On the other point we lean to the opinion of Professor Owen. We agree with him in thinking that where a rational correlation (in the highest sense of the term) can be made out, it affords a better basis for deduction than an empirical correlation ascertained only by accumulated observations. Premising that by rational correlation is not meant one in which we can trace, or think we can trace, a design, but one of which the negation is inconceivable (and this is the species of correlation which Cuvier's principle implies); then we hold that our knowledge of the correlation is of a more certain kind than where it is simply inductive. We think that Professor Huxley, in his anxiety to avoid the error of making Thought the measure of Things, does not sufficiently bear in mind the fact, that as our notion of necessity is determined by some absolute uniformity pervading all orders of our experiences, it follows that an organic correlation which cannot be conceived otherwise, is guaranteed by a much wider induction than one ascertained only by the observation of organisms. But the truth is, that there are relatively few organic correlations of which the negation is inconceivable. If we find the skull, vertebræ, ribs, and phalanges of some quadruped as large as an elephant; we may indeed be certain that the legs of this quadruped were of considerable size--much larger than those of a rat; and our reason for conceiving this correlation as necessary, is, that it is based, not only upon our experiences of moving organisms, but upon all our mechanical experiences relative to masses and their supports. But even were there many physiological correlations really of this order, which there are not, there would be danger in pursuing this line of reasoning, in consequence of the liability to include within the class of truly necessary correlations, those which are not such. For instance, there would seem to be a necessary correlation between the eye and the surface of the body: light being needful for vision, it might be supposed that every eye must be external. Nevertheless it is a fact that there are creatures, as the _Cirrhipedia_, having eyes (not very efficient ones, it may be) deeply imbedded within the body. Again, a necessary correlation might be assumed between the dimensions of the mammalian uterus and those of the pelvis. It would appear impossible that in any species there should exist a well-developed uterus containing a full-sized foetus, and yet that the arch of the pelvis should be too small to allow the foetus to pass. And were the only mammal having a very small pelvic arch, a fossil one, it would have been inferred, on the Cuvierian method, that the foetus must have been born in a rudimentary state; and that the uterus must have been proportionally small. But there happens to be an extant mammal having an undeveloped pelvis--the mole--which presents us with a fact that saves us from this erroneous inference. The young of the mole are not born