IN ROCKS OF DISSIMILAR ORIGIN. 329 
coarsely porphyritic rocks. Macroporphyritic and microporphyritic 
are convenient terms by which to indicate such differences. 
The next question which seems worthy of consideration is the 
importance which should be attached to the rounded contours of the 
fragments and crystals which enter into the constitution of many 
rocks, attaching importance to the question mainly because a just 
solution of the various cases which present themselves may serye to 
throw considerable light on the origin of the rocks in which such 
rounded fragments and crystals occur, and because we may thus be 
enabled in time to distinguish the different causes which Hane been 
instrumental in obliterating angularity. 
We find fragments and crystals which have unquestionably been 
rounded by attrition. The attrition has in some instances been 
produced by the motion imparted to these bodies either by running 
water or by the action of waves on coasts. In other instances it is 
due to the ejection of fragments and crystals from volcanic vents, 
operating both during their upward journey and their descent. 
Again, we find rounded erystals occurring in volcanic rocks, 
notably in those of a vitreous character, and we also meet with them 
occasionally in rocks of a totally different origin (e.g. the coccolith 
grains in certain limestones and the glauconite grains in some 
sedimentary rocks; the latter are, however, often the internal casts 
of Foraminifera). 
In some cases crystals present rounded forms, due to aborted 
development, as instanced by Prof. Renard in the orthoclase crystals 
of the porphyroids of Mairus and Laifour in the Ardennes. These 
crystals have, externally, the aspect of water-worn pebbles; but 
when broken open they are all seen to be twinned on the Carlsbad 
type. This is probably a much more frequent cause of the rounded 
appearance of crystals than is generally supposed. 
The roundish amygdaloidal kernels of calespar which occur in 
vesicular rocks sufficiently demonstrate their secondary origin by 
their polysynthetic character ; but in the case of other minerals 
which sometimes occupy such cavities, it is not always so easy to 
demonstrate that they have been subsequently infiltered, although 
there can, as a rule, be little doubt on the subject. The difficulty 
in such demonstration lies chiefly in showing that their origin is 
not concretionary, since in radial crystallization and in concentric 
zonal structure they often simulate coneretions and spherulitic 
bodies. It appears, indeed, an open question at times whether the 
amygdaloidal character of some rocks is not rather due to conere- 
tionary developments than to the filling in of vesicles. 
If the amygdaloids in such rocks consist of calearcous or other 
readily soluble matter, the rocks, when weathered or when treated 
with acids, present the character of true vesicular rocks, and, without 
duo precaution, very erroneous conclusions may be formed concerning 
their real nature and origin. Some of the vesicular Bonelicine 
afford good examples of this kind; and a highly vesicular character 
is sometimes imparted to certain micaceous eruptive rocks by the 
weathering out of the crystals and scales of mica, which, at first, 
