Volume 34 1925 > Volume 34, No. 133 > The Korekore Pa. An ancient Maori fortress, by R. W. Firth, p 1-18
[See map at the end of the previous section]
THE KOREKORE PA. AN ANCIENT MAORI FORTRESS.
Situation—Defences—Vestiges of Habitation—Storage places—An Artificial Cave—A Stone Pillar—Carvings on a Pit Wall—Burial Places.
THE stockaded Maori pa is now a thing of the past, gone with many another of the works of the men of old, and naught but its crumbling vestiges remain. Of these ruins of olden days, one that is in better preservation than most is that known as the Korekore pa, situated on the western coast of Auckland, about two and a-half miles north of the well known Muriwai pleasure resort. See Fig. 1. It has as yet happily escaped the attentions of the tourist and the picnicker, but its proximity to the haunts of such will not long leave it undisturbed.
A record of the earthworks and features of note of this old stronghold before they are irreparably damaged may, therefore, be of interest to the ethnologist. The pa crowns the summit of a thickly-wooded outstanding hill about five hundred feet high, rising steeply from the drifting sands in front, but linking up at the rear by a high ridge with the main heights which fringe the coast. The stronghold is excellently situated for defence, being rendered almost impregnable on three sides by the exceeding steepness of the slope. On the fourth the narrow ridge which joins it to the other hills is very well adapted to the construction of fortifications. In addition, there is a view for miles in every direction, the coast to the north being visible on a clear day almost as far as the Kaipara Heads.
The strategic position of the fort is a good one inasmuch as it dominates the highway of the beach. It is essentially a stronghold for defence, however, and in time of peace the greater portion of the hapu probably lived in the kainga built on a fairly level stretch of ground some three hundred feet below the pa on the north side. One informant, Mr. Wilson of the Kaipara, pointed out to me a grass covered - 2 green patch of ground which now marks the site of the ancient kainga, where, according to tradition, a great slaughter of the former inhabitants took place. The same man said that about 1880 he was told by an old chief that the pa was abandoned about five generations or one hundred and twenty years before that time, which would place the date of its evacuation about 1760. One name of the district he had heard to be Wharekura. The inhabitants of the pa
seem to have been the earlier inhabitants of the district before Ngatiwhatua swept down from the north. (See Mr. Graham's Historical Account which follows.)
It is certain that the pa has been abandoned for a great number of years. The hill is covered with thick bush of a fairly open character, containing pohutukawa, karaka, and nikau of some size. Just outside the earthworks are two immense trees, one being a Metrosideros of a circumference of twenty-four and a-half feet, the other a puriri eighteen feet round. Inside the pa on the tihi or summit, there is another puriri measuring fifteen feet in girth, while a Metrosideros of about the same dimensions projects from the side of one of the pits. The general character of the bush all over the pa indicates that a great number of years must have elapsed since it was last occupied.- 3
It is interesting to note that the ngaio shrub, which is somewhat uncommon hereabouts, and which is favourite article of diet with cattle, is fairly abundant on the western or seaward slope of the pa. I have heard also that near the southerly fringe of the fort is a tree of tawapou (Sideroxylon costata), the black shiny seeds of which were formerly used by the Maori to make a necklace. This tree and some few at Karekare are the only specimens I have knowledge of in the district.
The pa was well situated as regards food supplies. Birds could be readily obtained from the large area of bush inland at no great distance, and a fair expanse of rich warm soil lying at the back of the ford and facing the rising sun, would be conducive to the raising of good kumara crops.
Several lakes were also close at hand, and these provided as well as eels, the succulent kakahi (fresh water mussel). One is called Okaihau while the other bears the suggestive name of Paekawau (shag's perch). The rocks two miles down the coast are to this day an excellent fishing ground for line work, while netting in the surf generally results in a good haul. On Oaia, a mile or so out from the beach, there is a gannet rockery, which could provide young birds in their season. Toheroa from the beach sands and shell-fish from the rocks would make a welcome addition to the bill of fare, while the examination of the various species of mollusca represented in the shell middens round the pa, reveals that some of them could only have come from a mudflat, which in this case would be either the reaches of the Kaipara or the Manukau. Amphibola crenata, a species favoured by the natives of old as food, is never found in such exposed situations, and must in this instance have been transported from one or other of the adjacent harbours. Under ordinary circumstances, then, the inhabitants of Korekore need never be troubled by scarcity of food.
