Researching Roman Dividers

For the last twelve months I was lucky enough to work as an Assistant Finds Liaison Officer in the Durham Portable Antiquities Scheme office. This was a part-time role supported by the Headley Trust for the first six months and I am grateful that it was extended for another 6 months for one day a week. Working on the Portable Antiquities Scheme has been a wonderful experience. I have enjoyed talking to finders (mostly metal-detectorists) about their finds and the chance to research artefacts I was less familiar with. I used to be really intimidated at the prospect of identifying coins, but having access to training and the right resources allowed me to grow my confidence. Let’s just say, I may actually like coins!

Roman dividers DUR-83CC6C (© Durham County Council CC BY)

One of the objects that I came across while working through a finder’s collection was a pair of dividers (DUR-83CC6C). Dividers are a measuring device that allow the user to measure one thing and compare it to something else. They could be used by carpenters, blacksmiths and stone masons, but also on written documents or drawings (Manning 2011, 83). Dividers were even depicted on the tombstone of a carpenter P. Beitenos Hermes (shown in this blog) along with other tools. One of our volunteers had already written a good description of the dividers and I was checking over their work before making the record ‘live’ and publicly accessible. I was really happy with the descriptive work our volunteer had done and they had suggested that they dated to the Roman period. I was only planning to add an extra line or two about other comparable examples, but the more I read, the more I was intrigued. Apparently, Roman dividers are a rare object (Worrell 2005).

I am going to show in this post the process I went through to confirm the identification and write up the database entry. I hope that this will be useful to readers.

Confirm the material

Visual examination showed that the object was made from a copper alloy. The metal is in really good condition, but there were a couple of patches of powdery green corrosion. There was a spot on one of the arms where the dividers were recently scratched, which showed a lovely metallic orange-yellow. A little area of iron staining could be seen on the pivot area, but because it wasn’t substantial there was no reason to think that the surviving portion was made up of more than one material. I was happy that this object was indeed made from a copper alloy.

Confirm the object type and period

It may seem obvious, but it is worth confirming you’re happy with the object identification. When I work with artefacts that come off excavation sites they sometimes have initial identifications. Sometimes they’re right and sometimes they’re wrong (this is totally expected and okay!). In this case, I started by looking at records for all other dividers on the Portable Antiquities Scheme website by period. I then checked online museum collection databases (the British Museum and Museum of London online collections are excellent research tools). I also checked the finds sections of excavation reports that cover the Roman period to the post-medieval. I searched for both ‘dividers’, ‘calipers’ and ‘callipers’ in case sources used different terminology.

Some examples of other Roman dividers from the Portable Antiquities Scheme website. Left: NMS-DEB0C2 (© Norfolk County Council CC BY-SA), Centre: BERK-8A377C (© Oxfordshire County Council CC BY), Right: NARC-B85204 (© Northamptonshire County Council CC BY-SA)

I made a list of all dividers I found on the PAS database, museum collections databases and excavation reports I could find. I also compiled all the images I could find of the dividers. In this particular instance, I experimented with using OneNote for organising the information about each example and created a separate page for the images so that I could see them all together and move them around. After reading through the description of each example, I made  list of any citations mentioned in the descriptions or discussions and followed up on those until I couldn’t find any more information. If I had more time I would have followed up with a library visit to check published excavation reports for additional examples, but in this instance I had already compiled enough information for the purposes of a PAS database entry.

I mostly found images of Roman and post-medieval dividers, which were very distinctive. Most examples were made from copper-alloy, but there were a couple made from iron (British Museum: 1870,0402.459 from Colchester; Cunliffe 1971, no.186, fig. 53, Bushe-Fox 1913, LON-960EA3, and possibly Casey and Davies 1993, 189, no. 282, fig. 10.13). The copper-alloy examples seem to be made from several parts: the two arms, the spindle with decorated heads at either end, and an iron wedge that was inserted into the spindle hole that kept the arms locked.

DUR-83CC6C has faint decoration along the surviving arm. Other Roman examples were similarly decorated, but it seems that the actual style and motifs might be individually unique. Many other examples also had animal heads on one or both ends of the spindle, which DUR-83CC6C did not have. Instead, the head was dome shaped with a radial design. As the other head was missing, it is possible that it had one animal head or none at all.

Based on the similarity in form with other examples considered to be of Roman period date, I was happy to suggest that this one also probably dated to that period.


Once I was confident that the object was in fact a set of dividers and dated to the Roman period, I then wrote some discussion text to supplement the database record. This is placed after the general description of the dividers. I decided to focus the discussion on comparing DUR-83CC6C to the other examples. I would have liked to have used part of the discussion section to address the dividers’ date, as ‘Roman’ is quite vague. However, even with the excavated examples taken into account, there just wasn’t the evidence to be able to narrow this down more.

Illustration of DUR-83CC6C (© Durham county Council CC BY-NC-SA)


I decided to do an illustration of the dividers because the decoration on the arms is quite faint and difficult to see on the photographs. I have found learning to illustrate artefacts to be a useful process, as I notice details that I didn’t see during my initial examination. Taking a methodical approach to illustrated artefacts really forces you to look at every millimetre and break it down into simple lines. I also noticed that, curiously, the design on one edge didn’t match the design on the opposing edge, which I though was rather odd! I had assumed that the decoration on one edge would be mirrored on the other edge. I’m not sure why this would be the case and it is certainly something I’ll look out for in the future.

Final thoughts

I hope that this provides a useful overview of the process I went through to research an artefact that I was unfamiliar with. Going through the process reminded me that there is so much we don’t know about Roman artefacts and highlights that there is so much more that we can learn. It would be interesting to compare the examples found in Britain with those found elsewhere in Europe and especially with the depictions of tools on tombstones and paintings. I’m curious to know whether many examples were decorated and whether the decoration may indicate use.


Bushe-Fox, J.P. (1913) Excavation on the Site of the Roman Town at Wroxeter, Shropshire. Society of Antiquaries of London Research Committee Report.

Casey, P.J., Davies, J.L. (1993) Excavations at Segontium (CAERNARFON) Roman Fort, 1975-1979. CBA Research Report 90.

Cunliffe, B. (1971) Excavations at Fishbourne 1961-1969, vol. 2: the finds. Rep. Res. Comm. Soc. Antiqs London 27.

Manning, W.H. (2011) ‘Industry’ in Allason-Jones, L. Artefacts in Roman Britain. Their purpose and use.  Cambridge: Cambridge University Press, 68–88.

Worrell, S. (2005) ‘Finds reported under the Portable Antiquities Scheme’, Britannia 36, 447–472.

Image notes

In the composite photo above, the only alternation made to the original images was that they were cropped and set to a similar scale for comparability. Copyright attributions for the individual images are given above.