Volume 54 1945 > Volume 54, No. 1 > The acoustics of three Maori flutes, by Ernest S. Dodge, p 39-61
                                                                                             Previous | Next   

- 39

MUSEUM labels of primitive musical instruments and their catalogue descriptions are not always happily expressed. Any bamboo stem one or two feet long with holes near one end may get called a nose flute though it is equally well blown with the mouth. Three or four inches of bored out whale's tooth also exhibits itself as a nose flute, though it looks more like a mouth whistle and no responsible observer has ever seen anything like it actually sounded through anybody's nose. What are obviously gourd whistles figure as “calabash trumpets”, though all their mouth parts are too small to be trumpet-blown by any creature more than a foot high. The whole wind group, in short, tends to be discussed without much regard to any acoustic principles or much concern for the way in which any particular pipe can actually be blown and the notes which it will actually yield.

It has, therefore, seemed to the authors of this paper to be worth while to start all over from the beginning and restudy both the acoustics and the operation of three primitive Maori wind instruments of different types and known history from the collection of the Peabody Museum of Salem, Massachusetts. One of us, as a museum curator, has the run of specimens and books. The other, in addition to teaching physics, has from boyhood played various sorts of modern flute. Our special thanks are due to Mr. and Mrs. Joseph N. Ashton of Andover, Massachusetts, the one an organist, the other a fiddler, for invaluable help in correcting our own amateurish judgments of pitch and quality.


In general, any wind instrument, sounding a note, has its tube occupied by a standing wave. If the tube is closed at one end, this standing wave takes off from the end wall. In this region, therefore, there forms a node, where the air is at rest. Away from this node, the air particles swing more and more widely, to reach their maximum to-and-fro motion at the open end of the tube. At this antinode, therefore, starts the moving outside wave that brings the sound to the ear.

But a pipe, open at both ends, can pretty obviously not form a node at either. Instead, the node forms at the middle of the pipe, with an antinode at each end. Therefore does an open pipe sound one octave higher than a stopped pipe of the same length.

One must, however, always bear in mind that a pipe may be open at both ends and still be acoustically “stopped”. A clarinet in particular, though it ends in a wide bell, behaves in all respects like a - 40 stopped organ pipe. But a saxophone, with a similar mouth-piece and a not dissimilar bell, is an “open” pipe. Some confusion has arisen concerning primitive instruments from the fact that a pipe, structurally open, may be functionally stopped.

For either sort, shortening the tube raises its pitch, a full octave if the length is halved, a fifth if one third is taken off, with other intervals in proportion. It is convenient to have in mind, as a standard from which to reckon, that a modern concert flute, sixty centimeters or two feet from blow-hole to distal end sounds middle c, and that shortening a pipe by a ninth raises its pitch by one tone.

Unfortunately, the general rule that the pitch of a pipe depends on its length is strictly true only for ideal tubes whose diameters are negligible. In all actual instruments, the standing wave, so to say, projects somewhat from the tube end, the more so as the bore is enlarged or the distal end flares into a bell. The result is that for a stopped pipe the functional length which determines the pitch is close to the measured length plus one and one-third diameters, and for an open pipe the measured length plus one and two-thirds diameters. (For this and related topics see among other authorities Culver, especially p. 115ff.) This correction varies somewhat with pitch and does not hold for short wide pipes, which may sound, stopped, not one octave but nearer three lower than slender open pipes of the same length. An ocarina for example, with a length three times its diameter, sounding middle c, is only eight inches over all. Moreover, since the diameter correction is slightly larger for open pipes than for stopped, the same pipe, blown both stopped and open, gives its open note a little flat.

We may, then, following Sachs (p. 166-8) distinguish, somewhat vaguely, between tubular and globular pipes. The former have lengths ten or twenty times their diameters and conform in all respects to the familiar laws of both open and stopped pipes. The latter run from strict globularity to, let us say, lengths of three or four times their diameters. Examples are ocarinas, many high-pitched whistles, and bottles blown across their mouths. All such are more commonly blown stopped and always blow more easily thus. What little is known of their acoustics has been set forth by Wead (pp. 428-33). In general, if a globular flute is really globular, its pitch depends on its volume and the area of its blow-hole, added vents sharpening the notes according to their size, without much regard to their location. A short wide pipe more or less compromises between the two types.

All pipes, as they warm up with the player's breath, sharpen their pitch, the change sometimes amounting to a quarter-tone. Most, in addition, can have their pitch more or less “nursed” by way of the embouchure. This also amounts, in general, to something like a quarter-tone, but with certain short and especially wide pipes may be a good deal more. Both these properties of wind instruments introduce uncertainties into determinations of pitch.

An open pipe, overblown, forms two nodes, each a quarter of the way in from the nearest open end, with an antinode at each open end and another at the middle, where the single node was when the pipe sounded its fundamental. The standing wave is, therefore, half its former length, the pitch is one octave higher, and the pipe behaves in all respects as if it were halved and the two parts blown separately. A stopped pipe, on the other hand, does not when overblown divide its aircolumn into quarters, but into thirds. Its lowest overblown note - 41 is, therefore, not an octave but a twelfth—that is to say, an octave and a fifth—above its fundamental. In general, all these notes are more easily blown on stopped than on open pipes.

The standing wave inside a wind instrument which initiates the travelling wave outside, is itself initiated in three different ways. In the “brass winds” the player's lips, held somewhat rigid and pressed against the mouth-piece, are vibrated by the player's breath and pass on this vibration to the air inside the tube. Most commonly such trumpets and horns are made long, ten or even twenty feet, and do not in general use their fundamentals, though short primitive conch shells and animal horns more commonly do.

A second type of wind instrument initiates its standing wave by means of one or two vibrating reeds. These also are set in motion by the player's breath, and they may by their own vibration rate somewhat modify the normal pitch of the pipe beyond. Such reeds may vibrate freely, often in a slot, as in the harmonica, or they may “beat”, either against one another, as in the oboe, or against a slotted plate, as in the clarinet. The human voice is commonly described as a “reed instrument”. Actually it is of the lip-sounded or trumpet type, the vocal cords being internal lips and the mouth and nose cavities the horn beyond.

The third type of wind instrument sounds by means of air streaming past a hole. Humming tops are one sort. Flutes are another.


Calling any musical instrument a flute that is operated by an air-reed, from primitive gourd whistles to the silver tubes of a modern orchestra, a flute will sound only under three essential conditions. There must be present and in the right relations:

A wind-way, through which passes the thin stream of air that forms the air-reed.

A blow-hole, across which the air-stream passes.

A wind-cutter, lip, or tone-edge, upon which the air-stream impinges.

In addition, most flutes have their proximal ends at least partly closed by some sort of more or less separable Apple, whence the name flpple-flute for a somewhat primitive instrument with an obvious wooden plug.

No flute will sound at all without the first three features. Practically, nearly all flutes do have also some sort of fipple, the panpipe or syrinx being the one notable exception.

Take for .example the common toy flageolet that has a metal tube plugged with a wooden fipple. The narrow passage between the tube wall and the upper side of the fipple is the wind-way. Such a wind-way, more or less built into the rest of the instrument, whether or not there is a separable fipple, is a beak. “Beak flute” is, therefore, a suitable name for the type, while “whistle flute” is objectionable since “whistle” suggests also a shrill pitch. Since, however, a few primitive instruments lack the shaped beak, “wind-way flute” is perhaps the best term.

The familiar orchestral flute, along with the simpler fife of the marching band, is side-blown and without beak. The player's lips, therefore, form the wind-way.

- 42

The following classification of all wind instruments exhibits their relation to one another, to the flute group, and of the different flutes among themselves.


