Popular Science Monthly/Volume 34/November 1888/Paleolithic Man in America: His Antiquity and Environment
By W J McGEE,
OF THE U. S. GEOLOGICAL SURVEY.
DURING the unlettered youth of the race there were no written records from which the antiquity of man can be read. So the anthropologist on the one hand, and the geologist on the other, have sought to construct an early human history from prehistoric relics, and from the formations in which they are imbedded or the fossils with which they are associated. Lubbock divided prehistoric time into four great epochs, viz.: 1. The Paleolithic or rough stone epoch, during which primitive man flourished; 2. The Neolithic or polished stone epoch, during which higher development was reached; 3. The Bronze age; and 4. The Iron age. These divisions, at first supposed to represent successive eras, are now regarded as representing cultural phases rather than periods of time (in fact, all are found among the present population of the world), and they are accordingly valueless as measures of the antiquity of man upon the globe. The geologist classifies later geologic time as Cenozoic or the era of modern life, divides it into Tertiary and Quaternary (or Pleistocene), and subdivides the Tertiary into Eocene, Miocene, and Pliocene. But no part of the geologic record, as hitherto interpreted, is more indefinite than that of the transition from Tertiary to Quaternary, or from Pliocene to Pleistocene; and this indefiniteness is especially unfortunate for American anthropology, since it was about this period that the autochthon—the primeval inhabitant of the continent—first appeared. It is, indeed, customary to recognize the geologically recent glacial period, during which northern United States was overspread by an ice-sheet extending southward to the thirty-eighth parallel, as the initial episode of the Quaternary; but it is becoming apparent that this period is too long and too vaguely defined to satisfy inquirers for a date of man's origin.
Recent researches in the Great Basin of western America, in the Mississippi Valley, and in the Atlantic slope, have shown (1) that the glacial period consisted of two epochs of humid climate and glaciation (the later comprising two or more sub-epochs); (2) that these cold and wet epochs were preceded, separated, and followed by climatal conditions much like those of to-day; (3) that the intervening a-glacial epoch was of considerable duration; and (4) that the earlier epoch, of cold was the longer, though the cold was the more intense and the climate more variable during the later epoch.
The latest researches in the Atlantic slope have shown more definitively (1) that the interval of mild climate separating the two cold epochs was from five to ten times as long as the post-glacial interval; (2) that the cold epochs themselves were brief in comparison with the inter-glacial and post-glacial intervals; and (3) that the earlier and longer epoch of cold was attended by continental submergence reaching four or five hundred feet in the latitude of New York and extending to South Carolina, while the land depression of the later refrigeration was but forty or fifty feet at New York, and scarcely extended beyond the great terminal moraine of Long Island and northern New Jersey.
During the first epoch of cold and wet local glaciers formed in the Rocky Mountains and in the Sierras, the Great Basin (which these ranges bound) was flooded and the now extinct Lakes Bonneville and Lahontan were formed, and into the lakes great volumes of sand and silt—the lower lacustral beds of Gilbert and Russell—were swept by the flooded rivers; at the same time the northern ice-sheet stretched down into the Mississippi Valley as far as the Missouri River, the land was depressed, and both glacial and aqueo-glacial deposits were laid down; and it was at the same time, too, that the Atlantic coast was depressed until the high hills overlooking New York, Philadelphia and Washington were half submerged, and that the rivers built great deltas of gravel and loam along the shore of the expanded ocean, while the waves dropped shallow-water sediments all over the lowlands. With the interglacial warmth the glaciers of the Western mountains were melted, the lakes were dried, and river-gravels were deposited by the shrunken streams over and canons were cut into the old lake-bottoms; in the Mississippi Valley the glaciers retreated and the drift-plains became forest-covered; while in the East the land underwent re-elevation, and there was erosion of such extent as to afford a rough measure of the duration of the warm interval. During the later epoch of cold and wet glaciers again formed in the Rocky Mountains and in the Sierras, Lakes Bonneville and Lahontan were refilled—the former to overflowing—and the upper lacustral beds were laid down within them; the northern ice again invaded the Mississippi Valley and formed two or more drift-sheets, together with the peculiar glacial-mud deposit (or loess) into which they graduate, as well as the great terminal moraine stretching from Ohio to Dakota; and in the East the ice again overrode the Adirondacks and the New England ranges, crept southward to Long Island and northern New Jersey, and heaped up the eastern extension of the terminal moraine, and, as it melted, gave origin to the Champlain deposits of the New England rivers and to certain distinctive aqueo-glacial gravels, of which those of the Delaware at Trenton are the type.
