Roman Finds Group

Welcome Page

Join the RFG

Contact the Committee



New Finds

Londinium Exhibition


New Books


Related Links

New Finds

Two lead bull heads from Cambridgeshire
Button and loop fasteners in a Roman Province
Wooden Combs and niche markets
Pottery Counters
Help Needed

Finds Archive

Two lead bull heads from Cambridgeshire

Models of bulls as cult objects are well known throughout the Celtic world and bull symbolism is associated with many aspects of Celtic and Romano-Celtic religion.
The two bull heads illustrated here are of particular interest as they come from the same general locality. The first (Fig 1) is from the southern outskirts of Peterborough, the second (Fig 2) from Flag Fen. The two models differ in style, the first being realistically modelled whilst the second is more na´ve in execution.
A bronze model of a bull's head was found with the Willingham Fen hoard, and it may be significant that this is only some 20 miles to the south of Flag Fen and Peterborough.

Click to Enlarge (2470x1420)
Fig 1. A naturalistic three-dimensional bull's head with forward-thrusting curved horns and delicately defined ears (the left ear is broken). The nostrils and mouth are neatly modelled and the eyes are inset with blue paste. The top of the head is flat and projects at the back to form a shelf which is pierced vertically for suspension.

Click to Enlarge (3151x2179)
Fig 2. A flat-backed bull's head with outward-pointing horns and a mane of shaggy hair falling over the crown. The ears remain as stubs and the protruding ring-and-dot eyes stare boldly forward. Detailing of the mouth and nose is obscured by the raggedness of the model's edges. A fragmentary suspension loop springs from the centre of the head.

Barry Carter
Radnorshire LD7 1UG

Button and loop fasteners in a Roman Province

A step towards a regional typology?

This piece is very much a work-in-progress, my first look at a class of object that has fascinated me since I took up the post of 'Finds Liaison Officer' a little over three years ago. Looking rather like ancient duffle coat toggles, button-and-loop fasteners are aesthetically pleasing. They also come in fairly discrete types, classified by the shape of their heads and attachment loops. In the wake of the Portable Antiquities pilot scheme, these objects are more plentiful, or certainly more recorded, than ever before.
I have recorded 16 examples from Yorkshire so far. Using the typologies put forward by Wild (1970) and MacGregor (1976), building upon the earlier work of Gillam (1958), I have simply looked at what types have been found where in the region, as recorded through the 'Finding our Past' scheme.
The majority of button and loop fasteners in my survey have come from East and North Yorkshire, with 8 and 5 examples respectively. I have not seen any examples from West Yorkshire and only 2 have come from the South Riding.

Distribution of button-and-loop fasteners in Yorkshire.

The overwhelming majority have been found with the aid of a metal detector. Only one has been found by eye, a beach find from East Yorkshire. This reflects the fact that 96% of all portable antiquity finds recorded from Yorkshire have been recovered in this way.
As with Wild's and MacGregor's surveys, the predominant class of button-and-loop fasteners in my sample group so far, is the boss and petal type (Class III). Another similarity with earlier surveys has been the dominance of triangular shaped loops. All of the complete examples have triangular loops, but the loop is missing on 5 of the 16 examples recorded. Surprisingly though, and in contrast to other studies, this survey has found a very even distribution of finds through Wild's other classes. It is, of course, possible that this profile will change as more examples are recorded.

Button-and-loop fasteners from Yorkshire shown by Wild's Classes.

Two examples of the double boss and petal type (Wild's Class I) have been recorded, together with one T-shaped (Class IX), two square headed (Class V), and three simple dome shaped fasteners (Class IV).
There have been two examples of the ring shaped Class II, the type given a possible pre-Roman date by MacGregor (1958). It is certainly present during the Iron Age in what is now North Yorkshire, as evidenced by the Stanwick Hoard. I had, at the outset of this survey, expected to see rather more of this type.
A trend in the Yorkshire region, certainly over the last year, has been an increase in Iron Age and early Romano-British finds (coins and artefacts) being brought in by finders. Far more surprising has been the emergence of an unusual type. One double headed/ended boss and petal variant has been found in North Yorkshire. It begs the question, 'Is this really a new type or simply an idiosyncrasy?'

