[i]

TENTH ANNUAL REPORT
OF THE
BUREAU OF ETHNOLOGY
TO THE
SECRETARY OF THE SMITHSONIAN INSTITUTION
1888-’89

BY
J. W. POWELL
DIRECTOR

WASHINGTON
GOVERNMENT PRINTING OFFICE
1893

[ii]

[iii]

[iv]

[v]

[vi]

[vii]

Smithsonian Institution, Bureau of Ethnology,

Washington, D. C., October 1, 1889.

Sir: I have the honor to submit my Tenth Annual Report as Director of the Bureau of Ethnology.

The first part of it presents an exposition of the operations of the Bureau during the fiscal year 1888-’89; the second part consists of a work on the Picture-writing of the American Indians, which has been in preparation for several years.

I desire to express my thanks for your earnest support and your valuable counsel relating to the work under my charge.

I am, with respect, your obedient servant,

Prof. S. P. Langley,
Secretary of the Smithsonian Institution.

[viii]

[ix]

TENTH ANNUAL REPORT
OF THE
BUREAU OF ETHNOLOGY.

By J. W. Powell, Director.

Research among the North American Indians, in obedience to acts of Congress, was continued during the fiscal year 1888-’89.

The explanation presented in several former annual reports of the general plan upon which the work of the Bureau has been performed renders a detailed repetition superfluous. The lines of investigation which from time to time have appeared to be the most useful or the most pressing have been confided to persons trained in or known to be specially adapted to their pursuit. The results of their labors are presented in the three series of publications of the Bureau which are provided for by law. A brief statement of the work upon which each one of the special students was actively engaged during the fiscal year is furnished below; but it should be noted that this statement does not specify all the studies made or services rendered by them.

The assistance of explorers, writers, and students who are not and may not desire to be officially connected with the Bureau is again invited. Their contributions, whether in suggestions or extended communications, will always be gratefully acknowledged and will receive proper credit. They may be published as Congress will allow, either in the series of annual reports or in monographs or bulletins. Several valuable papers of this class have already been contributed and published.

[x]

The report now submitted consists of three principal divisions. The first relates to the publications made during the fiscal year; the second, to the work prosecuted in the field; the third, to the office work, which chiefly consists of the preparation for publication of the results of field work, with the corrections and additions obtained from exhaustive researches into the literature of the subjects discussed and by correspondence relative to them.

The publications actually issued and distributed during the year were as follows, all octavo:

Bibliography of the Iroquoian Languages, by James C. Pilling; pages i-vi + 1-208. Facsimile reproductions, at pages 44 and 56, of title pages of early publications relating to Indian languages, and, at page 72, of the Cherokee alphabet.

Textile Fabrics of Ancient Peru, by William H. Holmes; pages 1-17, Figs. 1-11.

The Problem of the Ohio Mounds, by Cyrus Thomas; pages 1-54, Figs. 1-8.

The field work of the year is divided into (1) mound explorations and (2) general field studies, the latter being directed chiefly to archeology, linguistics, and pictography.

The work of exploring the mounds of the eastern United States was, as in former years, under the superintendence of Mr. Cyrus Thomas. The efforts of the division were chiefly confined to the examination of material already collected and to the arrangement and preparation for publication of the data on hand. Field work received less attention, therefore, than in previous years, and was mainly directed to such investigations as were necessary to elucidate doubtful points and to the examination and surveys of important works which had not before received adequate attention.

[xi]

The only assistants to Mr. Thomas whose engagements embraced the entire year were Mr. James D. Middleton and Mr. Henry L. Reynolds. Mr. Gerard Fowke, one of the assistants, ceased his connection with the Bureau at the end of the second month. Mr. John W. Emmert was engaged as a temporary assistant for a few months.

During the short time in which he remained with the division, Mr. Fowke was engaged in exploring certain mounds in the Sciota valley, Ohio, a field to which Messrs. Squier and Davis had devoted much attention. Its reexamination was for the purpose of investigating certain typical mounds which had not been thoroughly examined by those explorers.

Mr. Middleton was employed from July to the latter part of October in the exploration of mounds and other ancient works in Calhoun county, Illinois, a territory to which special interest attaches because it seems to be on the border line of different archeologic districts. From October until December he was engaged at Washington in preparing plats of Ohio earthworks. During the next month he made resurveys of some of the more important inclosures in Ohio, after which he resumed work in the office at Washington until the latter part of March, when he was sent to Tennessee to examine several mound groups and to determine, so far as possible, the exact locations of the old Cherokee “over-hill towns.” The result of the last-mentioned investigation was valuable, as it indicated that each of these “over-hill towns” was, with possibly one unimportant exception, in the locality of a mound group.

Near the close of October Mr. Reynolds, having already examined the inclosures of the northern, eastern, and western sections of the mound region, went to Ohio and West Virginia to study the different types found there, with reference to the[xii] chapters he was preparing on the various forms of ancient inclosures in the United States. While thus engaged he explored a large mound connected with one of the typical works in Paint creek valley, obtaining unexpected and important results. The construction of this tumulus was found to be quite different from most of those in the same section examined by Messrs. Squier and Davis.

Mr. Emmert devoted the few months in which he was employed to the successful exploration of mounds in eastern Tennessee. Some important discoveries were made and additional interesting facts were ascertained in regard to the mounds of that section.

Early in the month of July Col. Garrick Mallery proceeded to Maine, Nova Scotia and New Brunswick to continue investigation into the pictographs of the Abnaki and Micmac Indians, which had been commenced in 1887. He first visited rocks in Maine, on the shore near Machiasport, and on Hog island, in Holmes bay, a part of Machias bay. In both localities pecked petroglyphs were found, accurate copies of which were taken. Some of them had not before been reported. They are probably of Abnaki origin, of either the Penobscot or the Passamaquoddy division, the rocks lying on the line of water communication between the territories of those divisions. From Maine he proceeded to Kejemkoojik lake, on the border of Queens and Annapolis counties, Nova Scotia, and resumed the work of drawing and tracing the large number of petroglyphs found during the previous summer. Perfect copies were obtained of so many of them as to be amply sufficient for study and comparison. These are incised petroglyphs, and were made by Micmacs. The country of the Malecites, on the St. Johns river, New Brunswick, was next visited. No petroglyphs were discovered, but a considerable amount of information was obtained[xiii] upon the old system of pictographs on birch bark and its use. Illustrative specimens were gathered, together with myths and legends, which assisted in the elucidation of some of the pictographs observed elsewhere.

Mr. W. J. Hoffman proceeded in July to visit the Red Lake and White Earth Indian reservations in Minnesota. At Red lake he obtained copies of birch bark records pertaining to the Midē'wiwin or Grand Medicine Society of the Ojibwa, an order of shamans professing the power to prophesy, to cure disease, and to confer success in the chase. The introductory portion of the ritual of this society pertains particularly to the Ojibwa cosmogony. At the same place he secured several birch bark records of hunting expeditions, battles with neighboring tribes of Indians, maps, and songs. He also investigated the former and present practice of tattooing, and the Ojibwa works of art in colors, beads, and quills.

At White Earth Reservation two distinct charts of the Grand Medicine Society were obtained, together with full explanations by two of the chief midé or shamans, one of whom was the only fourth-degree priest in either of the reservations. Although a considerable difference between these three charts is apparent, their principles and the general course of the initiation of the candidates are similar. The survival of archaic forms in the charts and ritual indicates a considerable antiquity. Some mnemonic songs were also obtained at this reservation. In addition to the ritual, secured directly from the priests, in the Ojibwa language, translations of the songs were also recorded, with musical notation. On leaving the above reservations, Mr. Hoffman proceeded to Pipestone, Minnesota, to copy the petroglyphs upon the cliffs of that historic quarry.

He then returned to St. Paul, Minnesota, to search the records of the library of the Minnesota Historical Society for copies of pictographs reported to have been made near La Pointe, Wisconsin. Little information was obtained, although it is known that such pictographs, now nearly obliterated,[xiv] existed upon conspicuous cliffs and rocks near Lake Superior, at and in the vicinity of Bayfield and Ashland.

Mr. Hoffman afterward made an examination of the “pictured cave,” eight miles northeast of La Crosse, Wisconsin, to obtain copies of the characters appearing there. These are rapidly being destroyed by the disintegration of the rock. The colors employed in delineating the various figures were dark red and black. The figures represent human beings, deer, and other forms not now distinguishable.

Mr. H. W. Henshaw spent the months of August, September, and October on the Pacific coast, engaged in the collection of vocabularies of several Indian languages, with a view to their study and classification. The Umatilla Reservation in Oregon was first visited with the object of obtaining a comprehensive vocabulary of the Cayuse. Though there are about four hundred of these Indians on the reservation, probably not more than six speak the Cayuse tongue. The Cayuse have extensively intermarried with the Umatilla, and now speak the language of the latter, or that of the Nez Percé. An excellent Cayuse vocabulary was obtained, and at the same time the opportunity was embraced to secure vocabularies of the Umatilla and the Nez Percé languages. His next objective point was the neighborhood of the San Rafael Mission, Marin county, California, the hope being entertained that some of the Indians formerly gathered at the mission would be found there. He learned that there were no Indians at or near San Rafael, but subsequently found a few on the shores of Tomales bay, to the north. A good vocabulary was collected from one of these, which, as was expected, was subsequently found to be related to the Moquelumnan family of the interior, to the southeast of San Francisco bay. Later the missions of Santa Cruz and Monterey were visited. At these points there still remain a few old Indians who retain a certain command of their own language, though Spanish forms their ordinary means of intercourse. The vocabularies obtained are sufficient to prove, beyond any reasonable doubt, that there are two[xv] linguistic families instead of one, as had been formerly supposed, in the country above referred to. A still more important discovery was made by Mr. Henshaw at Monterey, where an old woman was found who succeeded in calling to mind more than one hundred words and short phrases of the Esselen language, formerly spoken near Monterey, but less than forty words of which had been previously known. Near the town of Cayucas, to the south, an aged and blind Indian was visited who was able to add somewhat to the stock of Esselen words obtained at Monterey, and to give valuable information concerning the original home of that tribe. As a result of the study of this material Mr. Henshaw determines the Esselen to be a distinct linguistic family, a conclusion first drawn by Mr. Curtin from a study of the vocabularies collected by Galiano and Lamanon in the eighteenth century. The territory occupied by the tribe and linguistic family lies coastwise, south of Monterey bay, as far as the Santa Lucia mountains.

On July 5 Mr. James Mooney started on a second trip to the territory of the Cherokee in North Carolina, returning after an absence of about four months. During this time he made considerable additions to the linguistic material already obtained by him, and was able to demonstrate the former existence of a fourth, and perhaps even of a fifth, well-marked Cherokee dialect in addition to the upper, lower, and middle dialects already known. The invention of a Cherokee syllabary which was adapted to the sounds of the upper dialect has tended to make that dialect universal. A number of myths were collected, together with a large amount of miscellaneous material relating to the Cherokee tribe, and the great tribal game of ball play, with its attendant ceremonies of dancing, conjuring, scratching the bodies of the players, and going to water, was witnessed. A camera was utilized to secure characteristic pictures of the players. Special attention was given to the subject of Indian medicine, theoretic, ceremonial, and therapeutic. The most noted doctors of the tribe were employed as informants, and nearly five hundred specimens of medicinal[xvi] and food plants were collected and their Indian names and uses ascertained. The general result of this investigation shows that the medical and botanical knowledge of the Indians has been greatly overrated. A study was made of Cherokee personal names, about five hundred of which were translated, being all the names of Indian origin now remaining in that region. The most important results of Mr. Mooney’s investigations were the discovery of a large number of manuscripts containing the sacred formulas of the tribe, written in Cherokee characters by the shamans for their own secret use, and jealously guarded from the knowledge of all but the initiated. The existence of such manuscripts had been ascertained during a visit in 1887, and several of them had been procured. This discovery of genuine aboriginal material, written in an Indian language by shamans for their own use, is believed to be unique in the history of aboriginal investigation, and was only made possible through the invention of the Cherokee syllabary by Sequoia in 1821. Every effort was made by Mr. Mooney to obtain all the existing manuscripts, with the result of securing all of that material which was in the possession of the tribe. The whole number of formulas obtained is about six hundred. They consist of prayers and sacred songs, explanations of ceremonies, directions for medical treatment, and underlying theories. They relate to medicine, love, war, hunting, fishing, self-protection, witchcraft, agriculture, the ball play, and other similar subjects, thus forming a complete exposition of an aboriginal religion as set forth by its priests in their own language.

Early in October Mr. Jeremiah Curtin left Washington for the Pacific coast. During the remainder of the year he was occupied in Shasta and Humboldt counties, California, in collecting vocabularies and data connected with the Indian system of medicine. This work was continued in different parts of Humboldt and Siskiyou counties until June 30, 1889. Large collections of linguistic and other data were gathered and myths were secured which show that the whole system of[xvii] medicine of these Indians and the ministration of remedies originated in and are limited to sorcery practices.

