without auxiliaries. This was most likely the 9th, which, we are told, was the weakest, " ut maximè invalidam." The situation was such as Vegetius says was preferred by the Romans, being a gentle eminence, near the conflux of the water of Ruchil and the Earn. The Prætorian gate seems to have been on the east side, and the Decumen gate toward Ruchil water; so that the front of the camp looked down Strathearn. Of the two principal gates, that on the north side is not exactly opposite to the gate on the south side, being somewhat nearer to the rear, owing perhaps to the camp being pitched in a hurry. As formerly mentioned, the gates are protected by semicircular claviculæ, one overlapping the other, so as to render the defences particularly strong. Just within the principal gate on the south, there is a large upright stone, and beside it two smaller ones, which have fallen down: hence the name. Dealg, in Gaelic, signifies a needle, or skewer, and, figuratively, an obelisk: Dealgin Ross, the monument on the peninsula. Besides these stones, tumuli, cairns, and circles of various kinds, are scattered about the plain.

On the north side of this camp, there are the remains of another work, which might perhaps contain the auxiliaries of the legion, or a detachment sent to join it. Great part of this work has been washed away by the Ruchil; but the ramparts that remain are considerably stronger than those of a temporary camp usually are, and the intrenchments are not parallel to those of the little camp. There are vestiges of a still stronger rampart within the outer one; and there are two ways leading from it, one south, or south-west, toward Ardoch, and the other toward the east, leading

to Strageth. The last mentioned road cuts obliquely the intrenchments of the camp before it on the south, which shews it to have been a subsequent work; and, as we find no such roads leading out of temporary camps, this seems to have been a station which, we conclude, was Victoria. The Romans appear to have had time to execute these roads only to a short distance.

As the works at Dealgin Ross would not hold more than 10,000 men, they could not be occupied by the army of Agricola, consisting of about 26,000, before the great battle with Galgacus: but it seems very probable, that, on his return from that expedition, in honour of the victory recently obtained, he founded Victoria, on the spot where the 9th legion had been attacked the previous year, and where he had also gained a signal victory. Dealgin Ross is also called Galdachan, which signifies the town of the strangers.

The natives, who were probably hovering on the neighbouring hills, while Agricola occupied the great camp at Ardoch, seem, upon the advance of a division of his army to Dealgin Ross, to have retreated, in a north-west direction, toward Glen-Artney, allowing the Romans to fortify their camp without molestation. Another division advanced to Strageth, while Agricola remained with the third in the small camp at Ardoch. In the mean time, the Caledonians called in their detachments; and, issuing from Glen-Artney in the night, fell in a body on the camp of the 9th legion at Dealgin Ross, having learned that this was the weakest division of the Roman army. They killed the guards, and, having stormed the intrenchments, the battle began in the very camp. Roused from sleep, and in confusion, from the darkness; surprised by an attack so

unexpected, and in terror from not knowing the extent of their danger, the Romans fought under the greatest disadvantage, and nothing but the courage and discipline of veterans could have saved the legion from destruction. From the monumental stones, already mentioned, being within the intrenchments, we conclude, that a person of great distinction among the Caledonians, and two inferior chiefs, had fallen in this night attack, after storming the south principal gate of the Roman camp. From this gate being forced, it would appear that the assault had been made on that side, and that the Caledonians descended to the attack from Glen-Artney, interposing themselves between the 9th legion and the division at Ardoch under Agricola. Their rear-guard seems to have been posted at Blair-in-roan, 2 miles to the southward of Dealgin Ross; and it is probable they would send a detachment to the neighbourhood of Strowan, to intercept the communication with Strageth.

Agricola, having learned that the Caledonians had united their forces, in order to overwhelm the 9th legion, advanced, by a forced march, to its relief. We suppose that, leaving his small camp at Ardoch, he ascended the right bank of the Knaik, and, turning the hill Mealcorry-our, fell upon the rear-guard of the Caledonians at Blair-in-roan, from which place the shouts of his soldiers might be heard in the camp at Dealgin Ross. Part of the forces of the Caledonians would hasten to sustain their rear-guard, and in the vicinity of Blair-inroan the principal action with Agricola seems to have been fought. Near this village there are six or seven upright monumental stones, called by the country people, to this day, Roman stones. The names of places hereabout refer to an engagement. Blair-in-roan,

the spotted battle-field; Toi-na-blair, the stream of the battle-field, being one of the sources of the Machany, near the above village; and in Glen-Artney, three miles to the westward, we find Alt-na-chask, the burn of action. The victory of the Romans does not appear to have been so complete as Tacitus insinuates. The 9th legion had been roughly handled in the camp; and, when the Caledonians were defeated by Agricola at Blair-in-roan, they seem to have rallied in Glen-Artney, and, after making a stand at Alt-na-chask, to have retreated among the fens and forests in the neighbourhood. They had shewn considerable address in spreading reports, that induced a commander, so skilful as Agricola, to commit the error of dividing his forces,—an error which was like to have cost him dear; though foiled in their attempt upon the 9th legion, and obliged to retire, they did not consider themselves conquered, as we may infer from both armies being said by Tacitus to have left the field mutually exasperated,-"irritatis utrinque animis discessum."

The next station is Hierna, or Strageth upon the Earn, or Erne, near to Inverpeffery. It consists of two parts; and the original work, which is square and of small dimensions, seems to have been that next the river. To this the Romans had made, or intended to make, an addition, or procestrium, westward, and divided again into two parts. From the abrupt manner in which these last works are broken off, there is reason to conclude, that they had never been fully executed. The Roman road from Ardoch crosses the Earn at a short distance above the Station, and a branch leads into it. The situation is excellent, and there are strong reasons for supposing, that, in the vicinity, a division of

the army of Agricola pitched a temporary camp, the intrenchments of which have been ploughed down. They existed in 1757; for Maitland says, - "The Roman road intersects the camp at Strageth," which must refer to the intrenchments of a temporary camp not now to be seen. Hist. Scot. vol. i. p. 196.

Following the Roman way to the eastward of Strageth, and distant 5 English miles, we come to a square redoubt, adjoining to the south side of the road, and very near to Gask House: 2 miles farther, and on the opposite side of the road, we come to a similar redoubt, but of considerably less dimensions than the former: between them, they might contain a cohort, or about 600 men. For several miles in this quarter, the Roman way is still used as a common road; it needs but little repair, and is commonly dry in the wettest seasons. A small circular British fort lies 1 mile to the north-west of Gask House.

A. D. 84.

Agricola, in his last campaign, is supposed, on leaving Ardoch and Strageth, to have crossed the Tay by Derder's Ford, at the present mouth of the Almond, and to have encamped with his whole army at Grassy Walls, half a mile to the northward, on a gentle eminence in every respect suitable for a camp. To the westward it was protected by the Tay, which here makes a fine sweep, and it has the advantage of being watered by two or three rivulets. This camp is, in mean breadth 1950 feet, corresponding exactly with that of the great camp at Ardoch; and, though the length cannot now be ascertained, there is reason to suppose it might be about 2800 feet, and consequently would hold 26,000 men. Its sides are also a little deflected, like the Ardoch camp, yielding to the ground, particularly at Donald's