It is an almost universal characteristic of sedimentary accumulations, whether modern deposits or ancient rocks, that they are stratified, that is, divided into more or less parallel layers or beds. Indeed, the terms secondary, derivative, sedimentary, and stratified rocks are but different names for the same thing. Stratification is due to the sorting power of water, or of the wind, by which, so long as conditions remain the same, particles or fragments of similar size and weight are thrown down at the same spot. If sand, gravel, mud, and clay are shaken together in a jar of water and then allowed to stand, the various materials will settle to the bottom in the order of their coarseness, the finest coming down last. Yet the change from one kind of material to another will be so gradual that no well-defined layers will appear, and thus no true stratification results. Layers clearly demarcated from one another may be produced in either one of two ways: (i) by such a change of conditions that the material deposited changes abruptly, though perhaps only as a mere film of a different substance, or (2) by a pause, however brief, in the process of deposition.

In the latter case, each layer represents a time of deposition broken by an interval which allows the surface particles to arrange themselves somewhat differently from the position they would take were the deposition continuous. The planes of contact between the successive layers, which may be indistinct or very sharply defined, are called the bedding or stratification planes, and each one of these formed the surface of the lithosphere, either as a land surface or the bottom of some body of water, for a short time. The thickness of each layer indicates the length of time during which the deposition of similar material went on without interruption, and varies from hundreds of feet to a small fraction of an inch.

The power of ordinary winds to transport material is much less than that of water, and wind-borne debris is, on the average, much finer than water-borne sediment, and furthermore the winds are less constant in direction and subject to greater and more sudden changes of velocity. Consequently, stratification by the wind is, as a rule, less even and regular than that which is due to water; but still wind-made deposits are stratified, and it is not always practicable to distinguish with certainty between the two classes. Fine volcanic ash and dust may be spread by the winds over immense areas and in very regular beds or strata.

The sorting power of water or wind results in the concentration of similar material, so that, as a rule, each bed is made up of some predominant substance in a state of greater or less purity, such as gravel, sand, clay, etc., and thus heterogeneous material is separated into its constituent parts, though the separation is rarely quite complete, and sometimes there is hardly any separation at all. On examining a thick series of deposits, we find that the materials are apt to change both vertically and horizontally. Changes in the vertical direction imply changes of conditions, in accordance with which different kinds of material are successively laid down over the same area, so that gravel is deposited on sand, sand on mud, or vice versa. Such changes are usually abrupt, so that each stratum is sharply demarcated from the one above and the one below it. On the other hand, changes in material in the horizontal direction are usually gradual, and a bed of sand may pass by imperceptible transitions into one of gravel or of mud. This is because of the gradual change in the velocity of the transporting agent and therefore of its carrying power.

In the sea or a large lake the material on the bottom grows finer outward from the shore, while a river, whose velocity diminishes from head waters to mouth, lays down material of decreasing coarseness, from the boulders and cobbles of the head waters to the fine silt of the lower course.

Each agent of reconstruction, or deposition, has its own characteristic manner of accumulating material, and, in typical instances, it is easy to distinguish between them, but there are also many similarities and, as we have already learned, it is sometimes exceedingly difficult to determine which of several possible agents was the actual means of forming a given series of deposits. If no fossils (i.e. recognizable traces of animals or plants) are present, it is not always easy to determine, for example, whether a given sandstone was laid down in the sea, or in a lake, or heaped up by the winds in a desert. This uncertainty is, however, largely due to our ignorance concerning all the minute details of structure which characterize the work of each agent, and may be expected to disappear as knowledge of these details advances. Of late years great progress has been made in these matters, and systematic study, it may reasonably be hoped, will remove an ignorance which is owing chiefly to a neglect of the subject and to certain preconceptions inherited from the early days of geology.