The result of the writer’s studies upon petroglyphs as distinct from other forms of picture writing may now be summarized.

Perhaps the most important lesson learned from these studies is that no attempt should be made at symbolic interpretation unless the symbolic nature of the particular characters under examination is known, or can be logically inferred from independent facts. To start with a theory, or even a hypothesis, that the rock writings are all symbolic and may be interpreted by the imagination of the observer or by translation either from or into known symbols of similar form found in other regions, were a limitless delusion. Doubtless many of the characters are genuine symbols or emblems, and some have been ascertained through extrinsic information to be such. Sometimes the more modern forms are explained by Indians who have kept up the pictographic practice, and the modern forms occasionally throw light upon the more ancient. But the rock inscriptions do not evince mysticism or esotericism, cryptography, or steganography. With certain exceptions they were intended to be understood by all observers either as rude objective representations or as ideograms, which indeed were often so imperfect as to require elucidation, but not by any hermeneutic key. While they often related to religious ceremonies or myths, such figures were generally drawn in the same spirit with which any interesting matter was portrayed.

While the interpretation of petroglyphs by Indians should be obtained if possible, it must be received with caution. They very seldom know by tradition the meaning of the older forms, and their inferences are often made from local and limited pictographic practices. There is no more conscientious and intelligent Indian authority than Frank La FlÊche, an Omaha, and he explains the marks on a rock in Nebraska as associated with the figures of deceased men and exhibiting the object which caused their death, such as an arrow or ax. This may be a local or tribal practice, but it certainly does not apply to similar figures throughout the Algonquian and Iroquoian areas, where, according to the concurrent testimony for more than two centuries, similar figures are either designations of tribes and associations, or in their combinations are records of achievements.

Lossing (b) gives the following explanation of markings on a well known rock:

This tale may be true, but it surely does not account for the turkey-foot marks which are so common in the northeastern Algonquian region, extending from Dighton rock to Ohio, that they form a typical characteristic of its pictographs. They have been considered to be the sign for the bird, the turkey, which was a frequent totem. Lossing’s story is an example of the readiness of an Indian, when in an amiable and communicative mood, to answer queries in a manner which he supposes will be satisfactory to his interviewer. He will then give any desired amount of information on any subject without the slightest restriction by the vulgar bounds of fact. It is dangerous to believe explanations on such subjects as are now under consideration, unless they are made without leading questions by a number of Indian authorities independently.

Specially convenient places for halting and resting on a journey, either by land or water, such as is mentioned supra, on Machias bay, generally exhibit petroglyphs if rocks of the proper character are favorably situated there. The markings may be mere graffiti, the product of leisure hours, or may be of the more serious descriptions mentioned below.

Some points are ascertained with regard to the motives of the painters and sculptors on rocks. Some of the characters were mere records of the visits of individuals to important springs or to fords on regularly established trails. In this practice there may have been in the intention of the Indians very much the same spirit which induces the civilized man to record his name or initials upon objects in the neighborhood of places of general resort. But there was real utility in the Indian practice, which more nearly approached to the signature in a visitor’s book at a hotel or public building, both to establish the identity of the traveler and to give the news to friends of his presence and passage. At Oakley springs, Arizona territory, totemic marks have been found, evidently made by the same individual at successive visits, showing that on the number of occasions indicated he had passed by those springs, probably camping there, and the habit of making such record was continued until quite recently by the neighboring Indians. The same repetition of totemic names has been found in great numbers in the pipestone quarries of Minnesota, on the rocks near Odanah, Wisconsin, and also at some old fords in West Virginia. These totemic marks are so designed and executed as to have intrinsic significance and value, wholly different in this respect from names in alphabetic form, which grammatically are proper but practically may be common.

Rock carvings are frequently noticed at waterfalls and other points on rivers and on lake shores favorable for fishing, which frequency is accounted for by the periodical resort of Indians to such places. Sometimes they only mark their stay, but occasionally there also appear to be records of conflict with rival or inimical tribes which sought to use the same waters.