The accompanying plan, the result of an accurate survey of the hill, shows the general character of the fort.
On the smaller and southerly ridge the only indications of earthworks are terraces which may or may not have been used for purposes of defence, but more probably were constructed purely for house sites. All the really defensive works are situated on the larger, higher and more northern - 4 ridge. This, in conjunction with the fact that all the underground rua referred to later, are on the same ridge, leads one to conclude that the actual defensive stronghold was on the latter ridge alone, and that the first one was used almost solely for dwellings. The shape of the ridge renders any elaborate scheme of defence unnecessary. On the north side the cliff falls sheer for some three hundred feet, and renders the pa impregnable from that side unless by a stealthy and arduous night attack—a toilsome piece of work. On the beach front also there is an exceedingly steep slope, and a palisade at the brink of the hill would probably be ample fortification. The upward climb on this side is also a heart-breaking one, as the writer knows to his cost. On the southerly side there are three narrow terraces which would probably be defended by the usual palisades, and below these the hill again falls steeply to the watercourse beneath. The only really practicable way of attack would be by way of the connecting saddle from the hills at the back, or over the sand-hills from the same direction. In both cases one strikes the hills somewhere near the join of the two ridges, and the way to the citadel lies straight along the main ridge. To obstruct this—the only easy method of approach—ditches have been dug across the ridge at its narrowest parts. The scheme of defence consists of three fosses, the first and third comparatively small, the second a veritable gap in the crest, which would prove no mean obstacle to hostile visitors.
VESTIGES OF HABITATION.
There are the remains of several distinct terraces on the more gently sloping side of the fortress toward the south-east. Not many house sites occupy these terraces, however, most of them being on the crest of the ridge.
There are a large number of rectangular pits in the pa altogether, and it is thought that the larger ones indicate former house sites, while those of lesser size represent storepits for the kumara, aruhe, or other food crops. If the assumption as to house sites is correct then the huts have evidently been of a pit-dwelling nature—the site being excavated to the depth of a few feet and then roofed over, - 6 the entrance being by means of a small hole which served as doorway. In his work on Maori storehouses (Dom. Museum Bul. 5) Mr. Elsdon Best gives full description of the rua tahuhu and similar pits of large size for the storing of the kumara. Some of the large pits at Korekore may have been used for that purpose, but many of them are undoubtedly the relics of ancient dwellings. Thus on page 85 of the same work he notes: “The curious subterranean dwelling-places occasionally made and used by the Maori in former times. … . They were made by excavating a rectangular pit about four feet deep, over which was erected a Λ-shaped roof of timber which was then covered with a thick layer of earth.” Very often a layer of sand was spread on the floor and served to render the whare more comfortable. This was evidently the main type of dwelling in use at Korekore, and appears to have been a form of house somewhat common with the people of that district, though not often found in other parts. It is comparable with the pit-dwellings discovered by Mr. Joshua Rutland in Pelorus Sound, though I would not suggest that there is any necessary connection between the two. The northern people on this coast, as far as can be gathered, seem to have a preference for dwellings and storehouses of the excavated type. (See Mr. George Graham's paper on Rua kopiha in this Journal, Vol. XXXI., page 120.)
In all, the pits of rectangular shape now to be found in the pa are sixty-three in number. Doubtless there are some others which the process of time has obliterated. Of those remaining approximately twenty are the sites of dwellings or are of such size as to be used as hut sites. In addition there are four excavations which are undoubtedly the levelling of the slope to allow of the building of whares. About forty-three of the pits are of smaller size and are probably the common type of rua, originally roofed over. Besides these there are also eight pits on the saddle connecting the pa ridge with the hills, and beyond again on the cleared ground are still to be seen numerous other depressions of the same type.
