For instance, from Canis, the Dog, Canidae has been formed, to designate the whole Family of Dogs, Wolves, Foxes, etc.  Nothing can be more superficial than such a mode of classification; and if these groups actually exist in Nature, they must be based, like all the other divisions, upon some combination of structural characters peculiar to them.  We have seen that Branches are founded upon the general plan of structure, Classes on the mode of executing the plan, Orders upon the greater or less complication of a given mode of execution, and we shall find that form, as determined by structure, characterizes Families.  I would call attention to this qualification of my definition; since, of course, when speaking of form in this connection, I do not mean those superficial resemblances in external features already alluded to in my remarks upon Parallel or Collateral Types.  I speak now of form as controlled by structural elements; and unless we analyze Families in this way, the mere distinguishing and naming them does not advance our science at all.  Compare, for instance, the Dogs, the Seals, and the Bears.  These are all members of one Order,—­that of the Carnivorous Mammalia.  Their dentition is peculiar and alike in all, (cutting teeth, canine teeth, and grinders,) adapted for tearing and chewing their food; and their internal structure bears a definite relation to their dentition.  But look at these animals with reference to form.  The Dog is comparatively slender, with legs adapted for running and hunting his prey; the Bear is heavier, with shorter limbs; while the Seal has a continuous uniform outline adapted for swimming.  They form separate Families, and are easily recognized as such by the difference in their external outline; but what is the anatomical difference which produces the peculiarity of form in each, by which they have been thus distinguished?  It lies in the structure of the limbs, and especially in that of the wrist and fingers.  In the Seal the limbs are short, and the wrists are on one continuous line with them, so that it has no power of bending the wrist or the fingers, and the limbs, therefore, act like flappers or oars.  The Bear has a well-developed paw with a flexible wrist, but it steps on the whole sole of the foot, from the wrist to the tip of the toe, giving it the heavy tread so characteristic of all the Bears.  The Dogs, on the contrary, walk on tip-toe, and their step, though firm, is light, while the greater slenderness and flexibility of their legs add to their nimbleness and swiftness.  By a more extensive investigation of the anatomical structure of the limbs in their connection with the whole body, it could easily be shown that the peculiarity of form in these animals is essentially determined by, or at least stands in the closest relation to, the peculiar structure of the wrist and fingers.