PHILLIPS ACADEMY AN DO VER, MA SSA CH USE TTS ?©cpattmcixt of larc^acolog^ BULLETIN I THE EXPLORATION OF JACOBS CAVERN Mcdonald county, Missouri CHARLES PEABODY and W. K. MOOREHEAD 1904 NORWOOD, MASS. PRINTED FOE, THE ACADEMY BY WciZ Nortoool) 5Prcss CORNELL UNIVERSITY LIBRARY MARSHALL H- SAVILLE COLLECTION W €L £4^ St^ 'vw/ V^ CORNELL UNIVERSITY LIBRARY 3 1924 101 546 384 PHILLIPS ACADEMY AN DO VER, MA SSA CHUSETTS ?^epartment of arci^aeologt BULLETIN I THE EXPLORATION OF JACOBS CAVERN Mcdonald county, Missouri CHARLES PEABODY and W. K. MOOREHEAD 1904 NORWOOD, MASS. PRINTED FOE THE ACADEMY BY Efje NorSnooi ^ress The original of tiiis book is in tine Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924101546384 PREFACE The Department of Archaeology of Phillips Academy, Andover, Massachusetts, through the courtesy of the news- papers ^ of the United States, published, in 1902, a circular letter. The purpose was to direct the attention of those especially who reside in the remoter districts to the importance of finding, preserving, and studying specimens of archaeological value. The letter was read by Mr. E. H. Jacobs, an archaeologist of Bentonville, Arkansas, whose kindness prompted him to impart to the Department his knowledge of the existence, near Bentonville, of caves, caverns, and rock-shelters affording traces of former human habitation. He also at the time sent a gift of specimens to the Depart- ment from his own collection, and showed an eager interest and a desire to assist in furthering the plans of the Department for excavation. Accordingly, exploration was decided upon, and during April and May, 1903, cursory examinations of several caves and shelters and a more thorough examination of Jacobs Cavern were accomplished. The Curator of the Department was in this field from April 19 to May 23 ; the Honorary Director from May 12 to May 23 ; after their departure the work was carried on under the direction of Mr. Jacobs. All operations ceased on June 1, 1903. It is with great cordiality that the acknowledgments of the Director and the Curator are paid to Mr. Jacobs for his con- tinued and cheerful assistance in every branch of the work ; 1 Among others of the Sun of Bentonville, Arkansas, iri iv PREFACE to Mr. Samuel Prater and his family for their hospitality; to Mr. J. H. Foster, the owner of the land, for his free permis- sion to excavate ; to Mr. J. L. B. Taylor for his unceasing vigilance at the cavern ; to Mr. W. N. Smith for his accurate photographic work and valuable advice in difficult mechanical situations, and to all the men who worked with the explorers for their good nature and perseverance. Particularly are thanks due to Professor Charles N. Gould of the University of Oklahoma for the gift of his valuable time in spending a week at the Cavern and in preparing a report on its geological features. The preparation of the present Bulletin has been made easier by the information furnished by Professor W. J. McGee of the Smithsonian Institution of Washington and by the cooperation of Professor Frederick W. Putnam of Harvard University and of Dr. H. C. Hovey of Newburyport, Massachusetts. It is also a pleasure to speak of the facilities offered by the Peabody Museum and by the Library of the Department of Geology of Harvard University, which have been freely made use of. The metric system is used throughout the report on Jacobs Cavern. Andover, Massachusetts, January 28, 1904. I. THE ARCHAEOLOGICAL EXPLORATION OF CAVES, IN GENERAL In order to determine the importance relative to archaeology of the exploration of Jacobs Cavern, notice should be taken of work done elsewhere of a similar nature. In Great Britain true palaeolithic man is proved to have existed in many caves. As an illustration, that one called Robin Hood's Cave^ may be cited ; its situation is near the common boundary of Derby- shire and Nottinghamshire. A vertical section of the cave is here given, for purposes of demonstration and comparison : ROBIN HOOD'S CAVE Vertical Section from Roof to Flock Stratum Composition Thickness Age Contents X Stalactite (sic) unit- ing roof to floor . Dark layer of earth 5" to 6" Recent _ ^^ r Mediaeval Pottery J. „ [ Roman f Bones and implements A Stalagmitic breccia 1 Imported flint flakes, (very hard) . . 0" to .36" Pleisto- cene? points, etc.; scrapers Bones broken by man B Cave earth . . . 21" to 52" . Charcoal Flint and quartzite imple- ments f Bones of lion, wild boar, C Red, clayey sand . 24" to 48" ■ Pleisto- cene J etc., marked by hyenas Quartzite hammers and D Light-colored sand, I splinters with limestone blocks .... 24" / No traces of man or wild I animals E Rocky floor . . . 1 Boyd-Dawkius, Early Man in Britain, p. 178. 1 2 EXPLORATION OF JACOBS CAVERN Here traces of man's occupancy were found in strata 5' to 9' in aggregate depth. Remains of palaeolithic man have also been found in Kent's Hole, Devonshire,^ Wookey Hole, Somer- setshire,^ in the cave of Pont Newydd, St. Asaph, North Wales,^ and elsewhere. In France an illustration is furnished by the Grotte de I'Eglise, Dordogne,^ whose section follows : GROTTE DE L'EGLISE Section from Top to Bottom Stratum Composition COKTENTS; Desckiption X Sheet of stalagmite . . "Solutrian"* Bones of reindeer, horse, and ox Bone awls and points . Flint flakes, saws, and scrapers A Layer of black earth . . I "Moustierian," and older B Red earth j Remains of reindeer, cave bear, and horse [ Chipped implements C Yellow sand ... r Remains of bear and bison 1 " Choppers " and flakes of " jasper " Remains of cave man have been found in Belgium in the valleys of the Meuse and the Lesse, in Switzerland on the Saleve and near Schaffhausen, and on the northern continent as far south as Styria.^ Caves in Austria- Hungary are described by L. Karl Moser." A vertical section of one from top to bottom follows : '' ' Boyd-Dawkins, Early Man in Britain, p. 194. 2 Ibid. p. 192. 8 Ihid. p. 198. * Mortillet, Congres. Int. d'Anthrop. et d'Archeol. Trehist., Brussels, 1872. 5 Boyd-Dawkins, Early Man in Britain, p. 204. ^ Der Karst imd seine Ilolile, Triest, 1899. ' Doline und Hohle Vtascajama. EXPLORATION OF CAVES, IN GENERAL Stratum Thickness Dksceiption Contents A 15 cm. Earth Pottery B 30 cm. Ashes Pottery C 15 cm. Stones J Charcoal I Implements {1°^"^^^ D E 40 cm. 40 cm. Red-brown gravel Red earth and limestone Flint chips Few implements Palaeolithic F 1 Stalagmites, etc. So-called "bone-caves" are found in Moravia. ^ Cliff-dwellings in France, unknown till recently, have been noted in the Departments of Aveyron and Ardeche and near the Jura Mountains, — all of the neolithic period ; others of later date are also known. ^ In Crete terra-cotta figurines have been taken from the stal- agmitic formation in the Dictaean Cave ; it does not appear, however, that they have been embedded in the material. ^ In America reports are at hand from Honduras and Yuca- tan. G. B. Gordon* mentions in "Cave 3, Chamber 2" the remains of fire and burnt bones, under a layer of stalagmite about six inches in thickness. E. H. Thompson,^ in his re- searches in Yucatan, found nothing of palaeolithic period, a conclusion also accepted for that country by H. C. Mercer.