The Arunta are totemistic—that is, they believe in close relations which bind up the groups of their society with certain plants and animals. But they differ vastly from other totemistic races all over the world, and even in Australia. So much do they differ that it may be doubted whether their totems can properly be called totems at all. Elsewhere a man of a given totem—say the emu—cannot marry a woman of that stock; it is incest. The children inherit their totem, either from the mother, or, less frequently, the father. Any local group in a given region contains persons of various totems. People may not kill, eat, or make any use of the plants and animals which, in each case, are their totems.

Among the Arunta all is otherwise. A child's totem may be that of his father, of his mother, or different from that of either parent. A man may marry a woman of his own totem, which elsewhere is incest, and capitally punished. Thus, father is a grub; mother is a grub; one child may be a grub, another an emu. Moreover, here totems are local; almost every one in a given place will be, for example, a lizard or a plum tree. Usually people do eat their own totems, though sparingly, at magical rites, intended to multiply the animal or plant with which it is associated, in the interests of the general food supply. The Grubs work a rite to cause plenty of grubs, and they give the other groups a lead by eating sparingly of the first fruits of the grubs. This bears, in my opinion, no strong analogy to the so-called 'totem-sacrament.' To work the magic, the men of the grub or other totem must eat a little of it. This probably confirms their relation to the grub, but involves no religious element. They do not adore the grub. If any one likes to call this a 'totem-sacrament,' he is rather easily satisfied. Nor does it agree with the notion that a man's totem is the receptacle of his 'life' or 'soul;' if so, why should he encourage his neighbour to kill and eat it? Nay, he even helps them to destroy it.

Whether Arunta totemism is the most archaic kind, from which all other totemism has varied, or whether it is a private 'sport' from the main stock, does not concern us here, and is matter of conjecture. The Arunta, and other Central Australian tribes, look back to a mythic past, when ancestors, closely connected with this or that plant or animal, perhaps transformations of such animal or plant, roamed the country in groups, each of the same totem name, each feeding freely on its own totem.

This was 'the Alcheringa time,' and existing rites are explained by 'ætiological myths,'stating how such or such a mummery, still practised, was originally practised in the Alcheringa. Nothing of the sort, of course, need have been the case, and such myths cannot tell us what the manners and customs of that dim age really were. Demeter was a woman of the Greek Alcheringa, and the Eleusinian rites were explained by the Greeks as originating in her Alcheringa adventures. But these obviously were invented purely to account for the rites themselves, not vice versâ.

Now, among the Arunta the blacks of to-day are regarded as reincarnations of the Alcheringa fabulous ancestors. Each of these carried about (both men and women) churinga, the portable decorated stones. When an Alcheringite died, a rock or tree rose to mark the place, but his or her spirit 'remained in the churinga.' Plenty of churinga were dropped at different sites, and round these now hover the spirits associated with them. In one place is a crowd of wild cat ghosts; at another, a mob of frog or lizard or emu ghosts. These want to be reincarnated. Consequently, a woman who desires to have a baby goes to one of them (in Argyll she would slide down a cup-marked rock!), a woman who does not want to have a baby keeps away. A child's totem is derived, not from father or mother, but from the totem of the ghosts at the place where the woman thinks she conceived it. When the baby is born her relations hunt the spot, and find for it the churinga left by the spirit which is reincarnated in it.

Thus, first there is the fabulous Alcheringite, himself a transformation of an emu, lizard, water, fire, or what not. Then there is his spirit haunting, after his death, a spot where churinga of his totem were deposited. That spirit enters into and is born again from a passing woman, and the spirit's churinga is found and is henceforth the child's churinga—an oval plate of stone, with cup and ring or other decorations.[12] All these churinga are kept at sacred central stores, caves, or crevices. Each member of the tribe is represented by her or his 'churinga nanja' in these repositories. Women may not go near these sacred stores, nor may they see a churinga.[13] If they do, their eyes are burned out with a fire stick. A man's churinga is not, to him, like the egg in which was the life of the giant in the fairy tale. If it comes to grief, he does not die, but expects bad luck, as we do if we break a mirror. Not till he has been through the mysteries and the most cruel mutilations, and just before he has been painted with the pattern on the sacred rock of his totem, can a man see the store-houses of the churinga. Now, in the witchetty grub totem this sacred painting tallies with the lines incised, under concentric circles, on the covering of a stone kist at Tillicoutry.[14] There are circles above the lines in the Australian example, or rather circular dabs of paint, called 'the decorated eyes,' painted on the rocks; the corresponding patterns are incised on the portable churinga. In Scotland the patterns are incised both on fixed rocks and portable stones; the latter at Dumbuck and Dunbuie.

