FIRST ANNUAL REPORT: -f^^^f^- 2; 913.773 F519 AMERICAN BOTTOMS ARCHAEOLOGY JULY I, 1961- JUNE 30, 1962 by MELVIN L. FOWLER, EDITOR, Southern Illinois University CHARLES BAREIS, University of Illinois DONALD LATHRAP, University of Illinois BRIAN CARSS, University of Illinois JOSEPH VOGEL, Illinois State Museum WARREN WITTRY, Cranbrook Institute JAMES PORTER, Southern Illinois University JAMES SCHOENWETTfR, Southern Illinois University A report of research under the auspices of The National Science Foundation (G17922) And the Illinois Archaeological Survey Based upon field excavations carried out under the auspices of The Bureau of Public Roads, U. S. Department of Commerce The Division of Highways, Illinois Department of Public Works and Buildings The Illinois Archaeological Survey 137, Davenport Hal 1 University of Illinois, Urbana, Illinois Cooperating Institutions; The Illinois State Museum The University of Illinois Southern Illinois University Springfield. Illinois Urbana, Illinois Carbondale, Illinois :ll!MO!s historical survey FIRST ANNUAL REPORT: AMERICAN BOTTOMS ARCHAEOLOGY JULY I, 1961 - JUNE 30, 1962 ALTON . ILLIWCI*^ WOOD BIVEB H0X4N4 MARTFO«0 POAG N4ME0KI GRANITE CIIV MADISON VENICE BROOKLYN NATIONAL CITY FAIRMONT CITY COLLINSVILLE CASEYVILLE WASHINGTON PARK EAST ST LOUIS BELLEVILLE CAHOKIA EAST CARONDELET OUPO, ILLINOIS LIMITS ^,,,, ',j 'M-^S^Mi;^EM:^MM^^!mm^^^\o ..^.^Lj ^...^.r F 5/'? INTROmiCTTON This report sutninartzes the findings accruing from the first year of research under National Science Foundation Grant 17922. This grant was given to the Illinois Archaeological Survey for the purpose of ana- lyzing data that was accumulating on one of the most significant archnco- logical areas in North America. This area, known in the Midwest as the American Bottoms, is a large pocket of fertile land in the Mississippi river valley centering around the area of St. Louis. In prehistoric times, as it Is today, this was a large population center. The American Bottoms area is best know archaeologically for the site of Cahokia or Monks Mound. The site, known by either of these names, is a large cluster of platform mounds of Mississippian culture located in the central portion of the American Bottoms (see Figure 1). Central to this large mound group is Monks Mound itself which covers about 16 acres and rises over 100 feet above the valley floor. Scattered throughout the American Bottoms are archaeological sites composed of mounds and habitation areas that relate culturally to this central group. The Mitchell Group on the north near Granite City, Illinois, and the Pulcher Group on the south near the town of Dupo, Illinois, seem to represent the outliers of this occupation and demark the zone of popula- tion concentration (see Figure 1). It has been known for many years that this was an area of sig- nificance archaeologically but its tremendous size and the lack of money for field work have kept archaeologists from detailed investigations. Moorehead (1928) did some cursory field work in the area and he was followed by several others in the thirties. Griffin (1949) documented the basic ceramic sequence that had developed out of these studies. In the past five years there has been an intense campaign of highway construction in this area due to the fact that several of the major east-west and north-south Interstate highways of the new Federal highway system cut through the St. Louis area. These modern highways often have right-of-way areas several hundred feet in width. Because of the many other highways, railroads, and streams In this area, and because the Interstate highways are all limited access highways, it has been necessary to have large interchanges and overpasses at rather close Intervals. These overpasses and Interchanges require thousands of cubic yards of fill to raise the elevation of the highway sufficiently to pass over the obstructions. This type of construction has already destroyed completely some archaeological sites. Fortunately, in Illinois an Archaeological Salvage Program for the Interstate Highway areas had been established before the construction in the American Bottoms area had gotten very far along. This program carried out by the Illinois Archaeological Survey in cooperation with the U. S. Bureau of Public Roads, United States Department of Commerce and the Division of Highways of the State of Illinois Department of Public Works and Buildings has yielded tremendous quantities of inval- uable archaeological data. This destruction of archaeological sites and the consequent accumulation of data due to the cooperative archaeological salvage pro- gram, coupled with the fact of almost daily obliteration of archaeologi- cal data due to industrial and suburban expansion in the area, prompted the members of the Illinois Archaeological Survey to request support of the National Science Foundation to gather and analyze the archaeological data before it was completely lost. While reservoirs merely inundate archaeological sites, highway construction and industrial and urban expan- sion are total destruction in terms of archaeology. Fortunately, the National Science Foundation agreed, and thus was launched a program of an integrated study of the archaeological data from the American Bottoms. In- the Archaeological Salvage Program, the three cooperating institutions carried out field work in different areas of the American Bottoms. The Illinois State Museum Field Parties, under the direction of Warren Wittry, worked in two major areas, known as Tracts 15A and 15B, both within 3000 yards west of Monks Mound. The University of Illinois Field Parties, under the direction of Donald Lathrap and Charles Bareis, worked in the vicinity of the old Powell Mound about two miles west of Monks Mound. The Southern Illinois University Museum Field Parties, under the direction of James Porter, worked at the Mitchell Site about seven miles north of Monks Mound. In the following sections these individuals summarize the material recovered from those excavations. Analysis of these data is really just beginning so detailed published reports on them must be deferred to a later date. Under the National Science Foundation grant each of the cooper- ating institutions undertook to carry out certain specialized research working with materials recovered throughout the American Bottoms. The University of Illinois undertook a study of clay-minerals used in pot- tery manufacture. This work is reported on by Brian Carss in Chapter III, The Southern Illinois University undertook to establish a pollen labora- tory and a lithic laboratory. James Porter reports on the Lithic Labor- atory and James Schoenwetter reports on the results of the first year of operation of the Pollen Laboratory in Chapters V and VI, The Illinois State Museum undertook to set up a punch card system of ceramic analysis, and a preliminary report on the success of this venture is given in Chapter IV. This report summarizes all of the above-mentioned projects and presents some data