^ Caves in Peru containing human remains are reported,'^ and the " burial caverns " of Alaska have been visited by Dall.^ 1 R. Trampler, MiUheil. d. Prdhist. (Jomm. dex K. Akad. der Wissenschaften, Vienna, B. 1. 3. 189.3. 2 E.-A. Martel, La Speleologie, Paris, 1900, p. 115. s R. B. Richardson, The Nation, June 18, 1903. * Caverns of Oopan, Mems. Peahody Museum, Vol. I, No. 5, Cambridge, 1898, pp. 7 (143) and 11 (147). 5 Cave ofLoltun, Mems. Peabody Museum, Vol. I, No. 2, 1897, p. 22. « " Cave-hunting in Yucatan," Mass. Inst. Tech. Quarterly, December, 1897, Vol. X, No. 4, pp. 370 and 355. ' Boston Herald, January 24, 1904. 4 EXPLORATION OF JACOBS CAVERN In tlie United States proper widespread efforts have been made, and are at present making, in cave exploration. H. C. Mercer,! in his many investigations, has found nothing truly palaeolithic. O. C. Farrington^ has examined caves in Indiana, and made observations on the rate of growth of sta- lagmites. He has visited Wyandotte Cave, Crawford County ; Marengo Cave, Crawford County; Shiloh Cave, Lawrence County; Cran's Cave, Monroe County. J. R. Nissley^ has described a cave in Hancock County, Ohio. Stalagmitic deposits have been found in Luray Cave, Virginia, and F. W. Putnam* has reported on the Mammoth Cave, Ken- tucky. Stalagmitic deposits are also said to exist near Cavetown, Washington County, Maryland, and in southern California. The article above mentioned, in the Boston Herald,^ inspired by W. H. Holmes, contains a resume of the previous season's work in American caves. Early human occupation is sug- gested as possible in the caves of Grand Gulch, Utah ; mention is made as well of explorations under the auspices of the Smithsonian Institution, in Indiana, Kentucky, Tennessee, Ala- bama, Virginia, Maryland, and Pennsylvania. In addition to the work mentioned, the University of Cali- fornia is at present conducting cave examinations in that state, a work of the utmost detail, under the direction of Professor Frederick W. Putnam. Completing the list of published work, the following titles represent what has already been published on Jacobs Cavern : C. N. Gould : Science, July 31, 1903, pp. 151 S. E. H. Jacobs : Archaeological Edition of Benton County Sun, June 11, 1903. E. H. Jacobs: Quoted in American Antiquarian, September-October, 1903, Vol. XXV, No. .5, pp. 312 ff. 1 Cave Explorations, Eastern United States, Univ. of Penn. Dept. of Amer. and Prehist. Archaeplogy, 1894. Indian House and Durham Gave, Univ. of Penn., as above. Vol. VI, 1897, pp. 147 and 173. 2 Observations on Indian Caves, Field Col. Mus. Public. No. 53, Geol. Series, Vol. 1, No. 8, Chicago, February, 1901. " Americaji Antiquarian, 1888, Vol. 1, p. 40. '' Peabody Museum Ann. Rep. No. 8. ^ January 24, 1904. GEOGRAPHY, ETHNOLOGY, AND HISTORY ■ 5 C. Peabody: American Anthropologist, New Series, Vol. V, No. 3, July- September, 1903, pp. 579-580. C. Peabody: Abstract of paper read at meeting of Archaeological Insti- tute of America, Cleveland, Ohio, December, 1903 ; in American Journal of Archaeology, 2d Series, Vol. VIII, 1904, No. 1, p. 81. It may be said here that work more detailed and complete has been accomplished in the caves of Europe than in those of America, with the result of proving for the former a very much earlier occupation by man than for the latter. In one respect the continents agree, — in the widespread and not unnatural use of cliff excavations, caves, caverns, and rock- shelters as habitations by primitive man. II. GEOGRAPHY, ETHNOLOGY, AND HISTORY Jacobs Cavern is situated on the north bank of Little Sugar Creek, two miles east of Pineville, the county-seat of McDonald County, Missouri. The County is included in the so-called " Ozark Uplift," or elevated lands extending from near Carbondale, in southern Illinois, southwestward through Missouri and Arkansas to near South McAlester, in the Indian Territory. Some of the heights attain an altitude of seven or eight hundred metres, but in McDonald County much of the land forms a plateau of about three hundred metres altitude, intersected by valleys seventy or eighty metres deep. The large number of the valleys, the steepness of the slopes, and the beautiful woods upon them, combine with numerous rushing, clear, cold streams to give a charm to the landscape that is a reminder of some parts of New England. The climate is moderate, rather colder than would be inferred from the latitude, 36° 35' north ; the soil is stony, but some- what fertile in the river bottoms. Fruit raising is, however, the characteristic agricultural pursuit. Freshets of violent and widespread destructive power are not infrequent, the creeks and rivers doubling or trebling their 6 EXPLORATION OF JACOBS CAVERN depth in a few hours. Such floods add an element of uncer- tainty to farm work and the chances of success. Good water and the altitude render the region comparatively ' healthy. Jacobs Cavern may be reached by carriage from Pineville, itself accessible by stage from Lanagan, a station 9 km. distant on the Kansas City Southern Railway ; or the cavern may be approached by wagon road from Bentonville, Arkansas, 30 km. to the south- east. This city is the county seat of Benton County, Arkansas, on the Bentonville Branch of the St. Louis and San Francisco Rail- road. The absence of bridges and the occurrence of the floods above mentioned make communication on occasion uncertain. Who were the first inhabitants of the Ozark region -is not known. The historical inhabitants may have been the Kiowas after their migration southward from the Black Hills. The cavern is included in the district ceded by the Great and Little Osages to the United States on June 2, 1825.^ These Osages, of the Siouxan linguistic stock, are further said to have lived on the Osage River, a hundred miles to the north, and on a tributary of the Arkansas River, perhaps somewhat nearer .