Here a wonderful and one would say unaccountable transference takes place in a short time. As Steenstrup showed, one eye actually passes through the head while the young fish is growing. We ask how this comes about; and we are told, truly enough, that it takes place in each generation because it did so in the parents and in the whole line of ancestors. Why offspring should be like parent is more than any one can explain; but so it is, in a manner so nearly fixed and settled that we can count on it; yet not from any absolute necessity that we know of, and, indeed, with sufficiently striking difference now and then to demonstrate that it might have been otherwise, or is so in a notable degree. This transference of one eye through the head, from the side where it would be nearly useless to that in which it may help the other, bears all the marks of purpose, and so carries the implication of design. The case is adduced as part of the evidence that Darwinian evolution supersedes design. But how? Not certainly in the way this goes on from generation to generation; therefore, doubtless in the way it began. So we look for the explanation of how it came about at the first unintentionally or accidentally; how, under known or supposed conditions, it must have happened, or at least was likely to happen. And we read, "A spontaneous variation occurred, consisting in the passage of one eye to the opposite side of the head." That is all; and we suppose there is nothing more to be said. In short, this surprising thing was undesigned because it took place, and has taken place ever since! The writer presumes, moreover (but this is an obiter dictum), that the peculiarity originated long after flounders had fixed the habit of swimming on one side (and in this particular case it is rather difficult to see how the two may have gone on pari passu), and so he cuts away all obvious occasion for the alteration through the summation of slight variations in one direction, each bringing some advantage.