It is not to be expected that Roman place names are to be found in Moray At the time of the Romans it is believed that the people here were p-Celtic speaking: therefore, only the few p-Celtic place names still extant can be considered as possibly contemporary with the Romans.

A most significant p-Celtic place name element is Cathair meaning a fort. According to Watson, Cathair Mhaothail or Muthil Fort was the old name for the important Roman fort at Ardoch, while Watson also compares Catter with Cadder at Kirkintilloch, the site of a Roman fort. Watson also mentions Stracathro, but as he did not know of the existence of a Roman fort, which was only discovered after his death, he postulated that the Cathair must have been the White Caterthun.

In Moray, near Forres, there is the place name Sanquhar. This is not an imported name. There is charter evidence from 1341 of the forest of Sanquhare. Watson says that Sanquhar comes from Sean-cathair, Old fort.  Johnston gives the same derivation. The farm of Sanquhar Mains (NJ 043567) is about 35 m above sea level on a well drained gravel plateau, bounded on one side by the Mosset burn.

Mrs. Lesley Joughin has drawn my attention to the field name, Ker or Curmore (NJ l95584). This site is 65m above sea level with views all round and a particularly good prospect across the Laigh of Moray to the sea. The soil is very sandy and naturally well drained. Curmore might possibly be derived from Cathair Mohr or great fort. The field of Curmore is on the farm of Wester Manbeen and at least one of the other field names dates back to the Middle Ages.

If the two Cathair names have come down from the Roman period then they could possibly indicate the sites of Roman forts. There are marching camps at Keith and Fochabers, but a fort would add a whole new dimension on our knowledge of the Roman presence in the North.

About 500 AD the q-Celtic speaking Scots came across from Ulster and eventually spread their Gaelic language all over Scotland. As the Scots arrived long after the Romans had left they had no first hand knowledge of Roman forts, but they may have seen the remains of such forts and correctly conjectured what they were. Rothiemay comes from the Gaelic Rath a'mhaigh, fort in the plain. Ancient Britons were not in the habit of building forts on plains; particularly when there were convenient hills around. But the Romans had a predilection for siting their forts on river gravels, so there is a possibility that there was once a Roman fort in the vicinity of Rothiemay. However, the Normans, like the Romans, were quite prepared to build in the plain and as there is a castle at Rothiemay the stronger possibility is that the fort referred to was once a Norman motte.

The trenches of Roman camps could last for hundreds of years and Scottish speaking farmers referred to them as dykes. Therefore, the presence of place names, beginning or ending with dyke, may, in some instances, give an indication of possible Roman sites. In Scotland, Dyke names sometimes betray Roman sites.

There is the Antonine fort at Castiedykes near Lanark. Then there is Cleaven Dyke, Battledykes near Oathlaw, Battle Dykes at Reithock and Raedykes and Normandykes near Aberdeen - all proven Roman sites.

In and around Moray there are a dozen Dyke names:

• Dykeside, NJ 023523 is sited on gravel, some seven kilometres south of Forres. The ground here is undulating and unlikely to be the location of a Roman camp.
• Dykeside NJ 118600. This is on a gently sloping plain with good view to the West, East and North across the Moray Firth. A possible site.
• Parks of Dykeside NJ 113582. In Monaughty forest at an elevation of 160m. An unlikely site.
• Dyke NH 990585. A small village on the West side of the Findhorn. Flat ground, gravel with the Muckle Burn as a suitable source of water.
• Dykeside NJ 209589. On sandy soil near old Birnie Kirk. There is a local tradition of a Roman camp about l km south of Dykeside.
• Dykeside NJ 306563. On the edge of a broad Speyside gravel terrace. A glass bead, circa 200 AD, was found here some years ago.
• Holladyke NJ 469523. This is a farm some 500m from one of the sides of the Roman marching camp at Auchinhove.
• Hollodyke NJ 471458. This is 250m from the A96 road near Coachford and 6 km. from Auchinhove. If Agricola was marching with three independent columns then one of the columns may have camped near this site.
• Nether Dykehead NJ 578622. Nether Dykehead, Dykehead and Strypeside are all within 500m of each other, On flat heavy soil and within 2 km. of the summit of Durn Hill where there is a hill top "fort". ~ The Dykes are most probably drainage ditches.
• Green Dykes NJ 588557. On mainly clay soil but with patches of gravel. Some 1250m south of Park of Cornhill where the Colonels Gordon—Duff played, as boys, on the "Roman Ramparts". Green Dykes is 2 km. from the croft of Battlefield, but this latter name probably dates from the Napoleonic or Crimean wars.
• Dykehead NJ 521468 and Dykehead NJ 537462. On either side of the Deveron near the confluence of the rivers Isla and Deveron.

REFERENCES:-

1. Johnston, J.B. Place Names of Scotland London 1934
2. Macdonald, J. Place Names in Strathbogie Aberdeen l89l
3. Maxwell, H. Scottish Land Names Edinburgh 1926
4. Nicolaisen, W.F.H. Scottish Place Names London 1976
5. Tacititi, C. De Vita Agricolae Oxford 1967
6. Watson W.J., History of the Celtic Place Names of Scotland Edinburgh 1926