A considerable number of odd or abnormal forms are met with in collections of Maori stone adzes. It is fairly clear, from the evidence of roughly fashioned, unground specimens, and of half-ground ones, that such oddities of form have often been caused by the peculiar shape of the rough stone, whether water-worn stone or a piece of a broken boulder. Some stones seem to break into triangular forms when broken up by hammering. Again, when a stone is so battered or hammered it often splits along a natural line of cleavage that leaves one face or side of a peculiar shape, possibly a concavity, that is not wholly removed by reducing the ends or by grinding. Some of these odd forms will now be described.

Subtriangular Form

Chisel-point, but bevelled apparently on face only. This curious specimen is not definitely triangular, but has flat sloping sides that merge imperceptibly into a rounded back. Thus the cross-section might perhaps be more appropriately styled semicircular than subtriangular. The peculiarity of this specimen lies in the fact that the bevel, or facet, that forms the cutting-edge is on the curved face of the tool. This would make it a very difficult matter to secure the implement to a haft, unless a semicircular groove was gouged out of the foot of the handle, and in which the tool might be fitted prior to being lashed on. We notice this peculiarity in one of the sharply triangular forms described elsewhere. In only one case out of a great many adze-handles observed have we noted any interference with the foot of the handle. They are almost invariably flat, or, as occasionally seen, with a rectangular shoulder. But it must be remembered that these handles do not date from the stone age of New Zealand, which may be said to have passed away by about the year 1830, but were made for lashing steel blades thereto, often plane-irons.

This specimen (Fig. 82, Plate XVII) is 7 in. long, and 2 in. wide at the cutting-edge, whence it narrows towards the poll, which is page 2951¼ in. wide. It is slightly over 1 in. thick at the shoulder, and becomes thinner towards the butt end. Its weight is 1 lb. The face is straighter (flatter) longitudinally than is common, and is convex transversely at the cutting-edge and for some distance back from it. Thus it was what Evans terms a "bastard gouge." But to fasten this tool on to a handle would be a puzzle, if it were to be used as was the ordinary adze. The back of the tool being semicircular, it would not lie flat on the face of the haft for lashing, and hence it would be difficult to secure it firmly. If let into a groove in the foot of the handle the difficulty would disappear. The butt end of the back has not been flattened in order that it might lie better on a flat surface, but has, on the contrary, been chipped so as to render it more triangular in section than the other part. If hafted in the usual manner, with the flat face against the handle, the tool would be a most awkward one to work with, for this would bring the face with the shoulder and deep bevel in contact with the timber being worked, and the prominent shoulder would interfere with the work. Moreover, the curve of the cutting-edge would be upward, and this would make it an impossible sort of tool to work with. The two corners of the cutting-edge would dig into the timber being worked, and prevent a chip being taken off. The cutting-edge has been considerably chipped, and never reground. This specimen is a very odd form, and in order to make it a useful tool there must have been some such accommodating peculiarity about the handle as that described above. As to the cause of this odd form, it seems to have been entirely due to the shape of the stone when struck off from the boulder. The tool is but partially ground, and shows much of the bruising or hammering process by which the surface was prepared for grinding. The flat face shows an undoubted natural line of cleavage, and it is clear that the stone was a piece struck off a boulder having a flat face and a rounded one, the one end being thicker than the other. At this thick end it fell away abruptly in thickness, thus giving a naturally formed bevel to form a cutting-edge. All the workman had to do was to chip the sides down to a desired form and commence grinding operations.

Another specimen of a similar nature, but with a concave face and only roughly chipped, is undoubtedly a spawl struck off the face of a water-worn boulder, the cause of the longitudinal concavity of the face being a flaw in the stone, along which flaw the stone split. The tool has been badly chipped, and is probably a reject. The back is convex longitudinally to a marked degree, and has been chipped at the lower end to form the blade. Had the blade been ground it would have been a very thin one.

page 296

In Fig. 82a, Plate XXIV, we have an implement of curious form and fine finish, one of the most interesting items in the Buller Collection. A cross-section of this tool might be termed subtriangular, but approaches an ovoid form. The sides have been so worked down as to be represented merely by somewhat sharply defined edges, which implies that both face and back are transversely convex to an unusual extent. This tool is 8 in. long, 1⅞ in. wide across the cutting-edge, whence it narrows to 1¼ in. at the poll. Thickness in centre, ⅞ in. Weight, 12 oz. Material, black aphanite. The form is a peculiar one, owing to a marked longitudinal curve of the tool from a side view, the face being convex longitudinally to a most marked extent, and the back concave in a like direction. The blade-bevel on the back is 2¼ in. long, and markedly concave transversely. The angle of the same is low, from 40° to 20°. This tool must have been used for light work only, and, on account of its form, would cut a groove or channel like that formed by a shallow gouge. The butt end has been reduced on the face and sides for the lashing, and the poll is rough but all the rest of the tool is ground to a very fine smooth surface, save a strip on the back, which the grinder has not touched on account of that part being slightly hollow.

