Volume 35 1926 > Volume 35, No. 138 > The evolution of Maori clothing. Part IX, by Te Rangi Hiroa (P. H. Buck) p 111-149
                                                                                             Previous | Next   

- 111
THE EVOLUTION OF MAORI CLOTHING.
IX. [Continued from Vol. 34, No. 4.]
CONNECTING MAORI TECHNIQUE:

SINCE the last section went to press, the author had the good fortune to examine a unique garment in the Otago University Museum. It was a very old South Island rain cape with tags of tussock grass, Poa caespitosa.

There were two rain-capes in the Otago University Museum with old labels stating that they were made of tussock grass. On examining the first, it was obvious that tussock had not been used for the rain tags. These tags consisted of flax that had been soaked and then lightly scraped as in the North Island rain-cape method already described. Every here and there, wide tags occurred that had not had the inter-fibrous material completely removed. The warps and wefts were made of scutched flax-fibre whilst the lower border thrum commencement and the three-ply neck finish were quite orthodox.

It was therefore with feelings of suspicion that the second cape was examined. Here, however, all doubt was happily dispelled, for whilst the warps and wefts were of dressed flax-fibre, the rain tags throughout were of tussock. Apart from the tussock tags, some points in the technique were so unique that the cape merits a full description.

The garment was found in a cave on Mount Benger in Central Otago. Though the neck border was torn on the right, the garment was otherwise well preserved, and the technique could be clearly followed.

Its dimensions were 40 inches in width across the lower border and 45 inches across the mid-horizontal line. Owing to the tearing of the neck border, it was impossible to determine the exact width in this line, but it was little if anything greater than the mid-horizontal measurement. The - 112 depth was 29 inches at either side border and down the mid-vertical line. (See Fig. 134.) The same depth down the

FIG. 134.
Dimensions of tussock rain cape.

middle line was due to the significant fact that there were no inserts, the number of weft rows being 63 throughout. The increasing width was due to extra warps being gradually added on either side edge between the bottom border, a, and the mid-horizontal line, b. Otherwise, owing to there being no insert, the garment would have been perfectly rectangular.

The warps consisted of well-dressed flax-fibre without any trace of extra or inter-fibrous material that is so common in the warps of the rain capes. From the whiteness and clean appearance of the material, it was evident that the fibre had been carefully washed and beaten as in the preparation of the warps of the dress cloaks. The warps were fairly thick, there being 7 to one inch.

The weft material also consisted of scutched flax-fibre. The weft rows were .46 in. apart and consisted throughout of regular, well executed single-pair twining.

The rain tags were entirely of tussock. Each tag contained a dozen or more of the long filiform leaves naturally united together at the root end.

Technique.—Following the order of manufacture from the bottom border, the tags were attached to every third weft row down to the mid-horizontal line. Each tag was attached to the outer surface of the warps in the same manner as feathers. The tuft with the long free end pointing upwards, was placed on the warp with the part one to two inches from the closed root end on the line of the weft row. The anterior weft element was passed over it - 113 in the half turn round the warp. The lower root end of the tuft was then bent upwards on the next warp and fixed by the half turn of the single-pair twine. A warp was then missed before the next tag was attached (see Fig. 135). The free ends of the tags were from 12 to 13 inches

FIG. 135.
Attachment of tussock tags in first or lower part of tussock rain cape.

in length, whilst the short root ends projected upwards beyond the weft row for from one to two inches. In the Figure the long ends of the tags are cut away in the lower weft row to show the untagged weft rows.

From the mid-horizontal line of the cape, the tags were attached to every alternate row with no warps missed between the the tag attachments, Fig. 136. The tags are

FIG. 136.
Attachment of tags in second or shoulder part of tussock rain cape.
- 114

cut at the right of the lower row to show the technique. Near the neck border, the tags were longer and there was no clear weft row between the last two rows of tags.

When the cloak was reversed for wear, the long free tags overlapped from above downwards and effectively turned the rain. The closer thatch shown in Fig. 136 covered the neck and shoulders.

The commencement technique of the cape is quite unorthodox. Two two-ply twisted cords were evidently stretched horizontally, one above the other. A warp was doubled over them as in Fig. 137a. The near limb of the warp was then passed back between the horizontal cords, x and y, and the posterior limb brought forward between them on the right of the other limb as in Fig. 137b. The

FIG. 137.
Two-cord commencement of tussock rain cape.

other warps were treated in the same way. A single-pair twine was then run along immediately below the lower horizontal cord, each stroke including both limbs of the doubled over warp as in Fig. 137c. The ends of the two horizontal cords were knotted, twisted together with a two-ply twist for a short length, and ended with an overhand knot.

The neck finish was very primitive. The ends of the warps, instead of being finished off with a twist or braid, were simply turned back on the next warp as in Fig. 138.

FIG. 138.
Neck finish of tussock cape.
- 115

It must be remembered that a close row of tussock tags was at the same time attached along this last weft row, but for the sake of clearness they are not shown in the Figure. Thus the warp ends, about 7 to 8 inches in length, form part of the last layer of tags at the neck border of the cape.

Special neck fringe.—An extraordinary feature of the cape was a special neck fringe that had been made separately and then attached. One row of long tufts of tussock were joined together by a single-pair twined weft row. A long tussock tuft had a half-turn of the weft pair made

FIG. 139.
Technique of neck fringe, tussock cape.

round it about 3 inches from the root end. Fig. 139a. The next tuft was placed beside it, the root end of the first tuft turned up on it and both included in the next weft stroke, Fig. 139b. This technique was continued, each stroke including a fresh tuft and the turned up root end of the previous tuft as in Fig. 139c. The fringe was made the same width as the neck border of the cape.

Attachment of neck fringe.—The completed neck fringe was laid along the outer side of the neck border of the cape, with its single weft row corresponding to the last weft row of the cape. They were fastened together by a two-ply twisted cord making a chain knot by passing round

FIG. 140.
Attachment of separate neck fringe, tussock cape., a—neck fringe, b—cape neck border, c—cord of attachment.
- 116

both weft rows with overhand knots at regular intervals as in Fig. 140. On the right of the Figure the neck fringe is turned upwards out of the way to show the neck border of the cape.

Link with earliest technique.—At first sight, there is little to distinguish the general appearance of the tussock cape from the ordinary rain-capes made from flax or kiekie. The long thread-like or almost cylindrical leaves of the Poa caespitosa are very similar to the coarse fibres of flax or kiekie that have been treated by the rough method of light scraping after soaking in water. The tussock rain tags, however, form the least important feature of the cape. The unique features are the commencement, the finish, and the neck fringe.

The commencement or setting up of the warps by doubling them over two horizontal cords is unlike any method seen in the developed craft of the North Island. For any affinity, we must refer to the kilt commencement already described for Niue and Rarotonga. Though the exact method of doubling the warps over is different from either, the main principle of using two cords is the same.

The finish at the neck border is about as simple as it could possibly be, and again finds no resemblance in the technique of the North Island. To some extent, perhaps, it resembles the selvedge commencement of cloaks except that the warp ends are left long and free. There is, however, a vast difference between the desired simplicity of a commencement technique and the lack of finish at a neck border.

The method of attaching a separate neck fringe by a knotted cord is somewhat crude and primitive. The absence of any attempt at an insert also adds to the more primitive nature of the technique of the garment.

The question to decide is whether the more primitive characteristics of technique enumerated above are really old or whether they are due to the work of an unskilled and inexperienced craftswoman. Unskilled work in modern times is particularly marked by lack of inserts and poor finish especially about the neck band. Both these faults are present in the tussock cape. The writer knows of a case where a woman does really good work but has to get someone else to finish off the neck band. However, the preparation of the warps, the regularity of the weft rows, and the neat fixation of the tags, show that the garment was carefully made by an experienced craftswoman and could not - 117 have been the amateur attempt of a modern tyro. In addition, the fact that the garment was found in a shelter cave in a district not inhabited or visited by Maori for a considerable time, supports the conclusion that it was made before closer European settlement caused degeneration in the craft of weaving.

