).

association with boring mill technology, and the as the water-power arrangements for the forge,
personalization of the token itself, it is tempting and various improvements were made during this
to suggest that this was a deliberate deposit by a period. Perhaps the most significant was the aban-
former Wilkinson employee bringing knowledge donment of the old well, the area around which
gained under the ‘iron mad’ entrepreneur to was being encroached upon by the northern forg-
Wednesbury Forge. ing hearths and the new boring mill installation.
The well was deliberately filled by a mixed sandy
OTHER INDUSTRIAL ADDITIONS silt containing rubble fragments and ceramics
suggesting a date post-1770. Water supply by this
Other additions included an extension of the earlier time appears to have been piped directly from the
flue network to the south, presumably serving northern forge pool through a series of elm pipes
furnaces located over and beyond the southern
(Fig. 25). These ran under the floor of the gun-
culvert. This extension [1074] abutted the earlier
finishing building (and may have been installed
flue [1073] and was straight-sided with a shallow
arched roof with spread brick footings. At its when that building was constructed). A length of
northern end was an irregular series of roughly- 9m survived [2043], consisting of five sections of
built small brick flues or drains [1294], [1297] and hollowed out elm trunks with tapered ends. Each
[1298]. Meanwhile, the area around the Phase VI tapered end was wrapped in fabric and covered
hand-forge buildings (C and D) was improved by with pitch, before being slotted into the next
the laying of a blue-brick floor surface [1750]. section; the join was then bound with a wrought-
iron strap to create a watertight junction.
The cellar walls were buttressed and thickened
FURTHER DOMESTIC IMPROVEMENTS in the areas of subsidence [1523] and [1541]. Other
The redoubtable Mrs Willetts no doubt had some modifications included blocking the western
input into the management of the house as well entrances and their associated stairwells [1562]

44-1-1-PMA 01.indd 28 3/1/2010 7:35:33 PM

 WEDNESBURY FORGE 29
and [1437]. Rubbish from this period was also
being dumped to the east of the house, in a large pit
[1321], dug in what had originally been the front
garden. Finds from both the garden soil [1312] and
the pit fill [1322] included ceramics, glass bottles
and clay pipes (Fig. 26); it is tempting to associate

FIG. 25
Wednesbury Forge: elm water pipes supplying Willetts’
house from the New Pool, seen here as they passed
beneath Building H (scale 2m). FIG. 26
Wednesbury Forge: finds from Willetts’ House.

44-1-1-PMA 01.indd 29 3/1/2010 7:35:35 PM

 30 PAUL BELFORD
the mark ‘BW’ on some of the bottles with the the existing watercourse wall [1581] and brick
late Benjamin Willetts. This evidence suggests that flooring [1692], which were incorporated into
the former back yard on the western side of the a larger set of gates. Three openings were con-
building was no longer in domestic use, presum- structed, consisting of a set of two brick piers
ably due to the continuing enlargement of the on either side of an open space containing two
forge. freestanding cast-iron uprights. The brick piers
had cast-iron beams set into them to hold planks or
sluice gates to regulate the height and flow of the
PERIOD 4: ELWELL’S FORGE (1817–1929) water (Fig. 29). The northern set of sluices [1583]
were constructed in a slightly different fashion;
The name Elwell is still synonymous with Wednes- comprising four openings created using five brick
bury Forge, despite Elwell-branded garden and columns with inset cast-iron beams (Fig. 30). The
agricultural tools having been out of production reconstruction of both sets of sluices involved
since the 1970s. Various branches of the Elwell the insertion of a rough stone floor [1690] between
family were involved in the West Midlands metal the windmill building and the sluices themselves.
trades during the 18th century. Edward Elwell was This seems to have been created in order to make
the son of West Bromwich ironfounder William the two sluice floors level with one another. A
Elwell.68 After a spell as an army surgeon with 1.1m-high cast-iron filter grille was set on the pond
the Royal Artillery during the Napoleonic wars, (western) side of the north and south sluice gates to
Edward set up as an edge tool manufacturer. His prevent debris from passing into the water-power
business prospered, and having already moved system.
The sluice arrangement reflected the historic
from small premises in Walsall to a forge in
division between north and south ponds, and
Wednesbury, he jumped at the chance to take
north and south wheelpits. Thus the northern and
over the Wednesbury Forge lease when it became
southern sluices were divided by a low wall [1584]
available in 1817.69 attached to the western elevation of the windmill
building. This east–west oriented wall was 2.9m
PHASE IX: EARLY 19TH CENTURY long and 0.5m wide, and the 19th-century rebuild-
Post-Waterloo peace reduced demand for Wednes- ing incorporated a cast-iron capping beam con-
taining sockets for uprights. Part of the groove
bury flintlocks, guns having been the forge’s prin-
used as the runners for the sluice gates was made
cipal product during the later 18th century, since
up of reused grindstones. Thus the division between
the decline of the Midlands saw trade.70 Elwell
the ponds was controllable, and presumably this
had begun to take advantage of the growing arrangement meant that the water supply could
market at home and abroad for good quality be directed from either pond to either wheelpit
agricultural tools, in particular edge tools such as through the manipulation of the various sets of
axes, billhooks and pruning shears, and such items sluices.
required different production methods. Thus Some additional culverting also took place in
Elwell embarked upon a substantial programme association with the construction of buildings in
of investment, most of which probably took place the southern area (see below). However, the main
after he bought the property outright in 183171 emphasis was in expanding the site to the east,
(Fig. 27). which resulted in the construction of two further
large brick culverts [1357] and [1359]. The principal
FURTHER CHANGES TO THE culvert ran roughly parallel to the existing main
WATER-POWER SYSTEM culvert. This originated in the south-west corner
of the site [1355] and ran north-east, reducing in
Significant changes were made to the sluices during diameter from 2.35m to 1.15m after approximately
the early 19th century, resulting in the arrangement 40m [1359]; it then continued out of the site [1752],
shown on the 1847 Tithe Map, and which survived presumably feeding back to the river Tame,
more-or-less unaltered until its early 20th-century although only c. 62m of the length of this culvert
abandonment. Their absence on the 1860s litho- was extant. This was linked to the main culvert by
graph (see Fig. 39 below) is probably the result a short length of culvert oriented north–south
of artistic licence, suggested by other inaccuracies [1357], which was 5.7m long and 1.6m wide.
in the illustration.
The 19th-century reconstruction involved
the creation of a more extensive system of sluice MODIFICATIONS TO THE BORING AND
gates which enabled very sensitive adjustments GRINDING MILL
to be made to the water flowing through the site The northern end of the grinding mill was extended
(Fig. 28). The southern set of sluices [1582] reused to the east during this period, truncating and

44-1-1-PMA 01.indd 30 3/1/2010 7:35:38 PM

 WEDNESBURY FORGE 31

FIG. 27
Wednesbury Forge: Period 4, Phases IX and X. Overall site plan showing 19th-century features.

partly removing the Phase VII hearths and flues construction of this north-eastern set of grinding
[2010]. This extension [1296, 1448, 1468, 1486, pits meant the demolition of the forging hearth
1682] included a further set of three grinding structure [2010] associated with the 18th-century
pits [1500, 1501, 1502], constructed in a similar flue system (see Phase VII). The southern end
manner to their earlier counterparts (Fig. 21). The of the grinding mill [1580] was converted to a

44-1-1-PMA 01.indd 31 3/1/2010 7:35:38 PM

 32 PAUL BELFORD

FIG. 28
Wednesbury Forge: Phase IX sluices and associated features. Overhead view facing south-west and looking at the
sluices from the pond side with the metal grilles in the foreground. The modified Phase V windmill building is clearly
visible; beyond it the Phase VII grinding pits and the chimney for the Phase X steam engine. The large square
structure at left is the modified inlet for the Phase XII turbines (scale 2m).