accruing from these. It should be considered current as of 1 July 1962. Considerable work has been carried out on these pro- jects since then and will be discussed in future reports. This report should be considered informative but not exhaustive. II UNIVERSITY OF ILLINOIS PROJECTS by Charles Bareis and Donald Lathrap Processing and Study of Archaeological Materials Work, under the auspices of the grant, did not begin until September 1, 1961, since one of the principal investigators (Bareis) was engaged in archaeological excavations in the Cahokia area. All archaeological excavations undertaken by the University of Illinois in the American Bottoms during 1960 and 1961 were in the vicinity of the Cahokia Mounds, The specimens obtained from these excavations have been washed and sorted and, with the exception of the Phase 2 (Highway Sal- vage Exc.) work at site Ms-2-2 (below), they have also been catalogued or otherwise bagged for study and analysis. The Excavations 1. Ms-2-2, Phase 2 Highway Salvage Excavations. Site area located in Madison County, east of Illinois 111, and ca . 200 yards NE of the old Powell Mound area. Excavated by Lathrap 1960. 2. Ms-2-2, Phase 3 Highway Salvage Excavations. Site area same as 1 (above). Excavations initiated by Lathrap 1960 and completed by Bareis 1960, 3. Ms-2-2, Groves Borrow Pit Excavations, Site area adjacent to and north of 1 and 2 (above). Excavated by Bareis and Lathrap 1960 (assisted by Porter). 4. S-34-1, University of Illinois Field School Excavations, Site area located in St. Clair County on western border of Collinsville Airport property, and ca, one-half mile SE of Monks Mound. Excavated by Bareis 1961. 5. S-34-2, Mound 51, University of Illinois Field School Ex- cavations. Site area located in St, Clair County ca. 100 yards SE of Monks Mound. Excavated by Bareis 1961 (as- sisted by Porter and Wittry). ^ H lU _J (T) Q < O O Totals From Excavations of Specimens Processed and Features Under Study (Totals not available for Ms-2-2, Phase 2, and totals for Ms-2-2, Phase 3, in progress.) Ms-2-2, Phase 3 Specimens Totals Chert Chips 8538 Chipped Chert 3028 Sherds 12826 Ground Stone 291 Rough Rock 1284 Limestone 2900 Animal Bone 591+ Plant 71+ Clay 908 Misc. 263 Features Totals Structures 29 Wall Trenches 3 Refuse Pits 21 Post Pits 6 Corn Caches 2 Fireplaces 10 Misc. 9 Specimens Totals Chert Chips 1820 Chipped Chert 1089 Sherds 2862 Ground Stone 50 Rough Rock 502 Limestone 1338 Animal Bone 1107+ Plant + Clay 209+ Shell 48+ Human Bone + Misc. 4 30700+ Ms-2-2, Groves Borrow Pit Features 9029+ 80 Totals Structures 6 Refuse Pits 155 Corn Caches 1 Fireplaces 2 Burials 2 Misc. 11 177 S-34-1 Specimens Totals Chert Chips 6156+ Chipped Chert 1033 Sherds 4406 Ground Stone 77 Rough Rock 502 Limestone 562 Features Totals Structures 8 Refuse Pits 6 Com Caches 9 Rock Features 2 Burials 2 27 Specimens Animal Bone Plant Clay Shell Human Bone Misc, s-3/4- 1 - - (Con t Inupf] ) Totals Features 183+ + 147 5 48+ 77+ Totals 13196+ Specimens Totals Chert Chips 210 Chipped Chert 72 Sherds 691 Ground Stone 3 Rough Rock 121 Limestone 409 Animal Bone 130 Clay 28 Shell 12 Human Bone 7 Bone Artifacts 8 Misc. 11 1702 S-34-2, Mound 51 Features Refuse Pits Wall Trench Fireplaces Misc. Totals 5 1 2 J_ 15 A number of the limestone and rough rock specimens recovered during the excavations have been transferred to the Southern Illinois University Museum for incorporation in the lithic project. Likewise, some of the 156 soil samples collected during the excavations have also been transferred for incorporation into the pollen studies. Six of the 152 charcoal samples obtained from the sites were processed for radiocarbon dating by the Uni- versity of Michigan (below). Shell Analysis : Several conch shells obtained from cuts by heavy machinery on the west side of the Alton and Southern Railroad during the 1960 Phase 3 excavations at site Ms-2-2 were submitted for identification to Dr. Fritz Haas of the Zoology Department of the Chicago Natural History Museum. The shells have been identified as Busycon perversum and are from the eastern coast of the United States between Cape Hatteras and Florida. Ceramic Analysis : The ceramic analysis in progress at the Univer- sity of Illinois during 1961-62, under the supervision of Lathrap, has the following aims: 1. The several ceramic complexes at Cahokia will be defined as total cultural patterns on the basis of the modes present (Qualitative Attributes in Spaulding's terminology) and the rules for mode combination. Such a pattern will be determined for each of the synchronic units present. In order to facilitate the study of the rulr-^ of mode combination, I.B.M. cards will be used. 2. Once these several cultural paL terns have been defined, they can be ordered temporally on tlir basis of the considerable stratigraphic data and the several C 14 dates available. Once this ordering is made, a comparison of these several cultural patterns should yield a very clear picture of ceramic change at Cahokia, not in terms of quantitative shifts in gross, rather ill-defined types, but in terms of the shift in the nature of individual modes, the api^earance or disappearance of individual modes, or the appearance or disappearance of particular mode combinations. 3. The end result of these procedures should be a master chron- ological chart for the ceramics in which the co-occurrence and temporal range of all recognizable modes and temporal variants of modes are clearly stated. Such a chart would not only serve as a powerful dating tool but would permit meaningful discus- sions of culture change at Cahokia in terms of operationally defined units. The form of this master chart would be very similar to the charts of qualitative seriation discussed by Rowe (Archaeological Dating and Cultural Process. South - western Journal of Anthropology , Vol. 15, No. 4, pp. 317-24, 1959). Radiocarbon Dating :^'" Six radiocarbon samples, obtained from exca- vations in two areas of the Cahokia site during 1960-1961, were processed by the Phoenix Project Radiocarbon Laboratory at the University of Mich- igan. Submitted by Charles J. Bareis, Department of Anthropology, Univer- sity of Illinois. Me-2-2, Subseries B. P. A, D. M-1292 1055t150 755-1055 M-1293 1190tl50 620-920 M-1294 1125+150 685-985 M-1295 1915+150 105B.C . -195A.D, S-34-i Subseries M-1296 725tl50 1085-1385 M-12y7 675+150 1135-1435 *A11 radiocarbon determinations are discussed in detail in Chapter VII of this report. New Excavations During the summer of 1961 archaeological excavations were conducted by Barcls on the western margin of the Collinsville Airport property (S- 34-1) southeast of Monks Mound. The area of excavation was located on the eastern slope of a ridge with an elevation of A15' above sea level. The surface of this portion of the airport property and the large Listerman tract to the west, extending from U.S. 40 south, is replete with evidence of the Mlssissippian occupation at Cahokia. The primary purpose of the work during 1961 was to test