Celtic places of worship: these are temple and grove (temenos), or consecrated allotment. The Celts buried their dead like the other Aryan nations; they burnt the bodies, like the Greeks of Homeric times, and built the tumulus over the ashes, though it is just as likely that the common people buried by inhumation as in Greece and Rome. The tumuli, also, were erected only over chiefs and great men, as among the pre-historic and contemporary tribes. The Homeric burial existed, according to Roman and Greek writers, in Gaul. Homer represents Achilles as placing the fat of many sheep and oxen, whose carcases were heaped round the pyre, about the body of Patroclus, from head to foot. He set vessels with honey and oil slanting towards the bier, and then threw horses, pet dogs, and captive Trojans, after slaying them, on the pile, to be burnt along with the body of his friend. Celtic burial tumuli are not easy to identify. Burials of the pre-Roman and pre-Christian period of Celtic occupation are very few indeed, and in archæological works are continually confused as "AngloSaxon," being, indeed, of a similar type. Like their houses and temples, they were of no lasting character. One thing is clear: they had no rude stone circles around them.

The Celts made use of iron ever since they appear in history. It is not likely on a priori grounds that they would build rude stone circles for worship or even for burial, nor can it be understood from their religious beliefs what use they could make of circles of rude stone. The Druids again were merely the Celtic priests a priesthood of more than ordinary influence and power, but their doctrines contained little else than was believed in then by other Aryan races in Europe or Asia. It is sheer improbability that they could have worshiped in stone circles,

If the Celts did not build these rude stone monuments, then some race previous to them, and in a more barbaric state of culture, must have been the builders. Various facts go to show that there existed previous to the Celts another people or rather other peoples. Professor Rhys has proved from the evidence of language and mythology that there was a previous race; while Mr Elton, founding on a study of customs and on the researches of archeologists, has still further proved the fact. Following Canon Greenwell and Dr Thurnam, and extending the significance of their conclusions, Mr Elton is able to prove that two races at least existed previous to the Celtic race. There was,

first, the small dark-skinned, long-headed race of the Neolithic and later cave age, whom Mr Elton calls Iberians, whose descendants survived in Siluria of Wales, in Ireland, and in Aquitania, and

who spoke a language probably like the Basque. They were the builders of the oval barrows. The second race was tall, roughfeatured, strong-limbed, round-headed, and fair-haired, and Mr Elton calls them Finnish or Ugrian. They appear to have been in their Bronze Age, whereas the Iberians were in their Stone Age. The Finnish race may have had an alphabet, if we can attribute to them the numerous unreadable inscriptions-rock-carvings and sketchings of the Bronze Age-which appear in Ireland, Scotland, and Scandinavia. They appear to have subdued the previous race. They built the round barrows, and we have every reason to believe that they were also the builders of the rude stone circles, their crowning effort being the temple (?) of Stonehenge. We have besides their burial customs, glimpses possibly of their social condition. Cæsar and other writers continually and persistently refer to races in Britain who had community of wives, and there can be no question that there was some foundation for the rumour. Nor can we have much doubt that the nation referred to was this Finnish one, for it is quite certain that it was not a Celtic or Aryan nation, among whom monagamy was the strict rule. The Pictish custom of succession through the female also establishes among them low ideas of marriage, quite consonant with community of wives; and from this we must conclude that the Picts were strongly intermixed with, if not altogether, a nonAryan race. The nakedness and blue paint of historians is another feature which, as knowledge of the races of Britain advanced, the classical writers learned to locate among the inhabitants of Northern Scotland.

The long barrows were built by a race anterior to this Finnish race; the Finnish race built the round barrows, chambered cairns, and rude stone circles. They were probably also the builders of the brochs. The theory that brochs are of Norse origin arises from ethnological confusion; for the Norse were Aryans possessing iron implements, and builders, like the Germans and Celts, of wooden and not of stone buildings. The Picts were Finnish. Mr Joseph Anderson says that stone circles attain their principal development in Pictland proper, and are most abundant in the district between the Moray Firth and the Firth of Tay. "Those of the Scottish circles that have been examined," he says, "have yielded interments of the Bronze Age." This better development of the circles in Pictland goes to prove that the circle builders lasted longer in Pictland than anywhere else, and, in fact, that the builders were the ancestors of the historic Picts, and possibly the historic Picts themselves. Mr Fergusson,

from quite another standpoint, suggests that Clava is the burial place of the royal family of Brude Mac Maelchon, the King whom Columba visited on the banks of the Ness. The suggestion is not at all a bad one.

