152 ANNUAL OF SCIENTIFIC DISCOVERY. 



made experiments on rolled iron plates, of the same kind as those of which 

 ships were generally built, and had ascertained that the magnetism in 

 these also was changeable and controllable like that in bar iron, under the 

 requisite change of position, by vibratory or percussive action. lie had 

 also made experiments on a portion of a plate cut out of the side of a ship 

 recently built, and the result of his observations was to establish the fact 

 that, besides the two denominations of magnetism ordinarily received, that 

 of simple terrestrial induction and that of permanent independent mag- 

 netism, there was another denomination corresponding with neither ; not 

 being absolutely controllable, like the former, by terrestrial influences, nor 

 capable, like the latter, of resisting all kinds and modes of mechanical vio- 

 lence. To this third denomination he gave the name of "retentive mag- 

 netism." Dr. Scoresby then exhibited experiments with three sets of 

 plates, two of iron and one of steel, for the illustrating of these several 

 qualities of magnetism : 1. That of simple terrestrial induction by 

 iron plates free from polarity, which became magnetic, or changed their 

 magnetism, according to the position in which they were held. 2. Reten- 

 tentive magnetism, as illustrated by similar plates, which had been pre- 

 viously magnetized by bending and blows, such magnetism appearing 

 as if permanent wheii the plates were moved about, without being vibrated 

 or bent. And, 3rdly. Permanent magnetism, as illustrated by an elastic 

 steel plate, which, however violently it was bent, or struck, or vibrated, or 

 in whatever position, still preserved its magnetism unaltered. >Tow, this 

 retentive magnetism was the quality which had been prevalently considered 

 as permanent ; which he was prepared to show, both by experiments on 

 iron and facts of experience, was by no means a fixed quality. The 

 vibration of a ship in a heavy sea was sufficient to change the original 

 magnetism developed and augmented in the course of her construction. 

 A great deal depended on the position in which the ship had been built. 

 In the case of the Tayleur, when he first heard of the catastrophe and read 

 the evidence, he had stated to some friends at Torquay that he would ven- 

 ture to predict that she had been built with her head to the north. He had 

 found, on inquiry, that she had been built with her head nearly northeast. 

 Here, then, were the precise circumstances for expecting a change in the 

 ship's magnetic distribution. Having been built with her head to the 

 northeast, she had a certain magnetic distribution ; and when she began to 

 strain with her head to the southwest, that distribution was necessarily 

 changed, and the first effect of it had been to alter the two compasses 

 adjusted by fixed magnets. If the captain had been aware of the changes 

 which might, and most probably would, take place when the ship began to 

 strain in a different position from that in which she had been built, if he 

 had known that the compasses might vary as much as two, or three, or 

 even four points, he would have known, of course, that he must place no 

 reliance upon them. It did not follow, however, that compasses were of 

 no use because, under circumstances, they were liable to change. They 

 ought to be, and were, of great use for all that. But what he wished to 