The field work of Mr. Albert S. Gatschet during the year was short. It had been ascertained that Mrs. Alice M. Oliver, now in Lynn, Massachusetts, formerly lived on Trespalacios bay, Texas, near the homes of the Karánkawa, and Mr. Gatschet visited Lynn with a view of securing as complete a vocabulary as possible of their extinct language. Mrs. Oliver was able to recall about one hundred and sixty terms of the language, together with some phrases and sentences. She also furnished many valuable details regarding the ethnography of the tribe. Ten days were spent in this work.

Mr. J. N. B. Hewitt was occupied in field work from August 1 to November 8, as follows: From the first of August to September 20 he was on the Tuscarora reserve, in Niagara county, New York, in which locality fifty-five legends and myths were collected. A Penobscot vocabulary was also obtained here, together with other linguistic material. From September 20 to November 8 Mr. Hewitt visited the Grand River reserve, Canada, where a large amount of text was obtained, together with notes and other linguistic material.

Mr. Victor Mindeleff left Washington on October 23 for St. John’s, Arizona, where he examined the Hubbell collection of ancient pottery and secured a series of photographs and colored drawings of the more important specimens. Thence he went to Zuñi and obtained drawings of interior details of dwellings and other data necessary for the completion of his studies of the architecture of this pueblo. He returned to Washington December 7.

Mr. A. M. Stephen continued work among the Tusayan pueblos under the direction of Mr. Victor Mindeleff. He added[xviii] much to the knowledge of the traditionary history of Tusayan, and made an extensive study of the house lore and records of house-building ceremonials. He also reported a full nomenclature of Tusayan architectural terms as applied to the various details of terraced-house construction, with etymologies. He secured from the Navajo much useful information of the ceremonial connected with the construction of their conical lodges or “hogans,” supplementing the more purely architectural records of their construction previously collected by Mr. Mindeleff. As opportunity occurred he gathered typical collections of baskets and other textile fabrics illustrative of the successive stages of their manufacture, including specimens of raw materials and detailed descriptions of the dyes used. These collections are intended to include also the principal patterns in use at the present time, with the Indian explanations of their significance.

Major J. W. Powell, the Director, devoted much time during the year to the preparation of the paper to accompany a map of the linguistic families of America north of Mexico, the scope of which has been alluded to in previous reports. This report and map appear in the Seventh Annual Report of the Bureau.

Mr. Henshaw was chiefly occupied with the administrative duties of the office, which have been placed in his charge by the Director, and with the completion of the linguistic map.

Col. Mallery, after his return from the field work elsewhere mentioned, was engaged in the elaboration of the new information obtained and in further continued study of and correspondence relating to sign language and pictography. In this work he was assisted by Mr. Hoffman, particularly in the sketches made by the latter during previous field seasons, and in preparing a large number of the illustrations for the paper on Picture-writing of the American Indians which appears in the present volume.

Mr. J. Owen Dorsey did no field work during the year, but devoted much of the time to original investigations. Samuel[xix] Fremont, an Omaha Indian, came to Washington in October, 1888, and until February, 1889, assisted Mr. Dorsey in the revision of the entries for the Ȼegiha-English Dictionary. Similar assistance was rendered by Little Standing Buffalo, a Ponka Indian from the Indian Territory, in April and May, 1889. Mr. Dorsey also completed the entries for the Ȼegiha-English Dictionary, and a list of Ponka, Omaha, and Winnebago personal names. He translated from the Teton dialect of the Dakota all the material of the Bushotter collection in the Bureau of Ethnology, and prepared therefrom a paper on Teton folklore. He also prepared a brief paper on the camping circles of Siouan tribes, and in addition furnished an article on the modes of predication in the Athapascan dialects of Oregon and in several dialects of the Siouan family. He also edited the manuscript of the Dakota grammar, texts, and ethnography, written by the late Rev. Dr. S. R. Riggs, which has been published as Volume VII, Contributions to North American Ethnology. In May, 1889, he began an extensive paper on Indian personal names, based on material obtained by himself in the field, to contain names of the following tribes, viz: Omaha and Ponka, Kansa, Osage, Kwapa, Iowa, Oto and Missouri, and Winnebago.

Mr. Albert S. Gatschet’s office work was almost entirely restricted to the composition and completion of his Ethnographic Sketch, Grammar, and Dictionary of the Klamath Language of Oregon, with the necessary appendices. These works have been published as Parts 1 and 2, Vol. II, of Contributions to North American Ethnology.

Mr. Jeremiah Curtin during the year arranged and copied myths of various Indian families, and also transcribed Wasco, Sahaptin, and Yanan vocabularies previously collected.

Mr. James Mooney, on his return from the Cherokee reservation in 1888, began at once to translate a number of the prayers and sacred songs obtained from the shamans during his visit. The result of this work has appeared in a paper in the Seventh Annual Report of the Bureau entitled “Sacred formulas of the Cherokees.” Considerable time was devoted also to the elaboration of the botanic and linguistic notes obtained in the field. In[xx] the spring of 1889 he began the collection of material for a monograph on the aborigines of the Middle Atlantic slope, with special reference to the Powhatan tribes of Virginia. As a preliminary, about one thousand circulars, requesting information in regard to local names, antiquities, and surviving Indians, were distributed throughout Maryland, Delaware, Virginia, and northeastern Carolina. Sufficient information was obtained in responses to afford an excellent basis for future work in this direction.

Mr. John N. B. Hewitt, from July 1 to August 1, was engaged in arranging alphabetically the recorded words of the Tuscarora-English dictionary mentioned in former reports, and in the study of adjective word forms to determine the variety and kind of the Tuscarora moods and tenses. After his return from the field Mr. Hewitt classified and tabulated all the forms of the personal pronouns employed in the Tuscarora language. Studies were also prosecuted to develop the predicative function in the Tuscarora speech. All the terms of consanguinity and affinity as now used among the Tuscarora were recorded and tabulated. Literal translations of many myths collected in the field were made, and free translations added to four of them. In all appropriate instances linguistic notes were added relating to etymology, phonesis, and verbal change.

Mr. James C. Pilling gave much time to bibliographies of North American languages. The bibliography of the Iroquoian languages was completed early in the fiscal year, and the edition was issued in February. In the meantime a bibliography of the Muskhogean languages was compiled, the manuscript of which was sent to the Public Printer in January, 1889, though the edition was not delivered during the fiscal year. Early in March, 1889, Mr. Pilling went to Philadelphia to inspect the manuscripts belonging to the American Philosophical Society, the authorities of which gave him every facility, and much new material was secured. In June he visited the Astor, Lenox, and Historical Society libraries in New York; the libraries of the Boston Athenæum, Massachusetts Historical Society, American Board of Commissioners for Foreign Missions, and the Boston Public Library, in Boston;[xxi] that of Harvard University, in Cambridge; of the American Antiquarian Society, in Worcester; and the private library of Dr. J. Hammond Trumbull, in Hartford. In Canada he visited the library of Laval University, and the private library of Mr. P. Gagnon, in Quebec, of St. Mary’s College and Jacques Cartier School in Montreal, and various missions along the St. Lawrence river, to inspect the manuscripts left by the early missionaries. The result was the accumulation of much new material for insertion in the Algonquian bibliography.

Mr. William H. Holmes continued to edit the illustrations for the publications of the Bureau, and besides was engaged actively in his studies of aboriginal archeology. He completed papers upon the pottery of the Potomac valley, and upon the objects of shell collected by the Bureau during the last eight years, and he has others in preparation. As curator of Bureau collections he makes the following statement of accessions for the year: From Mr. Thomas and his immediate assistants, working in the mound region of the Mississippi valley and contiguous portions of the Atlantic slope, the Bureau has received one hundred and forty-six specimens, including articles of clay, stone, shell, and bone. Mr. Victor Mindeleff obtained sixteen specimens of pottery from the Pueblo country. Other collections by members of the Bureau and the U. S. Geological Survey are as follows: Shell beads and pendants (modern) from San Buenaventura, California, by Mr. Henshaw; fragments of pottery and other articles from the vicinity of the Cheroki agency, North Carolina, by Mr. Mooney; a large grooved hammer from the bluff at Three Forks, Montana, by Mr. A. C. Peale; a large series of rude stone implements from the District of Columbia, by Mr. De Lancey W. Gill. Donations have been received as follows: An important series of earthen vases from a mound on Perdido bay, Alabama, given by F. H. Parsons; ancient pueblo vases from southwestern Colorado, by William M. Davidson; a series of spurious earthen vessels, manufactured by unknown persons in eastern Iowa, from C. C. Jones, of Augusta, Georgia; fragments of pottery, etc., from Romney, West Virginia, given by G. H. Johnson; fragments of a steatite pot from Ledyard,[xxii] Connecticut, by G. L. Fancher; an interesting series of stone tools, earthen vessels, etc., from a mound on Lake Apopka, Florida, by Thomas Featherstonhaugh; fragments of gilded earthenware and photographs of antiquities from Mexico, by F. Plancarte; fragments of gold ornaments from Costa Rica, by Anastasio Alfaro. Important specimens have been received as follows: Articles of clay from a mound on Perdido bay, Alabama, loaned by Mrs. A. T. Mosman; articles of clay from the last mentioned locality, by A. B. Simons; pottery from the Potomac valley, by W. Hallett Phillips, by S. V. Proudfit, and by H. L. Reynolds; articles of gold and gold-copper alloy from Costa Rica, by Anastasio Alfaro, Secretary of the National Museum at San Jose.

Mr. Thomas was chiefly occupied during the year in the preparation of the second and third volumes of his reports upon the mounds. He also prepared a bulletin on the Circular, Square, and Octagonal Earthworks of Ohio, with a view of giving a summary of the recent survey by the mound division of the principal works of the above character in southern Ohio. A second bulletin was completed, entitled “The Problem of the Ohio Mounds,” in which he presented evidence to show that the ancient works of the state are due to Indians of several different tribes, and that some, at least, of the typical works were built by the ancestors of the modern Cherokees.

Mr. Reynolds after his return from the field was engaged in the preparation of a general map of the United States, showing the area of the mounds and the relative frequency of their occurrence. He also assisted Mr. Thomas in the preparation of the monograph upon the inclosures.

Mr. Victor Mindeleff, assisted by Mr. Cosmos Mindeleff, was engaged in preparing for publication a “Study of Pueblo Architecture” as illustrated in the provinces of Tusayan and Cibola, material for which he had been collecting for a number of years. This report has appeared in the Eighth Annual Report of the Bureau.

Mr. Cosmos Mindeleff with the force of the modeling room at the beginning of the fiscal year completed the exhibit of the Bureau for the Cincinnati Exposition, and during the early[xxiii] part of the year he was at Cincinnati in charge of that exhibit. Owing to restricted space it was limited to the Pueblo culture group, but this was illustrated as fully as the time would permit. The exhibit covered about 1,200 feet of floor space, as well as a large amount of wall space, and consisted of models of pueblo and cliff ruins, models of inhabited pueblos, ancient and modern pottery, examples of weaving, basketry, etc.; a representative series of implements of war, the chase, agriculture, and the household; manikins illustrating costumes, and a series of large photographs illustrative of aboriginal architecture of the pueblo region, and of many phases of pueblo life. Upon Mr. Mindeleff’s return from Cincinnati he resumed assistance to Mr. Victor Mindeleff upon the report on pueblo architecture, and by the close of the fiscal year the two chapters which had been assigned to him were completed. They consist of a review of the literature on the pueblo region and a summary of the traditions of the Tusayan group from material collected by Mr. A. M. Stephen. Work was also continued on the duplicate series of models, and twelve were advanced to various stages of completion. Some time was devoted to repairing original models which had been exhibited at Cincinnati and other exhibitions, and also to experiments in casting in paper, in order in find a suitable paper for use in large models. The experiments were successful.

Mr. J. K. Hillers has continued the collection of photographs of prominent Indians in both full-face and profile, by which method all the facial characteristics are exhibited to the best advantage. In nearly every instance a record has been preserved of the sitter’s status in the tribe, his age, biographic notes of interest, and in cases of mixed bloods, the degree of intermixture of blood. The total number of photographs obtained during the year is 27, distributed among the following tribes, viz: Sac and Fox, 5; Dakota, 6; Omaha, 6, and mixed bloods (Creeks), 10.

Mr. Franz Boas was employed from February to April in preparing for convenient use a series of vocabularies of the several Salish divisions, previously collected by him in British Columbia.

[xxiv]

Mr. Lucien M. Turner was for two years stationed at the Hudson Bay Company’s post, Fort Chimo, near the northern end of the peninsula of Labrador, as a civilian observer in the employ of the Signal Service, U. S. Army. He was appointed to that position at the request of the late Prof. Baird, Secretary of the Smithsonian Institution, in order that his skill might be made available in a complete investigation of the ethnology and natural history of the region. Mr. Turner left Washington in June, 1882, and returned in the autumn of 1884. During the last year he was engaged in the preparation of a report which will appear in one of the forthcoming annual reports of the Bureau.

The officers of the Bureau of Ethnology and all persons interested in researches concerning the North American Indians were this year called to lament the death of Mr. James Stevenson, who had made regular and valuable contributions to the publications and collections of the Bureau.

