NO. 19 NORSE VISITS TO NORTH AMERICA BABCOCK 23 



ward wall is divided by the Strait of Cabot ; and the great estuary of 

 the St. Lawrence, dividing the opposite side, might well be thought a 

 continuation of that channel and to lead out again to the sea. Just 

 this was in fact supposed down to Carder's voyages or later. We are 

 now aware that only the front of this elbow of the continent is insular 

 (Newfoundland and Cape Breton), but it was inevitable that in all 

 the centuries before the seventeenth the whole tract, if known at all, 

 should be regarded as an island. The circular external outline may 

 have been some mariner's guess from the curvature of Xewfoundalnd 

 and Nova Scotia considered together, and the scollops, serrations, or 

 indentations of this outline presented by many maps may indicate a 

 memory of real bays and inlets, though fancy would be ample for sup- 

 plying them. As to the mountains, there are considerable elevations 

 along these ocean- fronting regions, and they grow distinctly impres- 

 sive beyond the Bay of Fundy, still within the land-wall of the St. 

 Lawrence Gulf. 



We have, then, in a real region, and in only one, the several 

 peculiar features above stated, each offered also by a group of old maps 

 as though every observer had individually contributed what most 

 particularly impressed each of them, and was most vividly remem- 

 bered : and there is nothing in geography or in the circumstances of 

 those times to make predecessors of Cabot, crossing as he crossed, 

 impossible or very improbable. Indeed, that particular part of Amer- 

 ica always held itself out conspicuously, tempting discovery. The 

 coincidences may be nothing more ; but the speculation has probably 

 a sounder basis than any other advanced thus far concerning this 

 very suggestive " island." 



f Some investigators, considering Brazil a reality of the past, have 

 explained it in another way, making it a lesser Atlantis of more 

 gradual submergence, a veritable " sunken land," which went slowly 

 down, leaving no more to show for it now than the lonely, bare, 

 granite peak of Rockall, best described by Mr. Miller Christie in 

 The Scottish Geographical Magazine for 1898. He does not, how- 

 ever, suggest its identity with Brazil. According to a globe which 

 he has found, there seems to have been a sand-bank visible (at least 

 sometimes) on the spot three or four centuries ago ;but nothing could 

 have been there in the historical period to warrant belief in the great 

 Brazil ; its crags must have been frequently in sight of those who 

 sought the latter ; and the situation must always have been too incle- 

 ment. Porcupine Bank has also been presented in this connection, 

 but with even less plausibility, being too near the Irish coast, too 

 ancient in its visibility, too much out of the right direction from 



