MELANDRA CASTLE, DERIiYSHlKE. II 



Mr. Watson, it is well known, was a trustworthy authority, 

 being careful not to make any statement unless he had convincing 

 evidence of its truthfulness, therefore, his letter is most important 

 to us, as it gives a better idea of the station than we can form 

 from its present condition. " The plough had not then defaced 

 it, and its form could not be mistaken." 



The form of the station was a parellelogram, with the corners 

 rounded off; its extent, 122 yards by 112 yards; and the walls 

 were twelve feet thick, their height, of course, not now being 

 ascertainable, but sufificient, we may depend, to resist any sudden 

 onslaught. 



There were four entrances, and, most probably, an inscribed 

 stone over each, as one was found by the farmer who farmed 

 the land, a few years previous to Mr. Watson's visit. The 

 farmer had this stone built over the doorway of his house, 

 where it remained until 1842, when the farmhouse was pulled 

 down and re-erected, the stone being again built in the wall, 

 over a doorway, exposed to the weather, and there it remains 

 to this day. By the way, the site of the station seems to 

 have been a veritable quarry to the farmers in the vicinity 

 for dressed stone, for most of the farm-houses, barns, out- 

 buildings, and fence-walls in the neighbourhood contain con- 

 siderable quantities of stone obtained from this Roman station, 

 the nature of the stone and the peculiar manner in which 

 it was dressed — by a kind of pick — making them easily dis- 

 tinguishable. 



I have brought with me a plaster cast of the inscribed stone, 

 also a photo of same. If you compare them with Watson's 

 engravings in Vol. III. of Archcelogia, you will perceive a 

 slight difference — for instance, there is no dot in the centre of the 

 O ; Watson has a dot between L • V : it should be on the other 

 side of the V ; and the conjunction of the tt are separated. I am 

 not in a position to say it alters the translation, but it may be 

 a point worth noting ; however, Watson's translation has never 

 been questioned, and it is thus : " Cohortis, Primae, Frisian- 

 orum, Centurio, Valerius Vitalis " — that is, " Valerius Vitalis, 



