have been derived from the wolf, or to belong to the same family as the wolf, it is, of course, impossible to produce, by any crossing of different breeds of dogs, an animal that would not belong to the class of the _Canidæ_. Indeed, it is conceded by Darwin, with all the array of facts which he adduces in regard to the domesticated animals, that by crossing we can only get forms in some degree intermediate between the parents; and that although a race may be modified by occasional crosses, if aided by careful selection of the individuals which present the desired character, yet to obtain a race intermediate between two distinct races would be very difficult, if not impossible. If this is so, how much more remote must be the possibility, by any selection, or by any crossing to which Nature will allow the different animals to submit, to produce an animal of so distinct a type that it would amount to a different species from its known progenitors! From all that has been brought about in the efforts of man to improve or to vary the breeds of domestic animals--a kind of selection that is supposed to be analogous to what takes place in Nature, although under different conditions--it is apparent that there are limitations to the power of selection in regard to the effects that are to be attributed to it. A line must be drawn somewhere. It will not do in scientific reasoning, or in any other reasoning, to ignore the limitations to which all experience and observation point with unerring certainty, so far as experience and observation furnish us with facts. It is true that the lapse of time during which there has been, with more or less success, an intentional improvement in the breeds of domestic animals carried on with recorded results has been very short when compared with the enormous period that has elapsed since the first creation of an animal organization, whenever or whatever that creation was. But history furnishes us with a pretty long stretch of time through which civilized, half-civilized, and savage nations have had to do with various animals in first taming them from a wild state and then in domesticating so as to make them subservient to human wants, and finally in improving their breeds. But there is no recorded or known instance in which there has been produced under domestication an animal which can be said to be of a different species from its immediate known progenitors, or one that differed from its remote known progenitors in any but minor and adventitious peculiarities of structure. If in passing from what has been done by human selection in the breeding of animals to what has taken place in Nature in a much longer space of time and on a far greater scale, we find that in Nature, too, there are limitations to the power of that agency which is called natural selection--that there is an impassable barrier which Nature never crosses, an invincible division between the different species of