If through a mutation a character appears that is neither advantageous nor disadvantageous, but indifferent, the chance that it may become established in the race is extremely small, although by good luck such a thing may occur rarely. It makes no difference whether the character in question is a dominant or a recessive one, the chance of its becoming established is exactly the same. If through a mutation a character appears that has an _injurious_ effect, however slight this may be, it has practically no chance of becoming established. [Illustration: FIG. 94. Evolution of elephant's trunk. (After Lull.)] If through a mutation a character appears that has a _beneficial_ influence on the individual, the chance that the individual will survive is increased, not only for itself, but for all of its descendants that come to inherit this character. It is this increase in the number of individuals possessing a particular character, that might have an influence on the course of evolution. This gives a better chance for improvement by several successive steps; but not because the species is more likely to mutate again in the same direction. An imaginary example will illustrate how this happens: When elephants had trunks less than a foot long, the chance of getting trunks more than one foot long was in proportion to the length of trunks already present and to the number of individuals; but increment in trunk length is no more likely to occur from an animal having a trunk more than one foot long than from an animal with a shorter trunk. The case is analogous to tossing pennies. At any stage in the game the chance of accumulating a hundred heads is in proportion to the number of heads already obtained, and to the number of throws still to be made. But the number of heads obtained has no influence on the number of heads that will appear in the next throw. [Illustration: FIG. 95. Evolution of elephant's trunk: above Maeritherium, in the middle Tetrabelodon (After Lancaster); below African elephants (After Gambier Bolton).] Owing then to this property of the germ plasm to duplicate itself in a large number of samples not only is an opportunity furnished to an advantageous variation to become extensively multiplied, but the presence of a large number of individuals of a given sort prejudices the probable future result. The question may be raised as to whether it is desirable to call selection a _creative_ process. There are so many supernatural and mystical implications that hang around the term creative that one can not be too careful in stating in what sense the term is to be used. If by creative is meant that something is made out of nothing, then of course there is no need for the scientist to try to answer such a question. But if by a creative process is meant that something is made out of something else, then there are two alternatives to be reckoned with.