^ Of the early explorers, Coronado, on his expedition to Kan- sas in 1540, probably passed to the westward of the Ozark Mountains ; and there is a tradition that De Soto's men, after his crossing the Mississippi, ascended the White River as far as what is now Benton County, Arkansas. The first recorded white settlement near the district was that of the French near the mouth of the St. Francis River, in 1670. A certain Adam Batie was, it is said, the first white man to enter a claim to government land in the neighborhood ; this was shortly after 1819, when Arkansas Territory was organ- ized. It lay near Maysville, northwestern Benton County. In later times the region was in the track of military opera- tions in the war of 1860-65, and the caves were probably of ^ C. C. Royce, Indian Land Cessions, Bur. Eth. Rep. No. 18. 2 Lewis and Clark, Journals, Vol. I, p. 43, Ed. New Amsterdam Book Co., New York, 1902. MINOR CAVERNS AND SITES 7 use in the guerilla warfare characteristic of border sections. A Minie ball was found in the ashes at the rear of Jacobs Cav- ern. At present the immediate valley is somewhat sparsely inhabited by farmers, of whom a considerable proportion are of northern birth. Negroes are almost entirely absent. III. MINOR CAVERNS AND SITES The " Ozark Uplift " abounds in recesses and caverns varying from rock-shelters a few metres in depth to true caves with a length of several kilometres. EDEN BLUFF Under the guidance of Mr. Jacobs, Mr. Moorehead examined some of the caverns of the White River Valley, Arkansas. One of them, known as Eden Bluff, is on the north bank of White River, in the western half of Section 34, Township 19 north and Range 29 west, Benton County, Arkansas. In the bluff proper is a rock-shelter, with an opening about 100 m. in length in the side of the cliff and with a depth inward of perhaps 17 m. ; the bluff has altogether a height of nearly 100 m. The rock-shelter was used in early times for burials, and five or six skeletons have at different times been exhumed; a feature was the " wild hay," dried, found in connection with the burials. This long grass is said to have been abundant on the hills at a time when the country was less heavily wooded. Great damage has been done in the interior by modern searchers after "Spanish gold." The same condition of affairs exists two kilometres up the White River from Eden Bluff. A small cavern at this point has been tampered with at different times, and among the earth and stone excavated were many bones of animals, some of them showing the first stages of becoming fossils. A more detailed examination was undertaken of McElhaney Cavern, situated \ km. east of Monte Ne, Benton County, 8 EXPLORATION OF JACOBS CAVERN Arkansas,^ 8 km. southeast of Rogers. It is on the farm of J. H. McElhaiiey, in the southeastern quarter of Section 27, Township 19 north, Range 29 west. It faces southwest, and its dimensions are as follows : Height from floor to roof, at the opening 3 m. 35 cm. Height from deposits on floor to roof, 6 m. within the cavern . 2 m. 44 cm. Height from deposits on floor to roof, 8 m. 30 cm. within the cavern 1 m. 52 cm. Immediately adjoining McElhaney Cavern to the eastward is a small rock-shelter, the floor of which is 2 m. 13 cm. above the present valley level, but whose roof is a prolongation of the stratum of rock forming that of McElhaney Cavern proper. A deposit of gravel 75 cm. in thickness has been laid down in a cave of which the present small rock-shelter is probably a remnant or successor ; water dripping upon this has caused it to become a firmly cemented breccia. Above the breccia, stalactitic and stalagmitic material filled up much of what was left of space in the cave. The average thickness of the two strata of breccia and of stalactitic material is not far from the same, 75 cm. or more. Behind these and parallel to them, the small shelter extends for some distance in the form of a cave. The filling in of the small shelter in all likelihood took place before the floor of McElhaney Cavern was established, hence, possibly, the com- plete absence thus far of evidences in the breccia stratum of liuman presence. The floor of the larger (McElhaney) Cavern was found cov- ered with black earth 66 cm. to 1 m. in depth; ashes and debris, abundant near the walls, were intermingled. There were no stalactites long enough to reach the floor, and no stalagmites. Animal bones, flint implements, and fragments of pottery were found, not giving evidence of great age. One fragmentary skeleton was found near the east wall, 40 cm. down. 1 Abstract of Report by E. H. Jacobs. GEOLOGY 9 It is probable that the occupation of this cavern was more recent than that of Jacobs Cavern. The abundance of the village sites in the valleys, and in the terraces of the White River region, is shown by the frequent find- ing of projectile points, knives, hammer- and rubbing-stones, etc., and points to a long and full inhabiting by man ; game and fish are not far to seek, and the soil is of a moderate richness. Pictographs are found in the neighborhood of Eden BlufC ; cosmic symbols, tribal and totemic signs, etc., are in evidence. IV. GEOLOGY 1 The hills along Sugar Creek have been caused by the erosion by water of the massive limestone rock. This limestone contains a large amount of flint or chert, usu- ally in the form of nodular concretions, although sometimes occurring in definite layers. This is the Boone chert of geolo- gists and belongs to the Subcarboniferous Age. The limestone is otherwise known as the Mississippian limestone ; it is the rock that covers the greater part of southwestern Missouri, northeast- ern Indian Territory, and northern Arkansas. The lead and zinc mines of Kansas, Missouri, and Arkansas are in this formation. In the lower part of the limestone, flint is often absent, and the massive gray rocks are arranged in definite layers, usually from 10 cm. to 50 cm. in thickness. This part of the formation, known as the St. Joe limestone, is sometimes 20 m. thick, and often is wrought by weathering into characteristic precipitous or overhanging bluffs, extending, it may be, for miles along a stream. The St. Joe limestone is cut at irregular intervals by two series of vertical joints or fissures that divide the rook into rectangular blocks. The principal series of these joints, which may be termed the master- joints, trend approximately north 35° west, and the other series north 20° east. I From the complete report of Professor C. N. Gould. Cf. also his article in Science, July .31, 1903, p. 151. 