an area well beyond the rights-of-way of Federal Aid Interstate Highways in order to obtain additional information pertaining to the settlement plan in the central Cahokia area. The field party, consisting of 8 University of Illinois students, exposed a tract of approximately 3200 square feet which contained a large quantity of archaeological specimens and 27 features. The eight structures excavated at the site were all wall trench houses, probably occupied by nuclear families. It is particularly signifi- cant that seven of them were either superimposed or placed adjacent to one another in an area of 35' x 35'. Five of the houses were oriented with the long axis east-west, and two were constructed with the long axis north- south. A radiocarbon date of 725+150 years ago has been obtained for one of the latter. The compact arrangement of the houses is indicative of an inten- sive reoccupation of areas suitable for agricultural activity. This pattern is quite similar to the type of settlement plan exposed by the University of Illinois and Illinois State Museum Highway Salvage Excavations in other sections of the Cahokia site during 1960 and 1961 (ISM). Another project undertaken by the University of Illinois during the summer of 1961 was the testing of Mound 51 (S-34-2) southeast of Monks Mound. Since Mound 51 has been gradually removed during the past two years for subdivision purposes, it was deemed necessary to undertake excavations in order to obtain archaeological information before complete destruction. The excavations were conducted by Bareis, who was assisted by Porter and Wittry on weekends when their labor crews were not working. A 60-foot test trench 10 feet wide was established across the top of the mound. This trench was excavated to a depth of approximately five feet. The testing indicated that the mound had been constructed in at least two sand-filled stages. The most significant features excavated in the mound were two large fireplaces located on the western slope which was under constant destruction by a heavy equipment operator. Clay Minerals Project Work on the clay minerals project was initiated in September, 1961, under the supervision of Bareis with the analyses undertaken by Mr. Walter Parham of the Illinois State Geological Survey. Mr. Parham worked on the project until February, 1962, and during this time analyzed 28 pottery and raw clay specimens. During this time the Illinois Geological Survey kindly provided their facilities and Mr. Parham's working time for the project. Subsequent to February, 1962, the project was continucnl by Mr. Brian Carss of the Department of Geology. By June of 1962, Mr „ Carss had analyzed an additional 47 samples. The 75 samples currently processed were submitted by the three participating institutions in the Grant. Breakdown of the number of samples submitted by each institution is as follows: UI, 37; SIU, 18; ISM, 20. Following is a progress report by Mr. Carss on the clay minerals project, and a comment on points 1 and 2 in the Conclusion and Recommenda- tions section of his report by Mr. James Porter of the Southern Illinois University Museum, 10 III PROGRESS REPORT ON THE ClAY MINERAL ANALYSES OF POTTERY AND SOILS RECOVERED FROM THE AMERICAN BOTTOMS by Brian Carss Department of Geology, University of Illinois Introduction To date approximately 75 different soils and pottery fragments have been analyzed for their clay mineralogical content by means of X- ray Diffraction analysis using a General Electric X/R D-4 Dif fractometer and a Phillips recorder. Experimental Technique Each specimen was first crushed and ground to pass a 200 mesh sieve. A small part of this specimen was placed in a holder and an X- ray powder diffraction obtained. A second part, approximately 3 grams, was placed in a 50 mis. beaker which was then half filled with distilled water, stirred for two minutes, and then allowed to stand for one hour. A small tube was inserted to a depth of 1 cm. and about 5 mis. withdrawn. This suspension was then flowed onto a glass slide and allowed to dry overnight. This is the standard procedure for the preparation of an oriented aggregate. Three slides of each specimen are prepared, so that they could be subjected to various treatments before X-raying. One slide was X-rayed without any further treatment, to determine the major clay mineral constituents. The second slide was placed in an atmosphere of ethylene glycol for approximately 24 hours. This has the effect of expanding the montmorillonoid minerals to a basal d-spacing of 17^. This test is a diagnostic one. The third slide was treated with concentrated hydrochloric acid or, in some of the earlier specimens, heated to approximately 550°C for a period of 30-45 minutes. This is a diagnostic test for distinguishing between Kaolinite and Chlorite. One sample (CS-38) was taken and placed in a specially designed furnace so that it could be continuously X-rayed as the temperature was raised. The aim of this experiment was to try and determine the sequence of changes that take place in the mineralogy of a soil as the temperature is increased. The sequence of changes that do take place are as follows: 11 a) Up to approximately AOO C, organic constituents are burned. b) 400 -550 C, destruction of the kaollnite structure. c) Up to 950°C, gradual destruction of the illite structure. This gradual destruction can be followed on the dif fractometer trace as a steady decrease in the first order reflection intensity and a loss of the sharpness and symmetry of the reflection, d) Above this temperature there is the beginning of the develop- ment of the high temperature phases. This last stage requires rather sophisticated firing techniques, and from examination of the diffractometer traces, no high temperature phases have been developed. Mineralogy of the Soil Samples By far the most dominant mineral present in the soil is quartz, ranging in size from a coarse sand to what is called amorphous silica. This latter shows on the diffractometer traces as a high background in- tensity in the approximate position of the first order reflection for quartz. Associated with the quartz and usually in very minor quantities is cristobalite. In the majority of cases it is just detectable, so that there is probably less than 1 per cent, as this is about the limit of detection of the X-ray machine. The second major non-clay mineral group present is that of the feldspars, with the plagioclase feldspars dominating over the potassium feldspars, and in general these minerals tend to be well defined crystals, as they rarely appear present in the less than 2 micron fraction. The third non-clay mineral group is that of the carbonates. Their presence appears to be quite random and in the soils is probably directly connected to the amount of shell material that has been added. In the sherds, calcite and dolomite commonly occur in pieces that can be readily picked out with a needle. However, further investigation shows that the calcium carbonate has been