of JStatural Philosophy. 139 



Book IV. Prop. 1. Scholium. The editor of the Lou- 

 don edition of 1799, left the question undecided, as Mr. 

 Cavallo had done, whether any other than ferruginous bodies 

 are capable of magnetic properties. This scholium has been 

 retained ; although the experiments of Tourte of Berlin,* 

 and Larapadius, place it beyond a doubt, that nickel and 

 cobalt possess the same capacity, in this respect, with iron, 

 except in an inferior degree. According to Lampadius, (see 

 Thomson's Annals, 1815,) the " magnetic energies" of iron, 

 nickel, and cobalt are as the numbei's 55, 35, and 25, re- 

 spectively. 



Prop. 2. Schol. If it were worth the while to retain 

 the experiments of Musschenbroek on the variation of the 

 magnetic force at all, in place of the far more important 

 ones of Coulomb, the numbers ought at least to have been 

 corrected. Had the editor of the second edition taken 

 these experiments from Musschenbroek himself, instead of 

 taking them from Cavallo, he would have avoided the error 

 of making the distances ail ten times too large.f The de- 

 nomination employed in the original statement (see Philos. 

 I. 206. Ed. 1744,) is tenths of inches, instead of inches. 



Prop. 10. To illustrate the mode of finding the declina- 

 tion of the needle by amplitudes of the heavenly bodies, 

 the following example is stated : " If the magnetic ampli- 

 tude is 80° eastward of the north, and the true amplitude is 

 82° towards the same side, then the variation of the needle 

 is 2° west." This statement cannot be reconciled to any 

 definition of the term " amplitude ;" and it cannot be re- 

 conciled with the usual definition and the one given in the 

 Astronomy, except by making all the alterations which fol- 

 low: " If the magnetic amplitude be 10° N. of E. and the 



mates of altitudes from the barometer, tliat the ascent correspoi)ding to 1-10 

 in. fall of the mercury, instead of being one hundred and three feet, is 

 at a mean, (that is, when the barometer itself is at 30 in. and the thermom- 

 eter at 60°;) onh' about ninely-three feet. 



* Nicholson's Journal, Vol. 26. 



t If perfect exactness in such a case were of any importance, it would 

 be necessary to recollect, that Musschenbroek's denominations were Rhin- 

 land measure ; which are greater than the English in the ratio of 1,03 to 1. 



We acknowledge ourselves indebted, for the above remark, to the sug 

 gestion of a scientific friend. 