chimney. This ‘grinding mill . . . had formerly constructed at the same time, since elements of
been a windmill’ according to a description in some of the flue superstructures also formed parts
1831, and the windmill served as the base of the of the wall foundations. However, it was not
chimney. This is clearly evident on a drawing used possible to determine precisely original above-
as the company letterhead (not illustrated here).72 ground functions and layout.
In 1858 a visitor noted that ‘the tall and finely The pool edge wall [1140, 1165, 1215] was up
tapered chimney of the extensive Wednesbury to 0.9m thick and built of dense semi-engineering
Forge . . . and the adjoining capacious sheet of bricks bonded in a cement-rich mortar. The west-
water (called Elwell’s Pool) form attractive objects ern (i.e. pool-side) edge of the wall was packed with
of observation’.73 redeposited clay. The southern wall [1215] was the
western wall of a substantial square building which
extended outside the excavation area. Both this
SOUTHERN AREA
and the curved wall to the west are clearly shown
The most significant addition during this period on the 1847 Tithe Map. The remaining walls varied
was the southward expansion of the site (Fig. 31). between one brick (230mm) and three bricks
Several buildings were constructed, partly overly- (690mm) thick, bonded using cement mortar. In
ing the Phase VII culverts, and creating a new places original brick and tile floor surfaces survived
eastern edge to the southern pool. These buildings [1034, 1105].
occupied a total area of c. 30x30m. It was clear Within these buildings there was a complex
that above- and below-ground features had been network of flues. This appeared to be divided

44-1-1-PMA 01.indd 32 3/1/2010 7:35:39 PM

 WEDNESBURY FORGE 33

FIG. 29
Wednesbury Forge: southern sluices (scale 2m).

into northern and southern elements, the southern walls [1136] survived together with its connection
part being served by a small chimney inside the to the flue [1132]. The furnace room [1133] was
building. The chimney [1217] was circular in plan, a sub-basement reached via four brick steps; it
2.5m in diameter and 3 bricks thick; it survived was oriented east–west and survived to a depth
to a height of at least 1.06m and was filled with of 0.92m, retaining its original floor surface (com-
cinders and ash. The northern part of the flue prised of a mixture of reused bricks and a broken
system may have been connected to a second chim- grindstone). An iron door regulated access to
ney (since destroyed), which also served Building the flue system (Fig. 32). This area also contained
F; this area, and the area where the two flue sys- bases for heavy equipment, constructed with
tems met, was reconfigured during Phase X and reused grindstones. A cemented platform of small
is dealt with below. The flues were straight-sided grindstones in association with brick wall [1156]
with a shallow arched roof; in some places the and [1158] was interpreted as a machine base. A
brickwork was sealed with a skim of reddish more complex structure involved a setting of two
mortar to provide an airtight construction. The large grindstones [2153, 2154], supporting a pad
flues were all between 1.1 and 1.4m in diameter, of reused timbers; this was the base for a hammer
and the foundations consisted of double brick or press. Two large grinding pits were located
footings sat upon a thin layer of crushed brick and at the western end of this part of the site [1145,
mortar. 1879]; these were brick structures over 3.5m long
The nexus of the southern flue system was a and up to 2m wide, which survived to a depth
large rectangular structure [1133], interpreted as of 0.8m, and were strengthened in the corners
a stoking/ashpit area serving a possible puddling with wrought-iron straps.
furnace to the east. The furnace (being above The southern area was a purpose-built forging
ground) had been destroyed, but its foundation and grinding facility, specifically created for the

44-1-1-PMA 01.indd 33 3/1/2010 7:35:41 PM

 34 PAUL BELFORD

FIG. 30
Wednesbury Forge: northern sluices (scale 2m).

manufacture of edge tools and thus largely replac- of rooms arranged around a central courtyard,
ing the earlier grinding and boring mills to the and was built from a mixture of handmade and
north which had been designed for saw and gun machine-pressed bricks. The original floor surfaces
manufacture. Although the grinding and boring inside the building appear to have been tiled (a
mills were subsequently refitted and adapted (see fragment of this, [1779], survived in the southern-
Phase X below), the associated gun-finishing work- most room), and the courtyard was originally
shop (Building H) was abandoned and demolished paved in blue brick [1769, 1788]. Both internal and
during this period. external floors were later resurfaced with concrete.
Building E served as the main office range until the
BUILDING E late 1960s (see Phase XIV), whereupon most of it
was demolished, although the western part (latterly
The new southern area rendered the now century- in use as a toilet block) outlived the forge itself
old hand-forge buildings C and D (erected in Phase and was not demolished until July 2006. A detailed
VI) redundant. Consequently Elwell, determined plan of the building74 made in the 1960s closely
to improve the efficiency of administration as well matched the layout revealed by excavation.
as technology and process, used the area to create
a central administration building. Buildings C and
D were demolished and the ground levelled with a HOUSING
combination of demolition rubble and sandy silts In 1831 the forge included thirteen cottages ‘which
and gravels. had been workshops but which Edward Elwell had
As with the other activity ascribed to Phase reconverted into dwellings’.75 These appear to have
IX, Building E was depicted on the 1847 Tithe been the group of small terraced houses excavated
Map. Building E (Fig. 18) consisted of a series during the evaluation stage of the project in 2005

44-1-1-PMA 01.indd 34 3/1/2010 7:35:43 PM

 WEDNESBURY FORGE 35

FIG. 31
Wednesbury Forge: the southern area. Top: overhead photograph looking north (scale 2m). Bottom: plan of features
as excavated.

44-1-1-PMA 01.indd 35 3/1/2010 7:35:45 PM

 36 PAUL BELFORD

FIG. 32
Wednesbury Forge: the furnace room [1133]. View from the west showing entry to flue [1132] on the left and the iron
door on the right (scale 2m).

(Trench 7). Lightly constructed of handmade brick to its predecessor; here the west-facing elevation
bonded with lime mortar, they had brick floors and was the front, with bay windows on the ground
small sub-basements. The houses are shown on the floor looking out over a lawn.
1860s lithograph (see Fig. 38 below), and remained Most of the evidence for the superstructure
in use until the early 20th century, when part of the of the house comes from photographic and docu-
row was demolished and the remainder turned into mentary sources, since it was comprehensively
the ‘Forge Arms’ public house. The census records removed during Phase XIII. However, parts of the
John Marlow (grinder) and his young family, Isaac quarry-tiled brick-built cellars were incorporated
Ashton (engineer), Henry Russell, Benjamin and into the basement of the 20th-century gatehouse
Richard Challoner (all blacksmiths) living in these building, and were latterly used to store ‘seconds’
houses in 1841.76 (Fig. 33). The floor of the cellar incorporated a
The Elwell family and their servants also lived series of brick drains [2040] which fed into the
at the Forge. At first they occupied the Willetts’ old northern wheelpit culvert [2012].
house (Building A), but between 1831 and 1847
Elwell constructed a new house slightly to the
PHASE X: MID-19TH CENTURY
north. The new house was built in two stages,
partly reusing earlier foundations; the evidence of Edward Elwell handed over the business to his
the surviving cellar walls suggested that the new namesake son c. 1850, but the younger Edward
house originally connected with the northern end died in 1857 and so the father resumed his role
of Building A. The southern end of the house was as director. The new house was extended quite
built first, and was a five-bay three-storey structure soon after it was built, with a two-storey three-bay
of brick. The new house was in reverse orientation addition to the north. The northern wall of this

