SiA 
SECOND REPORT- 1832 . 
Among the properties which have been proposed as bonds 
of connexion between the members of plesiomorphous groups, 
may be mentioned Naumann’s formulas expressing certain 
simple algebraical relations between the dimensions of the dif¬ 
ferent axes ; thus the right prismatic primary forms of carbonates 
of baryta, strontia, &c. have all the relation 2 a = b + c, though 
the magnitudes of a, b, c, somewhat vary. 
II. Systems of Classification .—While the lowest member of 
our mineralogical classification, the Species, is in this state of un¬ 
certainty and confusion, it cannot be surprising that the superior 
departments should be not yet satisfactorily adjusted. The fact 
has been, that in England the imperfection and inutility of the 
systems commonly put forwards have been so obvious, that a ge¬ 
neral impression has established itself among our mineralogists, 
that a system is a useless source of perplexity, and that any 
system, however arbitrary, is nearly as good as any other. On the 
Continent, however, the case is widely different; and an extraor¬ 
dinary number of mineralogical systems, published of late years in 
Germany, Sweden, and France, show how earnestly foreign philo¬ 
sophers have struggled with the difficulties of the problem. Much 
indeed of the past failures and present apathy of our country¬ 
men on this subject, is to be attributed to their having under¬ 
taken the task somewhat hastily and lightly. If any one en¬ 
deavours to construct a complete classification on that which is 
perhaps the most obvious principle, the leading ingredient, he 
will soon be led into endless inconsistencies, and will obtain 
none of the great advantages which a good system will certainly 
afford. It might easily be shown by examples that this is what 
has occurred ; and any attempts to patch up the manifest defects 
of such undertakings would probably lead us, after various wan¬ 
derings and struggles, to the point at which the Swedish, Ger¬ 
man, and French chemists have already arrived. 
There has taken place in Germany a somewhat vehement 
dispute respecting the merits of two apparently opposite me¬ 
thods of classification ;—that which proceeds by external cha¬ 
racters, and that which depends on chemical composition. 
In Werner’s system, chemical differences were recognised as 
the great leading divisions ; but the subordinate distinctions 
vrere established by means of that nice discrimination of external 
characters which Werner so successfully applied and taught. 
It was, I believe, with Professor Mohs, the successor of 
Werner at Freiberg, that the design originated of demonstra¬ 
ting the possibility and the propriety of founding our minera- 
logical classification on external resemblances and differences 
alone , borrowing nothing from chemistry; so that mineralogy 
