What I have written is not in any degree inconsistent with the belief that, by the dispensation of Providence, moral causes have a natural and often overwhelming influence upon happiness and upon success, nor yet with the belief that our moral nature enters into a very real, constant, and immediate contact with a higher power. Nor does it at all disprove the possibility of Divine interference with the order even of physical nature. A world governed by special acts of intervention, such as that which mediæval theologians imagined, is perfectly conceivable, though it is probable that most impartial enquirers will convince themselves that this is not the system of the planet we inhabit; and if any instance of such interference be sufficiently attested, it should not be rejected as intrinsically impossible. It is, however, the fundamental error of most writers on miracles, that they confine their attention to two points—the possibility of the fact, and the nature of the evidence. There is a third element, which in these questions is of capital importance: the predisposition of men in certain stages of society towards the miraculous, which is so strong that miraculous stories are then invariably circulated and credited, and which makes an amount of evidence that would be quite sufficient to establish a natural fact, altogether inadequate to establish a supernatural one. The positions for which I have been contending are that a perpetual interference of the Deity with the natural course of events is the earliest and simplest notion of miracles, and that this notion, which is implied in so many systems of belief, arose in part from an ignorance of the laws of nature, and in part also from an incapacity for inductive reasoning, which led men merely to collect facts coinciding with their preconceived opinions, without attending to those that were inconsistent with them. By this method there is no superstition that could not be defended. Volumes have been written giving perfectly authentic histories of wars, famines, and pestilences that followed the appearance of comets. There is not an omen, not a prognostic, however childish, that has not, in the infinite variety of events, been occasionally verified, and to minds that are under the influence of a superstitious imagination these occasional verifications more than outweigh all the instances of error. Simple knowledge is wholly insufficient to correct the disease. No one is so firmly convinced of the reality of lucky and unlucky days, and of supernatural portents, as the sailor, who has spent his life in watching the deep, and has learnt to read with almost unerring skill the promise of the clouds. No one is more persuaded of the superstitions about fortune than the habitual gambler. Sooner than abandon his theory, there is no extravagance of hypothesis to which the superstitious man will not resort. The ancients were convinced that dreams were usually supernatural.