Volume 33 1924 > Volume 33, No. 131 > The evolution of Maori clothing. Part III, by Te Rangi Hiroa (P. H. Buck), p 185-197
                                                                                             Previous | Next   

- 185

[See also pictures at the end of the previous article]

III. [Continued from Vol. XXXIII., No. 2.]

(b) GARMENTS NOT MADE OF BARK CLOTH.—Though bark cloth was so widely used, it was not suitable during wet weather or whilst fishing. For ordinary rough wear in the bush, very little was used. Linton 1 says of the Marquesans that the simplest form of clothing consisted of a few fau leaves wrapped round the genitals but he does not describe how they were kept in position. On festive occasions and in certain ceremonies, material other than bark cloth was also used.

As in the case of bark cloth, these other garments may be grouped in three forms according to usage: Loin girdles, kilts and cloaks. The loin girdles retain the name of maro. These maro were worn by the highest chiefs as a badge of rank. As the technique of manufacture is not so simple as that of the kilts, they will be discussed last.

Kilts.—The usual material for rough kilts throughout Polynesia were the leaves of the ti, Droecena terminalis. The Polynesian ti leaves were longer, broader and softer than the New Zealand ti, Cordyline—and were thus eminently suited for providing a set of vertical elements for a kilt. Turner, 2 writing of Samoa, says that during the day a covering of ti leaves was all that either sex thought necessary. “They sewed ti leaves together and made themselves aprons.” Though he does not describe the manner of sewing, his illustration shows a kilt of finely split leaves joined together at the upper border by a three-ply braid. He also states that the men wore a smaller one about a foot - 186 square and the women one with longer ti leaves reaching from the waist to below the knee and made wide so as to form a girdle covering all round. These kilts were called titi.

Mr. J. F. G. Stokes found similar kilts as far East as Rapa. They were made of ti leaves which were either split into narrow widths or attached whole to the supporting band of three-ply braid. By kind permission of Mr. Stokes, the kilt made with unsplit leaves is reproduced in Plate VII.

Rarotongan Kilt.—A kilt that I have from Rarotonga, is of a different nature. The material consists of strips of the bast of the purau, Hibiscus—Plate VIII. It is 78 cm. wide and 48 cm. in depth. The supporting part is comprised of two separate twisted cords stretched side by side. The bast strips, about 86 cm. in length, are doubled under the two cords. Each half is brought up on the outer side of its respective cord, both turned over the upper edge, passed down between the two cords and drawn taut as in Fig. 8, c.

Attachment of strips to upper cords.
- 187

This is continued along the required width, the strips thus being looped closely along the cords. Four centimetres below the supporting cords, the vertical strips are further kept together by a single weft line of two strands worked with a plain twine. Each half turn encloses two of the doubled vertical strips. (Fig. 9). The two weft strands are knotted over the marginal strips at either edge and then tied to the two supporting cords above. The ends of the two supporting cords are prolonged sufficiently to permit of tying round the waist.

Showing line of plain twined weft.

The use of the two strand plain twine should be noted. It is identical in technique with the plain twined weave in basketry.

Niue Island Kilt.—Another type of bast kilt comes from Niue or Savage Island. It is made of wider and longer strips of the inner bark of the fau, Hibiscus. They were used in dances and on festive occasions. The one I have is 137 cm. wide with a band 26 cm. deep. From the lower border of the band hang bast strips 61 cm. in length. See Plate IX.

- 188

The bast strips are attached to a double cord as in the Rarotongan kilt, but the fixation is slightly different. The two supporting cords rest one above the other. The bast strip is passed over both cords, A. in Fig. 10. The part B.

Strip attachment to upper cords (front).
Back view of figure 10.

of the strip is carried over the upper cord, X., behind both cords, up under the lower cord, Y., crossed upwards over the first part A., passed back between the two cords X. and

- i
Photo by W. Revell-Reynolds. Kilt of hibiscus bast from Niue or Savage Island.
Photo by W. Revell-Reynolds. Fine plaited mat from Samoa.
- ii
Photo by W. Revell-Reynolds. Fibre mat from Samoa. (University Museum, Dunedin.)
- 189

Y. and brought down obliquely across its own loop at the back. Fig. 11. shows the back view. The two parts of the strips are kept diverged below the double cord, one part passing obliquely to the right and the other part to the left. This is done all along the double cord for the required width of the kilt.

The band 26 cm. deep is now plaited and the parts A. and B. of the strips are treated as crossing wefts, dextral or sinistral as the case may be. About six of these strips are brought together to form a single plaiting weft and the plaiting proceeds with a twilled two stroke, i.e., the composite wefts containing six strips of material pass under and over two crossing wefts of a similar nature. When the plaiting is 26 cm. deep, the wefts are knotted together and the ends hang down vertically as a fringe.

