It is interesting to follow this ancient dispute, particularly its course during the last few decades. In 1880, Roskoff wrote a book entitled "The Religion of the Most Primitive Nature-Peoples." In this work he assembled all the available facts, and came to the conclusion that no peoples exist who have not some form of religion. About ten years ago, however, the two Sarasins, students of Ceylon and of the primitive Veddah tribes, summed up their conclusions in the proposition: The Veddahs have no religion. If, however, we compare Roskoff's facts concerning primitive peoples with those reported by the Sarasins concerning the belief of the Veddahs in demons and magic, it appears that the facts mentioned by these investigators are essentially the same. What the former calls religion, the latter call belief in magic; but in neither case is there a statement as to what is really meant by religion. Now, we cannot, of course, come to an understanding with reference to the presence or absence of anything until we are agreed as to what the thing itself really is. Hence, the question under dispute is raised prematurely at the present stage of our discussion; it can be answered only after we have examined more of the steps in the development of myth and of the preconditions of the religion of later times. We shall therefore recur to this point in our third chapter, after we have become acquainted with such religions as may indubitably lay claim to the name. Postponing the question for the present, we will designate the various phenomena that must be discussed at this point by the specific names attaching to them on the basis of their peculiar characteristics. In this sense, there is no doubt that we may speak of ideas of magic and of demons even in the case of primitive peoples; it is generally conceded that such ideas are universally entertained at this stage of culture. But the further question at once arises as to the source of this belief in magic and in demons, and as to the influences by which it is sustained. Now, in respect to this point _two_ views prevail, even among the ethnologists who have made an intensive study of primitive peoples. The one view may briefly be called that of nature-mythology. It assumes that even far back under early conditions the phenomena of the heavens were the objects that peculiarly fascinated the thought of man and elevated it above its immediate sensible environment. All mythology, therefore, is supposed originally to have been mythology of nature, particularly of the heavens. Doubtless this would already involve a religious element, or, at least, a religious tendency. The second view carries us even farther in the same direction. It holds that the ideas of primitive man, so far as they deal with the supersensuous, are simpler than those of the more highly developed peoples. Just for this reason, however, it regards these ideas as more perfect and as approaching more nearly the beliefs of the higher religions.