66 leigh ("Phil. Mag." [5], vii, p. 153) would be particularly suitable. The method depends upon the behaviour of a sensitive flame," when a high note is sounded. The coal-gas flame issuing from a pin-hole burner becomes "sensitive" if the gas is supplied at a sufficiently high pressure (say 10 inches of water, the ordinary gas supply pressure being about 1 inch), so that a flame about 18 inches long is obtained just on the point of flaring. A high note, such as those caused by rattling a bunch of keys, or a hiss, makes the flame flare, and it continues to flare as long as the note sounds, recovering its steadiness when the sound ceases. If a continuous high note is made by blowing a whistle or “birdcall" by means of a weighted gas bag, and at some distance from the whistle a plane vertical surface is erected, so that the waves of sound are reflected normally, the interference of the reflected waves with the subsequent incident waves produces a series of nodes with intervening loops in the space between the whistle and the wall. Lord Rayleigh has shewn that if the sensitive flame is moved into various positions in this space, the whistle will flare everywhere except at the nodes, and there the flame will be, comparatively speaking, undisturbed. The position of the nodes can therefore be identified by means of the sensitive flame, and, as the distance between consecutive nodes is half a wave-length, the wave-length of the note in free air is easily deduced from observation of the internodal distance; if λ be the wave-length so obtained, N the vibration frequency of the note, and v the velocity of sound, which may be quite safely assumed as known, N = , and no correction is required. λ A number of observations were taken by Mr. Turner on this plan with three whistles, denoted A, B, and C respectively, the gas for the sensitive flame was supplied from a gas-holder that could be loaded at pleasure, and a gas-bag furnished air to the whistles. A plate of glass served as a reflecting wall. The pressures of the air and gas were measured by means of U tubes containing water. This method proved to be applicable to notes with a pipelength between 3 mm. and 7·6 mm. The practical reasons against its use for other notes are as follows:-A particular flame is not sensitive for all high notes within the range of a whistle at the same time, but the gas pressure can always be adjusted and a flame obtained which flares when a given note is blown on the whistle, within the range of 9 mm. pipe length to half mm. or less, far beyond the limit of audibility. My own ears cannot appreciate a note with a pipe-length less than 3.7 mm. Mr. Under favourable conditions the behaviour of the sensitive flame when the whistle is blown as in actual practice, by squeezing the india-rubber bladder, is very striking. As the length of the pipe is gradually shortened, the flame gives a short flare for each puff of air until the piston is pushed quite home, when no Яlare occurs. It would, however, not be safe to make any inference as to the pitch of the note for lengths less than half a millimetre. Turner's limit is 38, so that the flame responds to continuous sounds at least three octaves above the highest sounds commonly audible. But it is not an easy matter to get the reflexion nodes well marked as points of minimum flaring; the pressure of the gas requires careful adjustment in any case, and we were unable with any adjustment to get nodes for pipe lengths greater than 7.6 mm. or less than 3 mm. This range embraces notes from 8,000 to 21,000 complete vibrations per second. I cannot safely assign any reason for our inability to get nodes for higher pitches; it may be due to the comparatively large area of the section of the flame at its sensitive part, or to the continuous effect of disturbing vibrations which, being inaudible, cannot otherwise be perceived; but even as the case stands we have measured the nodal distances of notes a major third higher than anything that either of us could hear, and it is possible that with other flames a higher limit may be reached. For wave-lengths longer than 34 mm. the flame had to be made so sensitive that it was unstable and no satisfactory observations could be made. In each determination of an internodal distance the positions of a large number of points of minimum flaring at different distances from the reflecting wall were read many times over and a mean result deduced. The following specimen will exemplify the agreement between the observations, and show how the mean value is deduced. WHISTLE A. Length of whistle 7·1 mm. Pressure of air blowing the whistle = 26.4 cm. of water. This may In this and some other cases the length calculated from the first node is far greater than that from any of the others. be due to the heating effect of the flame, which cannot well be allowed for. We thought better therefore to leave out the first position in all cases. The mean result of this series then will beHalf wave-length=16.63 mm. All the mean results obtained are given in the following table : It appears from these results (1) that the different whistles give practically the same note at the same pressure; (2) That the note sounded varies considerably with the pressure, the values at 5 mm. differing by about a "whole tone" for a variation of pressure between 19.3 and 32.8 cm. of water; (3) That the true wave-length is greater than four times the length of the pipe. It is difficult to suggest a general law which will meet the case, in consequence of the variation of the note with the pressure. The pressure of the air in actual practice with the whistle is obtained by squeezing the little india-rubber bladder attached to the whistle. The pressure is variable, but the ear recognises only one note. This may be connected with a fact that we observed in connection with them, namely, that the whistles required a certain definite pressure, different for each, in order to produce a clear note, if the pressure was not correct the sound produced might be called a hiss, yet it gave good nodes. It appears that, speaking very roughly, the correction to be applied to the observed length of pipe in order to obtain the the true quarter wave-length for the higher pressures does not differ much from 1 mm., and is therefore nearly equal to 1.5 × diameter of pipe, the correction found by Wertheim to be applicable in the