416 Mr. Lyon's Observations on Magnetic Substances. 



other element whose values come very near or fall within 

 those of the three magnetic elements. This is a fact or ob- 

 servation which appears hitherto not to have been brought 

 forward, but which is certainly deserving of notice, both for 

 itself, and for the considerations to which it may lead. 



If, for the several chemical elements, we divide the specific 

 gravities by the atomic weights respectively, we shall obtain 

 a series of numbers, which are the relative numbers of atoms 

 of the several elements contained in a given space or volume. 

 In this way, and making use of the values stated above, we 

 have for the three magnetic elements: 



« .«. *t .. *. • Aur • u.l Atoms contained 



Specific Gravity. Atomic Weights. 

 1 J & in a given space. 



Nickel 8-27 ... 739*51 ... 1118 



Iron 7*207 ... 678*43 ... 1062 



Cobalt 7*8 ... 738* ... 10.57 



It will scarcely need to be noted, with reference to this cal- 

 culation, that we are at liberty to remove the decimal points, 

 provided it be done uniformly for all. 



On considering the two series, i. e. the atomic weights and 

 the ratios of atoms contained in a given space, respectively, 

 for the several chemical elements, we shall be conducted to 

 speculations relative to the atomic bulks ; or in other words, 

 to investigate what may be the relative bulks or sizes of the 

 solid nucleus of the atoms of the several elements. But for 

 our present purpose, still attending solely to the three mag- 

 netic elements, when we consider the approximation of their 

 numbers in these two series, there appears no ground for he- 

 sitation in allowing that their atomic bulks must be but little 

 different from one another. 



We come then to the conclusion that the values of the 

 atomic weights, and likewise of the atomic bulks of the three 

 magnetic elements, are near together*. And we shall find 

 further, that no other element approaches them in both these 



* The atomic weights stated above are those published by Berzelius in 

 his Essai sur la Thi'orie des Proportions Chimiques, and copied in the fifth 

 edition of Thenard's Traite de Chimie. In a subsequent list, published by 

 Berzelius in the Annates de Chimie et de Physique, tom.xxxviii. (1828), and 

 copied in Dr. Daubeny's Introduction to the Atomic Theory, some modifi- 

 cations are made in the atomic weights. The metals generally have num- 

 bers assigned to them, which are one half of those in the former list. But 

 they have not all been so reduced; there are exceptions in some few in- 

 stances. Now according to this later series, it is further remarkable, on 

 forming as above indicated a table of the numbers of atoms contained in a 

 given space, that of all the chemical elements, with the exception of car- 

 bon, the three magnetic elements and copper are those which have the 

 greatest numbers in that table. 



