growth, recall this gradation. He has a fish-like, a reptile-like stage before he shows unmistakable mammal-like features. We do not on this account suppose a quadruped grows out of a fish in our time, for this simple reason, that we live among quadrupeds and fishes, and we know that no such thing takes place. But resemblances of the same kind, separated by geological ages, allow play for the imagination, and for inference unchecked by observation."
I do not believe that Prof. Agassiz's worst enemy—if he ever had an enemy—could have been so hard-hearted as to wish for him the direful catastrophe into which this wonderful piece of argument has plunged him irretrievably. For the question must at once suggest itself to every reader at all familiar with the subject, If Prof. Agassiz supposes that the development theory, as held nowadays, implies that a quadruped was ever the direct issue of a fish, of what possible value can his opinion be as regards the development theory in any way?
If I may speak frankly, as I have indeed been doing from the outset, I will say that, as regards the Darwinian theory, Prof. Agassiz seems to me to be hopelessly behind the age. I have never yet come across the first indication that he knows what the Darwinian theory is. Against the development theory, as it was taught him by the discussions of forty years ago, he is fond of uttering, I will not say arguments, but expressions of dislike. With the modern development theory, with the circumstances of variation, heredity, and natural selection, he never, in any of his writings, betrays the slightest acquaintance. Against a mere man of straw of his own devising, he industriously hurls anathemas of a quasi-theological character. But any thins: like a scientific examination of the character and limits of the agency of natural selection in modifying the appearance and structure of a species, any thing like such an examination as is to be found in the interesting work of Mr. St. George Mivart, he has never yet brought forth.
Now, when Prof. Agassiz fairly comes to an issue, if he ever does, and undertakes to refute the Darwinian theory, these are some of the questions which he will have to answer: 1. If all organisms are not associated through the bonds of common descent, why is it that the facts of classification are just such as they would have been had they been due to such a common descent? 2. Why does a mammal always begin to develop as if it were going to become a fish, and then, changing its tactics, proceed as if it were going to become a reptile or bird, and only after great delay and circumlocution take the direct road toward mammality? In answer to this, we do not care to be told that a mammal never was the son of a fish, because we know that already; nor do we care to hear any more about the "free manifestations of an intelligent mind," because we have had quite enough of metaphysical phrases which do not contain a description of some actual or imaginable process. We want to know how this state of things can be sci-