The lower lacustral deposits of the Great Basin, the aqueo-glacial deposits of the Mississippi Valley, and the ancient deltas of the Atlantic slope are correlated, partly because (1) each attests a great and similar climatal episode, because (2) it is evident that each of these episodes was so extreme as to affect the entire breadth of the continent, and because (3) there are no indications among American geologic deposits of other episodes with which these might be confused; the upper lacustral beds of the Great Basin, the upper glacial and aqueo-glacial deposits of the Mississippi Valley, and the glacial deposits of the Atlantic slope, are correlated upon similar grounds; and the harmony among the various records gives cumulative proof of the accuracy of each.
By these researches of the last decade the earlier conceptions of Quaternary history are greatly expanded, and the hitherto obscure relation between the Tertiary and Quaternary is made clear. Where they contain vertebrate fossils, the earlier and even the later of these deposits are, it is true, referred to the Pliocene by paleontologists, while physical geologists refer the entire series to the Quaternary; but thisis one of classification only, and in no way affects the phenomena classified.
The sequence of events made out independently in the three widely separated regions may be depicted as in the accompanying diagram (Fig. 1), representing the temperatures and land-altitudes
Fig. 1.—Graphic Representation of Climate and Changes in Altitude of the Land.
during late Tertiary and Quaternary time. By such a chronograph alone is it possible to accurately measure the antiquity of paleolithic man. Marsh has well shown that only long æons can be measured by plant fossils, that somewhat shorter periods may be measured by the records of invertebrate life, and that the vertebrates afford by far the most delicate of the paleontologic time-measures; but the swing of even the vertebrate life-pendulum is long as compared with, that marking the transitory stages of human development, and must give place to the still more precise chronometer afforded by the brief and sharply defined climatal episodes of later geologic time; the human records of diverse regions can only be correlated in terms of these brief episodes; and in ascertaining the relations of paleolithic man to the two best known climatal episodes of the past, it is immaterial whether these be called Tertiary or Quaternary. It is the special merit of the graphic method that it exhibits quantitative relations (for while verbal language is commonly qualitative, graphic language is always quantitative); and in the present case it affords a means of measuring the consistency of the evidence, and of instantly detecting the inconsistent records, of human antiquity.
There are several well-authenticated discoveries of human relics in this country in geologic deposits whose places may be fixed in the graphic time-record forming Fig. 1. Aughey records two chipped implements from the loess of the Missouri Valley, one of them coming from immediately beneath an elephantine vertebra; Miss Babbitt has found great numbers of quartz-chips in a Champlain terrace of the Mississippi at Little Falls, Minnesota, which are regarded by many archæologists as unquestionably artificial; N. H. Winchell records polished stone and copper implements as well as human bones from the same aqueo-glacial terrace of the Mississippi near Minneapolis; Belt several years ago found a fossilized human skull in what appears to be the westernmost extension of the loess in Colorado; Gilbert has shown that the geologic position of an ancient hearth found in excavating a well in northern New York indicates that it was constructed during the closing episodes of the last glacial epoch; a few years since McGee discovered a chipped obsidian implement imbedded in the upper lacustral marls of western Nevada; McAdams notes the finding of a stone axe in loess seventy feet beneath the surface in Illinois; and among the most recent and satisfactory archæologic discoveries of this country are those of two chipped implements of black flint found in Ohio by Dr. C. G. Metz, at Madisonville and Lovelands respectively, in deposits of loess and aqueo-glacial gravel which G. F. Wright has shown to represent a closing episode of the later glacial epoch. But it is in the aqueo-glacial gravels of the Delaware River at Trenton, which were laid down contemporaneously with the terminal moraine one hundred miles farther northward, and which have been so thoroughly studied by Abbott, that the most conclusive proof of the existence of glacial man is found; and it is here, too, that the most satisfactory evidence is obtained concerning the conditions by which paleolithic man was surrounded. It is significant that in all these cases the human relics were found in deposits representing the closing episodes of the later epoch of Quaternary cold.
