Hills from Balthayock to Invergowrie, to the southward of which was the influx of the Earn.* Immediately to the northward of the Roman station, there has been a lake, the site of which is still distinguished by the name of Loch-ee; and the stream flowing from Loch-ee, after being joined by the burns of Fowlis and Balruddery, falls into the Firth of Tay, about half a mile to the westward of Invergowrie House. The mouth of this stream would form a harbour; and the stream may have anciently been called Gowrie. General Roy had not heard of Cater Milley, and finding the situation of Dundee to be such as the Romans would not have chosen for a camp, the Saxon or Danish name of Burgh-Tay, the Fort of Tay, seems to have engaged his attention, and induced him to suppose the station ad Tavum to be Burgh-Tay Castle; which, by some oversight, he states to be 18 English miles from Orea; whereas the true distance is fully 24, equal to 26 Roman miles, being 7 more than the Itinerary distance: Errare humanum est. It is rather a singular coincidence, that Cater Milley is just the same distance north-west from Dundee, that Orea is from Perth.

The next station is Æsica, distant from Tavum XXIII Roman miles, equal to 21 English; and Tina follows Esica at the distance of vIII Roman miles, equal to 7 and 1 furlong English. There can be no doubt of Esica having been a station on the South Esk, and

* The Gaelic inver is of the same import with the British aber, and denotes the confluence of rivers, or streams; the influx of a small stream into a greater, or the fall of a stream into the sea, or a lake; the mouth of a river; and, in a few instances, inver has been transferred to the river itself; metaphorically, a port or harbour; and Invergowrie may mean, the port of Gowrie.

C

Tina another on the North Esk: the Gaelic Esk, and the British Tine, being of the same import, and signifying a river, or water; which last is a name given, in Scotland, to streams inferior to rivers in magnitude. Montrose, situated at the mouth of the South Esk, has been supposed to be the former station, by General Roy; and he places the latter at the Bridge of Morphy, near the embouchure of the North Esk. From Cater Milley, going across the country to Montrose, it is 28 English miles, and even from Burgh-Tay Castle, 24; from Montrose to the Bridge of Morphy, it is 34; and these distances cannot be reconciled to those given us by Richard. Dr Stukely and Chalmers agree in placing Esica at Brechin, and Tina at King's Ford, on the North Esk, below the confluence of West Water. From Cater Milley to Brechin, by Caerbuddo and Kirkden, that being the most obvious route, the distance is 24 English miles; and from Brechin to King's Ford, 4. If we regard the distances in the Iter, we cannot admit Brechin to be Esica; and we are not aware of any remains having been discovered at this town that could indicate a Roman station. By following a route from Cater Milley, not much different from the last, but going by Claverhouse, Pourie, Gagie, the Hare Cairn, passing between Lower and the camp at Haerfaulds, then by Rescobie and Aberlemno, 21 English miles bring us to the South Esk, where a peninsula, formed by the influx of the Noran, is such a situation as the Romans often chose for their castra stativa; and we would suggest that Esica perhaps stood here. It commands a ford on the South Esk; and on the isthmus at the west end of the peninsula, there is a house called Ward-end, which seems to imply that this place had been occupied for

military purposes. From South Esk, at the influx of the Noran, to the confluence of West Water and North Esk, at King's Ford, the distance is 71⁄2 miles and 1 furlong English, corresponding exactly with the Iter. We agree with Chalmers in thinking Tina was situated at this ford.

The next station northward is Devana, distant from Tina XXIII Roman miles, equal to 21 English. It is clear that Devana was situated on the Dee; it is not so easy, however, to determine the particular spot where the Romans had their station. General Roy is of opinion that it was at Old Aberdeen; but, from the Bridge of Morphy to that town, it is not less than 33 English miles, and from King's Ford, 36; distances that seem to put Aberdeen out of the question. Chalmers fixes upon the neighbourhood of Peter Culter, where there is a large Roman camp, that commands the fords of the Dee. From King's Ford to this camp, going by the How of the Mearns, and turning northward near Stonehaven, by the camp at Urie, called Rae Dykes, the distance to Peter Culter is 28 English miles; going from King's Ford by Fordun, where there is both a temporary and a permanent camp, and crossing the hills to Peter Culter, the distance is 26 English miles. The former route is preferable; but even this is 5 miles more than the Iter allows. From King's Ford, on the North Esk, going by Fettercairn, Clattering Brig, where there is a Roman post, Cairnie Mount, and Cuttie's Market, 21 English miles bring us to Kincardine O'Neil upon the Dee. The first portion of this name implies a fortress; and that the Romans crossed the Cairnie Mount is probable, from their having established a post at Clattering Brig.