Standing wave within the pipe initiated by:


When it comes to assigning any particular musical instrument to its place in whatever classification one adopts, the problem may not be so simple as it appears. For the fact is almost any tube a scant inch in - 43 bore and a few inches long, can be made to sound as a trumpet. Almost any tube within a wide range of diameters and lengths, provided the two are duly related, can be blown as a flute, and that, rather commonly, either open or stopped. Virtually all nose-flutes can be blown also with the mouth, while any side-blown mouth-flute can also be blown with the nose, if only the blow-hole be near enough to the end and not too large. In other words, primitive flutes being mostly just straight tubes, open often at both ends, each of them can commonly be sounded in more than one way, sometimes in four or five.

It is, therefore, not easy, in the absence of direct testimony, to make out by experiment just how certain primitive wind instruments were meant to be blown. Indeed, it is always possible that a particular pipe may have had more than one use. So the test is not the instrument's possibilities, but that special one of them all which lets the note come more easily and gives the most agreeable or the loudest tone. This principle will govern the discussion which follows.


Since putorinos commonly run from twenty-five to fifty centimetres in length, mostly of the longer sorts, our Salem example (Peabody Museum catalogue, E5, 515), at 41.2cm., is just above average. The mid-hole, fairly large, has its centre 18.5cm. from the blowing end. As many such instruments do not (Andersen, p. 274) this one has a minute vent at the distal end, 2mm. bore, widening only gradually into the main tube, which at its widest is 1.3cm. by 3.2. In effect, then we are dealing with a wooden tube, 1.7 by 1.9cm. inside at its blowing end, that expands slightly for a scant twenty centimetres to the mid-vent, and then, after continuing at about this diameter for another five or ten centimetres, contracts to almost nothing. In other words, our putorino, with its mid-vent open, is nothing more than a slightly conical open pipe just under twenty centimetres long. With its mid-vent closed, it becomes a more nearly cylindrical stopped pipe, with an effective length of something less than forty centimetres, the bore at the distal end being too small to count. Other, sometimes remarkable, features of a putorino are acoustically unimportant.

How such an instrument was actually blown is still an unsolved problem. It has been called a flageolet, though this it most certainly is not. A flageolet has a built-in wind-way; and though a putorino appears, externally, as if it had such, a glance into its interior proves that it has nothing of the kind. The putorino as a flageolet may therefore be written off the books.

Most authorities, Andersen for example (pp. 273-4) call it bugle or trumpet, and its mouth-piece is indeed just about like those of other Maori instruments that are trumpets without any possible doubt. Furthermore, all putorinos appear to be blowable trumpet-wise. Our Salem example, the mid-vent open, blows a rather agreeable a', that is to say, the a above middle c. Andersen (p. 277) records for a somewhat larger instrument “A fine clear mellow trumpet-note is produced by blowing bugle-fashion into the upper aparture”, the note being f # the f # below middle c. Apparently, then, a putorino is some sort of trumpet, whatever else it may turn out to be.

Even so, it is a strange sort of trumpet. For one thing, it is too short. Proper trumpets are at least two or three feet long, and the Maori themselves (Andersen, p. 287) made them five and seven feet. But the smallest putorinos are only nine inches long. And since, as - 44 trumpets, they have to speak through the mid-vent, their effective length is only four or five inches. In general, too, the mid-vent is too small for a trumpet mouth. A trumpet should flare, not contract, distally. Finally, if they are primarily trumpets, why the distal portion at all? Our Salem instrument as a trumpet, performs exactly as well with its lower half stuffed with cotton, while the minute distal opening has, for a trumpet, no discoverable function. One cannot but feel that if any Maori desired merely a five to ten inch trumpet he would have made himself the upper end of a putorino and stopped there.

Maori carvings (Andersen, figs. 59, 61, 62 at p. 277; Best, p. 128, fig. 53; p. 132, fig. 62, A, B, C.) exhibit unmistakable putorinos being handled, not as trumpets, but as flutes, the method of blowing coming out especially well in a slab in the Auckland Museum (Andersen, fig. 61; Best, fig. 62A). A putorino has too large a bore to be blown like a pan-pipe, something like one centimetre beiiig about the limit for this method. Larger pipes than this, end-blown, if they have no special blow-hole, require the player to form such, commonly by means of his own lower lip acting as a fipple.

This general method of blowing a fairly large tube at the end is widely distributed. Piggott (p. 152) and Sachs (p. 213) describes it for the Japanese shakuhachi. Kirby (plate 36 B and 42 A and B) pictures for South Africa both the method of blowing and types of the instruments themselves, which are bevelled where they go against the lip to make this lip-fipple device easier and any other method virtually impossible. Kaudern (fig. 7, p. 46) has some admirable diagrams that show the lip-fipple technique for two different end-blown flutes and brings out clearly the difference between this and the ordinary way of blowing into a pan-pipe or a bottle. Buck (pp. 175 and 177) figures five pan-pipes from Tonga that are bevelled at their proximal ends like the South African single pipes and evidently designed to be blown in the same way. The Salem collection has two like specimens, which do not blow in the ordinary pan-pipe fashion, although these and Buck's examples are the right size for the common method. This sort of pan-pipe seems to have been reported only for Tonga; but Worcester (pl. 57) pictures a larger pan-pipe, thus blown, from the Philippines.

Variants of this lip-fipple method, used especially with long tubes, are illustrated by Kirby (pl. 40A and B and 41B) and in many pictures of the ancient Egyptian oblique flute which survives as the Arab nai (see also Kirby, pp. 119-120). Such long pipes, instead of being held in front of the body like a clarinet are carried to one side to bring the distal end nearer the operating hand. Later in this paper we shall discuss Maori short flutes that seem, at least sometimes, to have been blown lip-fipple and at the end, but held out at one side horizontally about like our ordinary fife. Most of these pipes have their mouth ends bevelled. The fact that the Maori instruments do not, suggests that they really were intended to be used also as trumpets.

This somewhat peculiar lip-fipple method of blowing- a flute seems to have been more often taken for granted than described. The pipe is held, not vertically like a pan-pipe, but thirty or forty decrees from the horizontal, roughly half way between the positions of a trumpet and a clarinet. The open end of the bore rests against the player's lower lip in such wise that the lip closes the hole except for a narrow lune, a couple of millimetres across and twice as wide, which now becomes the blow-hole. This blow-hole should come about the middle of the - 45 red skin of the player's under lip. Above this, about where the skin of the lip joins the mucous membrane that lines the mouth, between the upper and lower lips, is formed just about such a wind-way as would be used for any transverse flute. The player's breath does not strike across the opening of the tube as for a pan-pipe, but even more into the tube than for a transverse flute.

All this sounds somewhat complicated. It really is easily learned with a little practice before a mirror. The main point is that the player's lower lip has two functions. Much of it, well down toward the chin, acts as a Apple to help form a blow-hole. At the same time, upper and lower lips between them form a wind-way about as if this blow-hole were that of almost any flute that does not have a built-in beak. Our Salem specimen, which has a somewhat elliptical bore, is more easily sounded if the wind-cutter is taken at an end rather than a side of the oval and at one of the least damaged spots in the rim. Naturally, these ancient wooden pipes will not blow as easily as a modern scientifically designed instrument. Our own, we confess, is distinctly difficult.

There is, however, yet another way in which a wide pipe can be end-blown, and the putorino quite possibly was.

For this, the blowing end of the tube, instead of being partly closed by the player's lower lip, rests against the tip and the under side of his somewhat extruded tongue, which thus becomes a fipple, while the wind-way is formed between the upper surface of the tongue and the upper lip. Kirby (pi. 33B and apparently a variant, 38B) pictures this method from South Africa. It seems to occur nowhere else, and seems to be correlated with the thick African lip, which if used as a fipple, puts the tone-edge too far from the wind-way. To be sure, the Africans do employ also, quite freely, the lip-fipple method. But if one may judge from Kirby's figures (especially pls. 40A and 42A) they pretty generally offset the thick lip by bevelling the pipe at its upper end. Both the bevelled pipe and the tongue-fipple method are characteristic of Africa, though a bevelled pan-pipe occurs in Tonga (Buck, fig. 1; Mead, fig. 2a; and two examples from Tonga in the Salem collection).

