On the other hand, let it not be thought unscientific to advocate the claims of an unproved hypothesis. It is the nature of hypotheses to be unproved. As they gather proof, the hypothetical becomes a theory. At length the theory goes on to demonstration. The use of hypotheses has often been explained. The human mind is easily exhausted by the observation of numerous incoherent facts. It is impelled to arrange and classify, to find some thread or threads of association on which the facts may be strung, some principle on which they may be parcelled out into groups. The arrangement may be erroneous, the explanatory principle untrue; they may be so plausible, so apparently satisfactory, so fascinating withal in general aspect, as for a long time to hinder the real solution of great problems; and nevertheless it may be judged that their services in the advancement of knowledge far outweigh the hindrances caused by the too servile acceptance accorded them. The foibles of a great writer may long infect the literature which his greatness has ennobled. A constitution grandly conceived in proportion to the moral and social ideas prevailing at the time of its conception may be clung to with servility long after it has been outstripped by the progress of civilization. But neither the genius of the poet nor the skill of the lawgiver could be spared in its own day and generation; neither could have been sacrificed to prevent the follies of the plagiarist, or the dulness of eyes that in after ages might read the letter without discerning the spirit. We may almost say that a bad hypothesis is better than none; but a wrong hypothesis need not be a bad one. It may tend so to group around it the facts contributed by supporters and opponents, that when the real explanation of them all comes to be suggested, the fraternization of the confronting armies may be easy, and the truth be greeted and acknowledged with general acclaim. THE IMPERFECTION OF THE GEOLOGICAL RECORD.