10 EXPLORATION OF JACOBS CAVEBN The joints may often be distinguished along the face of the cliffs, where they form the division planes between the rock that is still in place and the blocks that have fallen off. Immediately beneath the limestone is a mass of shales known as the Eureka shales, attaining at times a thickness of 15 m. The shales, usually black and papyraceous and worn by weather into thin flakes and tablets, are impervious to water, while the limestone and chert just above are more or less porous; thus there arise thousands of springs which issue between the St. Joe limestone and the Eureka shales ; at this horizon are the famous Eureka Springs of Arkansas. Also between the limestone and tlie shale are numerous caves, formed usually by the wearing away of the soft shales beneath the harder medium. Jacobs Cavern is entirely within the St. Joe limestone, thus differing from the other caves. Immediately above the mouth of the cave the bluff overhangs for about 3 m., and then gradu- ally slopes for 100 m., more or less, to the top of the hill. The cave is truly a rock-shelter, with floor, roof, and walls of limestone, irregularly V-shaped ; it is throughout natural, no marks of human workmanship being visible in the walls or roof. The flat top is composed of a single stratum of limestone, while along the sides of the cave stratification lines are well exhibited. The rock floor is covered to a depth of 1 m. with clay, usually a homogeneous mass, yellowish brown, containing fragments of limestone. Above this was a deposit of ashes. There seems no reason to doubt that the clay is a residual result of the disintegration of the limestone, for, so far as noticed, it has never been dis- turbed, and the line of separation between it and the ashes above is generally sharply marked. Pits dug in different places showed essentially the same clay structure. Near the bottom of the clay the small limestone fragments are more nximerous than above, while at the top they are practically wanting. GEOLOGY 11 At the back of the cave is a fissure, extending upward from the roof to a height of 3 m., separating the roof of the cave from the rear wall. The fissure, probably a master-joint of the series described above, is from 1 m. to less than 1 m. wide, and continues along the back of the cave beyond the main part> forming a narrow recess, which in turn extends for about 5 m.i Along this fissure, and also along the back of the cave where the fissure does not extend, are stalactites, stalagmites, and pilasters, formed by the action of water dropping from the roof. In places the entire fissure above the level of the roof is choked with stalactitic material. The continued dropping of the water carrying CaCOg in solution upon the ashes cov- ering the floor has formed a sort of stalagmitic ash-breccia, often enclosing fragments of flint or sandstone, flint imple- ments, bones, charcoal, and other material similar to that found buried elsewhere in the ashes. To the mind of the writer, there is no doubt that the ash- breccia was formed very slowly during and after the deposition of the ashes. The peculiar toadstool-like shape of some of the pillars, the like of which he has never observed in any other cave, appears to point indisputably to this conclusion. After a careful ex- amination the writer assumes that the following process has taken place : Ashes mingled with bones, flint, and charcoal were deposited upon the floor ; then the water dropping from the roof formed the stalagmitic breccia, which spread out in a mass in the shape of a toadstool. Around or upon this sta- lagmite other ashes and charcoal were deposited, and a second toadstool was formed by the water; in some cases even a third may be seen. Finally, when the deposition of the ashes ceased, the stalag- mite continued to grow until it joined the stalactite from above, forming a pilaster. Near the back of the cave, particularly underneath the fis- 1 "Bone-recess" on the map. 12 EXPLORATION OF JACOBS CAVERN sure, the greater part of the ashes and some of the clay cover- ing the limestone floor have been cemented by the action of CaCOj, forming ash-, clay-, and limestone-breccia, often very firm and solid. In other parts of the cave both ashes and clay are soft and easily moved. A number of blocks and slabs of limestone were found on the surface of the ashes, or embedded in them or in the clay beneath. They have evidently fallen from the roof, some before man's occupancy, others during it, and still others quite recently. Of the sandstone fragments and flint flakes in the ash stra- tum, there seems no doubt that all were carried into the cave from outside. (The possibility of their having entered from above through the fissure at the back is rendered small, first, by their great number, second, by their even distribution throughout the cavern, i) The nearest sandstone outcrop on the surface is, so far as could be determined, 6 km. distant, near White Rock, although small sandstone boulders and pebbles are occasionally found on the gravel bars of Sugar Creek. Whatever the source of supply, man has necessarily brought the sandstone specimens into the cavern. As to the thousands of flint flakes, varying from small "spalls" to pieces the size of the hand, it was at first thought that they might have fallen from the roof ; careful search, however, failed to detect the presence of flint in roof or walls. Hence (outside of the slight possibility of their having en- tered by way of the flssure) it is believed that the flakes and implements have all been carried into the cave or produced within it by human agency. Much of the flint was obtained from the hills near by ; but judging from the lithological character of other pieces, it is evident that they have been brought from a distance, some of them, probably, from the flint hills of central Kansas. 1 Editors' note. DESCRIPTION OF JACOBS CAVERN 13 V. DESCRIPTION OF JACOBS CAVERN METHODS OF EXCAVATION Jacobs Cave opens on a hill slope forming the northwestern boundary of the valley of Little Sugar Creek. It is 12 m. in perpendicular distance above the valley floor and a former main channel of the creek. The opening, like that of most of the caverns of the district that show occupation, is toward the southwest. The length of the opening from northwest to southeast is 21 m. 20 cm.; the greatest depth inward, measured from south- west to northeast, is 14 m. The " bone-recess " extends northwestward from the north- western angle of the cavern for 5 m. The height of the roof above the undisturbed surface of the ash deposit varied according to the place where measurements were taken. In front at the eastern end of the opening the height was . . 2 m. 10 cm. In front at the "western end of the opening the height was . 