32

The existence of a line of fortifications between Colunio (Colne) and Mancunium (Manchester) on the lower range of hills which divides Lancashire from Yorkshire, has been known to antiquaries ever since Dr. Whitaker completed his History of Whalley in 1818. He supposed the line to extend from Colunio to Cambodunum, but this I conceive is a mistake; for neither the locality nor the Ordnance maps give indications of any fortifications farther east than those on Twist Hill; whilst on the other hand they represent the line as continuous in the direction of Manchester, from Colne, by Littleborough and Rochdale. Several of these intrenchments were unknown to the historian of Whalley; but a recent examination of the principal portion of these interesting remains, (October 25th, 1856,) in company with my friend Benjamin Chaffer, Esq., has enabled me to enter more fully into their peculiarities and situation. For the sake of clearness I shall take them in order from northeast to southwest, in accordance with the numbers on the accompanying sketch-map of the district.

1. CASTERCLIFF is a formidable looking fortification, at a short distance from Colne, in a portion of what was once the Forest of Mereclesden (Marsden), and is supposed to have been the castra astiva of Colunio. Many Roman coins have been found in the valley below, and it is situated on the Roman road between Colunio and Cambodunum, and at the junction of this with the vicinal way from Rigodunum (Ribchester) to Alicana (Ilkley). The intrenchments form a parallelogram measuring about 550 feet long by 520 feet broad; but the walls appear to have enclosed an area of about 380 feet in length by 340 in breadth. The camp has been protected on the southwest front by a deep gully and also by a double vallum and foss, which are still entire about the whole crest of the mound. We were informed that many hundreds of tons of stones have been carted away from the walls within the last thirty or forty years, all of which appeared to have been subjected to intense heat. Large quantities still remain half buried in the soil, many of them completely vitrified, and others presenting a singularly mottled appearance from having been only burnt half through. The burnt sandstone and lime form excellent manure, and at the time of our visit a luxuriant crop of corn and cabbages had just been gathered from the broad ditches of this Roman camp. A less elevated plateau of considerable extent bounds the northeastern slope, which is again protected

by a steep cliff sloping down to the Calder, near Waterside. This would afford a convenient space for the exercise of large bodies of troops, or for the protection of the cattle belonging to the garrison; and it has probably been used for such purposes by the successive masters of the fortification. Being almost inaccessible on all sides except the east, where they are skirted by the Roman road, these defences, when complete, must have constituted one of the strongholds of the North, since they overlook the whole of the Forest of Trawden, Emmot Moor, a great portion of Craven, with the valley of the Calder, and terminate the eastern limit of the ridge on which Saxifield is situated. Castercliff has evidently been the key of this portion of Lancashire in the hands of the Romans, and its importance would undoubtedly not be overlooked by the Saxons and the Danes *

2. Crossing Gib Clough, and passing along the Roman road, portions of which still remain entire, a distance of less than a mile brings us to SHELFIELD, on the summit of which the late R. T. Walton, Esq., erected a tall granite cross. This erection probably stands in the centre of a large circular encampment, of which some portions of the ditch are indicated by the undulations of the surface, and by a swampy part of the ground on the western slope.

3. RING STONES HILL, at a short distance, formerly contained a large circle of stones, erected either for purposes of burial, worship, or defence; but they were dug up a few years ago, and have since been used for repairing the roads.

4. After crossing Float Bridge Beck, and Catlow Brook, we next arrive at BROAD BANK, near Ridihalgh, on the summit of which are the remains of a circular intrenchment, measuring about 150 feet in diameter. There is no appearance of walls, but both the vallum and foss are perfect throughout the whole circumference. This station occupies a very commanding position, overlooking the whole of the Valley of Thursden, with the heights of Boulsworth in the distance, but it does not appear to have been of a permanent character.

5. Passing through Thursden Valley, to a corresponding crest on the opposite ridge called BONFIRE HILL, at the distance of about a mile, we find another circular intrenchment, 130 feet in diameter, and very similar

* In the map, C indicates a camp, B a beacon, T a tumulus, and F an entrenchment.

C

in character to the last. This encampment is surrounded by an earthwork rampart, which is still comparatively perfect on three of its sides, and easily traceable on the fourth. The rampart measures 700 feet in length by 450 in average breadth; thus constituting a most formidable post in an almost inaccessible position to an attacking force. At a short distance, on what is now called Delf Hill, there is a circular tumulus of loose stones, about forty feet in diameter, and two feet in average height. Both the station and the tumulus occur on heights which overlook the Valley of Swinden; the slopes of Briercliffe lie to the northwest, and Burnley is visible in the distance.

6. PIKELAW, on the same ridge, has much the appearance of a large tumulus; but, as its name indicates, it has long been used for the purposes of a beacon. There are two circular tumuli on Beadle Hill, within a short distance, but they have almost disappeared in course of cultivation. They have been composed as usual of loose stones mixed with earth, and have not measured more than thirty feet in diameter when perfect.

7. From this point the line diverges a little to the west, and on arriving at Twist Hill, which overlooks Holden Clough, Rogerham, and Swinden Valley, we find the remains of a square camp, which has long been known as TWIST CASTLE. How it came to have this designation is by no means clear. Dr. Whitaker supposes the name to have reference to the ancient forests which once covered the district, and to indicate the " boundary of oaks." Should this conjecture prove correct, the castle will probably have given name to the township of Extwistle, in which it is situated. Each side of the encampment measures about 150 feet in length, within the trenches; the remains of the foss and vallum are very distinct; the latter being elevated considerably above the surface of the moor; and spaces still remain for the Pretorian and Decuman gates. A smaller enclosure occurs at the southeast corner, measuring about sixty feet square; and at a short distance is a circular tumulus of loose stones, about 45 feet in diameter. A few years ago this was elevated five or six feet above the surface, but its height has been much diminished in consequence of the stones having been carted away to form the road to the new quarry near Monk Hall. No remains were found in the tumulus during the excavations, but this may perhaps be accounted for from the fact that as yet the top only has been removed. Twist Castle occupies a commanding position on the western