dehydrated to calcium oxide and then slaked again to calcium hydroxide and then back to calcium carbonate. These series of reactions verify a temperature of firing greater than 650°C, (CS-73 was the specimen used in this investigation.) The clay mineral composition of the soils remains reasonably con- stant in variety in all the samples. The soils contain the following clay minerals in order of abundance: Montmorillonite, Illite, Kaolinite, and some Chlorite, the first being by far the most abundant. Mineralogy of the Sherds The mineralogy of the sherds is very much simpler than that of the soils, as the temperatures reached in firing have been great enough to completely collapse most of the montmorillonites. Illite is present, but its crystallinity decreases steadily with increasing temperature and 12 can be used as a rough guide as to the firing temperatures. The feldspars and tlie carbonates are not greatly altered except for the carbonates which are burned hut revert to carbonate again on weathering. C onclusion and Recommendations At this early stage in the investigation, two main generalizations can be made that seem to hold true in all cases. 1. The manufacturers of the pottery did not seek any special clays for their pottery, such as the kind that occurs directly under the coal seams in the area, but just used the soil in the immediate vicinity. 2. The range in firing temperatures seem to be nearly constant and never above 900°C , There are a number of cases in which the firing temperature of the particular specimen never exceeded 550°C as shown by the presence of kaolinite. This may be a result of the uneven firing of the vessel. It would be expected that the pottery placed in a fire, the base of the pot would reach a higher temperature than the neck or rim. It is recommended that material from the Cahokia site be continued to be processed so that a complete picture of the character of pottery mineralogy be completed, and it is hoped that this can be directly linked to the mineralogy of the parent material. It would be very helpful to obtain both soil samples and sherds from entirely different areas to see if there is any distinct difference either in mineralogy or firing temperatures. The manufacture of pottery under conditions similar to those being tested with thermocouples included in the pot before firing so that the firing temperatures can accurately be determined. In this manner it may be possible to obtain an idea of the temperature gradient within the pot itself. Investigations of the slips have proved to be of little use in determining if it was painted on or is the color of the burnt clay material, Comments on "Conclusions and Recommendations ^*^ in Report by B. Carss, James Porter , Southern Illinois University Museum Item 1. To the geologist using only X-ray methods, the alluvial clays are not special clays. It is advantageous to know the aboriginal potters did not use Pennsylvanian under- clays, but now other methods, thin-sections for example, must be brought into the analyses to check on the charac- teristics of the potter's clay. 13 Item 2. An important fact here is that the low-fired specimens (under 550°C) are "stumpware."^ We do not know the function of this pottery but we do know it is always grog-tempered and usually with two or three pottery types being crushed and added as temper for a given piece of "stump." 14 Grid Axis (Not FAI 255 Center Line) i 800 + A05 ASU Mop of Excavated Area Archaeological Salvage Project FAI 255, Section 60-6-1. ISA Project 1-70-1(40)4 ^Sto 213+65 The Cohokio Site 3000 Fflet West of Monks Md June 6 - Sept. 'S, 1961 FIGURE 3 15 IV ILLINOIS STATE MUSEUM PROJECTS October 1961 to June 1962 by Warren L. Wittry and Joseph 0. Vogel Introduction This paper outlines the results of the first year of analysis of the data and materials from the Cahokia site. The basic field work was conducted during the summers of 1960 and 1961 under the direction of Wittry, then Curator of Anthropology, Illinois State Museum. This research was undertaken in co-operation with the Illinois Division of Highways and was financed by the United States Bureau of Public Roads. It is part of the Coordinated Archaeolog- ical Investigation of the American Bottoms, a project of the Illinois Archaeological Survey, Subsequent laboratory analysis was financed by the National Science Foundation under Grant NSF-G17922, In May, 1962, Wittry resigned his position at the Illinois State Museum to become Assistant Director and Curator of Anthropology at Cran- brook Institute of Science, Bloomfield Hills, Michigan. In order to facilitate his continued participation in the American Bottoms Research Project, all of the field notes and maps of excavations were microfilmed and xerox prints were made from the films. He will receive a small portion of Grant NSF-G23998 for the purpose of engaging an illustrator to assist in his studies of architecture, settlement pattern and ceremonial- ism. Basic Data In brief summary, the data obtained from the field investigations consists of the following plus the pertinent notes, maps, photographs, and other records: I. Summer of 1960 . Tract 15B, 800 feet west of Monks Mound, and north of Highway 65. 16 A. The remains of 131 prohtstoric houses. B. 13 burials. C. 465 storage/refuse pits. D. TliG outline of 4 large circular structures. E. A large rectangular compound of small circular structures connected by walls. F. Over 100 cubic feet of artifacts, faunal refuse, botanical specimens, skeletal remains, soil and carbon samples. II. Summer of 1961 . Tract 15A, 3000 feet west of Monks Mound and north of Highway 66. A. The remains of 176 prehistoric houses. B. One burial. C. 384 storage/refuse pits and post pits. D. The outline of 4 large circular structures. E. A palisade which may relate to Mound 44. F. Over 150 cubic feet of artifacts, faunal refuse, botanical specimens, skeletal remains, soil and carbon samples. In this report no mention is made of the non-ceramic artifacts obtained in the field investigations. These have been relatively scanty in number and to date little beyond cataloging has been done with them. Ultimately, these non-ceramic materials will be included in the analysis. Analysis Most of the first year has been taken up with the preliminary processing of the material obtained, e.g. — washing, sorting, cataloging and establishing the ceramics analysis program, for which a 5 x 8 inch single-row edge-punched card has been programmed by Wittry. Tracts 15A and 15B were two of perhaps the three or four remaining good areas in the central Cahokia site. Preliminary testing on these tracts indicated they contained much desired information about ceramics, house types, settlement pattern and the temporal changes which had taken place in these. The excavations revealed that Cahokia, already heavily occupied in Late Woodland times, had made a shift to a Mississippian kind of culture with an astonishing development of ceremonial structures. 