44-1-1-PMA 01.indd 36 3/1/2010 7:35:48 PM

 WEDNESBURY FORGE 37
were also bone china transfer-printed teabowls,
saucers and cups, and large black-glazed platters
and storage jars (Fig. 34). The building was then
extended to the east [1335, 1815], partly over the

FIG. 33
Wednesbury Forge: cellars of Elwell’s house.

was encountered during the evaluation (Trench 9),

cutting through 18th-century deposits of forge
waste and general rubbish. The extension was cer-
tainly completed by the 1860s, when it is depicted
in a photograph showing a game of croquet in
progress between some of the younger Elwells and
their friends.77 The 1861 census records Edward
Elwell (then aged 77), his wife Carew, and four of
their grandchildren, together with three domestic
servants and a governess.78

BUILDING A
The construction of the new house meant that
Willett’s former house (Building A) could be con-
verted to industrial use. This process had probably
already begun in Phase IX. Some remodelling
of the building took place, beginning with the
deliberate infilling of the southern cellar and the
former cistern. This seems to have been an oppor-
tunity to get rid of various domestic items. The
fill of the cellar [1421] included early 19th-century
creamwares and a Caughley porcelain teabowl
from the 1780s. The fills of the cistern [1825] FIG. 34
included a large quantity of creamware plates, Wednesbury Forge: finds from Phase IX clearance of
dishes, sauceboats, a teapot and chamber-pot; there Willetts’ former house, dumped in the cistern.

44-1-1-PMA 01.indd 37 3/1/2010 7:35:50 PM

 38 PAUL BELFORD
former cistern, and also to the west, sealing earlier ‘was killed about 2.30 p.m. by a grinding stone’.
rubbish deposits including gun flints. The eastern Georgiana Elwell, Edward’s granddaughter, took
extension included a brick platform [1337, 1819], time off from her croquet-playing to break the
oriented north–south, which may have been the news to his wife and three young children.79 The
base of a hearth. The western extension [1099] 49-year-old William was the son of the grinder
walls were heavily truncated, but the floor surface John Marlow noted above; the accident did
[1093] survived. not deter his son from working at the forge later
The north-eastern end of Building A, con- in life, nor indeed subsequent generations: his
structed over the infilled former cistern, began to great-grandson was still working there in 1972.
subside. In response, the northern and eastern
walls were strengthened with new brickwork [1338,
EASTERN AREA
1816], and the original southern wall was demol-
ished and rebuilt slightly to the north [1338]. At The southern forging facility built in Phase IX was
the same time, a new floor was inserted using a mirrored in Phase X with the construction of a
combination of old grindstones and new machine- similar and linked set of features in the eastern
made Staffordshire blue engineering bricks [1004, part of the site. The eastern area included at least
1005] (Fig. 35). The grindstones varied between two buildings, the eastern end of which was posi-
0.65m and 1.72m in diameter and were made tioned over the culverted tailraces. Its foundations
from local sandstone; this type of stone was infe- incorporated a series of arches to carry the weight
rior to ‘millstone grit’ and was prone to problems. over the culverts. The north wall [1375] was 15.8m
Thus on 14 June 1869 the grinder William Marlow long and incorporated a single shallow relieving

FIG. 35
Wednesbury Forge: grindstone floor inserted into Building A during Phase X. Overhead view facing west. Note to the
left the remains of the southern cellar (filled in during Phase IX), and also the Phase XI grindstone trackbed to the
south (scales 2m).

44-1-1-PMA 01.indd 38 3/1/2010 7:35:51 PM

 WEDNESBURY FORGE 39
arch [1387] running its full length. The western north–south flue [1132] was extended northwards,
wall [1392] had originally been similar, although and the two systems met at a complex junction
the superstructure had been rebuilt more than once just to the south of the culverted southern tailrace
and later insertions made. (Fig. 36). Here a third flue [1047, 1177] entered
The primary flue in the eastern area [1367] was from the west, and there was also an above-ground
oriented east–west and was 1.05m wide and 0.64m network connecting on the eastern side and
deep; it was built with vertical side-walls and a regulated by moveable iron plates [1077, 1345].
shallow arched roof bonded in lime mortar. The Above-ground arrangements in the eastern
flue served a series of smaller flues [1368, 1371, area were not easily determined. Only a small frag-
1377, 1405] inside the building. These flues were ment of brick floor survived [1378], together with
constructed in a similar fashion but were only 0.4m the truncated remains of below-ground features.
wide and were not so well built. The primary flue These included brick pillars for machine bases
ran west into another forging area — presumably [1372] and [1373], the ashpit of a small forging
also within a building — but only the foundations hearth [1069], as well as the base for a lightweight
of the northern and eastern walls [1311] and [1366] steam hammer constructed of four grindstones
survived. To the north, and apparently outside [1086, 1088, 1089, 1090] with a brick platform
the building, was a substantial rectangular brick [1092] surrounding the anvil, which had been later
structure [1340], which may have been the base of removed.
a chimney serving hearths in this area. This seems
to have been relatively short-lived, since the flue
had been blocked and the main east–west flue was STEAM POWER
extended west and south [1072] to connect with Many of the extensions to the forge seem to have
the flue system from the southern area. There the involved hand operations. However, the massive

FIG. 36
Wednesbury Forge: junction of flues [1072] and [1132] with iron grilles [1077] (scale 2m).

44-1-1-PMA 01.indd 39 3/1/2010 7:35:53 PM

 40 PAUL BELFORD
extension of grinding operations required more been severely truncated during Phase XI. A small
power, and during the mid-19th century a steam circular chimney [1593] was situated to the west
engine was installed. As with the earlier attempt of the engine room. This was 2.43m in diameter
to supplement water power using the windmill, and originally fed into the large chimney built on
the new power-source was located centrally. This the base of the windmill. Wheelpits for the engine
meant that the works could be run ‘with steam and flywheel and power drive wheel were situated
water jointly, or . . . when there is a good supply north-east of the chimney. These were housed in a
of water, independently of each other’.80 The new single brick structure 3.58x3.42m in plan and
steam engine was built partly over the old boring 1.77m deep [1264], which contained two pits each
mill complex, so that power could be transmitted 2.45m long, the northern one (for the drive wheel)
north and south to the two grinding mills. The being 1.1m wide and the southern one (for the
steam engine complex included a boiler house, flywheel) 0.62m wide. An iron pipe located in the
cylinder bed (truncated) and wheelpits from which north-west corner of this structure supplied oil to
power was transmitted around the site (Fig. 37). the wheelpits; excess oil was drained down into the
The boiler house consisted of two parallel watercourses. A brick floor to the north was also
brick walls running east–west [1197, 1261], each associated with this structure.
constructed of machine-pressed brick bonded in
cement mortar. These walls were up to 1.2m wide
PHASE XI: LATE 19TH CENTURY
and survived to a height of over 1m; they were
over 8.5m long. The boiler was situated at the In the same year that William Marlow was killed
eastern end of the boiler house building, stoked by a grindstone, Edward Elwell himself passed
from the eastern end (subsequently truncated). A away, albeit under more peaceful circumstances.
Lancashire-type cylindrical boiler was mounted According to the Walsall Observer, his funeral was
horizontally on a specially-made curved brick base attended by about 200 ‘stalwart wielders of the
[1260], with flues to both sides and an access space hammer, who gave vent to their sorrow in tears’.81
beneath [1262] which contained some clinker and The forge thus came under the management of his
ash [1283]. 23-year-old grandson, Alfred.82 A lithograph was
The original engine was mounted on a brick produced at around this time, showing the Elwell
base directly to the west of the boiler. It was prob- concern at its Victorian peak (Fig. 38).83 Despite
ably a horizontal engine, but its precise form could minor inaccuracies and exaggerations the litho-
not be determined since the brick base had later graph provides a fair overall impression of the

FIG. 37
Wednesbury Forge: steam engine
complex, showing original Phase X
installation (black) and later Phase XI
additions and modifications (light grey).

