The excavations and studies of the next few years may not provide evidence as startling as King’s, but radiocarbon promises to date any site of early man where a few ounces of charcoal can be found. Among the first dates provided by Libby was one between 6,300 and 6,900 years ago for trees charred by the eruption of pumice from Mt. Mazama, Oregon, and beneath this pumice lie four or five sites where L. S. Cressman found artifacts of early man and the fossils of horse and camel.[73] By means of radiocarbon and, here and there, newer and perhaps equally refined techniques, anthropologists are beginning to learn a little more about the relationships in time and culture between the prehistoric makers of the weapons and tools we have described. Yet out of more than 150 sites, there are only a few where datings are fully reliable. From the Glacial to the Archaic The record of early man in North America may be dealt with on five levels, each overlapping another. The most recent is sometimes called the Archaic Period—the stage before native civilizations took shape—and it lies outside the range of early man as we define him. The earliest of the other four levels is the most significant and the least known. We have no suitable name for this period, which stretches back from about 15,000 years ago into the very dim past. It may have begun early in the last glacial age or even in the third interglacial; this would take our earliest man back 50,000 to 100,000 years. There are no firm dates for this. The best we have is a radiocarbon estimate of 30,000 years ago, when man may have killed dwarf elephants on an island off the coast of California. Let us start with some dates and associations that follow the end of that mysterious first level 15,000 years ago, and carry us up to the brink of the Archaic. Let us then go back to the few hazy glimpses we now have of the very beginnings of man in the Americas. We must start with the fluted points of Clovis. They may range from 10,000 to 15,000 years back—more likely 11,000 to 13,000. It is still uncertain whether the shouldered points of Sandia fit into this time span or precede it. For various technical reasons, we must throw out a date of more than 37,000 years ago for Clovis points from Lewisville, Texas, and a date of 8,500 at the Lehner site in Arizona. Points remarkably like those from the Clovis-type site appear in such places as Bull Brook, Massachusetts,[74] and Alaska,[75] but dating is either lacking or insecure, and a 10,000- to 15,000-year age may not agree comfortably with other geological or cultural associations. Archaeologists have been tempted to associate this horizon with the hunting of mammoths.[76] The Folsom period is next, and fairly well dated. The span of time during which these classic fluted points were made seems to fall mainly between 9,000 and 11,000 years ago. The most noted big game of the Folsom hunters was _Bison antiquus_.[77]