both of the vast amount of material and of water for sweeping it into the valleys, which appear in most cases to have been thus filled to the level of their highest terraces. The prevailing horizontal stratification of these deposits shows that they were spread over large areas by the current of the floods which held them in suspension. The modified drift thus increased in depth in the principal valleys through a long period, which may have continued till the last of the ice at the heads of the valley and of its tributaries had disappeared.

During the recent or terrace period the rivers have been at work excavating deep and wide channels in this alluvium. The terraces mark heights at which in this work of erosion they have left portions of their successive flood-plains. As soon as the supply of material became insufficient to fill the place of that excavated by the river, a deep channel was gradually formed in the broad flood-plain. This process was very slow, allowing the river to continue for a long time at nearly the same level, undermining and wearing away its bank on one side, and depositing the material on the opposite side, till a wide and nearly level lower flood-plain would be formed, bordered on both sides by steep terraces. When the current became turned to wear away the bank in the opposite direction, a large portion of this new flood-plain would be undermined and redeposited at a lower level; but the direction of the current's wear might be again reversed in season to leave a narrow strip which would then form a lower terrace. In this way the Merrimack River through New Hampshire has excavated its ancient high flood-plain of the Champlain period to a depth of seventy-five to one hundred and fifty feet, for a width varying from an eighth of a mile to one mile. In Canterbury and Concord we see the highest plain is being now undermined by the wear of the current, forming steep bluffs.

The very fine character of the materials which compose the lowest terraces and the interval or present flood-plain is due to this wearing away and redeposition by the river, which have been many times repeated, till what may have been at first gravel becomes very fine sand or silt. By each removal this alluvium is made one degree finer, and is deposited at a lower level and farther down the stream. The end of its slow journey is the ocean, where it will help to make the sedimentary rocks of this epoch. It has completed a great cycle of changes, ending where the upheaved and contorted ledges from which it was derived had their remote beginning.

ON CRITICAL PERIODS IN THE HISTORY OF THE EARTH, AND THEIR RELATION TO EVOLUTION; ON THE QUARTERNARY AS SUCH A PERIOD.1

BY JOSEPH LECONTE.

IN the series of rocks representing the history of the earth there occur at different horizons unconformities. In most cases these are not found at the same horizon in different places; but 'there are a few which seem to be very general. Associated with these unconformities, as is well known, there is nearly always a marked change in the fossil species. The greatness of this change is invariably in direct proportion to the generality of the unconformity. These general unconformities attended with very great changes in organic forms are the natural boundaries of the great divisions of time, and the less general unconformities attended with less sweeping change of organic forms, of the subdivisions of time.

The earlier geologists, under the influence of the then dominant idea of frequent supernatural interference with the course of nature, imagined that these unconformities marked the times of instantaneous cataclysm which disturbed the rocks and destroyed all living things, sometimes locally, sometimes generally, and that these exterminations were followed by re-creations of other and wholly different species at the beginning of the subsequent period of tranquillity. Now, however, we believe that no such instantaneous general exterminations and re-creations ever occurred. We know that unconformity simply indicates eroded land-surface, and therefore marks a period of time during which the observed place was land and received no sediment; that two series of rocks unconformable to each other denote two periods of comparative quiet, during which the observed place was sea-bottom, receiving sediment steadily, separated by a period of oscillation producing increase and decrease of land, during which the observed place was raised into land-surface, with or without crumpling of the strata, deeply eroded, and then sunk again below sea level to receive the second series of strata. The length of the two periods of repose is roughly measured by the thickness of the two conformable series. The length of the period of commotion is roughly measured by the amount of erosion at the line of unconformity.

1 Read before the National Academy of Sciences, April 18, 1877. (Reprinted from the American Journal of Science and Arts, for August, 1877.)

Evidently, therefore, every case of unconformity marks a period of time often a long period during which there was no record made in strata and fossils at the observed place; certain leaves-frequently very many-are there missing from the Book of Time. Is it any wonder, then, that skipping over these pages when we commence reading again we find the matter entirely new? Evidently, the suddenness of the change in organic forms is only apparent. If we could recover the record, which was doubtless carried on elsewhere, the break would disappear; if we could find the missing leaves the reading would be continuous. In every such instance, therefore, there is a lost interval of history. In cases of local unconformity we recover the lost record in other places, and thus fill up the blank in the history. But in some cases of very general unconformity, such as those which mark the great divisions of time, the loss is not yet recovered, perhaps is irrecoverable, though doubtless the more complete knowledge of the geology of the whole earth surface will go far toward filling blanks and making the record continuous.