There are several cases in which traces of human activity have been reported from older deposits, but in which the discoveries are not so well authenticated. E. g., there is Dr. Koch's well-known record of the finding of mastodon remains in the Osage Valley, Missouri, associated with human implements and traces of fire, in deposits probably contemporaneous with those of the earlier ice-sheet; but the geologic relations have never been clearly made out, and the verity of the discovery has always been (perhaps unjustly) questioned. The finding of a fossilized human bone at Natchez, Mississippi, apparently associated with an early Quaternary fauna, is equally well known; but the attendant circumstances were not such as to carry conviction to the minds of contemporary students. Lewis, also, has described a paleolithic implement from aqueo-glacial gravels at Philadelphia; but he did not personally witness the discovery, and was not certain that the object came from the older (earliest glacial) and not the newer (latest glacial) gravels. It is significant that in all these cases the testimony is internally defective; and, since its acceptance would many times multiply human antiquity as established by collateral evidence (as clearly shown in Fig. 1), it would seem especially wise to reserve judgment upon it.
There are other cases in which human remains have been found in such position as to indicate great antiquity measured in years—e. g., the shell-heaps of Damariscotta River in Maine and St. John's River in Florida, representing a fauna now extinct or displaced; the enormous shell-heap at San Pablo on the Bay of San Francisco, which evidently represents a vast period of building; the shell-beds and superimposed deposits of the Aleutian Islands, which have been shown by Dall to represent at least three thousand years of accumulation, etc.; but in none of these cases is it possible to reduce the historic time-units to definite geologic time-units.
There are still other cases in which human relics have been reported from deposits of considerable geologic age—e. g., in Calaveras County, California, near Golden, Colorado, etc.; but, while the archæologic evidence would seem conclusive in at least one of these instances, it is impossible to confidently transmute the paleontologic record of the age of the deposits into the physical record which alone is sufficiently refined for the measure of human development; and it would thus seem wise to reserve judgment in these cases, also, with respect to the correlation of the deposits as well as to the association of the relics.
Excluding all doubtful cases, there remains a fairly consistent body of testimony indicating the existence of a widely distributed human population upon the North American Continent during the later ice epoch. The records are not equally decisive, it is true; the artificial character of Miss Babbitt's quartz-chips has been questioned, and they represent a stage of culture widely different from that represented by Winchell's objects from the same deposit; it has been suggested that McAdams's axe may have been an adventitious inclusion; Belt's untimely death prevented final statement of the geologic position of the Colorado skull; the apparently conclusive structural evidence of the antiquity of the Nevada obsidian is opposed by its fresh aspect and modern form; and Gilbert's hearth was not seen by the geologist who studied its relations. Yet, however the doubtful cases may be weighted, the testimony is cumulative, parts of it are unimpeachable, and the proof of the existence of glacial man seems conclusive. But the evidence of man's existence during the earlier epoch of glacial cold is not conclusive; and the evidence of still earlier human occupancy of the continent is not reducible to the terms of definite geologic chronology. Moreover, there is a body of negative evidence which is worthy of consideration. The lower lacustral deposits of the Great Basin have been as carefully explored as the upper, but have yielded no trace of human remains; the oldest glacial and aqueo-glacial deposits of the Mississippi Valley have been explored in Nebraska, Illinois, Missouri, and Ohio, as carefully as the later deposits, but (if Dr. Koch's famous find be excluded) no traces of human occupancy have been found; and, most significant of all, the deposits of the earlier cold epoch throughout the District of Columbia have been scanned for years by a dozen trained collectors and not a single object of human manufacture has been found within them, though thousands have been found on the surface, and though it might be shown that the conditions for savage life were as favorable on the Potomac during this epoch as they were on the Delaware during the later.