17 The present paper Is not exhaustive: the problems of Mississippian ceramics are barely touched; the sequence of archltccturn 1 types is drawn in the boldest of outlines; the ceremonial structures nrc merely described; the section on Woodland ceramics contains preliminary doscriptions where additional analyses are underway. The Late Woodland "Bluff Culture" Occupation A result of considerable importance of the recent work is the revelation that the American Bottoms were heavily populated during Late Woodland times. We have become accustomed to referring to this Late Woodland mani- festation on the American Bottoms by the term "Bluff Culture." Tlirough certain crossties in ceramics, we feel certain that many of the mounds of Tittertngton's Jersey Bluff focus and others on the edge of the bottoms area contained burials of people who lived on the Bottoms. The same ves- sel types are present, but the ones with the burials are smaller than the culinary vessels of the Bottoms. The Late Woodland Occupation, which occurred during the approxi- mate period of A.D. 500 to 800, is characterized by a particular pottery type and a house form markedly different from the latter Mississippian occupation. "Bluff Culture" houses were usually constructed in a rectangular, steep-sided basin dug to a depth of 1.5 to 3,0 feet below the surface. The walls were formed of single poles, averaging 0.4 ft. in diameter, set into individual holes spaced at intervals of 0.5 to 1.0 ft. The earliest of these dwellings average 75 square feet (ca, 7 x 10 feet) and gradually increase in size to about 150 square feet (ca. 9 x 17 feet). Usually one-half of the house was taken up by a sleeping platform. There was no evidence of hearths associated with these structures. We interpret that each house was occupied by a nuclear family and that there were no strong differentiations of individuals beyond those of age and sex. There is an average of three to four storage pits, usually rec- tangular in outline, associated with these houses. In every instance of intersection between "Bluff Culture" and Mississippian houses, the latter have proved to be superimposed upon the former. Several houses appear to be transitional between the typical "Bluff Culture" dwelling and the typical Mississippian house, and there appear to be some pottery correspondences between the late "Bluff Cul- ture" material and the Mississippian; i.e., persistence of limestone- tempered, red-slipped vessels, "stump ware," the use of shell temper in some of the late Bluff pottery and the existence of otherwise typical Bluff culture jars with smooth or smoothed-over-cordmarked bodies. Whether these represent stages in a gradual change from one form to the oth( )■, nr t h(^ inixinp, of tf-chniquoR di.ir i.i pari In My coiitcmporniinoiiR 1 iv- ini: l^v the I w'l ciiltiii'i^'^, IK n f|urR(l.>ii v/c hopr wLI I br r(^s(i i v'l-d Inter in L hi' 'inn lysis. Latr Woodland Ceramics Wo arc presently conducting a very detailed analysis of Cahol0« POLLEN N ZONES •IkljM WkT •LACK auMto e •l*c« 8 • «MIO : VtLLO« %ii.1 s • !.»<.« • U«*0 •ITH ROrttD VIMTATIX ■Lt^X «uu»o FIGURE 12 41 Before passing on to the next profile, I wish to point out that artifacts of Mississippian culture--in this case Old Village pottery types--were discovered in direct association in both pollen zones. Since Pollen Zone I is so shallow, however, it is likely that the recovery of prehistoric pottery from this surface level is due to redcposition after the prehistoric period. The second profile (Figure 12) was collected from tlie same slough but at a point approximately 2,5 kilometers distant. The uppermost sample in this profile shows an unusually high frequency of arboreal pollen. Almost all of it is pollen of Salix and it undoubtedly indicates over- representation. The second level gave a spectrum similar to those of Pollen Zone I at the first locality. Below this and extending to a depth of about 36 inches the spectra are essentially similar to those of Pollen Zone II at the first locality. Below 38 inches, however, the percentage of Compositae pollen decreases and grass pollen is the only type which contributes more than 30 pe r cent to the profile. This change is con- sidered sufficient for the establishment of Pollen Zone III. The cultural material from this locality is modern brick, wood and metal. It will be noted that this refuse occurs at the top of Pollen Zone II. Old Village pottery was recovered from Pollen Zone II at the first locality. The third profile (Figure 13) was collected about 200 meters from the first locality. Pollen Zones I, II and III are in evidence and in addition a Pollen Zone IV may be recognized. In this last zone the per- centage of cheno-am pollen is greater than 30 per cent and the percentage of grass pollen is also 30 per cent. Old Village pottery was recovered in association with pollen spectra representing Zones II, III and IV. Thus if Old Village pottery represents a particular time phase in Missis- sippian culture history, the lowermost pollen spectra are not older than the oldest date which can be assigned to Old Village. Also, if the changes in the pollen profiles represent changes in environment, the Old Village phase of Mississippian culture is seen to have continued through at least three periods of different environmental conditions and two environmental changes. Figure 14 shows the results of sediment samples collected at the Mitchell Site approximately 11 kilometers north of Cahokla. Because there is only a thin deposit of artifact-bearing alluviums the samples were not collected in superposition. Thus we have no pollen profile, only a series of samples whose provenience is known relative to cultural remains. A lake is located at the southern margin of the Mitchell Site. The three samples at the top of the diagram were collected from the soil surface along the lake shore. All show the characteristic features of Pollen Zone I. It is particularly interesting that the proportion of arboreal pollen is low in these samples, since the margin of the lake is densely wooded with willow, oak, elm, Prunus and other trees. Two of the three samples were from moss polsters growing at the roots of trees; the third was taken beneath oak leaf litter. None, however, shows more than 25 per cent arboreal pollen. .<>- A2 ei SFDiMtNT 40 '* <* I'* '"■>• N ZONf. . I r,.i r 1 r . . 