The South Island has been settled by successive waves of people from the North Island. Each wave has pushed the previous one further south, so that Otago, the southern end of the South Island, may be regarded as a remote area. The Ngaitahu tribe, which formed the last wave, permanently occupied the east and south coasts of Otago and must have pushed the surviving remnants of their predecessors into Central Otago and the fastnesses of the west. It might therefore be expected that the technique of some of the earlier waves of people would be at the stage of evolution reached in the North Island at the time of their departure. Isolation in a remote area would form an additional factor in preserving some, at least, of the characteristics of the earlier stages of technique. We must therefore regard the tussock cape found in Central Otago as an important link with the past. Its peculiarities in technique are thus due, not to degeneration or defective craftsmanship, but rather to retention of methods marking the earlier stages in the evolution of clothing technique.

THE CLOTHING OF THE MORIORI OF THE CHATHAM ISLANDS.

The material culture of the Moriori of Chatham Islands has been shown by Mr. H. D. Skinner 1 to have affinity with that of the southern portion of the South Island of New Zealand. Among the facts cited by him to prove the New Zealand origin of Moriori culture is that they appear to have brought with them to the Chathams “weapons and ornaments made by the usual Maori methods from New Zealand nephrite and conforming in shape to common Maori types.” This makes the working of greenstone very old, for according to Moriori genealogical evidence, their last migration to the Chathams must have occurred somewhere about A.D. 1175. The Moriori were thus separated and short circuited from New Zealand for over a century and a-half - 118 before the arrival of the great Hawaiki migration from Eastern Polynesia in 1350.

Regarding the Chathams as a remote area, we would expect Moriori clothing to shed some light on the technique used in New Zealand before the coming of the Hawaikians. Unfortunately exact details are lacking. Shand, 2 who made the most use of his opportunities for study, dismissed the subject of clothing in less than one page of his Memoir on the Moriori. He states that: “Originally, i.e., from the date of their arrival at the group, the people used mats for clothing, the general name of which was WERUWERU. These were made of scraped flax (muka), and were fine in texture and warm, but, owing to the number of seals to be found there, this kind of clothing was abandoned and sealskin universally adopted, so that the art of making mats became lost.”

Mr. Elsdon Best supplied the author with the following note in Maori: “The clothing of the Moriori in ancient times was the pakirito. The making (whatu) of that garment was like that of the parawai of the Maori—the fibre was scraped—it was not twisted or beaten like the Maori method, with two beatings—but the garment was somewhat stiff. Because of the abundance of sealskin that form of making garments was abandance of sealskin that form of making garments was abandoned, and they took to those skins. After Europeans arrived, the seals were exterminated and the Moriori plaited the tukou; some flax was split into narrow strips, others were wider. Änother thing was a net that was almost worn out, scutched fibre was filled in and birds' feathers sewn on.” This note was obtained from John White's unpublished manuscript, but no mention is made of his source of information. Owing to the slack use of terms applying to exact detail, the reference to the similarity to the Maori parawai cloak cannot be taken seriously.

From the above, there seems little prospect of learning what the original Moriori technique was. Shand, though he lived in the Chathams, had to rely on traditional accounts regarding the clothing of the pre-European period. Mr. William Baucke, 3 better known under the pen name of “W.B.”, was also brought up in the Chathams and knew - 119 the Moriori well. He is opposed to much in Shand's writings. He maintains that the Moriori aped their Maori conquerors as far as possible and that many of their traditional accounts are adapted from what they learnt from their Maori masters. It would thus be only natural for the Moriori, after attributing the loss of their original clothing technique to the plentiful supply of sealskin, to say that their original technique resembled that of the Maori. The first European to discover the Islands was Broughton in 1791. Fortunately, both he and Johnston, who accompanied him, mention that the Moriori wore other garments beside sealskin. This was in the period before European sealers had arrived and killed off the supply of seals. From their first-hand evidence it is proved that the loin-cloth (maro) and the cape or cloak made from Phormium tenax were in use during the period that seals were plentiful. These alternative garments should indicate the main principle of the original technique. It is evident that the sealskin material did not do away with the need of a technique to deal with flax material.

MORIORI GARMENTS.

Sealskin.—The transition to skin clothing in the colder climate of the Chathams is interesting in view of what happened to other races under similar circumstances. Broughton 4 described the sealskin as being worn with the fur inside, tied with sinnet round the neck and falling to below the hips. The sinnet referred to would probably be a three-ply braid of scutched flax fibre which was called muka as in New Zealand.

Loin-cloth.—The loin-cloth or maro was used throughout Polynesia. No matter what the material, the common usage was that it was brought back under the crutch and fastened by a turn round the waist or to a separate belt. John Biscoe 5 in 1831, in speaking of a short shoulder mat worn by three Moriori who came aboard his vessel, finishes with the words: “And a strap of the same material passed under the crutch.” Hunt, 6 another early observer, says: “The females girded their loins with a band of plaited flax.” This most probably also refers to the maro though it may possibly - 120 have been a kilt. The old men of the Ngati-Mutunga tribe, which took part in the conquest of the Chathams, say that the Moriori made a maro from a grass called pouaka which was softer material than flax.

Shand 7 described the marowhara as a war girdle “about five yards in length, worn criss-crossed over the shoulders and round the waist, with the ends brought through the Tahei, or girdle, to allow of one end hanging in front and the other at the back, and coming down nearly to the knees. These were supposed to be worn by people of rank.” We have already seen that the Hawaiian feather malo was worn in a similar way by chiefs of the highest rank. It was evidently the ordinary maro that Johnston 8 referred to when he said that the Moriori had no covering but a mat, “neatly wrought, like the marro of the Sandwich Islands.”

Shand7 further states that the Moriori marowhara was made of scraped flax—not scutched like muka. The scraping of flaxen strips to whiten and soften the material was the usual Maori technique for fine plaiting. It is thus quite evident that the technique of the marowhara was plaiting and not twining.

The word marowhara is also used by the Maori to denote a war girdle. The name itself conveys the meaning of a plaited band used as a maro. It occurs in the Ngapuhi saying applied to the rough seas of the West Coast—

Te moana i tauria ki te marowhara.
The sea that was girded with the war belt.

Both Maori and Moriori recited incantations whilst the war belt was being girded on.

The kilt.—Broughton 9 recorded that “Some were naked, excepting a well-woven mat of fine texture, which, being fastened at each end by a string round their waists, made a sort of decent garment.” As, however, Broughton and Johnston were together and saw the same persons, it may be that Broughton was referring to the garment that Johnston likened to the “marro” of the Sandwich Islands.

Cape or cloak.—Broughton9 says that in addition to sealskin, they wore “mats neatly made, tied in the same manner, which covered their backs and shoulders.” Biscoe9 records: “They were quite naked, with the exception of - 121 a short mat over the shoulders which seemed to be used as a roof to them to turn the water off.” Hunt, 10 of a later period, added more detail. “The only garment worn by them in early times was made from the leaves of the flax split into three or four strips and interwoven into each other like a kind of stuff between netting and cloth with all the ends, which are eight or nine inches long, hanging down on the outer side. It was suspended from the shoulders like a cloak; tied round the neck and extended a little below the knee.”

The technique of splitting the flax leaf into three or four strips indicates that the garment was made by plaiting. Shand 11 supports this by saying that they “also plaited a rough kind of mat called tukou, from broad strips of flax leaves, which on shrinking formed a very indifferent protection from the cold.”

