‘SEXING’ SMALL FINDS
Lindsay Allason-Jones

John Clayton in 1880, when discussing why Coventina’s Well was full of
artefacts, put forward the hypothesis that ‘lovesick damsels cast into the
Well their spare trinkets in the hope of obtaining the countenance of the
Goddess in their views’. ‘To these interesting ladies we are doubtless indebted for the brooches, rings, and beads found in the Well’ (Clayton
1880: 31). In 1979 Charles Daniels wrote of the buildings, referred to as
‘chalets’, discovered during his excavations at Housesteads: ‘both Mr
Gillam and the writer were struck by the preponderance of brooches and
other trinkets at Housesteads XIII. To the writer this suggested that the
chalets had been married quarters of some sort’ (Daniels 1980: 189). Both
archaeologists, separated by one hundred years of improved archaeological
techniques, were basing their hypotheses on the premise that there were
certain small finds which could be identified as having been used or worn
by men and others only by women. With the recent development of interest in the topic of gender in archaeology many researchers are being
tempted to base their theories of space allocation/role/status, etc., on the
evidence of small finds. This paper looks at some groups of finds from
Romano-British contexts, traditionally regarded as ‘female’, in order to see
if there really are classes of artefact which might be assigned to gender –
might be termed ‘male’ or ‘female’.
Many archaeologists have continued to follow Clayton and Daniels’ view
that brooches are indicative of a female presence despite the overwhelming
evidence for men wearing brooches throughout the period of Roman
Britain. Almost every military and civilian tombstone where the deceased is SEXING SMALL FINDS 23
shown wearing a cloak has a brooch clearly visible fastening the garment
(see for example: Ribchester: Shotter 1973a: p1.16; Housesteads: Coulston
and Phillips 1988: nos 202-203; Vindolanda: ibid. no. 212). Philpott, in his
extremely useful survey, Burial Practices in Roman Britain, has pointed out
that ‘sexual determinations of cemeteries have shown that males are often
provided with a brooch at a number of sites, while the evidence of sexed
late Iron Age cremations from King Harry Lane indicates that brooches
were buried equally often with males and females …. The provision of
brooches with males in cremations is mirrored in the mid-late 1st century
in Dorset and Wessex, where brooches were largely confined to males and
children in native inhumations’ (1991: 123).
Having clarified the point that both men and women wore brooches is it
possible to discover whether certain types of brooches were worn only by
women while other types were confined to male costume? Theoretically
one should be able to use the evidence from the cemeteries with this precision. Unfortunately, few cemetery reports are detailed enough. Even when
the sex of the skeleton is given in association with a list of grave goods
many of the older reports record merely ‘a fibula’ with no indication as to
type.
One of the few cemetery reports which might be used to attribute gender
to a brooch type is the Lankhills report (Clarke 1979) as only the male
graves contained brooches, all of which were the 3rd – 4th century massive
crossbow type, which have occasionally been described as insignia of rank
for military or civilian officials. On the continent, however, such brooches
have been found in female graves and it is possible that the gold example
from The Wincle, Cheshire, found with necklaces and ear-rings, may be
from a female grave (Johns, Thompson and Wagstaffe 1981).
Is it possible to make a judgement on the reasons why brooches were
worn? Brooches would have been an essential feature of some RomanoBritish women’s wardrobe as a few ensembles were literally pinned
together. The costume now named after a woman called Menimane
from Mainz-Weisenau (Esperandieu 1922: no. 5815) is the most obvious
example. This was worn in the northern provinces in the pre-Roman Iron
Age and the Roman period up to the early 2nd century AD, reappearing
briefly in the 5th century. Menimane’s costume included a closely fitting
bodice fastened at the front by a brooch; over this she wore a loose tubular
tunic, pinned at the shoulders by a pair of brooches with a third at the
breast. Women wearing a similar outfit would have needed at least three
brooches and many preferred more (see Wild 1968: fig. 21). 24 L. ALLASON:lONES
Women from the area of the Danube wore overtunics caught by a pair of
brooches which were linked across the chest by a chain (Wild 1968: 207);
so is it possible to attribute matching pairs of brooches to women, particularly those chained together? As men would have only needed a single
brooch to fasten a cloak one might conclude that pairs of brooches linked
by a thread or a chain were worn exclusively by women, although the evidence is not irrefutable. If the hypothesis does hold good then can one presume that any brooch which has been seen as one of a pair can be claimed
as a female type? Unfortunately this is unlikely. Headstud brooches,
trumpet brooches, disc brooches, and innumerable other types are found
with chains or with the loops for attaching chains. The headloop is a
feature which is rarely found on the Continent, despite the evidence for
linked brooches on the Danubian tombstones, but it is a common element
in the 1st and 2nd century British brooches and is ubiquitous on the
trumpets, headstuds and other bow brooches which some specialists in the
past have described as ‘military’.