In a few cases the face is concave longitudinally from the poll to the beginning of the blade part. Longitudinal concavity is, however, more frequently seen on the back of the tool, where it extends from the shoulder to the poll. In cases where one face, front or back, is concave, it is usually noted that a marked convexity occurs on the opposite face. This latter feature accentuates the concavity to the eye of the observer, and represents a peculiar and uncommon form. Such a shape may have been caused by the peculiar cleavage of the stone of which the implement was made.

Petaloid Form

Oval section. An unusual type of stone adze is represented by several specimens of an almost petaloid form, reminding the observer of West Indian specimens, and also of several British examples described by Evans. Of the few New Zealand specimens of this form, however, none are so sharply pointed at the poll as is fig. 52 in Evans's work.

One of this type in the Museum is thought to have been found at the Chatham Islands, the stone being decomposed feldspar. In cross-section it is almost a perfect oval, and carries a curved cutting-edge (see Fig. 83, Plate X). It has an adze-like blade, and has been ground over the whole of its surface, which, however, seems to be page 297much weathered. It is widest at the cutting-edge, from which the rounded sides curve convergently to an almost conical poll. Length, 7 in.; width, 2¾ in. across cutting-edge; thickness, 1⅜ in. Weight, 22 oz.

Another specimen (Fig. 84, Plate X), a local one, is less petaloid in shape, the rounded sides not being so convex longitudinally, nor is the poll so pointed. The material is mudstone. The face also is somewhat flattened on the lower part of the tool, hence that portion does not show an oval section, though the upper part does. The blade is of adze-form, the cutting-edge somewhat curved, and slightly oblique. The length of this specimen is 6½ in., and it is 2⅜ in. wide at the shoulder, its widest part. About one-half of its surface has been ground. Weight, 16 oz. Another tool of similar form is 8½ in. long. In specimens where the transverse convexity is considerable and persistent on the face, it follows that its intersection with the bevel-facet of the blade produces a curved cutting-edge, like that of a shallow gouge. (For another example see Plate LI.)

The truly conical poll is seen in a specimen in Mr. A. H. Turnbull's collection. This tool has the sharpest, most pointed poll seen in any of the hundreds of stone adzes under inspection. It is most unlike the ordinary Maori form. The wide straight cutting-edge detracts from its petaloid appearance, but were it somewhat rounded the tool might well pass for a West Indian implement, or a European type mentioned by Sir John Evans. This specimen is well formed, but only partially ground, and the cutting-edge is broken. Its length is 7 in. Width across cutting-edge, 3 in., from which its straight rounded sides taper off to the pointed poll. The face is convex longitudinally and much rounded transversely. At the butt end a cross-section would be almost cylindrical. A slight ridge has been left 2½ in. from the poll. The back is straight and flat, the blade very thin, and carries an angle of 30° near the cutting-edge and of about 20° higher up. It would be of interest to know how this tool was hafted.

In the same collection is the butt end of another peculiar form, showing a cross-section somewhat pyramidal in form. The sides slope inwards to a narrow flat face, which is thus much narrower than the back. This flattened face runs out to a point at the butt shoulder, 2½ in. from the poll, on which shoulder-point is a curious little knob, from which point the face has been much pecked down to near the poll, whereon another, but smaller, knob has been left.

In Fig. 85a, Plate XXII, we have a large specimen with an almost oval cross-section throughout the greater part of its length. The back and sides bear out the ovoid aspect, but the face is flattened, save at page 298the butt end, whereat it is much rounded. This tool is 13¾ in. long, 4 in. wide across the cutting-edge, narrowing to 2⅜ in. across the poll. Thickness, 1¾ in. Weight, 6¼ lb. Material, fine black aphanite. So rounded are the sides of this item that they present no resemblance of an edge or ridge, and no marked shoulder-line exists. Angle of blade, about 45° near the cutting-edge, dropping to 30° higher up. The blade and face are well ground, and present a smooth surface, albeit faint striæ caused by grinding operations are visible at several places. On the sides and back grinding had just been commenced, those parts presenting surfaces reduced in a very even manner by bruising rather than pecking. At the poll, on the face side, are two slight projections that would serve to contain the lashing.