On the outer surface of the kilt, the plaited part is ornamented by adding rows of wide strips of bast arranged in vertical loops and rows of narrower strips loosely twined in a two-ply twist. The latter has a crinkled appearance and is run transversely across the kilt in a continuous series of small rings or circles. This ornamentation has no significance as regards technique for they are attached with rough stitches of bast material. The two supporting cords are split at the ends of the garment and plaited into a three-ply braid for tying round the waist. At the lower end of the plaited band, some elements of the weft are knotted into meshes resembling netting. The knot, however, is not the netting knot, but is made by bringing dextral and sinistral elements together and tying both with an overhand knot. Below the first line of knots, the two elements of each knot are diverged to form a fresh combination. Another row of overhand knots is tied and the process continued for five rows of knots. The result is the formation of a series of open meshes. Below the last row of knots, the strips hang down with the others not used in the knotting, to form a fringe. Plate IX. shows the inner surface on the left where the twilled two plait is not obscured by the ornamentation.

These garments are also called titi or sisi as it is pronounced in Western Polynesia. This type of kilt is also found in Samoa. The points to note are the fixation of the bast strips to the supporting cords and the fact that the garment is plaited with a twilled-two stroke.

- 190

Tahitian Mats.—Ellis 3 described mats in Tahiti as being made from the bast of the purau, Hibiscus. After the bast was prepared it was split into narrow strips frequently less than eighth of an inch wide. “They were woven by hand and without any loom or machinery. They commenced weaving at one corner, and having extended it to the proper width, which was usually three or four feet, continued the work till the mat was about nine or ten feet long, when the projecting ends of the bark were carefully removed and a fine fringe worked round the edges. … When first finished, they are of a beautifully white colour and are worn only by the men, either round the loins as a pareu or with an aperture in the centre as a tiputa, and sometimes as a mantle thrown loosely over the shoulder.”

This is a case in which the use of the term weaving cannot be accepted as being technically correct. The lack of any mechanical contrivance and the commencing in one corner must be considered. Though weaving must commence in one corner, no time is spent there as the weft works quickly across. In plaiting, however, some time is spent in the corner in separating the dextral and sinistral wefts and interlacing a sufficient number to form the working diagonal edge by which plaiting proceeds across the article being manufactured. There seems little doubt that these mats were plaited.

Pandanus Mats.—Brigham 4 states that mats made from pandanus leaves were used by the Hawaiians, not only as a wrap but as a waist cloth, malo. The Tongans used pieces of old floor mats tied round the waist with yards of sinnet braiding as a sign of mourning. These were plaited with the usual Polynesian check stroke.

Samoan Fine Mats.—A higher development of the plaited mat for clothing purposes is seen in the ie taua of Samoa. They were considered the most valuable form of clothing. Stair 5 describes them as being made from the leaves of the ie, Freycinetia. The prickly edges were cut off with a shell and the leaves baked in a native oven. The inner or finer part of the leaf was separated from the outer - 191 and treated with sea water until bleached. Turner 6 says the pandanus was used and after being scraped as thin as writing paper, was slit into strips, sometimes as fine as 1-16 inch in width. The plaiting stroke used was the check. The mats were made from two to four yards square. The commencing and finishing borders were fringed with the unin-terlaced ends of the wefts and often ornamented at the lower border with scarlet feathers. See Plate X. The mats were thin and as flexible as calico. They were thus quite suitable as a garment. They were worn round the waist as a kilt.

In Tonga, similar mats were made but not usually as fine as the Samoan article. Mariner 7 mentions the Tongan fine mat as gnafi gnafi.

Samoan Fibre Mats; ie sina or sealoa.—According to Brigham, 8 these mats (?) were made of unbleached bast strips of fau, Hibiscus or more “usually of fine thread made by pounding the bast and then bleaching the fibre until it is as white as well cleaned bannana fibre.” The garment was more than doubled in thickness by a “pile put on after the mat was woven by passing a parcel of the fibres with a full turn about a mesh of the mat at suitable intervals, and these can be pulled out only by loosening the loop formed about the mesh; pulling on the ends only tightens the hold on the mat.” The length of the pile varies up to seven inches. They were highly valued as presents and were also used “in the curious tokens of virginity custom.”

Through the kindness of Mr. H. D. Skinner in lending me the specimen from the Otago University Museum, I have been able to examine the technique. The garment is roughly 82 cm. by 148 cm. The material is evidently unbleached strips of Hibiscus bast about two mm. in width. On the inner surface, the garment looks like coarse sacking, there being four strips to one cm. The technique is plaiting with a check stroke. The wefts thus run diagonally to the edges. On the outer surface, strips of the same material are attached and give a rough shaggy appearance to the garment. These strips are attached by passing one end - 192 under a crossing weft and drawing it through to half its length. The end is brought over the crossing weft, again passed under it and drawn taut, Fig. 12. The strip thus forms two tags. The tags are placed close together and are

Attachment of tags to outer surface.

