The substitution in that case was largely by elements drawn from other regions and peoples. But a similar substitution and consequent change of innate mental qualities may go on slowly within any people which has been formed, as have almost all the present European nations, by an incomplete blending of two or more racial stocks; it may be effected by internal selection without any introduction of new elements from any other region. Before considering an example of the process, let us note certain facts which show that there may well have taken place, throughout the historic period, changes of the composition of peoples by internal substitution or changes of the mental constitution by internal selections—that is to say, by the more rapid multiplication of certain mental types and the relative infertility of other types. Consider first the striking fact that the populations of the various European countries seem for the most part to have remained almost stationary as regards numbers, or even in some cases to have diminished greatly in numbers, throughout the period between the Roman domination and the later part of the eighteenth century. The population of Spain is said to have declined from forty millions under the Roman rule to only six millions in the year 1700 A.D. The population of Great Britain is said to have increased from five millions to six millions only during the seventeenth century; and it is certain that in the main it had increased at an even slower rate, or not at all, in the preceding centuries since the Norman Conquest; whereas in the nineteenth century it increased from thirteen millions to nearly forty millions; that is to say, it trebled itself in the century; and even that rate of increase is considerably less than the possible maximal rate. The same is roughly true for most of the European countries; their populations, throughout great stretches of the historic period, remained stationary or increased only very slowly. Now when, during any period, a population does not multiply at the maximal physiological rate, changes of its character may well be taking place; for, in proportion as the rate of increase falls below the maximal, there is a lack of fertility in the population or in some part of it; if this relative infertility affects equally all parts and classes of the population, it will produce no change of its composition; but if it is selective, if for any reason it affects one class, or persons of some one kind of temperament or mental type, more than others, then this class or this temperament or this form of ability tends rapidly to diminish and to disappear from among that people.