Click to Enlarge (481x379)

Double-headed button and loop fastener from North Yorkshire.

As with all finds research in its early stages, my initial survey has thrown up far more questions than answers, which is why I would like to appeal to RFG members. Do you have any provenanced examples of button and loop fasteners from Yorkshire?
It is possible that Yorkshire produces similar types and distributions of these wonderful objects to other areas, but as yet, we just do not know! The portable antiquities data certainly gives us a wealth of new data, ideally suited to looking at finds on a regional scale.
Ceinwen Paynton
Senior Finds Liaison Officer
Portable Antiquities Scheme ('Finding our Past')

Gillam, J P 1958 'Roman and Native AD 122-97' in Roman and Native in North Britain, ed. I A Richmond, 60-90
MacGregor, M, 1976 Early Celtic Art in North Britain 1, 129-34
Wild, J P, 1970, 'Button and Loop Fasteners in the Roman provinces', in Britannia I, 137-56

Wooden combs and niche markets

The majority of boxwood combs of Roman date from Britain show a rather poor level of workmanship which supports the hypothesis that they were mainly utilitarian artefacts, frequently crafted by non-specialists. Moreover, experimental work has confirmed that serviceable wooden combs can be produced with a minimum of training. The result may not be too impressive but it will do the work.
This short contribution considers those comparatively few instances of combs of outstanding quality: those that might have been intended for use, but were certainly meant to be seen. This aim was achieved in a variety of ways: some quite obvious, other more subtle. There were a number of strategies.
The first one must be to patronise an established combmaker (a pectinarius), perhaps one "Dignus" or "Marcellinus Lugrac..." whose impressed maker's mark can still be read on a couple of artefacts respectively from London and Carlisle. Both combs are plain, but the cutting of the teeth which was certainly done in Roman times by eye, is very regular. Experimental work has shown that while cutting a few good fine teeth is not too difficult, sustaining that from end to end of a comb must require a trained hand, and, what is more, mistakes cannot be remedied. No wonder in medieval times combmakers' training lasted years!
Fancy designs like a comb in the shape of a lyre (Fig 1) and one with curly terminals (Fig 2) were also put on the market. In the latter the design of the terminals matched the grip of the box in which it was housed, yet another refinement.

Click to Enlarge (1912x1768)

Fig 1. Lyre-shaped comb from Praeneste in Italy - now at the Villa Giulia Museum in Rome.

Another approach was to do something with the central area. This is the band between the two rows of teeth which is sometimes decorated with cordons and grooves. Experimentally such ornamentation proved extremely easy to make. Boxwood, and practically all wooden combs in antiquity were in this wood, is extremely homogeneous and has a natural excellent finish. Cutting a groove or two with a small saw, or raising a cordon proved to be very simple operations. On the other hand alternative forms of ornamentation like fine marquetry, or a dedicatory inscription in openwork (Fig 2) certainly betray a very expert hand, very possibly of a worker in ivory as well as in wood.

Click to Enlarge (3825x2170)

Fig 2. Comb with openwork inscription from the Crimea.

A different approach was to use two craftsmen instead of one, a combmaker and a metalworker for instance. There is only one known example (from Carlisle with an early 2nd century date) and what is most interesting, the artefact in question was undoubtedly meant for show: with teeth barely 7mm, long it could hardly be of any use to anyone. It must have looked quite a sight on the lady's dressing table, though. The ends have a fretted design, on one side it is all chip carving, while on the other three bronze plates embossed with classical scenes, take up practically all the space.
The last on my list, the one that may have been the ultimate statement of what the well-to-do could afford, is distinctly underwhelming in the simplicity of its design (Fig 3). In this particular case there are no fancy shapes, no decorative tours de force. Cyparenis' comb may be plain, though quite elegant, but it was certainly not within everyone's reach because of the size of the raw material required. As stated above Roman wooden combs were made in boxwood and this one was no exception. There are 12 instances of this design all from Britain and they are all in that material.

Click to Enlarge (1840x1288)

Fig 3. The stele of the hairdresser Cyparenis.