1 r ; i TOP SOIL BROWN- BLACK BROWN SANOr' 3JMBC YELLOW GUMBO FIGURE 13 iOIL LIRFACE NEAR LAKC EOIMENT lURFACE UNDER LAKC ^ /^«« II ■ III 114 III ■ irt ti* 1/4 ■ II i ■ t«0 a <<• III ■ tl» ■ 43 The next four samples wcro collectpd from the sediment surface n few cpntlmetcrs beneath the v/alrrs "f the Inke. Of the four, two show the characLertstlc features of Pollen Zone I nnd two show those of Pollen Zone If. Below this tliere are a group of samples from Misslssi pplan archi- tectural features, and the final group of samples was collected from the cultural sediment horizon but has no provenience relative to features. About half of the samples In these two groups show Zone II characteristics and about half show those of Zone III. Zone IV Is not represented In the few samples so far processed from this site. The recovery of consistent features of form and superposition In the three pollen profiles from Cahokla strongly suggests that environmental fluctuations have occurred. The recovery of similar pollen spectra from a site seven miles distant strongly suggests that these fluctuations were regional rather than local. Tii70 basic questions now present themselves: First, what environmental changes seem evidenced by these data, and Second, to what extent can the series of fluctuations be developed Into an absolute chronology. Today a large portion of the surface area of the American Bottoms is the water surface of lakes, ponds, swamps and sloughs--each with distinc- tive plant communities. Most timber Is to be found along the water's edge, though fairly dense stands of temperate trees may be found as local remnants- most of them near houses or on terraces at the outer margins of the bottom land. But the ecological system of most of the surface of the American Bottoms is primarily Influenced by man's Industry. The majority of the land Is either planted to maize and wheat or consists of fallow, weedy plots, roads, railway embankments and levees or house sites. In the sense of being unforested, the region Is a prairie; but because of the vast disturbance of the land by agricultural and earth- moving machinery, It Is a prairie composed of crop plants and weeds rather than wild grasses. The most prevalent weeds are members of the Chenopo- dlaceae, Amaranthaceae and Composltae. This situation appears to be ad- mirably reflected by the samples of Pollen Zone I (see Figure 13). Because of the recovery of modern refuse In Pollen Zone II, we may be confident that Pollen Zone II ended after white settlement began In the area. Thus the boundary between Pollen Zones I and II must have been laid down less than 150 years ago. Surveyors' records of the 1830's and 40' s Indicate that at that time the region was mostly unforested, but the amount of surface area covered by swamps, sloughs and lakes was much greater. The surveyors' reports are quite valuable In reconstructing the ecology, for they describe conditions for 5 points within each square mile. In order to improve the agricultural potential of the land, a great amount of filling and draining was undertaken during the last part of the 44 19th century and the 20th century. 1 brlicvc that the cheno-am rise indi- cates sediment disturbance and thus that it is the combination of filling, draining and the general use of tractors after 1900 A.D. which accounts for the high frequency of cheno-am pollen in Zone 1, Climatic fluctuation cannot be eliminated as a possible cau'^e for the Z.me Il-Zone I shift, but in view of the present ecological situation it seems less probable, I am tentatively dating the Zone I-Zone II boundary at 1900 ± 25 years A,D., but even if I am incorrect in this absolute date, the distinc- tive characteristics of Pollen Zone II must be a reflection of the ecolog- ical system in effect at the time of white settlement and before. The historic record describes this ecological system as "swampy prairie." Swampy prairie was a common form among the various prairie types found in Illinois which survived as relics and remnants into the 20th century when they were mapped and described by ecologists. Through the pollen record of Zone II we can observe the presence, not to say the dominance, of pollen of the anemogamous Compositae on the American Bottoms in historic time. Yet in the vast majority of ecological descriptions of wet prairie types, plants of anemogamous genera are usually listed as rare if they are noted to occur at all: Sampson's ecological survey of the prairie communities of Illinois (1929) indicated that Ambrosia (the only anemogamous composite which he lists) is only to be found in xeric locales, though he concantrated on the piairies of the northeastern part of the state. I have been able to locate only one record of swampy prairie in which anemogamous composites played any important role. In 1930 Turner (1934) collected frequency curves on swampy prairie stations from the Illi- nois River valley and the Mississippi River valley. The basic difference in the floristics of the two areas was that on the floodplain of the Mis- sissippi there was only one dominant plant: Spartina Mlchuaxiana Hitchc, or slough grass. But on the floodplain of the Illinois there were two dominants: slough grass and Iva clllata Willd.--one of the anemogamous Compositae. Since we know from historic record that the region was swampy rather than xeric, it appears that the pollen record supports a reconstruc- tion of an aberrant, but not unknown, prairie form in the American Bottoms at the beginning of white settlement. In support of this reconstruction I correlated the frequency of occurrence per pollen zone of the non-arboreal swamp indicators Sparganium , Cyperaceae and Polygonum . Since the majority of the samples came from a known slough, it is not surprising that pollen of swamp indicators should be found in all zones, but it seems significant that it is more frequent from Pollen Zone II than any of the other zones. What are the environmental conditions reflected by the pollen curves of Pollen Zone III? On the basis of the reduction of the Compositae curve and the reduced frequency of non-arboreal swamp indicators, I consider that Zone III represents a shift towards aridity relative to Zone II, I am extremely hesitant to make any definitive statements concerning temper- ature changes between Zones II and III, and in fact wish to refine and qualify my statement about a shift from wetter to drier conditions. The use of non-arboreal pollen curves in the reconstruction of climatic conditions is, it seems to me, far more dangerous than the use 45 of arboreal curves. Because most of the non-arboreal vegetation is com- posed of species of annuals, the stand for any given year represents mainly the meteorological conditions of that year. The stand of arboreal vegetation, especially mature arboreal vegetation which releases pollen, represents climatic conditions for a reasonable length of time--at least a couple of decades and not uncommonly for some centuries. Thus when a change can be recognized in a pollen profile constructed on the basis of non-arboreal types, it must be recognized that the changes involved might well be those which are of little climatic significance. Also, it must be recognized that such changes could not represent more than conditions existing during the growing season. Throughout this paper I have refrained from mentioning climatic change in reference to the pollen chronology developed for the American Bottoms. I do this because I consider it quite injudicious to attempt to reconstruct climatic conditions when no data is available for the four or five months of the year when the non-arboreal flora is dead or dormant. On the other hand, the inferences that may