It is also possible to overemphasise the need for women to wear
brooches to fasten their clothing. Not all women in Britain wore Menimane’s costume or Danubian dress, the majority apparently preferring the
Gallic coat worn by men, women and children throughout the Roman
period. As this was a T-shaped roomy garment with a slit neck it required
no brooches to hold it in position (Wild 1968: fig. 1). Wild has stated that
clothing which needed more than one brooch to fasten it was rare in the
north and that which needed a linked pair of brooches was confined to
women who followed Continental styles of dressing.
As brooches continued to be manufactured and worn when Menimane’s
costume was out of fashion it must be presumed that some people wore
brooches purely for decorative rather than functional purposes. The small
triskele brooches and enamelled animal brooches are too small to fasten
bulky woollen cloth and could be described as decorative only, but other
types could be used for either purpose. Clearly more work is required in
this area but the results so far seem to indicate that brooches were sexless;
that they were bought by men and women alike, according to personal
taste. If a man wanted to wear a small openwork triskele he did so, and if a
woman wanted to wear a large crossbow brooch there was little to stop her.
Loops were provided in case the purchaser needed to wear the brooches in
a pair or attach a safety chain.
If it is difficult to determine whether a particular type of brooch was
worn by men or women is it any easier with other types of jewellery? SEXlNG SMALL FINDS 25
Among the inhabitants of the city of Rome the wearing of ear-rings was
purely for decorative purposes and strictly confined to women. To their
historians and geographers the idea of men wearing ear-rings was simply
evidence of the barbarity of foreigners and worthy of outraged comment.
Isodorus (Orig. XIX.31.10) mentions that it was the fashion for Greek
youths to wear a single ear-ring in one ear. Pliny (Nat. Hist. XI.50) was horrified that ‘in the East, indeed, it is considered becoming even for men to
wear gold in that place’ (i.e. in the ear lobe). Other authors felt it was
worthwhile to mention foreign men wearing ear-rings: Xenophon (Anab.
III.1.31) refers to the Lydians; Juvenal (1.104) to the Babylonians; Macrobinus (Sat. VI1.3) to the Libyans; and Plautus (Poen. V.2.21) to the Carthaginians.
If one looks at the coinage of the eastern kings, such as Phraates and
Bahram, however, it is noticeable that they all wear ear-rings, from Phraates
in 3 BC to Yazdgard in AD 457 (Toynbee 1978). In fact, throughout the
period of the Roman empire the male rulers of the east and their subjects
wore ear-rings and may have regarded them as a sign of rank.
The Roman army included many men from the eastern provinces and
Africa where it was not considered improper or unusual for men to wear
ear-rings. Despite the classical authors’ disapproval of the practice it is possible that some of these troops continued to wear ear-rings during their
military service in Britain. Unfortunately, there are no literary references
which state categorically whether a serving auxiliary was or was not allowed
to wear ear-rings. No ear-rings have been found in an indisputable relationship with a male skeleton in a Romano-British context, although bearing in
mind the earlier remarks about brooches in cemeteries this is hardly surprising. Lankhills cemetery, which might have been of assistance, produced
no ear-rings at all, even in the female graves (Clarke 1979), while the northern military cemeteries, such as Petty Knowes (Charlton and Day 1984),
have produced very few metal artefacts other than coffin nails, hobnails and
coins for Charon’s fee. Ear-rings are not depicted on military tombstones
but neither do they appear on the tombstones of women – the only exception being Regina from South Shields, who is depicted in the Palmyrene
tradition and invariably is the exception to every rule (CSIR 1.1. no. 147).
On the other hand, a large number of ear-rings come from forts and fortresses. Unfortunately, few come from firm contexts and many may have
come from the vici or other areas where the presence of women would not
have been unusual. A few, however, come from what might be seen as
good military contexts. For example, two penannular copper alloy ear-rings 26 L. ALLASON:JONES
come from Longthorpe (Frere and St. Joseph 1974: 62 fig. 32 no. 7S; Allason:Jones 19S9a: nos .. 317 and 31S). This fortress was occupied from c. AD
44/ 8 to c. AD 62, an early period for women to be present in any numbers.