The distribution of the pits seems to indicate no definite order or arrangement. There are thirty-two on the smaller ridge and thirty-one on the larger, the latter having twelve house sites, whereas the former shows only seven. The pits are generally in groups of four or so, some lying side - 7 by side, others running end to end. On the smaller ridge the majority of excavations lie on the eastern side. This is easily understood when it is remembered that for rua kumara a warm dry soil was much to be desired; hence the slope of the hill facing the sun was usually chosen.
SIZE OF PITS.
The excavations vary considerably both in size and in proportion of length to breadth. The dimensions of a few are given below to indicate the character of the pits in the northern district. The typical pit here seems to be rather larger than in some southern districts.
On one of the terraces at the point of the first ridge is a group of four:—One pit measuring 20 feet by 9 feet is evidently a former hut, and has three narrow rua, 20 feet by 5 feet; 16 feet by 4 feet; and 10 feet by 4 feet alongside it. Other dwelling pits on the same ridge measure 17¾ feet by 10 feet by 4 feet and 16 feet by 8 eight by 4 feet, and are in very good preservation. It may be noted, however, that the depths as given here are probably much less than was originally the case, owing to the accumulated deposit of leaves, etc., through a hundred years or so. The largest pit in the whole pa is on this ridge and is 28 feet long, 21 feet wide and 8 feet deep. This pit is remarkable for its depth and also for the fact that there are certain carvings on the sandstone walls.
Other smaller pits have the dimensions of:—15 feet by 8 feet; 15 feet by 6 feet; 14 feet by 7 feet; 14 feet by 6½ feet by 3½ feet; 11 feet by 6 feet by 2½ feet.
On the second ridge the largest excavations measure 24 feet by 12 feet and 20 feet by 10 feet, and the levelled whare site near the crest also has a 20-feet side. The pit on the far side of the main ditch is peculiar in that it is square, measuring 14 feet by 14 feet. Two of the pits by the first trench have dimensions as under:—14 feet by 8 feet by 4 feet; 12 feet by 5 feet by 1½ feet. The depression near the fork of the ridge is 12 feet long, 8½ feet wide and 2 feet deep. Various store pits have the following dimensions:—12 feet by 4 feet; 15 feet by 5 feet; 15 feet by 8 feet; while two small excavations near the point of the ridge are each 10 feet long by 5 feet wide.
One thing was noticeable when measuring these pits—that in general the proportion of length to breadth was - 8 approximately two to one. In about ten examples, this was the exact proportion, while in many others the difference was only some six inches or so, either way.
In addition to the rectangular pits there are also many storage chambers of the true underground type. These are fairly common throughout New Zealand and have been described by Elsdon Best (op. cit. p. 79 et seq.) but mainly with reference to Whanganui and southern districts. A few notes on those in this northern pa may, therefore, be of interest for comparative purposes.
Entry is made by a round hole in the fairly level ground of about two feet diameter to a dome shaped cavity. One series of pits was on sloping ground on the crest of the ridge; another series lay on the second terrace right at the base of the join of the two terraces. The pits were circular in form, the walls rising and drawing in to meet at the entrance hole overhead. They were entered by means of a step or two projecting from the lower side of the pit, i.e., the down-slope side. Most of the pits were divided in two by a sandstone partition running full width across, which had been left standing when the pit was first excavated. In addition to serving as a wall this also acted as an extra step.
One pit near to the third trench is a typical example of the type without the dividing partition. The diameter of the entrance hole was two feet six inches; the only step was two feet from the surface and seven inches wide, projecting from one wall which was practically vertical, and the inside diameter of the pit was seven feet nine inches. Its depth was four feet nine inches. The floor of the pit was circular in shape and had traces of shell lying around.
Several other pits were measured. A series of five marked in the plan were situated just beyond the third trench. Pit F. is a hole immediately on top of the cliff storehouse, and beside the main ditch. Pits Y. and Z. are beyond the third trench and members of the series of twelve marked on the plan.