One of the most interesting types of stone adzes noted is a curious subtriangular form from the Chatham Islands, which are situated 536 miles east of Lyttelton, in the South Island of New Zealand. The specimen here described is in the collection of Mr. A. H. Turnbull (see Fig. 85b. Plate XXI). It is a small item, but well formed and extremely well finished, every part, including the poll, having been ground smooth. Its only blemish is a small fracture on the face. This item is essentially an adze, and is 6 in. long and 1 13/16 in. wide across the cutting-edge, from which point it decreases evenly in width back to the poll, where it shows a width of 1¼ in. on the face and ¾ in. on the back. Its thickness at the shoulder is ¾ in., which is preserved backward to within 1¼ in. of the poll, where the face has been reduced in order to facilitate lashing to a haft. The face is, as is usual in such forms, convex longitudinally and transversely; the back is straight longitudinally, but very slightly convex transversely. The sides are slightly concave longitudinally, but to so small an extent that it is scarcely noticeable until a straight-edge is applied thereto. The sides slope inward from face to back to a marked extent, so that the latter is ½ in. narrower than the face. Thus a cross-section would appear as somewhat of a truncated pyramid-form. Had the tool been ¾ in. thicker its form would have been triangular in section. One side is slightly thicker than the other, which destroys absolute symmetry of form, but such are matters of close observation. The blade shows an angle of about 50° on its lower part, and is well finished, the length of the blade from cutting-edge to shoulder being 1¼ in. On the face a short bevel of ¼ in. has been ground to form the cutting-edge, an unusual feature, and it may be the result of grinding out a former gap in the edge, which is keen and flawless. The chief peculiarity of this tool is at the poll, whereat two horns or projections have been left, one at each face corner, in order to contain the lashing. The butt page 299end of the face has been ground down for a length of 1¼ in., with the exception of these two lugs, which would certainly much assist the retention of the tool by the lashing, albeit an unusual feature. The hollow space between these lugs can only have been formed by rasping out the material between them with a narrow piece of sandstone, as we use a file or rasp. The material is a fine-grained black stone, and the whole of the tool has been carefully ground to an even surface, but has later been exposed to drifting sand which has slightly affected the surface, except the blade, in a curious manner, imparting to it somewhat the aspect of a piece of worm-eaten wood that has been planed so as to expose slight channels or grooves. The weight of the tool is 9 oz.

The Dominion Museum possesses a cast of an unusual type of adze in the Otago University Museum that might be termed either subtriangular or ovoid in cross-section, for it approaches both. This implement has been ground except the poll, which is rough (see Fig. 85c, Plate XXX). Its length is 15 in.; width across cutting-edge, 4¼ in.; at butt end, 2 in. Thickest part, at shoulder, is about 1¾ in. The unusual thinness of body and blade for so large an item show that it was not used for heavy work. The sides of this tool are merely edges, to which the face and back fall away. The more pronounced transverse convexity of the back prevents a cross-section being truly ovoid. This pronounced roundness of the back is continued to the poll; hence we infer that the foot of the handle must have been hollowed somewhat, in order to accommodate itself to the form of the adze. The face has been slightly reduced for 3½ in. at the butt end, to facilitate lashing. The blade is 4 in. from shoulder to cutting-edge, and is markedly concave transversely, being also unusually thin. This is the form of adze that was used in finishing off the interior of a canoe, &c., the result being not a plane surface, but a series of long, parallel, shallow grooves or channels. This ovoid form is unusual. This tool is from the west coast of the South Island, the material being a grey stone.

The New Zealand forms mentioned above much resemble in outline fig. 68 in Evans's "Ancient Stone Implements of Great Britain," 1897 edition, and also in section, though the form of blade differs from English types. The expansion of Evans's fig. 68 at the blade is not met with in the New Zealand specimens.

Wide Cutting-edge in Proportion to Length

There does not appear to be any New Zealand form that expands abruptly at or near the cutting-edge, as illustrated by Evans and page 300other writers, European and American. But there are in the Museum a few specimens that are unusually wide in proportion to their length, and the widest part is at the cutting-edge. One such is a beautifully finished adze of hard black stone (black veined aphanite) (see Fig. 86, Plate X). It is 7½ in. long, and slightly over 3½ in. wide at the cutting-edge, but only 2 in. wide at the butt end. The sides are convex longitudinally and transversely, as also are the face and back, but not to any marked extent, hence the longitudinal edges are quite sharply defined. The blade is of adze-form, having no prominent shoulder-limit. The cutting-edge is very slightly curved, and the whole tool has been carefully ground and polished right back to the poll. The only flaw is one in the stone, on the back of the implement. Why so much trouble should be taken to carefully smooth the portions of the surface concealed by the lashing it is hard to say. The angle of inclination of the blade is about 50° near the cutting-edge, and 30° higher up the blade. The weight is 24 oz.

The Museum contains a cast of a very fine stone adze in the Auckland Museum. It was found at Karamea, in the Nelson district. The finish of this tool is superb. The only part that has not been carefully ground and smoothed is the poll, or extremity of the butt end. It is also exceedingly symmetrical, and has been very carefully fashioned (see Fig. 86a, Plate XXXII). It is also remarkable for the width of its blade, and for the most prominent transverse ridge at the shoulder-line of the back of the tool that we have noted on any specimen. The material is black aphanite. This adze is 9 in. long, 3| in. wide across the cutting-edge, narrowing back to 2⅛ in. across the poll. Thickness behind shoulder, 1⅛ in. The sides are somewhat convex both ways, as also are the face and back, but not to such a marked extent as is often seen. It looks as if it might be such an adze as was used for the final dressing of house-battens, as described by Te Whatahoro. That it was not used for heavy work is shown by the remarkably low angle of the blade. The back is somewhat narrower than the face, and across it at the shoulder-line is a remarkably prominent supplementary ridge about ¼ in. higher than the plane of the back. To form such a ridge must have involved a considerable amount of extra care and labour. In form and finish this adze is perfect.

