The form, the habits, and the instincts of the pointer have been made by a slow process of artificial selection of divergent varieties of offspring, and by training of individuals continued and its effects accumulated through many generations. But this form and these habits and instincts, so laboriously produced, would be quickly destroyed by Nature. The pointer, left to himself, must either change or become extinct, because not adapted to the wild state. Such instincts and habits would not only be of no use, but would be incompatible with success in the struggle for life. But suppose for a moment that these habits and instincts were useful to the animal in a wild state; evidently they would be instantly seized upon by natural selection, and not only perpetuated but intensified until a very distinct species would be produced. The same is true of all other races of dogs. If the Newfoundland, the greyhound, and the pug were all turned loose in a forest, and if each of these kinds were admirably adapted to some place in the economy of Nature--for some special mode of food-getting without corresponding disabilities in other directions (as must be the case if made by natural selection)--there can be no doubt they would each survive, and their characters intensified; intermediate forms would disappear (for reasons which we shall see presently), and we would soon have three distinct species, or perhaps we would even call them distinct genera. =Second Difference, Intermediate Forms.=--Natural species are distinct--marked out with hard and fast lines--while artificially-made races, even though in their typical forms they differ as much or more than natural species, shade into one another by insensible gradations. In answer and explanation of this difference we remark: If species or modified forms of any kind, whether natural or artificial, are made by natural causes, and not at once out of hand by supernatural creation, then of course there must have been gradations in the process of making. Now, in the artificial case, the whole process as well as the result lies within the limits of observation, while in the natural case only the final result. But it will be asked, Why are the gradations not seen also in the final result? We answer, because the intermediate forms are eliminated in the struggle for life, and not reproduced by cross-breeding. If artificial races always bred true--i. e., without crossing, as natural species do--they would probably soon be as sharply demarked. Cross-breeding is the great cause of the shadings between domestic races. This brings me to the third and most important difference.