Some peculiar branches of the camel stem appeared in North America. In the Pliocene arose a llama with the long neck and limbs of a giraffe, whose food was cropped from the leaves and branches of trees. Far more generalized in structure was the Oreodon, an animal related to the camels, but with distinct affinities also with other lines, such as those of the hog and deer. These curious creatures were much like the peccary in appearance, except for their long tails. In the middle Eocene they roamed in vast herds from Oregon to Kansas and Nebraska.

THE RUMINANTS. This division of the artiodactyls includes antelopes, deer, oxen, bison, sheep, and goats,--all of which belong to a common stock which took its rise in Europe in the upper Eocene from ancestral forms akin to those of the camels. In the Miocene the evolution of the two-toed artiodactyl foot was well-nigh completed. Bonelike growths appeared on the head, and the two groups of the ruminants became specialized,--the deer with bony antlers, shed and renewed each year, and the ruminants with hollow horns, whose two bony k.n.o.bs upon the skull are covered with permanent, pointed, h.o.r.n.y sheaths.

The ruminants evolved in the Old World, and it was not until the later Miocene that the ancestors of the antelope and of some deer found their way to North America. Mountain sheep and goats, the bison and most of the deer, did not arrive until after the close of the Tertiary, and sheep and oxen were introduced by man.

The hoofed mammals of the Tertiary included many offshoots from the main lines which we have traced. Among them were a number of genera of clumsy, ponderous brutes, some almost elephantine in their bulk.

THE CARNIVORES. The ancestral lines of the families of the flesh eaters--such as the cats (lions, tigers, etc.), the bears, the hyenas, and the dogs (including wolves and foxes)--converge in the creodonts of the early Eocene,--an order so generalized that it had affinities not only with the carnivores but also with the insect eaters, the marsupials, and the hoofed mammals as well.

From these primitive flesh eaters, with small and simple brains, numerous small teeth, and plantigrade tread, the different families of the carnivores of the present have slowly evolved.

DOGS AND BEARS. The dog family diverged from the creodonts late in the Eocene, and divided into two branches, one of which evolved the wolves and the other the foxes. An offshoot gave rise to the family of the bears, and so closely do these two families, now wide apart, approach as we trace them back in Tertiary times that the Amphicyon, a genus doglike in its teeth and bearlike in other structures, is referred by some to the dog and by others to the bear family. The well-known plantigrade tread of bears is a primitive characteristic which has survived from their creodont ancestry.

CATS. The family of the cats, the most highly specialized of all the carnivores, divided in the Tertiary into two main branches.

One, the saber-tooth tigers (Fig. 351), which takes its name from their long, saberlike, sharp-edged upper canine teeth, evolved a succession of genera and species, among them some of the most destructive beasts of prey which ever scourged the earth. They were masters of the entire northern hemisphere during the middle Tertiary, but in Europe during the Pliocene they declined, from unknown causes, and gave place to the other branch of cats,--which includes the lions, tigers, and leopards. In the Americas the saber-tooth tigers long survived the epoch.

MARINE MAMMALS. The carnivorous mammals of the sea--whales, seals, walruses, etc.--seem to have been derived from some of the creodonts of the early Tertiary by adaptation to aquatic life.

Whales evolved from some land ancestry at a very early date in the Tertiary; in the marine deposits of the Eocene are found the bones of the Zeuglodon, a whalelike creature seventy feet in length.

PRIMATES. This order, which includes lemurs, monkeys, apes, and man, seems to have sprung from a creodont or insectivorous ancestry in the lower Eocene. Lemur-like types, with small, smooth brains, were abundant in the United States in the early Tertiary, but no primates have been found here in the middle Tertiary and later strata. In Europe true monkeys were introduced in the Miocene, and were abundant until the close of the Tertiary, when they were driven from the continent by the increasing cold.

ADVANCE OF THE MAMMALIA DURING THE TERTIARY. During the several millions of years comprised in Tertiary time the mammals evolved from the lowly, simple types which tenanted the earth at the beginning of the period, into the many kinds of highly specialized mammals of the Pleistocene and the present, each with the various structures of the body adapted to its own peculiar mode of life.

