For instance, those languages which have grammatic gender almost necessarily divide their deities according to sex[14]; those in which the passive voice is absent or feebly developed, will be led to associate with their deities higher conceptions of activity than where the passive is a favourite form: those which have no substantive verb cannot express God as pure being, but must associate with Him either position, action, or suffering. In the speech of the Algonquin Indians, there is no grammatic distinction of sex; but there is broad discrimination between objects which are animate and those which are inanimate. When the Catholic missionaries brought to them the rosary, the natives at first spoke of it as inanimate; but as their reverence for it grew, it was transferred to the animate gender, and was thus on its way to a personification.[15] The third source of information is that which is called folk-lore. Its field of research is to collect the relics and survivals of primitive modes of thought and expression, beliefs, customs, and notions, in the present conditions of culture. It is, therefore, especially useful in a study like the present, the more so on account of the extraordinary permanence and conservative character of religious sentiments and ceremonies. Among the peasantry of Europe, the paganism of the days of Julius Cæsar flourishes with scarcely abated vigour, though it may be under new names. “The primitive Aryan,” writes Professor Frazer,[16] “is not extinct; he is with us to-day.” And another English writer does not go too far when he says: “There is not a rite or ceremony yet practised and revered among us that is not the lineal descendant of barbaric thought and usage.”[17] It is this which gives to folk-lore its extremely instructive character for the student of early religion. The fourth source of information is the description of native religions by travellers. You might expect this to be the most accurate and therefore valuable of all the sources; but it is just the reverse. Omitting the ordinary tourist and globe-trotter, who is not expected to know anything thoroughly, and never deceives the expectation, even painstaking observers, who have lived long with savage tribes, sometimes mastering their languages, are, for reasons I shall presently state, constantly at fault about the native religions. We must always take their narratives with hesitation, and weigh them against others by persons of a different nationality and education. Indeed, of all elements of native life, this of religion is the most liable to be misunderstood by the foreign visitor. Bearing in mind these various sources of information, what tribes, about which we have sufficient knowledge, could fairly be considered as examples of primitive conditions?