The results of carbon 14 tests on materials falling within the early man category are in some respects as surprising as a few of those in other fields, but on the whole they are reasonably satisfactory. Because of the regional distribution of the samples and the number of people who sent them in, several nominally belonging in this group are discussed elsewhere. The latter include Gypsum cave, Leonard rockshelter, the early horizons in California and Oregon and the material from South America.

The age of the well-known Folsom complex caused considerable comment when the figure pertaining to the type site was released. The average of 4283 ± 250 years obtained from sample 377 was entirely too low in the opinion of many archaeologists and geologists and was completely out of line with dates for other materials known to be later stratigraphically. Because of this, Harold J. Cook, who provided the charcoal used in the test, returned to and carefully reexamined the Folsom quarry in 1950.

Three separate dates were determined by Arnold and Libby from materials submitted by L. S. Cressman from Oregon (samples 247, 428, 430). Because of the significance of the date of the Mt. Mazama (Crater Lake) eruption to archaeological dating due to Mazama pumice having covered certain archaeological cave deposits, the eruption date of 6453 ± 250 (based upon an average of four runs of sample 247 ranging from 5938 ± 400 to 7318 ± 350) is important. Williams (Cressman and Williams, 1940, p. 76. See also Hansen, 1942, p. 105; Cressman, 1942, pp. 135, 24) estimated this eruption as occurring not less than 4000 nor more than 10000 years ago. Williams (1942, pp. 212-4) later revised this estimate to approximately 5000 years with a minimum of 4000 and maximum of 7000 years.

The Chronological relationship of the Adena and Hopewell cultures and their placement within the temporal column of North American prehistory was considered of sufficient importance by the Committee on Radioactive Carbon 14 to warrant a special effort to obtain dates. Individuals and institutions in the Mississippi Valley were requested to provide samples, with the result that specimens were submitted from an Adena site in Kentucky (126) and one in Ohio (214). Two specimens from a Tchefuncte culture midden in Louisiana thought to be of comparable age and typology to Adena (150 and 151) were processed in order to provide information on a common cultural level over a considerable area. A total of six specimens from the Hopewellian level were dated. These comprise runs on three different kinds of organic material from mound 25 of the Hopewell group (137-139), Ross County, Ohio; a specimen from mound 9 of the Havana group in Mason County, Illinois (152); material from Bynum, a Hopewell period site in northeast Mississippi (154); and from the Crooks site, a Marksville mound in LaSalle Parish, Louisiana (143). It was believed, before the samples were dated, that this would afford satisfactory information not only on the time position of Adena and Hopewell but might also clear up the question as to the priority of appearance in the north or the south of those cultural influences which produced the general stages of early to middle Woodland and their local representations.

While the sites of the Archaic manifestation (shell middens) in the Southeast are numerous and extensive, in general they do not furnish clean specimens of wood or charcoal suitable for radiocarbon dating. The use of shell or deer antler, which occurs in great quantity, for carbon dating is justified only by experimental proof of their ability to give consistent results. With the exception of sample 180, four other assays from three other samples, 116, 251, 254, three of antler and one of shell, yielded results which differ from the average by less than the range of assigned error and may thus be regarded as consistent among themselves. This places the age of the Archaic of Green River, Kentucky, at from 4900 to 5300 years, which, while greater than was anticipated, may be considered reasonable, particularly since it agrees well with the age of the Archaic manifestations in New York State, where the assays were made from charcoal.

Radiocarbon dates on charcoal samples from New York State have added significant new data to our knowledge of previously little known factors of chronology and continuity in the area. The dates have provided the lower portions of an apparently valid time scale upon which may be reared a more realistic three-dimensional picture of the prehistoric events of this portion of the United States.

Sample 191 was excavated by the writer from a hearth situated in the deepest refuse level in trench 4, section 4, of the Frontenac Island site, Cayuga County (Ritchie, 1945, pp. 6, 115, Plate 115). At Frontenac Island, an occupational component referable to the Lamoka focus appeared to precede the coming of Laurentian brachycephals with their specific cultural complex. The subsequent interaction of the two discrete peoples and cultures was evident in the physical remains, burial modes, and character of grave goods. The hearth, found on the lower level of the site, is presumed to pertain to the initial Lamoka- Laurentian contact. The radiocarbon dating yielded an antiquity of 4930 ± 260 years of elapsed time.