The view above presented is now held by all geologists, but there seems to be danger, under the influence of the now dominant views of evolution, of erring on the other extreme. Assuming a uniform rate of evolution, many, it seems to me, commit the mistake of measuring the amount of lost interval by the amount of change of organic forms, and thus discredit the real value of the geological record by exaggerating greatly its fragmentary character. On the contrary, there appears good reason to believe that the evolution of the organic kingdom, like the evolution of society and even of the individual, has its periods of rapid movement and its intervals of comparative repose and readjustment of equilibrium. Geological history, like all other history, has its periods of comparative quiet, during which the forces of change are gathering strength, and periods of revolution, during which the accumulated forces manifest themselves in conspicuous changes in physical geography and climate, and therefore in rapid movement in the march of evolution of organic forms, — periods when the forces of change are potential, and periods when they become active. Conformable rocks represent the intervals of comparative quiet, during which organic forms are either permanent or change slowly; unconformity represents a time of oscillation, with increase and decrease of land, and therefore of rapid changes of physical conditions and correspondingly rapid movement in evolution. The general unconformities, of

course, mark times of very general commotion, of wide-spread changes of physical geography and climate, and consequently of exceptionally rapid and profound changes in organic forms.

These periods of revolution in all history are critical, and hence are of especial interest to the philosophic historian and to the evolutionist; but they are also in all history periods of lost record. And as in human so also in geological history,the farther back we go the longer are the lost intervals and the more irrecoverable the lost records. We will now give examples of such lost intervals, and show their significance in evolution.

The first and by far the greatest of these is that which occurs between the Archæan and the Palæozoic. In every part of the earth where the contact has yet been observed the Primordial lies unconformably on the upturned and eroded edges of the Archæan strata. This relation was observed first in Canada, then in various parts of the Eastern United States, then in Scotland, the Hebrides, Bavaria, Bohemia, Scandinavia. Unconformity in such widely separated localities indicates wide-spread changes in physical geography, and therefore, presumably, of all those physical conditions included in the word climate. These changes of physical geography are best illustrated in the United States. The break between the Archæan and the Primordial has been observed in very many places all over the wide area of the United States, east and west: not only in Canada, in New York, in the Appalachian region, in Wisconsin, Missouri, Arkansas, and Texas, but also all over the Rocky Mountain region, in Nebraska, Montana, Idaho, Wyoming, Colorado, Utah, Nevada, New Mexico, and Arizona. As upturned, eroded, outcropping strata mean land surface, it is evident that there was at that time a very large area or else several large areas of land in the place now occupied by the American continent. In compar ison with the subsequent Silurian it was a continental period. This land is often spoken of as Archæan land. It was indeed land of Archæan rocks, but for that very reason not of Archæan times, for these rocks were, of course, formed at the bottom of the sea in Archæan times, and therefore these localities were all seabed receiving sediment at that time. We know absolutely nothing of the land of Archæan times, and never can know anything until we find still older rocks, from the débris of which Archæan sediment was formed. The land spoken of above was land of the Lost Interval. That the interval was immensely long is evident from the prodigious erosion. That it was a period of wide

spread oscillation is also apparent, for all the places mentioned were sea-bed in Archæan, land during the interval, and again sea-bed during the Silurian. But of this long interval not a leaf of record remains.

Evidently, then, at the end of the Archæan an enormous area of Archæan sea bottom was raised up and crumpled, and became land. After remaining land for a time sufficiently long to allow enormous erosion of crumpled strata it again went down to the old Primordial shore line, and the Silurian age commenced. This time of elevation is the lost interval.

Now, when the record closed in the Archæan, as far as we know, only the lowest forms of Protozoan life yet existed. The beginnings of life had not yet differentiated into what might be called a fauna and flora. When the record again opened with the Primordial we had already a varied and highly organized fauna, consisting of representatives of many classes and of all the great types of animal structure except vertebrates. Nor were these representatives the lowest in three several departments, for Trilobites and Orthoceratites can hardly be regarded as lower than the middle of the animal scale as it now exists. It is certain, therefore, that all the great departments except vertebrates, and most of the classes of these departments, including animals at least half-way up the animal scale, were differentiated during the lost interval. The amount of evolution during this interval cannot be estimated as less than all that has subsequently taken place. Measured by the amount of evolution, this lost interval is equal to all the history of the earth which has since elapsed. We escape this very improbable conclusion only by admitting a more rapid rate of evolution during critical periods.

It is one of the chief glories of American geology to have first established the Archæan as one of the primary divisions of time. It is even yet reluctantly admitted as such by many European geologists. And yet it is seen that from every point of view, whether of the rock system or of the life system, it is by far the most widely and trenchantly separated of all the eras.

The next greatest lost interval (though far less than the preceding) is that between the Paleozoic and the Mesozoic. Here we have the next most general unconformity, indicating the next most wide-spread changes of physical geography and climate, accompanied by the most sweeping changes in organic forms, not only in species and genera, but also in families and orders. This change is the more striking as it occurs in the midst of an abun