238 THE ROMANS. 



the places round to Longovicus and Derventio would be in 

 geographical sequence. 



Concangium, by the title of its company and its position be- 

 fore three known stations on the great road, may claim place 

 somewhere on the Tees, as at Pierse Bridge (Coniscliffe) or Greta 

 Bridge. 



For the situations of Maglovse, Magae, and Longovicus, we 

 have only conjecture ; nor is it clear what Derventio here signi- 

 fies. It seems not likely that a company of foot should be per- 

 manently established at and named from the station supposed 

 to be situated only seven miles from Eburacum ; the camp near 

 Derby seems to have a better claim ; but Pap Castle, or Cocker- 

 mouth on the Derwent of Cumberland, should not be forgotten. 

 Longovicus, Magse and Maglovse may be in the vicinity of the 

 Lakes, to which Galava, Alone, and Galacum of the Itinerary 

 seem also referable. 



The Notitia gives further the distribution of troops along the 

 line of the Hadrian Wall (' per lineam valli'), but on this subject 

 the reader may be referred to the recent volume of Mr. Bruce. 



The last of the Documents of Roman date which has been 

 referred to, l Anonymi Ravennatis Britannise Chorographia/ will 

 not detain us long. This geographical compilation of the names 

 of places, &c. in the first, second, third and fourth parts of Bri- 

 tain, though but a tasteless performance, is rather less confused 

 than is commonly thought. If a little pains be taken to unravel 

 it, the descriptions are seen to arrange themselves in the order 

 of the British Tribes, beginning with Cornwall and the Dam- 

 nonii, passing on to the Durotriges and Belgae ; Silures, Dobuni, 

 Atrebatii, Regni, Cantii ; Cornavii, Ordovices, Coritani, Catyeu- 

 chlani, Trinobantes, Iceni; and again by the Coritani to the 

 Brigantes. It then proceeds beyond the Hadrian Vallum, and 

 afterwards beyond the Wall of Agricola. The names appear often 

 to have been taken from a Greek copy ; and the terminations are 

 in different cases, as if from different authors. The spelling is 

 often thus known to be erroneous. 