The only evidence that the putorino was ever blown tongue-fipple rests on Maori carvings of the human face. These, in general, merely as a stylistic convention, often show a protruded tongue. One (Andersen, fig. 59; Best, fig. 53) shows a player of what may be a putorino with his tongue out but not inserted in the pipe. A house carving (Best, fig. 62c) shows the player's tongue apparently more or less closing the tube almost exactly as in Kirby's photograph. On the other hand, in the clearest picture of all, another house carving (Andersen, fig, 61; Best, fig. 62A) a putorino is being blown lip-fipple. Possibly the Maoris, like the South Africans, used both methods. We have, ourselves, throughout this study, employed the more hygienic device.

Our Salem instrument, flute-blown, when it sounds at all, yields a flute-like and agreeable tone that is never in the least like the “harsh shrill sound” of Parkinson's early account (Andersen, pp. 9-10). The note comes easier with the mid-vent closed, and easier also with the minute end-hole open. Pitches are uncertain within about one interval, since the blow-hole, as always with flutes, acts also as a vent, raising the pitch when enlarged. We record the pitch at which the note comes clearest and most easily. It is, however, impossible to sound different - 46 notes, each at its best, with the instrument at the same angle and with the same lip position.

Our Salem pipe, sounding clearest, mid-vent open, gives f #″ exactly as it should considered as eighteen centimetres long and sounding its open pitch. This is with the end-hole closed. With this open, the pitch rises to g″, the note coming rather more easily and the tone improved. But each pitch is a little off, so that the effect of opening or closing the end-hole really amounts to hardly more than a quarter tone. Evidently, then, mid-vent open, the lower half of the instrument is little more important for our putorino blown as a flute than it proved to be blown as a trumpet. Partly closing the mid-vent flattens the pitch to f″. But the note is hard to get, and can hardly have been of much use, since the same result can be had more easily by nursing at the embouchure. Here again, stuffing the lower part with cotton has no effect.

Yet that peculiar mid-vent is always there, usually with a small tongue projecting into it from its upper lip, sometimes with projecting tongues from both sides, and occasionally with the two tongues joining so as to cut the mid-vent in two. But some instruments have their mid-vents almost circular, and some have them elongated up and down instead of crosswise (see Andersen, figs. 55-60 following p. 276; Best, pp. 132 A, B, C, D). Nothing in the least like this putorino mid-vent occurs in any other known wind instrument, either flute or trumpet.

But if this mid-vent concerns the acoustics of the putorino, either as trumpet or flute, it should be standardized as to both size and shape, as repeated experiment works out the most efficient arrangement and convention fixes this once for all. But this mid-vent is highly variable, commonly too large for a flute, always too small for a trumpet, usually not at all the shape one would expect for either. One cannot but suspect some non-acoustical reason, most probably nothing more than the conventional rendering of the human mouth in other Maori carvings.

With the mid-vent closed, the acoustical problem becomes more complex. The instrument now becomes a stopped pipe, of uncertain functional length, because of the contracting bore, but something more than 30cm. from mouth-hole to stopped end. It should, therefore, in theory, yield a note somewhat below middle c, and opening the end-hole should slightly raise this pitch.

This last it does—the same scant half-tone shift as for the open pipe. But there is nothing at all like the expected middle c. Instead, mid-vent and end-hole both closed there comes and that easily for this instrument, a loud, clear, agreeable, and unmistakable g″, actually a half tone higher than the open note, though it should be more than a full octave below.

The anomaly is easily resolved. Our putorino, everything closed, is over-blowing, and yielding, not its fundamental, but the twelfth above this. In other words, our putorino, at 18.5cm. as an open pipe, is operating in two segments each of about nine centimetres. As a stopped pipe, its effective length is a little short of 30cm. and it is sounding in three segments, its working length, therefore, about the same as when it was open. That this is the correct explanation is proved by the fact that very careful blowing, mid-vent stopped, brings a faint but unmistakable a, the tube's fundamental as a stopped pipe and one octave below the note it gave as a trumpet. Theoretically, this note should be middle c, since the fifth above the octave of this - 47 is the g″ which the pipe gives over-blown. But the difference is only one and a half tones, and that is taken care of by uncertainties of embouchure and the peculiar shape of the tube. It all goes to show how tricky flutes are.

Evidently, the relation between the open-pipe fundamental and the stopped-pipe harmonic is controlled by the relative lengths of the upper and lower segments of the instrument taken from the mid-vent. Pictures show this to be pretty uniform, the mid-vent, whatever its size or shape being always at about the middle, though one cannot, of course, make out its relation, both ways, to the taper of the bore. Apparently, then, all putorinos, whatever their size and design, are intended to give two notes not far apart.

Such as have the little end-hole can get something like a half-tone shift out of that, trilling easily, rapidly and prettily, and as either open or stopped pipes. The half-tone flattening by partly closing the mid-vent does no more, is harder to do, and merely duplicates effects obtained more easily in at least two other ways. One may well question whether it was ever actually used.

Altogether, then, as a musical instrument, the putorino can hardly have amounted to much. Two or three notes of our scale, with a halftone trill if it had an end-hole, are about its limit, and even these had to be blown in different ways. But Polynesian tunes are also simple. Burrows (pp. 7ff) records from the Tuamotu Archipelago ancient magical chants that run a dozen or twenty syllables on one note and then make a shift or a trill one or two tones above. In his own words (p. 6) “. . . there is a trill or quavering of the voice between the main pitch and the one slightly higher. . . . The tonal and rhythmic material of these chants is so simple, and so moulded by the words, that it is on the borderline between formal declamation and music proper”. A putorino, used thus in an incantation, might well both help to fix the pitch and alter the quality of the chanted sounds. It would, however, be two octaves above a man's voice.

Andersen (pp. 114ff) along with more complex tunes, records some that are equally simple. One in particular (p. 117) consists solely of an f′ repeated thirty-nine times. Still another (pp. 117-18) mostly on the same f′ and nearly a hundred measures long, merely drops from time to time to one, two, or three d′'s A putorino with segments less alike than in our example or with a larger end-hole should manage this. Here is a problem that somebody might well take up who has access to several different instruments. One must, besides, not forget that a putorino sounding only a few notes or only one, could still play a drone accompaniment to voices or other flutes and need not even then join them on every note.

But the problem always remains; why, if any Maori wanted a flute, or a pipe to serve either as flute or trumpet, he did not simply make one, with or without finger-holes, just as other primitive people did and he himself was entirely competent to do? A putorino, admirable as primitive wood carving in a museum case, is an indifferent trumpet and a very poor flute. The Maori had both that were much better. But there is a hint of magic properties in Parkinson (Andersen, p. 10) “. . . these they frequently held up when they approached the ship: perhaps it may be the figure of some idol which they worship”, and Best (p. 133) points out that only the putorino of all musical instruments appears in Maori carvings.

- 48

Acoustically, however, all one can be sure of is that a putorino will blow about equally well either as a flute or as a trumpet, whatever its original owner may or may not have intended it to do. “Bugle-flute” for a name, favoured by Andersen and others, is, then, appropriate. We suggest, however, “trumpet-flute” not to introduce needlessly a dubious term. It seems to us, in addition, that the practice of the Metropolitan Museum of Art, among others, of classifying putorinos with flutes is better than that of Andersen and Best who lump them in with trumpets. After all, whatever their makers thought them, they do look like flutes.

