45 
president’s address. 
crystallisation. The matter seems to have a ver^ important 
bearing on certain points in the nomenclature of rocks. 
Ought such a mass to receive different names in its different 
parts, or should it all be called by the name descriptive of its 
original state. In this case, where proof is at hand, it is 
easy to call the rock both massive and schistose, an altered 
dolerite ; but how are we to do with cases where there is no 
certainty of the kind? Very frequently the probability is very 
great that something of the sort has gone on. I may 
mention the cases of rocks from south Devon, Dolgelly, and 
Arran, which I have examined ; and indeed one of the rocks 
from Nuneaton has a very suspicious appearance. A recent 
writer on the other side of the Atlantic, however, answers 
unhesitatingly that the original state of a rock is that which 
should give it its name ; and he expresses his hope that the 
name diorite may soon be dropped as being only an alteration 
product of the plagioclase augite rocks, which he classes 
together as basalt, stating his conviction that hornblende is 
not a product of consolidation from fusion, but of changes 
subsequently induced ; meeting such cases as the hornblende 
in the andesites and trachytes by the proposition that it is 
only what is left of a previously altered rock on being melted 
up. That hornblende is the result of processes which differ 
from simple crystallisation by the cooling of a molten mass 
at ordinary pressures is very probable indeed, as shown by 
the experiments of MM. Levy and Fonque, which I 
mentioned last year, but that it is not an orignal crystal¬ 
lisation product of a rock mass may well be doubted, when 
we consider how (1) it occurs so regularly, and apparently 
unaltered in syenites and some granites ; and (2), that the 
argument from the failure of laboratory experiments would 
equally prove orthoclase and mica to be secondary. 
The question as to the formation of minerals in rocks, and 
their subsequent changes, presents some interesting points 
when viewed from another direction. The mineral leucite only 
occurs, so far as is at present known, in the recent, or at most 
late, Tertiary-lavas ; and the question naturally arises whether 
there have been old leucite rocks which, by change of arrange¬ 
ment, have lost their characteristic mineral. The crystalline 
form of the mineral has presented many difficulties. Appa¬ 
rently belonging to the regular or cubic system of crystals, it 
was found on examination in polarised light to possess feeble 
double refraction, and indeed to be twinned in a highly com¬ 
plicated pattern. Hence a grave doubt was thrown on its 
true position in the crystallographic system ; and careful 
measurements seem to establish that it is really tetragonal, 
