EXPEDITION TO FUNK ISLAND, 499 



now and then given that on June 25 he must have been somewhere in 

 the vicinity of the Bird Eocks. 



It is also difficult to resist the temptation of suggesting that there 

 has been a mistake in translating Cartier's log, and that " 15 leagues 

 southeast" should be 15 leagues southwest. This is certainly a some- 

 what radical change, but the difierence between sudest and sudouest is 

 not very great to the eye, and the translator might easily have gone 

 astray there. 



Moreover why Oartier should have run dead before the wind to the 

 eastward when he seems to have been trying to work to the westward, 

 and could have held his own simply by keeping the wind abeam, is rather 

 strange. 



Finally, if he was seven and a half leagues northwest by west from 

 some point on the west coast of Newfoundland, a run of 15 leagues 

 southeast would have carried him plump ashore, owing to the great 

 amount of westerly variation. 



In this connection I desire to express my indebtedness to the courtesy 

 of Commander J. E. Bartlett, Chief of the Hydrographic Office, and to 

 Mr. G. W. Littlehales, of the Division of Chart Construction, for very 

 kindly supplying me with the necessary data for solving this problem. 



Fortunately, too, there is a reference to the island of Brion, giving its 

 distance and direction from the lies des Margaulx, and this is alone suf- 

 ficient to identify the spot, as they harmonize with existing facts. 



Brion Island, like Blanc Sablon and Chateau Bay (the Bay of Castles 

 in Hakluyt), has luckily retained its name unchanged, while so many 

 other places have either been re-named or had their original appellations 

 anglicized out of existence. 



Further confirmation is found in the Margaulx themsel ves,these birds, 

 " which bite even as dogs," being gannets, whose descendants, in spite 

 of centuries of persecution, may still be found breeding where their 

 ancestors did before them. Ordinarily the presence or absence of any 

 given species of bird might seem of small value as a factor in the 

 identification of a locality, but the gannet is extremely critical in the 

 choice of a breeding place, and extremely pertinacious in clinging to it 

 when once selected. 



Once established, nothing short of complete destruction appears to 

 drive them away, and unless carefully protected this curiously conserv- 

 ative spirit will eventually result in extermination. 



Thus, while there are many points along the coast from Maine to Lab- 

 rador where the Gannets might breed, they are found, so far as I have 

 been able to ascertain, only at three places, an island in the Bay of 

 Fundy, the Bird Eocks and Bonaventure Island at Perce, Canada, the 

 colony at Mingan being too small and too nearly exterminated to be 

 taken into consideration. * 



*Dr. Stejueger tells lue that the same thing occurs in Europe, where the Gannets 

 cross the North Sea to breed on the Scottish coast, although there are numerous 

 favorable localities on the coast of Norway. 