The only complete one is from London: it is 190 mm long with a maximum width/length ratio of approximately 1:3, like the comb on the stele. Since combs are always manufactured with the grain in the direction of the teeth, this specimen must have required a piece of boxwood with a width in the excess of 200 mm. Finding such a billet cannot have been easy.
For a start, boxwood in antiquity was not as wide spread as it is nowadays, since it has been propagated for ornamental purposes. Its main growing areas were the Pyrenees, the western slopes of the Alps, Corsica and the southern shores of the Black Sea. Secondly its growth rate is painfully slow: the slower the growth, the better the quality of the wood.
The analysis of a batch of boxwood logs from the Comacchio wreck (dated to the end of the 1st century BC) gives an idea of what sort of tree had to be felled to make Cyparenis' comb. The logs are the right size, between 150 to 220 mm in diameter. The ring count gives an average growth of 276 years with a maximum of 513, for a 170 mm diameter. The logs originated all from the same stand, which had apparently been managed to produce the best material, ie straight, with no knots. It is hardly surprising that wood of that quality was the object of long distance trade, in this case, from the Pyrenees to Italy. It is the people in whose circles a professional hairdresser like Cyparienis moved who could afford it.
Paola Pugsley,
St George's Well,
Devon EX15 1AR

Castelleti, L, Maspero, A, Motella, S, Rottoli, M, 1990 'Analisi silotomiche e tecnica di lavorazione del legno' in F Berti 1990, Fortuna Maris, la nave di Comacchio. 136-153
Kuniholm, P I, Griggs, C B, Tarter, S L, Kuniholm, H L, 1992 'Comacchio (Ferrara). A 513-Year Buxus dendrochronology for a Roman ship', in Bollettino di archeologia July-December 1992. Istituto poligrafico dello Stato, Roma, 291-299
Lloyd-Morgan, G, forthcoming, 'Comb with inlaid bronze plaque' in I Caruana et al forthcoming, The Roman forts at Carlisle: excavations at Annetwell St, 1973-84
Vaulina, M, & Wasovicz, A, 1982 Bois grecs et romains de l'Hermitage. Warsaw

Toy storey (sic)

Various uses have been proposed in the past for the ubiquitous pottery counters, from game counters to pot lids. However, a remarkable set of eleven counters from a site near the Mercury Theatre in Colchester provide a new explanation for these items. Stacked together in order of decreasing diameter, they form a neat tower, just like a set of stacking bricks suitable for a young child (CA 11, 34). All are made from sherds of large grey ware vessels, with the edges ground smooth and the curved parts of the faces ground away. Another seven counters were found with the set, but do not fit into the tower.
Intriguingly, another group of counters from Colchester, found in the early 1990s, was discovered in a stack with the smallest at the bottom and the largest at the top (CAR 6, 166, fig 5.24, 683-7. Most of these counters were also made of grey ware, but included one made from a BB1 sherd.
Nina Crummy

CA The Colchester Archaeologist
CAR 6 Excavations at Culver Street, the Gilberd School, and other sites in Colchester 1971-85, Colchester Archaeological Report 6 by P Crummy

Help Needed

Excavation this season, carried out by the English Heritage Centre for Archaeology at the sites of several milecastles along Hadrian's Wall, has yielded this object. At first thought to be a fragment of a stone gaming board, subsequent research has shown up no comparanda. Lindsay Allason-Jones has suggested that it may be a table-top fragment.
It is a roughly rectangular fragment of iron-rich micaceous sandstone. One end of the upper face bears two incised cross-hatched squares, highlighted in a dark red pigment. Wear suggests that the bottom edge is an original surface; the fragment appears to be broken at the other edges. There is some evidence of burning. Max length 105 mm, max width 94 mm, depth 15 mm.

Click to Enlarge (857x979)

Illustration by Chris Evans, CFA Graphics Studio.

If anyone has any ideas as to the function of this object, or any similar objects that they could point me in the direction of, all information will be very gratefully received, and of course acknowledged.
Nicola Hembrey
EH Centre for Archaeology
Fort Cumberland, Portsmouth PO4 9LD
+44 02392 856716