One ear-ring was found in the praetentura whilst the other was found in
building X, which Frere and St. Joseph suggested might be an auxiliary
praetorium. This last example was found in association with an armlet and a
nail cleaner and might suggest the presence of an officer’s wife or daughter
– the Vindolanda writing tablets have shown that some officers were accompanied by their families at a very early stage despite official disapproval
(Bowman and Thomas 1986: 122; Tacitus Annals 111.33). An armlet and a
nail cleaner are not very convincing evidence for the presence of women
either, as will be discussed later.
The majority of forts which have produced ear-rings are in the area of
Hadrian’s Wall, although none of the milecastles or turrets of that frontier
have produced examples (Allason:Jones 1988). All three units raised in
areas known to have favoured ear-rings, which are attested in Britain, were
stationed in the north: the Hamian archers from Syria at Carvoran and Bar
Hill, the numerus Maurorum Aurelianorum at Burgh-by-Sands, and the Tigris
Bargemen from Mesopotamia at South Shields and possibly Lancaster
(Breeze and Dobson 1978; but see Shotter 1973b for the latter). Of these
only South Shields has produced any ear-rings (Allason:Jones 1989a: nos.
480-88) but equally only South Shields could be regarded as having been
extensively excavated. The evidence for eastern or Mrican troops in Britain
wearing ear-rings is, therefore, slight but should not be disregarded. The
literary sources are silent on whether the men of Gaul or the Germanies or
even Iron Age Britain wore ear-rings but it would be rash to take this as
firm evidence that it was not done. It is only possible to say that in Britain
some foreign troops or merchants may have continued their native tradition of wearing ear-rings as they continued to prefer their native costume.
After all, if the Syrian archers did not give up wearing their flowing robes
when they were sent on foreign postings it is unlikely that they gave up
their ear-rings.
Solid neck-rings or torcs were worn as symbols of power and status in
pre-Roman Britain and as such had magico-religious significance. Dio
Cassius (LXII.2.4) tells us that Boudica wore ‘a great twisted gold necklace’
when she led the Iceni into battle, indicating that she had taken on the
authority of a Celtic warrior chieftain symbolically. During the Roman
period torcs were awarded to soldiers for acts of bravery but later came to
be regarded merely as good luck symbols (Maxfield 1974) and were fre-SEXING SMALL FINns 27
quently worn as such by women (e.g. Regina CSIR I.i. no. 247). Necklaces
made from beads seem to have been worn by women purely for decorative
purposes (see Volusia Faustina: Allasonjones 1989b: p1.23) but they were
also worn by children of both sexes to support amulets. Melon beads of
blue glass are usually found individually and one discovered attached to a
dolabra sheath in Bonn Museum indicates that they too had an amuletic significance and may have been worn around the necks of either sex on
leather thongs or copper alloy wire. In Rome itself the practice of men
wearing necklaces and bracelets was considered to be on a par with wearing
ear-rings: Diodorus Siculus (VA5) was disparaging about the Panchaeans
who wore ‘ornaments of gold, not only the women but the men as well,
with collars of twisted gold about their necks, bracelets on their wrists, and
rings hanging from their ears, after the manner of the Persians’. Clearly the
arguments rehearsed previously about men of eastern origin wearing earrings can be extended to necklaces and bracelets, both of which can be
seen on male Romano-Egyptian mummy portraits from the Fayum
(Bowman 1990: pI. 9). There is evidence for gold armlets being male accessories in the early Celtic world (Strabo IVA.5), and the massive armlets
found in Iron Age contexts may well be a reflection of the use of torcs as
symbols of the warrior class (see Anderson 1904). Unlike ear-rings bracelets
have been found in male graves, both worn: Whitcombe, Dorset (grave 6:
Whimster 1981: 271), Langton, N. Yorks. (Corder and Kirk 1932: 59, 66)
and Cirencester (grave 179: McWhirr et al. 1982: 129); and unworn: Oakley
Cottage, Cirencester (Reece 1962: 51).
Apart from being purely decorative, bracelets could secure amulets
around the wrist, as at Cirencester (Crummy 1983: no. 1610) – where
should the line be drawn between a bracelet worn for aesthetic reasons and
something worn in order to fasten an amulet into position?