Andersen (p. 273) quotes Williams as saying there were two sorts of putorino, “A kind of flute blown by the mouth, the end being placed below the chin. Another form was played by blowing through it, and was a kind of flageolet”. Certainly the putorino is not a flageolet; and it looks like a better guess that Williams did not see two sorts of putorino, but one sort blown two different ways — as flute and as trumpet.

In any case, the putorino, both as an objet d'art and as a musical instrument, is unique in the world and quite without transitions to anything else or any discoverable history. Not only is it unique in form, but nothing else is even suspected of being sounded in so many different ways.

Even so, there is in South Africa a primitive flute, which although it has no side vent and cannot be blown as a trumpet, does in certain other respects suggest the putorino. Kirby describes these (pp. 106-111) pictures several (pls. 37, 39) and photographs one being played (pl. 38A). They differ greatly in detail from place to place and have different names. Essentially, they are antelope horns or horns imitated in wood, blown flutewise at the larger end, with a small pore at the smaller end that is sometimes left open and sometimes closed by the player's finger, to yield two different notes. The natural horns are, of course, curved. But the wooden imitation is straight, though preserving the taper of the horn, so that the whole instrument has a good deal the general shape of a putorino. In length, they run rather smaller than putorinos, around 15cm., but they go up to twice this to match the smaller Maori pipes.

The only one which Kirby describes in much detail (pp. 106-7) is 13.3cm. long, 1.8cm. in diameter at the blowing end, with the diameter of the end pore 3mm. Its pitch is g' with the end pore stopped and bb' with it open, the larger hole on the smaller tube making more difference than in our putorino. In general, the change of pitch between the two notes is from a half tone to two tones. Kirby (p. 109) seems to state that these pipes were blown tongue-fipple, “. . . laid across the hollowed tongue in the usual way”.

An obvious interpretation of all this is that the putorino simply continues the evolution which the South African instruments have already begun. New Zealand had, indeed, no horned mammals. But the ancestors of the Maori may well have come from some continental spot which did or have borrowed elements of their culture from some people acquainted with animal horns. But any folk blowing an animal horn as a stopped flute, if it loses contact with the creature which provides this, is likely to proceed much as the South Africans have now that game is becoming scarce, reproduce the animal horn in wood, making it straight to suit the rifted wooden block, but still retaining the tapering form and the solid distal portion of the original. The end pore, not always present in putorinos, is an incidental detail.

- i
The three Maori flutes in the Peabody Museum of Salem. The catalogue numbers and sources of the instruments are as follows: The putorino (E 5, 515) and the nguru (E 5, 520) were given to the Museum in 1807 by Captain William Richardson, the koauau (E 5, 521) was given by Captain William Putnam Richardson in 1812.

- ii Page is blank

- 49

But the Maoris went a step further and made their flutes into a trumpet. This may well have been at least partly accident. Given the Maori practice of carving grotesque human faces on any wooden surface, the extraordinary mouth which they favoured is pretty likely to be cut through into the bore of a decorated pipe. After that, it is only a matter of time before somebody discovers that the pipe will blow also as a trumpet.


The nguru of the Salem collection (Peabody Museum catalogue, E5, 520) is the one pictured by Best (p. 148a) and undamaged, the other having somehow disappeared. It enjoys the distinction of being the only wooden instrument of the sort known except the single example in the British Museum. Otherwise it is entirely commonplace. Ngurus run from eight centimetres to sixteen; this at 13.5 is typical. It has also the four conventional vents, one at the little end, one at the turn, and two smaller openings twenty-five and fifty millimetres from the large end.

Ngurus have often appeared in museum catalogues and travellers' tales as nose flutes, it even being assumed at times that the small curved end goes up the player's nostril (Best, pp. 146-147). This last is quite impossible. There would be no vibrating lips nor fluttering reed nor air-stream striking across a blow-hole, No flute, thus blown, could possibly sound at all, and all this also should be written off the books.

There is, in fact, not a little in print concerning nose flutes that rests on misunderstandings concerning them. Roberts, for example (p. 346) opines that: “The peculiar custom of playing a flute with the nose is in itself sufficiently unnatural to afford a feature by which to trace connections”, and that (p. 388): “Its presence in Brazil may be due, as Tylor suggests, to a lip-plug which prevented mouth blowing”. Best (p. 147) thinks it “a puzzling question why anyone should prefer to sound an instrument with the nose, when, presumably, they might effect their purpose much more easily by using it as a mouth flute; and Miss Roberts quotes Tylor's explanation that the high-cast Hindu does not want to risk touching his mouth to anything that may be contaminated. Sachs (p. 46-47) calls such theories “so foolish that they are not worth contradicting”, and offers instead, “the origin of the nose flute is found in the association of nose breath with magic and religious rites”, since it is the nose breath rather than that from the mouth that contains the soul. We have already suggested magic functions for the putorino. Furthermore, Sachs suggests (loc. cit.) that nose flutes are commonly blown by the right nostril because the right side of the body has in general superior mana.

Now all this may, of course, be true, though some of it, as Best (p. 147) wisely points out, is far-fetched. Certainly no one will question that flutes in general do tend to take on magic powers beyond other musical instruments. There remains, notwithstanding, another and quite simple explanation: People blow flutes with their noses because, unless there is a built-in wind-way, that is the easiest method!

A mouth-blown pipe, whether direct or transverse, demands an embouchure that is maintained only by the strong and accurate contraction of many small lip muscles, which soon tire. Even a modern concert instrument cannot be played indefinitely like a fiddle, and primitive tubes are a great deal harder to keep going. But for a nose flute the wind-way lies between the rim of the nostril and the wall of the pipe, cartilage on one side, wood on the other, and no small - 50 muscles anywhere to become fatigued. The single advantage of mouth-blowing results from the fact that different notes require different embouchures for the best tone, lips altering quickly and surely, as noses and built-in wind-ways cannot. So the best flute music is mouth-blown. In addition, it comes out of transverse pipes, because these, always mouth-blown, permit changes of embouchure that are both quick and easy merely by rotating the tube on its long axis.

Curiously, too, nose-blowing is also easier to learn than the familiar method. One has only to get in front of a mirror, set the edge of the nostril close to the proximal margin of the blow-hole, and press the instrument gently upward until the resulting lune becomes one or two millimetres high and about twice as wide, not very different, in fact, from the opening one would make between the lips. The air-reed will be horizontal rather than somewhat downward as for mouth blowing.

Aside from the location of the blow-hole close to the end of the pipe and its somewhat smaller size, the only essential for a nose flute is that the curve of its wall where it forms the floor of the wind-way shall be flatter than the curve of the player's nostril so as to form a proper slit. Part of the difficulty with nose flutes seems to be that the long European nose requires the pipe to be swung sideways in the horizontal plane, instead of held straight out in front of the face as in the pictures of more fortunate peoples. With these remarks in mind, nose-blowing should come at the first try—as mouth-blowing never does.

As for the special mana of the right side, nearly all flutists put the right hand distal to the left to bring the more dexterous fingers on the vents that are most used. That brings the left thumb or the left fore-finger nearest the face and handiest to stop the unused nostril. But the left digit most conveniently stops the left nostril and leaves the right one to sound the pipe.

Parenthetically, we note that in Kaudern's drawings (p. 46) to which we have already referred, the nose flute, as pictured, has no wind-way and therefore cannot sound at all. It should be set a few millimetres higher, so as almost to close the nostril, as we have explained.

