Barlow's Essay on Magnetic Atlraclion. 137 



deviation caused by the guns is to be determined : this being 

 known, a pin is to be fixed experimentally to the binnacle, iu 

 such a situation, that when a ball ov plate of iron is slipped on, 

 and the binnacle turned about, the deviations arising from the 

 ball or plate shall be the same as those produced in the like si- 

 tuations by the guns. 



This being dene, the plate (which Mr. B. adopts in preference 

 to the ball) is to be laid aside ; then, at any time when it is de- 

 sirable to ascertain the local effects of the guns, &c. the plate is 

 to be applied in its assigned situation, and it is to be observed 

 how much it draws the needle out of its prior direction ; and just 

 so much will the guns have deflected it before the experiment. 

 This being ascertained, and the course of the vessel corrected ac- 

 cordingly, the plate is to be again removed, and laid aside till 

 some new circumstance renders its application again necessary. — 

 Several experiments are then reported as they were made with 

 an attached plate and the 24 -pounder above mentioned, which 

 were lepcated before some of the Lords of the Admiralty, and 

 apparently much to their satisfaction. 



Nothing can be desired more simple than this method of cor- 

 recting the errors arising from local attraction ; and we are glad to 

 find that the Board of Longitude has ap})rovcd of the principle, 

 and has recommended its adoption to tiie Admiralty, who it seems 

 are about to give it aproper trial in both hemispheres of the globe. 



Having dismissed this su!)ject, Mr. B. turns his attention to 

 an explanation of the cause of the daily variation of the compass; 

 a phenomenon which has never yet been explained in a manner 

 entirely satisfactory, and which certainly appears to have an in- 

 timate connexion with the laws and princijjles discovered in the 

 early part of the experiments above referred to. Mr. B., after 

 enumerating all the known circumstances attending this plieno- 

 menon, observes : " It appears then, that any hypothesis which 

 may be advanced to account for the diurnal variation of the 

 needle ought to be competent to the explanation of the follow- 

 ing facts, viz. 



I. That the general character of the daily change in the di- 

 rection of the needle shall be westerly in the morning and easterly 

 in the evening. 



'^. That although the above is the general phenomenon, yet 

 this is sbl)ject to certain modifications, indicated by tlic letters 

 W.E.; t'WE, cAVEk', where the small italics denote ii less 

 change than the Roman letters. 



.'J. That the variation in the night, or while the sun is l)elow 

 the horizon, is very inconsiderable with respect to that which 

 takes place in the ^ay-timc. 



'\. That 