Finger rings are very difficult to attribute to a male or female wearer unless they are found in situ on a finger. Size alone is not an adequate criterion: some women have large hands, some men very small hands, and the
wearing of rings on the second joint of the finger confuses the issue. It has
often been supposed that only men wore intaglio rings but this is not so.
Intaglios could be worn for decoration alone and women would also have
required them for business purposes such as sealing letters and documents.
Jewellery, therefore, is not as clear an indicator of gender as might be
thought.
A survey of the small finds from the turrets on Hadrian’s Wall revealed
other artefacts, which had traditionally been taken to be female, in male 28 L. AuAsON:JONFS
contexts (Allason:Jones 1988). The turrets were only occupied for a total of
forty years in two separate stages and only by the military. None so far excavated show any sign of squatter occupation and several were either
demolished or had their doorways blocked up when the army withdrew. It
might, therefore, be presumed with some safety that only men attached to
the army were present on these sites. The presence of needles, nail
cleaners and tweezers – all traditionally used as indicators of a female
presence – may come as a surprise. Soldiers in the 2nd century, no doubt,
had to mend their clothes, clean their nails, and remove splinters. Recent
excavations at Vindolanda and Carlisle show that Roman soldiers, like
British soldiers in the Second World War and in the modern army, were
issued with sewing kits – known nowadays as ‘housewives’.
Among the commonest finds on Roman sites, whether civilian or military, are pins, made from a variety of materials. These artefacts have been
the subject of a number of articles, both classifying the types and debating
their use. MacGregor (1976: 13) expressed doubts as to whether they could
all be identified as hairpins, offering an alternative suggestion that they
were used to fasten garments. Cool, in 1991, presented the evidence for
decorated pins having been solely for hairdressing, but the large number of
roughly fashioned, undecorated, bone pins found on sites remain ambiguous. Philpott (1991) in discussing the possibility of attributing function to
pins by their position in graves, emphasised the difficulties that ensue
when the exact position is not recorded but drew attention to groups of
pins found at the feet of inhumations. He pointed out that metal pins ‘are
rarely found close to the skull but are usually found lower down the body’.
He continued, ‘it may be significant that the deposition of metal pins which
were apparently functional in the grave occurs at a time when there is a decline in the use of brooches in graves for fastening garments. Metal pins
may have partially replaced brooches as shroud pins in the 3rd and
perhaps 4th century’ (1991: 151).
Much is known of hair fashions during the Roman period and while it is
clear that both long and short hairpins were used, indeed required, in
female hairstyles, there is no evidence that male hairstyles needed pinning.
The Italian fashion was for men to wear their hair short and while male
skeletons have been found with long hair in the Celtic provinces, for
example at Poundbury (Green, Paterson and Biek 1981), none appear to
have affected pins. We must, therefore, remain open minded as to whether
the discovery of pins indicates a female presence.
So are there any groups of artefacts which clearly indicate the presence SEXING SMALL FINDS 29
of women? Few seem to be good indicators – medical instruments with a
purely gynaecological purpose form a rare group. There is also the possibility that items made of jet had special significance for women.
Jet was first worked in Britain in the Bronze Age but, although the
Romans were aware of the properties of jet when they invaded Britain, the
manufacture of jet objects at such places as York does not appear to have
gathered momentum until the late 3rd or early 4th century. It was then
used to make jewellery – beads, betrothal pendants, bracelets and fine
finger rings, as well as hairpins, spindles, and spindlewhorls.
Jet is rarely found in a male grave in Britain and when it is, as at Oakley
Cottage, Cirencester (Reece 1962: 51), it is unworn. Jet artefacts regularly
appear in female graves at York and elsewhere in the country – mostly in
the eastern counties. Knife handles and a few fragments of furniture inlay
have also been found but these may have belonged to women. One clearly
male item is the scabbard chape from Bonn in Germany (Hagen 1937) but
this stands out as an exception, the rest all appearing to be biased towards
the female.
The reasons for this bias may be religious as most of the British finds
have been from graves or other religious contexts, or because jet, like
amber, had a particular significance for women. Pliny related that ‘the
kindling of jet drives off snakes and relieves suffocation of the uterus. Its
fumes detect attempts to simulate a disabling illness or a state of virginity’
(Nat. Rist. XXXVI: 141-42). There is, however, a noticeable lack of fertility
amulets made from jet.
