Mohs’* System of Crystallography 
SO. Cwinection between the Species . — From the union of ho- 
mologous individuals by transitions, it is plain, that an admi- 
rable connection reigns throughout these natural historical species 
(4.) This co7inection is a proof of the correctness with which 
the species have been fixed. It does not therefore seem advi- 
sable to divide the species ; to distinguish them into subspecies, 
(species in the Wernerian system). At all events, such a subdivi- 
sion merits entire reprobation, whenever, as has frequently hap- 
pened, it proceeds without logical foundation, without logical 
Ibrm, and according to nothing but the author’s caprice. 
31. The principle of a natural system is natural histor ical si- 
milarity . — The principle which natural history follows in erecting 
a natural system, is natural historical similarity. An artificial 
system, on the other hand, rests upon distributions according to 
particular marks or characters, among which natural historical si- 
milarity is commonly lost sight of. But how different soever the 
two systems may be in their natures, they have yet this in com- 
mon, that the natural historical species^ as it has been explained 
above, must lie at the foundation of both, if they are to be use- 
ful in natural history. It appears, then, to have been a great 
error to give the name of artificial systems to certain injudici- 
ous distributions of minerals ; for example, that of Brunner ; 
and from their failure, to infer the impossibility of such under- 
takings in mineralogy. 
32. Possibility of artificial systems . — When the species 
are determined, the possibility of an artificial system depends 
upon the first distribution, so that no species may, by means 
of it, be again split asunder. Hitherto no basis of mineralogi- 
cal distribution has been discovered, sufficiently general to effect 
this purpose. The systems of crystallisation are fitted to ef- 
fect it ; and may, therefore, on the supposition that the form of 
certain species, which is yet unknown, will hereafter be disco- 
vered, serve as the gi’oundwork of an artificial system. 
33. Artificial systems do 7iot fulfil all the requisitions (f mi- 
neralogy . — An artificial system, however, is of little service in 
fnineralogy. It can, at best, be applied only to designate va- 
rieties that may occur, as plants are designated in botany ac- 
cording to the method of Linnaeus. After the acute Werner s 