And yet, in spite of all the above, our Salem nguru can be nose-blown. It happens to have a large and uncommonly flat region around the end-hole, which is itself fair-sized and sharp edged. This flat region forms the floor of the wind-way, with the pipe about horizontal and the hole at the outer side of the curved tip stopped against the player's upper lip. Thus sounded, it gives the g″ which is reasonable for an open pipe of that length, but the tone, though clear, is faint and not at all what one would expect from a prized and expensive instrument. Apparently all this is accidental. Most ngurus, as pictured, do not have the flat region around the end-hole and therefore probably cannot be nose-blown. In any case, they cannot have been pushed up the nose or blown straight into as has been supposed.

Furthermore, if the end-hole of a nguru is designed as a blow-hole, both it and the region around it should be standardized. Actually, these vents are widely different in size, sometimes with sharp edges and sometimes with rounded, sometimes set in flat surfaces and sometimes at the bottom of sizable craters. There is no such uniformity as in the blow-holes of other flutes. But a finger-hole may have a wide range of sizes, shapes, and positions. In addition, as a nose flute, the vent next the small end, nearly always present, has to be stopped most - 51 inconveniently against the player's upper lip. The smallest leak there and the pipe will not sound at all. It is, of course, conceivable that this vent at the turn is not acoustic at all, but being in general just in line with the bore of the tube was useful in fashioning this. But Fisher's study (pp. 111-15) of a series of unfinished ngurus goes to prove that the boring was always done from the two ends. Finally, the two vents near the large end are so small and so close to the end, that opening or stopping them has virtually no effect at all on either tone or pitch.

Altogether, then, we agree most heartily with Andersen's opinion (pp. 262, 264) that the nguru was blown at the large end and by the mouth. Andersen finds, as we do, that “it may be played with the nose, provided the position of the top hole allows of it”. Andersen (p. 195) lists no sort of nose-blown flute for New Zealand, though he does cite them for Hawaii, Tahiti, and Tonga, and opines (p. 260) that “nose-playing lingered as an occasional method”.

Our Salem specimen, mouth-blown at the large end, by the same lip-fipple or tongue-Apple method already discussed for the putorino, for which its sixteen millimetre bore is exactly right, with all finger-holes stopped, gives a beautiful, clear bb', the same note as the ordinary band fife, flute-like but with the characteristic stopped-pipe quality, and quite as loud as is to be expected from anything of this size. Moreover, this highly agreeable sound is so easy to obtain that one cannot help thinking that exactly this was the original intent. That the pitch is less than the octave below the nose-blown note is explained by the difference in embouchure and the inevitable greater functional length of open over stopped pipes of the same measurements.

Most authors, dealing with short Maori flutes, have stressed the precise location of the finger-holes, which in ngurus are most commonly four but may number from six down to none. Much of all this has been waste labour. With an open pipe, as most flutes are, the location of a finger-hole determines the working length of the pipe, everything beyond the open hole influencing the pitch only partly or not at all. But even with open pipes the location of a finger-vent is by no means the whole story: the size also has its effect.

The modern orchestral flute has all its side-vents closed with pads worked by levers. These vents may, therefore, be given the full size of the bore and be located where they theoretically belong. But a vent closed by a finger has to be fairly small, commonly something like a quarter of the cross-section of the tube bore. So the instrument “speaks” both through the uppermost finger-hole, the end-opening, and all finger-holes in between, compromising the effective length. But the smaller the finger-hole the less it influences the pitch and the further up the tube it must be placed to have the same effect. All this can easily be made out by examining any modern fife or flageolet. Sizes of finger-holes are varied, to put the player's fingers at convenient positions. But old wooden flutes without more than one or two metal keys used to have finger-holes of the same size but irregularly spaced. Even so, with all open pipes, the location of a vent does affect decidedly the pitch.

But a stopped pipe is a different matter. Here the side-holes must be kept small of the order of a tenth or less of the cross-section of the bore, otherwise opening one of them would convert the instrument into an open pipe. A stopped pipe may have many vents, an ocarina has ten, all of which, opened at once, may mave a combined area well - 52 up to that of the bore. Yet so long as no one vent is too large, the instrument remains acoustically a stopped pipe like a clarinet.

For stopped pipes, with their small vents, it makes small difference where these vents are placed. Opening a little hole, located almost anywhere, elevates the pitch a little. A larger hole has more effect, no matter where it is. Two, three, or four little holes, opened at once, have about the same effect as a single larger one to the area of which they sum up. In brief, ngurus, ocarinas, and the like are essentially globular flutes in Sach's sense; their pitch depends more on volume than on tube-length; and their finger-vents, located almost anywhere, are very much smaller than the cross-section of their tubes.

Much has been recorded of Maori flutes with finger-holes plugged and others opened near them (see, for example, Andersen, p. 242 and fig. 45; Best, p. 147) and it has generally been assumed that the shift was solely of location for a change of pitch, as if the instrument were an open pipe. But the original holes, according to all accounts, were carelessly located, as if their positions were small matter. We suggest that such vents were placed more or less after a conventional pattern, and then enlarged cautiously until they delivered the right note. But if this enlarging was overdone and the note made too sharp, then the maker simply plugged the hole and started over.

The Salem nguru, all vents closed except either of the two small ones twenty-five and fifty millimetres from the blowing end, gives c″, one full tone above the bb″. Hole number one, nearest the end, gives the pitch slightly flat. Number two makes it just right. Both vents measure two millimetres in diameter, but number one is, if anything, slightly smaller.

Alas, here is always the problem of embouchure. The finger-vent is two millimetres across. But the blow-hole, only one or two finger-breadths away, is several times as large, with half its margin formed by the player's lower lip, and the angle at which the breath enters alterable within wide limits. Our instrument, held as nearly horizontal as it will sound, and then turned as far as possible toward the vertical, pulling the player's lips away to enlarge the blow-hole and at the same time setting the air-reed more across the hole and less into it, has its pitch raised, with all finger-holes open, by a major third, and with all finger-holes closed by a full fifth. It is, then, quite possible to play a simple tune on a nguru without using the finger vents at all. Even with the best embouchure for each pitch, there is a range of a whole tone without appreciable change of quality. Although we have been as objective as possible in recording pitches, we have little doubt that we have unconsciously nursed our notes to make them agree with European intervals. Maori performers probably blew somewhat different sounds.

The third finger vent of our pipe is, as almost always, at the turn of the curved tip, the full length of the instrument away from the blowing end and only two centimetres from the tip of the instrument. This third vent is larger than the first two, 4.5mm., and it expands inward toward the main bore, instead of contracting as do the small vents. Its cross-section is, therefore, at least five times theirs. Opened alone, it sounds the d″ that we might fairly expect, since with globular pipes, as vents are opened, it takes, on the whole, somewhat larger holes to produce the same effect.

The end-vent, largest of all, 6.5mm. and enlarging promptly to become the interior bore, gives e″ slightly sharp. Vents one, two, and - 53 three, opened together, vent four closed, give a correct e″. The areas of one, two, and three together sum up to just about that of four alone.

Vents three and four, open together, one and two closed, give e#″; three and four with either one or two give f″, with the former slightly fiat, to match the slightly flat c″ when one is opened alone. All four vents open give f#″, which can be rather easily nursed at the embouchure to g″. We need not ring the changes on other combinations, since no two ngurus seem to be alike and any one of them can probably get any pitch within its range by small alterations of the player's lips.

On the other hand, the bb, c, d, e, and f# come so simply and obviously that one cannot help feeling that the maker of the pipe intended them. Here, then, are five notes just about one tone apart in our scale. But Buck's pan-pipes from Tonga (p. 175), which as we have noted are bevelled to be blown lip-fipple, if one may judge from the numbers and the relative lengths of the pipes, and the similar example at Salem, which we know by trying, also sound either exactly or about five notes and not far from one tone apart. Worcester's Philippino (pl. 57) blowing lip-fipple a pan-pipe that is too large to be sounded in the ordinary way, has, to be sure, seven pipes, but two of them seem to be the same length, with the shortest the octave of the longest, and the rest apparently about a tone apart in pitch. Furthermore, the Maori long flute, as pictured, for example, by Andersen (p. 254) and Best (p. 132g) commonly has only three finger-holes, and can therefore, unless over-blown, play only four notes. These, moreover, so far as one may judge from the size and location of the vents, can hardly spread over more than a half octave or be more than a tone apart. So the nguru seems fairly comparable with other Polynesian flutes.