The various archæologic discoveries of America display striking diversity in the degree of development exhibited in the relics, which range from the rudest "turtle-backs" to finely chipped flint, polished stone, and even copper; but whether this disparity indicates adventitious inclusion in certain cases—and thus weakens the chain of evidence of human antiquity—or heterogeneity in the primitive population, can not yet be decided. Whatever interpretation be placed upon the questionable cases, however, there is convincing proof not only of man's existence, but of the definite stage of culture called paleolithic, in the later cold epoch of the glacial period. It is indeed obvious that the autochthon must have found birth anterior to this epoch, but the objective evidence of pre-paleolithic art has not been ascertained; and, since the date of origin of a higher culture is unknown, it can only be said that the paleolithic stage began toward the close of the later cold epoch and extended well toward the historic period, probably overlapping far upon the neolithic stage. Thus the place of paleolithic man in the chronograph afforded by geology is that shown in Fig. 1.
It should be pointed out that the human period of America can not be synchronized with that of Europe, since the geologic chronometer employed abroad is not sufficiently sensitive. It is true that Penck and others have recently read from the glacial and associated deposits of the Alps a climatal record coinciding exactly with that recognized in this country (save that the duration of the episodes is less closely measured), and that Mortillet and others have inferred a definite climatal sequence from the
Conspectus of Quaternary History.
relic-bearing deposits of central France; but, as shown in the accompanying table, the records are not accordant in their entirety, and can only be reduced to common terms by juxtaposing the earliest recognized Quaternary episode of the lowlands with one of the episodes of the later Quaternary in the mountains. This allocation harmonizes the evidence as to the antiquity of man on opposite sides of the Atlantic, but runs counter to current opinion and appears inconsistent with certain cavern phenomena, and can therefore be set forth as only a possible one. In this as in other cases, paleontologic correlation is incompetent if not utterly meaningless, since the episodes dealt with were so brief that chorologic diversity among the higher animals was unquestionably more important than chronologic variation—indeed, the latest lacustral (and relic-bearing) deposits of the Great Basin, which are referred to the Pliocene upon paleontologic grounds, appear to have an older facies than the oldest relic-bearing river-deposits of France.
The chipped implements found by Aughey appear to have been dropped on the bottom of the shallow lake or muddy swamp within which the loess was accumulated; since the loess itself consists of glacial mud, and since the basin in which it was deposited was bounded on the north by the Quaternary mer de glace, the climate must have been cold; and the associated elephantine remains prove the association of man and mammoth. The relics themselves throw little light upon the habits of their makers, but suggest that they were well advanced in the fabrication of chipped implements. If the obsidian implement from the Nevada lake-beds was really in situ (as all appearances indicated), it must have been dropped in a shallow and quiet bay of the saline and alkaline lake Lahontan, and gradually buried beneath its fine mechanical sediments and chemical precipitates; as indicated by the associated fossil bones and teeth, its makers must have been contemporary with the indigenous horse, an elk or deer, an elephant or mastodon, the camel, a gigantic ox, and other extinct animals commonly referred to the later Pliocene; but the single implement tells little of the habits and customs of the people it represents, save that they had advanced far in the art of stone-chipping. Gilbert's hearth was located on the southern shore of Lake Ontario, when it was greatly expanded by continental tilting and obstruction of its present outlet by the later Quaternary glacier, and was buried beneath lacustral deposits when further tilting of the land altered the position of the lake-shore; and since the lake was confined on the north and east by the mer de glace, the temperature of the times must have been low and the surface of the water dotted by floes and icebergs in spring-time, if not all the year round. The hearth itself tells only that its builders knew the uses of fire, and constructed rude fireplaces, but is silent as to their knowledge of water-craft, as to their implements and utensils, as to whether they were hunters or fishermen, and so as to nearly all of their habits and customs. Miss Babbitt's quartz-chips appear (though the geologic relations are somewhat obscure) to represent the site of a primitive workshop or rendezvous on the banks of a river heading in the ice-sheet a few miles or scores of miles up-stream. The artificial origin of the chips has been disputed, and is indicated by their concentration in a certain local stratum and their absence from contiguous strata and other localities rather than by their form—the distribution being apparently explicable only on the hypothesis that they were artificially accumulated, whether or not they were artificially fabricated. The rude chips throw no light on the habits, customs, or environment of the men
Fig. 2.—Margin of the later Quaternary Ice-Sheet in the Valley of Delaware River.