Mr. Stevenson was born in Maysville, Kentucky, on the 24th of December, 1840. When but a boy of 16 he became associated with Prof. F. V. Hayden, and accompanied him upon expeditions into the regions of the upper Missouri and Yellowstone rivers. Although the main objects of these expeditions were geological, his tastes led him chiefly to the observation of the customs and dialects of the Indians, and the facilities for such study afforded him by the winters spent among the Blackfoot and Dakota Indians excited and confirmed the anthropologic zeal which absorbed the greater part of his life.

After military service during the civil war he resumed, in 1866, the studies which had been interrupted by it, and accompanied Prof. Hayden to the Bad Lands of Dakota. From this expedition and the action of the Congress of the United States in 1866-’67, sprang the Hayden survey, and during its existence Mr. Stevenson was its executive officer. In one of the explorations from 1868 to 1878, which are too many to be here enumerated, he climbed the Great Teton, and was the[xxv] first white man known to have reached the ancient Indian altar on its summit.

In 1879 the Hayden survey was discontinued, the Bureau of Ethnology was organized, and the U. S. Geological Survey was established. Mr. Stevenson, in addition to his duties as the executive officer of the new survey, was detailed for research in connection with the Bureau of Ethnology. In the subsequent years he devoted the winters—from the incoming of the field parties to their outgoing in the spring—chiefly to business of the survey; his summers to his favorite researches. He explored the cliff and cave dwellings of Arizona and New Mexico; he unearthed in the Canyon de Chelly two perfect skeletons of its prehistoric inhabitants; he investigated the religious mythology of the Zuñi, and secured a complete collection of fetich-gods, never before allowed out of their possession; he studied the history and religions of the Navajo and the Tusayan, and made an invaluable collection of pottery, costumes, and ceremonial objects, which are now prominent in the U. S. National Museum. But in the high mesas which were the field of his explorations in 1885 he was attacked by the “mountain fever” in its worst form. It was his first serious illness, and his regular and temperate life saved him for the time. But a visit to the same region in 1887 brought on a second attack of this peculiar and distressing disease. He came home prostrated, with symptoms of serious heart failure.

He died at the Gilsey House, in New York city, on the 25th of July, 1888, and was buried in the cemetery of Rock Creek church, near Washington.

For the first time in the series of the Annual Reports of this Bureau a single paper is submitted to exhibit the character of the investigations undertaken and the facts collected by its officers, with the results of their studies upon such collections. But while the paper is single in form and in title, it includes, in its illustrations and the text relating to them, nearly all topics into which anthropology can properly be divided, and therefore shows more diversity than would often be contained[xxvi] in a volume composed of separate papers by several authors. Its subject-matter being essentially pictorial, it required a large number of illustrations, twelve hundred and ninety-five figures being furnished in the text, besides fifty-four full-page plates, which, with their explanation and discussion, expanded the volume to such size as to exclude other papers.

The papers accompanying the Fourth Annual Report of this Bureau, which was for the fiscal year 1882-’83, included one under the title “Pictographs of the North American Indians, a Preliminary Paper, by Garrick Mallery.” Although that work was of considerable length and the result of much research and study, it was in fact as well as in title preliminary. The substance and general character of the information obtained at that time on the subject was published not only for the benefit of students already interested in it, but also to excite interest in that branch of study among active explorers in the field and, indeed, among all persons engaged in anthropologic researches. For the convenience of such workers as were invited in general terms to become collaborators, suggestions were offered for the examination, description, and study of the objects connected with this branch of investigation which might be noticed or discovered by them. The result of this preliminary publication has shown the wisdom of the plan adopted. Since the distribution of the Fourth Annual Report pictography in its various branches has become, far more than ever before, a prominent feature in the publications of learned societies, in the separate works of anthropologists, and in the notes of scientific explorers. The present paper includes, with proper credit to the authors quoted or cited, many contributions to this branch of study which obviously have been induced by the preliminary paper before mentioned.

The interest thus excited has continued to be manifested by the publication of new information of importance, in diverse shapes and in many languages, some of which has been received too late for proper attention in this paper.

[xxvii]

Col. Mallery’s studies in pictography commenced in the field. He was stationed with his military command at Fort Rice, on the upper Missouri river, in the autumn of 1876, and obtained a copy of the remarkable pictograph which he then called “A Calendar of the Dakota Nation,” and published under that title, with interpretation and explanation, in Vol. III, No. 1, of the series of bulletins of the U. S. Geological and Geographical Survey of the Territories, issued April 9, 1877. This work attracted attention, and at the request of the Secretary of the Interior he was ordered by the Secretary of War, on June 13, 1877, to report for duty, in connection with the ethnology of the North American Indians, to the present Director of this Bureau, then in charge of the Geographical and Geological Survey of the Rocky Mountain Region. Upon the organization of the Bureau of Ethnology, in 1879, Col. Mallery was appointed ethnologist, and has continued in that duty without intermission, supplementing field explorations by study of all accessible anthropologic literature and by extensive correspondence. His attention has been steadily directed to pictography and to sign-language, which branches of study are so closely connected that neither can be successfully pursued to the exclusion of the other, but his researches have by no means been confined to those related subjects.

The plan and scope of the present work may be very briefly stated as follows:

After some introductory definitions and explanations general remarks are submitted upon the grand division of petroglyphs or pictures upon rocks as distinct from other exhibitions of pictography. This division is less susceptible of interpretation than others, but it claims special interest and attention because the locality of production is fixed, and also because the antiquity of workmanship may often be determined with more certainty than can that of pictures on less enduring and readily transportable objects. Descriptions, with illustrations, are presented of petroglyphs in North America, including those in several provinces of Canada, in many of the states and territories of the United States, in Mexico, and in the West Indies. A large number from Central and South America[xxviii] also appear, followed by examples from Australia, Oceanica, Europe, Africa, and Asia, inserted chiefly for comparison with the picture-writings in America, to which the work is specially devoted, and therefore styled extra-limital petroglyphs. The curious forms called cup sculptures are next discussed, followed by a chapter on pictographs considered generally, which condenses the results of much thought. The substances, apart from rocks, on which picture-writing is found are next considered, and afterwards the instruments and materials by which they are made. The subjects of pictography and the practices which elucidate it are classified under several headings, viz: Mnemonic, subdivided into (1) Knotted cords and objects tied, (2) Notched or marked sticks, (3) Wampum, (4) Order of songs, (5) Traditions, (6) Treaties, (7) Appointment, (8) Numeration, (9) Accounting; Chronology, in which the charts at first called calendars, but now, in correct translation of the Indian terms, styled winter-counts, are discussed and illustrated with the care required by their remarkable characteristics; Notices, which chapter embraces (1) Notice of visit, departure, and direction, (2) Direction by drawing topographic features, (3) Notice of condition, (4) Warning and guidance; Communications, including (1) Declaration of war, (2) Profession of peace and friendship, (3) Challenge, (4) Social and religious missives, (5) Claim or demand; Totems, titles, and names, divided into (1) Pictorial tribal designations, (2) Gentile and clan designation, (3) Significance of tattoo marks, which topic is discussed at length, with ample illustration, and (4) Designations of individuals, subdivided into insignia or tokens of authority, signs of individual achievements, property marks, and personal names. Some of the facts presented are to be correlated with the antique forms of heraldry and others with proper names in modern civilization.

The topic Religion, considered in the popular significance of that term, is divided into (1) Symbols of the supernatural, (2) Myths and mythic animals, (3) Shamanism, (4) Charms and amulets, (5) Religious ceremonies, and (6) Mortuary practices. Customs are divided into (1) Cult associations, (2) Daily life and habits, (3) Games. The chapter entitled Historic[xxix] presents (1) Record of expeditions, (2) Record of battle, which includes a highly interesting Indian pictured account of the battle of the Little Big Horn, commonly called the “Custer massacre,” (3) Record of migration, (4) Record of notable events. The Biographic chapter gives too many minutiæ for particularization here, but is divided into (1) Continuous record of events in life and (2) Particular exploits or events. Ideography permeates and infuses all the matter under the other headings, but is discussed distinctively and with evidential illustrations in the sections of (1) abstract ideas expressed objectively, and (2) symbols and emblems. In the latter section the author suggests that the proper mode of interpretation of pictographs whose origin and significance are unknown is that they are to be primarily supposed to be objective representations, but may be, and often are, ideographic, and in a limited number of cases may have become symbolic, but that the strong presumption without extrinsic evidence is against the occult or esoteric symbolism often attributed to the markings under discussion. The significance of colors is connected with ideography and examples are given of the colors used in many parts of the world for mere decoration, in ceremonies, for death and mourning, for war and peace, and to designate social status. The depiction of gesture and posture signs is next discussed, showing the intimate relation between a thought as expressed without words by signs, and a thought expressed without words by pictures corresponding to those signs.

Conventionalizing is divided into conventional devices, which were the precursors of writing, and the syllabaries and alphabets evolved. The pictographic origin of all the current alphabets of the world, often before discussed, receives further explanation.

While comparison by the reader between all the illustrations and the facts recorded and the suggestions submitted about them is essential to the utility of the work, the author gives, as representing his own mode of study, found to be advantageous in use, a chapter on Special Comparison, divided into (1) Typical style, (2) Homomorphs and symmorphs,[xxx] (3) Composite forms, (4) Artistic skill and methods. This chapter is followed by one with which it is closely connected, styled Means of Interpretation, divided into (1) Marked characters of known significance, (2) Distinctive costumes, weapons, and ornaments, (3) Ambiguous characters with known meanings, the latter being chiefly a collection of separate figures which would not be readily recognized without labels, but which are understood through reliable authority. Finally, under the rather noncommittal title of Controverted Pictographs, the subjects of fraud and error are discussed with striking examples and useful cautions.

From this brief paraphrase of the table of contents, it is obvious that nearly all branches of anthropology are touched upon. It is also to be remarked that the work is unique because it presents the several anthropologic topics recorded by the Indians themselves according to their unbiased conceptions, and in their own mode of writing. From this point of view the anonymous and generally unknown pictographers may be considered to be the primary authors of the treatise and Col. Mallery a discoverer, compiler, and editor. But such depreciative limitation of his functions would ignore the originality of treatment pervading the work and the systematic classification and skillful analysis shown in it which enhance its value and interest.

Classification of expenditures made from the appropriation for North American ethnology for the fiscal year ending June 30, 1889.

[1]

[2]

[3]

ACCOMPANYING PAPER.

SMITHSONIAN INSTITUTION—BUREAU OF ETHNOLOGY.

BY
GARRICK MALLERY.

[4]

[5]

[10]

[11]

[25]

PICTURE-WRITING OF THE AMERICAN INDIANS.

By Garrick Mallery.

An essay entitled “Pictographs of the North American Indians: A Preliminary Paper,” appeared in the Fourth Annual Report of the Bureau of Ethnology. The present work is not a second edition of that essay, but is a continuation and elaboration of the same subject. Of the eighty-three plates in that paper not one is here reproduced, although three are presented with amendments; thus fifty-one of the fifty-four plates in this volume are new. Many of the text figures, however, are used again, as being necessary to the symmetry of the present work, but they are now arranged and correlated so as to be much more useful than when unmethodically disposed as before, and the number of text figures now given is twelve hundred and ninety-five as against two hundred and nine, the total number in the former paper. The text itself has been rewritten and much enlarged. The publication of the “Preliminary Paper” has been of great value in the preparation of the present work, as it stimulated investigation and report on the subject to such an extent that it is now impossible to publish within reasonable limits of space all the material on hand. Indeed, after the present work had been entirely written and sent to the Public Printer, new information came to hand which ought to be published, but can not now be inserted.

It is also possible to give more attention than before to the picture-writing of the aboriginal inhabitants of America beyond the limits of the United States. While the requirements of the acts of Congress establishing the Bureau of Ethnology have been observed by directing main attention to the Indians of North America, there is sufficient notice of Central and South America to justify the present title, in which also the simpler term “picture-writing” is used instead of “pictographs.”

Picture-writing is a mode of expressing thoughts or noting facts by marks which at first were confined to the portrayal of natural or artificial objects. It is one distinctive form of thought-writing without[26] reference to sound, gesture language being the other and probably earlier form. Whether remaining purely ideographic, or having become conventional, picture-writing is the direct and durable expression of ideas of which gesture language gives the transient expression. Originally it was not connected with the words of any language. When adopted for syllabaries or alphabets, which is the historical course of its evolution, it ceased to be the immediate and became the secondary expression of the ideas framed in oral speech. The writing common in civilization may properly be styled sound-writing, as it does not directly record thoughts, but presents them indirectly, after they have passed through the phase of sound. The trace of pictographs in alphabets and syllabaries is discussed in the present work under its proper heading so far as is necessary after the voluminous treatises on the topic, and new illustrations are presented. It is sufficient for the present to note that all the varied characters of script and print now current are derived directly or mediately from pictorial representations of objects. Bacon well said that “pictures are dumb histories,” and he might have added that in the crude pictures of antiquity were contained the germs of written words.

