13: RECENT WORK
COINS
A total of 586 coins was recovered, of which 189 were not
attributable to a particular emperor due to their advanced
state of wear. A small group of these, comprising a total of
166 coins, was commented upon in 1989 (Casey 1989).
Unfortunately, once identified, these coins were not kept
separate from the material found after 1989 and it was
therefore necessary to catalogue and comment on them a
second time. Some of the coins identified by Casey were
no longer present in the assemblage, most notably, a
denarius of Titus, raising questions about the completeness
and integrity of the assemblage.
Initial identification of all coins has been completed and,
although minor details may be subject to change, summary
tables are provided showing the incidence of coin by
emperor and denomination (TABLES 13.1-13.2). Roger
Bland identified the radiates whilst Ian Leins provided
assistance with the remainder of the assemblage. The coins
have been assigned Reece periods (Reece 1987, 71-6)
which enables graphical representation of the material (FIG.
13.4), and comparison with other sites (TABLE 13.3).
It is evident that the composition of the river assemblage
is quite different to that of the excavation material reported
on by Brickstock in this volume and some specific areas of
variation are worthy of note. From the excavations, there
TABLE 13. l: COINS FROM THE RIVER AT PIERCEBRIDGE BY EMPEROR AND DENOMINATION
THE FINDS FROM THE RIVER
Philippa Walton
INTRODUCTION
Since the mid 1980s, divers Bob Middlemass and Rolfe
Mitchinson have recovered hundreds of objects from a site
on the bed of the River Tees at Piercebridge centred on NZ
21301565. The objects, ranging in date from the late Iron
Age to the medieval period, were recovered through a
combination of underwater metal detecting and 'eyes only'
retrieval. Whilst some were easily retrievable from the
riverbed, others were found encased within a hard, black
concretion comprising iron corrosion products and organic
material. The divers were able to recall the underwater
distribution of some of the material, and have created plans
of some associated wooden posts and structures. However,
they have not kept detailed findspot records for all objects
and therefore only general comments on their spatial
distribution are possible.
In 2003, the divers approached the author, then Finds
Liaison Officer for the North East, to record their finds
with the Portable Antiquities Scheme. This is a voluntary
initiative to record archaeological objects found by
members of the public in England and Wales. Since its
introduction in 1997, the Scheme's network of Finds
286
Liaison Officers have recorded more than 350,000 objects.
For more information, visit www.finds.org.uk. The riverbed
finds assemblage has been repackaged, sorted and
quantified, and the cataloguing of much of the material
has been started on the Scheme database
(www.findsdatabase.org.uk). However, significant
further study and analysis is necessary and work is
ongoing.
This summary aims to provide an overview of the
material from the river. Due to the large body of material
and the early stage of research, it is by no means exhaustive
but will concentrate on the coins and small finds belonging
to three functional categories - personal adornment,
military and religious. Other finds will be commented on
at a later stage and include medical or cosmetic instruments,
seal boxes, weighing equipment, household equipment and
fixtures and fittings. It should also be noted that there is
also a large assemblage of fine and coarseware vessels in
both imported and Romano-British fabrics including many
near complete vessels. At present, Lucy Cramp of Reading
University is undertaking lipid analysis on the mortaria from
the river deposit, the results of which will be published in
due course.
