358 
SECOND REPORT - 1832 . 
the place of another with no definite change of external charac¬ 
ter, renders an arrangement, such as he had before proposed, 
impossible in practice. Thus, in this new system one of the 
families is sulphur, and under this are arranged the sulphurets 
of zinc, of iron, of cohalt, of nickel, of copper, &c. He ac¬ 
knowledges, however, that by this means the difficulty arising 
from isomorphism is only diminished and not removed : for 
Mitscherlich had found that electro-negative as well as electro¬ 
positive elements may replace each other isomorphously ;—for 
instance, arsenic and phosphoric acid. 
In 1824 Beudant also published his Traite Elementaire, 
which professes to found its arrangement on tw T o leading max¬ 
ims : 1. The electro-negative element imparts its character to 
the combination more frequently than the electro-positive one, 
and hence is rather to be taken as the principle of classifica¬ 
tion ; 2. The electro-negative elements are to be arranged in 
a circular series according to their natural relations. The cir¬ 
cular series thus adopted, is founded upon certain views of Mr. 
Ampere, according to which elementary substances are divided 
into three great classes, termed, 1°, Gazolytes; 2°, Leucolytes \ 
3°, Croicolytes ; according as they form gases, uncoloured solu¬ 
tions, or coloured solutions. In this system also th eform of the 
chemical symbol of composition is taken into the account. Thus 
under the family of sulphurets we have (a) simple sulphurets, 
(5) double or multiple sulphurets. 
The doctrine of isomorphism, however imperfectly developed 
at present, undoubtedly promises much fairer to disclose to us 
the true chemical relations of minerals than the views previously 
entertained. It was probably in consequence of this promise 
that an extraordinary number of new systems of mineralogical 
classifications on chemical principles were published about the 
period of which we are speaking. Gmelin (1825) was the 
author of one which appeared about the same time as that of 
Berzelius, and like that founded its leading distinctions on the 
electro-negative or formative element of bodies ; but besides 
this, it took into account the numbers which occur in the che¬ 
mical formula, distinguishing, for instance, simple, double, and 
triple silicates. The method of Nordenskiold (1827) has a 
similar groundplan, but proceeds still more decidedly upon the 
proportions of the number of atoms of the different elements ; 
and thus, as Berzelius observes, (Jahr Bericht, viii. 188,) pre¬ 
supposes a complete knowledge of the chemical composition of 
each species. 
Bonsdorff’s Essay on this subject was printed at Abo in 
Finland, in 1827 ; but its publication was prevented, the im- 