2 m. 60 cm. Inward 7 m. from the opening at the east side the height was 1 m. 40 cm. Inward 7 m. from the opening at the west side the height was 1 m. 30 cm. At the northwest corner in front of the fissure the height was 1 m. 20 cm. The " bone-recess " varied from 50 cm. to 80 cm. in height. In the exploration of Jacobs Cavern four things were made the subjects of study : the layer of fine ashes covering the floor, the underlying clay stratum, the stalactites and stalag- mites, and the "bone-recess." The layer or stratum of ashes was from 50 em. to 1 m. 50 cm. in thickness. The method of excavation was, first, to divide the surface of the ashes into square metres by rows of stakes, and then to remove the ashes in order, front to back, using the lines of stakes as coordinates to determine the position of any objects found. The linear distances of 1 m. were numbered from northwest to southeast in Arabic numerals from 1 to 21, and lettered from southwest to northeast from A to Q. The amount and the contents of the ash stratum will be 14 EXPLOUATION OF JACOBS CAVERN discussed later. Their character was such as when disturbed during the work seriously to affect the throats and lungs of most of the workmen. Wet sponges tied over the mouth proved effective as preventives against a recurrence of the trouble. A difficulty during the removal of the ashes was presented by the large blocks of limestone, fallen from the roof since the beginning of the formation of the ash stratum. Some of them weighed over 1000 kg., and a block and tackle were required for their removal, as well as the services of strong and expert stone-breakers. The exceedingly irregular presence and the partial clearing away of these blocks interfered greatly with a regularity of excavation, readily adhered to in mounds and sites of homo- geneous composition. The presence of skeletons and of large bones and specimens could, of course, be noted on the first disturbance of the ashes. For the detection of small implements and fragments of bone, stone, and pottery, men were stationed with hand trowels to examine the ashes as they were placed on the outside slope from the wheelbarrows used in their removal. To determine the composition of the clay stratum underly- ing the ashes, pits were sunk thi-ough it to the true limestone floor of the cave at metres B/C-9/10, J-10, and G-14. The same succession and the same composition of strata were observed in all cases. A typical vertical section may be thus shown: Top to Bottom [ a. Ashes 71 cm. b. Ashes with limestone blocks . . ... 66 cm. 2. Stratum B. Clay mixed with limestone fragments . . 94 cm. 3. True limestone floor. 1. Stratum A. j A trench at the back of the cavern in front of the row of stalagmites was dug in the clay stratum (B). From the evidence of the trench and the pits the conclusion was arrived at that no traces of human occupancy were con- cealed in the clay or lower down than the ash stratum (A). At the back of the cavern is a row of stalactites, stalagmites, and stalactite-stalagmites. DESCBIPTION OF JACOBS CAVERN 15 The following table gives the dimensions and some of the characteristics of the most important of these ; most of the measurements were taken by Mr. Jacobs.^ No. 'J II O O K K b IS. B £ S 2 * H a 5 2 2 H R Notes S.-S. 1 . . . . 1-18 m. cm. 2 80 m. cm. 1 20 m, cm. 1 00 m. cm. 1 10 With an upper and a lower "stool" 60 cm. below a cor- Stal™-2 . . . J-17/18 70 30 70 responding sta- lactite S.-S. 3 ijower "stool" K-16 2 50 80 00 93 S.-S. 3 Upper K-16 40 40 60 Entirely below ash- level. The summit Stal"-4 . . . K-14 4 70 80 00 1 50 is 1 m. 25 cm. be- low a correspond- ing stalactite Entirely below ash- level. The summit Stal™-5 . . . L-11 6 30 70 00 2 20 is 1 m. 20 cm. be- low a correspond- ing stalactite ■ Entirely below ash- level. The summit Stal™-6 . . . L/M-9/10 4 70 55 00 1 70 is 1 m. 30 cm. be- low a correspond- ing stalactite S.-S. 7 . . . . N-6 2 60 Stal»- 1 00 00 1 20 f Stalagmite entirely [ below ash-level S.-S. 8 . . . . 0-6 1 05 1 50 Stalagmite entirely below ash-level r "Stools" of stalag- S.-S.9 . . . . 0-2/3 1 55 mites 10 cm. to 30 cm. below ash- level 1 Abbreviations : Stal™- stalagmite. : Stalagmite ; Stal"'- = stalactite ; S.-S. = stalactite- 16 EXPLORATION OF JACOBS CAVERN Some of the stalactite-stalagmites were attached on one side to the wall, pi'esenting the appearance of half-columns. One stalagmite (No. 5) was taken to pieces and taken away, as well as one pendent stalagmitic "stool." At the close of operations most of the stalagmites were thickly covered again with soil and stones ; this was a precau- tionary measure of preservation. The following abstract from Mr. Jacobs's report on ' the "bone-recess" is based on the work done after the departure of the directors. 1 The " bone-recess " extends 5 m. in a northwesterly direction from the northwestern corner of the cavern. At the entrance are several irregular stalagmites numbered 7 and 8 ; their sum- mits are all below the original level of the ashes in the cavern. Outside of this stalagmitic barrier the ashes, when examined, showed the same contents as elsewhere in the cavern, but with an increasing percentage of animal bones near the barrier. Specimens of the material and contents of stalactite-stalag- mites 7 and 8 were obtained, and then work was begun upon a soft deposit within the recess just back of the barrier ; this was found to contain decayed organic matter, but fewer ashes ;' what bones and implements there were, were similar to those from the larger cavern. In square metre 0-4/5 a pit was sunk down to the underlying clay (Stratum B); the soft deposit above was 1 m. 5 cm. in depth, and contained stones of various sizes up to a thickness of 30 cm. At a distance of 1 ra. 50 cm. back from the entrance a heavy stratum of animal bones was met, embedded in the soft deposit above a large flat stone. This continued for nearly 1 m. backward into the recess. One bone awl and a few flint chips formed part of the bone stratum, which itself was dry and in probably the driest part of the recess. In the rear there were many stones, 10 cm. to 40 cm. down in the soft deposit. This part is damper, and the ^ The complete report may be consulted at the Department of Archaeology of Phillips Academy. HUMAN BUBIALS 17 decomposition of the bones may account for their lessened fre- quency here. A stalactitic-stalagmitic mass adhering to the south wall was numbered stalactite-stalagmite 9. A cross-sectipn of the " bone-recess," 80 cm. in from the en- trance, shows the recess already described to be' at that point but the lower of two chambers. The second chamber is perpendicularly above the first, and is separated from it