Date
1
Republican
Juba II
Nero
Galba
Vespasian
Domitian
Nerva
Trajan
Sabina
Hadrian
Aelius
Antoninus Pius
Faustina I
Marcus Aurelius
Faustina II
Lucius Verus
Lucilla
Commodus
Septimius Severus
JuliaDomna
Caracalla
Geta
Julia Maesa
Elagabalus
Julia Soaemias
Julia Mamaea
Severus Alexander
Orbiana
Maximinus
Otacilia Severa
Gordian III
Philip I
Volusian
Valerian
Gallienus
Salonina
Claudius II
Quintillus
Postumus
Victorinus
Tetricus I
Tetricus II
Victorinus or Tetricus I
Constantine I
Helena
Constantius II
Eugenius
Arcadius
12
2
5
2
9
6
I
7
I
19
16
9
3
9
I
2
6
23
7
20
6
4
12
9
28
2
3
3
14
3
7
I
10
3
5
1
6
3
5
4
5
I
2
5
2
11
2
2
I
2
3
4
3
7
II
6
7
8
9
10 Total
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
I
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3
4
7
7
2
3
9
1
4
2
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
2
I
2
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
2
8
4
4
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3
3
I
4
n/a
n/a
n/a
n/a
n/a
12
1
6
2
12
6
2
22
2
29
45
14
5
18
3
9
29
10
25
6
7
16
3
21
42
3
3
4
14
2
2
3
9
6
1
2
11
7
5
5
2
I
2
Key: I. Denarius; 2. Denarius (copy); 3. Sestertius; 4. Dupondius; 5. As; 6. Other; 7. Radiate; 8. Radiate (copy);
9. Nummus; 10. Siliqua
287
13: RECENT WORK
COINS
A total of586 coins was recovered, of which 189 were not
attributable to a particular emperor due to their advanced
state of wear. A small group of these, comprising a total of
166 coins, was commented upon in 1989 (Casey 1989).
Unfortunately, once identified, these coins were not kept
separate from the material found after 1989 and it was
therefore necessary to catalogue and comment on them a
second time. Some of the coins identified by Casey were
no longer present in the assemblage, most notably, a
denarius of Titus, raising questions about the completeness
and integrity of the assemblage.
TABLE
Initial identification of all coins has been completed and,
although minor details may be subject to change, summary
tables are provided showing the incidence of coin by
emperor and denomination (TABLES 13.1-13.2). Roger
Bland identified the radiates whilst Ian Leins provided
assistance with the remainder of the assemblage. The coins
have been assigned Reece periods (Reece 1987, 71-6)
which enables graphical representation of the material (FIG.
13.4), and comparison with other sites (TABLE 13.3).
It is evident that the composition of the river assemblage
is quite different to that of the excavation material reported
on by Brickstock in this volume and some specific areas of
variation are worthy of note. From the excavations, there
13. l: COINS FROM THE RIVER AT PIERCEBRIDGE BY EMPEROR AND DENOMINATION
Date
1
Republican
Juba II
Nero
Galba
Vespasian
Domitian
Nerva
Trajan
Sabina
Hadrian
Aelius
Antoninus Pius
Faustina I
Marcus Aurelius
Faustina II
Lucius Verns
Lucilla
Commodus
Septimius Severns
JuliaDomna
Caracalla
Geta
JuliaMaesa
Elagabalus
Julia Soaemias
Julia Mamaea
Severns Alexander
Orbiana
Maximinus
Otacilia Severa
Gordian III
Philip I
Volusian
Valerian
Gallienus
Salonina
Claudius II
Quintillus
Postumus
Victorinus
Tetricus I
Tetricus II
Victorinus or Tetricus I
Constantine I
Helena
Constantius II
Eugenius
Arcadius
12
2
5
2
9
6
1
7
1
19
16
9
3
9
1
2
6
23
7
20
6
4
12
9
28
2
3
3
4
5
14
3
7
1
10
3
5
1
6
3
5
1
2
5
11
2
2
2
1
2
3
4
3
7
11
6
7
8
9
10
Total
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
1
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3
7
2
3
9
1
4
2
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
12
1
6
2
12
6
2
22
2
29
4
7
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
2
I
2
2
8
4
4
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
n/a
3
3
I
4
n/a
n/a
n/a
n/a
n/a
45
14
5
18
3
9
29
10
25
6
7
16
3
21
42
3
3
4
14
2
2
3
9
6
1
2
ll
7
5
5
2
l
2
Key: I. Denarius; 2. Denarius (copy); 3. Sestertius; 4. Dupondius; 5. As; 6. Other; 7. Radiate; 8. Radiate (copy);
9. Nummus; 10. Siliqua
287
ROMAN PIERCEBRIDGE
TABLE
13.2: UNATTRIBUTED COINS FROM THE RIVER AT
PIERCEBRIDGE
TABLE
13.3 THE COINS FROM THE RIVER AT PIERCEBRIDGE BY
REECE PERIODS
BY DENOMINATION
Type
Reece
Period
Number
Denarius
Denarius (eopy)
Sestertius
Dupondius or As
As
Radiate
Radiate (copy)
Nummus
Siliqua
Total
17
15
21
59
18
7
49
5
2
189
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
is a complete absence of coinage prior to the reign of
Vespasian (69-79) apart from a single denarius of Mark
Antony. In contrast, however, the river assemblage includes
twelve Republican denarii, a bronze of Juba II and eight
Neronian issues. It is significant that all the identifiable
Republican issues were of Mark Antony and that they were
recovered by just one of the divers, Rolfe Mitchinson. This
suggests that they represent the contents of a single
dispersed hoard, which on a dry site might be considered a
'foundation hoard'.