The swift feet of the horse, the horns of cattle and the antlers of the deer, the lion's claws and teeth, the long incisors of the beaver, the proboscis of the elephant, were all developed in Tertiary times. In especial the brain of the Tertiary mammals constantly grew larger relatively to the size of body, and the higher portion of the brain--the cerebral lobes--increased in size in comparison with the cerebellum. Some of the hoofed mammals now have a brain eight or ten times the size of that of their early Tertiary predecessors of equal bulk. Nor can we doubt that along with the increasing size of brain went a corresponding increase in the keenness of the senses, in activity and vigor, and in intelligence.

CHAPTER XXII

THE QUATERNARY

The last period of geological history, the Quaternary, may be said to have begun when all, or nearly all, living species of mollusks and most of the existing mammals had appeared.

It is divided into two great epochs. The first, the Pleistocene or Glacial epoch, is marked off from the Tertiary by the occupation of the northern parts of North America and Europe by vast ice sheets; the second, the Recent epoch, began with the disappearance of the ice sheets from these continents, and merges into the present time.

THE PLEISTOCENE EPOCH

We now come to an episode of unusual interest, so different was it from most of the preceding epochs and from the present, and so largely has it influenced the conditions of man's life.

The records of the Glacial epoch are so plain and full that we are compelled to believe what otherwise would seem almost incredible, --that following the mild climate of the Tertiary came a succession of ages when ice fields, like that of Greenland, shrouded the northern parts of North America and Europe and extended far into temperate lat.i.tudes.

THE DRIFT. Our studies of glaciers have prepared us to decipher and interpret the history of the Glacial epoch, as it is recorded in the surface deposits known as the drift. Over most of Canada and the northern states this familiar formation is exposed to view in nearly all cuttings which pa.s.s below the surface soil. The drift includes two distinct cla.s.ses of deposits,--the unstratified drift laid down by glacier ice, and the stratified drift spread by glacier waters.

The materials of the drift are in any given place in part unlike the rock on which it rests. They cannot be derived from the underlying rock by weathering, but have been brought from elsewhere. Thus where a region is underlain by sedimentary rocks, as is the drift-covered area from the Hudson River to the Missouri, the drift contains not only fragments of limestone, sandstone, and shale of local derivation, but also pebbles of many igneous and metamorphic rocks, such as granites, gneisses, schists, dike rocks, quartzites, and the quartz of mineral veins, whose nearest source is the Archean area of Canada and the states of our northern border. The drift received its name when it was supposed that the formation had been drifted by floods and icebergs from outside sources,--a theory long since abandoned.

The distribution also of the drift points clearly to its peculiar origin. Within the limits of the glaciated area it covers the country without regard to the relief, mantling with its debris not only lowlands and valleys but also highlands and mountain slopes.

The boundary of the drift is equally independent of the relief of the land, crossing hills and plains impartially, unlike water-laid deposits, whose margins, unless subsequently deformed, are horizontal. The boundary of the drift is strikingly lobate also, bending outward in broad, convex curves, where there are no natural barriers in the topography of the country to set it such a limit. Under these conditions such a lobate margin cannot belong to deposits of rivers, lakes, or ocean, but is precisely that which would mark the edge of a continental glacier which deployed in broad tongues of ice.

THE ROCK SURFACE UNDERLYING THE DRIFT. Over much of its area the drift rests on firm, fresh rock, showing that both the preglacial mantle of residual waste and the partially decomposed and broken rock beneath it have been swept away. The underlying rock, especially if ma.s.sive, hard, and of a fine grain, has often been ground down to a smooth surface and rubbed to a polish as perfect as that seen on the rock beside an Alpine glacier where the ice has recently melted back. Frequently it has been worn to the smooth, rounded hummocks known as roches moutonnees, and even rocky hills have been thus smoothed to flowing outlines like roches moutonnees on a gigantic scale. The rock pavement beneath the drift is also marked by long, straight, parallel scorings, varying in size from deep grooves to fine striae as delicate as the hair lines cut by an engraver's needle. Where the rock is soft or closely jointed it is often shattered to a depth of several feet beneath the drift, while stony clay has been thrust in among the fragments into which the rock is broken.

