the earth's surface and in its atmosphere, alternations of day and night, an equable temperature, a sufficient density of the atmosphere to retain the gases which are necessary to the support of life. In order to maintain these terrestrial conditions, the following astronomical conditions must obtain: The proper distance of earth from the sun, the mass of the planet falling within certain defined limits, the obliquity of the ecliptic, the amount of water as compared with land, the surface distribution of land and water, the permanence of this distribution, dependent probably on the unique origin of our moon i.e., its being a detached portion of the earth leaving behind suitable ocean basins, an atmosphere of sufficient density and composed of suitable gases, an adequate amount of dust in the atmosphere, and atmospheric electricity. Finally, none of the other planets of the solar system combine all these complex conditions, which, upon the earth, work harmoniously to the production and the support of life; therefore, it is reasonable to suppose the other planets to be uninhabited; and moreover, the probabilities are almost as great against any other sun possessing inhabited planets. Such being the line of argument, it will be readily seen that the force of Mr. Wallace's contention depends upon the exceedingly great complexity of living organisms, and the nice balancing of conditions which it is necessary to maintain in order to produce and preserve such organisms on any planet, and the improbability that such correlated conditions exist anywhere in the universe except upon our earth. It may be well, perhaps, to have before us Mr. Wallace's position as expressed in his own words: "The combinations of causes which lead to this result [the presence of living organisms] are so varied, and in several cases dependent on such exceptional peculiarities of physical constitution that it seems in the highest degree improbable that they can all be found again combined either in the solar system or even in the stellar universe" (p. 310).
This method of reasoning from known conditions which produce known results to the conclusion that the absence of these conditions renders the same or similar results impossible, must be regarded as possessing cogency only when extended to adjacent cases. As regards the cases which are necessarily so far removed from the sphere of direct observation, the unknown so far overbalances the known that the inference as to what must be considered impossible is exceedingly precarious. What seems to be impossible in a setting which is completely within the compass of our knowledge, may be quite possible in a setting which transcends our knowledge. No one has
