Take any analogous case. The selective agency of specific gravity which is utilised in gold-washing does not create the original differences between gold-dust and dust of all other kinds. But these differences being presented by as many different bodies in nature, the gold-washer takes advantage of the selective agency in question, and, by using it as a cause of segregation, is enabled to separate the gold from all the earths with which it may happen to be mixed. So far as the objects of the gold-washer are concerned, it is immaterial with what other earths the gold-dust may happen to be mixed. For although gold-dust may occur in intimate association with earths of various kinds in various proportions, and although in each case the particular admixture which occurs must have been due to definite causes, these things, in relation to the selective process of the washer, are what is called accidental: that is to say, they have nothing to do with the causative action of the selective process. Now, in precisely the same sense Darwin calls the multitudinous variations of plants and animals accidental. By so calling them he expressly says he does not suppose them to be accidental in the sense of not all being due to definite causes. But they are accidental in relation to the sifting process of natural selection: all that they have to do is to furnish the promiscuous material on which this sifting process acts. Or let us take an even closer analogy. The power of selective breeding by man is so wonderful, that in the course of successive generations all kinds of peculiarities as to size, shape, colour, special appendages or abortions, &c., can be produced at pleasure, as we saw in the last chapter. Now all the promiscuous variations which are supplied to the breeder, and out of which, by selecting only those that are suited to his purpose, he is able to produce the required result--all those promiscuous variations, in relation to that purpose, are accidental. Therefore the selective agency of the breeder deserves to be regarded as the cause of that which it produces, or of that which could not have been produced but for the operation of such agency. But where is the difference between artificial and natural selection in this respect? And, if there is no difference, is not natural selection as much entitled to be regarded as a true cause of the origin of natural species, as artificial selection is to be regarded as a true cause of our domesticated races? Here, as in the case of the previous illustration, if there be any ambiguity in speaking of variations as accidental, it arises from the incorrect or undefined manner in which the term "accidental" is used by Darwin's critics. In its original and philosophically-correct usage, the term "accident" signifies a property or quality not essential to our conception of a substance: hence, it has come to mean anything that happens as a result of unforeseen causes--or, lastly, that which is causeless.