It seems likely that there was a finer kind of mat neatly made as described by Broughton and a rougher rain-mat plaited of broader strips as described by Hunt and Shand. To the former would belong the fragment of an ancient mat found by Percy Smith in a cave on Chatham Island. This he informed Mr. H. D. Skinner was more finely plaited than any Maori mat he had ever seen and called to mind the siwa mats of Fiji and Samoa. Though the fragment was unfortunately lost, Mr. Smith's comparison settles the question of technique, for the Samoan fine mats were plaited and not twined. It is interesting in view of the affinity between the Moriori and the South Island cultures that the finest Maori plaiting that the author has seen was in a deep almost bag-shaped satchel discovered in a cave in Otago.

Net garments.—Shand 12 describes one kind of garment as being made from scutched fibre in a fine kind of net. It was called kupenga, which is the Maori and Moriori name for a fishing net. Shand states that it was used as a substitute for sealskin and thus implies that it was a recent invention. White's informant said that the net was an old one that was nearly worn out or rotten (pirau). It would seem from this that the net was not specially made as a garment but that disused fishing nets were utilised as a covering. A Maori once told the author that old fishing nets - 122 were sometimes used as garments but it could only have been among the poor and neglected who had no women to attend to them. It was used as a substitute for want of something better. White's informant, however, goes a step further when he says that the net was repaired with scutched fibre and feathers were sewn on. It is most improbable that there is any affinity here with the Hawaiian feather cloak. The Moriori was seeking a clothing fabric whilst the Hawaiian used netting as a basis upon which he could develop featherwork.

OTHER GARMENTS.

Kakaponga is given by Shand12 as the name of another garment used as a defence against spears in the period before mortal combat was abolished by the wise enaction of Nunuku. The Maori used broad plaited belts, tatua kotara, for a similar purpose. It is probable that the Moriori kakaponga had affinity with it in technique as well as use.

Pokipoki is mentioned by Skinner 13 as a coarse kind of mat. The elders of the Ngati-Mutunga say of the Moriori clothing: “He pokipoki te kakahu,” the garment was a pokipoki. The pokipoki is the ordinary rough form of plaited floor-mat but the Maori who have lived in the Chathams, say that the Moriori article had even wider strips than were used by the Maori. They further stated that the main garment of the Moriori was the fire. They squatted over their fires to such an extent that their shins and abdomens were quite darkened and blistered with the heat. The pokipoki was worn over the shoulders to protect the part away from the fire. The Ngati-Mutunga and Ngati-Tama tribes of North Taranaki took the Chathams in the early part of the 19th century. They use the word pokipoki for the coarse floor-mat in addition to the common Taranaki name of taka. As however they are the only people who use the word pokipoki in this sense, they must have adopted it from the Moriori after the conquest.

CONCLUSIONS.

From the enumeration of the foregoing garments, it will be seen that the Moriori resembled his Polynesian kinsmen in the use of the maro and the cape or cloak. He - 123 resembled his nearest geographical neighbour, the Maori, in the use of the Phormium tenax with which their environment provided them both. The scraping of strips and the scutching of fibre was common to both. In this, however, there is nothing extraordinary, as they are the universal methods in Polynesia for preparing material. Slight modification may occur according to the material available. The common use of the word muka for scutched fibre and weruweru for garments, is of importance. The fact that the word taveru is used by the people of Rapa to denote a raincape, shows that the root weru or veru is an old Polynesian word for clothing that must have been in use before the separation of the people of Rapa, New Zealand, and Chatham Islands. Muka must have been in use on the mainland of New Zealand before the Moriori migration to the Chathams unless it was subsequently adopted from the Maori. It, however, merely means scutched flax-fibre. If not an old Polynesian word it could still have been in use before the Moriori migration as applying to the material from which cords and fishing lines were made. These must have been in use before the development of Maori clothing from scutched fibre. The use of the word muka by the Moriori cannot be taken as evidence that they were acquainted with the Maori technique of making garments from scutched flax-fibre.

The direct evidence of Broughton, Johnston, and others has shown that the sealskin garments did not entirely oust the garments of flax as Shand has stated. Even when seals were plentiful, as when the Islands were first discovered by Europeans, the Moriori had alternative garments. Whilst alternative garments continued to be made from flax, it seems highly improbable that the main principle of a known technique for dealing with flax should have been abandoned. Whilst the necessity for using such garments continued, how could the most suitable known technique be forgotten by disuse? The presence of such clothing points in the direction of chronological sequence. It is natural, therefore, to suppose that the technique in use on the discovery of the Islands by Europeans would be that handed down from previous generations. Though woodcraft and canoe-building degenerated, lack of suitable material had something to do with it. In the case of clothing, there was no lack of suitable material, for the Phormium tenax grew in abundance. Though isolation may have prevented progress - 124 and even caused degeneration in craftsmanship, it does not seem feasible that the main principle of a clothing technique could have been utterly forgotten. As the principle of the technique of the alternative garments in use before and after the advent of Europeans points to the craft of plaiting, with the more recent addition of netting, we are led to the conclusion that it is extremely improbable that the Moriori ever had any knowledge of the extended use of the single-pair twine or the two-pair interlocking weft in the manufacture of clothing. Taking the available data into consideration, we seem justified in assuming that Maori clothing technique developed its characteristic features in New Zealand after the Moriori had departed for the Chathams.

THE CLOTHING OF THE NORTH PACIFIC AREA, NORTH AMERICA.

The North Pacific area ranges northward from California to the Alaskan Peninsula. Attention has been drawn to the resemblances between the material culture of this area and that of New Zealand. As early as 1778, Captain James Cook 14 noticed that the natives of Nootka Sound had amongst their weapons “short truncheons of bone, something like the patoo patoo of New Zealand.” In mentioning the carved woodwork, he does not seem to have noticed any resemblance to Maori work. The following, 15 however, is interesting. “In most of the houses were women at work, making dresses of the plant or bark before-mentioned, which they executed in exactly the same manner that the New Zealanders manufacture their cloth.” In speaking of flaxen or woollen garments, he goes on to say: 16 “The former of these are made of the bark of a pine-tree, beat into a hempen state. It is not spun, but, after being properly prepared, is spread upon a stick, which is fastened across to two others that stand upright. It is disposed in such a manner, that the manufacturer, who sits on her hams at this simple machine, knots it across with small plaited threads, at a distance of half an inch from each other.” As regards the actual garments, he says: 17 “Their common - 125 dress is a flaxen garment, or mantle, ornamented on the upper edge by a narrow strip of fur, and, at the lower edge, by fringes or tassels. It passes under the left arm, and is tied over the right shoulder, by a string before, and one behind, near its middle; by which means both arms are left free; and it hangs evenly, covering the left side, but leaving the right open, except from the loose parts of the edges falling upon it, unless when the mantle is fastened by a girdle (of coarse matting or woollen) round the waist, which is often done. Over this, which reaches below the knees, is worn a small cloak of the same substance, likewise fringed at the lower part. In shape this resembles a round dish cover, being quite close except in the middle, where there is a hole just large enough to admit the head; and then, resting upon the shoulders, it covers the arms to the elbows, and the body as far as the waist.”

Clark Wissler 18 has classified the clothing technique of this area as downward or finger weaving as opposed to upward or loom weaving. “In the loom, the weaver begins at the bottom and builds the fabric upwards, driving the weft home with a downward stroke; in the other, the warp threads are hung loosely from a horizontal support and the fabric built from the top, the weft being pushed upwards into place. In loom weaving, a sword or batten is used to beat down the weft and also as a shedding device, though an additional shedding device may be used. In downward weaving, there are neither battens nor shedding devices, the fingers taking their place, though a bodkin or other pointed instrument may be used to force the weft into position.”