44-1-1-PMA 01.indd 40 3/1/2010 7:35:56 PM

 WEDNESBURY FORGE 41

FIG. 38
Wednesbury Forge: lithograph c. 1869, from an Elwell’s catalogue. The large chimney is founded on the former
windmill base; details of the house and pond arrangement are broadly accurate, including the east–west oriented
water-wheel. The overall outline of the forge buildings is also correct, as is the railway line leading to Bescot Junction
in the background, although scales are distorted. The curve of the railway viaduct bottom left is entirely imaginary.

site. The later Elwell period was characterized by base was replaced by a more substantial effort in
a series of minor improvements to the site itself concrete [1196]. Associated with this change was
(Fig. 39), but several significant interventions in the demolition of the spectacular 67m-high chim-
the surrounding landscape. These included the ney, said at the time to be the tallest in England.
construction of a church (St Paul’s) and housing Unfortunately (but perhaps unsurprisingly, given
for the workforce (Wood Green Square), together that it was founded on a 120-year-old windmill
with a playing field and other recreational facilities built in a pond) it had suffered subsidence. The
— unfortunately, outside the scope of this paper. demolition of the chimney coincided with the
Documentary evidence, in the form of census abandonment of the southern grinding pits within
returns, directories and newspaper accounts, the former windmill building. They were replaced
provides more detail about the forge and its by a new facility comprising a series of lightly-built
surrounding landscape during this period. walls and associated floor surfaces [1594, 1595,
By 1889 the site was said to have ‘steam 1622, 1636, 1677, 1678 and 1683] to the north and
engines of 200 horse power’.84 Although new west of the steam engine chimney, which evidently
engines were added elsewhere on site (outside the survived as a freestanding structure.
excavation area), this power increase also involved At the same time a new grinding wheel pit
alterations to the steam engine serving the grinding [1160] and associated machine base [1203, 1204]
mills. The boiler and engine house walls were were inserted into the forging complex in Building
strengthened by adding an external skin of blue I. Another lightweight steam hammer or press
engineering bricks [1194]. The original brick engine base was also constructed, employing reused

44-1-1-PMA 01.indd 41 3/1/2010 7:35:56 PM

 42 PAUL BELFORD

FIG. 39
Wednesbury Forge: Period 4, Phases XI and XII. Overall plan showing late 19th- and early 20th-century features.

grindstones, to the west of Building A [1409, 1412, lands by workers building the Grand Junction
1413]. Railway.85 The South Staffordshire Railway fol-
The arrival of the railway network also lowed in 1850, which was routed along a wooden
affected the forge. In 1834 Edward Elwell had viaduct over the pool. It was rebuilt in brick fol-
mobilized his workforce to prevent access to forge lowing an accident in 1859 when a goods train fell

44-1-1-PMA 01.indd 42 3/1/2010 7:35:59 PM

 WEDNESBURY FORGE 43
into the pool, by which time the two routes were PHASE XII: EARLY 20TH CENTURY
linked via the Bescot Curve (Fig. 2). Amalgamated Alfred Elwell died in 1902, and the firm became a
into the London and North Western Railway in private limited company (Edward Elwell Limited)
the 1840s, the Wednesbury railway network partly under the direction of Alfred’s nephew, Charles
encircled the forge. Bescot Junction, located less Edward Elwell. Freshly returned from the Boer
than a kilometre away, developed as a significant War, Charles apparently ‘restored discipline’ to
freight handling site, with the up marshalling yard the organization.88 He also made further improve-
completed by 1881 and further additions in the ments to the forging facilities, and modernized
1890s.86 A line was built from the forge to Bescot the water-power system by installing two simple
Junction during this period; it is shown on the single-jet Pelton-type turbines. The turbines were
lithograph (Fig. 38) and on subsequent mapping. installed from c. 1904,89 and represented the final
It was also encountered archaeologically, in the phase of water power at Wednesbury Forge.
form of two parallel lines of reused grindstones The northern wheelpit was modified to
accommodate the turbine by refacing most of the
[1100, 1807] running east–west across the site
eastern elevation with machine-made engineering
(partly shown in Fig. 35). These grindstones were bricks [1733]. A cast-iron bearing box [1734] was
all between 0.67m and 0.84m in diameter, and were set into the wheelpit wall, together with two iron
made of millstone grit; some had letters carved into I-section beams [1721] which crossed the wheelpit
them indicating their quarry of origin (M, O, P, T and supported the turbine superstructure. Two
and W).87 The grindstones were bedded in a further parallel slots for timber beams were evident
cindery levelling layer and formed the base for the in both eastern [1736] and western [1719] eleva-
track, which was also braced with wrought-iron tions. Modifications were also made to the sluice
straps, some of which survived in situ. structure and water management system (Fig. 40).

FIG. 40
Wednesbury Forge: northern turbine inlet shaft. Overhead view with south at the top of the photograph (scale 2m).

44-1-1-PMA 01.indd 43 3/1/2010 7:36:01 PM

 44 PAUL BELFORD
A wall [1728] was built at the northern end of cover for inspection and repair of the turbine
the sluices to divert the water down into the turbine mechanism.
via a brick-built shaft. This shaft [1586] was con-
structed on the western side of the wheelpit and
cut through the earlier floor surface [1729]; it PERIOD 5: WEDNESBURY FORGE
was 2.35x2.80m in plan and 3.45m deep, with (1929–2006)
an iron grille and timber sluice gate built into
its western side. The turbine inlet sat at the base of In 1929 Elwells merged with Chillingtons, a
this shaft and took the form of an inverted cone, Wolverhampton-based tool maker with a similar
made of thick rolled iron or steel plates riveted range of products and markets. The two firms
together. This channelled water through the wheel- operated as Edge Tool Industries Limited, although
pit wall and into the vertically-mounted body of maintaining their separate production facilities
the turbine. and brand identities. During this period the site
The water regulation mechanisms for the underwent extensive modernization, with large-
southern wheelpit had been destroyed; however scale demolition of earlier buildings and the reori-
the lower part of the turbine casing again survived entation of the site away from the historic core
more or less intact (Fig. 41). Again iron beams of the forge (Fig. 42).
had been inserted across the wheelpit in order to
support the turbine superstructure. The turbine
PHASE XIII: EARLY TO MID-20TH
casing [1646] was 1.5m in diameter and constructed
CENTURY
from wrought iron or steel plate riveted together.
As in the northern turbine a curved inlet pipe led to The turbine-based swansong of water power was
a vertical turbine casing; this had a detachable short-lived, and both turbines had been abandoned
by the early 1930s.90 The sluices controlling the
flow from the northern pool were demolished and
the pool filled in, shortly after the Great War.91 It
is of course possible that water from the south
pool could have been routed round into the tur-
bine, but by 1934 this too had been filled in. The fill
of the northern turbine sluice, and the southern
wheelpit both contained early 20th-century arte-
facts. Thus ended four centuries of water power
at Wednesbury. The sites of both pools were
encroached upon by housing and, later, the playing
fields of Wood Green School.
The former sluices and windmill were buried,
and a brick wall was built [1587, 1701] along the
western edge of the site. Everything to the east
of this was cleared to below-ground level. Thus
Building A was finally demolished and its northern
cellar infilled [1539, 1540]; most of Elwell’s house
was also removed, along with what was left of
the former boring and grinding mill. The area was
then concreted over and a series of new buildings
erected. These included a stamp shop (of which
some of the brick-paved internal floor surface
[1122] survived), stores, cycle sheds and a garage.92
A new gatehouse and time office were also built,
partly incorporating the cellars of Elwell’s house,
as well as a small first aid building attached to the
stores (Fig. 43).
Some alterations were also made to forging
operations in the eastern area. New flues [1024,
1062, 1070] were added at the western end, and new
FIG. 41 internal partitions were created at the eastern end.
Wednesbury Forge: Southern turbine casing (Phase XII) These walls [1370, 1380] were built of machine-
in the Phase VIII iron-framed wheelpit (scale 2m). pressed brick on lightweight spread concrete