A series of dimensions are given in tabular form for purposes of comparison- 9
The size and number of steps varies, no absolute uniformity of structure being observed even in pits of the same series. The ancient craftsman evidently worked mainly by eye and rule of thumb in these matters. In pit 1 the step is four inches wide, in pit Y. it is eight inches wide. Pit 3 has neither step nor dividing ledge, but pit 2 has a second step twenty inches below the first, while pit 4 has four steps in all, the topmost six inches from the ground, the others at intervals of one foot three inches, three inches, and one foot down. In this case these steps serve the purpose of the notched pole which was often used for entrance into and exit from rua kai.
The position and dimensions of the dividing ledge or partition, which acts in most cases as a bottom step, also vary considerably. By far the greater number of pits examined possessed this partition, which was always carefully chiselled and smoothly finished with vertical sides and level top. There were no signs of board covering such as mentioned in Mr. Best's description, nor does the nature of the soil render any necessary. Four of the ledges were measured and gave the following dimensions:—
In pit 5 the two sections separated by the partition are of unequal dimensions, one segment measuring 6 feet by 3 feet 4 inches while the other is 6 feet 8 inches by 4 feet 10 inches. In nearly all other pits the segments are symmetrical about the dividing ledge.
One cannot be too dogmatic about the purpose for which these pits were intended. It seems probable, however, that certain of them served as storage chambers for water rather - 10 than as rua kai for kumara or other viands. They correspond very closely in type with the sketch and description given by T. W. Downes and recorded in the Dominion Museum Bulletin No. V., p. 81 et seq. The shape is identical, and the raised earthen partition is present in both cases. At Korekore though, the dome-shaped chamber commences at once from the opening, whereas at Tunu-haere pa “the depth of soil from the surface down to where the chamber widens out, is in some cases as much as three feet.” The fact that the nearest stream is some fifteen or twenty minutes' walk away with a steep ascent on returning, into the bargain, would seem to render some sort of water storage imperative. When it is considered that as far as could be ascertained all the underground chambers of this particular type were on the main ridge and well within the fortifications, it seems probable that some at least acted as water reservoirs. It may be objected that the light soil of the nature of volcanic tuff is incapable of retaining water, at any rate, for any period of time. But on the very summit of the ridge, some few feet above the pits, is a large house site which to the best of my knowledge has remained full of water for the last four years. If the water only remained in the pits for a week or more, even so it would richly repay the defenders of the pa for the labour in excavating those underground chambers. None of the pits contain water at the present day, but several are exceedingly damp, which would have militated against their use as rua kumara. Others, however, show remains of shells and are undoubtedly rua kai.
Several of the pits are connected together by small openings at the base. Thus in the group of five, numbers 1 and 2 are joined by a hole 2¾ feet wide and 10 inches high; numbers 3 and 4 are linked by an opening 4 feet wide and 2 feet high. This series are all close together and in line along the slope, and are about five feet apart. The distribution of these underground chambers is interesting. Some 50 yards or so along the crest from the group of five is a single isolated one, then comes a group of twelve all in line, having step and earthen partition, many of them being connected as in the manner described above. Quite a number of this series are in a state of decay, only the opening and a portion of the chamber remaining.- 11
Further on again, is a group of three pits of the same type as before, while over to the right of the ridge are five, very much filled with accumulated debris. These last are excavated in an old house site, one pit having been dug at each of three corners and two others along one of the sides. A similar group occurs in another place, but here only the depression remains to mark the type of excavation.
In the sides of the last trench a slightly different type of rua is met with, corresponding somewhat to the storehouses mentioned by Mr. Cowan as existing at Te Pehu (v. J.P.S., Vol. XVII., p. 222) and those described and figured by Elsdon Best (op. cit., p. 88). The entrances to these are made in the sides of the ditch. These are ordinary rua kumara. A somewhat quaint theory has been put forward that these pits were constructed as places of refuge for the women and children of the pa in time of siege. If so all that can be said is that the quarters must have been exceedingly cramped, as the writer was compelled to unnaturally contract himself while lying solus in one to examine it.
There were originally six of these rua, three in each bank of the ditch. Now only three remain, the sides of the fosse having slipped in and covered the others. The three on the eastern side were connected by small openings made in the soft stone as in the case of the other pits; those on the other side appear to have been separate. A rectangular mark round the doorway of one shows where a tatau or wooden door has been fitted. The floor level of all these pits is several inches below the entrance sill.
