This paper summarizes certain facts of observation, as reported by a number of investigators, concerning the nature of climatic variations on the earth's surface during observational time, during postglacial or historical time, and during the geological past. Particular emphasis is placed on possible interpretations or explanations of these changes. The discussion of the secular variations of the world weather patterns during observational time stresses on the one hand apparent relationships between these variations and solar activity, as indicated by sunspots, and on the other hand the essential similarity of these secular variations to those which occur from week to week and month to month in the fluctuation of the general-circulation pattern between more zonal (high index) and more cellular (low index) characteristics. The discussion of the variations of world climate during geological time is concerned primarily with the relative merits of the three principal theories which currently attempt to account for glacial and interglacial climates. Of these three—(l) continental uplift combined with volcanic activity, (2) the geometrical variations of insolation due to variable ellipticity of the earth's orbit, precession, and variable tilt of the earth's axis of rotation, and (3) variations in the solar constant—emphasis is placed on the advantages of the solar variability theory. Simpson's explanation of the probable role of an increase of the solar constant as a favorable condition for glaciation is stressed, and in particular the advantages which are gained for Simpson's argument by the assumption that the solar-constant change occurs principally in the ultraviolet portion of the spectrum. The essential similarity of the changes of climate during postglacial time as intermediate between the geological changes on the one hand and the secular changes on the other is attributed to the basic change of the general-circulation pattern which evidently characterizes all of these climatic variations. This basic change apparently consists essentially of a contraction or expansion of the circumpolar zonal weather pattern, accompanied by decrease and increase, respectively, of cyclonic activity in the low-pressure belts. It is emphasized that solar variability of the sunspot variety is the only possible single factor of climatic control which might be made to account for all of these variations.