by two blocks of fallen limestone. Of these the lower, 55 cm. thick, forms the roof of the lower chamber and a part of the floor of the upper ; the upper block, lying partially on the lower, forms the remainder (on a differ- ent level) of the floor of the upper chamber. The walls of the upper chamber are lined with stalactitic deposits ; there are none depending from the roof. Animal bones, "bone dirt," and stalactites enclosing bones were found in the upper cham- ber, but no stone implements. The total height from the clay (Stratum B) to the roof of the upper chamber is 3 m. 20 cm. at a point 80 cm. within the entrance. VI. HUMAN BURIALS Six human burials were found in the ashes of Jacobs Cavern, from 45 cm. to 1 m. 30 cm. down. They were in poor preservation, and in cases incomplete. The known types of Indian burial called the "bundle" type and the " scissors " type were represented ; these types have also been observed in the mounds of Mississippi. A detailed description is subjoined. Skeleton No. Position Depth Orientation by End at which THE Skull was found m. cm. 1 A-15 80 West 2 D-14/15 1 30 South 3 P-11 45 North 4 F-15 70 Northeast 5 J-8 50 Southwest 6 I-IO 1 30 North 18 EXPLORATION OF JACOBS CAVERN VII. THE BONES OF ANIMALS Seven-tenths of a cubic metre of animal bones were removed from the cavern; they are whole or in fragments ; most of the latter have the appearance of having been split for marrow, others of having been boiled or otherwise cooked. The following animals have been identified ^ from bones in the collection at Andover ; they are contemporary with man's occupancy of the cavern : Deer (Cariacus virginianus). Gray Wolf (Canis nccidentalis). Raccoon {Procyon lotor). Elk (Cervus canadensis). Opossum {Dldelphys virginiana'). Buffalo (Bos americanus). Bear (Ursus americanus). Woodchuok (Arctomys monax) . Beaver {Caslor canadensis) . Tortoise {Cistudo Carolina). Turkey (Meleagris gallopavo americanus). Hog (Sus scrofa domestica). VIII. THE IMPLEMENTS OF STONE The collection at Andover may be considered characteristic, though not complete ; there are in the Department of Archae- ology a collection from Jacobs Cavern, two surface collections, one carefully made by Mr. J. L. B. Taylor, the other gathered by the boys of Pineville, and a small collection from McElhaney Cavern. The remainder of the specimens from Jacobs Cavern are, the majority of them, in the possession of the Smithsonian Institu- tion of Washington or in that of Mr. Jacobs himself. From Jacobs Cavern the larger implements were almost wanting. Three stone metates (one of sandstone and one of them broken), one stone axe, one celt, and fifteen hammer-stones were found. In projectile points and knives the collections are rich. A rough classification follows. In many cases a rejected, broken, or incomplete implement of one kind may have been used as 1 By Dr. W. C. Farabee of Harvard University. THE IMPLEMENTS OF STONE 19 a perfect implement of another. The hard and fast lines of division vary with the personal equation of the examiner, but the general proportion of types is a correct one. THE COLLECTION FROM JACOBS CAVERN 1. Implements, knives, or projectile points, when complete, 5 cm. or more in length : A. Without stem : a. More or less circular in form ^ 50 b. Oblong, generally without shoulder I "^ Complete ... 13 l p. Incomplete ... 10 c. Flakes, generally long and thin 53 d. Pointedj^^'f °'^'',^"'^, 31 [ p. At both ends ^ 3 e. Scrapers, " humped type " 11 Total 1. A 171 A subdivision of 1. A. rf. a (pointed at one end) : f aa. Base concave 4 More or less triangular J "^- ^^^'^ straight 3 ay. Base convex 3 I aS.' Base irregular 3 Leaf-shaped. ...["'■ (S«i|j^s expanding from both Total 31 B. With stem : a. Stem tapering: T, f . /Barbed 8 a. Base of stem concave -^ „ , , , , L Not barbed 11 B. Base of stem straight -I „ , , ' ' ^ ° [ Not barbed 18 T, , , f Bai-bed 1 ■y. Base oi stem convex ]-.,,,■,-, ' [ Not barbed 12 S. Base of stem pointed | „ , , ! ' ^ [ Not barbed 7 £. Base of stem irregular 2 Total 1. B. a 56 1 Some of these, doubtless, were scrapers. 2 sides convex. 8 "Barbed" is defined as having an acute or reentrant angle between the stem and the shoulder or shoulders. 20 EXPLORATION OF JACOBS CAVERN b. Stem with sides straight : „ * . ^ r Barbed u. Base of stem concave i „ , , , , o L Not barbed o B. Base of stem straight -i „ ^ , i i n '^ ^ l Not barbed „ „ , /Barbed 1 V- Base of stem convex i -.r ^ ■, ■, -, o ' L Not barbed . 3 8. Base of stem irregular 1 Total l.B.b 10 c. Stem expanding : „ ,. ^ r Barbed 4 u. Base 01 stem concave i -^t , , •■ , L Not barbed 11 B. Base of stem straight i „ , , ' ' ^ ^ I Not barbed 2 „ f , f Barbed. . . . i 25 v. Base or stem convex \ „ , , , , [ In ot barbed 8 8. Base of stem irregular S Total 1. B. c 59 C. Shouldered knives : „ , , fa. Chipped on both sides 8 a. Broad type -i „ „ „ 1 1 „ i • [ p. " Spalls, or chipped on one side only . . 3 f nu- J 1. ii_ • 1 f Two cutting edges . 37 , T J. a. Chipped on both sides -^ „ , , . , 6. Long type i ^'^ [ One cutting edge . 6 [ /3. " Spalls " or chipped on one side only . . 4 Total 1. C 58 D. Perforators 5 E. Doubtful form 1 2. Implements, when complete, less than 5 cm. in length : A. Without stem : T^ . J , fa. Chipped on two sides 4 a. Knives, round type -^ „„,•,„,■ , [p- " Spalls, or chipped on one side only , 9 b. Knives, long type {"t Chipped on two sides 1 LP- "Spalls," or chipped on one side only 6 a. Concave /3. Straight y. Convex . c. Points, triangular or leaf-shaped with sides Incomplete form 1 Total 2. A 29 THE IMPLEMENTS OF STONE 21 B. With stem : r a. Base of stem concave ] a. Stem tapering U- Base of stem straight 1 [ Barbed . . j y. Base of stem convex . 5 | Not barbed . 7 I 8. Irregular 1 J Stem with sides straight a. Base of stem concave 1 „ , li. Base of stem straight 1 I- ^, , ' ' '^ -Q f , 1 I J^ ot barbed . 1 y. Base of stem convex . 1 J ( a. Base of stem concave 5 ] „ , , c. Stem expanding \ B. Base of stem straight 6 I ..^ '' , , " , ' I -D . , £. Not barbed . 4 [ y. iiase 01 stem convex . 6 J d. " Bunts " 4 Total 2. B 30 Summary of the Collection 1. Implements 5 cm. or more in length : A. Without stem 171 B. With stem : u. Stem tapering 56 b. Stem straight 10 c. Stem expanding 59 125 C. Shouldered knives 58 D. Perforators 5 E. Doubtful 1 Total 1 (A, B, C, D, E) 360 360 2. Implements less than 5 cm. in length : A. Without stem 29 B. With stem : u. Stem tapering 7 b. Stem straight 2 c. Stem expanding 17 26 d. " Bunts " ' . . . . 4 Total 2 (A, B) 59 59 Total Jacobs Cavern collection i 419 1 In addition to these, hundreds of cores, fragments, "spalls," and "rejects" of flint were brought to Andover for purposes of study and comparison. 