James Teit 19 in his Memoir on the Thompson Indians of British Columbia shows that spaced single-pair twining was used in their basketry. The same technique was used with sagebrush bark in making socks 20 to use with moccasins, quivers 21 for arrows and cloaks and ponchos. The poncho 22 figured in Fig. 141 is evidently the type of garment described by Cook at Nootka Sound as resembling a round dish cover with a hole to admit the head. In a specimen seen by the author in a private collection, there - 126 was no doubt as to the spaced rows of single-pair twining. Teit says that willow bark and cedar bark were also used. Wissler 23 states that cedar bark was used in the same way by the Kwakiutl and neighbouring tribes and that wild goat wool was sometimes treated in a similar fashion by the Salish.

FIG. 141.
Thompson Indian poncho with spaced rows of single-pair twining. (After Teit.)

Ling Roth 24 has described in detail a cedar bark mantle from this area that is in the Perth Museum. Its depth down the middle is 98.5 cm. or practically the same as the type Maori dogskin cloak which is 98.8 cm. Its width is 140.5 cm. or about the same width as the Maori tag cloak. The weft rows are worked in single-pair twining as in the Maori rain capes except that the rows run in pairs close together. The pairs are two separate single-pair twined rows and not a two-pair interlocking weft row. Of these paired rows, there are 54 with an average weft row interval of 1.82 cm. They are thus wider apart than in Maori work, where the weft row interval may be only .59 cm. The warps average 6.4 to the inch. They are much coarser than the Maori warps which average from 7.5 in the kilt to 27.5 per inch in the finest dress cloaks.

- 127
FIG. 142.
American cedar bark mantle, with single-pair twining. (After Ling Roth.)

This mantle is shown in Fig. 142. It has a collar marked A which is from 15.5 to 18 cm. in depth. The weft rows of the collar are in single rows and closer together than those of the body. It is ornamented at the top and the bottom with a zig-zag pattern in black and yellow hair which Roth says is similar to that in a Chilcat blanket in the British Museum.

There are no inserts in the body, but the double rows are made to converge by goring at the point B in the Fig. so as to make the bend for the collar. The bottom border is finished off by turning up the ends of the warps over a horizontal strand and fixing them with a double row of single-pair twining. A strip of greenhide has been twisted round the last pair of twined weft rows.

Fig 143 shows more detail of the parts marked A and B in the previous Fig. The single weft rows of the collar A are shown with the space occupied by the zig-zag pattern, and below that the goring of the paired weft rows of the body.

Blankets were made in this area with a spun yarn, and goat and other hair was used. The technique was close single-pair twining. The best known of these garments is

- 128
FIG. 143.
Cedar bark mantle, showing details of single-pair twining and goring. (After Ling Roth.)

the Chilcat blanket, but unfortunately G. T. Emmons's memoir on The Chilcat Blanket was not available to the writer. Goldenwiser 25 states that the blanket was woven in sections and afterwards sewn together. Designs were woven in colour which were copied from the patterns on the wood-work. “The fact that the weaving was done in small sections which were afterwards sewn together, enabled her to follow the painted designs with great accuracy, and the change of the technique in this case has exercised no visible effect on the character of the design.” The Chilcat blanket in the British Museum has had a totemic design painted on it in red and black.

Comparison with Maori technique.—At first sight, it almost seems as if Captain Cook's remarks that the two techniques were executed in exactly the same manner were correct. Both areas used unspun fibre and simple contrivances to suspend the warps. In both, the technique was based on the single-pair twine in basketry and the weft rows were spaced. The fabric was worked downwards with the fingers. The method of wearing the cloak or mantle was also similar. Roth 26 says that the lengthening of the warps by overlapping was also common to both as was also the bifurcation of the warp to increase width. Both peoples used the close single-pair twine as in the Chilcat blanket and the dogskin cloak.

- 129

When a close survey of the two crafts is taken, the differences are marked. Besides an unspun fibre, the Americans also made a spun yarn, employed animal hair and wool and used the spindle. The Chilcat blanket was made throughout of spun yarn.

For warp suspension, both peoples used two wooden uprights. The Americans, however, used a wooden crossbar over which the warps were spread. This necessitated a different commencement technique from that of the Maori. In the latter case, the first weft row was completed before the warps could be suspended between the wooden uprights.

If the Americans bifurcated the warps to increase width, then another difference exists; for we have already drawn attention to the fact that the Maori did exactly the opposite.

With regard to weft technique, the Americans stopped at single-pair twining with the weft rows spaced or close. The Maori went further. Spaced single-pair twining was used for the rougher capes and cloaks and close twining for protective purposes. The two-pair interlocking weft was used with the best dress cloaks. It was a marked advance, as was also the wrapped twined stitch of the taniko bands. Twilled weaving and the spaced half-hitch weft were also used. Thus, whilst the Americans had only one variety of weft stroke, the Maori had no less than five.

In decoration, the two techniques were entirely different. Roth 27 says that the zig-zag and other geometrical borders were produced in the cloaks of the Americans by covering the bast warp with black and yellow fur-hair. If a close twine is used, it is difficult to see how the patterns would show on the surface when the warps are covered by the weft twine. In the twined basketry of the Thompson Indians described by Teit, colour is introduced by wrapping the coloured element over the weft elements for the requisite number of strokes. It is only in this manner that the colour can show on the surface. To avoid the disjointed use of short lengths of differently coloured elements, the Maori made use of the wrapped twine or birdcage stroke already described in taniko work. The colour was obtained by dyeing the fibre. Making all allowance for the difference of material, the Maori technique shows a higher standard of inventive genius.

- 130

The sagebrush bark poncho was used in wet weather, but there was no attempt by the Americans to turn the rain by means of a thatch of overlapping layers of material attached to the outer surface of the garment as in the case of the Maori rain cape or cloak.

The Americans do not seem to have thought of decorating the outer surface of their garments by means of separate attachments. The Maori, by attaching tags, rolls, dogs'-hair tassels and feathers, besides using coloured strands in loops, spirals, pompoms, and twists, opened up a much wider field of ornamentation and created a greater variety of garments.

The American method of shaping the collar by goring was occasionally used by the Maori in their kilts. The insert, unknown to the Americans, was universally used by the Maori, and various forms were adapted to the bulge required, the size of the garment and the manner in which it was worn. This alone shows that the Maori had more initiative in the field of twined work.

The Americans used greenhide strips to strengthen the lower border of the cloak by twisting the strip along the lowest weft row. In New Zealand, the few strips used for ornamentation at the bottom corners of some dress-cloaks, and the complete covering of dogskin cloaks, were attached by separate strands of dressed fibre.

The Chilcat blanket seems to be the highest development of the clothing technique of the North Pacific area. It owes much of its pleasing appearance to the unique designs worked upon them. The working of sections and subsequent joining together was not done by the Maori. Though the garments were sometimes daubed with red ochre there was no painting of patterns.

On reviewing the resemblances and differences, it seems evident that the only main feature in common between the two techniques is the use of the single-pair twine. Single-pair twining in soft material necessitates suspension and downward finger weaving. The use of short lengths of unspun fibre also necessitates the joining of warps by overlapping. Thus the main resemblances are primarily due to the use of the twining technique. Single-pair twining is so widely spread that there is no need to account for its presence in the two areas by attributing it to remote contact or later diffusion. The differences are so great that we must - 131 come to the conclusion that both techniques were independantly based on the single-pair twine of basketry and, influenced by available material, developed along totally different lines.

RECAPITULATION.

Hawaiki.—The last ancestral Hawaiki from which the Maori came, may be fairly definitely stated to be Tahiti, Raiatea and perhaps other islands of the Society Group in Eastern Polynesia. Rarotonga in the Cook Group seems to have been a rallying point for some of the voyaging canoes.