case of organ pipes of corresponding shape. NOTE on the DIEYERIE TRIBE of SOUTH AUSTRALIA. By Mr. SAMUEL GASON. (Communicated by J. G. FRAZER, M.A.) Mr. Frazer writes as follows: I enclose a copy of a letter received by me from Mr. Samuel Gason in reply to some enquiries which I had addressed to him concerning the Dieyerie tribe of aborigines, South Australia. Mr. Gason, in the course of his duties as police trooper, has been for many years familiar with the tribe in question, whose manners and customs he has described in a very valuable little work, included in the volume, "Native Tribes of South Australia." The following letter supplements on some important points the information contained in that work. In particular it shows that the Dieyerie belongs to that rare class of cases, intermediate between mother kin and father kin, where the sons take their totem from the father and the daughters from the mother. This is not, as I hope to point out elsewhere, to be confounded with the sex totem, of which examples are to be found in Australia, but (so far as I know) nowhere else. In view of Mr. Gason's letter the statement of Mr. Howitt ("Journ. Anthrop. Inst.," XIII, p. 457) that descent in the Dieyerie tribe is uterine, needs correction. The following is Mr. Gason's letter, dated from Beltana, South Australia, March 6th, 1887 : I beg to acknowledge the receipt of your letter, and to send the following remarks in reply to your inquiries, re branches of the aborigine Dieyerie tribe of South Australia. 1st. As to whether children of the father inherit the father's branch or class name, I reply yes, the sons take the father's class, the daughters the mother's class, e.g., if a dog (being the man) marries a rat (being the woman) the sons of the issue would be dogs, the daughters of the issue would be rats. 2ndly. As to whether the father is the head of the family, I say most certainly. 3rdly. As to whether the father eats of his children at the burial ceremony, my reply is that the father does not eat of his offspring; the reason assigned is that, being the head of the family, he has sufficient command, and being a man (not weak like a woman) he can resist the deep grief occasioned by the loss of his child, and not be perpetually crying, causing a nuisance to the camp and tribe ; whereas, a mother and other female relatives are compelled to eat of their offspring and dear departed relatives, for by so doing, they are supposed to have a presence of their departed in their liver (they feel from their liver, not from their heart as we do). The man will eat of his brother, his uncle, his cousin, or dear friend, but not of his father, nor his grandfather, nor his offspring. 4thly. The members of each class-name do not pay any particular respect to their branch, further than each class thinks that they are of the oldest families. They eat the animals or plants of which they derive their class names. 5thly. On all deaths, either from natural causes or otherwise, an inquiry or inquest is held immediately before burial, and in case of the departed being a person of note or influence, the result of the inquiry is a verdict of murder against some person or persons of the same tribe or of the neighbouring tribe, even if the deceased died from natural causes, they having a superstitious belief that any man who is a Koonkie (doctor) has the power to cause any person's death by sickness at any distance by the use of a human bone, carried out by a superstitious charm. FOURTH ANNUAL REPORT OF THE BUREAU OF ETHNOLOGY-- THE Fourth Annual Report of the Bureau of Ethnology is not quite as large as its predecessors, neither is it, perhaps, quite as full of interest, although the papers are of high scientific value, from the systematic manner in which the work of exploration has been carried out, and the large amount of material collected and classified. The recorded field-work consists of explorations among the mounds of the Mississippi Valley, in the course of which four thousand one hundred specimens were added to the National Museum, including a large collection of pottery, skulls, stone, copper, and sheli implements and ornaments, also some articles shewing contact with Europeans, such as hammered iron, bracelets, brooches, and crosses of silver, and a hog's tooth. The full report of these explorations will be of great utility to future workers in this extensive field; but the explorations carried on by Mr. James Stevenson, among the cliff dwellings in the cañons of North Mexico, are of still greater interest, for these cliff dwellings are an extraordinary development of the primitive cave dwellings of early man, consisting of huge chambers tunnelled into the solid rock, in some cases 300 feet above the bed of the cañon. Some of these village chambers are of enormous dimensions, one being described as 1,500 feet from side to side, and about half that space from the back to the edge of the cliff; calculated to have been the home of between a thousand and fifteen hundred persons. The floor of this cave, probably a natural cavity enlarged, was studded with dwellings built of square stones laid in mortar, and houses three stories high are found, filling up spaces between rocky projections, whilst frequently houses have been built jutting out from the cliff, reminding one of swallows' nests on a large scale. The most singular part of these extraordinary dwellings is, that they appear to have been constructed entirely by users of stone implements; but who, judging from their works, must have attained to a civilisation superior to that of the neolithic peoples of Europe, for they had much artistic skill, the dwellings being painted in various colours, generally in bands, but also in a pattern resembling the Greek fret, and with many curious designs of unknown meaning introduced. They seem, also, to have cultivated Indian corn, and to have made garments and matting from the fibre of the Yucca; they also wore finely woven sandals of peculiar pattern, and skeletons have been found, buried in a sitting posture, with the flesh and skin dried to the hardness of stone. Mr. Frank Cushing continues his interesting researches among the Zuñis, and Mr. Victor Mindeleff is prosecuting the same work among the Moki, and we may hope soon to see the result of their investigations. Good work has also been done in classifying the languages of |