In truth the distinction between the conclusions of science and those of religious creeds might be more properly expressed by saying that the former are ‘working hypotheses’ which never fail, while the latter are ‘working hypotheses’ which frequently fail. Thus, the fundamental hypothesis of Cardinal Newman and his school of a one infinite and eternal personal Deity, who regulates the course of events by frequent miraculous interpositions, so far from being a necessary and axiomatic truth, has never appeared so to the immense majority of the human race: and even at the present day, in civilised and so-called Christian countries, its principal advocates complain that ninety-nine out of every hundred practically ignore it. It is not so with the uniformity of the laws of nature. No palæolithic savage ever hesitated about putting one foot after another in chase of a mammoth from a fear that his working hypothesis of uniform law might fail, the support of the solid earth give way, and with his next step he might find himself toppling over into the abyss of an infinite vacuum. In like manner Greeks and Romans, Indians and Chinese, monotheists, polytheists, pantheists, Jews and Buddhists, Christians and Mahometans, all use standard weights in their daily transactions without any misgivings that the law of gravity may turn out not to be uniform. But religions theories vary from time to time and from place to place, and we can in a great many cases trace their origins and developments like those of other political and social organisms.