by whom they may have been fashioned, save that, if artificial, they exhibit the lowest known grade of culture; but this testimony of the quartz-chips is apparently antagonized by that of the polished-stone axe and disk, the copper spear-head, etc., recorded by N. H. Winchell from another part of the same terrace-plain.
The deposit in which the Madisonville implement was found by Metz was laid down by a turbulent ice-bearing river but a few miles from the glacier's margin. The implement, unlike those recorded by Aughey, McGee, and N. H. Winchell, is of the rude type commonly called paleolithic, and thus indicates primitive customs among its makers; but neither alone nor in conjunction with the similar implement found by the same individual under like conditions at Lovelands does it tell whether the inhabitant of the ice-front in the Ohio valley was hunter, fisherman, or husbandman, troglodyte, nomad, or house-builder; and only the geologic evidence suggests conditions of life approaching those of the modern Esquimau.
When the primitive man of Trenton flourished, the later Quaternary mer de glace covered New York and New England, and extended far into northern New Jersey and Pennsylvania. The ice was from five hundred to one thousand feet in thickness near its margin, and overflowed the highest mountains, though they somewhat impeded its progress; the land beneath was somewhat depressed and was tilted northward toward the ice-front; and flooded rivers, born upon and beneath the edge of the icesheet, swept into their lower courses and into the sea, glacial mud, sand, and pebbles, while upon their surfaces floated ice-floes laden sometimes with larger pebbles and anon with great bowlders. Among these rivers was the Delaware, which was transformed in its middle course from a constricted torrent rushing swiftly over a rocky bottom (as it is to-day and as it was anterior to the second ice-epoch) into a broad slack-water estuary, tidal probably to the mouth of the Lehigh. This estuary found its source at the edge of the ice, where now lies the terminal moraine (just below Belvidere); and at what is now the head of tide at Trenton it embouched into a broad, shallow bay. At the ice-front it gathered a harvest of cobble-stones which were washed down-stream and deposited in a series of terraces more than one hundred feet in height and two miles or more in width, extending ten miles down the river; the cobbles growing finer and finer and finally passing into beds of gravel and sand. There, too, the waters became charged with glacial mud—the rock-flour forming the grist of the glacial mill—which was more slowly deposited in the form of fine loam sometimes enveloping the cobbles and abundantly intermixed with the finer gravel and sand as far south as Philadelphia, but most abundantly above Trenton. There, also, the river gathered sand, fine and coarse pebbles, great bowlders, and heterogeneous débris, which were frozen into ice-floes, floated gently down-stream, and dropped as the floes melted, equally far southward, but most abundantly where the river embouched into the bay and where the floes lingered longest in the slackened current, These aqueo-glacial deposits extend continuously from the terminal moraine to Philadelphia. They are most conspicuous in the great gravel terrace just south of Belvidere, gradually diminish in volume and height and even merge into the modern alluvium by which they are in part overlaid between Easton and Trenton, become conspicuous again at Trenton (where they cover an area of fully fifty square miles, and are exposed in every natural and artificial excavation below their maximum altitude in and near the city), and finally disappear near Bristol, though the cobbles are largely dredged from the channel to and beyond Philadelphia. They are in part overlaid by modern alluvium, into which they appear to merge midway between the moraine and Trenton; and they repose unconformably upon the greatly eroded surface of the Columbia formation—the aqueo-glacial deposits of the earlier cold epoch of the Quaternary—notably at Trenton, where they fill a basin lined with Columbia brick-clays and gravels.
