causes in operation, though the matter is presented as though there were.
But passing over this, let us now represent to ourselves in detail this process which Prof. Weismann thinks will, in thousands of generations, effect the observed reduction of the eyes: the process being that at each successive stage in the decrease, there must take place variations in the size of the eye, some larger, some smaller, than the size previously reached, and that in virtue of the economy, those having the smaller will continually survive and propagate, instead of those having the larger. Properly to appreciate this supposition, we must use figures. To give it every advantage we will assume that there have been only two thousand generations, and we will assume that, instead of being reduced to a rudiment, the eye has disappeared altogether. What amounts of variation shall we suppose? If the idea is that the process has operated uniformly on each generation, the implication is that some advantage has been gained by the individuals having the eyes 1⁄2000 less in weight; and this will hardly be contended. Not to put the hypothesis at this disadvantage, let us then imagine that there take place, at long intervals, decreasing variations considerable in amount—say 1⁄20, once in a hundred generations. This is an interval almost too long to be assumed; but yet if we assume the successive decrements to occur more frequently, and therefore to be smaller, the amount of each becomes too insignificant. If, seeing the small head, we assume that the eyes of the Proteus originally weighed some ten grains each, this would give us, as the amount of the decrement of 1⁄20, occurring once in a hundred generations, one grain. Suppose that this eel-shaped amphibian, about a foot long and more than half an inch in diameter, weighs three ounces—a very moderate estimate. In such case the decrement would amount to 1⁄1440 of the creature's weight; or, for convenience, let us say that it amounted to 1⁄1000 which would allow of the eyes being taken at some fourteen grains each. To this extent, then, each occasional decrement
- I find that the eye of a small smelt (the only appropriate small fish obtainable here, St. Leonards) is about 1⁄180 of its weight; and since in young fish the eyes are disproportionately large, in the full-grown smelt the eye would be probably not more than 1⁄200 of the creature's weight. On turning to highly-finished plates, published by the Bibliographisches Institut of Leipzig, of this perenni-branchiate Proteus, and other amphibians, I find that in the nearest ally there represented, the caducibranchiate axolotl, the diameter of the eye, less than half that of the smelt, bears a much smaller ratio to the length of the body; the proportion in the smelt being 1⁄26 of the length, and in the axolotl about 1⁄56 (the body being also more bulky than that of the smelt). If, then, we take the linear ratio of the eye to body in this amphibian as one half the ratio which the fish presents, it results that the ratio of the mass of the eye to the mass of the body will be but one eighth. So that the weight of the eye of the amphibian will be but 1⁄1600 of that of the body. It is