Another interesting excavation is a small cave made in the sheer cliff overlooking the northern beach and ten feet below the brink. This is concealed from anyone on the cliff above and was discovered by pakeha (European) investigators quite by accident. The mouth was somewhat hidden by bushes. The photograph (Fig. 2) gives an idea as to the size and character of the hole. The diameter of the entrance is 1 feet 7 inches; and the depth of the pit from the opening is 4 feet 3 inches. It is four feet wide and the average height inside is 2 feet 5 inches. The sill or threshold is quite high, the level floor of the chamber being ten inches below it. The hole or pit as it might be called, is circular in shape. The roof is rounded as a dome, rising carved from the floor. The marks of the wooden or stone tools are still quite plain on - 12 the roof and sides and are of two kinds—some gouge-shaped, others small and round. Two kinds of tools were evidently in use to make the excavation—first, an instrument for boring or pitting the surface of the soft material; and, secondly, a tool of the adze or gouge type to chisel away the remainder of the soil and smooth off. The gouge marks were narrow and indicated the use of a fairly small implement.
There have been several conjectures as to the purpose of the cave. One is that it served as a watch hole for the look-out man, where he might lie secure from observation and obtain an uninterrupted view of the far-flung beach—an analogy to the modern sentry box, in fact. This seems hardly probable. It has been further suggested that it was used as a temporary dwelling place for a person under tapu as, e.g., a chief after the completion of tattooing, or a tohunga who had been engaged in burial rites. This, too, seems hardly likely, though the cave is large enough to accommodate a man under such circumstances.
A more feasible suggestion is that it served a similar purpose to that mentioned by Mr. Cowan as existing at the Whetengu, Rotorua, i.e., as a repository for sacred objects. The dimensions and situation are much the same, viz:—A recess in the face of the cliff, about four feet deep and wide, with an entrance somewhat over a foot in diameter. Such a sacred storehouse might well hold either an image of the tribal god as did that at Rotorua, portions of chiefs' hair after the sacred cutting, remains of priestly food, or similar tapu things. There such objects would be secure from disturbance and would be saved from any dangerous contact with the people of the pa.
Some 250 yards or so further along, past the highest point of the pa and away from all dwellings or store pits is a curious object, a natural rock pillar, set in the earth close to the cliff edge and apparently in quite a secluded portion of the pa. Since there is no similar rock of any kind within the fort, this has evidently been carried here by the inhabitants. The material is andesite and the formation is that of a pentagonal pillar of somewhat similar type to the well known basalt pillars of the Giants' Causeway. As far as could be ascertained, the nearest locality in which similar pillars occur in situ is a volcanic cliff the other side of Motutara and some three miles to the south. This stone, - 13 weighing several hundredweight, must, therefore, have been carried with enormous labour along the beach to be set up in the pa some five hundred feet above the sea. The accompanying photograph (Fig. 3) shows the stone as at present situated. It measures four feet in height with an approximate diameter of one foot, and is pentagonal in cross section, the various sides being respectively 10 inches, 6 inches, 10½ inches, 8½ inches, 8½ inches across, measured near the bottom of the stone. At present the pillar is in three segments. It is broken across the middle and a smaller piece has been knock off the side of the lower portion of the stone. Some fragments have also disappeared from the top. When the writer first saw it some years ago the stone was set deeply into the earth, but it has evidently been recently dug up, and now lies completely exposed. Its original purpose is unknown. Several suggestions have been made. The late Mr. Cheeseman, curator of the Auckland Museum, was of the opinion that it may have been a symbol of Rongo, god of the kumara, and hence used as guardian of the cultivations. An alternative suggestion was that it acted as a boundary stone for the demarcation of land. As the stone was set up in the midst of a pa, however, these suggestions seem hardly likely. It may have been a rock at which the ceremony of uruuru whenua was performed, but this again is not probable. The two suggestions which to me seem most probable are that it was either a material mauri symbolising the mana of the pa, examples of which are known to have existed in forts, or else it served as the tribal tuahu. The latter seems more likely. A mauri or talismanic symbol embodying the mana or power of the pa would, in all probability, be well concealed in some place which no enemy would discover, lest by potent karakia he destroy the efficacy of that mauri and so deprive the fortress of any virtue it might possess as a stronghold and place of refuge. Material mauri, as a rule, seem to have been buried or hidden within the precincts of the fort for this reason.