Another specimen (Fig. 87, Plate XI) that has a wide blade in proportion to its length is 6⅜ in. long and 3⅛ in. wide at the cutting-edge, but only 1⅞ in. wide at the butt end, the latter measurement being the chord of the arc described by the semicircular poll. The material is diorite. This adze is unusually flat and thin for a Maori implement, while it has a very short and thick blade, the facet forming page 301it being but ¾ in. from shoulder to cutting-edge, and its angle is almost 70°. The thickness of the tool in no part exceeds ¾ in. So flat a shape is seldom met with, the back being almost a plane surface. This item is not so well finished as the one described above. Weight, 20 oz.

The widest-bladed stone adze we have to describe is one in the Otago University Museum, a cast of which is in the Dominion Museum. It is probable that the butt end of this item has been broken off (see Fig. 87a. Plate XXXI). The original is nephrite, and was found 5 ft. below the surface, at the junction of Leith and Hanover Streets, Dunedin. This tool is such a one as described by Te Whatahoro as having been used for the final dressing of house-battens, &c. Its length is 11½ in.; width across cutting-edge, 5⅝ in., from which it narrows back towards the butt end. The thickest part is at the shoulder, where it is 1½ in. The sides slope inward to the back, which is nearly ¾ in. narrower than the face. All surfaces have been ground, save the rough fracture at the irregular poll and a flaw on one side. The cutting-edge is curved, and the blade is convex transversely on the back. From cutting-edge to the prominent shoulder the blade measures 3 in. The whole face, including the blade, is unusually flat, showing but slight convexity either way. The material (nephrite) contains many black veins and blotches.

The peculiar British form described by Evans as having a large knob on the poll is unknown in New Zealand; indeed, it seems to have been an axe, not an adze.

Form with Transverse Ridge at Shoulder-line of Blade

This peculiar ridge across the shoulder-line or top of the blade is not often met with; but there are half a dozen good examples in the Museum. These have prominent ridges, but some other specimens show a small ridge less sharply defined, and this type merges into the common form, which merely has the shoulder, or junction of the blade-bevel on the back with the axial plane of the back, sharply defined as an angle of greater or less obtuseness, but on which no supplementary ridge or raised part is seen. The only reason for the existence of this peculiar transverse ridge on the back of an adze at the shoulder-line, as obtained from natives, is that it served to force the chip upward and detach it from the baulk being worked, and also to prevent the adze-blade from being bound in the cut. This, however, was merely a supposition. It is well known that the uma of a stone adze was carefully designed for that object.

It is curious that this transverse ridge is often accompanied by straight backs and a marked slope inwards of the sides, causing the back of the tool to be considerably narrower than the face. Thus, in page 302a specimen (fig. 88, Plate X) 8½ in. long, 2½ in. wide at the cutting-edge, and 1¾ in. near the poll, the face just behind the shoulder is 2⅝ in. wide, whereas the back is only 2 in. wide on the same section. Also, the back, from the poll to the rise of the transverse ridge, is perfectly straight and flat, not being at all convex either way, a very rare thing, save in very small forms. The sides also are almost straight, indeed very slightly concave longitudinally, and unusually flat transversely. The ridge across the shoulder-line is prominent, and comes to quite a sharp point on top. The thickness at the ridge is 1¼ in., and in the middle 1 in. The tool has been well ground except on the back, and is disfigured only by two gaps in the blade and one near the butt end. Well-defined striæ show that one side has been rubbed obliquely on the grinding-stone. On the other side both oblique and longitudinal striæ are visible. Material, black-veined aphanite. Angle of blade, 45° to 30°. Weight, 1 lb. 14 oz.

Another specimen is 7½ in. long, and has a somewhat thinner blade than the one last described. It is not, however, so symmetrical, a long flake having been struck off one side, apparently in the blocking-out process, which has produced a disfiguring hollow in that side. A longitudinal curve on the opposite side has possibly been formed in order to strengthen the weakened part of the tool. The transverse ridge will be noted in several types.

Wide, Flat, and Thin Form, with Transverse Ridge

This is a lone specimen of black-veined aphanite, and an interesting one (see Fig. 89, Plate XIV). It is 11½ in. long, 4⅝ in. wide at the cutting-edge, whence it narrows to 2⅞ in. at the poll. The face is convex both longitudinally and transversely, but not markedly so. The back is straight, save for a natural hollow (apparently) of the original surface of the stone. The sides slope inwards to the back, and preserve almost the same thickness throughout, from 1⅝ in. to 1 in., the latter figure representing a thinning of the butt end. The transverse ridge is not so prominent as those of the other specimens described, and has a rounded top. It is the only specimen in which the ridge is not situated on the shoulder-line, where the maximum thickness of the tool breaks down into the blade-bevel. In this case the ridge is situated 1 in. back from the shoulderline, which, however, is not a sharply defined one. This specimen is not wholly ground. The angle of inclination of the blade is from 35° to 20°. Weight, 4 lb.