44-1-1-PMA 01.indd 44 3/1/2010 7:36:02 PM

 WEDNESBURY FORGE 45

FIG. 42
Wednesbury Forge: Period 5, Phases XIII and XIV. Overall plan showing 20th-century features.

foundations. New concrete grinding pits were PHASE XIV: LATER 20TH CENTURY
installed in the southern area [1143, 1149]. The The site survived the Second World War undam-
alterations to Building E noted above — insertion aged. Post-war aerial photographs show recently-
of concrete floors and minor blockings — were also infilled pools and the building arrangement of
made during this phase. 1937.93 Edge Tool Industries Limited was taken

44-1-1-PMA 01.indd 45 3/1/2010 7:36:04 PM

 46 PAUL BELFORD

FIG. 43
Wednesbury Forge: First Aid building constructed during Phase XIII and in use until 2006 (scale 2m).

over by Eva Industries Limited in 1962, and Peter grinding and stamping facilities were installed in
Elwell remained managing director until 1967.94 It the eastern area. These included prefabricated
was around this time that Edward Elwell Limited grinding wheel bases [1328] and concrete machine
merged with Spear and Jackson and Brades and bases [1021, 1025, 1065, 1356] for stamping and
Tyzack. The ‘Elwell’ brand began to be discontin- forging apparatus. In the later 20th century, pro-
ued in the garden tool ranges where it was super- duction in this area was abandoned altogether,
seded by the group brand ‘Spearwell’, although the and the main focus of the forge moved to the east,
‘Elwell’ name continued for agricultural tools and outside the area of excavation. The buildings in the
export lines.95 In 1970 Wednesbury Forge became eastern area were demolished, and a new mainte-
wholly owned by Spear and Jackson, who were nance facility erected in a steel-framed, steel-clad
later subsumed into the James Neill group.96 shed on a concrete base. This facility stored dyes
This newly-merged concern started life opti- and tools and also included facilities for their
mistically with the construction of an office block repair.
to replace the former Building E. This range (seen The basic organization of the site established
in the background of Fig. 31 above) was of rein- in the 1930s was retained through to its closure.
forced concrete construction, providing open-plan The processes and functions were recorded during
accommodation for design and marketing on the historic landscape appraisal phase in 2001
the ground floor and boardroom and executive (Fig. 44). However, between then and the field
facilities on the first floor. evaluation of 2005 the scale of forging operations
On the production side, the Phase XIII stamp on site had contracted significantly. Recording of
room was demolished, together with the Phase forging processes was undertaken in July 2005
XI buildings formerly in the southern area. New (Fig. 45). At this time the manufacture of spades,

44-1-1-PMA 01.indd 46 3/1/2010 7:36:05 PM

 WEDNESBURY FORGE 47

FIG. 44
Wednesbury Forge: plan of the site in 2001, showing forging operations and processes. The area of the 2006 and 2007
excavation is shaded.

for example, required a 13-stage process using five a hydraulic press to form the ‘dish’ in the blade and
or six men.97 Billets of Swedish carbon steel were the curve of the socket. Although presses had
heated to 1,200°C, then cut into four on die-set replaced water-powered hammers, and the furnaces
tools. The same press spread one end to form the ran on gas rather than charcoal, the 16th-century
blade; then the other end was rolled five times to forge workers would have recognized this process
form the socket. After reheating to about 1,100°C,
being carried on by their descendants five centuries
between five and seven blows with a forming tool
shaped the socket, and, with the metal still hot, the later.
blade end was widened further. The blade end was Forging at Wednesbury ceased at Christmas
reheated again, and then subjected to four or five 2005. The site continued in use for administrative,
passes in the blade rolling mill to thin the metal and warehousing and maintenance functions during
draw out the spade shape. This was then blanked 2006, but the remaining buildings were removed in
on another press, before being finally worked in 2007 and the site was cleared.

44-1-1-PMA 01.indd 47 3/1/2010 7:36:07 PM

 48 PAUL BELFORD

FIG. 45
Wednesbury Forge: making a shovel, July 2005.

DISCUSSION timber-framed wheelpit of 17th-century date, simi-

lar to Wednesbury’s Phase I northern wheelpit.99
The archaeology of Wednesbury Forge has raised Excavation of a 17th-century finery forge at
a number of interesting questions about post- Blackwater Green (Sussex) revealed two parallel
medieval ironworking technology. Key aspects timber-framed double tailraces slightly over 12m
include the long-standing use of water power, apart and analogous to the Phase II and III
the early development of integrated production, arrangements at Wednesbury.100
and the flexibility and adaptability of the site for The initial arrangement of water power
changing market conditions. determined the layout of the site for the next
four centuries. When the water-power system was
upgraded, great care was taken to ensure that
POWER AND SPACE
the forge could continue in operation as much as
Taken together, the evidence of the water-power possible. Fixed plant such hammers, anvils and
installations suggests a sophisticated approach to furnaces were therefore worked around rather
water management. Comparable features have than relocated, and supplementary power sources
been encountered at other 16th- and 17th-century (the 18th-century windmill and the 19th-century
excavated sites, albeit on a smaller scale. The forge steam engine) located at the heart of the forge,
at Ardingly (Sussex), in use c. 1570–1660, consisted where their power could be harnessed directly by
of two single parallel timber tailraces about 12m the existing infrastructure. A similar motive can be
apart, driving a hammer and bellows situated detected in the culverting process, where the exist-
between them.98 The layout and details of construc- ing base timbers were reused and the culverts built
tion echo the earliest phase at Wednesbury. Exca- in stages to minimize disruption. The culverting
vations at Chingley (Kent) revealed a curved-back process appears to have begun relatively early —