The tuahu or altar, however, was not concealed in this fashion. It was merely placed in a secluded part of the village at a little distance from dwellings. This seems to have been the case with the andesite pillar. It does not appear to have been concealed in any way, but simply to have been set firmly in the earth in a spot somewhat removed from habitation. A reference to the plan, where A. marks the - 14 position of the stone, shows this. It is reasonable to conclude then that this stone pillar was probably the old tuahu of the pa.
On the summit of Maungakiekie or One Tree Hill, is a similar pillar obtained from the Three Kings (v. Mr. Graham's paper, J.P.S.).
One of the most interesting features of the pa is the presence of carvings in the sandstone wall of one of the rectangular pits. This, marked K. on the plan, is the largest of all in the pa, and one wall is still in a good state of preservation. The original depth of the pit, considering the leaves and refuse that now lie there, must have been approximately ten feet. The carvings have been made chiefly on the south-western wall, that nearest the sea, with some minor figures on the north-western side. Two slabs from the latter wall are now in the Auckland Museum, having been cut out and deposited therein some three years ago by Mr. A. W. B. Powell and the writer, to save them from crumbling to pieces, as that section of the wall was in danger of collapsing. It has since fallen in and the south-western wall alone now remains intact. These carvings have been known for many years, but as they have not previously been described or figured, an account of them is included here.
Practically the whole of the south-western wall is covered with these incised lines. They are no mere scratchings on the surface, being in some cases more than an inch deep. The incisions are broad and the figures are well demarcated. The style of the carving is in general free and bold and the work is quite evidently that of no prentice hand, though whether or no it conforms to the canons of the pakeha artist it is not for me to say. Figure 6, especially, displays a commendable skill in delineation, and the carver has managed to convey quite a quaint effect by the use of a few sweeping curves.
The figures on the south-western wall are still in a good state of preservation. The greatest damage has been inflicted by the roots of ferns which ordinarily cover the side, and which had to be removed to allow of sketching and photography.- i
- ii Page is blank- iii
PLATE I.- iv
Fig. 2—Mouth of Cave in cliff at Korekore pa., Fig. 3—Rock Pillar at Korekore., Fig. 6—Figure Carved on wall of pit., Fig. 7—Incised Designs on pit wall.
PLATE II.- 15
Figs. 8 and 9—Designs incised on wall of pit, Korekore pa., Fig. 10—Carved block of pumice from Chatham Islands.
Besides partially obliterating the incisions, these ferns tend to confuse the design by making it difficult to distinguish between artificial lines and those formed by root action. The face of the north-western wall has already crumbled down. No trace of carving could be found on either of the other walls, or in any other pit in the pa.
An explanation of the plates may be necessary here. No photograph showing the complete scheme of carving on the south-western wall could be obtained. The sketch in Fig. 4, drawn by my cousin, Mr. C. W. Firth, gives, therefore, such a complete view. Figure 7 is a photograph taken 14 years ago, and shows a large section of the wall. This is reproduced by courtesy of Mr. Gilbert Archey, curator of the Auckland Museum, and bears the initials J.H.B. To this unknown photographer I wish to express my obligations. Figs. 5 and 6 give fuller views of details of the carving. Figs. 8 and 9 are photographs of the two slabs from the north-western wall, the originals of which are in the Museum. Fig. 10 is of a piece of pumice, carved, found at the Chatham Islands, and inserted for purposes of comparison. The originals of Figs. 8, 9 and 10 were photographed with the kind permission of Mr. Archey.