It will, apparently, play most Polynesian tunes, since these seldom range more than a fifth or a sixth above the lowest note. Andersen, for example (p. 123), records a chant of nearly twenty measures, that has only the notes b, e, and f, with a d-e trill that is just the sort of thing a nguru does to perfection. Our own pipe, trilling on the end-vent, bb to e, is as lovely as any bird. Burrows (p. 82) has another example, a long passage of some hundred and fifty notes, all between f and c and mostly between g and bb; while both Emerson and Roberts print page after page of chants and nose-flute tunes from Hawaii that a nguru can take equally well. Handy and Winne do the same for the Marquesas. Burrows (p. 147-8) sums up thus for all Polynesia: “Songs entirely in monotone are known from Hawaii, the Marquesas, and New Zealand. . . . The range of compass is narrow, commonly a fifth or even less. . . . Melodic intervals are usually small. . . . Some of the intervals, indeed, are smaller than the European tempered scale”. But there is a type of primitive tune (Andersen, for example, p. 220; Roberts, p. 284) that instead of closing with a trill, ends with an octave drop. This, of course, a nguru cannot take.

We trust we make clear that a nguru, unlike a putorino, is, within its limits, a real musical instrument, potentially correct in intonation, and with as pleasing a sound as any flute short of a modern concert instrument. It is, in fact, on a smaller scale, not unlike one of the mellowest pipes known, the Japanese shakuhachi, which is blown the same way. Both are primitive in form, sweet to hear, lovely to look upon—and very hard indeed to play.

All of which last does not agree at all with Andersen's (p. 263-4) “It was used as a signalling rather than as a musical instrument, and - 54 its sound was said by Cook, as noted above (p. 9), to be no more musical than a pea-whistle. . . . Colenso. . . . calls it a whistle”. Andersen continues with a quotation from Colenso “I suspect that these, like their trumpets, were not used for obtaining any proper tune, but only for the purpose of making a loud call—as from a chief to his followers”. We, on the other hand, suspect that the nguru, like the putorino, was blown two ways and did double duty—as an uncommonly sweet-sounding flute as we have described, and over-blown as a shrill whistle.

Our own example over-blows easily, giving as a stopped pipe should, the fifth above the octave of the fundamental, f″″. Opening holes one or two raises this note a half-tone, with hole one once more a little flat. Hole three gives g″″. Hole four, alone or with any of the rest in about any combination, all about alike, give close to a″″. So the finger-vents, so far as opening them has any effect at all, influence over-blown notes about half as much as at lower pitches.

All these tones are, of course, shrill, the a″″ being hardly more than an octave below the top note of piano or piccolo. But they are not the least unpleasant, are indeed rather better than most of our own high-pitched whistles and quite as good as any ordinary fife. We agree, however, that a nguru, over-blown, is not a musical instrument, and probably was never used for any proper tune, being much too shrill to accompany the voice, and yielding too small a range of notes to compete either with other flutes or with itself at lower registers. We suspect that our own specimen, most unusual in decoration and evidently expensive, has also an uncommonly good tone.

In brief, then, we interpret the nguru as essentially an ocarina without the built-in wind-way. Both instruments are short wide pipes, always sounding stopped notes. The obvious difference is that an ocarina has more finger-vents, and so runs its scale beyond its lower octave. The ocarina, also, over-blows easily; and in both registers the two instruments sound much alike. There is, in fact, now on the market as a child's toy, an ocarina with its beak on the end instead of at the side, that in size, proportions, general figure, pitch, tone, about everything in short except the number of finger-vents and the beak, fairly matches the larger ngurus. Both ocarina and nguru are essentially globular flutes, but far enough away from strict globularity to take on something of the attributes of the tubular group, conspiciously the richer tone. We like for the nguru the name by which it is so often called, “whistle-flute”, except that whistle-flute has also another meaning. Our example, at least, is a lovely flute in its lower register, a high-pitched and far-carrying whistle when overblown.

Globular flutes other than ngurus and ocarinas are, of course, common enough all over the world. Mostly they are cheap affairs, earthenware or dried gourds, with one or two finger-vents, and blown like pan-pipes. The nguru, laboriously drilled, carefully shaped, elaborately ornamented, is, like the putorino, unique in all the world, found nowhere outside New Zealand. Unique, also, in all the world is the nguru's curled-up tip.

The only thing anywhere the least like it is that strange terra cotta instrument from a pre-historic Peruvian grave described by Mead (pp. 337, 340) discussed, among others, by Andersen (pp. 264-6) and mentioned by Sachs (p. 199). It is a somewhat larger instrument than any ngurus, 22.5cm. over all, and the blowing end, instead of being cut off straight across as in all ngurus, has the bore contracted - 55 almost to a proper blow-hole. The sides of this blow-hole, moreover, are bevelled on both above and below, like the South African flute and the Tonga pan-pipes already noted. Evidently, then, it was blown lip-fipple like a nguru; but the upper bevel makes it a notched flute in Sachs's sense (p. 179) as the nguru is not. One cannot help feeling that resemblances to the nguru have been somewhat over-stressed.

Acoustically, so far as one can make out from pictures, differences are still greater. The Peruvian instrument has throughout the general look of an open pipe, the finger-vents, eight to ten millimetres across, being much too large for a stopped one. Unfortunately, there seems to be nothing in print concerning the acoustics of this Peruvian flute and we have had to fall back on the kindness of Professor George Herzog of Columbia University and Dr. H. L. Shapiro of the American Museum of Natural History for information not otherwise available. Neither, to our regret, succeeded in blowing the stopped note, which is probably not far from one tone below middle c.

However, with the end-hole open and the rest closed, this Peruvian instrument behaves like any open pipe, delivering a bb' as it reasonably should. From this the notes run up successively to b", just about as in any somewhat primitive six-hole flute. Our own model, without the curved tip and with the end opened to the full bore, behaves in the same fashion. So there can be no doubt that the flute in question, its end hole uncovered is always an open pipe. Its affiliations, therefore, are not with ocarina and nguru but more with, let us say, the Japanese shakuhachi, which also is end-blown against a bevelled wind-cutter and has a single thumb-hole underneath.

Evidently then, the curved tip of flutes from New Zealand and Peru are an unsolved problem. We offer, nevertheless, a suggestion concerning the Inca form.

It was, one gathers from various details already discussed, commonly sounded as an open pipe, blown always from the larger end. But open pipes contracted at the distal end are common enough and offer no problem. Sachs, for example (p. 41, fig. 20), illustrates one from Celebes, beaked to be sure, but with side-vents very much as in the Peruvian instrument, with main bore and end-vent exactly the same except that the tip does not curl up. He figures another (p. 193, fig. 45) essentially like this, from Mexico. The shakuhachi also is choke-bored, so was the European pre-Boehm orchestral flute along with Boehm's own early models, the smaller bore at the distal end giving better performance in the second octave where all the notes are over-blown.

Suppose then, a primitive flutist, his pipe already choke-bored, aspires to play such a melody as that which we have already cited (Andersen, p. 220) that runs along on c", sometimes shifts to d", and closes with an octave drop to a c'. He plays his tune in the ordinary fashion, open pipe. Then, for his final note, he stops the distal end of his instrument against his bare knee, makes his tube a stopped pipe, and thus gets his final note. If our hypothetical flutist doesn't like stopping his pipe against his knee, all he has to do it to curl up the tip so that he can reach the vent with his little finger.