In the presence of these glaciated surfaces we cannot doubt that the area of the drift has been overridden by vast sheets of ice which, in their steady flow, rasped and scored the rock bed beneath by means of the stones with which their basal layers were inset, and in places plucked and shattered it.

TILL. The unstratified portion of the drift consists chiefly of sheets of dense, stony clay called till, which clearly are the ground moraines of ancient continental glaciers. Till is an unsorted mixture of materials of all sizes, from fine clay and sand, gravel, pebbles, and cobblestones, to large bowlders. The stones of the till are of many kinds, some having been plucked from the bed rock of the locality where they are found, and others having been brought from outside and often distant places. Land ice is the only agent known which can spread unstratified material in such extensive sheets.

The FINE MATERIAL of the till comes from two different sources. In part it is derived from old residual clays, which in the making had been leached of the lime and other soluble ingredients of the rock from which they weathered. In part it consists of sound rock ground fine; a drop of acid on fresh, clayey till often proves by brisk effervescence that the till contains much undecayed limestone flour. The ice sheet, therefore, both sc.r.a.ped up the mantle of long-weathered waste which covered the coun try before its coming, and also ground heavily upon the sound rock underneath, and crushed and wore to rock flour the fragments which it carried.

The color of unweathered till depends on that of the materials of which it is composed. Where red sandstones have contributed largely to its making, as over the Tria.s.sic sandstones of the eastern states and the Algonkian sandstones about Lake Superior, the drift is reddish. When derived in part from coaly shales, as over many outcrops of the Pennsylvanian, it may when moist be almost black. Fresh till is normally a dull gray or bluish, so largely is it made up of the grindings of unoxidized rocks of these common colors.

Except where composed chiefly of sand or coa.r.s.er stuff, unweathered till is often exceedingly dense. Can you suggest by what means it has been thus compacted? Did the ice fields of the Glacial epoch bear heavy surface moraines like the medial and lateral moraines of valley glaciers? Where was the greater part of the load of these ice fields carried, judging from what you know of the glaciers of Greenland?

BOWLDERS OF THE DRIFT. The pebbles and bowlders of the drift are in part stream gravels, bowlders of weathering, and other coa.r.s.e rock waste picked up from the surface of the country by the advancing ice, and in part are fragments plucked from ledges of sound rock after the mantle of waste had been removed. Many of the stones of the till are dressed as only glacier ice can do; their sharp edges have been blunted and their sides faceted and scored.

We may easily find all stages of this process represented among the pebbles of the till. Some are little worn, even on their edges; some are planed and scored on one side only; while some in their long journey have been ground down to many facets and have lost much of their original bulk. Evidently the ice played fast and loose with a stone carried in its basal layers, now holding it fast and rubbing it against the rock beneath, now loosening its grasp and allowing the stone to turn.

Bowlders of the drift are sometimes found on higher ground than their parent ledges. Thus bowlders have been left on the sides of Mount Katahdin, Maine, which were plucked from limestone ledges twelve miles distant and three thousand feet lower than their resting place. In other cases stones have been carried over mountain ranges, as in Vermont, where pebbles of Burlington red sandstone were dragged over the Green Mountains, three thousand feet in height, and left in the Connecticut valley sixty miles away. No other geological agent than glacier ice could do this work.

The bowlders of the drift are often large. Bowlders ten and twenty feet in diameter are not uncommon, and some are known whose diameter exceeds fifty feet. As a rule the average size of bowlders decreases with increasing distance from their sources.

Why?

TILL PLAINS. The surface of the drift, where left in its initial state, also displays clear proof of its glacial origin. Over large areas it is spread in level plains of till, perhaps bowlder- dotted, similar to the plains of stony clay left in Spitzbergen by the recent retreat of some of the glaciers of that island. In places the unstratified drift is heaped in hills of various kinds, which we will now describe.