Evidence is presented in the present work that the characters on rock pictures sometimes were pointers or “sign-posts” to show the direction of springs, the line of established trails, or of paths that would shorten distances in travel. It has been supposed that similar indications were used guiding to burial mounds and other places of peculiar sanctity or interest, but the evidence of this employment is not conclusive. Many inquiries have been made of the Bureau of Ethnology concerning Indian marks supposed to indicate the sites of gold, silver, and copper mines and buried treasure generally, which inquiries were answered only because it was recognized as the duty of an office of the government to respond, so far as possible, to requests for information, however silly, which are made in good faith.

Petroglyphs are now most frequently found in those parts of the world which are still, or recently have been, inhabited by savage or barbarian tribes. Persons of these tribes when questioned about the authorship of the rock drawings have generally attributed them to supernatural beings. Statements to this effect from many peoples of the three Americas and of other regions, together with the names of rockwriting deities, are abundantly cited in the present work. This is not surprising, nor is it instructive, except as to the mere fact that the drawings are ancient. Man has always attributed to supernatural action whatever he did not understand. Also, it appears that in modern times shamans have encouraged this belief and taken advantage of it to interpret for their own purposes the drawings, some of which have been made by themselves. But notwithstanding these errors and frauds, a large proportion of the petroglyphs in America are legitimately connected with the myths and the religious practices of the authors. The information obtained during late years regarding tribes such as the ZuÑi, Moki, Navajo, and Ojibwa, which have kept up on the one hand their old religious practices and on the other that of picture writing, is conclusive on this point. The rites and ceremonies of these tribes are to some extent shown pictorially on the rocks, some of the characters on which have until lately been wholly meaningless, but are now identified as drawings of the paraphernalia used in or as diagrams of the drama of their rituals. Unless those rituals, with the creeds and cosmologies connected with them had been learned, the petroglyphs would never have been interpreted. The fact that they are now understood does not add any new information, except that perhaps in some instances their age may show the antiquity and continuity of the present rites.

A potent reason for caution in making deductions based only on copies of figures published incidentally in works of travel is that it can seldom be ascertained with exactness what is the true depiction of those figures as actually existing or as originally made. The personal equation affects the drawings and paintings intended to be copies from the rock surfaces and also the engravings and other forms of reproductions, and the student must rely upon very uncertain reproductions for most of his material. The more ancient petroglyphs also require the aid of the imagination to supply eroded lines or faded colors. Travelers and explorers are seldom so conscientious as to publish an obscure copy of the obscure original. It is either made to appear distinct or is not furnished at all, and if the author were conscientious the publisher would probably overrule him.

Thorough knowledge of the historic tribes, including their sociology, sophiology, technology, and especially their sign language, will probably result in the interpretation of many more petroglyphs than are now understood, but the converse is not true. The rock characters studied independently will not give much primary information about customs and concepts, though it may and does corroborate what has been obtained by other modes of investigation. A knowledge of Indian customs, costumes, including arrangement of hair, paint, and all tribal designations, and of their histories and traditions, is essential to the understanding of their drawings; for which reason some of those particulars known to have influenced pictography have been set forth in this work and objects have been mentioned which were known to have been portrayed graphically with special intent.

Other objects are used symbolically or emblematically which, so far as known, have never appeared in any form of pictographs, but might be found in any of them. For instance, Mr. Schoolcraft says of the Dakotas that “some of the chiefs had the skins of skunks tied to their heels to symbolize that they never ran, as that animal is noted for its slow and self-possessed movements.” This is one of the many customs to be remembered in the attempted interpretations of pictographs. The present writer does not know that a skunk skin or a strip of skin which might be supposed to be a skunk skin attached to a human heel has ever been separately used pictorially as the ideogram of courage or steadfastness, but with the knowledge of this objective use of the skins, if they were found so represented pictorially, the interpretation would be suggested without any direct explanation from Indians.