22 EXPLORATION OF JACOBS CAVERN THE SURFACE COLLECTION OF J. L. B. TAYLOK A. Implements without stem : a. Knives, round form 25 h. Knives, leaf-shape H c. Knives, long form 11 d. Knives or points, triangular 17 e. Points, triangular, small 3 f. Scrapers 3 70 B. Implements with stem : a. Stem tapering 39 h. Stem straight .... 17 c. Stem expanding' ... 99 155 C. Shouldered knives .... .... 41 D. Perforators 2 E. "Bunts" o F. Small stemmed points 10 Total J. L. B. Taylor collection 283 Incomplete Ob.jects Knives, round 11 Knives, leaf-shaped . 1 Knives, long form 12 Triangular forms 3 "Spuds" 2 Celt ' 1 30 SURFACE COLLECTION FROM THE VICINITY OF PINEVILLE A. Implements without stem : a. Knives, round 13 h. Knives, leaf-shaped 3 c. Knives, long form .... 2 il. Knives or points, triangular 3 e. Scrapers 3 24 B. Implements with stem : a. With stein tapering 33 h. With stem straight 7 c. With stem expanding 82 d. Lance-point, stem expanding 1 123 C. Shouldered knives 30 D. Perforators 4 B. Axe, chipped 1 Total Pineville collection 188 THE IMPLEMENTS OF STONE 23 THE COLLECTION FROM McELHANEY CAVERN Knives or points, triangular 4 r Tapering 9 Knives or points with stem \ Straight [ Expanding . 5 Total McElhaney collection 18 FROM EDEN BLUFF Knife or point with stem expanding 1 Noteworthy, from an examination of these lists, are the fol- lowing facts : First, the large number in proportion of knives of the round type; fiat, rough, often chipped to the edge on one side only, these objects seem yet to have the right of being considered complete implements. Second, the large proportion of knives, more or less rudely chipped, of the long, shouldered class ; often a natural chip struck off from a core or a mass of flint will take a form almost ready for use, as such, as an implement. Third, the comparative abundance of specimens with stems ; in the Jacobs Cavern collection the stemmed specimens are 155 out of a total of 419, or 37 per eent.i Fourth, the small proportion of specimens less than 5 cm. in length. Fifth, the fact that while in the Jacobs Cavern collection, among the specimens with stem, there are with stems ex- panding 76 out of a total of 151,^ or 50 per cent, in the J. L. B. Taylor collection they are 99^ out of 156, or 64 per cent, and in the Pineville collection they are 83 out of 123, or 67 per cent. This is a distinction possibly arising from a nai've selection in collecting surface specimens. 1 Shouldered knives are not considered stemmed. 2 " Bunts " are here not counted. 3 "Small tanged points " are here not counted. 24 EXPLORATION OF JACOBS CAVERN Besides the collection at Andover there have been taken from Jacobs Cavern the following : Knives . 5 Knives or projectile points 21 Projectile points (smaller) 7 Perforators 3 Hammer-stones, etc 24 Polishing stones 2 Metate, limestone 1 Metate, sandstone (broken) . . 1 Unfinished specimens 177 241 Also there were removed hundreds of chips, " spalls," and cores showing human workmanship. The rarity in this region is to be noticed of the larger " neolithic " implements, such as celts, grooved axes, pestles, " spuds," " plummets," and pipes ; also the total absence of the so-called "ceremonials." IX. POTTERY AND BONE Fragments of pottery were present, but not abundant, in the ashes of Jacobs Cavern. No complete vase, whole or in parts, was found. The type of pottery was rude, with primitive decoration and no attempt at coloring save by firing. The tempering was, as usual, of shell. From the shape of the fragments the vases seem to have been, many of them, bowls or " pots," swelling considerably in the middle. The most usual implement of worked horn or bone was the awl or needle of the long, tapering type. They were present in considerable numbers in the ashe§. The stalagmitic deposits, the " polished rocks," and the out- side evidences of habitation may better be considered later. EVIDENCES AS TO THE OCCUPATION BY MAN 25 X. EVIDENCES AS TO THE TIME, DURATION, AND CHARACTER OF THE OCCUPATION BY MAN OF JACOBS CAVERN The evidences of man's occupancy may be classified as follows : 1. The quantity of the ashes. 2. The type of human remains. 3. The type of animal remains. 4. The tjrpe of implements and fragments. 5. The stalagmitic deposits. 6. The polished rocks. 7. Outside evidences. 1. At an estimate probably much below the true figure, there were 115 cu. m. of ashes in Jacobs Cavern ; they covered the entire clay stratum (B) to a height of 50 cm. and more with a fairly homogeneous mass. The presence of ashes in such quantities precludes the assumption of purel)' natural causes. Fires produced by spontaneous combustion or by lightning, either inside or outside of the cavern, could not produce such a mass within and leave no trace without. While no exact calculation of the rate of accumulation of ashes may be hazarded, there are certain determining conditions. First, the daily occupancy of the cavern was probably longer in winter than in summer; hence a longer time would be required for the production of the ashes than were a constant fire to be assumed for the purpose of cooking and heating. Second, the fires must have been either outside or inside: if outside, but a part of the ashes have been preserved ; if in- side, the conditions of smokiness must have necessitated small fires ; in either case a long time of human occupancy must be assumed. Third, a singular persistency in occupation is observed. Be- fore excavation, the surface of the ashes was less than 2 m. 26 EXPLORATION OF JACOBS CAVEBN below the roof. The inhabitants must have found their later presence in the cavern to necessitate either a cramped position or a very limited floor space ; it is conceivable, therefore, that the later accumulation went on more slowly, owing to more infrequent occupation. The testimony of the ashes is, then, to establish long occu- pancy, not necessarily continuous, of the cavern. 2. Nothing indicating great antiquity has been inferred from examination of the human remains. It is to be considered, however, that among the fragments of skeletons found in Jacobs Cavern the maximum femoral length is 43 cm., not that of a large man. The Osages, the historical owners of the region, were accord- ing to Catlin the " largest and best-appearing " Indians seen by him. It is possible, therefore, that the burials are of an- other, hence earlier, people. ^ From the carelessness and irregularity of the burials com- pared to the usual carefulness in this regard of the Red Indian, it is unlikely that these were deposited during the time of the apparently leisurely occupancy while the ashes were accumulating. Occasional burials, much fewer in number than the deaths occurring in the course of nature, seem unlikely to have been made in the growing refuse-heap. It is more probable that the six burials are posterior to the deposition of the ashes. The occupants producing the ashes would then appear to have been earlier than those who made the burials, and these in turn to have preceded the Osages. The scant evidence, accordingly, of the skeletons is toward the greater antiquity of the earlier occupation. 