The “South Appalachian Province,” an area defined by the occurrence of the complicated stamp as a major mode of pottery decoration, was outlined by Holmes (1903). Almost nothing can be added to the area definition propounded by Holmes. Actually the area varied in size as it expanded and contracted through time (the temporal factor was necessarily omitted in 1898). Although the province varied in size, it should perhaps be reduced in size to include only the core area–the present state of Georgia with the addition of northern Florida. Areas outside this used complicated stamping only at certain time periods, and in all cases these usages appear to be stylistic imports from Georgia, not local developments.

Radiocarbon dates for the Arctic thus far determined by Arnold and Libby represent four time levels. The latest dates are for the Eskimo period. These are derived from wood samples from five sites in northern and northwestern Alaska ranging in age from 912 ± 170 (Ipiutak) to 2258 ± 230 years (Okvik). Johnson's site in Yukon Territory, Canada, dated at 1519±150 years, would also fall within this period if the assigned date is correct; however, the site is probably considerably older than this. Next in age is an early site in the Aleutian Islands with a date, from charcoal samples, of 3018±230 years. The age of the Trail Creek cave on Seward Peninsula is given as 5993±280 years. This is a pre-Eskimo site related to the Denbigh Flint complex; the latter has not as yet been dated. Two dates have been obtained from the Pleistocene muck deposits near Fairbanks, one sample at 12,622±750, and another at over 20,000 years (Arnold and Libby, 1951).

The problem of the origin and distribution of semi-subterranean structures is an intriguing one. It is generally accepted that the semi-subterranean structure, earth lodge, or pit house, was one of a number of relatively unmodified traits of Old World origin found in North America. This paper reviews some of the literature on the pit house, and summarizes its development and distribution in the Eurasian and North American continents.

Laymen, believing that early man habitually lived in caves, commonly use the stereotyped term “cave man” in referring to our early progenitors in the Old World. This assumption was an inevitable development since occupational sites in western Europe are predominantly caves or cave shelters. However, some European archaeologists have suggested such sites were seasonally occupied–that they were the winter habitations of man during the upper Paleolithic. There is reason to believe different types of shelters were used during warm seasons since engravings of tents or more substantial structures (tectiforms) have been found on the walls of caves and cave shelters. A number of European archaeologists including Breuil (1910), Clark (1939), Childe (1950), and others have accepted tectiforms as evidence of the existence of artificial structures in western Europe during the last glacial advance. Breuil has gone further and compared some of the tectiforms in the Font-de-Gaume, the Dordogne Valley of France, to the summer huts of Navahos described by Mendeleff in 1896.

Within recent years there appears to have been an increasing awareness on the part of archaeologists that certain statistical techniques offer economical methods of extracting information of cultural significance from archaeological data. The discussions of Kroeber (1940), Robinson (1951), and Brainerd (1951) have appeared in American Antiquity, and the last two even evoked a comment (Lehmer, 1951). In addition to these papers, which are primarily devoted to exposition of method, a considerable number of special applications can be found in the literature. Archaeological research inevitably brings the researcher face to face with the problems of ordering and comparing quantities of data and of sampling error. There seems little doubt that the best approach to these problems involves a search of statistical literature for appropriate methods.

The fallis village site is an area of some 4 acres, situated on Lot 2, Concession 4, in the Township of Tiny, Simcoe County, Ontario. Springs at the site are the source of a small stream, which joins the River Wye at the head of Mud Lake. At the foot of this small lake are the ruins of Fort St. Marie (Kidd, 1949) built in the opening decades of the seventeenth century, as the center of activities of the unfortunate mission of the Hurons. Thus the Fallis site is situated in the heart of the historic Huron country.

In A. F. Hunter's work (1899, p. 38), the site is listed as No. 38, “John Fraser's and William Smith's,” with the information that the author casually inspected the site and found fragments of pottery scattered about. Hunter does not report European goods at this site, nor does he say whether it had been cultivated.

The Gordon site has been mentioned previously by Ford (1935, 1936) as site 61, the type site for the Coles Creek ceramic complex originally identified by a surface collection of 175 sherds. In his initial survey of the site, Ford placed it uppermost in the Coles Creek horizon, although the exact relationship of that horizon to other Southeastern cultures was at that time uncertain. Ford states:

Although considerable archaeological excavation has been conducted throughout eastern Oklahoma during the past fifteen years, it is only recently that attempts have been made to present a chronological framework for the area (Orr, 1946; Krieger, 1946, 1947; Newell and Krieger, 1949; Bell and Baerreis, 1951). Within this framework of prehistory, the latter portions such as those represented by the Gibson and Fulton aspects are probably the best understood. "With a single exception (Baerreis, 1951), the earlier cultures remain relatively incognito. One of these early cultures is commonly referred to as Fourche Maline. Unfortunately, however, the Fourche Maline materials remain to be clearly described and identified.