be derived from the study of the non-arboreal pollen curves should be far more sensitive for the growing season than inferences that may be derived from the arboreal pollen curves--because an environmental fluctuation of almost any kind, such as erosion, drought, flooding, cultivation or shortening of the grow- ing season will probably be represented in the stratigraphic record if it persists for a decade. This is exactly the type of environmental record which the archae- ologist needs most desperately. For agricultural peoples such as those who dwelt in the Mississippian villages of the American Bottoms the basic question of economic survival was not whether the local ecology could support agricultural activity over the long time span of a climatic phase, but whether it could support agricultural activity of a certain level of yield year after year. I believe the limitations of using non-arboreal pollen curves require me to discuss not a general change from drier to wetter conditions but a change from effectively drier to effectively wetter conditions during the span of the growing season. In discussing moisture reception in this region f^e cannot consider just rainfall, for the yearly spring flooding of the American Bottoms allows the effective moisture reception to be appreciably higher at that time than might be recorded in rainfall figures. It is possible to account for the change in the pollen record on the basis of a change in total annual moisture reception, but I think it can also be accounted for on the basis of a fluctuation in the periodicity of maximal moisture reception, and I consider that this latter alternative is more acceptable in regard to the type of data I have collected. I believe that the difference between Zones II and III could be explained if Zone III was characterized as a phase in which the early part of the growing season was deficient in moisture reception, relative to Zone II, and the later part of the growing season was relatively superior in moisture reception. Turner's explanation of the growth of anemogamous Compositae in the Illinois River valley attributed this occurrence to the fact that there was poorer drainage there and the soil was saturated for a longer period during the growing season. Considering that grasses are. hi a shift In the pollen record occurred also, but that the date of the shift was 1000 A.D. + 100 years. Under no circumstancos could the pollen repre- senting the boundary zone have been laid down an late as 1200 A.D., and even 1100 A.D„ would be an extreme date from my f^/idcnce. I apologize that these data have not yet W-on published, but they are now in press and 1 have been assured that they \-!i\\ be out before the end of the year. To my knowledge this date of 1000 A,D. for the pollen fluctuation is the only one from the continent which cnn be said to have adequate documentation and precision for the time period of interest here. Other dates are based on radiocarbon determinations or on a guess corre- lation with the documented shifts in Greenland and Europe, It seems a little anticlimatic to return to the pollen diagram at this point, but there is still Pollen Zone IV to be reckoned with. At present I am hesitant to make particular claims as to the validity of Pollen Zone IV as a representation of environmental conditions. Pollen Zone IV is found at only one locality and it is possible that it repre- sents some local edaphic phenomenon and not a regional environmental one. It will be noted that its presence in the pollen profile is correlative with a layer of sandier sediment. I consider it quite possible tliat the increased percentage of cheno-am pollen reflects nothing more than the immense amount of sediment disturbance which may have resulted from the construction of mounds at Cahokia--perhaps Monks Mound which is only a couple of hundred tneters distant. The dating of the Zone Ill-Zone IV boundary is an unresolved matter, as is the dating of the base of this pollen profile. I can only say at present that the chronology is no older than the oldest date that can be assigned to Old Village pottery. According to archaeological esti- mates now prevalent this would be about 700 A„D. In summary, the oldest date that can be as yet assigned to the pollen chronology from the American Bottoms is in the neighborhood of 700 A.D., thus- the chronology goes back, roughly, 1500 years at most. All pollen zones show a primarily non-forested, or prairie, ecology though this is recognized as having undergone changes through time which may have had significant effects on the cultural phases of the area. The lowermost pollen zone may either be a local or a regional phenomenon. It appears to indicate a phase of sediment disturbance, but this could be due to cul- tural as well as ecological-edaphic factors. The upper three zones of the pollen chronology seem to reflect regional ecological occurrences, since they can be reproduced at localities up to 11 kilometers distant. Pollen Zone III, which may end at 1000 A.D., seems to reflect a relatively more arid environment. The shift from more arid to more moist conditions at the end of Pollen Zone III may be only a shift in the timing of moisture reception during the growing season and not a basic change in the climatic configuration of the American Bottoms region. Pollen Zone II dates from approximately 1000 A.D, to 1900 A.D. in my opinion. It is relatively more moist than the lower zones and seems to correctly answer descriptions of the area recorded during the early historic period. The uppermost pollen 48 7,one shows indications of sediment disturbance. From st ra tigraphic and ecological knowledge of this zone it evidently reflects the recent dis- turbance of the ecological balance of the region by mechanized agricul- turalists. 49 VII RADIOCARBON ASSAYS by Melvin L. Fowler Listed in the following section are the radiocarbon assays of age received to date on samples submitted under this project. All samples were collected in the field specifically for this purpose. All were sub- mitted to the University of Michigan Phoenix Memorial Project Laboratory for assay. Summary data on each of these lists are presented in the reports of the sub- projects above. In those reports the age estimates were given in terms of one standard deviation and the time range (one standard deviation) involved in years A.D. This range was obtained by subtracting the years B.P. from 1960. In the lists below both the before-present mean and the range in the years A.D. will be listed. The detailed de- scriptions of each sample are also given in this section. University of Illinois Samples Ms-2-2, subseries ^.