The excavations exhibit a 'normal' but depressed pattern
of coin loss throughout the remainder of the 1st and 2nd
centuries. The river assemblage, however, exhibits the exact
opposite with a steady increase through Reece periods 4 to
6 (69-138) and a significant peak in Reece period 7 (138161 ). A further 98 worn Roman bronzes dating from the
late 1st or 2nd centuries but not assigned to a particular
emperor would undoubtedly augment the pattern further.
Although it is likely that some of these very worn coins
were in circulation in the early 3rd century, it would be
unwise to attribute all to this date. In fact, this pattern
implies occupation and economic activity throughout the
2nd century, perhaps congruent with the existence of the
earlier, unlocated fort suggested by many scholars.
Total
Date range
Emperor
Pre-AD41
AD41-54
54-68
69-96
96-117
117-138
138-161
161-180
180-193
193-222
222-238
238-260
260-275
275-296
296-317
317-330
330-348
348-364
364-378
378-388
388-402
Pre-Claudian & Iron Age
Claudian
Neronian
Flavian
Trajanic
Hadrianic
Antonine I
Antonine II
Antonine III
Severns to Elagabalus
Later Severan
Gordian III to Valerian
Gallienus sole reign to Aurelian
Tacitus to Allectus
The Tetrarchy
Constantinian I
Constantinian II
Constantinian III
Valentinianic
Theodosian
Theodosian II
The excavations exhibit relatively normal proportions
of coins from the Severan period through to 260. This
contrasts sharply with the pattern shown in the river
assemblage where there is a peak in Reece period 10 ( 193222). This peak in coin deposition again indicates
significant activity at Pierce bridge and reinforces
Brickstock's suggestion of an early 3rd-century date for
the foundation of the second fort. Richard Reece (pers.
comm.) has noted how the river assemblage has an
exceptionally high number of coins for period 11 (22238), well above the national average.
The dramatic peaks in coinage in Reece periods 13 and
14 (260-75 and 286-96) from the excavations are not
present in the river assemblage. In fact, Reece period 13
100
90
80
70
Ill
C:
·o
....0
(.)
0
z
60
50
40
30
_111
20
10
0
I
2
3
4
5
6
7
I. 11. __
8
9
10
11
12
13
14 15
16 17
......
18
Reece periods
Fig. 13.4 Histogram of' the Piercebridge river coin totals using Reece periods
288
13
0
3
22
22
32
59
32
9
96
45
25
34
13
3
3
0
2
0
0
2
-
19 20 21
13: RECENT WORK
800
700
.....,._ Piercebridge
600
-Cov.Well
500
........ Bath
(/)
~
I..
a,
c.
(/)
C:
0
u
400
300
200
100
0
-100
1 2 3 4 5 6 7 8
9 10 11 12 13 14 15 16 17 18 19 20 21
Reece Periods
Fig. 13.5 A comparison of the coins from the river at Piercebridge, Coventina 's Well and the
sacred spring at Bath plotted against the British mean (see text.for discussion)
marks the beginning of the end for coin deposition in the
river and finds from period 14 are restricted to period 13
radiate copies, which have been dated by Sam Moorhead
and Richard Reece (pers. comm.) to the period 275-85. It
is possible that this downward trend is overemphasised by
the worn nature of the radiate assemblage there are seven
radiate and 49 radiate copies amongst the unattributed
coins. However, diminishing coin loss in periods 13 and
14 is seen in other votive coin profiles (see below) and
may reflect the dwindling importance of the cult worshipped
at Piercebridge. Alternatively, as coin became more
common in everyday life, other objects viewed as more
valuable, may have begun to be offered in a votive capacity.
