46 THE LABRADOR PENINSULA. chap, xxiir. 



of iron, not only in tlie gneissoid rocks which rise im- 

 mediately from the sea in bold hills 200 to 300 feet high, 

 but in the sand of the shore, and throughout the exten- 

 sive flats which occur between the hills when they retire 

 from the coast and the sea margin. 



In some places, such as Sawbill Eiver, twenty-three miles 

 and a half west of St. John, the black oxide is found in 

 nests and veins ; but the magnetic action of the ferruginous 

 minerals has not been observed to affect the compass five 

 or six miles from shore. Admiral Bayfield says, in his 

 ' Sailing Directions,' that ' an opinion is prevalent that the 

 compasses of vessels are disturbed in the Gulf and Eiver 

 St. La^vrence, and such disturbances have been attributed 

 to the magnetic ores of ii^on in the hills, particularly 

 those of the north coast. The magnetic oxide of iron 

 does exist abundantly, and attracts the needle very 

 powerfully at some points, particularly along the coast 

 from the Bay of Seven Islands eastward. Among the 

 Mingan Islands, we found the variation to vary from this 

 cause from 19° to 31° west. At Port Neuf and Mani- 

 couagan Point the needle was also disturbed ; but these 

 effects were only noticed when the instrument was placed 

 on shore. In two instances only, when sailing within 

 two miles from the shore, have we observed any effect of 

 the kind upon the compasses on board the Gulnare, and 

 then only to the amount of a few degrees. When 

 running from place to place, at greater distances from the 

 coast, nothing of the kind has been noticed ; so that I 

 feel sure, that in nine cases out of ten, where this source 

 of erroneous reckoning has been alleged as the cause of 

 accidents to vessels, they originated either in errors of the 



