108 Analysis of Scientific Books. 
tionably, the skeleton of a fish may be good evidence for the fish 
itself, as far as we may be satisfied without regaling on it, or are 
contented with guessing how it might have looked in a drawing, or 
skinned, varnished, and stuffed with plaster of Paris. This mode of 
assigning a species or a genus, will be still more satisfactory, when 
the naturalist has had the means of comparing the preserved skeletons 
of former days with those of existing species and genera to a sufhi- 
cient extent. But who need be told that there is such a simplicity 
and general uniformity in the skeletons of fishes, that the limits to 
this mode of investigation are very narrow indeed. They have no 
legs nor arms, no scapulee nor knee-pans, no os coccygis nor ster- 
.mim, nor phalanges, nor any of those multitudinous and ever varied 
parts from which the comparative anatomist derives so much facility 
in his researches on quadrupeds. There is something in the number 
of the spinal bones, there is something in their forms and propor- 
tions; and there is still more in the bones of the fins and in those of 
the skull. But all this is little; and while but little evidence can be 
derived from fragments, we are particularly determined to distrust 
Monsieur Blainville on points which neither he nor any one else 
could have ascertained, namely, the erection of new genera and new 
species from the contemplation of fragments, and these fragments 
often distorted by the effects of pressure and the other causes of 
change and injury to which fossil bones are exposed. 
When we said that Monsieur Blainville was ignorant of geology, 
we might also have said that he does not seem to have formed any 
conception of its nature and meaning, and of the relations of his own 
subject to it; considering this, as we do, rather a branch of zoology 
than of geology properly so called. He speaks as if the strata were 
only casual substances which might or might not be studied, but 
as being “* often useful.” We would gladly know how they are 
not always necessary instead of being often useful, at least in our 
view of the subject. If the object is merely to ascertain lost animals, 
they are neither necessary nor useful ; whence it is plain that when 
M. Blainville speaks thus, he is thinking of geology, not of his fishes, 
and thinking too, to little purpose. 
When he asserts that the nature of the organic remains offer the 
most ‘* unequivocal methods of establishing geology on indisput- 
able bases,” he is only saying what others have said before him, but 
which is not a bit the more true because it has been often said. In 
the first place, we will admit this, and then ask to what extent the 
science of geology can be based on the knowledge of organic re= 
mains? In many countries they do not exist; in many rocks they 
never occurred. ‘They are limited to a small portion of the geogra- 
phical globe, and they are confined to a small depth of the geological 
one. What would become of the theory and history of the primary 
rocks, of the trap rocks, of the volcanic rocks, if their history and 
theory depended on their organic remains? The organic remains 