Bark cloth.—Whilst in Eastern Polynesia, the Maori with the rest of his Polynesian kinsmen was a manufacturer and wearer of bark cloth. When the Hawaikian migration of A.D. 1350 came to colonise New Zealand, in addition to cultivable food plants, it brought the paper mulberry, Broussonetia papyrifera (aute) to provide material for clothing.

Alternative clothing, plaiting.—Besides the main supply of bark cloth, the various branches of the Polynesians used other material to provide alternative forms of clothing. In Samoa, Tonga, Niue, and other parts of Western Polynesia, as well as the Sandwich Islands in the Northern Hemisphere, these alternative garments were made by the craft of plaiting. The Sandwich Islands also fixed their feather work on fine netting as a foundation. Plaiting also seems to have been used in Tahiti, and may have been used in other parts of Eastern Polynesia.

Single-pair twine.—Single-pair twining, with spaced or close weft rows, was used with fish-traps and basket receptacles throughout Polynesia. Eastern Polynesia differs from Western Polynesia in that the single-pair twine of basketry was used with softer material as an alternative technique to make simple garments such as kilts. This use of the single-pair twine has been described for Rarotonga, Tuamotu, and Rapa. The principle of adapting the single-pair twine to the simple clothing technique of kilts was therefore probably known to the Eastern Polynesians before the Maori left Hawaiki.

Kilt commencement.—The kilt must have been the garment upon which experiments were conducted. An early and widely spread method of commencing the kilt was by twisting the vertical elements round two horizontal cords. - 132 It survived in the tussock rain-cape from a remote area in New Zealand. See Fig. 144.

FIG. 144.
Two-cord commencement., a.—Niue kilt, b.—Rarotonga kilt, c.—New Zealand cape

From some such commencement, the development that subsequently occurred was, at first, in the number of rows of spaced single-pair twining. In Rarotonga, we have the mere beginning with a single row; in Tuamotu and Rapa, the greater advance is marked by the increased number of weft rows. Mr. J. F. G. Stokes informed me that the lack of the paper mulberry in the Tuamotu Group and the colder climate of Rapa would both provide an incentive for the extra development of an alternative technique.

Climatic difficulties in New Zealand.—The Maori may be regarded as having arrived in New Zealand with his paper mulberry plants and a full knowledge of the Polynesian technique of making bark cloth. He also knew the two-cord kilt commencement and the rudiments of the single-pair twine technique as applied to friable material. Whilst the paper mulberry plants were maturing, he naturally experimented with the barks of local trees. We infer from tradition that he found the lace-bark, Hoheria populnea, the most suitable of a poor lot. He must have soon found, however, that bark material was inadequate as clothing in his new environment. As Banks pointed out, the paper mulberry trees did not do well in the colder climate. In any case, the plants do not seem to have grown much south of the Gisborne district. Traditional accounts from the Ngati-Kahungunu tribe of Hawke's Bay say that attempts to grow them at Waimarama and Te Aute failed. Even in the warmer North, the supply of material would have been quite inadequate to clothe the population. In the southern half of the North Island and the whole of the South Island, the paper mulberry tree would not grow. The cloth made from lace-bark was too poor to be of use as clothing. Thus bark cloth disappeared from the Maori clothing complex. - 133 In the regions where the climate was favourable, the paper mulberry continued to be cultivated for ornamental purposes and the making of kites. The finding of the two paper bark beaters in the Whangarei Harbour proves that the limited quantity of bark was still made with the Polynesian technique. The finding of strips of inferior bark cloth in caves in the South Island shows that the bark of the Hoheria populnea served, in the regions where the paper mulberry would not grow, a similar purpose to that in the North.

Alternative material.—Even had an adequate supply of paper mulberry bark been quickly available, bark cloth would have been found unsuitable in the severe climate of New Zealand. The Maori was forced to fall back on an alternative material. The friable strips of hibiscus bark or dracaena leaf found a better substitute in the fibre of the Phormium tenax. The material necessitated a suitable technique. What technique was first used in those far-off days, it is difficult to say. We know that the well known ancestors, Toi and Whatonga, came from Eastern Polynesia two centuries before the Fleet of 1350, and that they found people here before them. From traditional accounts, the Maori seems to have had a contempt for the crafts of their precursors. Taking the Moriori of the Chatham Islands as the direct representatives of the earliest immigrants, it would seem as if their clothing technique stopped short at plaiting. Still one would have expected the influx of Toi and Whatonga blood to have evolved something better in the two centuries that preceded the coming of the Hawaikians. Be that as it may, the advent of the greater influx of more vigorous and mentally superior Eastern Polynesians altered the complexion of affairs. They brought with them the knowledge of the single-pair twine as used with friable material and this technique was readily adapted to the local alternative material.

The Cape.—The colder climate necessitated a covering for the upper part of the body. The kilt could be worn over the shoulders just as well as round the waist, but to give more protection to the body, more weft rows were needed. Some of the Maori kilts have the twined part made deeper with extra wefts rows. In two varieties of such garments already described, there is nothing to distinguish the cape from the kilt except the situation of the tying cords. In the taveru garment from Rapa, Stokes wrote me that it - 134 was used as a kilt or as a cape. Thus there can be no doubt that the cape was derived from the kilt.

Suspension.—The Maori, however, was faced with the further problem of providing an efficient thatch for protection against the cold rain of a severer climate. The simplest technique was that of freeing one end of successive additions of short warp lengths to form an overlapping thatch as in the more primitive para rain-capes shown in Figs. 63 and 64. This necessitated an important change in technique. For ease of work, the first weft row was suspended horizontally between two sticks and thus became the aho tahuhu, the ridgepole or beam weft row.

Weaving sticks.—These were necessitated by the extra detail required for the cape. The simplest forms of Maori kilt with little twining beyond the waist band, were made on the ground without suspension between sticks. The author has no information as to whether the kilts of Rarotonga, Tuamotu, and Rapa were suspended between sticks or not. If they were not, the turuturu or weaving sticks become an independent invention in New Zealand. The few wefts rows of the simpler kilts did not urgently need suspension, but the more elaborate garments demanded some such device. Once the principle came into use, it could subsequently be applied to simpler garments.

Downward weaving.—The suspension of the work brought in another innovation. In plaiting, the work proceeds in a direction away from the body. Suspension of the first completed weft row with the warps hanging down with their lower ends free and unattached to any fixed object, necessitated the weft rows being worked successively from above downwards. This constituted downward or finger weaving as defined by Clark Wissler.

Bottom border upwards.—Furthermore, as the long ends of the lengths of fibrous material had to hang down to form the warps, the short free ends to form the thatch had to project upwards. On completion of the garment, the successive rows or layers of free ends overlapped each other from below upwards. To successfully turn the rain, the completed garment had to be turned upside down in order that the overlapping layers should, as in the shingles or thatch of a roof, overlap from above downwards. This instituted the important principle in Maori technique of commencing garments from what was to form the bottom - 135 border when worn. There was no choice in the matter. The technique demanded it.

Evolution of rain tags and warps.—The evolution of the rain tags can be clearly followed in the three types of rain-capes known as para, pureke, and pota, which culminate in the hieke rain-cloak. Intimately associated with this is the evolution of the warp. The stages are shown in Fig. 145.

FIG. 145.
Evolution of the rain tag and the warp., A—para cape, B—pureke cape, C—pota cape, D—hieke cloak, x—thatch tag, y—warp.