A partial view of petroglyphs has excited hope that by their correlation the priscan homes and migrations of peoples may be ascertained. Undoubtedly striking similarities are found in regions far apart from as well as near to each other. A glance at the bas-reliefs of Boro Boudour in Java, now copied and published by the Dutch authorities, at once recalls figures of the lotus and urÆus of Egypt, the horns of Assyria, the thunderbolt of Greece, the Buddhist fig tree, and other noted characters common in several parts of the world. If the petroglyphs of America are considered as the texts with which all others may be compared, it is believed that the present work shows illustrations nearly identical with many much-discussed carvings and paintings on the rocks of the eastern hemisphere, those in Siberia being most strongly suggestive of connection. But from the present collection it would seem that the similarity of styles in various regions is more worthy of study than is the mere resemblance or even identity of characters, the significance of which is unknown and may have differed in the intent of the several authors. Indeed it is clear that even in limited areas of North America, diverse significance is attached to the same figure and differing figures are made to express the same concept.

The present work shows a surprising resemblance between the typical forms among the petroglyphs found in Brazil, Venezuela, Peru, Guiana, part of Mexico, and those in the Pacific slope of North America. This similarity includes the forms in Guatemala and Alaska, which, on account of the material used, are of less assured antiquity. Indeed it would be safe to include Japan and New Zealand in this general class. In this connection an important letter from Mr. James G. Swan, respecting the carved wooden images of the Haidas, accentuates the deduction derived only from comparison. Mr. Swan says that he showed to the Indians of various coast tribes the plates of Dr. Habel’s work on sculptures in Guatemala, and that they all recognized several of the pictures which he notes. They also recognized and understood the pictures of the ZuÑi ceremonials, masks, and masquerades scenes published by Mr. F. H. Cushing.

Without entering upon the discussion whether America was peopled from east to west, or from either, or from any other part of the earth, it is for the present enough to suggest that the petroglyphs and other pictographs in the three Americas indicate that their pre-Columbian inhabitants had at one time frequent communication with each other, perhaps not then being separated by the present distances of habitat. Styles of drawing and painting could thus readily be diffused, and, indeed, to mention briefly the extralimital influence, if as many Japanese and Chinese vessels were driven upon the west American coast in prehistoric times as are known by historic statistics to have been so driven, the involuntary immigrants skilled in drawing and painting might readily have impressed their styles upon the Americans near their landing place to be thence indefinitely diffused. This hypothesis would not involve migration.

Interest has been felt in petroglyphs, because it has been supposed that if interpreted they would furnish records of vanished peoples or races, and connected with that supposition was one naturally affiliated that the old rock sculptures were made by peoples so far advanced in culture as to use alphabets or at least syllabaries, thus supporting the theory about the mythical mound builders or some other supposititious race. All suggestions of this nature should at once be abandoned. The practice of pictography does not belong to civilization and declines when an alphabet becomes popularly known. Neither is there the slightest evidence that an alphabet or syllabary was ever used in pre-Columbian America by the aborigines, though there is some trace of Runic inscriptions. The fact that the Maya and Aztec peoples were rapidly approaching to such modes of expressing thought, and that the Dakota and Ojibwa had well entered upon that line of evolution, shows that they had proceeded no farther, and it is admitted that they were favorable representatives of the tribes of the continent in this branch of art. The theory mentioned requires the assumption, without a particle of evidence, that the rock sculptures are alphabetic, and therefore were made by a supposititious and extinct race. Topers of the mysterious may delight in such dazing infusions of perverted fancy, but they are repulsive to the sober student.

The foregoing remarks apply mainly to rock inscriptions and not to pictographs on other substances, the discussion and illustration of which occupy the greater part of the present work. In that division there is no need of warning against wild theories or uncertain data. The objects are in hand and their current use as well as their significance is understood. Their description and illustration by classes is presented in the above chapters with such detail that further discussion here would be mere repetition.

