All this is merely a conceptual physical model of a process about which we know nothing at all. It is meant to illustrate the view that the organisation of a plant or animal is not something absolutely fixed and invariable. The organism in respect of each recognisable and measurable character oscillates about a point of stability, that is to say exhibits fluctuating variations about the mean value of this character. If the stability of the organisation is upset, so that it oscillates, or fluctuates about a new centre, that is, if the variations deviate in either direction from a new “type” or mean, a mutation has been established. A mutation is not, therefore, necessarily a large departure from “normality.” It is not necessarily a “discontinuous variation,” nor a “sport” nor a “freak.” It is essentially a shifting of the mean position about which the variations exhibited by the organism fluctuate. Such a mutation will, in general, involve the creation of an “elementary species.” We have considered only one character, say stature, in the above discussion, but it generally happens that the assumption of a new centre of stability involves _all_ the characters of the mutating organism. An elementary species therefore differs a little in respect of all its characters from the species from which it arose, or from the other elementary species near which it is situated. This is what we do usually find in the cases of the “races,” or “local varieties,” of any one common species of plant or animal. That we do not recognise that most, or perhaps all, of the species known to systematic biology are really composed of such local races is merely because such results involve an amount of close investigation such as has not generally been possible except in the few cases studied with the object of proving such variability; or in the case of those species which are studied with great attention to detail because of their economic importance. Thus the herrings of North European seas can be divided into such races, and it is possible for a person possessing great familiarity with these fishes to identify the various races or elementary species--that is, to name the locality from which the fish were taken--by considering the characteristics in respect of which the herrings of one part of the sea differ from those of other parts.