But, of course, none of this reasoning applies to the nguru, which is always a stopped pipe and cannot take that octave drop. There is, however, always the possibility, dwelt upon by numerous authors, of borrowing back and forth between the Pacific Islands and South - 56 America. Sachs, among others, argues the point at some length (pp. 202-3). The general idea will be that Polynesian voyagers, reaching South America, brought back at least the tradition of that curved distal end, forgot what it was for, and kept it as a convention or for some magical property. Possibly, also, the form of a whale's tooth, from which many ngurus were made, may have influenced the shape. Or it may well be that this curved tip is reminiscent of an original horn. If so, then nguru and putorino probably had a common origin, both are essentially ocarinas, and the large finger-vent at the tip of the ngui-u is the homolog of the minute pore of the putorino and the African pipes. The nguru retains its curve because it was bored out. The putorino loses its because it was rifted along the grain of the wood into two billets, which are separately grooved and then reunited.

Incidentally, we note Sachs's (p. 202) “. . . the notched flute, outside of America, is confined to the far east, . . .’—the next sentence, however, explains that he is, for the moment, neglecting Africa. We submit that Buck's pan-pipes from Tonga (p. 174, with figs. 1 and 2), of which the Salem collection has also two examples, are also notched flutes.

In sum, then, the nguru, like the putorino, offers a problem that is thus far unsolved. Primitive New Zealand had two unique flutes, unlike anything else anywhere on earth, and with virtually no transitions between either and any other wind-blown tube. Sounding either is a long-forgotten art. All that anybody can do now is to make the best guesses he can concerning both.


If two sorts of New Zealand flute are unique, a third seems common-place enough, merely the immemorial straight tube of pith-wood, animal bone, or bamboo shoot, with a few finger-vents bored through the wall. The Maori did not have bamboo, and employed largely the round limb-bones of birds and quadrupeds, favouring especially those from the legs or upper arms of revered ancestors or persons whom they did not like. Not seldom, however, they bored laboriously through wooden blocks, carving the outsides most elaborately. Possibly the absence of bamboo and the general shortness of the hollow portions of humori and tibiae, along with the difficulty of boring out wooden blocks without metal tools, accounts for the prevalence in New Zealand of so uncommonly short a pipe as the koauau.

The example in the Salem collection (Peabody Museum catalogue, E 5, 521) also pictured by Best (p. 148a), is of wood, uncommonly well carved, and obviously a choice instrument. Especially striking is its likeness to the nguru which we have just discussed. The two are of the same material, similarly decorated, and of almost exactly the same dimensions, 13.5cm. over all for both, bore of 15mm. on one end and 17mm. on the other for the koauau against 16mm. for the nguru, with the two vents next the blowing end in almost exactly the same location in both, although these vents are rather smaller in the koauau, two millimetres in diameter instead of three.

We are assuming that our koauau was blown at the larger end. The rim here has the sharper edge and is better finished; the bore continues at this size nearly to the distal end, as in the nguru; most significantly, perhaps, the tone comes easier from this end. Our koauau, in short, blown lip- or tongue-fipple, with the distal opening against the player's palm or finger tip to make a stopped pipe, most strikingly duplicates all the essentials of our nguru.

- 57

Blown thus, all finger-vents closed, the pitch is b'instead of the bb' of the nguru, where the curved tip gives a slightly greater working length. But the koauau is not quite so easily sounded, nor is its tone quite so pleasing, though it is on both counts superior to our putorino. That it also blows a b' trumpet-wise should be no more than one of those annoying accidents that beset a student of primitive pipes.

This particular koauau has its vents placed after one of the commonest fashions, two that divide the pipe into something like thirds and another, near the blowing end, that cuts the proximal third roughly in halves, with all vents the same size. In our instrument, however, the third hole, which is not present in the nguru, is very slightly smaller than the other two. In general, these vents affect pitch less certainly than in the nguru and there seems to be less systematic plan. Opening either the first or the second raises the pitch from b' to d" with the usual uncertantly introduced by differences of embouchure, which here amount to something like a minor third. The third vent alone gives e"; first and second together yield something between e" and f", second and third something between f#" and g". In general, then, the vents further along the pipe have somewhat more effect on pitch, either alone or together.

Evidently then, our koauau, blown lip-fipple as a stopped pipe, has about the same range of notes as our nguru and could play the same simple tunes, albeit with more difficulty and less certainty of pitch. It even trills prettily on the third hole. Altogether it is a cheaper instrument, easier by far to construct, and much the more common; yet essentially the same thing. Andersen (p. 244) discusses partial uncovering of the distal end, a device which would make a koauau still more like a nguru. This can, of course be done—is, in fact, done regularly on the finger-vents, for example, in the shakuhachi and the recorder. It does not impress us as workable on so large an orifice as the end-hole of a koauau, and we do not stress this likeness to the other pipe.

As one would expect after the analogy of a nguru, our koauau over-blows to the expected f#'", a good clear tone that is easily obtained. Fingering has about the same effect as with the nguru, all holes open giving an uncertain pitch that varies around g'". Briefly, when overblown, the koauau like the nguru ceases to be a musical instrument and becomes a whistle.

In one important respect, however, all koauaus differ from any nguru—the player can take his hand off the distal opening and make his instrument an open pipe.

Sounded thus, by the same lip-fipple method, though with a slightly different embouchure, its note is the expected b", a good enough tone but neither so clear, so loud, nor so easy to get as the stopped b'. Fingering, as with both it and the nguru over-blown, has something like half the effect as with the stopped fundamentals. All three finger-vents open give c#'". But the different combinations make changes of pitch that are within the uncertainties of embouchure, so that it is not worthwhile to record any of them.

Andersen (p. 244) states that a koauau, as an open pipe does not over-blow. This agrees in general with our own experience, so long as the pipe is held in the usual position, in the plane of symmetry of the body. Andersen, however (p. 249ff), discusses the koauau as a voice mask, noting especially (p. 250) that this explains “the fact that the Maori blew sideways across the end of the flute; not directly - 58 across as we do across the neck of a bottle”. Thus “the lips are free to move”. To Andersen's discussion of Maori flutes as voice masks, we add only that our instrument, held sidewise, will over-blow unstopped, and that rather easily.

For this blowing, the pipe, though horizontal, is not really transverse like a cross flute, but about forty-five degrees off the plane of symmetry of the body, in about the position that anybody with a European nose has to blow a nose-flute. The player's lips do not form the usual slit-like wind-way, but are puckered more as for a whistle, the side of both lips partly covering the end-vent as a lip-fipple.

Incidentally, and merely for the record, we note here that both our nguru and our koauau stopped can be doubly over-blown, and somewhat more easily when held sidewise as just described. The note is too high and thin to have any practical use.

This brings us to what is really the main problem of all this discussion: Was the koauau intended to be played as an open pipe or as a stopped one? On the one hand, the stopped pipe gives the better and stronger tone, is more easily blown, has the greater range of pitch, and can be utilized as a whistle. On the other hand, blown open, a koauau could play the octave drop at the end of a passage by the simple device of stopping the distal end as we have already discussed for the Peruvian antique. The instrument may well have been designed to play both ways. The fact that it so minutely duplicates the nguru, which cannot be sounded open, certainly does suggest the stopped pipe as the main object. Furthermore, there is that very common use of many primitive flutes to accompany the voice. Even stopped, a koauau is two octaves too high for unison with a man's singing. Open, it is two octaves too high for a woman's. This also argues on the side of the stopped pipe.

There remains, however, yet another possibility. May we not, in what we are calling koauaus, since we really do not know much about them anyway, be lumping together as the same thing what are really two different instruments?

