16 Biography of Berzelius. 



characters, or of compounds in which, though they possess 

 different characters, the relative proportion of the constitu- 

 ents is the same, but in which the atomic weights are not 

 equal, but twice, thrice, etc., times as great as that of each 

 other. Such bodies Berzelius termed, for the sake of anti- 

 thesis, Polymeric compounds. 



The other kind of isomerism Berzelius called Allotropism. 

 It refers solely to elementary bodies, which, owing to causes 

 not yet sufficiently understood, assume a different character 

 from that which is usual to them, and, as it appears, retain 

 this difference in many combinations, when it may be the 

 cause of differences in the character of these compounds. 

 When isomeric conditions are observed in compound bodies, 

 which consist of only two elements, combined in very simple 

 proportions, this isomerism is, according to Berzelius, to be 

 regarded less as owing to the different arrangement of the 

 elementary atoms than to the allotropic condition of one or 

 both of these elements ; nevertheless, instances may occur in 

 which both causes are simultaneously at work. 



It is possible that Berzelius may sometimes have gone too 

 far in his assumption of allotropic conditions, for there are 

 some grounds for believing that an apparent allotropism may 

 result merely from a different state of division. Thus, a few 

 years before the discovery of the first example of isomerism, 

 Magnus observed the interesting fact, that when the oxides 

 of iron, nickel, and cobalt, are reduced by means of hydrogen 

 to the lowest possible temperature, the metals obtained ignite 

 spontaneously, and oxidize when exposed to the atmosphere. 

 This pyrophoric character evidently results from the finer 

 subdivision of these metals, and it is destroyed when a higher 

 temperature is employed in their reduction, which causes the 

 particles to cohere together. The differences in platinum, 

 according as it is reduced from its salts by the humid process, 

 or obtained by igniting the ammonio-chloride : likewise the 

 unequal combustibility of silicium, and its variable solubility 

 in hydrofluoric acid, may probably be explained in the same 

 way. Nevertheless, Berzelius was inclined to ascribe all 

 these differences to allotropic conditions. 



Shortly after the appearance of the paper in which Berze- 



