seeing that, putting aside the name Volantium, all the names 

 applied to it are in this list. In fact, we might almost infer, from 

 its commanding position, its connections, its antiquities, etc., that 

 such would be the case. Now it is clear that all the names cannot 

 be applied to the station, if indeed any of them can. 



Let us then take the first name on our list, viz., AXelodunum, 

 and enquire (i) on what grounds this name has been given, and 

 (2) whether such reasons are satisfactory. 



The Notitia says that the First Cohort of the Spaniards was 

 stationed at Axelodunum. At Maryport, the spade of the antiquary 

 has unearthed some sixteen altars by these Spaniards. Hence the 

 inference has been drawn that therefore Maryport is Axelodunum; 

 and, on the face of it, this inference seems clearly the correct one' 

 But, we might ask whether we should be absolutely certain that 

 the mere finding of altars, even if taken in connection with the 

 Notitia, is proof positive. For my own part, I cannot see that 

 this fact taken per se can satisfactorily establish a claim ; for if it 

 could do so, then other places might have an equal claim. As it 

 is, the First Cohort of the Spaniards has left its record at Netherby, 

 at Ardoch,* and in North Britain,f besides Maryport. 



Clearly the Scotch stations are out of the running ; but the 

 claims of Netherby might be more tangible. 



Besides, if the number of altars is to be the main element, then 

 Maryport has the prior claim only so long as it maintains its 

 position in the front in this respect. In the same way any other 

 place near the Wall might, under like conditions, step into the 

 front rank : for who can say that all the Spanish altars have been 

 found ? Yet again, how is the matter to be decided when altars 

 by two of the cohorts named in the Notitia have been found in a 

 station, as has been the case at Moresby ? Clearly there would in 

 this case be a difficulty. Surely there must be a more satisfactory 

 method of allocation than this. A very simple argument will show 

 how very unreliable this method really is. Suppose a cohort to be 

 stationed at Maryport, and to signify its presence there by erecting 



* Celt, Roman, Saxon, p. 384. t Ibid p. 335. 



