170 Proceedings of the Royal Irish Academy. 
northern barbarian nations. The principal roads which invaders de- 
siring to reach any part of Italy from these islands should pursue 
were as clearly defined in the fifth century as they are now;’* for the 
passes through which alone roads could at any time be carried are 
limited in number and unchangeable in position. If, therefore, the 
nearest point at which sub-Alpine Roman territory could be reached 
was, as has been suggested, in the district of the Upper Rhine, there 
would be a reasonable presumption that the route by Coire and the 
passes of the Spleugen would be that entered on by Dathi in this expedi- 
tion. And this indeed is the route which early British and insular trayel- 
lers are best known to have frequented. Coire itself claims the British 
Lucius as founder of its church in the end of the second century, and 
still preserves evidence of early Irish influence in its remains of Chris- 
tian art.” This Alpine district also seems to have been known to the 
Irish legendary writers, as may be gathered from the passage in the 
Tain bo Fraich, where Conall Cernach and his companions, on their 
expedition to the Continent, are said to have gone ‘‘ over the Ictian 
sea to the north of the Lombards, till they came to Sliebte Ealpze ;*° 
and the tradition still preserved by the family of De Salis (Macari 
Exeid. 233), whose chief seat during the Middle Ages was at Marsch- 
lins, on the right bank of the Upper Rhine, that an Irish king on his 
journey to Rome on one occasion slept at their castle, evidences the 
continued user of that highway into Italy by the insular peoples. 
The Upper Rhine valley, to which we have been conducted by this 
concurrence of hints and inferences, debouches on the Lake of Con- 
stance at Bregentz (Brigantium), where the highway from Italy through 
ancient Rhetia divided, one branch leading northward to Augsburg 
and thence to the Lower Rhine, and the other, skirting the 
southern shore of the lake, westward and southward to Zurich 
(Turicum). A traveller to or from Coire might, however, adopt an 
alternative and shorter route by the defile of the Lacus Rivarius, 
18 The passes shown in the Peutinger map, and plotted out in the Itineraries 
are substantially the same as in a modern Bradshaw :— 
1. In Alpe Maritima, : : The Corniche road. 
2. In Alpe Cottia, . ‘ 4 Mont Cenis. 
3. In Alpe Graia, . . : The Little St. Bernard. 
4. In Summo Pennino, . ‘ The Great St. Bernard. 
5. By Curia and Clayenna, . The Splugen, with its branches. 
6. By the Noric Alps, . : The Brenner. 
19 Tt is impossible, at Coire, to contemplate the sculptured slabs dug up from the 
crypt of the cathedral, without agitating in one’s mind the problem whether that 
interlaced work, with its ancient grotesques, be an evidence of Roman design travel- 
ling northward, or of insular fancy reacting on the taste of the conquerors. Of the 
Trish design of the silver and ivory shrines preserved in the sacristy there can be 
no question. 
20 Do cumlat ass a triur tar muir tar Saxam tuascirt. tar muir h’icht co tuascert 
longbard corrancattar Sliebte Ealpe. (Book of Leinster Facs., p. 252, a). Where 
the designation Saxam, given to Britain, limits the age of the piece. 
