SALMON — ON ROCKS. 
231 
These new constituents, introduced in the course of time, are certainly foreign to 
tiic mineral in its original condition, and are on that account to lie deilucted. 
Since, however, alterations seldom take place merely hy addition, hut more fre- 
(lucntly hy loss of constituents, it is likewise retjuisite that, in the latter case, 
tlie (luivntitics lost should be added . . this is seldom possible. . . These examples 
will suttice to show the importance of the minute quantities of substances present 
in minerals, and generally considered as accidental. . . . They then no longer 
a])^)ear as accidental, but indicate the transition of one mineral into others, and lay 
before us clearly the genetic part of the conversion-process. ... It is but rarely 
that the chemist is able to produce artificially the changes observed in nature ; 
and, in order to trace the various stages of these natural processes, there remams 
no other course for him to pvu'sue than to ascertain by analysis the increase of the 
non-essential and the decrease of the essential constituents ; and, from the nature 
of the former, to draw conclusions as to the processes which were going on in the 
mineral when formed. . . . Cordierite is the starting-point of a whole series of 
alterations, finally ending in Mica. ... -It will scarcely ever be possible to convert 
Augite, Olivine, and Hornblende into Serpentine in our laboratories ; but, when 
we find Serpentine in the form of these minerals, this fact is sufficient evidence that 
such a conversion can take place. . . . The Silicates that are most readily decom- 
posed are generally those containing lime, protoxide of iron, and manganese. . . . 
Minerals consisting chiefly of Silicates of Alumina and Magnesia, which are the 
most stal)le of minerals, are less liable to decomposition. . . . The Silicates least 
liable to decomposition are chiefly such as have originated from the alterations of 
the less stable Silicates ; so that they may be termed the fnal products of altera- 
tion. They are not liable to undergo any fiu-ther alteration by means of the 
atmospheric agents. They may be either compomid or simple Silicates, — Mica, 
Chlorite, Serpentine, Asbestos ; Steatite, Talc, Clay, or Kaolin. * * * The most 
remarkable product of mineral alteration is unquestionably mica. ... It is scarcely 
i)iferior to any in stability. . . . These minerals are not the only final products of 
alteration ; there are, besides. Quartz in its various modifications, oxides, hydrates, 
and carbonates, incapable of higher oxidation. 
" The cyclical character which is so generally recognisable in the alteration of 
minerals suggests the question, whether the last-mentioned minerals, which have 
been spoken of as final products of alteration, may not really be particular stages 
of wider cycles of alteration. It is certain that there is a limit to theu duration ; 
those that are most stable among them — the silicates of alumina and magnesia — 
may under certain conditions become the starting points of other metamorphic 
processes. If the silicates of magnesia were dissolved and carried away by water, 
they wouM take part in the formation of new minerals. There are likewise means 
by which the peroxide of iron and quartz may be again brought within the cycle 
of alteration." 
Descriptive List of the Rock-forming Minerals. 
The oxygen^quotient is placed after the name of the species in square 
brackets, thus [ ]. 
The name of each species is followed by its synonyms and varieties. The 
synonym is separated from the name of the mineral by a comma ; the 
variety by a semicolon. 
The per-centage of constituents is put in brackets after the chemical symbol 
of the constituent. Thus K (5) means that the per-centage of soda 
is 5 per cent. 
A. Elements and Binary Compounds. 
1. Grapuite. The only regular rock-mineral formed of Carbon is Graphite, 
of which Diamond is the transparent variety. It often contains a 
variable quantity of iron, and even the purest varieties have traces of 