A shows the para cape where the short lengths are of rough fibre with no differentiation between the free thatch-ends, x, and the warp-ends, y. The warp is composed of successive additions of short lengths. B shows the pureke cape with the free thatch-ends unscraped and the warp-ends more carefully scraped than in the para. The warp is still composed of short lengths. C shows the pota cape with a distinct continuous warp but still reinforced with the scraped ends of the tags. The scraped ends of the tags have to enter into the composition of the warp owing to the method of fixation being adhered to. D shows the hieke cloak with an absolutely independent warp into which the tag-ends do not enter, owing to the new method of tag fixation. It is evident from the figures that the pota and the hieke cloak must still be commenced with the bottom border upwards in order that the rain tags may point upwards during manufacture and not interfere with the twining of the subsequent weft rows below them. The same working position was necessary, as we have seen, in all forms of cloaks with external attachments such as rolls, twisted cord tags and feathers.

- 136

Rain-cloak.—The evolution of the separate warp demanded the joining of the warp material, which was done as shown in Fig. 123. The rain cloak was merely a lengthening of the cape. It was used for warmth as well as protection from the rain. As a dress-cloak, the warps were more carefully dressed and prepared, whilst ornamentation was introduced by spacing yellow tags amongst the others.

Fine-weather cloaks.—Having solved the more immediate needs of providing against wet and cold, attention was turned to the exigencies of everyday life. For ordinary fine-weather use, rain tags were unnecessary. Merely dispensing with tags provided a simple plain garment which must have formed the basis for the better class cloaks. Such a cloak was needed by women in attending to their ordinary domestic duties, for the female sex amongst the Maori has been always more careful about exposure than the males. Just such a garment was the parakiri of the Whanganui tribes, worn by women in their household work and utterly devoid of ornamentation. Though the recent ones had the two-pair weft, the earliest must have been made with the single-pair twine.

Improved weft.—With the development of finer dress cloaks, an improved weft technique was discovered in the two-pair interlocking weft. Warp material was washed, pounded and rubbed between the hands to render it as soft and as white as possible. Finer warp threads were made by rolling on the bare thigh, and the weft rows were worked closer together.

Ornamentation.—The question of ornamentation now came into consideration. The principle of attaching material to the outer side of the warp by inclusion under the weft

FIG. 146.
Evolution of rain-tag fixation., A—pota cape, B—rain-cloak, C—toi cape
- 137

stroke had already been established in the rain-capes. Tags and ornamental elements being on the outer surface of the garment, necessitated that surface being toward the worker. The appearance of the fixation of rain tags on this surface is shown in Fig. 146. The more secure method of fixing the toi tag in C was primarily influenced by the necessity of holding the stiff toi fibre back in position.

The attachment of ornamental tags followed these methods as shown in Fig. 147. The fixation of the flaxen rolls of korirangi cloaks is shown in A and C, the ornamental cords of tag-cloaks in B, D and F and feathers in E.

FIG. 147.
Evolution of ornamental tag attachments., A and C—flaxen roll, B, D, and F—cord tags, E—feather

The Figure explains itself, for whilst the two-pair weft is used the fixations are all based on the rain tag fixations of the previous Figure. The exception is the more recent development in tag-cloaks shown in F. The necessity of keeping the ornamental elements out of the way of the work is obvious in the case of the stiffer feathers. This led to adherence to the rain-cape and rain-cloak method of commencing at what was to form the bottom border. In the case of the limp tags and cords which would fall downwards in any case, the same position of commencement was influenced by the fact that it was easier to make the more elaborate neck band last.

Rapa rain-cape.—Before leaving tag fixation, it is necessary to remember that the Rapa Island rain-cape found by Mr. Stokes in a mortuary cave, had the tags fixed in the manner shown in Fig. 146c.

Commencing edge of garments.—From the discovery that the two-cord commencement was used in the Otago tussock cape, it seems probable that this method was - 138 brought over from Eastern Polynesia and was the first form of commencement used in New Zealand. With the progress of time, the more simple technique of the thrum commencement was found quite adequate. With finer warps, a more finished appearance was given by the selvedge commencement, which supplanted the thrum method in the finer dress-garments except in some tag-cloaks where the warp ends were used to augment an ornamental fringe. See Fig. 148.

FIG. 148.
Evolution of cloak commencement., A—Two-cord commencement, tussock rain cape., B—Thrum commencement, rain cape and cloak., C—Selvedge commencement, dress cloaks.

Neck borders.—Whilst simplicity of technique was required with the first commencing weft row, no such restriction was needed with the warp ends left over at the dependent neck border. Again the Otago tussock cape provides the simplest type of neck finish where the warp ends are simply turned back in the neck fringe, Fig. 138. This was most probably the earliest form of neck finish. The possibility of utilising the warp ends to better advantage to form a definite neck band must have early occurred to the craftswomen and the single-pair twine, Fig. 128, and the three-ply braid. Fig. 129, were evolved in connection with rain-capes and rain-cloaks. With the finer warp ends of dress-cloaks, a further development occurred in the singlewarp spiral, Fig. 126, and later in the three-element twist, Fig. 130, and the four-element twist, Fig. 133.

Minor ornamentation.—With the development of the better types of cloaks, the artistic taste of the Maori found further expression in additional ornamentation to the side and lower borders. The principle of attachment to the outer surface of the warps by inclusion under the weft stroke was again brought to bear in the minor ornamentation by means of short tags, loops, spirals, circles, ovals and twists. - 139 This closer work was arranged into panels along the side borders and the bottom edge. It was called paheke by the Whanganui and according to them preceded the taniko borders. The minor ornamentation of coloured twists was also requisitioned to finish off edging and to conceal joins.

Protective fabrics.—With the increasing importance of military campaigns came the need for a protective fabric. The Maori believed in hand-to-hand fighting. He used no shield but parried with his weapon or side-stepped the blow or thrust. It was useful, however, to have some body protection against the thrust that eluded his guard. Hence the evolution of the war cloak with its technique of close single-pair twining. It was a very simple step from the spaced single-pair twining in everyday use. Further protection was afforded by attaching strips of dogskin to the outer surface of the war cloak to form the dogskin cloak, kahu kuri.

Decorative borders.—Again the artistic desire for decorative effect along the borders of dogskin cloaks, led to further invention. The simple paheke technique of minor ornamentation was not suited to the close twined technique of the body of the dogskin cloak. A twill technique, probably adapted from kilt bands, provided a decorative band that was used with some success, Plate 33c. Here the Maori reached the verge of true textile weaving.

Limitations of single-pair twine.—A further attempt to provide a narrow decorative border for dogskin cloaks by using colour with the single-pair twine, resulted in a type of border shown in Fig. 20c and Plate 29b. The effect was produced by having one coloured element in the single-pair weft. With the true single-pair twine, where each element makes a half-turn round each warp, the same colour must show on alternate warps as in Fig. 149A. This was the only result possible. The Maori with a taste for diagonal lines, probably establishel by the arbitrary diagonal lines in the coloured designs of the older craft of plaiting, arranged the thin coloured lines in the above technique to run diagonally as in Fig. 20c. For further development, he had to change his technique.

Development of wrapped twine.—It must readily have occured to the seeker that by keeping the same weft element on the same side of the warps, the coloured lines could be made broader. This was easily effected by twisting the anterior weft element round the posterior element and - 140 bringing it forward again to pass over the front of the next warp as in Fig. 149B. This we have alluded to as a full twined weft, Fig. 35c. A change in the relative positions of the weft elements could be effected on any warps by reverting to the half-turn for one stroke.

FIG. 149.
Evolution of the wrapped twine and the taniko., A—single-pair twine, half-turn, B—single-pair full twine., C—wrapped twine, D—taniko twine

In order, however, to keep the colour of the anterior weft element distinct, it was necessary to keep the posterior element from coming forward in the warp interspaces. This again was easily done by keeping the posterior element taut whilst the anterior element, after passing round it, was being pulled toward the worker to tighten the stroke. With the posterior element taut or rigid as shown in Fig. 144c, the full twine has changed into a wrapped twine which is the basis of the taniko technique.