So where does this leave Coventina’s Well and Housesteads? The
‘trinkets’ from the Well include ten brooches, fourteen finger rings, two
hairpins and five bracelets. These are the only objects which might be
regarded as female on traditional criteria other than a large number of
glass beads, all of which may be from the same necklace as they include
twenty-four gold-in-glass beads. Coven tina herself is unambiguously female
(AllasonJones and McKay 1985: pI. VI) although we have no clear evidence
as to her responsibilities and one must conclude that she was an ‘allrounder’, dealing with matters of healing among other human concerns.
All the inscriptions refer to male worshippers, but this may not be significant in itself as women very rarely dedicated stonework to a deity on their
own behalf – the ratio of female dedicators to male in Roman Britain being
about 1 in 10. Are the pieces of jewellery the female equivalent of an altar
dedicated by a man or would it be considered logical to present a female
deity with a feminine artefact – an item of jewellery might be seen to be an 30 L. ALLASON:JONES
appropriate offering with which to placate a female deity. There are no ex
votos which might indicate a female congregation, unlike the spring of Sulis
Minerva at Bath where breasts of ivory and bronze have been found (Cunliffe 1988: pI. 3). The Wheelers suggested that the discovery of hairpins and
bracelets at a temple indicated a shrine of healing which catered for women with gynaecological complaints (Wheeler and Wheeler 1932: 42), but
Coventina’s Well with only five bracelets and two hairpins compares badly
with the suggested temple of healing at Piercebridge, Co. Durham, where
over one hundred of each have been found (Scott forthcoming).
The artefacts which might be regarded as male are also unimpressive: a
strap-end, three studs, a seal-box, three belt buckles and seven bell-shaped
studs. Of these, only the buckles can be regarded as being exclusively male
with any confidence on the grounds that articles of female clothing of
Roman date involving buckles have not been discovered so far. Having said
that, women may have had buckled satchels, boxes or horse harness. The
finds from Coventina’s Well do not conclusively prove female devotees for
the cult, whether love-lorn or not.
As was mentioned earlier, the jewellery from Housesteads found in
barrack block XIII was considered proof of married quarters in the late 3rd
century to the early 4th century. It was said that there was ‘a preponderance of brooches and other trinkets’. The actual numbers of artefacts are
twelve brooches, only two of which are of late 3rd to 4th century date, four
finger rings, three bracelets, two ear-rings, and nine hairpins. This does not
seem excessive when compared to other sites, particularly as the area of
XIII covers the road outside as well as the interior of the building. The
same picture emerges at Wallsend, where chalets have also been found
(Daniels forthcoming). Wallsend has been almost fully uncovered and for
the whole site there are forty-two brooches of which four are late; there are
also ten bracelets, two ear-rings, six finger rings and thirty-six bone hairpins. Again, this does not imply a preponderance. Neither Housesteads nor
Wallsend has produced much jet. If these two factors are added to the difficulty of ascribing artefacts to gender, the argument of the chalets being
married quarters on finds evidence alone is considerably weakened.
This has been a brisk survey of only a small group of artefacts but it
should serve as a warning about the dangers of identifying objects and
their purpose from the limited viewpoint of modern values. The blinkered
view of the predominantly male, middle class, 19th century archaeologist is
still alive and well and even the most radical theorists can fall into the trap
of building their theory on the shifting sands of small find identification. SEXING SMALL FINDS
Bibliography
Allason:Jones, Lindsay 1988. Small Finds from the Turrets on Hadrian’s
Wall. In]. C. Coulston (ed.), Military Equipment and the Identity of Roman
Soldiers, 197-233. Oxford: British Archaeological Reports (IS 394).
Allason:Jones, Lindsay 1989a. Ear-rings in Roman Britain. Oxford: British
Archaeological Reports (201).
Allason:Jones, Lindsay 1989b. Women in Roman Britain. London: British
Museum Publications.
Allason:Jones, Lindsay and Bruce McKay, 1985. Coventina’s WeU: A Shrine on
Hadrian’s WaU. Chesters: Trustees of the Clayton Collection.
Anderson, J. 1904. A Note on a Late Keltic Armlet of Bronze, now
Presented to the National Museum. Proceedings of the Society of Antiquaries
of Scotland 38:460-466.
Bowman, Alan 1990. Egypt after the Pharaohs. London: Oxford University
Press.
Bowman, Alan and J. David Thomas, 1986. Vindolanda 1985: the New
Writing Tablets.Journal of Roman Studies 76: 120-123.
Breeze, David and Brian Dobson, 1978. Hadrian’s WaU. London: Penguin.