24 cm. long. As the strips are passed round an outer weft where it crosses an inner one, the latter prevents the appearance of the fixation loop on the inner surface of the completed garment. There is no knot in this fixation. As the nature of the material prevents slipping or running, pulling on the tags only renders them tighter as Brigham remarked. The free ends of the tags are frayed out and give the fleecy appearance mentioned by Turner. Plate XI. shows the outer shaggy appearance of the garment, whilst the turned back left side shows the inner check plait with the wefts running diagonally from the edges. Other details do not concern us here. The garment looks like a rough Maori rain cape but the technique of the two has absolutely nothing in common. Here we have another example of plaiting being used to provide a garment.

Mariner 9 mentions a garment in Tonga that must have been made in a similar way. His reference is as follows:— “Gie, stronger mats made of the bark of the fow or olonga, worn chiefly by people in canoes to keep out the wet, as the water does not damage them; they appear as if made of

- iii
Netting on inside of feather cloak from Hawaii. Dominion Museum Photo.
- iv
Outer feather surface of Hawaiian Cloak. (Dominion Museum, Wellington.) Dominion Museum Photo.
- 193

horse hair. Labillardiere mentions that he saw women of rank with a sort of mat made of the white hair of a horse's tail—he supposed from some horses that Cook had left there.” There is little doubt that Labillardiere was mistaken as to the nature of the material.

Kilt from the Tuamotu Archipeligo.—Ling Roth 10 has figured a very old kilt that is in the British Museum. It comes from Egmont Island (Vairatea) in the Tuamotus. Along the upper border it is 66 cm., along the lower 90 cm., and in depth 45 cm. down the middle. No mention is made of the material but the warps consist of “two independent strands which are so placed that the upper one goes from right to left, and left to right in the alternate space between the twined rows, and give a zig zag appearance to the work.” The weft contains two elements used with a plain twined weave. There are 52 rows thus making the spacing less than one cm. apart. Fig. 13 taken from Ling Roth shows up clearly that the twining technique has progressed considerably on that of the Rarotongan kilt.

Spaced weft lines of plain twine. (After Ling Roth.)
- 194

HAWAIIAN FEATHER CLOAKS, ahuula.—These magnificent feather cloaks are described by Brigham in the Memoirs of the Bishop Museum, Vol. I., No. 1, with additional notes in Vol. I., No. 5, and Vol. VII., No. 1. The basis is made from fibres of the scraped bark of the olona, Touchardia latifolia. The fibres are made into cords or threads by twisting on the bare thigh. A fine net, nae, is made with a shuttle and gage, the usual Polynesian netting knot being used. Stokes 11 says that the mesh varies from .05 to .27 of an inch and the twine varies from .025 in. in diameter. With fine meshes, the niao or net mender may be used instead of the shuttle and in some cases, the end of the cord is dipped in juice from the root of the ki, Cordyline terminalis, which on drying makes the fibre stiff. Pieces of netting were cut to shape and joined together. The pieces were often of different sized mesh as seen in Plate XII., which shows the inner surface of a cloak in the Dominion Museum.

Feathers from various birds were attached to the outer surface. Different colours were worked into patterns in which crescents appeared as well as triangles and lozenges. The most valuable colour was yellow from the feathers of the mamo, Drepanis pacifica, whilst the red from the iiwi, Vestiaria coccinea, gives a fine appearance. It will be noted that the Hawaiian name for these cloaks is ahuula. With the droping of the K, this corresponds to the Maori kahukura, the name given to red feather cloaks. Other colours as black and green were also used. The feathers were tied to the lower angles of the meshes of the net with fine threads or single fibres. A turn was taken round the shaft of the feather which was then bent down and another turn taken round it. This effectually fixed the feather and prevented it from being pulled out. The smaller feathers used give these cloaks a smoother appearance than Maori feather cloaks. See Plate XIII. Apart from the difference of shape and type of figure design, there is a certain resemblance which we shall see later, is quite superficial.

Shorter garments are used as capes.

- 195

Garments from Rapa.—Mention has been made of fragments of garments discovered by Mr. Stokes in Rapa to the south of the Austral group. In his preliminary report, he says, 12 “In Rapa, the mortuary customs have some interesting features in connection with the drying of bodies. The sepulchres yielded specimens of garments, one of which, a fragment of the early Rapa dress, is in technique identical with the Maori rain cloak.” In this fragment, the weft contains two strands, the stroke was the plain twined weave and the weft lines were spaced a little apart. This as Mr. Stokes remarked, is identical with the technique of some of the Maori rain capes or cloaks. Furthermore, the tags affixed on the outer surface to turn the rain, were put on in a similar manner to Maori work.