Development of taniko.—The above is the technique, in its simplest form, of the taniko band shown in Plate 31a and in Fig. 30, where there are two colours which are brought out in turn over eleven warps for a depth of four weft rows with the diagonal direction maintained. By some such simple process we have the evolution of the narrow decorative bands of the dogskin cloak from the single-pair twine of Fig. 20c and Plate 29b to the wrapped twine of the simple taniko in Fig. 30 and Plate 31a. Once the principle of using two colours at will by means of the wrapped twine was grasped, it was an easy matter to introduce additional colours. All that had to be done was to add the additional coloured weft strands to the posterior weft element and - 141 bring out whatever colour was required in exactly the same way as with two elements. The addition of a separate twisted white strand to the posterior elements for pulling to keep the posterior elements rigid, completed the evolution of the simple wrapped twine into the true taniko technique shown in Fig. 149D.

Data on taniko origin.—The following data support the contention that the evolution of taniko borders is intimately associated with the close single-pair twining of dogskin cloaks. Many tribes call the close twined work of dogskin cloaks taniko, and maintain that the name was afterwards applied to the technique of the decorative borders now known as taniko.

Again, owing to his contention that the migration of 1350 contained few women and that men have been known to practise weaving, Macmillan Brown held that weaving was introuduced from Eastern Polynesia by that ethnic wave. Apart from the fact that we can find nothing beyond the rudiments of the single-pair twine in the islands from which the Maori ancestors came, there are two important points against this theory. Men also plaited the fine floor-mats with coloured designs, but not the common varieties. From this it might as well be argued that the men of the same wave introuduced the ancient and wide-spread craft of plaiting. Which would be absurd. Furthermore, men did not go through the drudgery of making complete garments. They confined their attention to putting the finishing touches on dogskin cloaks by attaching the greenhide strips and to working taniko bands. Men were the art designers. They carved in wood, bone, and stone, painted the scrolls on wooden rafters, tattooed the intricate designs on the human face and body, and stitched the latticework of decorative house panels. The author's female relatives were at first unwilling to teach him the details of plaiting technique, because they said it was women's work and beneath a man's dignity to do the actual plaiting. At the same time, they admitted that the old chiefs of the tribe used to plait the finer floor-mats with coloured designs. Similarly the making of garments was the work of women, but the men practised the part that had to do with coloured design, namely, the taniko. Thus the men of 1350 neither introduced plaiting nor weaving; but the men of subsequent generations, because they were the artists, had much to do with anything pertaining to the technique and designs of the finer arts. It - 142 is reasonable to infer that men, in finishing off the dogskin cloaks, used their greater art experience in inventing a technique to provide a further decorative finish for the borders of those cloaks and so evolved the taniko technique.

Evolution of motives in taniko designs.—With the perfection of the taniko technique, a new field in decorative art was opened up. By increasing the depth of the bands, new art motives were possible, but the technique of working on the small squares made by the weft crossing the warp, limited the field to geometrical figures. Thus triangle and chevron motives were produced, and, by further increasing the depth of the bands, base to base triangles became lozenges. The evolution of the taniko motives is shown in Fig. 150. A, B, and C are designs worked on the very narrow borders of dogskin cloaks; A, the narrow diagonal parallel lines worked with single strokes of the single-pair twine; B, the wider diagonal bands worked with eleven strokes of the simple wrapped twine; and C, the change to a triangle motive. D, E, F, G, are variations of the triangle motive with internal enhancement more easily produced on the wider side bands of plain cloaks. In D the triangle is enhanced by parallel narrow bands; in E by repetition of the side lines of the triangle, giving a chevron effect; in F by inversion of the chevrons; and G by repetition of the triangle motive. It will be noted that the bases of the triangles must be alternately placed on the two borders of the decorative band in order to balance the design. H, I, and J show a further development of the motives on the still wider borders on the bottom of the plain cloak. H is a simple transition from G, for, by working two triangles apex to apex, the next two must come base to base and result in the lozenge motive. In the lozenge in H, the two small internal white triangles come apex to apex and thus give an hour glass effect which was a favourite motive with the Whanganui. I shows the continuous chevron, which is placed after the lozenge motive because the author has never seen it used on the side borders of cloaks but always on the wider lower border. As the triangle is the almost universal motive of the simpler designs of the side borders, it would seem that the continuous chevron is of later origin than the triangle. As the lozenge is directly derived from the triangle, and the continuous chevron is confined to the more elaborate lower borders, the chevron is placed after the lozenge in the order of evolution. The - 143 Maori artistic sense being against bare spaces, small triangles are used along both edges in I to break the monotony in the wide spaces between the points of the continuous chevron. The double depth of the lower border in J permits further variety in using two motives. J is really a combination of two borders with their individual designs. K, L, M, and N show the introduction of narrow bands to separate the design motives. In K the bands are plain with an equal number of strokes on each weft row. In L the dividing bands are improved by introducing narrow lines of another colour running in a direction opposite to that of the band. Here also a different lozenge enhancement is effected by combining two triangles of the variety shown in E. In M more variety is introduced by forming the band out of small triangles, whilst in N a still more artistic effect is produced by using smaller triangles in sets of three on each weft row. In M the large triangle motive is internally enhanced by smaller triangles along the right border combined with the parallel line treatment shown in D.

By using combinations of the above main motives and working them in black, white, reddish-brown and yellow, pleasing and artistic designs were worked that were highly prized. The principle already stated of running the lines diagonally was adhered to throughout. The geometrical figures were evolved with the technique and were never meant to represent anything in nature. The late Dr. Brigham 28 of the Bishop Museum, Honolulu, suggested that the triangles worked on a taniko satchel that he obtained in Taranaki were meant to represent the perfect volcanic cone of Mount Egmont but the triangle motive is universal amongst the tribes of New Zealand who have no Mount Egmont. The Maori evolved his design first and then gave it a name. The names of the designs vary with the different tribes. Von Steinan's theory of natural motives degenerating into geometrical figures does not apply to the taniko designs of the Maori.

Special cloaks to show off taniko bands.—The close technique of taniko made the fabric very stiff and rendered it unsuitable for the body of a dress-cloak. It remained purely decorative for the embellishment of the borders of dress-cloaks. To show the borders off to the best advantage,

- 144
FIG. 150.
Evolution of taniko motives.

special cloaks, with plain bodies made with the dress technique of spaced two-pair wefts, were evolved. External ornamentation of the body not being desirable, the established technique of commencing at the lower border was - 145 departed from. A taniko band being stiff and rough to the skin, they were confined to the sides and lower border. It was only in the dogskin cloaks, which were stiff throughout, that a narrow taniko band was used at the neck border. In the special plain body-cloak, it was easiest to commence the garment at the neck border and leave the taniko border for the last. Of the modes of commencement, the selvedge form gave the better finish and was accordingly used. The more elaborate neck bands characteristic of the other types of dress cloaks were not used, as they were a finishing and not a commencing technique. For security and neater appearance, the taniko was never attached to the cloak as separate bands, but the taniko warps were fixed in position during the manufacture of the body as already described. Further evidence of the æsthetic taste of the Maori craftswomen is shown by the fact that the warp material for the body of these cloaks was not subjected to successive poundings, washings, and rubbings as in the other types. This was in order that the yellowish silky sheen of the natural fibre, unobscured by external ornamentation, would show up on the body of the cloaks.

The last type of cloak to be evolved was the paepaeroa, with the weft rows running vertically when worn. This, as we have seen, was commenced with the left border above and consequently evolved a simple type of neck border consisting of a twisted cord over which the ends of the weft rows were knotted. The complexity of the taniko technique and the departure from the orthodox commencement and finish necessitated thereby, prove that the two types of plain body-cloaks with taniko borders were the latest to be evolved.