The importance of the study of picture-writing depends partly upon the result of its examination as a phase in the evolution of human culture. As the invention of alphabetic writing is admitted to be the great step marking the change from barbarism to civilization, the history of its earlier development must be valuable. It is inferred from internal evidence, though not specifically reported in history, that picture-writing preceded and generated the graphic systems of Egypt, Assyria, and China, but in America, especially in North America, its use is still current. It can be studied here without any requirement of inference or hypothesis, in actual existence as applied to records and communications. Furthermore, the commencement of its evolution into signs of sound is apparent in the Aztec and the Maya characters, in which transition stage it was arrested by foreign conquest. The earliest lessons of the genesis and growth of culture in this important branch of investigation may, therefore, be best learned from the western hemisphere. In this connection it should be noticed that picture-writing is found in sustained vigor on the same continent where sign language has prevailed and has continued in active operation to an extent historically unknown in other parts of the world. These modes of expression, i. e., transient and permanent thought-writing, are so correlated in their origin and development that neither can be studied to the best advantage without including the other. Unacquainted with these facts, but influenced by an assumption that America must have been populated from the eastern hemisphere, some enterprising persons have found or manufactured American inscriptions composed of characters which may be tortured into identity with some of the Eurasian alphabets or syllabaries, but which sometimes[27] suggest letters of indigenous invention. This topic is discussed in its place.

For the purposes of the present work there is no need to decide whether sign-language, which is closely connected with picture-writing, preceded articulate speech. It is sufficient to admit the high antiquity of thought-writing in both its forms, and yet it is proper to notice a strong current of recent opinions as indicated by Prof. Sayce (a) in his address to the anthropologic section of the British Association for the Advancement of Science, as follows:

When a system of ideographic gesture signs prevailed and at the same time any form of artistic representation, however rude, existed, it would be expected that the delineations of the former would appear in the latter. It was but one more and an easy step to fasten upon bark, skins, or rocks the evanescent air pictures that still in pigments or carvings preserve their ideography or conventionalism in their original outlines. A transition stage between gestures and pictographs, in which the left hand is used as a supposed drafting surface, upon which the index draws lines, is exhibited in the Dialogue between Alaskan Indians in the First Annual Report of the Bureau of Ethnology (a). This device is common among deaf-mutes, without equal archeologic importance, as it may have been suggested by the art of writing, with which, even when not instructed in it, they are generally acquainted.

The execution of the drawings, of which the several forms of picture-writing are composed, often exhibits the first crude efforts of graphic art, and their study in that relation is of value.

When pictures are employed for the same purpose as writing, the conception intended to be presented is generally analyzed and only its most essential points are indicated, with the result that the characters when frequently repeated become conventional, and in their later form cease to be recognizable as objective portraitures. This exhibition of conventionalizing has its own historic import.

It is not probable that much valuable information will ever be obtained from ancient rock carvings or paintings, but they are important as indications of the grades of culture reached by their authors, and of the subjects which interested those authors, as is shown[28] in the appropriate chapters following. Some portions of these pictures can be interpreted. With regard to others, which are not yet interpreted and perhaps never can be, it is nevertheless useful to gather together for synoptic study and comparison a large number of their forms from many parts of the world. The present collection shows the interesting psychologic fact that primitive or at least very ancient man made the same figures in widely separated regions, though it is not established that the same figures had a common significance. Indications of priscan habitat and migrations may sometimes be gained from the general style or type of the drawings and sculptures, which may be divided into groups, although the influence of the environing materials must always be considered.

The more modern specimens of picture-writing displayed on skins, bark, and pottery are far more readily interpreted than those on rocks, and have already afforded information and verification as to points of tribal history, religion, customs, and other ethnologic details.

A criticism has been made on the whole subject of picture-writing by the eminent anthropologist, Dr. Andree, who, in Ethnographische Parallelen und Vergleiche (a), has described and figured a large number of examples of petroglyphs, a name given by him to rock-drawings and now generally adopted. His views are translated as follows:

It appears, therefore, that Dr. Andree carefully excludes the picture-writings of the North American Indians from his general censure, his conclusion being that those found in other parts of the world usually occupy a lower stage. It is possible that significance may yet be ascertained in many of the characters found in other regions, and perhaps this may be aided by the study of those in America; but no doubt should exist that the latter have purpose and meaning. The relegation to a trivial origin of such pictographs as are described and illustrated in the present work will be abandoned after a thorough knowledge of the labor and thought which frequently were necessary for their production. American pictographs are not to be regarded as mere curiosities. In some localities they represent the only intellectual remains of the ancient inhabitants. Wherever found, they bear significantly upon the evolution of the human mind.

Distrust concerning the actual significance of the ancient American petroglyphs may be dispelled by considering the practical use of similar devices by historic and living Indians for purposes as important to them as those of alphabetic writing, these serving to a surprising extent the same ends. This paper presents a large number of conclusive[29] examples. The old devices are substantially the same as the modern, though improved and established in the course of evolution. The ideography and symbolism displayed in these devices present suggestive studies in psychology more interesting than the mere information or text contained in the pictures. It must also be observed that when Indians now make pictographs it is with intention and care—seldom for mere amusement. Even when the labor is undertaken merely to supply the trade demand for painted robes or engraved pipes or bark records, it is a serious manufacture, though sometimes only imitative and not intrinsically significant. In all other known instances in which pictures are made without such specific intent as is indicated under the several headings of this work, they are purely ornamental; but in such cases they are often elaborate and artistic, not idle scrawls.

This paper is limited in its terms to the presentation of the most important known pictographs of the American Indians, but examples from other parts of the world are added for comparison. The proper classification and correlation of the matter collected has required more labor and thought than is apparent. The scheme of the work has been to give in an arrangement of chapters and sections some examples with illustrations in connection with each heading in the classification. This plan has involved a large amount of cross reference, because in many cases a character or a group of characters could be considered with reference to a number of different characteristics, and it was necessary to choose under which one of the headings it should be presented, involving reference to that from the other divisions of the work. Sometimes the decision was determined by taste or judgment, and sometimes required by mechanical considerations.

It may be mentioned that the limitation of the size of the present volume required that the space occupied by the text should be subordinated to the large amount of illustration. It is obvious that a work on picture-writing should be composed largely of pictures, and to allow room for them many pages of the present writer’s views have been omitted. Whatever may be the disadvantage of this omission it leaves to students of the work the opportunity to form their own judgments without bias. Indeed, this writer confesses that although he has examined and studied in their crude shape, as they went to the printer, all the illustrations and descriptions now presented, he expects that after the volume shall be delivered to him in printed form with its synoptic arrangement he will be better able than now to make appropriate remarks on its subject-matter. Therefore he anticipates that careful readers will judiciously correct errors in the details of the work which may have escaped him and that they will extend and expand what is yet limited and partial. It may be proper to note that when the writer’s observation has resulted in agreement with published authorities or contributors, the statements that could have been made on his own personal knowledge have been cited, when possible, from[30] the printed or manuscript works of others. Quotation is still more requisite when there is disagreement with the authorities.

Thanks for valuable assistance are due and rendered to correspondents and to officers of the Bureau of Ethnology and of the United States Geological Survey, whose names are generally mentioned in connection with their several contributions. Acknowledgment is also made now and throughout the work to Dr. W. J. Hoffman, who has officially assisted in its preparation during several years, by researches in the field, in which his familiarity with Indians and his artistic skill have been of great value. Similar recognition is due to Mr. De Lancey W. Gill, in charge of the art department of the Bureau of Ethnology and the U. S. Geological Survey, and to Mr. Wells M. Sawyer, his assistant, specially detailed on the duty, for their work on the illustrations presented. While mentioning the illustrations, it may be noted that the omission to furnish the scale on which some of them are produced is not from neglect, but because it was impossible to ascertain the dimensions of the originals in the few cases where no scale or measurement is stated. This omission is most frequently noticeable in the illustrations of petroglyphs which have not been procured directly by the officers of the Bureau of Ethnology. The rule in that Bureau is to copy petroglyphs on the scale of one-sixteenth actual size. Most of the other classes of pictographs are presented without substantial reduction, and in those cases the scale is of little importance.

It remains to give special notice to the reader regarding the mode adopted to designate the authors and works cited. A decision was formed that no footnotes should appear in the work. A difficulty in observing that rule arose from the fact that in the repeated citation of published works the text would be cumbered with many words and numbers to specify titles, pages and editions. The experiment was tried of printing in the text only the most abbreviated mention, generally by the author’s name alone, of the several works cited, and to present a list of them arranged in alphabetic order with cross references and catch titles. This list appears at the end of the work with further details and examples of its use. It is not a bibliography of the subject of picture-writing, nor even a list of authorities read and studied in the preparation of the work, but it is simply a special list, prepared for the convenience of readers, of the works and authors cited in the text, and gives the page and volume, when there is more than one volume in the edition, from which the quotation is taken.

[31]

In the plan of this work a distinction has been made between a petroglyph, as Andree names the class, or rock-writing, as Ewbank called it, and all other descriptions of picture-writing. The criterion for the former is that the picture, whether carved or pecked, or otherwise incised, and whether figured only by coloration or by coloration and incision together, is upon a rock either in situ or sufficiently large for inference that the picture was imposed upon it where it was found. This criterion allows geographic classification. In presenting the geographic distribution, prominence is necessarily (because of the laws authorizing this work) given to the territory occupied by the United States of America, but examples are added from various parts of the globe, not only for comparison of the several designs, but to exhibit the prevalence of the pictographic practice in an ancient form, though probably not the earliest form. The rocks have preserved archaic figures, while designs which probably were made still earlier on less enduring substances are lost.

Throughout the world in places where rocks of a suitable character appear, and notably in South America, markings on them have been found similar to those in North America, though until lately they have seldom been reported with distinct description or with illustration. They are not understood by the inhabitants of their vicinity, who generally hold them in superstitious regard, and many of them appear to have been executed from religious motives. They are now most commonly found remaining where the population has continued to be sparse, or where civilization has not been of recent introduction, with exceptions such as appear in high development on the Nile.

The superstitions concerning petroglyphs are in accord with all other instances where peoples in all ages and climes, when observing some phenomenon which they did not understand, accounted for it by supernatural action. The following examples are selected as of interest in the present connection.

It must be premised with reference to the whole character of the mythology and folk-lore of the Indians that, even when professed converts to Christianity, they seem to have taken little interest in the stories of the Christian church, whether the biblical narratives or the lives and adventures of the saints, which are so constantly dwelt upon[32] throughout the Christian world that they have become folk-lore. The general character of the Christian legends does not seem to have suited the taste of Indians and has not at all impaired their affection for or their belief in the aboriginal traditions.

Among the gods or demigods of the Abnaki are those who particularly preside over the making of petroglyphs. Their name in the plural, for there are several personages, is Oonagamessok. They lived in caves by the shore and were never seen, but manifested their existence by inscriptions on the rocks. The fact that these inscribed rocks are now very seldom found is accounted for by the statement that the Oonagamessok have become angry at the want of attention paid to them since the arrival of the white people and have caused the pictures to disappear. There is no evidence to determine whether this tradition should be explained by the fact that the ingenious shamans of the last century would sometimes produce a miracle, carving the rocks themselves and interpreting the marks in their own way, or by the fact that the rock inscriptions were so old that their origin was not remembered and an explanation was, as usual, made by ascription to a special divinity, perhaps a chieftain famous in the old stage of mythology, or perhaps one invented for the occasion by the class of priests who from immemorial antiquity have explained whatever was inexplicable.

At a rock near the mouth of the Magiguadavic river, at the time immediately before the Passamaquoddy Indians chose their first governor after the manner of the whites, the old Indians say there suddenly appeared a white man’s flag carved on the rocks. The old Indians interpreted this as a prophecy that the people would soon be abandoned to the white man’s methods, and this came to pass shortly after. Formerly they had a “Mayouett” or chief. Many other rock carvings are said to have foretold what has since come to pass. Strange noises have also been heard near them.

The Omaha superstition is mentioned on pages 91-92 infra.

The Mandans had an oracle stone on which figures appeared on the morning after a night of public fasting. They were deciphered by the shaman, who doubtless had made them.

Mr. T. H. Lewis (a) gives the following tradition relating to the incised bowlders in the upper Minnesota valley:

Mr. J. W. Lynd (a) says of the Dakotas:

Mr. Charles Hallock, on the authority of Capt. Ed. Hunter, First Cavalry, U. S. A., furnishes the following information respecting the Assiniboin, Montana, rock pictures, which shows the reverence of these Indians for the petroglyphs even when in ruins:

Marcano (a) gives an account in which superstition is mixed with historic tradition. It is translated as follows:

The remarkable height of some petroglyphs has misled authors of good repute as well as savages. Petroglyphs frequently appear on the face of rocks at heights and under conditions which seemed to render their production impossible without the appliances of advanced civilization, a large outlay, and the exercise of unusual skill. An instance among many of the same general character is in the petroglyphs at Lake Chelan, Washington, where they are about 30 feet above the present water level, on a perpendicular cliff, the base of which is in the lake. On simple examination the execution of the pictographic work would[34] seem to involve details of wharfing, staging, and ladders if operated from the base, and no less elaborate machinery if approached from the summit. Strahlenberg suggests that such elevated drawings were made by the ingenious use of stone wedges driven into the rock, thus affording support for ascent or descent, and reports that he actually saw such stone wedges in position on the Yenesei river. A very rough geological theory has been presented by others to account for the phenomena by the rise of the rocks to a height far above the adjacent surface at a time later than their carving.