44-1-1-PMA 01.indd 48 3/1/2010 7:36:09 PM

 WEDNESBURY FORGE 49
certainly during the later 17th century — and the although the operations may not have been cen-
main culvert of the early 18th century was built on tralized. Angerstein noted ‘a number of workshops
an impressive scale. This enabled expansion of the for saw blades’ on his way to Wednesbury,102 and
site to take place. his fellow Swede Samuel Schröder, who had visited
Only in the 19th century did the operational west Midlands saw-makers in 1749, recorded how
focus begin to shift away from the historic core. forged blades were ‘sent to a grinding mill about
The steam engine installation found during the five miles away to be ground, then sent back for
excavation was too severely truncated to enable the teeth to be filed’.103 One characteristic of the
an accurate calculation to be made of its potential Midlands saw industry at this time was the adop-
power output; however, it was clearly smaller than tion of rolling mill technology for creating the
the ‘200 horsepower’ noted in 1889. A substantial saw plate, rather than hammering it out, as had
engine house is shown on the lithograph at the formerly been the case. This gave a much more
eastern end of the site, outside the area of excava- homogeneous product, less prone to failure. The
tion, probably driving a rolling mill. This installa- ‘plating mill’ at Wednesbury in the early 18th
tion provoked the eastwards migration of forging century would have been ideal for this process, and
processes, which accelerated during the 20th the newly-expanded site contained a wide range
century after the abandonment of the turbines. of small hand forges (Buildings C and D), as well
Non-production elements of the forge also as the grinding mills. It is therefore probable that
maintained a persistent layout. Access was proba- all, or almost all, of the processes of saw-making
bly always from the west, the later-named St Paul’s took place on site at Wednesbury Forge, making it
Road forming the main route from the town a very early example of an integrated saw factory.
of Wednesbury itself. Apart from the pools and Having established a site which could turn
chimneys, the approach would have been domi- iron bar into finished products, the Willetts con-
nated by more polite buildings. Willetts’ house, cern was well-placed to adapt to changing markets.
Elwell’s house and the 19th- and 20th-century The opportunity arose to turn the forging and
administrative and office buildings were all located grinding facilities to other uses, in this case gun-
within the same general area; indeed the 20th- making. Wednesbury’s 18th-century contribution
century gatehouse overlapped with the footprint to gun-making has traditionally been associated
of Elwell’s house, which partly incorporated the with barrels, leading to the town’s later 19th-
remains of Willetts’ house (Building A). There was century specialism in tube-making.104 However,
also a cluster of workers’ housing to the north of given the size of the Wednesbury Forge site and
the entrance during the 18th and 19th centuries. the wide range of processes which it was capable of
Assemblages of domestic artefacts from all periods undertaking, it is probable that the firm applied
and all social levels suggest very close connections its experience of fully-integrated manufacturing to
between work and home for most of the forge’s the new product. Barrel-making also involved a
history. series of forging, cutting and grinding operations
on steel plate,105 and the archaeological evidence
suggests that the Willetts concern was involved in
PRODUCTS AND MARKETS
making components other than barrels. The large
When the forge was part of the Foley concern, its quantity of gun flints may have been brought in
output was simply one part of a much wider chain simply for testing the barrels, but again the tech-
of production. During the 18th century however, nology existed on site for making flintlocks and
it developed as a site of increasing complexity. assembling lock and barrel. Stocks may have
The archaeological evidence shows that serious been sourced from outside, although saw-making
expansion of the site took place during the would have required woodturning capacity and
early 1700s, and this would appear to have been if it already existed on site then it could have
associated with saw manufacture. been adapted from saw handles to gunstocks. The
Saw-making involved a number of closely- installation of the post-1788 boring mill would
connected processes. Even in the 20th century have improved capacity, and it is tempting to think
many of these defied mechanization; Simon Barley that the new boring mill may have been for rifling.
records fifteen stages from steel ingot to finished As with the transition from saws to guns, so
product.101 One of the most crucial was creating the the forge was well suited to its 19th- to 21st-century
steel plate and then grinding it to produce a taper role as an edge-tool factory. For most of this time
from saw-back to saw-edge. The teeth were then Elwells had a product line of over 1,200 items,
cut, set and sharpened before backs and handles including sickles, scythes, billhooks, axes, spades,
were added. Wednesbury certainly emerged as shovels, forks, hoes and shears. Many of these were
a centre of this trade during the 18th century, specialized products for colonial markets, such as

44-1-1-PMA 01.indd 49 3/1/2010 7:36:11 PM

 50 PAUL BELFORD
the famous Elwell’s cocoa pruner, which was first of his daughter) during the three years in which he
listed in the 19th century and still in production in was engaged on the project. All the drawings in this
2005. paper were prepared by Sophie Watson, except
Figs 4, 11, 13, 14, 21, 22, 26 and 34, which were
drawn by Sophie Watson and Keith Hinton.
CONCLUSION Photographs were taken by Paul Belford and
William Mitchell. Staff who worked on the project
The programme of archaeology undertaken at over the years included Elizabeth Bishop, Emma
Wednesbury Forge has revealed a unique site of Dwyer, Richard Elliot, Keith Hinton, Matthew
national and international significance. This small Morgan, William Mitchell, Jonathan Prince, Kate
but powerful enterprise was built on trees that were Page-Smith, Suzanne Reeve, Jeremy Rogers,
growing when men were fighting at Agincourt. Simon Roper, Anna Wallis, Sophie Watson and
By the beginning of the 17th century it was already Alex Wilkinson. Specialist contributions have been
a substantial enterprise, and from the early 1700s invaluable in refining the dating and sequence
developed as an integrated factory making a of activities on the site, and particular gratitude is
diverse range of products that were sold around extended to Dr Ian Tyers (Dendrochronological
the world. Wednesbury-made saws, guns and edge Consultancy Limited) and Stephanie Ratkai
tools were literally at the cutting edge of imperial (ceramics). Local knowledge has been enhanced
expansion in the 18th and 19th centuries. Wednes- by Peter Knowles and the Wednesbury Local His-
bury Forge provided the means by which forestry, tory Society, and significant contributions were
agriculture and industry could be undertaken made in research and discussion by Dr Peter King,
around the world, and made the armaments with David Cranstone, Dr Simon Barley, Dr David
which these undertakings were defended. The story Dungworth and Ian Bott.
of Wednesbury Forge is a truly global one. Yet the
site is also firmly rooted in its locality. Generations
of Wednesbury families were associated with the NOTES
forge during five centuries of iron making, and it is 1
Greenslade & Jenkins 1967, 68–9.
their hard work, enterprise, courage and skill that 2
Gelling 1992, 91–3.
are reflected in the archaeology. The people of 3
Hackwood 1902; Hodder 1992, 98–112.
Wednesbury made the forge, and the forge made 4
Dilworth 1976,102.
Wednesbury one of the most important centres 5
Hodder & Glazebrook 1987, 75–6.
of English industry during its most heroic period. 6
Ede 1962, 26–7, 30; Dilworth 1976, 111.
Wednesbury forged the modern world. 7
Ede 1962, 108.
8
Hodder 1992; Stephanie Ratkai, pers. comm.
9
Belford et al. 2006; Hodder & Glazebrook 1987;
ACKNOWLEDGEMENTS Hodder 1992.
10
Dilworth 1976, 104–5.
Archaeological work was supported by the devel- 11
Dilworth 1976, 107.
opers Opus Land. Particular thanks must go to 12
Dilworth 1976, 108.
Geoffrey Barrett, of Crouch Butler Savage Limited, 13
Hackwood 1902, 36.
who facilitated archaeological investigations on 14
Dilworth 1976, 97.
behalf of his client. The wholehearted support 15
King 2006, 75.
of Sandwell MBC archaeologist Graham Eyre- 16
SHC: 1932, 298–9.
Morgan was essential and much appreciated, 17
Greenslade & Jenkins 1967, 113 fn.
together with that of his predecessor, Shane Gould. 18
Belford & Mitchell 2009.
All of the staff at Spear and Jackson were enor- 19
Belford & Reeve 2001.
mously helpful before, during and after the closure 20
Belford & Mitchell 2005.
of their long-established workplace. Documents 21
Greenslade & Jenkins 1967, 113 fn.
held in the Spear and Jackson offices at Wednes- 22
Ede 1962, 124.
bury Forge (SJWF) were seen whilst fieldwork was 23
King 1999, 59–76.
in progress 2005–7. Some were copied (copies are 24
Ede 1962, 124–5; Peter King, pers. comm.
with the site archive) but the fate of the originals 25
Lead 1977, 3–5.
after site closure is not known. 26
TNA: PRO C 2/James 1/C16/41; Ede 1962, 125.
Archaeological fieldwork was managed by 27
Dilworth 1976, 112.
the author, and directed on site by the author and 28
TNA: PRO C 2/JamesI/C16/41. It is possible that
William Mitchell, who endured many vicissitudes further disturbances in Wednesbury by Comberford
(including marriage, buying a house and the birth himself had caused disruption to the forge: he was

