DRIFT, BOULDERS, etC. 
427 
In ascending from facts to causes, it is necessary to classify whenever it is possible; to 
arrange together the phenomena which resemble each other, and to separate those which offer 
a dissimilarity in their origin. From the observations of geologists on both sides of the Atlan¬ 
tic, it is not difficult to see that it would be absurd to place all hills of sand and pebbles, or 
ridges of the same materials, and collections of boulders, in one class. The truth is, a variety 
of causes have operated in their production ; and geologists, in attempting to assign a cause 
which should cover the whole ground, have erred. In finding, in our long lines or ridges of 
rounded stone and gravel, evidences of glacial action, they have overlooked the fact that all 
bodies of water throw up such ridges. In explaining scratches or groovings of rocks by ice¬ 
bergs, they do not seem to have observed that the bottom of seas is covered with a thick bed 
of soft materials, capable of defending the rocks from such markings. In supposing the nor¬ 
thern hemisphere to have been capped with ice, converting all the water to a solid mass, they 
forget that such a condition is incompatible with life, that there must necessarily have followed 
an entire extinction of all organic beings, and that there would have been a gap in the series ; 
but we do not find this to be the case. There is a perfection apparently in the chain ; it is not 
broken just at the dawn of the present era, as is proved by the fossils of the tertiary. As we 
find, therefore, phenomena very similar, but produced by very different causes, so our theo¬ 
retic generalizations should correspond. In the grooving of rocks, we admit a variety of 
agents : so, in the accumulations of loose materials, we must do the same; and when we 
have sufficiently observed to classify correctly the results, we shall be able to comprehend 
the causes which have modified the surface, and to explain what each has effected in its own 
sphere and capacity. 
