Several sufficient reasons might be given. In the first place many people have, or think they have, so many other grounds for disbelief, that additional grounds, provided by the comparative method, are regarded rather as a luxury than as supplying a felt want. Again, but very few persons have leisure, or inclination, or power of mind enough to follow an elaborate argument through fifteen hundred pages, not to speak of other works on the same theme. Once more, only a minute minority are capable of testing and weighing the evidence, and criticising the tangled hypotheses on which the argument rests, or in which it is involved. But there is another and perhaps a sounder argument for indifference. The learned are aware that the evidence for all these speculations is not of the nature to which they are accustomed, either in historical or scientific studies. More and more the age insists on strictness in appreciating evidence, and on economy in conjecture. But the study of the evolution of myth and belief has always been, and still is, marked by an extraordinary use, or abuse, of conjecture. The 'perhapses,' the 'we may supposes,' the 'we must infers' are countless. As in too much of the so-called 'Higher Criticism' hypothesis is piled, by many anthropologists, upon hypothesis, guess upon guess, while, if only one guess is wrong, the main argument falls to pieces. Moreover, it is the easiest thing, in certain cases, to explain the alleged facts by a counter hypothesis, not a complex hypothesis, but at least as plausible as the many combined conjectures of the castle architects, though perhaps as far from the truth, and as incapable of verification. Of these statements examples shall be given in the course of this book. We are all, we who work at these topics, engaged in science, the science of man, or rather we are painfully labouring to lay the foundations of that science. We are all trying I to expedite progress. But our science cannot expedite progress if our science is not scientific. We must, therefore, however pedantic our process may seem, keep insisting on the rejection of all evidence which is not valid, on the sparing use of conjecture, and on the futility of piling up hypothesis upon unproved hypothesis. To me it seems, as I have already said, that a legitimate hypothesis must 'colligate the facts,' that it must do so more successfully than any counter hypothesis, and that it must, for every link in its chain, have evidence which will stand the tests of criticism.