The paucity of coin finds from the river continues
throughout the late Roman period (Reece periods 15 to
21 : 2 96-402) perhaps suggesting diminishing votive
activity. It is however worth noting other possible
explanations. The latest coins recovered are silver siliquae
rather than the more common nummi and this may be a
reflection of the divers' collection method which inevitably
favours the recovery of large or visible coins. Work
comparing different recovery methods underwater would
seem to support this hypothesis. On the Liri river, one
season (Liri I) used divers to recover material from the
riverbed while a second season (Liri II) employed a suction
dredge. When the objects recovered from both seasons were
compared it was apparent that Liri I overemphasised the
importance of larger, easily recoverable coins (Metcalf
1074, 42).
The striking dissimilarities between the river and
excavation assemblages points towards a different process
of coin loss and hence function. Indeed, the river location
and the composition of the assemblage suggests a votive
site. When compared with the cumulative frequency coin
profiles of two well known votive deposits Coventina's
Well and the Sacred Spring, Bath - the similarities in coin
deposition are striking (FIG. 13.5) (Allason-Jones and
McKay 1985; Walker 1988). Although Coventina's Well
has its sharpest rise in coin loss in the 2nd century as
opposed to a peak at Piercebridge in the early 3rd century,
all three sites decline in a similar manner. Coin loss drops
off between periods 12 and 14, levelling off from period
14 to 16 before sharply diminishing to period 21. This
shared pattern confirms a votive function for the river
assemblage and has major implications for the study of
Romano-British cult practice at Piercebridge and in
northern Britain.
More study will increase our understanding of the
material. However, even at this stage, there are two features
amongst the coin assemblage which may illuminate
something of the nature of the cult worshipped at
Piercebridge. First, 14% of the coins in the assemblage are
of imperial women. This seems an unusually high
proportion and could indicate a deliberate selection of coin
for deposition, perhaps illuminating a feminine element to
the cult being worshipped or the sex of its devotees. By
comparing the percentage proportion of empress to emperor
coin from Piercebridge with mean percentage values for
each emperor, we can go some way to establishing whether
there is anything significant happening (Duncan-Jones
2006, 224). At present mean percentage values exist for
Hadrian, Antoninus Pius, Marcus Aurelius and Commodus
and further work is necessary to compile those for emperors
of the early 3rd century. When these values are compared
with the Piercebridge river coin data it becomes evident
that the volume of empress coin is unusually high only in
the reign of Marcus Aurelius (TABLE 13.4). More analysis
is necessary to investigate what this may imply.
Second, there are 15 fragments of denarii amongst the
assemblage. Seven of these fragments cannot be attributed
to a particular emperor, whilst the others are from Reece
periods 5, 7, 10 and 11. It appears that a proportion have
been deliberately cut, although it is possible that some have
experienced post-depositional damage. One from period 5
approaches being an exact cut half coin; the others are
closer to being thirds. Halved Roman bronzes are known
from several Roman sites (Kraay 1964; Strack 1904) and
have been interpreted as ad hoe small change, or tokens
between friends or in commercial contracts (Buttrey 1972).
However, incidences of cut denarii in any proportion are
289
....
ROMAN PIERCEBRIDGE
TABLE 13.4: FEMALE DEDICATE COIN AS A PERCENTAGE OF ALL COIN
Emperor
% Silver Mean
% Bronze Mean
% Silver
% Bronze
Hadrian
A. Pius
M. Aurelius
Commodus
Septimius Severus
Elagabalus
Severus Alexander
14
39
34
12
7
30
42
15
5
32
No information
No information
No information
No information
No information
No information
10
19
80
0
0
0
33
elusive. In Britain, the only other example is also from a
riverine context, the Thames at London Bridge, where two
broken denarii and nine bent bronzes were recovered
amongst the large assemblage of coins from the site (Rhodes
1991, 184). Richard Reece has suggested (pers. comm.)
that cut denarii offerings may have been used to represent
fractions of sestertii when there were none available in
circulation. An alternative hypothesis is that this defacement
is a foreign ritual practice introduced by soldiers garrisoned
at the fort, 'killing' the coins so that they are put beyond
use before deposition.