There is, on the one side, a type, of which our Salem specimen is a pretty average example, commonly of wood, highly decorated and with a good deal the same patterns, and with the finger-vents small and located as we have described, two dividing the pipe into thirds and the third a sixth of the distance from the blowing end. Andersen (pp. 235-243, figs. 41, 42c, 43, 45 and 46) and Best (figs. 66, 67A and B, 68A, 69, 70, 71, 73) picture this type. The resemblance of all these to certain ngurus is most striking (Cf. for example, Andersen, figs. 50 and 52; Best, p. 148a).

The other type, pictured, for example, by Andersen (fig. 47) and Best (figs. 75, 76, 77, 78, contrast especially A and B of fig. 68) that are hardly altered from natural bones, are little decorated if at all, and have their finger-vents fairly large, with two of these located much as in the other type but the third between the other two, to make the spacing about equal and bring all three vents near the middle of the pipe. This sort does not in the least suggest any nguru. On the contrary, outside New Zealand, it would rate as an ordinary bone flute sounded open. Andersen, indeed, though he treats these along with proper koauaus, labels his figures simply “bone flutes”.

There are, however, transitions (Andersen, fig. 42A and B; Best, fig. 64), that have the holes spaced much as for a proper koauau but too large for a stopped pipe; and the carving, which is pretty crude, - 59 only partly obscuring the natural bone. Andersen's legend notes these as “Said to be modern workmanship”, and they certainly do look like something made to sell. Indeed, it is hard to see how the original of figure 42B, any finger-vent open, can be persuaded to sound at all, since it is too short and fat for an open pipe, and the finger-vents are too large for the stopped one.

If we count these out, and treat the long animal bones with the large vents and the general open-pipe look as ordinary bone flutes, rating only the other type as a proper koauau, we come close to having, for the ancient Maori still a third unique wind instrument, a cruder, cheaper, and somewhat inferior nguru, but nevertheless something that is not quite matched anywhere else on earth.

As for the other type, the ordinary short bone instrument with three finger-holes evenly spaced, what is this, essentially, more than the distal half of an ordinary long flute, either direct or transverse and blown either by nose or mouth, that occurs everywhere, all over the world? People who have hollow reeds or bamboo make these up to three feet long, with finger-vents toward their distal ends. Those who have not use marrow-bones, mostly human, and have to make their pipes only five or six inches long. So the Maori played about the same tunes as other Polynesians but an octave higher. This for the common-place bone flutes, which we assume to have been played open. The point of this discussion is that we suppose the proper koauau to be essentially a nguru designed to be played stopped. The Maori, we suggest, developed the stopped pipe, because their bone flutes had to be short and by blowing them stopped the pitches came down by an octave, to about where they are on reed and bamboo long-flutes.

There remains one more acoustical problem, not especially important, that involves both nguru and koauau.

Consider, for a single example out of many the ordinary six-hole European flute which we know also as fife, and diverse flageolet. It starts its scale with all finger-vents closed, opening them by single additions as it runs up its first octave. All these instruments are so planned that with all holes uncovered they give a note one half tone below the octave above the pitch with all holes closed. The next note is over-blown, all holes closed, one octave above the lowest pitch, as with all open pipes, and the second octave continues as before, all notes over-blown.

But that same lowest note of the second octave is obtained more easily and of better quality by uncovering a single vent, always in our European instruments the one nearest the blow-hole, though this is not the case everywhere. In other words, that proximal vent, which always comes about the middle of the pipe, is at just the region where, as we explained at the beginning of this paper, is located the internal antinode when the pipe is sounding its first over-tone. Opening a vent just there aids the formation of the antinode and promotes the octave shift. The clarinet has a special vent for the same purpose.

We have already pointed out that Maori bone flutes of the type which we do not reckon as koauaus and suppose in general to be blown as open pipes, nearly always have a finger-vent at just about the middle, exactly where it should be to serve also as an antinodal release to assist over-blowing.

But the proper koauaus, of which our own is an example, do not have a vent near the middle—still another reason for thinking they were blown as stopped pipes. But they do pretty generally have a - 60 vent, the one which we have called number three, one third of the way in from the distal end, just where the antinode comes when the pipe is over-blown stopped. To be sure, our particular nguru like many-others, does not have this hole, but some do, for example one of Andersen's (fig. 50C). There is, to be sure, an antinode at the same spot in an open pipe doubly over-blown. But so short and wide a tube as is a koauau does not in general doubly over-blow, and our suggestion of that third vent as a hint of a closed pipe still stands.

But a stopped flute doubly over-blown, forms one node a fifth of the way in from the stopped end and another two-fifths in from the open one. One would like to interpret the vent at the turn of the nguru and vent number two of both instruments as this antinodal release. Alas, the situation of these is far from invariable and at best they are only rather near their theoretic positions. Experimentally, all one can say is that the over-blown notes come at least as easily when the proper vents are uncovered, and though the difference at most is small, so too is it not very great in our own modern instruments where one vent is especially planned to do double duty.


In brief, then we interpret the putorino as both a trumpet and a flute, that is blown as a flute both open and stopped, the functions of which were both musical and magic. We look upon the nguru as always mouth-blown and essentially an ocarina, which is at the same time both a rather admirable musical instrument and a convenient whistle. We regard the koauau as something quite separate from the common primitive flute, being essentially another sort of nguru, that differs from the curved form in having its distal end stopped by the player instead of by the maker, and like the nguru, both a flute and a whistle. We recognize, however, that all koauaus in spite of their short tubes, may also have been sounded as open pipes like any primitive flutes.

  • Andersen, Johannes C.—Maori Music with its Polynesian Background, New Plymouth, New Zealand, 1934.
  • Best, Elsdon—“Games and Pastimes of the Maori”, Dominion Museum Bulletin No. 8, Wellington, New Zealand, 1925.
  • Buck, Peter H.—“Pan-pipes in Polynesia”, Journal of the Polynesian Society, Vol. 50, No. 4, December, 1941.
  • Burrows, E. G.—“Native Music of the Tuamotus”, Bishop Museum Bulletin 109, Honolulu, 1933.
  • Culver, C. A.—Musical Acoustics, Philadelphia, 1941.
  • Emerson, Nathaniel B.—“Unwritten Literature of Hawaii”, Bureau of American Ethnology Bulletin 38, Washington, 1909.
  • Fisher, V. F.—“The Material Culture of Oruarangi, Matatoki, Thames”, Records of the Auckland Institute and Museum, Vol. 2, No. 2, December, 1937.
  • Handy, E. S. Craighill and Winne, Jane Lathrop — “Music in the Marquesas Islands”, Bishop Museum Bulletin 17, Honolulu, 1925.
  • Kaudern, Walter—“The Passage of Air through a Flute”, Ethnologiska Studier 10, Gottenborg, 1940.
- 61
  • Kirby, Percival R.—The Musical Instruments of the Native Races of South Africa, London, 1934.
  • Mead, Charles W.—“The Musical Instruments of the Incas”, Anthropological Papers of the American Museum of Natural History, Vol. XV, Part III, New York, 1924.
  • Piggott, Frances T.—The Music and Musical Instruments of Japan, London, 1909.
  • Roberts, Helen H.—“Ancient Hawaiian Music”, Bishop Museum Bulletin 29, Honolulu, 1926.
  • “Melodic Composition and Scale Formation in Primitive Music”, American Anthropologist, Vol. 34, No. 1, January-March, 1932.
  • Sachs, Curt—The History of Musical Instruments, New York, 1940.
  • Wead, Charles Kasson — “Contributions to the History of Musical Scales”, Report of the United States National Museum for the year ending June 30, 1900, pp. 419-442, Washington, D.C.
  • Worcester, Dean C.—“The Non-Christian Tribes of Northern Luzon”, Philippine Journal of Science, Vol. 1, No. 8, October, 1906.