3. No bones of extinct animals were found ; all are of com- paratively late animals of historical times. The presence of the bones of domestic animals need not appear surprising; dogs or wolves can at any time readily 1 Cf. Catalogue Baisonne de la Galerie Indienne de Mr. Catlin, Paris, 1845, p. 8, s.v. " Osages-" EVIDENCES AS TO THE OCCUPATION BY MAN 27 carry bones away from the farmyards and deposit them in the cavern. The evidence, then, of the , animals fails to show any great antiquity in any part of the ash stratum, and, from the absence of testimony to man's occupancy below this stratum, the ashes in their gradual accumulation may safely be taken to be repre- sentative of the whole of such occupancy from beginning to end. 4. The types of stone implements are quite different from those of the neighboring Arkansas-lower-Mississippi basin. They are here ruder in form and finish, and the small arrow- and spear-points of the lower region are almost absent. The large proportion of very rough knives — round, oblong, shouldered, and not shouldered (often by haphazard) — char- acterize the Ozark district, and are almost sufficient in them- selves to determine a race of occupants different from the so-called "mound-builders." This distinction is enforced by the absence of the finer pot- tery, as characteristic itself of the Arkansas-Missouri culture as are in Jacobs Cavern the knives. To one versed for years in excavation, there comes a certain inexplicable feeling that the specimens from Jacobs Cavern "look old" in comparison to the mound specimens. While again no proof is forthcoming, the tendency of what the specimens have to say is toward the greater antiquity. 5. The nine stalagmites, so far as observed, contained, from their base on the limestone clay up to the level of the ashes at the time of excavation, charcoal, animal bones, and flint chips in such numbers and embedded in such a way as to prove to the excavators the presence of man in the cavern during the period requisite for the formation of that part of the stalagmites. The deepest specimen found was buried 50 cm. perpendicularly down from the stalagmite's surface. It remains to investigate the rate of growth of stalagmites, and unfortunately absolutely nothing can here be adduced with precision. 28 EXPLORATION OF JACOBS CAVERN Observations made elsewhere are pertinent. In Ingleborough Cave, Yorkshire, in the case of the stalag- mite "Jockey Cap," through continual dropping the rate of increase was annually, from 1845 to 1873, 0.2946" or 0.75 cm. in height.-' A stalagmite in Wyandotte Cave,^ Indiana, has been observed to increase annually 0.01" or 0.0254 cm. Taking a stalagmite 50 cm. in height, if the rate of growth were that observed in Ingleborough Cave, 66 years would be required for its formation; if that in Wyandotte Cave is assumed, 1968 years would be required. Hence we are forced to agree with Boyd-Dawkins,^ Gordon,* and Martel^ that thickness of stalagmitic deposit is of com- paratively little moment in the determination of its age. That this is so, it is only necessary to consider the determin- ing factors in the growth of a stalagmite : Rainfall. Vegetation and the character of the surface soil. Composition of the limestone medium. Size of the crevices througli which trickling takes place. Rate of evaporation within the cave. Presence of diverting bodies under the water flow. Shifting of the stalactitic point of departure. 6. Outside and near the entrance to Jacobs Cavern are large rocks, presenting on large parts a brilliant polish, quite differ- ent from M'eathering and akin to that found especially on the " spades " and " hoes " of Tennessee, produced, we may sup- pose, by friction between the implement and skins or some other oily substance. That the rocks have been polished by the naked bodies or the skin clothing of human beings becomes more probable when ■■ Boyd-Dawkins, Cave-Hunting, pp. .30-40. 2 H.C. Hovey, Celebrated American Caverns, pp. 137, 130, 191. ^ Of ive- Hunting, p. 40. ■• Caverns of Copan, Mems. Peabody Museum, Vol. I, No. 5, p. 12 (148). ^ La Speleologie, Paris, 1900, pp. 102-103. CONCLUSION 29 we find that, though a few other rocks with a similar polish exist in the Ozarlv district, they are not present where other evidences of man's occupancy are lacking. The polished rocks indicate a long occupation. The only similar cases known to the explorers are provided by the walls of the stone gallery at Tiryns, where the polishing is said to be due to the herding of sheep for centuries in that celebrated place. 7. The historical owners of Jacobs Cavern were the Osage Indians, a Plains tribe owning and rejoicing in horses in white men's times. At Eden Bluff are pictographs upon the rocks, presumably done by the occupants of caves in the vicinity. These contain the usual cosmic symbols, but no suggestions of horses; the same absence of horses elsewhere, notably in the Southwest, has been noted. It is probable that either the pictographs are the product of another, and therefore earlier, race than the Osages, or that they were done in prehistoric or protohistoric times by the Osages themselves. XI. CONCLUSION The evidence from the quantity of the ashes, the types of implements, the stalagmitic deposits, is toward the assumption of a very early and protracted occupancy of Jacobs Cavern by man. That the occupants were different from the Osages and also from the lower Mississippi tribes is negatively suggested by the human remains, the pictographs, and again by the types of implements. The polished rocks point to a long occupation, and its date and length, while not supported, is not denied by the animal remains. An early inhabiting of the cavern by man, who continued to abide there, perhaps hundreds, perhaps thousands, of years, is all that may at present be asserted. o Q- o o o 5 o CO o o PLATE V JACOBS CAVERN: KNIVES OF STONE. SIZE i PLATE VI JACOBS CAVERN: KNIVES OF STONE. SIZE { a. D- I- o if) X o -< (- tf> >- _I -X. o CO CQ o o (3 Z Q Z -4 Q. I- cn I _J Q- O U X o O o