44-1-1-PMA 01.indd 50 3/1/2010 7:36:11 PM

 WEDNESBURY FORGE 51
70
fined by the Star Chamber in 1609 for misdemeanours Barley 2008, 96-102
71
causing personal injury to others (see CSPD James I, Dilworth 1976, 115.
72
1603–1610 (1857), 524–40). In the private collection of Ian Bott, Wednesbury;
29
Tyers 2007, 5–6; see also Belford & Mitchell 2009, see also Belford & Mitchell 2009, fig. 16.
73
Appendix 2. Glew, cited in Hackwood 1889, 51.
30 74
TNA: PRO Probate 11/148, ff. 335–6. SJWF; see also Belford & Mitchell 2009, figs 12
31
TNA: PRO Probate 11/239; Probate 11/246. and 13.
32 75
Ede 1962, 82–3; Hackwood 1902, 68. Ede 1962, 234.
33 76
Ede 1962, 112. TNA: PRO Census HO107/984/1.
34 77
Dilworth 1976, 113; Johnson 1952, 322–5. Elwell 1976, 26.
35 78
Schafer 1978, 23; Dilworth 1976, 113. TNA: PRO Census RG9/2032.
36 79
Schafer 1978, 23. Elwell 1976, 47–8.
37 80
HRO: F/VI/KG/1-2. Hackwood 1889, 52.
38 81
HRO: F/VI/KG/8. Elwell 1987, 7.
39 82
Ede 1962, 126. TNA: PRO Census RG10/2987.
40 83
Schafer 1978, 23. Illustrated in an Elwell catalogue of c. 1869
41
Dilworth 1976, 113. (reproduced by kind permission of Ian Bott).
42 84
Greenslade & Jenkins 1967, 118. Hackwood 1889, 52.
43 85
Peter King, pers. comm.; see also King 2003, 182– Elwell 1987, 6–8.
86
4. WLHC: 118/9, 23 Oct 1902; Greenslade 1976,
44
Dilworth 1976, 113. 11–14.
45 87
Johnson 1952, 326. Tucker 1985, 51.
46 88
Dilworth 1976, 114. Ede 1964, 286; Elwell 1987, 13–14.
47 89
Ede 1962, 126. Dilworth 1976, 117.
48 90
Dilworth 1976, 111. Elwell 1964, 88.
49 91
Angerstein 2001, 50. SJWF: letter dated 10 January 1968; see also
50
Hackwood 1889, 80. Belford & Mitchell 2009, figs 10–11.
51 92
Dilworth 1976, 114. SJWF: map of the site dated 1937.
52 93
Cited in Bagnall 1854, 110. NMR: CPE/UK/2466 (March 1948); 541/29 (May
53
Angerstein 2001, 48. 1948) and 82/780 (May 1953).
54 94
SRO: 5350/44–5. Abley 1990, 5.
55 95
Hackwood 1889, 51. Dreaper 1968, 36–9.
56 96
Barley 2008, 96-101. Abley 1990, 5.
57 97
Angerstein 2001, 49; Peter King, pers. comm. This description is taken from oral recording
58
A skelp is a plate of metal that has been curved and undertaken in 2005; there is insufficient space to
welded to form a tube (the basis for the gun barrels). provide more detail here, but this is preserved in
59
Johnson 1960, 71–2; Belford & Reeve 2001, 11–12. the fieldwork archive and will be available in the
60
SRO: D1798/617/87, D1798/617/92. monograph currently in preparation.
61 98
Wheelswarf is an accumulation of fine particles of Bedwin 1976, 38–50.
99
metal and grindstone, which are produced during the Crossley 1975, 6–29.
100
grinding process. Crossley 1994, 168.
62 101
Ede 1962, 165. Barley 2008, 397–9.
63 102
Dilworth 1976, 114. Angerstein 2001, 51.
64 103
Rees 1820 (3), Pl. 2 (many thanks to David Schröder 2008, 208–10.
104
Cranstone for pointing out this reference). Hackwood 1902, 15–25.
65 105
Belford & Reeve 2001, 13. Wilkinson 1971, 19–20.
66
Hackwood 1889, 52; SRO: Q/RPL/4/29.
67
This wheelpit was built only a few years after the BIBLIOGRAPHY
erection of the famous Iron Bridge at Coalbrookdale
in 1779, and anticipated the more ambitious Dither- Abley, R.W. 1990, ‘A Brief History of Wednesbury
ington Flax Mill of 1796. Forge and the Elwell Connection’, unpubl. MS.
68
William Elwell also had a significant share in the Angerstein, R.R. 2001 [1755], Illustrated Travel Diary
Coalbrookdale Company as it was building the Iron 1753–1755, trans. T. & P. Berg, London: Science
Bridge and was sometime Mayor of Walsall. See Museum.
Elwell 1964, 81–8; Greenslade & Jenkins 1967, 260; Bagnall, J.N. 1854, A History of Wednesbury in the
Abley 1990, 2; Baugh 1985, 36. County of Stafford, Wolverhampton: William
69
Abley 1990, 2; Elwell 1987, 6. Parke.