The use or purport of this peculiar transverse ridge is not understood by us. It can scarcely be left for the purpose of page 303strengthening the tool, for it is situated at the thickest and strongest part thereof. Whether the stone-worker of yore left it for use or ornament we wot not, but it assuredly gave him a considerable amount of extra labour to form such a ridge, and much lengthened the period of time occupied in the manufacture of the adze. In this wise: The tool could not be rubbed lengthwise on the grindingrock, or the ridge would be ground down and soon be no more. It had to be rubbed sideways on the grinding-stone, and the parts adjacent to the ridge would probably have to be rubbed down with a hand-stone, a slow process. Some of these forms, indeed, show plainly the transverse striæ that betoken rubbing sideways on the grinding-stone, or vice versa.

Were this ridge situated on the face of the adze it might be thought that it was designed as a sort of fulcrum to assist in forcing off a thick chip or sliver. But they are invariably on the back of the adze, and so situated that they would not come into contact with the timber being worked. The thickness of blade in the stone axe also precludes the idea that the ridge was left for the purposes of clearing the chip. No danger that so thick and blunt a blade would "bind" in timber.

The only idea that occurs to us in regard to the utility of this ridge is that the lower part of the handle may have been butted against it, so that when lashed tightly on it prevented the adze from being forced upward and displaced by the shock of the blow when used. But, apparently, no lashing would have been applied so low down, or it would come into contact with the timber being worked, and thus render a blow ineffective. The utility of the transverse shoulder-ridge is not proven. In those specimens that have the butt end worked down to accommodate the lashing the cord used has not extended more than about 3 in. from the poll.

The supplementary ridge is never seen, says Mr. H. D. Skinner, in any collection of stone adzes from the Taranaki district; and he is of the opinion that it was confined to South Island forms.

Form with Oblique Tang, or Butt End

This is an unusual form among the stone adzes of New Zealand, but a very common type in the Hawaiian Isles, as Mr. Brigham has shown in his "Hawaiian Stone Implements." The advantage of having the tang or butt end set at a considerable angle to the rest of the implement is not obvious. It is highly improbable that any workman would make a stone tool to fit a peculiarity in the shape of the handle, rather would the wooden handle be made to fit the page 304stone tool. It is quite possible that these few local specimens owe their peculiar form to the shape of the blocked-out stone, either to it splitting along a natural line of cleavage or to unskilful chipping and flaking. This, however, is merely conjecture.

One of these specimens (fig. 90, Plate XVI) is 9½ in. long and 2⅝ in. wide at the cutting-edge, from which point it narrows to 1⅞ in. at the poll. The thickness at the shoulder is 1⅝ in., and 1½ in. at the angle where the tang commences. This tang, or butt end, is not so thick, running from 1¼ in. to 1⅝ in. The blade is thick, but well formed and ground, with a flawless cutting-edge. The sides and face of this implement are partially ground, but the back is still untouched by the grinder, being in a roughly chipped state. The back is deeply concave lengthwise, and flat transversely. The face is slightly convex longitudinally from the cutting-edge to the angle formed by the junction of the tang and the main part of the adze. The side view shows the extent of this angle. The stone is a fine basalt, and the angle of the blade-bevel from 60° to 40°. Weight, 21b. 12 oz. The face has been rubbed in an oblique manner on the grinding-stone.

Te Whatahoro states that Figs. 90 91 92 and 93, Plate XVI, resemble implements termed by east-coast natives toki waro, or mata waro, or toki tata waro, that were used to chip off charred wood when fire was used as an aid to tree-felling, &c. Also, that implements such as Figs. 90 and 93, in which the lower part of the face is comparatively straight and flat, were used as toki hema for bevelling off timbers and adzing plane surfaces. Intensified convexity of the face would render a tool unsuitable for such work.

Another of these odd forms is of a still more curious shape (see Fig. 91, Plate XVII). It is 7⅝ in. long and 2⅛ in. wide across the angle, its widest part, whence it narrows to 1¼ in. at the cutting-edge, and to the same width at the poll. The back of the tool is almost straight longitudinally, being very slightly concave, but it is much curved transversely, hence a cross-section would be semicircular, and the bevel to form the blade is a triangular facet. The face is convex transversely to a marked extent throughout its whole length, from poll to cutting-edge, which implies that the latter is curved like a shallow gouge. The angle on the face side is quite pronounced, as seen in the side view. The implement has been worked smooth, save where in some places rather deeply chipped spots have not been entirely ground out. The stone is aphanite.

This specimen is of a very rare shape, and, though apparently hafted as an adze, might rather be said to do the work of a chisel or page 305gouge. The angle of inclination of the blade is about 40° near the the cutting-edge but, as usual, much less higher up the blade. Weight of specimen, 12 oz.