SMALL FINDS
A total of 526 objects have been catalogued although at
present a further 100 remain unidentified. It is evident that
a full range of functional categories is represented with
objects of personal adornment, military equipment and
religious material being particularly well represented.
In common with the coin assemblage, the majority of
small finds retrieved date to the Roman period and, where
more specifically datable to the 2nd and 3rd centuries.
However, there are also some late Iron Age and early
medieval finds amongst the assemblage. The earliest
objects include a lst-century cosmetic grinder and mirror
handle, whilst later finds include three square-headed
brooch fragments and a late Saxon strap end. Their presence
in the river may be the result of river bank erosion or casual
loss rather than deliberate deposition and requires further
investigation.
Objects of personal adornment or dress
Items of personal adornment form the biggest functional
category of finds recovered from the river and represent a
larger assemblage than recovered from the excavations.
At present, 72 brooches have been catalogued and although
the river assemblage possesses a similar date range and
many of the same typologies seen in the excavations, there
are some interesting differences.
First and 2nd-century brooch types are far more
prominently represented and account for 43% of the brooch
assemblage (e.g. FIG. 13.6, nos 1-2). Notable are six
repousse plate brooches, all bearing 'Celtic' triskele motifs,
dating to the mid 1st century (Bayley and Butcher 2004,
173). As a type, they are completely absent from the
excavation assemblage. There are also a higher proportion
of Fowler (1960) type 'A' penannulars amongst the river
finds. This indicates a different process of deposition at
the river and may suggest a much stronger native element
to the activity in the early Roman period.
290
77
0
26
38
32
Late 2nd to 3rd-century brooches account for 53% of
the assemblage (FIG. 13.6, nos 3-5). A variety of types are
represented including the divided bow and tutulus disc.
However the knee brooch is by far the most prolific with
20 examples recorded. This is unusual as the knee brooch
is not common in Britain, more frequently occurring along
the Rhine and Danube. At Catterick, where a similar
proportion of knee brooches were recovered, it was seen
as being indicative of an 'extraordinary influx ... [of]
people'. (Mackreth in Wilson 2002b, 154). Perhaps the
predominance of the knee brooch is a reflection of the
arrival of troops with links to the German limes.
Many of the brooches are undamaged with the pin and
spring mechanism intact (FIG. 13.6, nos 1 and 5), hinting at
deliberative votive deposition as opposed to casual loss or
discard. In this context, the presence of two types of
zoomorphic brooches is significant. Horse and rider
brooches, of which there are two, are often found in temple
and votive deposits and it has been suggested that they
have a link with Epona, a deity of Celtic origin adopted
throughout the empire by Roman cavalry troops (Johns
1993). Ifzoomorphic brooches can be interpreted as having
religious resonances, then it seems appropriate that two
enamelled fish brooches were amongst the assemblage.
Twenty-four finger-rings have been catalogued with a
late 1st to 4th-century date range. Unlike the excavation
assemblage, where early intaglio rings were not common,
there are six amongst the river finds in iron and silver. All
possess carnelian or imitation glass intaglios and their
engravings all indicate a high level of craftsmanship.
Notable amongst the other finger-rings are two silver
examples dating to the 2nd or 3rd century with inscriptions,
reading )IMI and DMART (to the god Mars) respectively,
and a gold ring with garnet setting and applied granules on
the shoulders representing bunches of grapes.
Precious metal objects are not confined to the category
of finger-rings. Other gold jewellery include three ear-ring
fragments of types popular in the late 2nd or early 3rd
century, as well as 18 gold necklace components including
clasps and chain lengths, the longest possessing 15 links.
Finds of Roman gold jewellery in Britain are relatively
rare and the existence of several gold items in the river
assemblage is exceptional in a votive context. It is indicative
of the status, wealth and significance ofboth the cult being
worshipped at Piercebridge and its adherents.
Ten bracelets have been catalogued. With the exception
of a single tinned fragment which may be part of a 1st to
2nd-century armlet, all are plain, two or three strand armlets
dating to the late 3rd or 4th century (Crummy 1983, 38).