44-1-1-PMA 01.indd 51 3/1/2010 7:36:11 PM

 52 PAUL BELFORD
Barley, S.L. 2008, ‘Hand Tool Manufacture During Hodder, M.A. & Glazebrook, J.M. 1987, ‘Excava-
the Industrial Revolution: Sawmaking in Shef- tions at Oakeswell Hall, Wednesbury, 1983’,
field c.1750–c.1830’, University of Sheffield PhD Trans. Staffordshire Archaeol. Hist. Soc. 27,
thesis. 64–77.
Baugh, G.C. (ed.) 1985, The Victoria County History Johnson, B.L.C. 1952, ‘The Foley Partnerships: the
of Shropshire 11, Oxford: University Press. iron industry at the end of the charcoal era’,
Bedwin, O. 1976, ‘The excavation of Ardingly Econ. Hist. Rev. 4:3, 322–40.
Fulling Mill and Forge 1975–76’, Post-Medieval Johnson, B.L.C. 1960, ‘The Midland iron industry in
Archaeol. 10, 34–64. the early eighteenth century’, Business Hist. 2:2,
Belford, P., Eyre-Morgan, G., Mitchell, W. & Roper, 67–74.
S. 2006, ‘Archaeological Investigations at Lower King, P. 1999, ‘The development of the iron industry
High Street, Wednesbury’, Ironbridge Archaeol. in South Staffordshire in the early seventeenth
Ser. 175, unpubl. rep. century: history and myth’, Trans. Staffordshire
Belford, P. & Mitchell, W. 2009, ‘Archaeological Archaeol. Hist. Soc. 38, 59–76.
Excavations at Wednesbury Forge, Wednes- King, P. 2003, ‘The Iron Trade in England and
bury’, Ironbridge Archaeol. Ser. 230, unpubl. Wales 1500–1815: the Charcoal Iron Industry
rep. and its Transition to Coke’, University of
Belford, P. & Reeve, S. 2001, ‘Historic Landscape Wolverhampton PhD thesis.
Appraisal: St Paul’s Road, Wednesbury’, King, P. 2006, ‘Perry Barr and its water mills’, Trans.
Ironbridge Archaeol. Ser. 106, unpubl. rep. Staffordshire Archaeol. Hist. Soc. 41, 65–78.
Crossley, D.W. 1975, The Bewl Valley Ironworks, Lead, P. 1977, ‘The North Staffordshire iron industry
London: Royal Archaeological Institute. 1600–1800’, Hist. Metall 11:1, 1–14.
Crossley, D.W. 1994, Post-Medieval Archaeology in Rees, A. 1820, The Cyclopaedia, or Universal Diction-
Britain, Leicester: University Press. ary of Arts, Sciences and Literature, London:
Dilworth, D. 1976, The Tame Mills of Staffordshire, Longman, Hurst et al.
London: Phillimore. Schafer, R.G. (ed.) 1978, A Selection from the Records
Dreaper, J. 1968, ‘Geared for export: three case of Philip Foley’s Stour Valley Ironworks 1668–74,
histories’, Design 234, 30–9. Worcestershire Hist. Soc. New Ser. 9.
Ede, J. 1962, History of Wednesbury, Wednesbury: Schröder, S. 2008 [1749], Dagbok rörande Handel,
Simmons. När ingar och Manufakturer m.m.Ut i Danmark,
Elwell, C.J.L. 1964, The Iron Elwells, Ilfracombe: Holland, England, Frankrike och Tyskland,
Stockwell. Kungliga Biblioteket, Stockholm, X303 (1), in
Elwell, C.J.L. 1976, A Lady of Wednesbury Forge, Barley 2008, 208–10.
Tipton: Black Country Society. Tucker, D.G. 1985, ‘Millstone making in the Peak
Elwell, C.J.L. 1987, ‘The workers of Wednesbury District of Derbyshire: the quarries and the
Forge’, The Blackcountryman 20:4, 6–14. technology’, Ind. Archaeol. Rev. 8:1, 42–58.
Gelling, M. 1992, The West Midlands in the Early Tyers, I. 2007, ‘Tree-ring Spot Dates from Archaeo-
Middle Ages, Leicester: University Press. logical Samples: Wednesbury Forge, Sandwell’,
Greenslade, M.W. (ed.) 1976, The Victoria History of Ironbridge Archaeology unpubl. rep.
the County of Stafford 17, Oxford: University Wilkinson, F. 1971, Flintlock Guns and Rifles,
Press. London: Arms & Armour Press.
Greenslade, M.W. & Jenkins, J.G. 1967, The Victoria
History of the County of Stafford 2, Oxford:
University Press.
ABBREVIATIONS
Hackwood, F.W. 1889, Wednesbury Workshops; CSPD Calendar of State Papers Domestic
or Some Account of the Industries of a Black HRO Herefordshire Record Office
Country Town, Wednesbury: Woden Press. NMR National Monuments Record
Hackwood, F.W. 1902, Wednesbury Ancient and SHC Staffordshire Historical Collections
Modern, Wednesbury: Ryder & Son. SRO Staffordshire Record Office
Hodder, M.A. 1992, ‘Excavations in Wednesbury, SJWF Spear and Jackson offices,
1988 and 1989: the medieval and post-medieval Wednesbury Forge.
settlement, and the 17th-century pottery indus- TNA: PRO The National Archives (TNA):
try’, Trans. Staffordshire Archaeol. Hist. Soc. 32, Public Record Office (PRO)
96–115. WLHC Walsall Local History Centre

44-1-1-PMA 01.indd 52 3/1/2010 7:36:12 PM

 WEDNESBURY FORGE 53
SUMMARY IN FRENCH, GERMAN, ITALIAN AND SPANISH

RÉSUMÉ RIASSUNTO
Cinq siècles de travail du fer: les fouilles de la Cinque secoli di lavoro del ferro: gli scavi nella
Forge de Wednesbury fucina di Wednesbury
Les fouilles archéologiques menées entre 2004 Gli scavi intrapresi fra il 2004 e il 2008 presso la
et 2008 à la « Wednesbury Forge » (une forge de fucina di Wednesbury, a Wednesbury, nelle West
feronnerie) à Wednesbury, West Midlands ont Midlands, hanno restituito abbondanti resti di
permis la mise au jour d’importants vestiges de legno e di strutture in muratura, nonché altri mate-
bois, de maçonneries et d’autres structures. Les riali. Le testimonianze storiche e archeologiche
indices historiques et archéologiques ont révélé hanno messo in luce un sofisticato complesso per
un complexe sophistiqué de travail du fer en place la lavorazione del ferro, già esistente nel 1600 ca.
depuis 1600 environ, qui fut successivement adapté
e che in seguito fu costantemente riadattato e
et redéveloppé de manière continue jusqu’à la
fermeture du site en 2005. Les procédés incluaient risanato fino alla chiusura del sito nel 2005. Il
des forges de raffinement et de chaufferie, des pro- processo di lavorazione comprendeva forni di
ductions de clous, de scies, d’armes et une taillan- puddellaggio e riscaldamento, la fabbricazione di
derie. Des développements tardifs comprenaient chiodi, seghe, fucili e utensili taglienti. Gli sviluppi
un moulin à vent pour broyer, des réseaux de successivi inclusero una mola azionata dalla forza
chemin de fer internes, des turbines à eau, des lami- eolica, una rete ferroviaria interna, turbine ad
noirs, des équipements domestiques et récréatifs acqua, presse a rulli, strutture abitative e ricreative
pour les travailleurs. Les études archéologiques ont per i lavoratori. L’indagine archeologica com-
consisté en des recherches documentaires, des prende la ricerca documentaria, lo scavo, nonché
fouilles, des relevés en élévation, l’enregistrement la documentazione degli edifici, delle testimonianze
de l’histoire orale et des procédés. orali e del processo di lavorazione.

ZUSAMMENFASSUNG RESUMEN
Fünf Jahrhunderte Eisenbearbeitung: Ausgrabun- Cinco siglos trabajando el hierro: las excavaciones
gen In der Wednesbury Schmiede en la Forja de Wednesbury
Archäologische Ausgrabungen, die zwischen 2004– Las excavaciones arqueológicas desarrolladas
08 bei der Wendnesbury Schmiede, Wednesbury, entre 2004 y 2008 en la Forja de Wednesbury en
West Midlands, stattfanden, erbrachten ausgiebige Wednesbury, West Midlands, descubrieron restos
Überreste von Holz- und Mauerstrukturen, sowie importantes de estructuras de madera y piedra
andere Elemente. Historische und archäologische junto con otros elementos de interés. La evidencia
Zeugnisse zeigen einen komplizierten Eisenbear- histórica y arqueológica descubrió un sofisticado
beitungskomplex, der ungefähr um 1600 begann
complejo para el trabajo del hierro; éste estuvo
und in der Folge bis zu seiner Schließung im Jahre
en existencia ya hacia 1600 y fue sucesivamente
2005 kontinuierlich angepaßt und neu entwickelt
wurde. Der Arbeitsablauf schloß Frischofen und adaptado y desarrollado hasta su cierre en el 2005.
Wulstbrand ein, Nägel- und Sägenproduktion, Las labores allí desarrolladas incluyen la forja fina
sowie die Herstellung von Gewehren und Schnei- y tosca, fabricación de clavos, sierras, armas y
dewerkzeugen. In späterer Zeit gehörten eine utensilios con filo. Algo más tardía fue la adición
windgetriebene Schleifmühle sowie ein internes de un molino de viento, una red de ferrocarril
Eisenbahnnetz, Wasserturbinen, Walzwerke, sowie interno, turbinas de agua, laminadores, y estructu-
Arbeiterwohnungen und Freizeiteinrichtungen ras de recreo y alojamiento para los trabajadores.
dazu. Die archäologischen Untersuchungen schlos- La investigación arqueológica incluyó la investig-
sen urkundliche Nachforschungen, Ausgrabungen, ación de las fuentes escritas, excavación, estudio de
Bauunterlagen, mündliche Überlieferung, und paramentos, historia oral y estudio de los procesos
schriftliche Herstellungsmethoden ein. de manufactura.

Ironbridge Gorge Museum Trust, Coach Road, Coalbrookdale, Telford, Shropshire TF8 7DQ, UK
[[email protected]]

This paper is published with financial assistance from Opus Land

44-1-1-PMA 01.indd 53 3/1/2010 7:36:12 PM