But in the many cases observed in America it is not necessary to propose either the hypothesis involving such elaborate work as is suggested or one postulating enormous geological changes. The escarpment of cliffs is from time to time broken down by the action of the elements and the fragments fall to the base, frequently forming a talus of considerable height, on which it is easy to mount and incise or paint on the remaining perpendicular face of the cliff. When the latter adjoins a lake or large stream, the disintegrated débris is almost immediately carried off, leaving the drawings or paintings at an apparently inaccessible altitude. When the cliff is on dry land, the rain, which is driven against the face of the cliff and thereby increased in volume and force at the point in question, also sweeps away the talus, though more slowly. The talus is ephemeral in all cases, and the face of the cliff may change in a week or a century, as it may happen, so its aspect gives but a slight evidence of age. The presence, therefore, of the pictures on the heights described proves neither extraordinary skill in their maker nor the great antiquity which would be indicated by the emergence of the pictured rocks through volcanic or other dynamic agency. The age of the paintings and sculptures must be inferred from other considerations.

Pictures are sometimes found on the parts of rocks which at present are always, or nearly always, covered with water. On the sea shore at Machias bay, Maine, the peckings have been continued below the line of the lowest tides as known during the present generation. In such cases subsidence of the rocky formation may be indicated. At Kejimkoojik lake, Nova Scotia, incisions of the same character as those on the bare surface of the slate rocks can now be seen only by the aid of a water glass, and then only when the lake is at its lowest. This may be caused by subsidence of the rocks or by rise of the water through the substantial damming of the outlet. Some rocks on the shores of rivers, e. g., those on the Kanahwa, in West Virginia, show the same general result of the covering and concealment of petroglyphs by water, except in an unusual drought, which may more reasonably be attributed to the gradual elevation of the river through the rise of the surface near its mouth than to the subsidence of the earth’s crust at the locality of the pictured rocks.

It must be admitted that no hermeneutic key has been discovered[35] applicable to American pictographs, whether ancient on stone or modern on bark, skins, linen, or paper. Nor has any such key been found which unlocks the petroglyphs of any other people. Symbolism was of individual origin and was soon variously obscured by conventionalizing; therefore it requires separate study in every region. No interpreting laws of general application to petroglyphs so far appear, although types and tendencies can be classified. It was hoped that in some lands petroglyphs might tell of the characters and histories of extinct or emigrated peoples, but it now seems that knowledge of the people who were the makers of the petroglyphs is necessary to any clear understanding of their work. The fanciful hypotheses which have been formed without corroboration, wholly from such works as remain, are now generally discarded.

There is a material reason why the interpretation of petroglyphs is attended with special difficulty. They have often become so blurred by the elements and so much defaced where civilized man has penetrated that they cease to have any distinct or at least incontrovertible features. The remarks relating to Dighton rock, infra, Chap. XXII, are in point.

Rock-carving or picture-writing on rocks is so old among the American tribes as to have acquired a nomenclature. The following general remarks of Schoolcraft (a) are of some value, though they apply with any accuracy only to the Ojibwa and are tinctured with a fondness for the mysterious:

In the Oglala Roster mentioned in Chapter XIII, Section 4, infra, one of the heads of families is called Inyanowapi, translated as Painted (or inscribed) rock. A blue object in the shape of a bowlder is connected with the man’s head by the usual line, and characters too minute for useful reproduction appear on the bowlder. The name is interesting as giving the current Dakota term for rock-inscriptions. The designation may have been given to this Indian because he was an authority on the subject and skilled either in the making or interpretation of petroglyphs.

The name “Wikhegan” was and still is used by the Abnaki to signify portable communications made in daily life, as distinct from the rock carvings mentioned above, which are regarded by them as mystic.

[36]

One of the curious facts in connection with petroglyphs is the meager notice taken of them by explorers and even by residents other than the Indians, who are generally reticent concerning them. The present writer has sometimes been annoyed and sometimes amused by this indifference. The resident nearest to the many inscribed rocks at Kejimkoojik Lake, Nova Scotia, described in Chapter II, Section 1, was a middle-aged farmer of respectable intelligence who had lived all his life about 3 miles from those rocks, but had only a vague notion of their character, and with difficulty found them. A learned and industrious priest, who had been working for many years on the shores of Lake Superior preparing not only a dictionary and grammar of the Ojibwa language, but an account of Ojibwa religion and customs, denied the present existence of any objects in the nature of petroglyphs in that region. Yet he had lived for a year within a mile of a very important and conspicuous pictured rock, and, on being convinced of his error by sketches shown him, called in his Ojibwa assistant and for the first time learned the common use of a large group of words which bore upon the system of picture-writing, and which he thereupon inserted in his dictionary, thus gaining from the visitor, who had come from afar to study at the feet of this supposed Gamaliel, much more than the visitor gained from him.

[37]

The information thus far obtained about petroglyphs in Canada is meager. This may be partly due to the fact that through the region of the Dominion now most thoroughly known the tribes have generally resorted for their pictographic work to the bark of birch trees, which material is plentiful and well adapted for the purpose. Indeed the same fact affords an explanation of the paucity of rock-carvings or paintings in the lands immediately south of the boundary line separating the United States from the British possessions. It must also be considered that the country on both sides of that boundary was in general heavily timbered, and that even if petroglyphs are there they may not even yet have been noticed. But that the mere plenty of birch bark does not evince the actual absence of rock-pictures in regions where there was also an abundance of suitable rocks, and where the native inhabitants were known to be pictographers, is shown by the account given below of the multitudes of such pictures lately discovered in a single district of Nova Scotia. It is confidently believed that many petroglyphs will yet be found in the Dominion. Others may be locally known and possibly already described in publications which have escaped the researches of the present writer. In fact, from correspondence and oral narrations, there are indications of petroglyphs in several parts of the Dominion besides those mentioned below, but their descriptions are too vague for presentation here. For instance, Dr. Boas says that he has seen a large number of petroglyphs in British Columbia, of which neither he nor any other traveler has made distinct report.

The only petroglyphs yet found in the peninsula of Nova Scotia are in large numbers within a small district in Queens county, and they comprise objects unique in execution and in interest. They were examined by the present writer in the field seasons of 1887 and 1888, and some were copied by him, but many more copies were taken in the last-mentioned year by Mr. George Creed, of South Rawdon, Nova Scotia, who had guided the writer to the locality. Attention was at first[38] confined to Fairy lake and its rocks. This lake is really a bay of a larger lake which is almost exactly on the boundary line between Annapolis and Queens counties, one of those forming the chain through which the Liverpool river runs, and called Cegemacaga in More’s History of Queens County (a), but according to Dr. Silas Rand in his Reading Book in the Micmac Language (a), Kejimkoojik, translated by him as “swelled parts,” doubtless referring to the expansion of the Maitland river at its confluence with the Liverpool river.

The Fairy rocks, as distinct from others in the lake, are three in number, and are situated on the east side of Kejimkoojik lake and south of the entrance to Fairy lake. The northernmost of the three rocks is immediately at the entrance, the westernmost and central rock showing but a small surface at high water and at the highest stage of the water being entirely submerged. Three other inscribed rocks are about 2 miles south of these, at Piels (a corruption of Pierre’s) point, opposite an island called Glodes or Gload island, so named from a well-known Micmac family. These rocks are virtually a continuation of the same formation with depressions between them. Two other localities in the vicinity where the rocks are engraved, as hereafter described, are at Fort Medway river and Georges lake. As they are all of the same character, on the same material, and were obviously made by the same people, they are all classed together, when referred to in this paper, as at Kejimkoojik lake. All of these rocks are of schistose slate of the Silurian formation, and they lie with so gentle a dip that their magnitudes vary greatly with a slight change in the height of the water. On August 27, 1887, when, according to the reports of the nearest residents, the water was one foot above the average summer level, the unsubmerged portion of the central rock then surrounded by water was an irregular oval, the dimensions of which were 47 by 60 feet. The highest points of the Fairy rocks at that date were no more than three and few were more than two feet above the surface of the water. The inclination near the surface is so small that a falling of the water of one foot would double the extent of that part of the surface which, by its smoothness and softness, is adapted to engraving. The inclination at Piels point is steeper, but still allows a great variation of exposed surface in the manner mentioned.

Mr. Creed first visited the Fairy rocks in July, 1881. His attention was directed exclusively to the northernmost rock, which was then more exposed than it was in September, 1887, and much of the inscribed portion seen by him in 1881 was under water in 1887. The submerged parts of the rocks adjoining those exposed are covered with incisions. Many inscriptions were seen in 1881 by Mr. Creed through the water, and others became visible through a water glass in 1887. His recollection of the inscribed dates seen in 1881 is that some with French names attached were of years near 1700, and that the worn appearance of the figures and names corresponded with the lapse of time indicated by[39] those dates. A number of markings were noticed by him which are not found in the parts now exposed, and were evidently more ancient than most of the engravings on the latter. From other sources of information it is evident that either from a permanent rise in the water of the lake or from the sinking of the rocks, they formerly showed, within the period of the recollection of people now living, a much larger exposed surface than of late years, and that the parts long since permanently submerged were covered with engravings. The inference is that those engravings were made before Europeans had visited the locality.

It is to be specially remarked that the exposed surfaces where the rocks were especially smooth were completely marked over, no space of 3 inches square being unmarked, and over nearly all of those choice parts there were two, and in many cases three, sets of markings, above one another, recognizable by their differing distinctness. It also seemed that the second or third marking was upon plane surfaces where the earlier markings had been nearly obliterated by time. With pains and skill the earlier markings can be traced, and these are the outlines which from intrinsic evidence are Indian, whereas the later and more sharply marked outlines are obviously made by civilized men or boys, the latest being mere initials or full names of persons, with dates attached. Warning must be given that the ancient markings, which doubtless were made by the Micmacs, will probably not only escape the attention of the casual visitor, but even that an intelligent expert observer who travels to the scene with some information on the subject, and for the express purpose of finding the incisions, may fail to see anything but names, ships, houses, and similar figures of obviously modern design. This actually occurred within the week when the present writer was taking copies of the drawings by a mode of printing which left no room for fancy or deception. Indeed, frequently the marks were not distinctly apparent until after they had been examined in the printed copies.

The mode in which the copies were taken was by running over and through their outlines a blue aniline pencil, and then pressing a wetted sheet of ordinary printing paper upon them, so that the impression was actually taken by the process of printing. During the two field seasons mentioned, with the aid of Mr. Creed, three hundred and fifty different engravings and groups of engravings were thus printed. Some of these prints were of large dimensions, and included from ten to fifty separate characters and designs.

On the parts exposed in 1887 there were dates from 1800 to the current year, the number for the last year being much the greatest, which was explained by the fact that the wonderfully beautiful lake had been selected for a Sunday-school excursion. Over the greater part of the surface visible in 1887 there were few levels specially favorable for marking, and when these were found the double or treble use was in some instances noticed.

[40]

After the writer had inspected the rocks and discovered their characteristics, and learned how to distinguish and copy their markings, it seemed that, with the exception of a few designs recently dug or chipped out by lumbermen or visitors, almost always initials, the only interesting or ancient portions were scratchings which could be made on the soft slate by any sharp instrument. The faces of the rocks were immense soft and polished drawing-slates, presenting to any person who had ever drawn or written before an irresistible temptation to draw or write. The writer, happening to have with him an Indian stone arrow which had been picked up in the neighborhood, used its point upon the surface, and it would make as good scratches as any found upon the rocks except the very latest, which were obviously cut by the whites with metal knives.

As is above suggested, the peculiar multiplication of the characters upon the most attractive of the slates affords evidence as to their relative antiquity superior to that generally found in petroglyphs.[41] The existence of two or three different sets of markings, all visible and of different degrees of obliteration or distinctness, is in itself important; but, in addition to that, it is frequently the case that the second and third in the order of time have associated with them dates, from which the relative antiquity of the faintest, the dateless, can be to some extent estimated. Dates of the third and most recent class are attached to English names and are associated with the forms of English letters; those of the second class accompany French names, and in some cases have French designs. Figs. 1 and 2, about one-fourth original size, are presented to give an idea of these peculiar palimpsests.

[42]

For examples of other copies printed from the rocks at Kejimkoojik lake, see Figs. 549, 550, 654, 655, 656, 657, 658, 717, 718, 739, 740, 741, 1254, 1255, and 1262. These offer intrinsic evidence of the Micmac origin of the early class of engravings.

The presence of French names and styles of art in the drawings is explained by a story which was communicated by Louis Labrador, whose great-grandfather, old Ledore, according to his account, guided a body of French Acadians who, at the time of the expulsion, were not shipped off with the majority. They escaped the English in 1756 and traveled from the valley of Annapolis to Shelbourne, at the extreme southeast of the peninsula. During that passage they halted for a considerable time to recruit in the beautiful valley along the Kejimkoojik lake, on the very ground where these markings appear, which also was on the ancient Indian trail. Another local tradition, told by a resident of the neighborhood, gives a still earlier date for the French work. He says that after the capture of Port Royal, now Annapolis, in 1710, a party of the defeated Frenchmen, with a number of Indians as guides, went with their cattle to the wide meadows upon Kejimkoojik lake and remained there for a long time. It is exceedingly probable that the French would have been attracted to scratch on this fascinating smooth slate surface whether they had observed previous markings or not, but it seems evident that they did scratch over such previous markings. The latter, at least, antedated the beginning of the eighteenth century.