In the Summer of 1949, I was asked to 1 cooperate in securing appropriate samples from important Mexican sites. These, I suggested, should be selected with a view of obtaining new data on three significant problems: early man and associated geologic formations, the Archaic cultures, and the monumental civilizations of Teotihuacan and Monte Alban. This choice was dictated by the prevailing chronologic uncertainties of prehistoric cultures in Mexico and by my previous geologic approaches to time sequences in the Basin of Mexico. Computations from glacial, lake and erosion phenomena had suggested that previous age estimates, notably those presented by Vaillant (1944) and other archaeologists, had been unduly conservative. This was especially evident in the case of the basic pottery levels, known as the early Archaic or early Middle cultures, then assumed to have existed about 2100 to 1600 years ago.

The Stratigraphic data on some of the Peruvian samples provide a fairly good basis for evaluating the radiocarbon date measurements in relation to each other. The accompanying chart (Fig. 1) was prepared to clarify what is bound to be tedious reading. On it the plotting of the radiocarbon dates is obvious. The insertion with them of sectional profiles of midden excavations may be confusing, but does show where the tested material was found. In one instance (382), although the tested sample did not come from this midden, it belongs chronologically at the level indicated. Knowledge of the nature of the structure between the various layers gives us further means of assessing the results. Thus it is possible to compare the date blocks, as we might call the total plus and minus time span, and to point out which portions cannot be reconciled and are therefore impossible and which portions are improbable. The remaining segments mark the periods within which the actual dates are most likely to fall.

The Lascaux Cave, one of the most important single sites of upper Palaeolithic art ever discovered, is located near Montignac (Dordogne), 16 miles up the Vézère River from the famous town of Les Eyzies. It was found in September 1940 by two French schoolboys out rabbit hunting. Called the “Versailles of Prehistoric Man,” Lascaux ranks among the world's oldest and most remarkable art galleries. It has been completely sealed off from the outside world since late upper Pleistocene times, and it is now generally believed that the majority of the paintings are upper Perigordian (Gravettian) in date—Phase 2 in the upper Palaeolithic art sequence of western Europe. That they were produced during a time of climatic amelioration, possibly the one known as the Achen retreat, is borne out by the fact that the animals depicted belong much more to a steppe and forest type of fauna, than they do to a colder tundra group. For example, there are no mammoths and no reindeer shown, but there are many horses, a considerable quantity of cattle, bison and several ibexes.

It is an unfortunate fact that the Near-Middle Eastern area, which has produced its share of the world's more spectacular antiquities, and which has been the focus of archaeological activity for so many years, has supplied but little material for the radiocarbon dating project. The dearth of specimens would seem to be due to two main factors. First, the general lack of attention hitherto given by excavators to non-artifactual materials (i.e., unworked wood, grain, etc.). Second, the understandable reluctance of museum curators to submit actual specimens of artifacts as samples, since they are destroyed in the process.

Results from the geologic material are necessarily spotty, owing to the difficulties involved in obtaining samples that were both suitable for assay and derived from horizons whose stratigraphic position is known unequivocally. Many samples assayed were taken from horizons whose stratigraphic position is subject to doubt.

Two general results have emerged from the assay of samples of primarily geologic significance: (1) The dates obtained fall roughly into the same order as the stratigraphic sequence of deposits from which the corresponding samples were collected. From this it is inferred that most of the dates are acceptable as to relative position. (2) The dates obtained are generally more recent than had been believed by many geologists. A result of this nature has been welcomed by some, who believe it is more consistent with the observed slight weathering and erosion of the latest drift sheets than are the older dates current hitherto.

In the 31 dates of specific interest to palynology contained in Table I there are some surprises for the student of pollen chronology, although there are only three (349, 433, and 450) that must be regarded as unacceptable. For two of these, the list itself contains evidence that the stratigraphic position of the samples was incorrectly determined; we are left with only one sample (433) suggesting that the radiocarbon assay is subject to errors of an unknown nature. It follows that the new method of dating has been placed on a firm footing, and that it will amply justify the high hopes originally held for it.

To say this, however, is not to say that the entire list of dates will or should meet with unqualified acceptance. All degrees of palatability are represented, some apparent contradictions are evident, and in nearly every case a certain amount of special pleading is required to harmonize the results with what was previously known or believed. While the difficulties are not considered at present to be serious, it would be most unwise to minimize them, and the fact is that confirmation is required at every point if the newly acquired dates are not to make postglacial stratigraphy more controversial than it was before.

The Preceding discussions have considered the radiocarbon dates in relation to a number of archaeological, geological, and palynological problems. It remains to consider the endeavor as a whole, and to reach some conclusions concerning its present and future value and usefulness. Of primary importance is a consideration of what the dates mean. Basic to this is an understanding of the statistics involved. At the request of the Committee, Arnold, who has been intimately associated with the project, has kindly contributed the following explanatory paragraphs:

It seems worth while to review briefly the physical scientist's notions about the errors inherent in such measurements, as they apply to radiocarbon dates.