There are also two of these angular-tanged adzes in an unfinished state, being half-ground. One of these is 9½ in. long, and weighs 2 lb. (see Fig. 92, Plate XVI). Width across the somewhat curved cutting-edge, 2¼ in.; across poll, 1½ in. The face is somewhat convex longitudinally from the cutting-edge to the break or angle of the tang, also slightly convex transversely. The back is markedly concave longitudinally, but almost flat transversely. The face is narrower than the back by ¼ in., an unusual occurrence. The tang lies at an angle of about 20° to the face of the tool, but is much straighter on the back. Thickness, 1⅜ in., though the tang averages but 1 in. in thickness. The blade is slightly concave transversely in its lower part, and the keen cutting-edge would thus cut as a shallow hollow gouge. Angle of inclination, 45° near cutting-edge, 30° on upper part of blade. The stone is aphanite.

Another less decided example (see Fig. 93, Plate XVI) has the tang at an angle of 10° or 12° to the face. This specimen is 8¼ in. long, and weighs 2½ lb. It is 3 in. across the cutting-edge, 2⅜ in. across the shoulder, 1⅞ in. at the middle, and 1¾ in. at the poll, thus showing an unusual expansion at the blade. Thickness at shoulder, 1½ in., which, with the short bevel, means an extremely thick blade. Angle of inclination near cutting-edge, 60°, but less higher up the blade. Material, mudstone with inclusive crystals. The last three width-measurements are across the back, which is somewhat narrower than the face. Both the last two specimens are but half-ground.

Yet another implement with an oblique tang, and of remarkably fine shape, lines, and finish, is represented by a cast in the Museum (see Fig. 94, Plates XIV and XV). The original must be a very fine tool. Its length is 13½ in. Width across cutting-edge, 3⅝ in., and at poll, 2⅛ in. Thickness at shoulder, 1⅞ in.; in middle, 1⅝ in.; at butt end, 1⅜ in. to 1⅛ in. From the butt-end shoulder to a point 2 in. from the cutting-edge the face is somewhat concave longitudinally, and is convex transversely throughout its entire length. From a point about 1½ in. from the cutting-edge the face curves slightly, to form that edge by its junction with the long bevel of the back. At the butt end for 3¼ in. from the poll the face and sides have been worked down deeply for the lashing, leaving a deep and unusually abrupt shoulder. The back of the tool is concave longitudinally to an extent but seldom seen, and convex transversely. One side is straight longitudinally, the other somewhat concave. Both are almost flat transversely, and page 306incline inward to the back to a marked extent. Thus the back is very nearly 1 in. narrower than the face. The poll is almost flat. From the highly prominent shoulder to the cutting-edge is a long curve of 4¾ in. This long blade is one of much strength, on account of the massive shoulder. The transverse convexity of the face imparts the usual slight downward curve to the cutting-edge, as seen in steel adzes, and which has the desired effect of preventing the corners from catching, or sticking, in the surface of the timber being worked, or marring its dressed face.

The longitudinal edges of this fine implement are well defined and regular, those of the face being quite sharp. The cross-section is not an uncommon form, though the inward trend of the sides to the back is more pronounced than in most items of this shape. The obliquity of the tang in regard to the axial plane of the face amounts to 15°. The angle of inclination of the blade is about 45° near the cutting-edge, and 30° to 25° higher up. A face view of this item is given in Plate XIV, and a side view in Plate XV.

In Fig. 94a, Plate XXIV, we note a similar form to Fig. 94, albeit the implement is much smaller. Fig. 94 a represents the cast of a finely formed and finished adze of black aphanite in Mr. Hamilton's collection. It is most symmetrical in form and finely polished. Its length is 6¼ in., width across cutting-edge 1½ in., and across butt end 1⅛ in. Thickness at shoulder-line, 1¼ in. The face is convex transversely, and straight longitudinally to within 1 in. of the end, from which point it is bevelled off slightly to form the cutting-edge, as is usual. The back is narrower than the face by ¼ in., and the prominent shoulder causes the marked appearance of longitudinal concavity. The sides are straight lengthwise and practically so transversely. The blade shows an angle of about 50° near the cutting-edge, which falls to 30° higher up. The butt end of the face has been cut deeply down to facilitate lashing, and this, in conjunction with the concave aspect of the back, imparts to the butt end an angular appearance. The original is a remarkably fine implement.

