In the literature of the science of religion, the prayers of the lower races of mankind have not been recorded to any great extent by those who have had the best opportunities of becoming acquainted with them, if and so far as they actually exist. This is probably due in part to their seeming too obvious and too trivial to deserve being put on record. It may possibly in some cases be due to the reticence the savage observes towards the white man, on matters too sacred to be revealed. The error of omission, so far as it can be remedied henceforth, will probably be repaired, now that savage beliefs are coming to be examined and recorded on the spot by scientific students in the interests of science. And the reticence of the savage promises to avail him but little: the comparative method has thrown a flood of light on his most sacred mysteries. There may however be another reason why the prayers of the lower races have not been recorded to any great extent: they may not have been recorded for the simple reason that they may not have been uttered. The nature and the occasion of the rite with which the god is approached may be such as to make words superfluous: the purpose of the ceremony may find adequate expression in the acts performed, and may require no words to make it clear. If a community approaches its god with sacrifice or offering, in time of sore distress, it approaches him with full conviction that he understands the circumstances and the purpose of their coming. Words of dedication--'this to thee' is a formula actually in use--may be necessary, but nothing more. Indeed, the Australian tribes, in rites analogous to harvest-offerings, use no spoken words at all. We cannot, however, imagine that the rites are, or in their origin were, absolutely without meaning or purpose. We must interpret them on the analogy of similar rites elsewhere, the purpose of which is expressed not merely, as in Australia, by gesture-language, but is reinforced by the spoken word. Indeed, we may, perhaps, go even further, and believe that as gesture-language was earlier than speech, so the earliest rites were conducted wholly by means of ritual acts or gestures; and that it was only in course of time, and as a consequence of the development of language, that verbal formulae came to be used to give fuller expression to the emotions which prompted the rites.