and volcanic rocks which they support, I have already stated, in the first chapter, that to pronounce an opinion on this matter is as difficult as at once to decide which of the two, whether the foundations or superstructure of an ancient city built on wooden piles may be the oldest. We have seen that to answer this question, we must first be prepared to say whether the work of decay and restoration had gone on most rapidly above or below, whether the average duration of the piles has exceeded that of the stone buildings, or the contrary. So also in regard to the relative age of the superior and inferior portions of the earth's crust; we cannot hazard even a conjecture on this point, until we know whether, upon an average, the power of water above, or that of fire below, is most efficacious in giving new forms to solid matter.

After the observations which have now been made, the reader will perceive that the term primary must either be entirely renounced, or, if retained, must be differently defined, and not made to designate a set of crystalline rocks, some of which may be newer than the secondary formations. In this work I shall follow most nearly the method proposed by Mr. Boué, who has called all fossiliferous rocks older than the secondary by the name of primary, which thus becomes a substitute for the term transition, so far as regards

the

aqueous strata.

To prevent confusion, how

ever, I shall always speak of these as the primary fossiliferous formations, because the word primary has hitherto been almost inseparably connected with the idea of a non-fossiliferous rock.

If we can prove any plutonic, volcanic, or metamorphic rocks to be older than the secondary formations, such rocks will also be primary, according to this system. Mr. Boué having with great propriety excluded the metamorphic rocks, as a class, from the primary formations, proposed to call them all "crystalline schists," restricting the name of primary to the older fossiliferous or transition strata.

As there are secondary fossiliferous strata, so we shall find that there are plutonic, volcanic, and metamorphic rocks of contemporaneous origin, which I shall also term secondary.

In the next chapter it will be shown that the strata above the chalk have been called tertiary. If, therefore, we discover any volcanic, plutonic, or metamorphic rocks, which have originated since the deposition of the chalk, these also will rank as tertiary formations.

It may perhaps be suggested that some metamorphic strata, and some granites, may be anterior in date to the oldest of the primary fossiliferous rocks. The opinion is certainly not improbable, and will be discussed in future chapters; but I may here observe, that when we arrange the four

classes of rocks in four parallel columns in one table of chronology, it is by no means assumed that these columns are all of equal length; one may begin at an earlier period than the rest, and another may come down to a later point of time. In the small part of the globe hitherto examined, it is hardly to be expected that we should have discovered either the oldest or the newest of all the four classes of rocks. Thus, if there be primary, secondary, and tertiary rocks of the fossiliferous class, and in like manner primary, secondand tertiary plutonic formations, we may not be yet acquainted with the most ancient of the primary fossiliferous beds, or with the newest of the plutonic, and so of the rest.

ary,

271

CHAPTER XIII.

ON THE DIFFERENT AGES OF THE AQUEOUS ROCKS.

On the three principal tests of relative age-superposition, mineral character, and fossils - Change of mineral character and fossils in the same continuous formation Proofs that distinct species of animals and plants have lived at successive periods — Test of age by included fragments — Frequent absence of strata of intervening periods - Principal groups of strata in western Europe — Tertiary strata separable into four groups, the fossil shells of which approach nearer to those now living in proportion as the formation is more modern - Terms Eocene, Miocene, and Pliocene-Identifications of fossil and recent shells by M. Deshayes-Opinions of Dr. Beck.

-

In the last chapter I spoke generally of the chronological relations of the four great classes of rocks, and I shall now treat of the aqueous rocks in particular, or of the successive periods at which the different fossiliferous formations have been deposited.

Now there are three principal tests by which we determine the age of a given set of strata; first, superposition; secondly, mineral character; and, thirdly, organic remains. Some aid can occasionally be derived from a fourth kind of proof, namely, the fact of one deposit including in it

fragments of a preexisting rock, which last may thus be shown, even in the absence of all other evidence, to be the older of the two.

Superposition. The first and principal test of the age of one aqueous deposit, as compared to another, is relative position. It has been already stated, that where the strata are horizontal, the bed which lies uppermost is the newest of the whole, and that which lies at the bottom the most ancient. So, of a series of sedimentary formations, they are like volumes of history, in which each writer has recorded the annals of his own times, and then laid down the book, with the last written page uppermost, upon the volume in which the events of the era immediately preceding were commemorated. In this manner a lofty pile of chronicles is at length accumulated; and they are so arranged as to indicate, by their position alone, the order in which the events recorded in them have occurred.

In regard to the crust of the earth, however, there are some regions where, as the student has already been informed, the beds have been disturbed, and sometimes reversed. (See pp. 113, 114.) But the experienced geologist will not be deceived by these exceptional cases. When he finds that the strata are fractured, curved, inclined, or vertical, he knows that the original order of superposition must be doubtful, and he will endeavour to find sections in some neighbour