This conception, to which we have been led by the study of totemism alone, has the additional recommendation that many scholars have recently adopted it quite independently of one another, as a conclusion from very different sorts of studies. There is a tendency towards a spontaneous agreement on this point which should be remarked, for it is a presumption of objectivity. As early as 1899, we pointed out the impossibility of making the idea of a mythical personality enter into the definition of religious phenomena.[660] In 1900, Marrett showed the existence of a religious phase which he called _preanimistic_, in which the rites are addressed to impersonal forces like the Melanesian mana and the wakan of the Omaha and Dakota.[661] However, Marrett did not go so far as to maintain that always and in every case the idea of a spirit is logically and chronologically posterior to that of mana and is derived from it; he even seemed disposed to admit that it has sometimes appeared independently and consequently, that religious thought flows from a double source.[662] On the other hand, he conceived the mana as an inherent property of things, as an element of their appearance; for, according to him, this is simply the character which we attribute to everything out of the ordinary, and which inspires a sentiment of fear or admiration.[663] This practically amounts to a return to the naturist theory.[664] A little later, MM. Hubert and Mauss, while attempting to formulate a general theory of magic, established the fact that magic as a whole reposes on the notion of mana.[665] The close kinship of the magic rite and the religious rite being known, it was even possible to foresee that the same theory should be applied to religion. This was sustained by Preuss in a series of articles in the _Globus_[666] that same year. Relying chiefly upon facts taken from American civilizations, Preuss set out to prove that the ideas of the soul and spirit were not developed until after those of power and impersonal force, that the former are only a transformation of the latter, and that up to a relatively late date they retain the marks of their original impersonality. In fact, he shows that even in the advanced religions, they are represented in the form of vague emanations disengaging themselves automatically from the things in which they reside, and even tending to escape by all the ways that are open to them: the mouth, the nose and all the other openings of the body, the breath, the look, the word, etc. At the same time, Preuss pointed out their Protean forms and their extreme plasticity which permits them to give themselves successively and almost concurrently to the most varied uses.[667] It is true that if we stick to the letter of the terminology employed by this author, we may believe that for him the forces have a magic, not a religious nature: he calls them charms (_Zauber, Zauberkräfte_). But it is evident that in expressing himself