Two armlets are very small in size (less than 50mm in
diameter) and could therefore be interpreted as being the
possessions of children or perhaps deliberate miniatures
13: RECENT WORK
1
5
---====---c===---•mm
0
50
4
llllJ
Enamel
Fig. 13. 6 A selection of the brooches recovered from the river at Piercebridge. Scale 1: 1. lllustrations by Mark Hoyle
intended for votive deposition. There are, perhaps
surprisingly, no strip bracelets, a typology seen in large
numbers amongst the excavation finds.
Fifty pins have been recovered, one in silver, 18 in bone
and the remainder in copper alloy. Although on stylistic
grounds they appear to possess a broad date range, with
Crummy type I and Cool types 2 and 3 particularly well
represented, the mean length of complete pins (106mm)
suggests that 1st to 2nd-century forms may predominate.
It is also notable that a large proportion of the pins do not
fit comfortably within the established typologies (Cool
1991) - though these were based on southern examples suggesting particularly local or regional manufacture. For
example 13 of the pins possess head loops or shank
perforations. Several of the shank perforations are
decorated with fine wire links and decorated with blue glass
beads, ruling out the possibility that these perforated pins
are some type of needle variant.
Military equipment
The substantial number of finds of a military nature attests
to the presence of the army at Piercebridge. With the
exception of six mid lst-century button and loop fasteners,
which may have been used by the military, the material is
overwhelmingly 2nd to 3rd-century in date. This accords
well with the date range of material from the excavations
although the definite 4th-century material seen there is
absent.
There are a large number of fittings from military
uniform and armour. These include 37 apron pendants of
varying styles, eleven openwork belt mounts, eleven
scabbard slides, two sword chapes and two phalerae. There
are also more than 30 fragments of scale armour although
at present no lorica segmentata fittings have been
catalogued. Amongst these finds, the number of apron
pendants is particularly striking, contrasting markedly with
the total of three recovered from the excavations, and
indicating a different process of deposition. Whether these
pendants are the offerings of individual soldiers or the result
of casual loss (retrieval being more difficult when lost from
a bridge into a watery context) is open to interpretation.
However, given the nature of the overall assemblage it is
tempting to suggest the former.
Unfortunately, these fittings provide little indication of
whether the troops garrisoned at Piercebridge were
legionaries or auxiliaries. However, the small assemblage
of lead sealings recovered goes some way to providing
some likely candidates. There are eight lead sealings
stamped LVI (Legio VI Victrix). Whilst the sealings do not
necessarily confirm the presence of the sixth Legion at
Piercebridge, they do compliment the epigraphic evidence
presented by the building stone built into Gainford Church
nearby (RIB I, no. 1025).
Three further sealings read OVA which can be read as
Ala Vocontiorum retrograde. The Ala Vocontiorum were a
cavalry unit, originally from Lower Germany and are known
from an inscription at Newstead (RIB I, no. 2121) and lead
sealings from Leicester (RIB II, 1, no. 2411.90) and South
Shields (Allason-Jones and Miket 1984, 328, 8.27). Again,
the sealings cannot confirm the presence of the unit, but it
might account for the quantities of 3rd-century horse
harness and equipment recovered from the river. Whilst
there are only four classic mid-Roman equine pendants,
291
ROMAN PIERCEBRIDGE
-GD
3
2
t
..
.
.
'
1
0
5
50
---====---===--•mm
6
Fig. 13. 7 A selection of the copper alloy (nos 1-3, 5-6) and lead alloy (no. 4) finds recoveredfrom the river at
Piercebridge. Scale 1:1. Illustrations by Mark Hoyle
there are seven strap slides and 23 stud mounts. These are
all of a size likely to be associated with horse harness rather
than personal strap fittings. The mounts are dominated by
circular or rectangular openwork examples bearing intricate
'Celtic' trumpet motifs.
Unusually, archery equipment is also represented with
two types of iron arrowhead present (trilobate and bodkinheaded) in addition to an incomplete bone ear-lath with
nock similar to those from Caerleon, all likely to be 3rdcentury ih date (Bishop and Coulston 2006, 166, fig. 105,
nos. 1 and 4).