In another mortuary robe fragment found by Mr. Stokes, the weft consisted of a single strand which was carried round the warp with a wrapped stroke as shown in Figs. 5 and 6 A.

In a rain cape called a taveru, Mr. Stokes sent me a diagram which shows twining of a more complicated nature at the neck band of the cape.

LOIN CLOTHS, malo.—The only information secured applies to Hawaii and Tahiti.

Hawaiian Malo.—These follow the technique of the feather cloaks and were only worn by the highest chiefs. One in the Bishop Museum, consists of a plain piece of netting with a .25 inch mesh. It is a long narrow strip, 15.3 feet in length and 7 inches wide. The malo of Kau-mualii described by Brigham 13 is made of a closely woven net of olona, 4.5 inches wide and 11 feet 10 inches long. This is covered on both sides with red feathers of the iiwi, the attachment being the same as in the cloaks. To both edges, a band or lei of yellow oo feathers is attached by frequent cords. This increases the width of the malo to six inches. The end which is to hang down in front, is ornamented with rows of human teeth and bundles of fish teeth. In addition to passing round the waist as a belt, one end is evidently carried up over one shoulder as a cordon or baldric. On the statue in Honolulu of Kamehameha I., - 196 Brigham points out that the malo is incorrectly shown on the outside of the cloak at the back.

Tahitian Maro.—In Tahiti, the maro or girdle of feathers was worn by certain Ariirahi 14 or great chiefs as a symbol of their pre-eminence. The heads of two families had the right to wear the Maro-ura or girdle of red feathers. These were the families of Vaiari and Punaauia. The Papara head-chief had alone the right to wear the girdle of yellow feathers, the Maro-tea. Cook 15 in his last voyage described seeing the maro with which, as he put it, the Tahitians invested their kings. It was kept rolled up in a bundle in the sacred place or marae. The girdle was “about five yards long and fifteen inches broad; and from its name, seems to be put on in the same manner as the malo, or piece of cloth, used by these people to wrap round the waist. It was ornamented with red and yellow feathers; but mostly with the latter, taken from a dove found on the island. The feathers were in square compartments, ranged in two rows, and otherwise, so disposed as to produce a pleasing effect. They had been first pasted or fixed upon some of their own country cloth.”

From this account, we see that the basis of the Tahitian maro was a strip of bark cloth and that the feathers were fixed to it by some sticky substance.


In considering the technique of the various garments not made of bark cloth, we find that in Western Polynesia, as shown by the garments of Samoa and Tonga, the art of plaiting was the only one requisitioned. In the northern area of Hawaii, netting was used as well as plaiting. In Eastern Polynesia, plaiting was evidently used in Tahiti. Mr. Stokes informs me that some Tahitian garments in the Bishop Museum have a knotted weave, whilst the wrapped weave has been recorded by him in Rapa. The important point, however, is the appearance of the simple twined weave with a two strand weft. This appears as a single line in the Rarotongan kilt, and as several spaced lines in the garments of the Tuamotus and Rapa. This technique is - 197 exactly similar to that in basketry and is no doubt derived from it. It is significant that it does not appear in Western Polynesia and in Hawaii, though the plain weave was used in the basketry of both areas.

In the garments available for the whole area, there is no trace of the technique of true textile weaving, and, as a consequence, no indication of any form of loom.

(To be Continued.)

1   R. Linton, 1923, Memoirs Bishop Museum, Vol. 8, No. 5, p. 416.
2   G. Turner, 1861, Nineteen Years in Polynesia, p. 202.
3   W. Ellis, 1831, Polynesian Researches, Vol. I., p. 186.
4   W. T. Brigham, 1906, Memoirs Bishop Museum, Vol. II., No. 1, p. 49.
5   J. B. Stair, 1897, Old Samoa, p. 144.
6   G. Turner, opt. cit., p. 202.
7   W. Mariner, 1827, Account of the Natives of the Tonga Islands, Vol. II., p. 206.
8   Brigham, opt. cit., p. 94.
9   Mariner, opt. cit., p. 206.
10   H. Ling Roth, 1923, The Maori Mantle, Bankfield Museum, pp. 118-9.
11   J. F. G. Stokes, 1906, Hawaiian Nets and Netting, Memoirs Bishop Museum, Vol. II., No. 1, p. 153.
12   Report of the Director for 1922, Bishop Museum, Bulletin 4, p. 11.
13   W. T. Brigham, 1918, Memoirs Bishop Museum, Vol. VII., No. 1, p. 33.
14   Memoirs of Ariitaimai, 1901, Paris, p. 7.
15   Cook, III. Voyage, Vol. II., p. 37.