A general remark may be made regarding the Kejimkoojik drawings, that the aboriginal art displayed in them did not differ in any important degree from that shown in other drawings of the Micmacs and the Abnaki in the possession of the Bureau of Ethnology. Also that the rocks there reveal pictographic tendencies and practices which suggest explanations of similar work in other regions where less evidence remains of intent and significance. The attractive material of the slates and their convenient situation tempted past generations of Indians to record upon them the images of their current thoughts and daily actions. Hence the pictographic practice went into operation at this locality with unusual vigor and continuity. Although at Kejimkoojik lake there is an exceptional facility for determining the relative dates of the several horizons of scratchings, the suggestion there evoked may help to ascertain similar data elsewhere.

Mr. Charles Hallock kindly communicates information concerning pictographs on Nipigon bay, which is a large lake in the province of Ontario, 30 miles northwest of Lake Superior, with which it is connected by Nipigon river. He says:

[43]

He also, later, incloses a letter received by himself from Mr. Newton Flanagan, of the Hudson Bay Company, an extract from which is as follows:

Mr. Hallock also furnishes information regarding a petroglyph, the locality of which he gives as follows: Roche Percée, on the Souris river, in Manitoba, near the international boundary, 270 miles west of Dufferin, and nearly due north from Bismarck. This is an isolated rock in the middle of a plain, covered with pictographs of memorable events. It stands back from the river a half mile.

Mr. A. C. Lawson (a) gives an illustrated account of petroglyphs on the large peninsula extending into the Lake of the Woods and on an island adjacent to it. Strictly speaking this peninsula is in the district of Keewatin, but it is very near the boundary line of Manitoba, to which it is attached for administrative purposes. The account is condensed as follows:

The sketches published as copies of these glyphs show spirals, concentric circles, crosses, horseshoe forms, arrow shapes, and other characters similar to those found on rocks in the southwestern part of the United States, and also to petroglyphs in Brazil, examples from both of which regions are presented in this work, under their appropriate headings.

Dr. Franz Boas (a) published an account of a petroglyph on Vancouver island (now presented as Fig. 3) which, slightly condensed, is translated as follows:

The accompanying rock picture is found on the eastern shore of Sproat lake, near its southern outlet. Sproat lake lies about 10 kilometers north of the upper end of the Alberni fiord, which cuts deep into the interior of Vancouver island. In former times this region was the territory of the Hōpetschisāth, a tribe of the Nootka or Aht, who even now have a village some miles below the lake, at the entrance of Stamp river into the main river. That tribe, according to the statement of some of its older members, was a branch of the Kowitchin, who occupy the east side of Vancouver island, some kilometers northeast of the upper end of Alberni fiord. At that time the Ts’ēschāáth, another tribe of the Nootka, are said to have ascended the fiord and mixed with the Hōpetschisāth. The present inhabitants of the region know nothing concerning the origin of the rock picture. According to their legend, the rock on which it is carved was once the house of Kwótiath. Kwótiath is the wandering divinity in Nootka mythology, and corresponds approximately to the raven of the Tlinkit and Haida, the Qäls of the Kowitchin. The picture is found on a perpendicular rock wall about 7 meters high, which drops directly into the lake, so that it was necessary to make the copy while standing in the water. The rock is traversed in the middle by a broad cleft, narrowing below, from which blocks have fallen out which bore part of the drawing. To the north and south of the rock wall the shore rises gently, but rocky portions are found everywhere. The lines of the drawing are flat[45] grooves, about two or three fingers’ breadth, and in many places are so weathered as to be hardly recognizable. They have been scraped into the rock probably by the points of sticks rubbing moist sand against it. No marks of blows of any kind are found. The figures are here given in the same relative position in which they are found on the rock, except that the upper one on the right hand is at a distance from all the others, at the southern end of the rock. The objects represented are evidently fishes or marine monsters. The middle figure to the left of the cleft may be a manned boat, the fore part of which is probably destroyed.

Dr. Boas says that the copy as found in the Verhandlungen is incorrect. The design on the right hand is reversed and is now corrected.

Mr. G. M. Sproat (a) mentions this petroglyph:

Drawings or paintings on rocks are distributed generally over the greater part of the territory of the United States.

They are found on bowlders formed by the sea waves or polished by ice of glacial epochs; on the faces of rock ledges adjoining lakes and streams; on the high walls of canyons and cliffs; on the sides and roofs of caves; in short, wherever smooth surfaces of rock appear. Yet, while they are so frequent, there are localities to be distinguished in which they are especially abundant and noticeable. They differ markedly in character of execution and apparent subject-matter.

An obvious division can be made between the glyphs bearing characters carved or pecked and those painted without incision. There is also a third, though small, class in which the characters are both incised and painted. This division seems to coincide to a certain extent with geographic areas and is not fully explained by the influence of materials; it may, therefore, have some relation to the idiosyncrasy or development of the several authors, and consequently to tribal habitat and migrations.

In examining a chart of the United States in use by the Bureau of Ethnology, upon which the distribution of the several varieties of petroglyphs is marked, two facts are noticeable: First, the pecked and incised characters are more numerous in the northern and those expressed in colors more numerous in the southern areas. Second, there[46] are two general groupings, distinguished by typical styles, one in the north Atlantic states and the other in the south Pacific states.

The north Atlantic group is in the priscan habitat of the tribes of the Algonquian linguistic family, and extends from Nova Scotia southward to Pennsylvania, where the sculpturings are frequent, especially on the Susquehanna, Monongahela, and Alleghany rivers, and across Ohio from Lake Erie to the Kanawha river, in West Virginia. Isolated localities bearing the same type are found westward on the Mississippi river and a few of its western tributaries, to and including the Wind river mountains, in Wyoming, the former habitat of the Blackfeet Indians. All of these petroglyphs present typical characters, sometimes undefined and complicated. From their presumed authors, they have been termed the Algonquian type. Upon close study and comparison they show many features in common which are absent in extra-limital areas.

Immediately south of the Kanawha river, in West Virginia, and extending southward into Virginia, Tennessee, and North Carolina, the pecked or sculptured petroglyphs are replaced by painted figures of a style differing from the Algonquian. These are in the area usually designated as Cherokee territory, but there is no evidence that they are the work of that tribe; indeed, there is no indication of their authorship. The absence of pecked characters in this area is certainly not due to an absence of convenient material upon which to record them as the country is as well adapted to the mode of incision as is the northern Atlantic area.

Upon the Pacific slope a few pecked as well as colored petroglyphs occur scattered irregularly throughout the extreme northern area west of the Sierra Nevada, but on the eastern side of that range of mountains petroglyphs appear in Idaho, which have analogues extending south to New Mexico and Arizona, with remarkable groups at intervals between these extremes. All of these show sufficient similarity of form to be considered as belonging to a type which is here designated “Shoshonean.” Tribes of that linguistic family still occupy, and for a long time have occupied, that territory. Most of this Shoshonean group consists of pecked or incised characters, though in the southern area unsculptured paintings predominate.

On the western side of the Sierra Nevada, from Visalia southward, at Tulare agency, and thence westward and southward along the Santa Barbara coast, are other groups of colored petroglyphs showing typical features resembling the Shoshonean. This resemblance may be merely accidental, but it is well known that there was intercourse between the tribes on the two sides of the Sierra Nevada, and the Shoshonean family is also represented on the Pacific slope south of the mountain range extending from San Bernardino west to Point Conception. In this manner the artistic delineation of the Santa Barbara tribes may have been influenced by contact with others.

Petroglyphs have seldom been found in the central area of the United[47] States. In the wooded region of the Great lakes characters have been depicted upon birch bark for at least a century, while in the area between the Mississippi river and the Rocky mountains the skins of buffalo and deer have been used. Large rocks and cliffs favorably situated are not common in that country, which to a great extent is prairie.

In the general area of these typical groups characters are frequently found which appear intrusive, i. e., they have a strong resemblance not only to those found in other American groups, but are nearly identical with characters in other parts of the world. This fact, clearly established, prevents the adoption of any theory as to the authorship of many of the petroglyphs and thwarts attempts to ascertain their signification.

Ensign Albert P. Niblack, U. S. Navy, (a) gives a brief account, with sketches, reproduced here as Fig. 4, of petroglyphs in Alaska, which were taken from rocks from the ancient village of Stikine, near Fort Wrangell. Others were found on rocks just above high-water mark around the sites of ruined and abandoned villages.

In the upper character the Alaskan typical style of human faces is noticeable. The lower gives a representation of the orca or whale killer, which the Haida believe to be a demon called Skana, about which there are many mythic tales. Mr. Niblack remarks:

[48]

Mr. G. K. Gilbert, of the U. S. Geological Survey, gives the following information concerning petroglyphs observed by him in the vicinity of San Francisco mountain, Arizona:

All of these figures partake of the general type designated as the Shoshonean, and it is notable that close repetitions of some of the characters appear in petroglyphs in Tulare valley and Owens valley, California, which are described and illustrated in this section.

The object resembling a centipede, in Fig. 6, is a common form in various localities in Santa Barbara county, California, as will be observed by comparing the illustrations given in connection with that locality. In other of the Arizona and New Mexican petroglyphs similar outlines are sometimes engraved to signify the maize stalk.

Mr. Paul Holman, of the U. S. Geological Survey, reports that eight miles below Powers butte, on a mesa bordering on the Gila river and rising abruptly to the height of 150 feet, are pictographs covering the entire vertical face. Also on the summit of a spur of Oatman mountain, 200 yards from the Gila and 300 feet above it, are numbers of pictographs. Many of them are almost obliterated where they are on exposed surfaces.

Lieut. Col. Emory (a) reports that on a table-land near the Gila bend is a mound of granite bowlders, blackened by augite and covered with unknown characters, the work of human hands. On the ground near by were also traces of some of the figures, showing that some of the pictographs, at least, were the work of modern Indians. Others were of undoubted antiquity. He also reports in the same volume (b) that characters upon rocks of questionable antiquity occur on the Gila river at 32° 38′ 13″ N. lat. and 190° 7′ 30″ long. According to the plate, the figures are found upon bowlders and on the face of the cliff to the height of 30 feet.

[50]

Lieut. Whipple (a) remarks upon petroglyphs at Yampais spring, Williams river, as follows:

Many of the country rocks lying on the Colorado plateau of northern Arizona, east of Peach springs, bear petroglyphs of considerable artistic workmanship. Some figures, observed by Dr. W. J. Hoffman in 1872, were rather elaborate and represented the sun, human beings in various styles approaching the grotesque, and other characters not understood. All of those observed were made by pecking the surface of basalt with a harder variety of stone.

Mr. Gilbert also obtained sketches of etchings in November, 1878, on Partridge creek, northern Arizona, at the point where the Beale wagon road comes to it from the east. He says: “The rock is cross-laminated Aubrey sandstone and the surfaces used are faces of the laminæ. All the work is done by blows with a sharp point. (Obsidian is abundant in the vicinity.) Some inscriptions are so fresh as to indicate that the locality is still resorted to. No Indians live in the immediate vicinity, but the region is a hunting ground of the Wallapais and Avasupais (Cosninos).”

Notwithstanding the occasional visits of the above named tribes, the characters submitted more nearly resemble those of other localities known to have been made by the Moki Pueblos.

Rock drawings are of frequent occurrence along the entire extent of the valley of the Rio Verde, from a short distance below Camp Verde to the Gila river.

Mr. Thomas V. Keam reports drawings on the rocks in Canyon Segy, and in Keam’s canyon, northeastern Arizona. Some forms occurring at the latter locality are found also upon Moki pottery.

[51]

Petroglyphs are reported by Lieut. Theodore Mosher, Twenty-second Infantry, U. S. Army, to have been discovered by Lieut. Casey’s party in December, 1887, on the Chiulee (or Chilalí) creek, 30 or 40 miles from its confluence with San Juan river, Arizona. A photograph made by the officer in charge of the party shows the characters to have been outlined by pecking, the designs resembling the Shoshonean type of pictographs, and those in Owens valley, California, a description of which is given below.

A figure, consisting of two concentric circles with a straight line running out from the larger circle, occurs, among other carvings, on one of the many sculptured bowlders seen by Mr. J. R. Bartlett (a) in the valley of the Gila river in Arizona. His representation of this bowlder is here copied as Fig. 8. His language is as follows:

In the collections of the Bureau of Ethnology is an album or sketch book, which contains many drawings made by Mr. F. S. Dellenbaugh, from which the following sketches of petroglyphs in Arizona are selected, together with the brief references attached to each sheet.

Fig. 9 is a copy of characters appearing in Shinumo canyon, Arizona. They are painted, the middle and right hand figures being red, the human form having a white mark upon the abdomen; the left-hand figure of a man is painted yellow, the two plumes being red.

The petroglyphs in Fig. 10 are rather indistinct and were copied from the vertical wall of Mound canyon. The most conspicuous forms appear to be serpents.

