68 
Stanwix, the striking similarity between the three lists strongly 
favouring such conclusion. 
If this allocation be correct, the garrison was the 1s¢ Cohort of 
the Spaniards. Now an equestrian cohort would be undoubtedly 
placed at such an important station, at any rate late in the empire. 
To go back a little in the history of this cohort; about the year 120, 
as we have shown, it was at Maryport, but removed a short time 
afterwards, apparently into North Britain, at Ardoch, in advance of 
the Antonine Wall. In the ordinary course of events it would then 
return, and doubtless take up its position at Stanwix. The position 
of this station seems to necessitate the presence of a cavalry cohort. 
True, the station was only a small one—too small, in fact, for so 
large a cohort; but, when we consider that it would probably have 
to assist in the defence of Luguvallum, as well as the advanced post 
at Netherby—the Castra Exploratorum of the Notitia—the difficulty 
at once vanishes ; at least one authority, Mc. Lauchlan, agrees with 
this allocation. 
That Bowness—or at any rate Skinburness—is Tunnocellum, 
is strongly supported by eminent antiquarians, including Lysons, 
Ferguson, Longstaffe, Horsley, and Wright. According to the 
Notitia, the garrison of ‘Tunnocellum, the 18th station in the Iter. 
“per Lineam Valli,” consisted of the rst Cohort Elia Classica, ie, a 
Cohort of Marines. Hence, we may very reasonably suppose that 
Tunnocellum was a town on the sea coast, and, following the order 
of sequence, on the West Coast. Although altars, etc., have been 
found here, still there are none that testify to the presence of the 
Elia Classica, but, as I have attempted to show, they are not needed. 
But etymology—without, however, placing too much trust in it— 
comes to our aid. The Rev. W. Lytteil, writing to me on the 
subject, says, ‘‘I regard Tunnocellum as derived from the Gaelic 
dun @ chaoil, signifying “the fort of the firth—or, the fortress of the 
strait.” By adopting this derivation the difficulty vanishes. How 
very aptly it suits the position of the camp at Bowness, the map of 
Cumberland bears abundant testimony; of the two places I should 
certainly prefer Bowness as the site of the ancient Tunnocellum, 
The claims of the Maryport camp to be G/anoventa (Glannibanta), 
