In examining this and other Roman camps, the spectator will, perhaps, be struck with the narrowness of the streets, and the very small capacity of the the dwellings. It is well to recollect that in their encampments the Romans studiously avoided occupying a larger space than was absolutely necessary. Gibbon observes that a modern army would present to the enemy a front three times as extended as a Roman one of the same force. In the field, ten men were apportioned to a tent, ten feet square;[105] a similar proportion would without doubt be followed in the stationary camp. [Sidenote: BORCOVICUS.] It is not easy to ascertain the precise character of the dwellings of the soldiers; the more perfect of the ruins in this and other forts, induce the belief that they were dark, bare, and cheerless. The roofs were probably formed of free-stone slate. Several thin slabs of this kind, with nail-holes in them, as well as some of the nails themselves, have been found in this and other stations.[106] On Plate XIII, figs. 1, 7, are drawn some door or window heads, found here; these most likely belonged to buildings of a superior class. The entrance into a chamber at HABITANCUM, recently excavated, was found to be only fourteen inches wide; it was rudely ‘stepped over’ at the top. Fragments of a sort of window glass are frequently found in some of the stations; this would probably be a rare luxury.[107] [Sidenote: THE CHAMBERS OF THE STATION.] At Housesteads, two or three of the ruined chambers will, above the rest, attract the attention of the visitor. Near the centre of the northern division is one which is seventy feet long and eight broad; it must have been a place of public concourse. In the front of it is a kiln which has probably been used for drying corn; near the southern gateway is another which was nearly destroyed in the endeavour made to extricate a cow which had fallen into it, and, in struggling to relieve herself, had thrust her head and neck into the flue. The Romans seem to have kiln-dried their corn at the close of the harvest; it would not have been safe to stack it in the open fields. They would the more readily do this, as it is still by no means unusual, in the central and southern parts of Europe, to thrash the corn at the close of harvest on the field where it grew. Three hypocausts have been found here, two within the station, and another to the east of it, on the Knag-burn; the flues of the latter were full of soot; very slight traces of any of them now remain.