[52]

In the foothills of California, wherever overhanging and rain-protected rocks occur, they are covered with paintings of various kinds made by Indians. Those on Rocky hill, some 15 miles east of Visalia, are especially interesting. The sheltered rocks are here covered with images of men, animals, and various inanimate objects, as well as curious figures. The paint used is red, black, and white, and wherever protected it has stood the ravages of time remarkably well. In many places the paintings are as vivid as the day they were laid on. Deer, antelope, coyotes, birds, and turtles are figured quite frequently, and may indicate either names of chiefs or tribes, or animals slain in the hunt. Here are also circles, spirals, crowns or bars, etc., signs the meaning of which is yet doubtful.

Mr. H. W. Turner, in a letter dated June 3, 1891, furnishes sketches (Fig. 11) from this locality, and a description of them as follows:

During the summer of 1882 Dr. Hoffman visited the Tule river agency, California, where he found a large rock painting, of which Fig. 983, infra, is a copy made by him. His description of it is as follows:

“The agency is upon the western side of the Sierra Nevada, in the headwater canyons of the branches of the south fork of Tule river. The[53] country is at present occupied by several tribes of the Mariposan linguistic stock, and the only answer made to inquiries respecting the age or origin of the painting was that it was found there when the ancestors of the present tribes arrived. The local migrations of the various Indian tribes of this part of California are not yet known with sufficient certainty to determine to whom the records may be credited, but all appearances with respect to the weathering and disintegration of the rock upon which the record is engraved, the appearance of the coloring matter subsequently applied, and the condition of the small depressions made at the time for mixing the pigments with a viscous substance, indicate that the work was performed about a century ago.

“The Indians now at Tule river have occupied that part of the state for at least one hundred years, and the oldest now living state that the records were found by their ancestors, though whether more than two generations ago could not be ascertained.

[54]

“The drawings were outlined by pecking with a piece of quartz or other siliceous rock, the depth varying from a mere visible depression to a third of an inch. Having thus satisfactorily depicted the several ideas, colors were applied which appear to have penetrated the slight interstices between the crystalline particles of the rock, which had been bruised and slightly fractured by hammering with a piece of stone. It appears probable, too, that to insure better results the hammering was repeated after application of the colors.

“Upon a small bowlder, under the natural archway formed by the breaking of the large rock, small depressions were found which had been used as mortars for grinding and mixing the colors. These depressions average 2 inches in diameter and about 1 inch in depth. Traces of color still remain, mixed with a thin layer of a shining substance resembling a coating of varnish and of flinty hardness. This coating is so thin that it can not be removed with a steel instrument, and appears to have become a part of the rock itself.

“From the animals depicted upon the ceiling it seems that both beaver and deer were found in the country, and as the beaver tail and the hoofs of deer and antelope are boiled to procure glue, it is probable that the tribe which made these pictographs was as far advanced in respect to the making of glue and preparing of paints as most other tribes throughout the United States.

“Examination shows that the dull red color is red ocher, found in various[55] places in the valley, while the yellow was an ocherous clay, also found there. The white color was probably obtained there, and is evidently earthy, though of what nature can only be surmised, not sufficient being obtainable from the rock picture to make satisfactory analysis with the blow-pipe. The composition of the black is not known, unless it was made by mixing clay and powdered charcoal. The latter is a preparation common at this day among other tribes.

“An immense granite bowlder, about 20 feet in thickness and 30 in length, is so broken that a lower quarter is removed, leaving a large square passageway through its entire diameter almost northwest and southeast. Upon the western wall of this passageway is a collection of the colored sketches of which Fig. 983 is a reduced copy. The entire face of the rock upon which the pictograph occurs measures about 12 or 15 feet in width and 8 in height. The largest human figure measures 6 feet in height, from the end of the toes to the top of the head, the others being in proportion as represented.

“Upon the ceiling are a number of well executed drawings of the beaver, bear, centipede (Fig. 12), and bald eagle (Fig. 13). Many of the other forms indicated appear to represent some variety of insects, several of which are drawn with exaggerated antennæ, as in Fig. 14. It is curious to note the gradual blending of forms, as, for instance, that of the bear with those resembling the human figure, often found among the Shoshonean types in Arizona and New Mexico, some of which are described and figured infra.

“Fig. 15 embraces a number of characters on the ceiling. The left hand upper figure is in black, with a narrow line of red surrounding it. The drawing is executed neatly and measures about 18 inches in length.[56] The remaining characters are in dull red, probably ocher, though the two on the left hand, beneath the one just mentioned, are more yellowish.

“The first three forms in Fig. 16 are copies of human-like figures painted on the ceiling. They are each about 12 inches in length. The other form in Fig. 16 is white and is on the southern vertical wall of the passageway facing the north. It resembles some of the human forms occurring elsewhere in the same series of petroglyphs.”

In the range of mountains forming the northwestern boundary of Owens valley are extensive groups of petroglyphs, apparently dissimilar to those found west of the Sierra Nevada. Dr. Hoffman, of the Bureau of Ethnology, hastily examined them in 1871 and more thoroughly in the autumn of 1884. They are now represented in Pls. I to XI. So large a space is given to these illustrations because of their intrinsic interest, and also because it is desirable to show for one locality what is true of some others, viz, the very large number of petroglyphs still to be found in groups and series. Even with the present illustrations, the petroglyphs in Owens valley are by no means exhaustively shown.

Dr. Hoffman’s report is as follows:

[57]

In a communication dated Saratoga Springs, at the lower part of Death valley, California, February 5, 1891, Mr. E. W. Nelson says that about 200 yards from the springs, and on the side of a hill, he found[61] several petroglyphs. He also furnished a sketch as an example of their general type, now presented as Fig. 18. The locality is in the lower end of Death valley. Mr. Nelson says:

Dr. C. Hart Merriam, of the Department of Agriculture, on his return from the exploration of Death valley, kindly furnished a photograph of a ledge in Emigrant canyon, Panamint mountains, which was received too late for insertion in this work. This is much regretted, as a large number of petroglyphs are represented in groups. The characters are of the Shoshonean type. Among them are “Moki goats,” tridents, the Greek Φ, many crosses, and other figures shown in this chapter as found in the same general region.

In the Mojave desert, about 2 miles north of Daggett station, according to the Mining and Scientific Press (a) is a small porphyritic butte known as “Rattlesnake rock,” “so named by reason of the immense number of these reptiles that find shelter in this mass of rock.” The accompanying Fig. 19 is a reproduction of that given in the paper quoted. The author states that “the implement used in making these characters was evidently a dull-pointed stone, as the lines are not sharp, and the sides of the indentation show marks of striation.”

Lieut. Whipple reports the discovery of pictographs at Piute creek,[62] about 30 miles west of the Mojave villages. These are carved upon a rock, “are numerous, appear old, and are too confusedly obscured to be easily traceable.” They bear great general resemblance to drawings scattered over northeast Arizona, southern Utah, and western New Mexico.

From information received from Mr. Alphonse Pinart, pictographic records exist in the hills east of San Bernardino, somewhat resembling those at Tule river in the southern spurs of the Sierra Nevada, Kern county.

Mr. Willard J. Whitney, of Elmhurst, Lackawanna county, Pennsylvania, gives information regarding nearly obliterated pecked petroglyphs upon two flat granite rocks, or bowlders, on the summit of a mountain 4 miles directly west of Escondido, San Diego county, California.[63] The designs are not colored, and are not more than one-eighth or one-fourth of an inch in depth. There is a good lookout from the eminence, but there are no indications of either trails or burials in the vicinity.

[64]

This may be the locality mentioned by Mr. Barnes, of San Diego, who furnished information relating to petroglyphs in San Diego county.

Dr. Hoffman reports the following additional localities in Santa Barbara and Los Angeles counties. Fifteen miles west of Santa Barbara, on the northern summit of the Santa Ynez range, and near the San Marcos pass, is a group of paintings in red and black. Fig. 20 resembles a portion of a checker-board in the arrangement of squares.

[65]

Serpentine and zigzag lines occur, as also curved lines with serrations on the concave sides; figures of the sun; short lines and groups of short parallel lines, and figures representing types of insect forms also appear, as shown in Figs. 21 and 22.

These paintings are in a cavity near the base of an immense bowlder, over 20 feet in height. A short distance from this is a flat granitic bowlder, containing twenty-one mortar holes, which had evidently been used by visiting Indians during the acorn season. Oaks are very abundant, and their fruit formed one of the sources of subsistence.

Three miles west-northwest of this locality, in the valley near the base of the mountain, are indistinct figures in faded red, painted upon a large rock. The characters appear similar, in general, to those above mentioned.

Forty-three miles west of Santa Barbara, in the Najowe valley, is a[66] promontory, at the base of which is a large shallow cavern, the opening being smaller than the interior, upon the roof and back of which are many designs, some of which are reproduced in Fig. 23, of forms similar to those observed at San Marcos pass. Several characters appear to have been drawn at a later date than others, such as horned cattle, etc. The black used was a manganese compound, while the red pigments[67] consist of ferruginous clays, abundant at numerous localities in the mountain canyons.

Some of the human figures are drawn with the hands and arms in the attitude of making the gestures for surprise or astonishment, and negation, as in Fig. 24.

The characters in Fig. 25 resemble forms which occur at Tulare valley, and in Owens valley, respectively, and insect forms also occur as in Fig. 26.

[68]

Other designs abounding at this locality are shown in Figs. 27 and 28.

One of the most extensive groupings, and probably the most elaborately drawn, is in the Carisa plain, near Mr. Oreña’s ranch, 60 or 70 miles due north of Santa Barbara. The most conspicuous figure is that of the sun, resembling a human face, with ornamental appendages at the cardinal points, and bearing striking resemblance to some Moki masks and pictographic work. Serpentine lines and anomalous forms also abound.

Four miles northeast of Santa Barbara, near the residence of Mr. Stevens, is an isolated sandstone bowlder measuring about 20 feet high and 30 feet in diameter, upon the western side of which is a slight cavity bearing designs shown in Fig. 29, which correspond in general form to others in Santa Barbara county. The gesture for negation appears in the attitude of the human figures.

Half a mile farther east, on Dr. Coe’s farm, is another smaller bowlder, in a cavity of which various engravings appear shown in Fig. 30. Parts of the drawings have disappeared through disintegration of the rock, which is called “Pulpit rock,” on account of the shape of the[69] cavity, its position at the side of the narrow valley, and the echo observed upon speaking a little above the ordinary tone of voice.

Painted rocks also occur in the Azuza canyon, about 30 miles northeast of Los Angeles, of which Fig. 31 gives copies.

Just before his departure from the Santa Barbara region, Dr. Hoffman was informed of the existence of eight or nine painted records in that neighborhood, which up to that time had been observed only by a few sheep-herders and hunters.

Mr. L. L. Frost, of Susanville, California, reports the occurrence of pictographs (undoubtedly petroglyphs) 15 miles south of that town, on Willow creek, and at Milford, in the lower end of the valley. No details were furnished as to their general type and condition.

On Porter creek, 9 miles southwest of Healdsburg, on a large bowlder of hornblende syenite, petroglyphs similar to those found in Arizona and Nevada are to be seen. They are generally oblong circles or ovals, some of which contain crosses.

Figs. 32 and 33 are reduced copies 1/32 of original size of colored petroglyphs found by Dr. Hoffman in September, 1884, 12 miles west-northwest of the city of Santa Barbara, California. The locality is almost at the summit of the Santa Ynez range of mountains; the gray sandstone[70] rock on which they are painted is about 30 feet high and projects from a ridge so as to form a very marked promontory extending into a narrow mountain canyon. At the base of the western side of this bowlder is a rounded cavity, measuring on the inside about 15 feet in width and 8 feet in height. The floor ascends rapidly toward the back of the cave, and the entrance is rather smaller in dimensions than the above measurements of the interior. About 40 yards west of this rock is a fine spring of water. One of the four old Indian trails leading northward across the mountains passes by this locality, and it is probable that this was one of the camping places of the tribe which came south to trade, and that some of its members were the authors of the paintings. The three trails beside the one just mentioned cross the mountains at several points east of this, the most distant being about 15 miles. Other trails were known, but these four were most direct to the immediate vicinity of the Spanish settlement which sprang up shortly after the establishment of the Santa Barbara mission in 1786. The appearance and position of these and other pictographs in the vicinity appear to be connected with the several trails. The colors used in the paintings are red and black.

The circles figured in b and d of Fig. 32, and c, r, and w of Fig. 33, together with other similar circular marks bearing cross lines upon the interior, were at first unintelligible, as their forms among various tribes have very different signification. The character in Fig. 32, above and projecting from d, resembles the human form, with curious lateral bands of black and white, alternately. Two similar characters appear, also, in Fig. 33, a, b. In a the lines from the head would seem to indicate a superior rank or condition of the person depicted.