By structure, composition, and topographic relations the Fig. 3.—Artificial Cliff of Trenton Gravels. deposits tell the story, as by their geologic relations they fix the date, of their origin. At Trenton the deposits consist of stratified gravels more heterogeneous than, but otherwise undistinguishable from, those of the terraces into which the terminal moraine merges, interspersed with bowlders up to one hundred cubic feet in volume, the whole imbedded in a matrix of sand and loam. The entire mass is unquestionably water-laid; its continuous bedding is indicative of wave-action, and thus of shallow waters; and the bowlders scattered throughout it are evidently ice-borne. Its structure is shown in Fig. 3, reproduced from a photograph taken in the extensive gravel-pit half a mile east of the depot at Trenton. The relations of these gravels to the subjacent Columbia formation are shown in Fig. 4, also reproduced from a photograph taken at Chambersburg—the coarse, stratified gravels. bowlders, etc., representing the later deposit, and the homogeneous loam passing downward into coarse gravels representing the older formation. The thickness of the deposit ranges from a trifling veneer to forty feet or more; and its surface, where it has Fig. 4.—Trenton Gravels Lying upon Columbia Loam and Gravels. escaped erosion, forms a plain inclined gently southward from an altitude of about forty-five feet in north Trenton to tide-level midway between Bristol and Philadelphia; this inclination of the deposit being the measure of the northward tilting of the land during the later ice epoch.
Within the Trenton gravels two types of implement are found—viz., "turtle-backs" and the rude "leaf-shaped" implements regarded by Abbott as of Esquimau pattern. Both types are chipped from a peculiar argillite which is found in the deposit only as (presumptively) finished implements or as large bowlders. The implements, which occur in such numbers that over twenty-five thousand have been collected by Abbott, are seldom water-worn though frequently weathered, while the bowlders are somewhat worn by water and similarly weathered. It is significant that the "turtle-back" type is found throughout the deposit from top to bottom, but most abundantly in the lower half and in progressively diminishing abundance from bottom to top of the upper half, while the "leaf-shaped" type is found only in the upper half and in progressively increasing abundance upward; it is also noteworthy that both types of implement are occasionally found over contiguous surfaces of the Columbia formation (which were above water-level when the Trenton gravels were deposited), commonly associated with chipped implements of higher type; and it is equally noteworthy that the implements of higher type occur over the surface of the Trenton gravels but never within them, while the ruder implements found within the gravels do not occur upon the gravel surface.
The geologic evidence of the environment of the celt-users of the Delaware River is complete and intelligible: the continuity of the deposits proves that the Trenton gravels were laid down while the northern part of the country was occupied by ice, and while
Fig. 5—Stereogram of the Present Head of Delaware Bay.
the terminal moraine was forming; their structure and composition prove that they were laid down in shallow floe-bearing waters; their distribution indicates unmistakably the geography of the period; and this physical testimony is corroborated by the association of remains of extinct or boreal animals (mastodon, reindeer, bison, etc.) with the human relics. So definite, indeed, are the data, that the geography of the period may be depicted
Fig. 6.—Stereogram of the Head of Delaware Bay in Late Quaternary Times
graphically. Fig. 5 is a bird's-eye sketch of the present head of Delaware Bay, based upon the admirable topographic surveys of New Jersey and upon personal studies, and represents with reasonable accuracy the general features of the region; and Fig. 6 is a similar sketch of the region as it existed during the Trenton gravel period, based upon the same surveys and upon the shore-lines brought to light thereby, and probably represents the configuration of the earlier period with equal fidelity.