The carving as a whole presents no uniform scheme of design. As regards the south-western wall, Figs. 4 and 7 show that there are several main lines which traverse most of the field, but no clearly defined pattern is recognisable from them. When the detail of the work is studied, however, several complete figures can be picked out. A weak and ill-defined double spiral is seen on the right of the wall at the base of what faintly resembles a human head. By digging below the level of the pit floor the crude representation of a human face may be uncovered. The greatest interest attaches to Figs. 5 and 6. (Before taking the photograph in Fig. 6—about three years ago—the incised lines were carefully traced out in white in order to ensure a good result, but this precaution proved to be unnecessary.) These figures are quite small, the larger one being about fourteen inches, the smaller (Fig. 5) about eight inches high. The larger one is clearly defined with a bold stroke, and the obliquely-set eyes and round nose lend it a distinct Simian appearance. The design is incomplete, deeper incision on the left terminating the figure, which lacks a right arm and leg. In spite of these defects it possesses the most character - 16 of any of the carvings. The triangular shape of the face is reminiscent of the figure in the Kaitaia piece of carving, and it may be noted that this is a shape of face unusual in the orthodox Maori treatment. The smaller figure on the left of the main design is similar in treatment to Figure 6, though much cruder in execution. It suggests lack of finish. No other definite figures can be picked out on this wall.
Imagination can supply many interpretations to the carvings, but such speculation is valueless.
The two slabs from the north-western wall are distinctive, being unlike any design on the other face of the pit, and also of totally different character to one another both in design and workmanship. Fig. 8 is evidently intended to depict a human face. The eyes, nose and mouth can be distinguished, but the object of the other lines is not apparent. Comparison with Fig. 6 shows that the carving is of an entirely different nature. The engraving of the eyes in particular, differs. In Fig. 6 the eye is formed by cutting an elliptical ring, leaving the centre in relief, whereas in Fig. 8 a deep semiovoid cut serves the same purpose. Fig. 8 may be compared with the carvings in a similar pit near Otakanini pa, described by Mr. R. Buddle. (Trans. N.Z. Inst., Vol. XXXXIII., p. 597). Unfortunately, his plate is not sufficiently clear to establish any precise similarity. In general, however, the carvings are of similar type—each was found in the Kaipara district in a former pa on the walls of a sandstone pit, and are probably related in nature if not in form. What Fig. 9 represents is somewhat uncertain. It is apparently anthropomorphic in shape, but even in this it is unsafe to be dogmatic.
The whole series of carvings, however, is quite unique. Many substances have been used by the Maori as the media of expression in carving, but the cutting of designs or figures in the walls of dwelling-pits appears to be of quite rare occurrence. As to their significance we can only conjecture. Mr. Buddle offers one explanation in his account just referred to—esoteric figures drawn by the tohunga—but this is by no means conclusive. It is certain, however, that the carvings were not made to while away an idle hour, and it is probable that they originally had some symbolical meaning.- 17
Projecting from the steep seaward spur and some two hundred feet below the pa is a huge crag towering above the sand, and, to one who looks from below, dwarfing the rising hill behind. The rock is of conglomerate, sometimes termed pudding-stone, and is deeply fissured where the crag juts out from the hill, looking as if at any moment it might fall outwards down the slope. In this mighty cleft were laid the bones of the chiefly dead. From this urupa, 1 a shaft descending some forty feet into the rocky chasm, several skulls have been removed by scientists.
The writer was recently informed that another wahi tapu of the tribe lies on a sandy knoll to the south of the pa.
ACKNOWLEDGEMENTS.—For the survey of the pa and the plans here reproduced, I have to thank my cousin, Mr. C. W. Firth, B.Sc. I am also indebted to Messrs. A. W. B. Powell and R. A. Falla for my first view of the pa; to Mr. Gilbert Archey, M.A., as aforementioned; and to Messrs. W. E. La Roche, J. Elliott and W. A. Macky, B.Sc., for allowing me to turn a camping holiday into an archæological investigation, and for actively assisting in the delving into and measuring of these works of the old-time Maori. For their co-operation I am grateful.
Mr. George Graham also has greatly increased the interest of this paper by the historical account which follows.
Remains of old pa are common, but the fortress of Korekore contains such a variety of relics of former days that some record of its most interesting features should be preserved as a monument to those neolithic builders of the historic past.
Maori terms employed in the above narrative:—
1 This last resting place of exhumed bones would be termed a toma.