[71]

At the private ethnologic collection of Mr. A. F. Coronel, of Los Angeles, California, Dr. Hoffman discovered a clue to the general import of the above petroglyphs, as well as the signification of some of their characters. In a collection of colored illustrations of old Mexican costumes he found blankets bearing borders and colors nearly identical with those shown in the circles in Fig. 32, d, and Fig. 33, c, r, w. It is probable that the circles represent bales of blankets which early became articles of trade at the Santa Barbara mission. If this supposition is correct, the cross lines would seem to represent the cords used in tying the blankets into bales, which same cross lines appear as cords in l, Fig. 33. Mr. Coronel also possesses small figures of Mexicans, of various conditions of life, costumes, trades, and professions, one of which, a painted statuette, is a representation of a Mexican lying down flat upon an outspread serape, similar in color and form to the black and white bands shown in the upper figure of d, Fig. 32, and a, b, of Fig. 33, and instantly suggesting the explanation of those figures. Upon the latter the continuity of the black and white bands is broken, as the human figures are probably intended to be in front, or on top, of the drawings of the blankets.

The small statuette above mentioned is that of a Mexican trader, and if the circles in the petroglyphs are considered to represent bales of blankets, the character in Fig. 32, d, is still more interesting, from the union of one of these circles with a character representing the trader, i. e., the man possessing the bales. Bales, or what appear to be bales, are represented to the top and right of the circle in d, in that figure. In Fig. 33, l, a bale is upon the back of what appears to be a horse, led in[72] an upward direction by an Indian whose headdress and ends of the breechcloth are visible. To the right of the bale are three short lines, evidently showing the knot or ends of the cords used in tying a bale of blankets without colors, therefore of less importance, or of other goods. Other human forms appear in the attitude of making gestures, one also in j, Fig. 33, probably carrying a bale of goods. In the same figure u represents a centipede, an insect found occasionally south of the mountains, but reported as extremely rare in the immediate northern regions. For remarks upon x in the same figure see Chapter XX, Section 2, under the heading The Cross.

Mr. Coronel stated that when he first settled in Los Angeles, in 1843, the Indians living north of the San Fernando mountains manufactured blankets of the fur and hair of animals, showing transverse bands of black and white similar to those depicted, which were sold to the inhabitants of the valley of Los Angeles and to Indians who transported them to other tribes.

It is probable that the pictographs are intended to represent the salient features of a trading expedition from the north. The ceiling of the cavity found between the paintings represented in the two figures has disappeared, owing to disintegration, thus leaving a blank about 4 feet long, and 6 feet from the top to the bottom between the paintings as now presented.

Petroglyphs are reported by Mr. Cyrus F. Newcomb as found upon cliffs on Rock creek, 15 miles from Rio Del Norte, Colorado. Three small photographs, submitted with this statement, indicate the characters to have been pecked; they consist of men on horseback, cross-shaped human figures, animals, and other designs greatly resembling those found in the country of the Shoshonean tribes, examples of which are given infra.

Another notice of the same general locality is made by Capt. E. L. Berthoud (a) as follows:

Mr. Charles D. Wright, of Durango, Colorado, in a communication dated February 20, 1885, gives an account of some “hieroglyphs” on[73] rocks and upon the walls of cliff houses near the boundary line between Colorado and New Mexico. He says:

Mr. Wright further reports characters on rocks near the San Juan river. Four characters represent men as if in the act of taking an obligation, hands extended, and wearing a “kind of monogram on breast, and at their right are some hieroglyphics written in black paint covering a space 3 by 4 feet.”

The best discussed and probably the most interesting of the petroglyphs in the region are described and illustrated by Mr. W. H. Holmes (a), of the Bureau of Ethnology. The illustrations are here reproduced in Figs. 34 to 37, and the remarks of Mr. Holmes, slightly condensed, are as follows:

It will be noticed that the last two petroglyphs are in New Mexico, but they are so near the border of Colorado and so connected with the series in that state that they are presented under the same heading.

The following account is extracted from Rafn’s Antiquitates Americanæ (a):

Charles C. Jones, jr., (a) describes a petroglyph in Georgia as follows:

The illustration given by him is here reproduced in Fig. 38. It will be noted that the characters in it are chiefly circles, including plain, nucleated, and concentric, sometimes two or more being joined by straight lines, forming what is now known as the “spectacle shaped” figure. The illustrations should be compared with the many others presented in this paper under the heading of Cup Sculptures, see Chapter V, infra.

Dr. M. F. Stephenson (a) mentions sculptures of human feet, various animals, bear tracks, etc., in Enchanted mountain, Union county, Georgia. The whole number of sculptures is reported as one hundred and forty-six.

Mr. Jones (b) gives a different résumé of the objects depicted, as follows:

[77]

Mr. G. K. Gilbert, of the U. S. Geological Survey, has furnished a small collection of drawings of Shoshonean petroglyphs from Oneida, Idaho, shown in Fig. 39. Some of them appear to be totemic characters, and possibly were made to record the names of visitors to the locality.

Mr. Willard D. Johnson, of the U. S. Geological Survey, reports pictographic remains observed by him near Oneida, Idaho, in 1879. The figures represent human beings and were on a rock of basalt.

A copy of another petroglyph found in Idaho appears in Fig. 1092, infra.

Petroglyphs are reported by Mr. John Criley as occurring near Ava, Jackson county, Illinois. The outlines of the characters observed by him were drawn from memory and submitted to Mr. Charles S. Mason, of Toledo, Ohio, through whom they were furnished to the Bureau of Ethnology. Little reliance can be placed upon the accuracy of such drawing, but from the general appearance of the sketches the originals of which they are copies were probably made by one of the middle Algonquian tribes of Indians.

The “Piasa” rock, as it is generally designated, was referred to by the missionary explorer Marquette in 1675. Its situation was immediately above the city of Alton, Illinois.

[78]

Marquette’s remarks are translated by Dr. Francis Parkman (a) as follows:

Another version, by Davidson and Struvé (a), of the discovery of the petroglyph is as follows:

A footnote connected with the foregoing quotation gives the following description of the same rock:

Mr. McAdams (a), of Alton, Illinois, says “The name Piasa is Indian and signifies, in the Illini, ‘The bird which devours men.’” He furnishes a spirited pen-and-ink sketch, 12 by 15 inches in size and purporting to represent the ancient painting described by Marquette. On the picture is inscribed the following in ink: “Made by Wm. Dennis, April 3d, 1825.” The date is in both letters and figures. On the top of the picture in large letters are the two words, “FLYING DRAGON.” This picture, which has been kept in the old Gilham family of Madison county and bears the evidence of its age, is reproduced as Fig. 40.

He also publishes another representation (Fig. 41) with the following remarks:

Under Myths and Mythic Animals, Chapter XIV, Section 2, are illustrations and descriptions which should be compared with these accounts, and Chapter XXII gives other examples of errors and discrepancies in the description and copying of petroglyphs.

Mr. A. D. Jones (a) says of the same petroglyph:

Mr. McAdams, loc. cit., also reports regarding Fig. 42:

Mr. McAdams suggests that the painted figure of the human form with the bow and arrows was made later than the sculpture.

The same author (b) says, describing Fig. 43:

Mr. James D. Middleton, formerly of the Bureau of Ethnology, mentions the occurrence of petroglyphs on the bluffs of the Mississippi river, in Jackson county, about 12 miles below Rockwood. Also of others about 4 or 5 miles from Prairie du Rocher, near the Mississippi river.

Mr. P. W. Norris, of the Bureau of Ethnology, found numerous caves on the banks of the Mississippi river, in northeastern Iowa, 4 miles south of New Albion, containing incised petroglyphs. Fifteen miles south of this locality paintings occur on the cliffs. He also discovered painted characters upon the cliffs on the Mississippi river, 19 miles below New Albion.

Mr. Edward Miller reports in Proceedings of the American Philosophical Society, vol. X, 1869, p. 383, the discovery of a petroglyph near the line of the Union Pacific railroad, 15 miles southeast of Fort Harker, formerly known as Fort Ellsworth, Kansas. The petroglyph is upon a formation belonging to No. 1, Lower Cretaceous group, according to the classification of Meek and Hayden.

[81]

The parts of the two plates VII and VIII of the work cited, which bear the inscriptions, are now presented as Fig. 44, being from two views of the same rock.

Mr. James D. Middleton, formerly of the Bureau of Ethnology, in a letter dated August 14, 1886, reports that at a point in Union county, Kentucky, nearly opposite Shawneetown, Illinois, petroglyphs are found, and from the description given by him they appear to resemble those in Jackson county, Illinois, mentioned above.

Mr. W. E. Barton, of Wellington, Ohio, in a communication dated October 4, 1890, writes as follows:

A number of inscribed rocks have been found in Maine and information of others has been obtained. The most interesting of them and the largest group series yet discovered in New England is shown in Pl. XII.

The rock upon which the glyphs appear is in the town of Machiasport, Maine, at Clarks point, on the northwestern side of Machias bay,[82] 2 miles below the mouth of Machias river. The rock or ledge is about 50 feet long from east to west and about fifteen feet in width, nearly horizontal for two-thirds its length, from the bank or western end at high water, thence inclining at an angle of 15° to low-water mark. Its southern face is inclined about 40°. The formation is schistose slate, having a transverse vein of trap dike extending nearly across its section. Nearly the entire ledge is of blue-black color, very dense and hard except at the upper or western end, where the periodical formation of ice has scaled off thin layers of surface and destroyed many figures which are remembered by persons now living. The ebb and flow of tides, the abrasion of moving beach stones or pebble wash and of ice-worn bowlders, have also effaced many figures along the southern side, until now but one or two indentations are discernible. Visitors, in seeking to remove some portion of the rock as a curiosity or in striving to perpetuate their initials, have obscured several of the most interesting, and until recently the best defined figures. It was also evident to the present writer, who carefully examined the rock in 1888, that it lay much deeper in the water than once had been the case. At the lowest tides there were markings seen still lower, which could not readily have been made if that part of the surface had not been continuously exposed. The depression of a rock of such great size, which was so gradual that it had not been observed by the inhabitants of the neighboring settlement, is an evidence of the antiquity of the peckings.

The intaglio carving of all the figures was apparently made by repeated blows of a pointed instrument—doubtless of hard stone; not held as a chisel, but working by a repetition of hammerings or peckings. The deepest now seen is about three-eighths of an inch. The amount of patient labor bestowed upon these figures must have been great, considering the hardness of the rock and the rude implement with which they were wrought.

There is no extrinsic evidence of their age. The place was known to traders early in the seventeenth century, and much earlier was visited by Basque fishermen, and perhaps by the unfortunate Cortereals in 1500 and 1503. The descendants of the Mechises Indians, a tribal branch of the Abnaki, who once occupied the territory between the St. Croix and Narraguagus rivers, when questioned many years ago, would reply in substance that “all their old men knew of them,” either by having seen them or by traditions handed down through many generations.

Several years ago Mr. H. R. Taylor, of Machias, who made the original sketch in 1868 and kindly furnished it to the Bureau of Ethnology, applied to a resident Indian there (Peter Benoit, then nearly 80 years old) for assistance in deciphering the characters. He gave little information, but pointed out that the figures must not all be read “from one side only,” thus, the one near the center of the sketch, which seen from the south was without significance, became from the opposite[83] point a squaw with sea fowl on her head, denoting, as he said, “that squaw had smashed canoe, saved beaver-skin, walked one-half moon all alone toward east, just same as heron wading alongshore.” Also that the three lines below the figure mentioned, which together resemble a bird track or a trident, represent the three rivers, the East, West, and Middle rivers of Machias, which join not far above the locality. The mark having a rough resemblance to a feather, next on the right of this river-sign, is a fissure in the rock. Most of the figures of human beings and other animals are easily recognizable.

Peckings of a character similar to those on the Picture rock at Clarks point, above described, were found and copied 600 feet south of it at high-water mark on a rock near Birch point. Others were discovered and traced on a rock on Hog island, in Holmes bay, a part of Machias bay. All these petroglyphs were without doubt of Abnaki origin, either of the Penobscot or the Passamaquoddy divisions of that body of Indians. The rocks lay on the common line of water communication between those divisions and were convenient as halting places.

In the Susquehanna river, about half a mile south of the state line, is a group of rocks, several of the most conspicuous being designated as the “Bald Friars.” Near by are several mound-shaped bowlders of the so-called “nigger-head” rock, which is reported as a dark-greenish chlorite schist. Upon the several bowlders are deep sculpturings, apparently finished by rubbing the depression with stone, or wood and sand, thus leaving sharp and distinct edges to the outlines. Some of these figures are an inch in depth, though the greater number are becoming more and more eroded by the frequent freshets, and by the running ice during the breaking up in early spring of the frozen river.

The following account is given by Prof. P. Frazer (a):

Fig. 45 is a reproduction of Prof. Frazer’s illustration.

This region was also referred to by Dr. Charles Rau (a), his cut from the specimen in the collection of the Smithsonian Institution (Mus. No. 39010) being here reproduced as Fig. 46.

During the autumn of the years 1888 and 1889 Dr. Hoffman visited these rocks, securing sketches and measurements, the former of which are reproduced in Figs. 47 and 48. The figures are deeply cut, as if rubbed down with sand and a round stick of green wood. The deepest channels, varying from three-fourths to 1¼ inches across and almost as deep as they are wide, appear as if cut out with a gouge, and for this reason bear a strong resemblance to the petroglyphs in Owens valley, California. In whatever manner these sculpturings were made, it is[85] evident that much time and great labor were expended upon them, as this variety of rock, locally termed “Nigger-head,” is extremely hard.