Vol. 58. | THE MINERAL ANALYSIS OF ROCKS. 171 
A greater difference appears in the mineral composition than the 
chemical analyses would at first suggest, owing to the excess of 
alkalies and calcium in the Leinster granite having found its ex- 
pression in the soda-lime felspar rather than in the micas. 
The transition from a typical granite, like that of Cornwall, to 
Tonale diorite would seem to take place through granites like 
those of Leinster, as will be seen by tracing the passage through 
them to the Tonale diorite, of which the mineral composition is 
given in the last column of the foregoing table. ‘There is a steady 
falling-off in the orthoclase. The albite-molecules, after reaching a 
maximum in the Aughrim granite, diminish in number, while those 
of anorthite increase. The impoverishment in quartz takes place 
more suddenly, and muscovite disappears altogether on passing from 
the Aughrim rock to the Tonale diorite. There is also a marked 
increase in the proportion of biotite and of the heaviest minerals, 
while a small quantity of hornblende occurs for the first time and 
links on the Tonale diorite to the more common varieties of this 
rock. The Aughrim granite occurs as an isolated exposure to the 
east of the main mass, and is of somewhat later date: its position in 
petrological classification is evidently on the borderland between 
true granite and Tonale diorite. It may be observed in passing 
that in the examples studied the post-Ordovician granites of Britain 
are more basic in character than the post-Carboniferous, and these 
are more basic than the Tertiary. 
A question will naturally arise as to the reasons by which we 
may be guided in making a choice of any particular one, from 
amid so many published analyses of a mineral species—and these 
generally so widely divergent in their results,—as that most 
likely to represent the composition of the gathering which we have 
obtained on the separation of a rock. There would seem to be 
mainly three. First, if possible, the analysis should be that of a 
mineral obtained from the same kind of rock as the one under 
investigation ; evidently it would not be wise to choose an analysis 
of a biotite taken from a limestone or a schist, when that which we 
have in hand has come froma granite. Itis not, however, necessary 
that the analysis should be that of the mineral in the form in which 
it generally contributes to the igneous mass, since the minerals 
segregated out in veins in many cases possess the same composition 
as those of the parent-rock. 
In the next place, the mineral should, if possible, have been 
obtained from the same locality or district as that of the rock under 
examination. And last, its physical properties, under which specific 
gravity may be particularly mentioned as one of the highest im- 
portance, should be similar to those of the separated mineral. 
When all these conditions are satisfied, very exact results may be 
obtained from a mineral separation; but I imagine that the future 
success of the method will depend in greatest measure on an exten- 
sion of our knowledge of the manner in which the physical properties 
of a mineral may be correlated with its chemical composition. 