We shall now draw the threads of our argument together. In our process of elimination we discovered that the Aryan races built no stone circles; the Celts, therefore, and their Druids, had nothing to do with them; they are pre-Celtic as well as pre-historic ; the circles are so often connected with burial that we may take it for granted that they all originally had to do with burial; but we found, also, that in modern times, circles and stones were connected with worship, more especially the worship of ancestors. Everything points to their having been places of burial; they surround dolmens and barrows, or even when alone yield at times burial deposits; but their peculiar character, coupled with the fact of modern and ancient worship of and at stones, must make us pause ere we set burial down as their sole purpose. Would savage or barbaric man, out of mere reverence for the dead, raise such monuments to their memory? It remains to consider what kind of worship could be held at places of sepulture, and why stone circles should be used. In the matter of worship, the old theories as to the sun and serpent worship may be dismissed as out of place in connection with burial, for the worship of the sun as the giver of light and heat has never had connection with death: Apollo must not be contaminated with death. The evident reference many of the circles have to the sun's course, as for instance that in this district the heaviest stones are to the south, or that the entrance may have a southerly aspect, only proves that the light and heat of the sun were regarded as necessary for the dead as well. That the existence of solar worship would add to and emphasise the "sunward" tendency-the sun reference of the circles-is freely allowed, but nothing more can be legitimately deduced from such a fact. As for serpent worship, it is plain that the advocates of that theory did not quite grasp the full significance of the serpent cultus and its connection with phallic worship. The only worship appropriate at the grave is that of deceased ancestors. A study of the beliefs and customs of savage and barbaric races makes it abundantly clear that this is about the earliest shape in which religion manifests itself. We must inquire what the anthopologist has to say in regard to people in this state of culture. Reverence for the dead and belief in the existence of the Deity are glibly asserted by theorists as existent among every race, but that is a delusion. Reverence and belief in the sense understood by a

civilised and educated person there are none, for savage belief is eminently practical and unsentimental. To project the highest feelings and opinions of civilised man--and these local, too—into the early state of man, is to overlook the long perspective of time with its evolution of ever higher feelings and beliefs. The lowest phase of belief has been named by Mr Tylor, "animism": it consists in believing that what is presented to us in our dreams and other hallucinations has a real objective existence. Savage man makes little or no difference between his dreaming and waking state. He sees the "shadows" of the dead in his sleep, and believes in their objective reality. But not merely the dead have shadows or spirits; the living, too, have a spirit duplicate of self. The reflection in water proves this no less than the presentiment of the living man in dreams. Hence it is that the savage dislikes the photographer. Animals and material objects, of course, have souls, on the same grounds, for the dead hero appears in dreams with ghost of hatchet, sword, and spear. "The Zulu will say that at death a man's shadow departs and becomes an ancestral ghost, and the widow will relate how her husband has come to her in her sleep, and threatened to kill her for not taking care of his children; or the son will describe how his father's ghost stood before him in a dream, and the souls of the two, the living and the dead, went off together to visit some far-off kraal of their people." The funeral sacrifice of historic nations, of early Greeks, Romans, and Celts, show how barbaric religion includes the souls of men, animals, and material objects; for what was useful to the dead when alive was burnt or buried along with them-chariots, arms, horses, dogs, and even wives and slaves were sacrificed in one mighty holocaust. The religious creed in which "animism" embodies itself is, of course, the worship of the dead, especially the worship of ancestors. Worship and reverence, here, have a different sense from our ideas of them. The dead are worshipped for protection, and repaid with reverence, not merely in feeling, but also in practical gifts and sacrifices at their tombs. It may quite as often happen that their wrath is deprecated. From the mere family ancestor, the worship may rise to that of great chiefs and kings that are departed, and from that it may rise to a conception of a supreme father-"The old old one of the Zulus, as they work back from ancestor to ancestor, thus arriving at an idea of a creator, akin to the conception of the "Ancient of Days." One's own ancestor may be good to one; other people's ancestors may be the reverse. Hence these last have to be propitiated; evil spirits are worshipped to avoid their wrath. Thus the ghost

of a British officer was not long ago worshipped in India as a god, and on his altar his demon-worshipping votaries placed what they thought would please and appease him, for it had pleased him in this life, namely, offerings of cheroots and brandy! In fact, all the ills that life is heir to are among some races attributed to evil spirits, while the good is the work of the beneficent spirits; and among such tribes it is through the medicine-man, with his exorcisims, there is the only means of escape. Let it be noted that ancestral ghosts may not merely exist in proper human form, but they often assume animal forms, and what is more, they may even take up their abode in material objects-trees, stones, or anything. Hence arises "fetish" worship-the worship of "stocks and stones." And it is also easy to see that we may, on the other hand, rise from ancestor worship, through this transmigration idea, to the height of polytheism, with its gods of sun, moon, and sky.

This reverence of the savage for the dead is therefore con nected with his regard for himself. His religion, as usually happens in higher phases of culture, is selfish. The dead are therefore cared for and their abodes become places of worship. Various ways are adopted for disposal and worship of the dead. The hut they lived in may be left as a dwelling for them; the body may be buried in a canoe or coffin; a strong tomb may be built over it or its ashes, and this tomb may be a chamber with access to it to enable the votaries to bring offerings. Great labour was bestowed on these burial mounds of earth and stone. Nor have we yet ceased from this display, though we now have different methods and far different feelings in our burial rites. Yet there are survivals of ancient forms. "In the Highlands of Scotland," says Mr Tylor, "the memory of the old custom [raising of mounds and cairns] is so strong that the mourners, as they may not build the cairn over the grave in the churchyard, will sometimes set up a little one where the funeral procession halts on the way." Our memorial stones over the graves are but the descendants of the old menhirs; nor are dolmen forms absent in the stone box structures often placed over graves. In the Churchyard of Rothiemurchus, on the grave of Shaw Cor-fhiaclach, the hero of the North Inch at Perth, there used to be a row of small pillared stones set round all the sides of the tombstone. Circles of stone other than such far-off imitations as this we do not use now.

Burial and worship in early society go hand in hand, and we, therefore, conclude that these stone circles were used for both burial and worship, but more especially for worship, since mere