We now come to one of the most peculiar forms to be found in the Museum (see Fig. 95, Plate XVI). It is 9¾ in. long, and weighs 2¾ lb. What we take to be the back is deeply concave longitudinally and concave transversely. The small poll has a curious lateral projection, which may have served to hold this singular tool in the lashing. Without it there would certainly be some difficulty in keeping the tool on the haft, so rapidly does the chipped-down butt end decrease in size towards the poll. The thickness of this implement is 2¾ in. in the middle, a most unusual depth in one of its page 307length. For 4 in. from the poll the alleged face has been worked down to an extent not seen in any other specimen, leaving an abrupt and high shoulder, from which the butt end narrows both ways to a small poll. The face is convex longitudinally to a marked degree, and the angle formed by its intersection with the blade-bevel of the back is as high as 70°. So thick a blade and cutting-edge is rarely met with. The face is so much narrower than the back that at its thickest part a cross-section would be almost triangular in form. The cutting-edge is a scant 1 in. in width. The small rounded butt end lends itself to a hand-grip remarkably well. This tool may be a potuki, or pounder, that was used in the hand, not hafted. When asked his opinion as to Fig. 95, Te Whatahoro replied at once that it was a potuki, a pounder, used endwise to crush the roots of the raupo plant (Typha augustifolia). These roots (koareare, syn., karito) were collected, dried, and stored for future use. They were sometimes kept a long time. When wanted for food a portion was taken and soaked in water for one night, then put in a patua (a wooden vessel) and pounded with a potuki, such as we see in Fig. 95, in order to separate the mealy or farinaceous matter from the fibrous parts of the roots. This meal was then mixed with hot water, being made into a sort of porridge, and so eaten. For such purposes water was heated in the following manner: A kind of bottle was made of a sea-weed known as rimu-rapa. Water was poured into this vessel, and the upper part of it was then squeezed in and tied. The vessel was then placed in an upright position in a hangi, or steam-oven, and so heated. Ere long the water would become quite hot, and ready for mixing with the koareare meal, or for any other purpose it might be needed for. The seeds of the raupo were made into cakes or bread, and these were often eaten with the porridge-like food described above. Such food was reserved for the use of the upper class alone.

We now come to an odd form that may well be termed abnormal if treated as an adze, but it is evidently a pounder, or potuki (see Fig. 96, Plate XIV). It is an intermediate form between some of the thick adzes and the round stone pounders so well known. It may have been used as a beater, and not as a pestle is used. This curious item is 13¼ in. long, and weighs 7½ lb. Although it cannot have been a cutting implement, yet we propose to use such descriptive terms in explaining its form as have been applied to the stone adzes.

In the first place, a cross-section of this tool would appear as an acute-angled triangle, with its angles much rounded. Its greatest thickness, from face to back, is 3½ in., and this is in the middle of the tool. Its greatest width, also in the middle, is a little under 3 in. The back is straight longitudinally from poll to shoulder, and page 308somewhat convex transversely. Its two longitudinal edges are much rounded. The face, so called, is convex longitudinally, and very much rounded transversely, so much so that it would appear semicircular in section. This segmental section holds throughout the entire length of the tool, from the poll to the end whereat the cutting-edge would be in an adze. At the butt end the face and sides have been chipped down so as to leave a prominent shoulder on the face, and light ones on the sides. Evidently this was a hand-hold. The poll is rounded, and seems to show an original, smooth, water-worn surface. All the rest of the surface has been finished in a remarkably even manner, apparently by bruising, but not made smooth. The sides are some-what convex longitudinally, and much rounded transversely. The blade, if it may be so termed, is formed in the usual manner by means of a decided bevel on the back, showing a distinct shoulder, and a long, less decided bevel, or rather curve, on the face. The cutting-edge is, in this tool, replaced by a blunt, much rounded end, which seems to have been worn partly smooth by use. It can only have been used as a sort of pestle or punching-tool. If intended as a pestle, then the bevel would presumably represent useless work. If as a beater for fibre or roots, one would suppose the bevel would also impair its utility.

Another unusual form, noted in Mr. A. H. Turnbull's collection, is an adze of light-coloured stone, 8½ in. long, 2⅝ in. wide across the chord of the curved cutting-edge, and 2½ in. across the butt shoulder. Thickness, 1¼ in. in the axial centre of the tool at the lower part, and decreasing to 1⅝ in. at the poll. The face is convex longitudinally, and excessively so transversely. The sides have been worked down somewhat deeply, and the face roughened for 3½ in. from the poll so as to show a shoulder on each side, but none on the face, a very rare occurrence. This butt end is much rounded; in fact, a cross-section from poll to shoulders would appear as a symmetrical oval. The back of the lower half of the tool is of a curious form, and subtriangular cross-section. The sides slope inward to the narrow flat back at so acute an angle to the face that they practically form a part of the back of the tool. Hence the sides may be said to be represented by sharp straight edges. The angle of the blade ranges from 45° to 25°. The tool is ground to a smooth surface, and is polished, save on the butt end, where the lashing passes over.

In the Museum is a cast of a peculiar little adze which would act as a gouge, and cut a semicircular channel about 1¼ in. wide in timber (see Fig. 97, Plate VII). Its length is 5¾ in. Width in middle, 1⅝ in. It is of a curious turtle-back form, the contour of which is interrupted, however, by a deep hollow to accommodate and contain the lashing. page 309A cross-section would be semicircular, the back being straight longitudinally, though slightly convex laterally; and from its longitudinal edges the tool is rounded so as to represent a semicircle in section. The humped face is hollowed deeply for the lashing at the butt end, leaving a shoulder unusually prominent for so small an implement, also a prominent knob at the extreme end. The bevel of the blade is 1¾ in. long, and its junction with the semicircular face produces a cutting-edge of a similar form. It was evidently helved as an adze, and used for hewing out channels or grooves.

