Dr. E. B. TYLOR remarked that Mr. Frazer's original and ingenious treatment of the evidence must materially advance the study of animistic funeral customs. His theory of the connection of purification by water or fire with attempts to bar the return of ghosts deserved, and would doubtless receive, the careful consideration of anthropologists. Dr. Tylor adduced from Mr. Yarrow's paper on Mortuary Customs a case of water burial carried out for the purpose of preventing the return of harmful ghosts. With regard to the entrance of the person supposed dead by the roof, he called attention to the fact that such entrance is adopted in some districts as a symbolic rite, perhaps indicating descent from heaven, which might possibly be the explanation of the Roman practice. Dr. Tylor concluded by expressing his satisfaction at the excellent results of Mr. Frazer's study of classical authors, not as mere ancient texts, but as repertories of real facts full of anthropological value.

Mr. F. T. HALL suggested that the idea of water as a barrier between the dead and the living might have originated with the primitive and indeed general belief that the souls of the departed are not at rest until they have passed to the other side of some great water, now referred to as "the river of death." The Chaldeans made their dead cross a mysterious sea, the Egyptian dead navigated across the infernal Nile; the Greeks and Romans had their Styx, over which the soul could not be ferried until proper funeral rites had been performed with the body, the unburied wandering on this side of these waters for twelve months before being allowed to cross. Even the waters of the firmament were considered to be interposed between earth and heaven. The general idea was that the earth, the abode of the living, was encompassed by water over which the dead souls had to pass before they reached the place of rest, and that until water was interposed between the dead and the living the soul could not be at rest and was apt to wander through the earth.

Mr. BEAUFORT observed that there was at all events one modern nation where water was not supposed to restrict the movements of ghosts, namely, Japan. On the evening that the speaker entered Nagasaki the Japanese were celebrating the annual return of the dead to visit the living. All the tombs were lighted by pretty coloured lanterns, and food was placed there for the use of the spirits. On the third day hundreds of miniature vessels were sent to sea freighted with food for the spirits on their return voyage. Thus the spirits make two voyages every year.

Mr. HYDE CLARKE said that in the consideration of the re-entry to the house it must be taken into account that in the Persian example, as in many others, the house would be terraced on the top with an approach from below. In most cases the houses are isolated, and as there is no exit elsewhere from the terrace it is naturally suggestive as an entry for the ghost. With regard to not mentioning the name of the dead, it must be borne in mind, there is equal superstition as to mentioning the name of the living, as of a husband. So also the sacred name of a city. The name is the spiritual essence

of the ghost and the Ka. A character for name is the round or circle, and this is perhaps the origin of the cartouche encircling names in hieroglyphics, &c. He might mention one legend as to the connection of the dead and the living in Slav countries, which he had learned from a Servian friend, in whose family an example had happened, and which he believed was included in the MSS. of the folk-lore of Servia prepared for the press by Madame Mijatovich. There is a superstition of a mysterious connection between those members of a family born in the same month, who are denoted in Slav as "Same month," and of whom of course there are many examples, as we may observe that even in a family of six the births will be severally in three or four months, and not in separate months for each. On a child dying there was great fear for the sister of the "same month," and it was considered necessary to preserve her from the danger or certainty of a similar premature death. A hobble was got in with which horses of the herd are hobbled on the plain, and the living was hobbled by the leg to the dead. An exorcist then repeated the necessary formula, and to him was handed a piece of silver money (about a shilling) which had been given or begged. The child lived, which is a testimony, and of course a confirmation, of the efficacy of the process.

Mr. FRAZER, in reply, expressed his deep gratification at the interest which Mr. Tylor had expressed in his paper. It was the writings of Mr. Tylor which had first interested him in anthropology, and the perusal of them had marked an epoch in his life. He fully agreed with an observation of the President, that it would be hazardous to assume that when in modern times a man dresses very carefully on such momentous occasions as going into battle (as General Skobeleff used to do), we had here a relic of the old feeling which prompted people to dress a dying man in his best clothes. On the other hand, he was inclined to think that in the modern reluctance to mention the name of a person recently deceased we had a relic (of course quite unconscious) of the old belief that a dead man will hear and answer to his name; there was a large substratum of savagery underlying all our civilisation. Replying to Mr. Tylor he said that he (Mr. Tylor) had laid his finger on the apparent inconsistency of the facts that ghosts could bathe in water, yet not cross it; but the author pointed out that men were exactly in the same predicament-that, in fact, in dealing with primitive ghosts we always had to regard them as being as nearly as possible the exact counterpart (only invisible) of men, and hence that though ghosts had the same difficulty which men had in crossing water, yet the difficulty was not insuperable for ghosts any more than for men. Thus Mr. Beaufort had informed them that Japanese ghosts could cross water in boats, and the author referred to the well-known story of King Gunthram, whose soul was seen to depart from him in sleep and to seek in vain to cross a stream till some one laid a sword across it, on which the soul immediately crossed over to the other side. With regard to the interesting Slavonic superstition mentioned by Mr. Hyde Clarke, that a child born in the same month with a child that had died was especially

likely to die, and that special precautions had to be taken to save it, the author suggested that we might get some light by comparing the Laosian beliefs with regard to children. The Laosians think

that an infant is the child, not of its parents, but of the demons; and hence they call on the demons to carry off their child within four and twenty hours after birth or else to leave it for ever. Moreover, they give the child a hideous name by way of frightening away the demon, and they sell it for a nominal price to a friend, under the impression that the demons are too honest to carry off what has been actually bought and paid for. Now if the demons had carried off a child born in a particular month, it might be thought that this gave them a special power over another child born in the same month, and that therefore special precautions were needed to prevent its dying. One of the speakers had suggested that in Persia the supposed dead man might have returned through a door in a terraced roof. In reply, Mr. Frazer said that there was evidence to show that in the case in question the entrance was made through the compluvium, an opening in the atrium or principal apartment of the house. Now as this atrium was distinctly stated by the ancients to have been originally sitting-room and kitchen in one, it is not unreasonable to infer that it represented the single apartment of the primitive house, and that the aperture in the roof (afterwards known as the complurium) was originally the smoke-hole or chimney.

The following paper was then read by the Director:

The SCULPTURED DOLMENS of the MORBIHAN, BRITTANY.
By Rear-Admiral F. S. TREMLETT, F.R.G.S.

[WITH PLATES III AND IV.]

THE tumuli of Brittany having been so frequently described, it will be unnecessary to give in this paper a lengthened description of them, or of their contents. Several of the most interesting monuments have disappeared, but there still remain a considerable number which will probably be untouched, from the fact that below the thin coat of humus the granite rock is found, and that stone is so easily procurable that the farmers find it more economical to establish a quarry on their fields than to blast with powder the megaliths on their land. To this circumstance may be attributed the preservation of those that remain. There is, however, an exception to this, and that is when the Church requires building stone; the parishioners are then exhorted as a religious duty to obtain and bring to the church whatever material is required, which service is invariably cheerfully rendered and gratis. To this circumstance may be attributed the gap which