^^ Location (38° 39' 46" N. Lat., 90° 05' 17" W. Long.) ca. 200 yards NE of the old Powell Mound area and 2 miles west of Monks Mound, Madison County, Illinois. Range B.P . Years A.D . M-I292 Powell-Zurkuhlen Site, Feature 234, House 26 1055+150 755-1055 Charcoal from charred layer on the floor of a rectangular wall trench structure (Fea. 234). Long axis of structure oriented NW-SE, Associated with Mississippian occupation at the site. Sample No. IAS-11; coll. 1960 by Charles J. Bareis. M-1293 Powell-Zurkuhlen Site, Feature 227 1190^150 620-920 50 Charcoal from fire basin (Fea, 227) £a. 9' in diameter located near the center of a large rectangular wall trench strncture (Fea. 202, House 15) with the long axis oriented NE-SW. Both fea- tures associated with the Mississippian occupation at the site. Stratigraphic evidence indicates, however, that M-1293 sliould be later than M-1292 and M-1294, Sample No. IAS-12; coll, I960 by Donald W. Lathrap. Range B.P . Years A.D . M-1294 Powell-Zurkuhlen Site, Feature 217, House 21 1125+150 685-985 Charcoal from the floor of a rectangular post structure (Fea. 217). Long axis of the structure oriented NW-SE, Al- though the method of construction is of the Bluff or Late Wood- land type, Mississippian pottery was present on the floor of this structure. Sample No. IAS-13; coll, 1960 by Charles J. Bareis . M-I295 Powell-Zurkuhlen Site, Feature 197 1915+150 105B,C ,-195A.D. Charcoal from a charred area (Fea. 197) in Square J-1-36, level 6 (depth 31 inches). A large post structure and a wall trench structure (both associated with the Mississippian occupa- tion at the site) were located nearby approximately .6 '-.9' above Feature 197. M-1295 is too early for the Mississippian occupa- tion at Cahokia but may indicate the time of the beginnings of the Late Woodland occupation. This sample may have been contami- nated. Sample No. IAS-14; coll, 1960 by Donald W. Lathrap. S-34-1, subseries Location (38° 39' 21" N. Lat., 90° 03' 15" W. Long.) at west border of Collinsville Airport property and ca. \ mile SE of Monks Mound, St. Clair County, Illinois, M-1296 Collinsville Airport Site, House 3 7251150 1085-1385 Charred wood from top of west wall of rectangular wall trench structure (House 3). Long axis of structure oriented N-S. Asso- ciated with Mississippian occupation at the site. Sample No, LAS-15; coll. 1961 by Charles J. Bareis. M-1297 Collinsville Airport Site, Feature 1 675+150 1135-1435 Charcoal and charred corn from a large refuse pit (Feature 1), non-stratigraphic cut. Associated with Mississippian occupation at the site. M-1296 and M-1297 date the latest occupation at IAS-16; coll. 1961 by Charles J. Bareis. 51 Sou thern Illinois University MusGum Samples Mitchell Sitr Scries, Illinois Vegetable matter from various locations at the Middle Missis- sippian Mitchell Site (20B2-3), Madison County, Illinois. Sub- mitted by James Porter, Southern Illinois University, Carbondale, Illinois. A report on the site is in progress. Range B.P. Years A.D. M-1298 Mitchrll Site, Mound B 785+150 1025-1325 Charred wood (wall post or roof support) from a burned struc- ture, Feature 44, on top of Mound B. Coord. 38° 45' 28" N. Lat., 90° 05' 27" W. Long. Sample thus represents latest occupation of Mound B. Stratigraphic evidence indicated that M-1298 should be later than M-1299. Sample No, IAS-21; collected October, 1960, by Eugene Fugle. M-1299 Mitchell Site, Pre-Mound B 875+150 935-1235 Charred wood (roof support) from house floor at NE corner of early ceremonial structure found under west flank of Mound B. Coord, as M-1298. Sample associated with Feature 32. Sample No. lAS-22; collected October, 1960, by James Porter, M-1300 Mitchell Site, Mound C 785+150 1050-1325 600 700 800 900 1000 1100 1200 700 FIGURE 15 Ranges (1 sigma) of radiocarbon assays from the Cahokia area, 57 The radiocarbon assays from the Cahokla area itself (Tracts 15A and 15B) present quite a different picture. There Is no tight clustering of the data, but rather there is a great spread in range from 385 + 90 years ago to 1025 + 110 years ago and what appears to be an even dis- tribution over this range. For example, the range of the Mitchell dates, disregarding M-1303, is from 785 + 150 years ago to 950 + 150 years ago. The difference between these two extreme dates from Mitchell is 165 + 225 years indicating the strong probability that all of these assays are contemporaneous. Even if one takes only those dates from Cahokla that were culturally labeled "Late Mississippian," M-1337, M-1339, M-1332, and M-133A, one finds that the difference between the two extremes of this is 420 + 131 years and that this indicates these dates statistically are significantly different. What the Cahokla dates suggest is a long range of occupation of the site areas through time. The earliest portion of this Cahokla range (see Figure 15) seems to overlap with the time range suggested for the Powell-Zurkuhlin area. Thus, if sample M-1340 is compared with the average for the Powell area, we find that the difference between the two is 100 + 140 years, thus sug- gesting that these two dates are roughly contemporary. The middle portions of the Cahokia range correspond well with the time suggested for the occupation of the Mitchell Site. The upper end of the range coincides with the dating of the occupation of the Collinsville airport site and the one late date from a pit on top of Mound H of the Mitchell group. The fact that a correspondingly similar date was found at Cahokia suggests that the sample from Mitchell (M-1303) may actually represent a cultural occupation rather than contamination from other sources. Taken altogether, these radiocarbon determinations suggest that the Mississippian occupation of the American Bottoms region began about 800 A.D. The radiocarbon data are not clear enough at the present time to determine if the "Bluff" occupation was contemporary with the early Mississippian or preceded it in part. The present radiocarbon data suggest contemporaneity, but these data remain for detailed analysis in terms of the archaeological associations involved before more can be said on this question. This is one problem where further radiocarbon analysis would be helpful. A series of samples collected from "pure"^ Bluff houses would be important ones to assay. The data indicate a very late Mississippian occupation of the area, lasting into the 16th century, and more samples could profitably be submitted dealing in detail with this little known end of the time range. 58 REFERENCES CITED Griffin, James B. 19A9 The Cahokia Ceramic Complexes. Procee dings of the Fifth P lains Conference for Archaeology (T- I, Champe, Editor), The University of Nebraska, Griffith, Roberta J, 1962 Ramey Incised Pottery. Unpublished Ilaster's Thesis, Southern Illinois University, Department of Art, Kelly, A. R, 1952 Early Georgia . Volume I, No. 2. Moorehead, Warren King 1928 The Cahokia Mounds. University of Illinois Bulletin , Vol. 26, No. A, Rowe, John H. 1959 Archaeological Dating and Cultural Process. Southwestern Journal of Anthropology , Vol. 15, No. 4, pp. 317-324, Sampson, H„ S. 1929 An Ecological Survey of the Prairie Vegetation of Illinois, Illinois Natural History Survey , 13:523-577. Turner, L. M. 1934 Grassland in the Floodplain of Illinois Rivers, American Naturalist, 15:770-780. UNIVERSITY OF ILLIN0I8UBBANA 3 0112 050754123