Charlton, D. Beryl and John Day, 1984. The Roman Cemetery at Petty
Knowes, Rochester, Northumberland. Archaeologia Aeliana (5th series)
12:1-31.
Clarke, Giles 1979. Pre-Roman and Roman Winchester. Oxford: Oxford University Press.
Clayton, John 1880. Description of Roman Remains Discovered near to Procolitia, a Station on the Wall of Hadrian Archaeologia Aeliana (2nd
series) 8:1-39.
Cool, Hilary E. M. 1991. Roman Metal Hairpins from Southern Britain.
ArchaeologicalJournal147 (1990):148-182.
Corder, Philip andJ. L. Kirk, 1932. A Roman Villa at Langton, near Malton, E.
Yorkshire. Leeds.
Coulston, Jon. C. and E. J. Phillips, 1988. Corpus Signorum Imperii Romani:
Great Britain, 1.6, Hadrian’s Wall West of the Norlh Tyne and Carlisle.
Oxford: British Academy.
Crummy, Nina 1983. Colchester Archaeological Reporl, 2, The Roman SmaU
Finds from Excavations in Colchester 1971-9. Colchester: Colchester
Archaeological Trust.
Cunliffe, Barry 1988. The Temple of Sulis Minerva at Bath, 2, The Finds from
the Sacred Spring. Oxford: Oxford Committee for Archaeology.
Daniels, Charles M. 1980. Excavation at Wallsend and the Fourth Century
Barracks on Hadrian’s Wall. In William S. Hanson and Lawrence J. F.
Keppie (eds), Roman Frontier Studies 1979, 173-194.
Daniels, Charles M. forthcoming. Excavation Report on the Roman Fort at
Wallsend.
Esperandieu, E. 1922. Receuil general des bas-reliefs, statues et bustes de la Gaul
Romaine. Paris.
31 32 L. ALLAsON:JONES
Frere, Sheppard S. and]. K. St.Joseph, 1974. The Roman Fortress at Longthorpe. Britannia 5:1-129.
Green, C. S., M. Paterson and Leo Biek, 1981. A Roman Coffin-Burial from
the Crown Buildings, Dorchester, with Particular Reference to the Head
of Well Preserved Hair. Proceedings of the Dorset Natural History and
Archaeology Society 103:67-100.
Hagen, Wilhelmine 1937. Kaiserzeitliche Gagatarbeiten aus den Rheinisches
Germania. Bonner Jahrbilcher 142:77-144.
Johns, Catherine, F. Hugh Thompson and P. Wags taffe, 1981. The Wincle,
Cheshire, Hoard of Roman Gold Jewellery. Antiquaries Journal 60:48-58.
Maxfield, Valerie A. 1974. The Benwell Tore – Roman or Native? Archaeologia Aeliana (5th series) 2:41-48.
McGregor, A. 1976. Finds from a Roman Sewer System and an Adjacent Building
in Church Street, York. York: York Archaeological Trust.
McWhirr, Alan, Linda Viner and Calvin Wells, 1982. Romaru;.British
Cemeteries at Cirencester: Ci7encester Excavations II. Cirencester: Cirencester Excavation Committee.
Philpott, Robert 1991. Burial Practices in Roman Britain. Oxford: British
Archaeological Reports (219).
Reece, Richard 1962. The Oakley Cottage Romano-British Cemetery, Cirencester. Transactions of the Bristol and Gloucester Archaeol. Society 81 :51-72.
Scott, Peter R. forthcoming. Excavation Report on the Roman Fort at
Piercebridge.
Shotter, David C. A. 1973a. Romans in Lancashire. Clapham: Dalesman
Books.
Shotter, David C. A. 1973b. Numeri Barcariorum: a Note on RIB 601.
Britannia 4:206-209.
Toynbee, Jocelyn M. C. 1978. Roman Historical Portraits. London: Thames
and Hudson.
Wheeler, R. E. Mortimer and Tessa V. Wheeler, 1932. Report on the Excavations oj the Prehistoric Roman and Post Roman Site in Lydney Park,
Gloucestershire. Oxford: Society of Antiquaries of London.
Whimster, Rowan 1981. Burial Practices in Iron Age Britain: a Discussion and
Gazetteer of the Evidence, c. 700 BC – AD 43. Oxford: British Archaeological Reports (90).
Wild,John Peter 1968. Clothing in the North-West Provinces of the Roman
Empire. Bonner Jahrbilcher 168: 166-240.