Taking the preceding considerations into account, the question will perhaps arise,--How then is a genus to be defined? To which I may reply that, were I asked whether genera had any real existence in the animate world, my answer would be that they undoubtedly have,--though not in the sense (which is so commonly supposed) of abrupt and disconnected groups. I conceive them to be gradually formed nuclei, through the gathering together of creatures which more or less resemble each other, around a central type: they are the _dilatations_ (to use our late simile) along a chain which is itself composed of separate, though differently shaped links,--the links being the actual species themselves, and the swellings, or nodes, the slowly developed genera into which they naturally fall. When I say "slowly developed," my meaning may possibly require some slight comment. It is simply therefore to guard against the fallacy, which I have so often disclaimed, that genera are abruptly (or suddenly) terminated on their outer limits, that the expression has been employed. Though I believe that a series of _species_, each partially imitating the next in contact with it, is Nature's truest system; yet we must be all of us aware that those species do certainly tend, in the main, to map out assemblages of divers phases and magnitudes, distinguished by peculiar characteristics which the several members of each squadron have more or less in common. So that it is only in the middle points that these various groups, respectively, attain their maximum,--every one of which (by way of illustration) may be described as a _concentric bulb_, which becomes denser, as it were, in its successive component layers, and more typical, as it approaches its core.