One line of thought, however, is so connected with several of the classifications that it may here be mentioned with the suggestion that the preceding headings, with the illustrations presented under each, may be reviewed in reference to the methodical progress of pictography toward a determined and convenient form of writing. This exhibition of evolution was arrested by foreign invasion before the indirect signs of sound had superseded the direct presentments of sight for communication and record. Traces of it appear throughout the present paper, but are more intelligently noticed on a second examination than in cursory reading. In the Winter Counts of Battiste Good there are many characters where the figure of a human being is connected with an object, which shows his tribal status or the disease of which he died, and the characters representing the tribe or disease are purely determinative.

The discrimination which is made between animals and objects portrayed simply as such, and as supernatural or mystic, is shown in the many illustrations of Ojibwa and ZuÑi devices, in which the heart is connected with a line extending to the mouth, and those of the Ojibwa and the Dakota, where the spirals indicate spirit or wakan. Animals are often portrayed without such lines, in which cases it is understood that they are only the animals in natural condition, but with the designations or determinatives they are intended to be supernatural. Among the Ojibwa animals connected with certain ceremonies are represented as encircled by a belt or baldric, an ornamented baldric of the same character being used by the participants in the ceremonial chant dance; so that the baldric around the animal determines that the figure is that of a supernatural and mystic, not an ordinary, animal. This is an indication of the start from simple pictography towards an alphabet by the use of determinatives as was done by the Chinese.

It is not believed that much information of historical value will be obtained directly from the interpretation of the petroglyphs in America. The greater part of those already known are simply peckings, carvings, or paintings connected with their myths or with their every-day lives. It is, however, probable that others were intended to commemorate events, but the events, which to their authors were of moment, would be of little importance as history, if, as is to be expected, they were selected in the same manner as is done by modern Indian pictographers. They referred generally to some insignificant fight or some season of plenty or of famine, or to other circumstances the interest in which has long ago died away.

The question may properly be asked, why, with such small prospect of gaining historic information, so much attention has been directed to the collection and study of petroglyphs. A sufficient answer might be submitted, that the fact mentioned could not be made evident until after that collection and study, and that it is of some use to establish the limits of any particular line of investigation, especially one largely discussed with mystical inferences to support false hypotheses. But though the petroglyphs do not and probably never will disclose the kind of information hoped for by some enthusiasts, they surely are valuable as marking the steps in one period of human evolution and in presenting evidence of man’s early practices. Also though the occurrences interesting to their authors and therefore recorded or indicated by them are not important as facts of history, they are proper subjects of examination, simply because in fact they were the chief objects of interest to their authors, and for that reason become of ethnologic import. It is not denied that some of the drawings on rocks were made without special purpose, for mere pastime, but they are of import even as mere graffiti. The character of the drawings and the mode of their execution tell something of their makers. If they do not tell who those authors were, they at least suggest what kind of people they were as regards art, customs, and sometimes religion. But there is a broader mode of estimating the quality of known pictographs. Musicians are eloquent in lauding of the great composers of songs without words. The ideography, which is the prominent feature of picture writing, displays both primordially and practically the higher and purer concept of thoughts without sound.

The experience of the present writer induces him to offer the following suggestions for the benefit of travelers and other observers who may meet with petroglyphs which they may desire to copy and describe.

As a small drawing of large rock inscriptions must leave in doubt the degree of its finish and perhaps the essential objects of its production, it is requisite, in every instance, to affix the scale of the drawing, or to give a principal dimension to serve as a guide. A convenient scale for ordinary petroglyphs is one-sixteenth of actual size. The copy should be with sufficient detail to show the character of the work. It is useful to show the lithologic character of the rock or bowlder used; whether the drawing has been scratched into the face of the rock, or incised more deeply with a sharp implement, and the depth of such incision; whether the design is merely outlined, or the whole body of the figures pecked out, and whether paint has been applied to the pecked surface, or the design executed with paint only. The composition of paint should be ascertained when possible. The amount of weathering or erosion, together with the exposure, or any other feature bearing on the question of antiquity, might prove important. If actual colors are not accessible for representation the ordinary heraldic scheme of colors can be used.

