MAGMATIC ALTERATION OF CERTAIN MINERALS 263 
otherwise go on being prevented by the acidity of the magma in 
the rhyolites and allied rocks. 
Turning to the physical conditions involved we find the erup- 
tive rocks divided into two great classes; the plutonic rocks 
being those which in general have solidified at a great depth 
and consequently slowly and under great pressure, while the 
volcanic rocks have solidified at, or near, the surface and hence 
more rapidly and under little pressure. We find on examina- 
tion of eruptive rocks in general that the two classes show a 
marked difference in the condition of their hornblende and bio- 
tite crystals. 
In the plutonic rocks the hornblende and biotite (which are 
among the oldest components) are almost invariably unaltered, 
the borders being as clear as the interior and not showing any 
opacity or granular augite opacite aggregate. It is evident 
from their fresh and clear appearance that since their formation 
no forces have come into play (atmospheric and dynamomet- 
amorphic action being left out of account), to change them from 
their original condition. This invariability of non-alteration in 
the plutonic rocks would be absolute (so far as my knowledge 
goes) were it not for two exceptions. Both of these are in 
nepheline syenite, one from Sierra di Monchique in Portugal,* 
and the other from Serra di Tingua in Brazil.2 These two rocks, 
it is well known, are connected with peculiar dike types, and 
the group in general presents peculiar features. These two 
exceptions cannot then invalidate the general law, deduced 
from very numerous and world-wide observations, that in the 
plutonic rocks the hornblende and biotite are unaltered. 
This constancy of non-alteration is in striking contrast with 
the behavior of the two minesals in the volcanic rocks. In these 
the alteration is extremely common, quite the rule, in fact, among 
the intermediate and basic members, though comparatively rare 
in the latter, owing to rarity of hornblende and biotite, and only 
failing completely (or practically so) at the extreme acid end. 
*V,. WERVEKE, Neu. Jahrb., 1880, II, 151. 
2GRAEFF, ditto, 1887, II, 236, 242, 244. 