With Ornamented Polls

There are five stone adzes in the Museum showing a peculiar mode of ornamentation on the poll, effected by means of a graver, apparently, in two cases. On the other three the patterns seem to have been produced by grinding, or rubbing with a narrow piece of stone. Three out of the five adzes are of the unusually thick type; that in which we note the poll has been ground to a symmetrical form, a process that is most unusual in respect to the ordinary type of adze, and also those of nephrite. Of the other two with patterned polls, one is of ordinary thickness, the other rather thinner than usual, but in all the five specimens the cutting-edge is practically in the axial centre of the tool, as viewed sideways. Three of these specimens show by their worked-down butt ends that they were hafted either as adzes, or axially, as are chisels. Two have already been described among the wedge or axe shaped forms, and but three remain to be mentioned. The largest of these carries the best example of a patterned poll (see Fig. 98, Plate XI). It is 10 in. long, and weighs 3¼ lb. Width across cutting-edge, 2⅞ in.; across poll, 1¾ in. Thickness, 1½ in. at butt shoulder, from which point it decreases both ways. Face and back are convex both ways, the former to a greater extent than the latter. The sides are straight and flat both ways, quite flat transversely, and uncommon feature in Maori stone adzes. There is no bevel shoulder, the two faces converging in long low curves to form the straight cutting-edge. The butt end of the face has been very slightly reduced to facilitate lashing, leaving a slight shoulder-butt. The angle of inclination of the blade is nearly 50° near the cutting-edge, but soon drops to 30°. This item comes from Nuhaka. The stone is grauwacke. The poll has been ground smooth; and a curious pattern has been fashioned thereon, either by a graving process or by rubbing with the small end of a piece of sandstone or some other kind of stone. The pattern is in relief and is well executed. This implement (Fig. 98) resembles Fig. 20 in form, but is wider in proportion to its length.

page 310

Another, and much smaller, adze has a similar pattern incised on the poll, but it is much fainter, and does not stand out in bold relief (see Fig. 99, Plate XI). This item is 6½ in. long, and weighs 1 lb. Width across cutting-edge, 2¼ in.; across rounded poll, 1¼ in. Thickness in middle, 1 in. On what we presume is the back of the tool the blade has a very short bevel of 5/16 in. Face, back, and sides very similar in form to the last specimen described. In both items the back is somewhat narrower than the face. Fig 99 is of dioritic sandstone.

Another specimen in dioritic sandstone is of the thick wedge-shaped type, described elsewhere, in which the thickest part of the tool is at or near the middle, the blade being formed by an equal convergence of the face and back toward the cutting-edge (see Fig. 100, Plate XXVIII). It is 7¾ in. long, and weighs 2 lb. Width across cutting-edge (allowing for a chipped corner thereof), 1¾ in.; across middle, 2⅜ in.; across poll, 1¼ in. Thickness in middle, 1⅝ in.; thence decreasing both ways, thickness at poll being 1 in. Sides, convex both ways. Faces, convex longitudinally, but almost flat transversely. Longitudinal edges, rounded. Blade, rough on one face, possibly a fractured surface. Angle, 30°. The ornamental figuring on the poll is similar to that on the reke or butt of a stone patu (a short stone weapon), consisting of curved grooves in the form of concentric ovals.

Two other thick wedge-shaped tools show this latter pattern of ornament incised on the poll, which is additional proof that they were not used as wedges.

One of the most interesting items that has come under our notice is a cast of a stone axe, or adze, the original of which was discovered in Peru (see Fig. 100a, Plate XLIII). Judging from its general form, the original was intended to represent an adze, but it seems probable that it was a ceremonial implement, possibly not used as an ordinary tool, but merely in some form of ritual, inasmuch as it is ornamented with a human head and arms in relief on that face that would have been brought into contact with the material being worked if used as an adze. The length of the cast is 10 in. Width, 2⅜ in. across cutting-edge; width at 3 in. back from cutting-edge, 3¼ in., from which part the tool narrows to a smooth conical poll. Thickness in middle, about 1¾ in. Sides, face, and back are much rounded. The blade is very thick, and at and near the cutting-edge shows an angle of about 70°, hence "cutting-edge" is somewhat of a misnomer in this case. The original must be a very symmetrical well-formed implement, showing a smooth surface, save for a gap in the cutting-edge. The sides and what we take to be the face are markedly convex, both longitudinally and page 311transversely. The back is much flatter, its aspect of longitudinal convexity being imparted to it by the falling-away of the blade and butt end, for it is straight in its central part, about 4½ in. This is what gives the implement an adze-like form. On the back of this item is a curious carving in relief, some 5⅜ in. long, representing a human head and the two arms, but without any body, unless the bulk of the implement is intended to represent the body. The arms are represented as extending downward from the sides of the head, and the hands turn inwards at right angles 1 in. below the chin, the tips of the fingers meeting. The ears are placed high, and all the features are of rude formation, the teeth being very prominent. Apart from the carving, the most noticeable feature is the beautifully formed and finished smooth conical poll. It does not seem likely that this item was intended as a tool. We can only surmise that it was some form of ceremonial implement.