That sketches, even by artists of ability, are not of high value in accuracy, is shown by the discrepant copies of some of the most carefully studied pictographs, which discrepancies sometimes leave in uncertainty the points most needed for interpretation. Sketches, or still better, photographs are desirable to present a connected and general view of the characters and the surface upon which they are found. For accuracy of details “squeezes” should be obtained when practicable.

A simple method of obtaining squeezes of petroglyphs, when the lines are sufficiently deep to receive an impression, is to take ordinary manilla paper of loose texture, and to spread the sheet, after being thoroughly wetted, over the surface, commencing at the top. The top edge may be temporarily secured by a small streak of starch or flour paste. The paper is then pressed upon the surface of the rock by means of a soft bristle brush, so that its texture is gently forced into every depression. Torn portions of the paper may be supplied by applying small patches of wet paper until every opening is thoroughly covered. A coating of ordinary paste, as above mentioned, is now applied to the entire surface, and a new sheet of paper, similarly softened by water, is laid over this and pressed down with the brush. This process is continued until three or four thicknesses of paper have been used. Upon drying, the entire mold will usually fall off by contraction. The edge at the top, if previously pasted to the rock, should be cut. The entire sheet can then be rolled up, or if inconveniently large can be cut in sections and properly marked for future purposes. This process yields the negative. To obtain the positive the inner coating of the negative may be oiled, and the former process renewed upon the cast.

The characters when painted with bright tints and upon a light-colored surface, may readily be traced upon tracing linen, such as is employed by topographers. Should the rock be of a dark color, and the characters indistinct, a simple process is to first follow the characters in outline with colored crayons, red chalk, or dry colors mixed with water and applied with a brush, after which a piece of muslin is placed over the surface and pressed so as to receive sufficient coloring matter to indicate general form and relative position. After these impressions are touched up, the true position may be obtained by painting the lines upon the back of the sheet of muslin, or by making a true tracing of the negative.

An old mode of securing the outline was to clear out the channels of the intaglios, then, after painting them heavily, to press a sheet of muslin into the freshly painted depressions. The obvious objection to this method is the damage to the inscription. Before such treatment, if the only one practicable, all particulars of the work to be covered by paint should be carefully recorded.

The locality should be reported with detail of State (or territory), county, township, and distance and direction from the nearest post-office, railway station, or country road. In addition the name of any contiguous stream, hill, bluff, or other remarkable natural feature should be given. The name of the owner of the land is of temporary value, as it is liable to frequent changes. The site or station should be particularly described with reference to its natural characteristics and geological history. When petroglyphs are in numbers and groups, their relation to each other to the points of the compass or to topographical features, should be noted, if possible, by an accurate survey, otherwise by numeration and sketching.

The following details should be carefully noted: The direction of the face of the rock; the presence of probable trails and gaps which may have been used in shortening distances in travel; localities of mounds and caves, if any, in the vicinity; ancient camping grounds, indicated by fragments of pottery, flint chips or other refuse; existence of aboriginal relics, particularly flints which may have been used in pecking (these may be found at the base of the rocks upon which petroglyphs occur); the presence of small mortar-holes which may have served in the preparation of colors.

With reference to pictographs on other objects than rock it is important to report the material upon which they appear and the implements ascertained to be used in their execution examples of which are given in other parts of this work.

With reference to all kinds of pictographs, it should be remembered that mere descriptions without graphic representations are of little value. Probable age and origin and traditions relating to them should be ascertained. Their interpretation by natives of the locality who themselves make pictographs or who belong to people who have lately made pictographs is most valuable, especially in reference to such designs as may be either conventional, religious, or connected with lines of gesture-signs.