In view of the present turmoil in North American archaeology caused by the continuing release of radiocarbon dates, it is most difficult to maintain an understanding of the preceramic occupations. What seems like a good guess today is tomorrow relegated to the realm of unwarranted speculation. The continual excavation of preceramic sites in North America and the constant revision in geological and climatological theories also force us to be cautious in making interpretations and to be willing to change any of them. Yet the evidence now available for reconstructing the preceramic traditions in some instances appears to be sufficient and reliable enough to justify certain interpretations, even though reservations must be kept in mind. In particular the relative chronology of several cultural complexes is slowly taking form.

Since its discovery by Jenness in 1925, the Dorset culture has presented students of Eskimo archaeology with a persistent problem. Apparently representing an Arctic land-hunting culture, centering in the eastern Arctic, and containing many cultural items also to be found in more southern Indian cultures, the Dorset culture has presented ample opportunity for conjectural reconstructions of its origins and affinities. Mathiassen, who collected Dorset materials from Bylot and Southampton islands, first considered Dorset merely as a localized and specialized phase of the Thule culture (Mathiassen, 1925, part 1, pp. 206-12, 258-60; part 2, pp. 164-65). Later, however, he came to regard Dorset as non-Eskimo, and as a culture both influenced by, and influencing, Thule (Mathiassen and Holtved, 1936, p. 130). Jenness originally regarded Dorset as a paleo-Eskimo form of the proto-Eskimo culture of the Arctic Interior (Jenness, 1933, p. 365), but later shifted to the viewpoint that Dorset was a genuine Eskimo culture–albeit showing some Indian influences–which was distantly related to the cultures of Western Alaska (Jenness, 1940, p. 9). That this was the case has been repeatedly assumed as proved by de Laguna (1946, p. 106; 1947, pp. 13-14, 284-85).

Kolomoki is about six miles from the Chattahoochee River, in extreme southwestern Georgia. One of the larger southeastern sites, it covers approximately three hundred acres, including a rectangular truncated pyramidal temple mound 56 feet high, a plaza about 40 acres in extent, and 7 dome shaped mounds ranging in height from 4 to 20 feet. Three of the latter mounds have been excavated, two of which, D and E, were burial mounds and supplied parts of the data for this study (Sears, 1951b). The third excavated mound probably covered the site of a ceremonial structure and crematory area (Larsen, n.d.).

In the past 30 years discoveries of great importance in the prehistory of Mexico have been made in the vicinity of Tepexpan, a town some 35 kms. northeast of Mexico City. Diaz Lozano (1922-23, 1927) disinterred a mammoth identified as Mammuthus (Archidisko-don) imperator Leidy, whiah is now seen partially mounted in the entrance hall of the Instituto Geologico de Mexico. Mullerried (1933, 1934) collected teeth and bones of the ground sloth Mylodon harlani Owen on lands of the Colonia Estrella on the northern side of Mexico City near the old exit road to Pachuca, which continues on to Tepexpan and Teotihuacan. Both Diaz Lozano and Mullerried simply described the sequence of strata at the fossil localities, mentioning the nature and depth of the deposits, but without applying specific names to the formations.

The MacHaffie site, in the foothills a few miles south of Helena, Montana, is an unusual open site in that it has three occupation levels. Folsom Fluted points occur in the bottom; Scottsbluff points in the middle; later types of points comparable to Signal Butte II come from the top level. In 1938 A. J. Harstad, an amateur mineralogist from Helena, discovered the site. He saw flint and bones projecting from a cut bank; he later pointed these out to E. D. MacHaffie, also of Helena. In a surface survey, MacHaffie discovered artifacts of a late type, and Mrs. MacHaffie also found a Clovis Fluted point in the small creek near the site and possibly from the site itself (Fig. 62, 12). On a later visit MacHaffie found a point of Scottsbluff type on the surface. At this time he dug into the cut bank and found bones, chips, and some flaked knives.

For a number of years the subject of California archaeology has undergone a great deal of discussion. A large measure of this discussion has been concerned with the problem of chronology. Heizer (1949) has postulated a classification of cultural and temporal relationships for central California. His chronology is based primarily on stratigraphic sequences together with intrasite comparisons and geological-chemical soil and bone analysis.

This paper is an attempt to examine a part of his sequences using a new method for the chronological ordering of archaeological deposits (Brainerd 1951; Robinson 1951). The material to follow will be divided into three categories: 1) theory and method; 2) a discussion of the sequences resulting from the application of the method used here; and 3) the nature of the conclusions which one may safely draw in the light of 1) and 2) above.

Archaeological excavations along coastal Alaska from Cape Prince of Wales south to the Alaska Peninsula, a distance of approximately 2000 miles, were not initiated until 1948, when Giddings (1949) began work in the north at Cape Denbigh, and Larsen (1950) began at Bristol Bay in the southern section of this great area (see Fig. 7). During the summer of 1950 the writer carried out excavations at Hooper Bay, in approximately the center of the unworked region. Although the Hooper Bay Village material includes about 2000 artifacts and 1500 pot and lamp sherds, this paper is limited to a preliminary analysis of the cooking pots. The importance of pottery in determining the sequence within Eskimo prehistory has been neglected, in the face of the abundance of organic material which the frozen middens and house remains yield. The full significance of Alaskan pottery was not emphasized until de Laguna (1947) made her detailed study of the ware; also, Larsen (1950, p. 186) pointed out that potsherds are particularly good time indicators in a region like southwestern Alaska, where the preservation is poor. In this paper the Hooper Bay sherds are considered with particular reference to their position within the Bristol Bay-Norton Sound region and their relationship to finds of similar pottery in the north.