I observe many patterns common to both regions. There are the concentric circles, the spiral, the marks like horseshoes, the tree pattern, the witchetty grub pattern, the volute, the long sinuous snake-like pattern, and a number of these recur in Brazil, on the banks of the Rio Negro.[15] Now, though we have those patterns on rocks in Ohio, Brazil, Australia, in this country, in France, in Asia Minor, I only know the patterns on portable small stones in Australia, at Dunbuie, on the Dumbuck site, and, I think, in a cairn near Lough Crew, in County Meath. The curious, for this last case, may consult 'Proceedings of Scottish Society of Antiquaries,' 1893, p. 299, where in figs. 6 and 7 he will see what in Australia would be called two stone churinga, with any number of Scoto-Australian patterns on large stones. On one the pattern is like that of a stone from Dunbuie.

In Australia members of each totem decipher the marks, purely conventional, as representative of the totem, and of adventures in the Alcheringa time. For example, a mark like two croquet hoops, or horseshoes, is 'an old woman gathering frogs.' The concentric circles are frogs; the dots round them are tracks of women; dull, often dirty, stories are told about the adventures of the Alcheringites commemorated by the patterns. At the sacred pattern-painted rocks, magic ceremonies, extremely puerile, are performed to ensure a supply of the edible totem which the pattern represents. Some event occurred there in the Alcheringa; the rite repeats what, in myth, was then done, and the stomachs of the men are rubbed with the churinga 'for luck.' Such are the uses of the churinga. Did they once exist wherever the similarly decorated fixed rocks exist? Did the makers of the decorations in Scotland decipher the churinga as the Central Australians do now? Were the dwellers by Clyde (much more advanced in culture than the Australians) totemists, looking on their small decorated stones as associated with the spirits of Alcheringa ancestors? Do women in Argyll slide down a cup-marked rock, in hope of offspring, because totemistic ghosts once hovered round it, eager to be reincarnated? The fact of the sliding is attested by a chief of Clan Diarmid.

Nobody can answer! I have shown these decorated rocks and small stones to have a living significance, a vital legendary symbolism, in Central Australia. I cannot prove that they had the same significance in County Meath or Dumbartonshire. The Australians may have begun with mere decoration, and later added a symbolism suited to their amazing theory of life. In our country the decorations may have quite a different symbolical sense, but probably they had some sense. Otherwise, why engrave them, not only on rocks, but on small stones pierced for suspension? Perhaps men believed in an Alcheringa time on the Clyde; perhaps they multiplied salmon and deer by magical mummeries at the engraved rocks; perhaps these were sacred places, tabooed to women. Or quite a different set of fables and customs may have crystallised in Scotland round marked rocks and inscribed small stones. I cannot prove that, as in Australia, Clydesdale boys of old, when initiated in the mysteries, were painted with the pattern on their sacred totem rock and stone or wood churinga. But, if not these rites, other rites were, I conceive, connected with the decorative patterns found in so many still savage countries.

One piece of evidence rather points in this direction. The Australian stone churinga are shaped like the wooden churinga, and these are shaped like the tundun, or 'bull-roarer.' Now the bull-roarer (which occurs in Australia where stone churinga do not) is a sacred oval piece of wood, not to be seen by women, which is whirled at the mysteries, and makes a windy, roaring noise. The same object is used, for the same purpose, at the mysteries in America, Africa, and, of old, in Greece.[16] The roaring noise is taken to be the voice of Tundun, son of Mungan-ngaur, 'Our Father' in the heavens, among the Kurnai, and of gods or culture heroes of other names in other tribes. Now, in Celtic Scotland (as also in England) this instrument, the tundun, occurs as a mere toy, in Gaelic named strannam. Does it descend from a sacred object of savage mysteries, and are the Australian stone churinga—in shape like the tundun, and like the tundun tabooed to women—mere lapidary modifications of the wooden tundun? However this may be, the strannam looks like a link in the long chain which binds us to the prehistoric past.