Fig. 7.—Stereogram of the Head of Chesapeake Bay
Fortunately, the Delaware Bay of the later Quaternary has an existing homologue by which conceptions of the local physiography and the attendant biotic conditions may be rendered tangible—viz., the head of Chesapeake Bay, shown in Fig. 7. This estuary is broad and shallow, as was the Delaware estuary during the glacial period; it is gorged in spring-time with ice-floes formed within its own area and swept into it by its great affluent, as was the ancient Delaware throughout a longer and more rigorous winter; and its bottom is a submerged terrace-plain of loam, sand, and gravel differing from the Trenton gravel only in the less proportion of ice-borne materials within it. In the shoal Susquehanna-Chesapeake estuary grow a great variety of aquatic plants, harboring multitudes of minute animals, which together furnish abundant food for fish and water-fowl, and just as it is now among white men a far-famed fishing and hunting ground, so it was a notable resort of the aborigines, as attested by the village sites about its shores; and since its shallow waters may be waded over half its area and the simplest water-craft outlives the low billows of its storms, the primitive spear-head and stone sinker doubtless underlie the cartridge-shell and leaden sinker of the present, just as the "turtle-back" of Trenton underlies the finely chipped flint of the surface. During the later ice epoch of the Quaternary the climate of the Delaware estuary was less tolerable than that of the present Chesapeake estuary, but other conditions were more favorable to concentration of piscine, avian, and human life within and about it. There the river-breeding fishes were stopped in their instinctive ascent toward former spawning-grounds to increase their kind; there the migratory birds must have ended their vernal journeyings to nest and hatch; there also the flora, forced southward before the advancing ice, must have grown mixed and varied; there the land fauna, pressed by the northern cold and attracted by the forage and carnage, must have lingered and multiplied; and there primitive men must have congregated and dominated over all. It is true but not surprising that the fragile remains of fish, fowl, plants, or even human bones have never been found in the porous and thoroughly leached Trenton gravels, associated with the implements and the more massive bones of mastodon, bison, and reindeer; but the locality was as distant from the ice-front as the arctic breeding-grounds of to-day, and moderately mild climate is attested by the wonderfully abundant implements and the numerous population they represent.
The artificial origin of the "turtle-backs" has been questioned, and their abundance has been regarded as proof of their natural origin; and it is therefore not a work of supererogation to point out that the Trenton gravels are largely wrought for railway ballast, and have been scanned by the thousand tons by eager workmen with the hope of reward before their eyes, and to repeat that the argillite of which the implements are fashioned rarely occurs in the deposit in the form of natural pebbles. Of any hundred bits of argillite selected at random from the gravel-bank or the collection, ten per cent to thirty per cent exhibit unmistakable traces of design, a somewhat larger percentage suggest but do not prove design, and not more than fifty per cent strike the student as natural, when the individual specimens are examined separately; and when examined collectively the correspondence in form and mode of fracture between symmetric "turtle-backs," "failures," "spawls," "chips," and miscellaneous fragments compels the cautious geologist to question whether any are demonstrably or even probably natural: the series is not from the certainly natural to the doubtfully artificial, but from the certainly artificial to the doubtfully natural.
The "turtle-backs" tell nothing of the customs of the makers, since their function is unknown; whether they were sinkers for nets, whether they were hammers or axes used either in the hand or attached to withes or handles, whether they were used as the bolas of the Patagonians, whether they were employed in fishing first for cutting the ice and then for eviscerating and scaling the fish, or whether they subserved a variety of purposes, remains undetermined. The environment read from geology indicates that the Trenton man was a hunter or fisherman who used and lost the primitive tools of his mysterious craft within the waters rather than upon the land, and thus appears to materially narrow the range of hypothesis as to his activities; but the extravagance in labor indicated by the vast numbers of unworn implements suggests that the rapid modification in environment and occupation accompanying the ice-invasion outran the resulting modification in appliances, and that the implements were really invented on land and were but ill adapted to the new conditions; and the introduction of a new type of implement during the brief epoch of gravel deposition gives support to the suggestion.
So the margin of the cloud enveloping the beginnings of human life in America is slowly lifting. Already there is definite and cumulative evidence of man's existence during the latest ice epoch, with a strong presumption against an earlier origin than the first Quaternary ice-invasion; already it is known that the primitive American haunted the ice front rather than the fertile plain, and must have been hunter or fisherman; already his environment is so well known as to partially elucidate his activities; but the first traces of the autochthon yet found tell of an intelligent being who dominated the animal world as does his descendant, and thus the mystery of man's ultimate origin remains enshrouded as darkly as ever.