(1) Estimate of the time needed to produce observed changes in culture. There is no exact basis for such estimates, and this method, though often the only one available, is extremely inaccurate and inadequate.
(2) Estimate of the time necessary for the performance of certain geological work. This is used to date geological horizons in which human remains occur. For example, the date of a high river terrace may be given if it is estimated that it must have taken the river at least one hundred thousand years to cut down its bed to the present level. Such estimates are almost equally uncertain and inadequate for the purposes of the prehistorian.
(3) Extent of the breakdown of radioactive minerals in certain rocks. This method is based on a process that can be measured with some exactness and has been used to determine the age of the great geological epochs but cannot yet be applied accurately to any period that is much under half a million years.
(4) Geochronology; that is, the counting of the annual layers (varves) of sediment deposited by the melt-waters of a retreating ice-sheet. This method, first devised by the Swedish scientist, De Geer, has been applied to the Scandinavian region as the basis of an absolute system of chronology covering the last twelve thousand years. Owing to the scarcity of archaeological remains found in situ in dated varves, this system is of value principally in correlating and dating the different stages of the retreat of the last ice-sheet and the major fluctuations in sea-level in the Baltic area. These in turn serve to date many deposits which, by pollen analysis, can be assigned to the various post-glacial climatic cycles. Thus indirectly the archaeological remains found in these deposits can be tentatively dated.
(5) Dendrochronology. This method, devised by Douglass and applied first in the southwestern United States, is based on the fact that certain species of trees, especially in arid regions, show by the thickness of their annual rings of growth the alternation of relatively wet or dry years. By matching many specimens from trees of various ages it has been possible to construct a time scale by which to date timbers found in prehistoric ruins.
(6) Historical evidence. Late prehistoric cultures in backward areas can sometimes be dated in a general way by the presence of imported objects from a known historical culture in some more advanced area, where written documents already exist.
It is thus seen that dates assigned to prehistoric periods are, for the present at least, almost entirely a matter of estimate. In some cases the estimates of most experts are in more or less general agreement, but often they vary widely, depending upon the person who makes them and the particular methods he relies upon as a basis for his dating.