Taniko a local invention.—Ere leaving taniko work, a few remarks are necessary. The American Indian, in the close twined work of their basketry, produced designs on the outer surface by overlaying with coloured material. Various branches of the Polynesians, both East and West, also produced coloured designs in their plaited mats by overlaying. But what was possible with splints or broad flat strips was not feasible with the strands of separated fine fibres that formed the weft elements of Maori garments. Overlaying could not be used in the twined work of Maori - 146 fabrics. Ling Roth 29 states that wrapped twined work was known in Tahiti and probably other parts of Polynesia. He thus inferred that Maori taniko work was a diffusion from that area. As against that, the author has been unable to find any trace of the wrapped twine technique in any of the Maori crafts outside of taniko work. If, for the sake of argument, the Maori was acquainted with wrapped twine technique in Eastern Polynesia, how did he keep the memory of it alive until generations later it came in useful as a technique for the locally developed decorative borders of a dogskin cloak? He had no written records whereby the forgotten technique of the past could be revived. Traditional records were handed on orally from generation to generation, and the technique of the crafts had to be taught personally down the ages. Without direct instruction, a forgotten technique may yet be revived if objects made with that technique are in existence. Surveying the whole field of the Maori crafts, no object exists, or seems likely to have existed, that was made with a simple wrapped twine technique. Where a similar technique exists in widely separated areas even occupied by the same race, chronological sequence must be established are diffusion can be proved. Where chronological sequence can be disproved, then a technique, no matter how similar, must be regarded as a case of independent evolution. We feel convinced that the wrapped twine technique of the Maori was derived from the single-pair twine of the coloured border of dogskin cloaks as shown in Fig. 149, and that therefore taniko work is an excellent example of independent evolution in New Zealand.

Shaping of clothing.—Some time during the development of the dress cloaks, short weft rows were put in to correct or balance the work, and later to shape the garments to the contour of the body. These interpolated rows became arranged into sets forming elliptical inserts to correspond mainly to the bulge of the shoulders and of the buttocks. With the development of the cloak worn with the weft rows vertical, the interpolated rows in these cloaks were arranged into wedge-shaped inserts to suit the particular circumstances of manufacture and position of wearing. These latter cloaks being a later invention, it follows that the wedge-shaped insert is of more recent development than the elliptical.

- 147
CONCLUSION.

We have described the technique of Maori clothing from the simple kilt to the complex dress-cloak with taniko borders. From a careful study of that technique with its details, we have endeavoured to follow the order of its evolution. We have sat beside Maori craftswomen and watched deft fingers separating the weft elements, placing the warp between and pulling the weft stroke home with a click. We have gone from village to village of the surviving experts and, on a sampler, been taught the intricacies of the neck bands and the complications of the taniko. Details and processes that might have been curtailed to more able ethnologists of another race, have been freely and patiently imparted to a kinsman. Saturated with a Maori atmosphere and environment, one could almost feel and share the mental processes of successive generations of crafts-women as they groped after the solution of the various problems in technical detail that assailed them during the evolution of the craft. The details in actual technique were given by Maori experts and are correct. The deductions are the author's, and, if they are wrong, the mistakes are his own.

In endeavouring to follow the evolution of Maori clothing technique, we have been influenced and guided by three things. Firstly, the simplest technique is found in the simplest garments. The crude combined rain tag and warp is found in the most primitive type of cape used for the roughest purposes. The single-pair twine, the thrum commencement, and the braid finish are shared by it with the improved rain-cape and rain-cloak. In the dress-cloaks, in addition to more care in the preparation of material, finer warps and closer weft rows, a more complicated technique is used, such as the two-pair weft, the selvedge commencement, and the more complicated neck bands. Ornamentation progressively improves.

Secondly the order of complexity in technique coincides with what we regard as the order in which the need for the various garments occurred during the period that the Maori was perfecting the clothing craft; kilts, rain capes, rain-cloaks, dress-cloaks, protective garments, and still more decorative dress garments.

Lastly, the use for particular garments did not cease when a superior garment with an improved technique was - 148 evolved. The Maori was not unduly conservative, but he seems to have been happy in having evolved a variety of technique appropriate to the particular order of his needs. The needs he experienced during the evolution of the craft remained the needs that had to be satisfied after the perfection of his technique. Improved technique did not lead to the abandonment of the simpler technique of the rougher garments, but added a richer variety to his forms of clothing. Thus the various forms of garments, with their particular technique remained in use to point the way down which his inventive genius had travelled since his ancestors cut off the sea-roads to Hawaiki.

Our Polynesian material has been meagre. From such as we had, we have concluded that the use of the spaced single-pair twine was brought to New Zealand from Eastern Polynesia. Its more extended use in the rain-cape and rain-cloak and in close twined work was stimulated by local conditions. The fixation of the rain tags in the Rapa cape would seem to be a case of independent evolution created by needs similar to those of the Maori. The fact that there was apparently no such garment in the Society and Cook Groups would show that there was no need for such a technique in the area from which the Maori ancestors came. At the same time, owing to the closer relationship between Maori and the dialects to the east of Tahiti, we must not overlook the possibility that the peoples now occupying remote areas such as Rapa and New Zealand may have been in very close connection in the Society Group before they were displaced by the ancestors of those who now occupy what was the original distributing centre of the Eastern Polynesians. Such a hypothesis would account for resemblances in the material culture of the marginal areas to the east and south being closer to one another than to that of the nearer central area. The data recorded by the late Percy Smith and others dispose of the possibility of the Maori having come from islands to the east of Tahiti. However, leaving this doubtful case for further investigation, we may definitely regard the two-pair weft, external major and minor ornamentation, feather work, the commencement and finish of dress-cloaks, the dogskin cloak and taniko work, as being cases of independent evolution influenced primarily by the changed environment of New Zealand. With the varied technique evolved, the Maori was - 149 enabled to produce a variety of garments and develop a range and taste in textile ornamentation that compares more than favourably with that of any people living in the new stone age.

HEOI, KA MUTU.

FINIS.

1   H. D. Skinner, 1923, The Morioris of Chatham Islands, Memoirs Bishop Museum, Honolulu, Vol. IX., No. 1, pp. 129, 130.
2   A. Shand, 1911, The Moriori People of the Chatham Islands, Polynesian Society Mem., Vol. 2, p. 8.
3   W. Baucke, in conversation.
4   H. D. Skinner, op. cit., p. 108.
5   H. D. Skinner, op. cit., p. 108.
6   H. D. Skinner, op. cit., p. 110.
7   Op. cit., p. 8.
8   H. D. Skinner, op. cit., p. 108.
9   H. D. Skinner, op. cit., p. 108.
10   H. D. Skinner, op. cit., p. 109.
11   A. Shand, op. cit., p. 8.
12   A. Shand, op. cit., p. 8.
13   H. D. Skinner, op. cit., p. 111.
14   Captain James Cook, A Voyage to the Pacific Ocean. …in the Years 1776, 1777, 1779, and 1780, in three vols; Vol. 2, p. 324.
15   Op. cit., p. 280.
16   Op. cit., p. 325.
17   Op. cit., p. 304.
18   Clark Wissler, 1922, The American Indian, New York, pp. 53-55.
19   James Teit, 1900, Memoirs, American Museum of Natural History, Vol. 2, Pt, IV., p. 189.
20   Op. cit., p. 212.
21   Op. cit., p. 243.
22   Op. cit., p. 219.
23   Op. cit., p. 55.
24   Ling Roth, 1923, The Maori Mantle, p. 112.
25   A. Goldenweisir, 1921. Early Civilization, Harrop and Co., p. 68.
26   Op. cit., p. 33 and p. 30.
27   Op. cit., p. 112.
28   W. T. Brigham, 1906, Memoirs Bishop Museum, Honolulu, Vol. 2, No. 1, pp. 92, 93.
29   Op. cit., p. 37.