A satisfactory hypothesis of the development of the hydrosphere and atmosphere depends upon evidence from many sciences and the solution of many other fundamental problems of earth history. But because it is so closely related to many other problems, any progress toward unravelling the history of the hydrosphere and atmosphere limits the range of permissible speculation about such distantly related questions as the origin of the solar system, continents, mountains, and living organisms. Several hypotheses of the source of the earth’s air and waters are examined for their consistency with established principles and observed geologic evidence, and special attention is given to the probable composition of the early atmosphere.

Hypotheses of the origin of the atmosphere and hydrosphere fall into two chief categories: (1) that all air and water of the earth are residual from a dense primitive atmosphere that once enveloped a molten globe; or (2) that they have accumulated at the earth’s surface by leakage from the interior.

The quantities of water, carbon dioxide, organic carbon, nitrogen, sulfur, etc., that have been or are now part of the earth’s atmosphere and hydrosphere may be estimated within reasonable limits of uncertainty and these “excess” volatiles afford a basis for testing chemical consequences of the alternative hypotheses. Several writers have suggested that the primitive atmosphere may have been composed largely of CH₄ and NH₃. However, the equilibrium constants for reactions of these and other gases, combined with the evidence of the “excess” volatiles, indicate that CO₂ and N₂ are much more likely. The stabilities of methane and ammonia depend upon the presence of free hydrogen; and the escape rate of hydrogen from the earth is such that methane probably could have persisted in significant amounts in the early atmosphere no more than 10⁶ to 10⁸ years. For all but a relatively brief period at the very beginning of earth history, the atmosphere probably contained CO₂ and N₂ rather than CH₄ and NH₃.

When the consequences of a dense atmosphere of CO₂ and N₂ (but with almost no free O₂ or H₂) are examined, it is found that several chemical effects (such as the quantity of rocks that would have to be weathered, of sodium dissolved in sea water, and of CaCO₃ deposited on the sea floor very early in early history) are not borne out by the observed geologic record. From this and other lines of evidence it seems extremely improbable that the present atmosphere and hydrosphere are residual from any such dense primitive atmosphere. Instead, it seems likely that the atmosphere and hydrosphere have accumulated gradually during geologic time by the escape of water vapor, CO₂, CO, N₂, and other volatiles from intrusive and extrusive rocks that have risen more or less continuously from the deep interior of the earth.

The amount of free oxygen in the early atmosphere is a separate problem that cannot be solved until the evidence of the earliest rocks has been appraised more fully. Current hypotheses of the origin of life appear to require a reducing atmosphere, yet it seems likely that oxygen has been accumulating from the photodissociation of water vapor ever since the earth was formed. The oxidation of ferrous iron and sulfides in the earliest sediments may have kept the oxygen content very low, and life may have begun in local reducing environments.