In addition to the usual range of military finds, there are
several that have a more ceremonial function. Two complete
iron 'standard points', dating to the 3rd or4th century have
been catalogued. The 'standard points' are identical to
examples excavated at Vindolanda where they were
interpreted as possible symbols of military might for use
in parades (Jackson 1985, 132, no. 6). If this interpretation
is correct, then their presence amongst the river material
hints at an official, military element to the religious activity.
A single spearhead with a worn and damaged perforation
through the blade was also recovered. Perforated spears,
usually in miniature form, are known from temple contexts
including Great Walsingham (Bagnall Smith 1999, 34) and
Uley (Henig 1993).
292
Objects associated with religious beliefs and practices
Where the finds from the excavations contribute little to
our understanding of religion at Piercebridge, those from
the river are far more instructive. Many of the finds already
described have been noted as possessing votive elements
including finger-rings with dedicatory inscriptions and
horse and rider brooches. There are also a significant
number directly associated with religious beliefs and
practices. No single deity dominates the assemblage
although it is notable that deities particularly popular with
the military, such as Jupiter, Mars and Mercury are
represented.
These include three copper-alloy figurines. Two depict
winged Cupids in similar dancing or flying poses. They
are very similar to examples known from Kirby Thore,
Cumbria, and Corbridge, Northumberland, and their high
quality of craftsmanship suggests a Mediterranean
provenance (Green 1978, pls 31-2). A further figurine is
in the form of a ram, the cult emblem of the god Mercury.
Mercury may also be the deity represented by a
fragmentary pipe-clay figurine. The fragment comprises
the upper torso and head of a partially naked figure. The
figure possesses a full head of hair and is wearing an
unusual circular hat, characteristic of those worn by
13: RECENT WORK
Mercury. Representations of Mercury in pipe clay are rare
(Green 1978, 11 ). Interestingly, its interior is covered in
fingerprints indicating that it has been hand pressed into a
mould rather than poured.
A small silver plaque probably depicting Jupiter
abducting Ganymede has also been catalogued. Ganymede
is shown wearing a Phrygian cap with his legs crossed,
perhaps to indicate flight. To his right is Jupiter in the form
of an eagle with outstretched wings. This scene is frequently
depicted in classical art and the plaque is almost identical
in execution to a mosaic from Bignor Roman villa, Sussex.
However, the miniaturised aspect of the plaque is
unparalleled and it is uncertain what its original function
may have been.
A lead openwork plaque depicting a draped but naked
male figure holding a sceptre has also been recorded (FIG.
13.6, no. 4). His head and right arm are missing and he has
been defaced deliberately prior to deposition with the legs
twisted together. Due to the fragmentary nature of the
figure, positive identification is difficult although it is
possible that it represents a deified emperor. Lead figurines
are unusual in a Romano-British context with parallels
instead clustered in central and eastern Europe, particularly
Hungary. However, a similar lead
accompanied by
a portable lead shrine was found during the 1978
excavations of a barrack block at Wallsend (Allason-Jones
1984 citing Hungarian references).
Further objects with votive significance include a single
miniature socketed axe decorated with rows of crescental
cells filled with degraded enamel. Miniature objects are
known from many Romano-British temple sites and it is
perhaps surprising that only one has been recorded here.
fragments of rolled sheet lead have also been
recorded. Although it is possible that a number are crude
fishing weights, a proportion may represent 'curse tablets'
similar to those known from Bath and Uley, as evidence
suggests a preference for their deposition in watery
locations. If this is the case, they have major implications
for our understanding of the site and it is the intention that
they should be unrolled, transcribed and read as soon as
possible.
CONCLUSIONS AND FURTHER WORK
It is evid13nt that the material from the River Tees at
Piercebridge represents a rich and important votive deposit,
on a par with those from the sacred spring at Bath and
Coventina's Well. Indeed, the assemblage has tremendous
potential to augment our understanding of Romano-British
religion and votive practice, in addition to providing a
window on both military and civilian activity at
Piercebridge throughout the Roman period.
Further work will concentrate on cataloguing and
analysis of the remainder of the assemblage. In addition,
archaeological investigation is recommended to ascertain
whether any context remains for the material. Although
much may have been destroyed by the removal of the
objects, the divers have reported the existence of various
stone and timber structures on the riverbed. These may
represent platforms on which objects were .collected or from
which they may have been cast into the water.
293