The purpose of this paper is to report the results of excavations conducted in 1933, as yet unreported, and to correlate the findings with present knowledge of Jamaican and other West Indian prehistory. Descriptions of the sites and specimens may be of value to archaeologists in this area.

Since 1933 great strides have been made in archaeological knowledge of the West Indies. Irving Rouse has correlated known cultures in most of the Greater Antilles and in some of the Lesser Antilles. He has established seven arbitrary time periods, I, IIa, IIb, IIIa, IIIb, IVa, and IVb, for the area (Rouse, 1939, 1941, 1948, 1951, 1952) on the basis of stratigraphy, sedation of modes, establishment of style sequences, and cross-dating of trade objects. During these periods four cultures existed. The period I culture is preceramic and is associated with the Ciboney Indians who may have come from North America. The first ceramic culture, Igneri, is associated with the Arawak Indians, who pushed north and east from the Orinoco Valley in period II. It lacks the ceremonial complex which distinguishes the two later ceramic cultures, sub-Taino and Taino, which developed in the Greater Antilles during periods III and IV, the former as a simpler variant of the latter.

Hawver cave, located about two miles from Cool, El Dorado County, California, is well known to paleontologists for its abundant and diversified Pleistocene fauna. The fossil animal remains have been described in several monographs (Furlong, 1907; Stock, 1918; Miller, 1912; Kellogg, 1912). The occurrence of human skeletal material has also been noted (Merriam, 1909; Stock, 1918, pp. 466-8), but an interesting series of artifacts recovered from the cave has received only brief mention in a newspaper article (Hamilton, 1910). These man-made objects are, on the whole, distinct from those manufactured by the historic Indians of the region and seem worthy of detailed analysis. Before describing the artifacts, however, it is necessary to review some of the facts pertaining to the cave and its contents.

It is now over half a century since Carnot (1893) published his pioneer work dealing with the chemical analysis of fossil bone. In the intervening decades the problem has been investigated by occasional students who have approached the field from different directions. Recently interest in chemical methods has been intensified and the attention of archaeologists has been brought to a focus by the attempt of Oakley (1951) to utilize the fluorine content of human bones as a criterion of age. Also during the past few years a joint project has been pursued by the Departments of Anthropology and Physiology at the University of California (Berkeley) through the generosity of the Wenner-Gren Foundation for Anthropological Research. Thus a considerable effort has been expended, and a reasonable volume of results are on record, with reference to the chemical changes which occur in fossil bone. At the present time it appears that a general survey of this work is in order, together with an appraisal of its value as a tool for the archaeologist and palaeontologist. Fairly comprehensive listings of published works on the subject of bone fossilization occur in Heizer (1950, 1952).

Although the American Philosophical Society–University of New Mexico Mackenzie Valley Expedition of 1938 (Bliss, 1939, p. 365) was not primarily concerned with Eskimo archaeology, the members felt, while at the trading rendezvous Aklavik on the lower Mackenzie River, that the opportunity to run down to the Arctic coast was too obvious to be neglected. The archaeology of the Western Eskimo of the Mackenzie area has never been well studied; little, as a matter of fact, has been added since 1930 when Mathiassen wrote the introduction to his Western Eskimo report. This paper will add somewhat to a meager store of fact.

In the collection from Lovelock cave, gathered by Loud in 1912, there recently came to light a piece of finger-woven textile which, at first glance, appeared similar to the Apocynum fabric from Humboldt cave described by O'Neale (1947). Loud did not describe or mention the textile, though he illustrates it (Loud and Harrington, 1929, Pl. 53a), and one suspects that, being reluctant to disturb the bead strands to which it was affixed, he did not recognize the specimen for what it was.

The sling (Fig. 67, a), made probably of Apocynum, was wrapped with a compound necklace of Olivella shell bead strands and was found about the neck of a partially mummified child about 6 years old. The description of the necklace, the burial, and a photograph of the necklace itself with the sling clearly apparent, occur in Loud and Harrington's report on Lovelock cave (1929, p. 105, Pls. 10, 53a).

The Bintz site is a small village site located on the left bank of the Ohio River at the mouth of Twelve-Mile Creek, Campbell County, Kentucky, on property belonging to Mrs. William C. Bintz. The village consists of two parts – the “lower village,” at the immediate edge of the river, just downstream from the mouth of the creek, and the “upper village” about two hundred meters from the river and the creek on a terrace about four meters higher than the flood-plain where the “lower village” is located (Fig. 87). The cultural material found at both areas is identical, and it appears that both areas were occupied by the same people and at the same time. With the exception of a small amount of Woodland material found on the surface and in the excavations at the “upper village,” all the material is definitely of the Fort Ancient type. Since the pottery types link the site specifically with the Madisonville and Fox Farm sites, we classify the Bintz site as a component of the Madisonville focus (Griffin, 1943).