the size and proportion of parts. It was found that nearly all that varied to any considerable degree from the average were dead, while most of those near the average in all respects were alive. This of course means that a certain size and proportion of parts are most favorable for a sparrow living in this climate, and that those birds that vary the most from the normal are likely to be destroyed and produce few or no descendants. In a different environment or living a different life, the size and proportion of parts of the sparrow might be different, but in any case environment determines through natural selection the size of sparrows living in a certain way in a certain place." We notice, first of all, that no definition is given of a 'part', and that the 'average' used is not specified; however, it is probable that external measurements only were made, and that the average is the arithmetical mean. We notice, secondly, the implication that sparrow size and sparrow proportion of parts may differ in different environments and with different modes of life; the implication is left vague, without reference to the facts of variation; so that the thoughtless reader might actually conceive of sparrows, in some outlandish quarter of the world, as large as geese or as small as flies. These are minor points of criticism. It is more important to note that the argument as a whole is unsound. Once sparrow-size has been fixed (whether by the direct action of environment, by the operation of natural selection, by intrinsic laws of growth, or by any other causal agency or combination of causal agencies), then deviation from the normal will probably bring with it a constitutional weakness; over-large sparrows, e.g., may have too small hearts, oversmall sparrows may have an inadequate musculature, and so forth. The experiment brings no evidence whatsoever that "a certain size and proportion of parts are most favorable for a sparrow living in this climate"; it brings no evidence, that is, that a sparrow as large as a robin (if such a bird existed) could not propagate its kind, in the climate referred to, as well as or even better than the existing species; it shows only that, when once size has been fixed, considerable departure from that size, on the part of the individual, means unbalance of the vital functions. The question "why sparrows are not as large as geese or as small as flies" is not touched; the illustration proves, at most, that a size once fixed is maintained.
The passage quoted continues as follows: "What is true of the size of the sparrow is true of all his other characteristics, including his behavior; for if he attempted to do what his structure does not fit him to do, he would be at a disadvantage and would be eliminated by natural selection." Yet on p. n we read of "changes in the mode of behavior of different groups of the common species, which necessitated corresponding changes in structure" (italics not in original,). Slips of this sort are not uncommon.
Consider again the following passages. "In studying the structure and behavior of various types of organisms we must interpret them as means of survival for the individual or for the species" (p. 10); "it is reasonable to suppose that this [consciousness] like all other characteristics of organisms, is the product of natural selection, and that it has been preserved in certain kinds of behavior of certain organisms because it has proved useful" (pp. 13 f.; italics not in original). Yet the biologists themselves are tending more and more to emphasize the occurrence of non-adaptive variations. And apart from the question of the all-sufficiency of natural selection, the author forgets that utility is not a scientific concept. The natural world is a world of causation; and when we have explained in terms of cause and effect, our explanatory task is done. Huxley says somewhere that
