Cleaveland on Mineralogy and Geology. 395 
Action of Acids, and other Tests.—Very brief, and very 
common. 
Chapter III. Systematic Arrangement of Minerals.—After 
some general, but pertinent, remarks, our author proceeds, with 
great clearness and ability, to give an account of the system of 
Werner. Its defects and merits are fairly, stated. A short 
sketch of Mohs’ Natural History Method follows; and in 
Section 3 is given the arrangement of minerals, according to 
their chemical composition. He reasons thus, for preferring a 
chemical arrangement : 
The species, the most important division, ought to be first formed. 
It must be extremely obvious, that those minerals which most resemble 
each other, belong to the same species. We are then to inquire, what 
constitutes the most perfect resemblance between two or more minerals. 
Can similarity of colour, form, fracture, hardness, &c., constitute a resem- 
blance so perfect as that which arises from identity of composition? Or 
can a difference of colour, form, fracture, &c., establish so important a dis- 
tinction between minerals, as that which is produced by dissimilarity of 
composition? Would not two minerals, both composed of phosphoric acid 
and oxide of lead, in the same proportion, belong to the same species 
although the colour of one should be brown and that of the other green ? 
Would not two minerals, composed of phosphoric acid and lime, in the 
same proportion, belong to the same species, although the forms of their 
crystals, essentially the same, should exhibit different modifications? In 
fine, can properties, liable to numerous variations from trivial and accidental 
causes, be supposed to establish the identity of two or more minerals, with 
that degree of evidence which is afforded by a well-ascertained similarity 
in composition? We hesitate not to answer these questions by saying, that 
the true composition of minerals ought to be the basis of arrangement; and 
by this only ought the species to be established. This only can’give perma- 
nence of character to the species. The composition of a mineral, that is, 
the ingredients proper and essential to its composition, may remain un- 
affected by the accidental presence of certain foreign ingredients, which 
materially change several of the external characters. 
Hence a species may be thus defined : a collection of minerals, which are 
composed of the same ingredients, combined in the same proportions.—p. 84, 
To the question, whether chemical analysis be yet sufficiently 
perfect and accurate for the determination of the species, our 
author answers, that zt is, as far as regards the alkaline and 
earthy salts, some species of combustibles, and almost every 
species of the ores of the metals. 
There remains, however, one class of minerals, composed chiefly of 
different earths, combined in various proportions, such as garnet, feldspar. 
&c., whose composition is not yet sufficiently understood to be employe 
as the basis of specific or even generic arrangement, 
For the present, therefore, some other mode or modes must 
be employed for the determination of the species of earthy 
minerals ; and, when it can be ascertained, the primitive form, 
and also the form of the integrant particles, mark the species 
of crystallized minerals with great precision. 
These principles (composition and primitive form) are the 
basis of Haiiy’s theory, ‘“ whose first object seems to have 
been to unite different crystallized varieties of the same species 