DRUMLINS. Drumlins are smooth, rounded hills composed of till, elliptical in base, and having their longer axes parallel to the movement of the ice as shown by glacial scorings. They crowd certain districts in central New York and in southern Wisconsin, where they may be counted by the thousands. Among the numerous drumlins about Boston is historic Bunker Hill.

Drumlins are made of ground moraine. They were acc.u.mulated and given shape beneath the overriding ice, much as are sand bars in a river, or in some instances were carved, like roches moutonnees, by an ice sheet out of the till left by an earlier ice invasion.

TERMINAL MORAINES. The glaciated area is crossed by belts of thickened drift, often a mile or two, and sometimes even ten miles and more, in breadth, which lie transverse to the movement of the ice and clearly are the terminal moraines of ancient ice sheets, marking either the limit of their farthest advance or pauses in their general retreat.

The surface of these moraines is a jumble of elevations and depressions, which vary from low, gentle swells and shallow sags to sharp hills, a hundred feet or so in height, and deep, steep- sided hollows. Such tumultuous hills and hummocks, set with depressions of all shapes, which usually are without outlet and are often occupied by marshes, ponds, and lakes, surely cannot be the work of running water. The hills are heaps of drift, lodged beneath the ice edge or piled along its front. The basins were left among the tangle of morainic knolls and ridges as the margin of the ice moved back and forth. Some bowl-shaped basins were made by the melting of a ma.s.s of ice left behind by the retreating glacier and buried in its debris.

THE STRATIFIED DRIFT. Like modern glaciers the ice sheets of the Pleistocene were ever being converted into water about their margins. Their limits on the land were the lines where their onward flow was just balanced by melting and evaporation. On the surface of the ice along the marginal zone, rivulets no doubt flowed in summer, and found their way through creva.s.ses to the interior of the glacier or to the ground. Subglacial streams, like those of the Malaspina glacier, issued from tunnels in the ice, and water ran along the melting ice front as it is seen to do about the glacier tongues of Greenland. All these glacier waters flowed away down the chief drainage channels in swollen rivers loaded with glacial waste.

It is not unexpected therefore that there are found, over all the country where the melting ice retreated, deposits made of the same materials as the till, but sorted and stratified by running water.

Some of these were deposited behind the ice front in ice-walled channels, some at the edge of the glaciers by issuing streams, and others were spread to long distances in front of the ice edge by glacial waters as they flowed away.

ESKERS are narrow, winding ridges of stratified sand and gravel whose general course lies parallel with the movement of the glacier. These ridges, though evidently laid by running water, do not follow lines of continuous descent, but may be found to cross river valleys and ascend their sides. Hence the streams by which eskers were laid did not flow unconfined upon the surface of the ground. We may infer that eskers were deposited in the tunnels and ice-walled gorges of glacial streams before they issued from the ice front.

KAMES are sand and gravel knolls, a.s.sociated for the most part with terminal moraines, and heaped by glacial waters along the margin of the ice.

KAME TERRACES are hummocky embankments of stratified drift sometimes found in rugged regions along the sides of valleys. In these valleys long tongues of glacier ice lay slowly melting.

Glacial waters took their way between the edges of the glaciers and the hillside, and here deposited sand and gravel in rude terraces.

Outwash plains are plains of sand and gravel which frequently border terminal moraines on their outward face, and were spread evidently by outwash from the melting ice. Outwash plains are sometimes pitted by bowl-shaped basins where ice blocks were left buried in the sand by the retreating glacier.

Valley trains are deposits of stratified drift with which river valleys have been aggraded. Valleys leading outward from the ice front were flooded by glacial waters and were filled often to great depths with trains of stream-swept drift. Since the disappearance of the ice these glacial flood plains have been dissected by the shrunken rivers of recent times and left on either side the valley in high terraces. Valley trains head in morainic plains, and their material grows finer down valley and coa.r.s.er toward their sources. Their gradient is commonly greater than that of the present rivers.

THE EXTENT OF THE DRIFT. The extent of the drift of North America and its southern limits are best seen in Figure 359. Its area is reckoned at about four million square miles. The ice fields which once covered so much of our continent were all together ten times as large as the inland ice of Greenland, and about equal to the enormous ice cap which now covers the antartic regions.