THE SUMNER CAVES.

Globe, Volume I, Issue 94, 18 September 1874, Page 3

THE SUMNER CAVES.

The following paper on excavations conducted in the year 1872, at the Moa Bone Point cave, was read by Dr Haast, Ph. D., F.R.S., at a special meeting of the Philosophical Institute, held on the loth instant. Dr Haast said, in the spring of the year 1872, Mr Edward Jollie suggested to me that the Moa Bone Point cave, near Sumner, if properly examined would, doubtless yield important and interesting results. My friend thought that by making there extensive and careful excavations, the question of the age and extinction of the moa would be, if not entirely solved, at least considerably advanced, and, that even the ground near the entrance of the cave, many acres in extent, if properly investigated would offer additional evidence for the elucidation of the subject. On my remark that no funds for such an undertaking were at my disposal, Mr Jollie headed at once a subscription list for the purpose, followed by a number of gentlemen who took an interest in the matter, and which, assisted by a grant from the Philosophical Institute of Canterbury, soon placed the greater portion of the necessary funds in my hands. After having obtained the permission from Mr Alfred Claypon Watson, Sumner road, on whose property the Sumner cave is situated, I began the work of excavation on Monday, September 23rd, and ended on Saturday, November 9th, 1872, the same having thus been accomplished in seven weeks, during which time I occupied always two laborers working under my directions. I may be allowed to present here my warmest thanks, on behalf of the Canterbury Museum, to the subscribers of the funds, of which the details of expenditure will be found in Appendix A., and to Mr Watson, the owner of the soil, for his permission to undertake the work.

I wish also to apologise to them that I am only enabled to-day, to lay the results of these excavations and researches before them, However, I must plead in extenuation), thai the bulk of this paper was written more than a year ago, but that I was then compelled, from want of room in the Museum, to repack the extensive collections made during these excavations, before I found the time to examine them thoroughly, and describe them in detail ; and only in the last few months I have managed to make the necessary space in one of the workrooms for doing so. Before entering into a description of the results achieved, I think it will be expedient to offer first a few general observations on the geological features of the cave, and of the surrounding country, as in the summing up it will be necessary for me to refer to them in elucidation of some of the points at issue. GEOLOGICAL FEATURES. Banks' Peninsula, an extinct volcanic system of large dimensions, stood as an island in postplioecnc times in the sea, showing by the configuration of its' base that an oscillation'averaging about 20 feet in vertical height has taken place, the country being depressed and afterwards raised to about the same altitude again. This line is well visible travelling round Bank's Peninsula to its western terminatioß, where, when we reach that altitude above the sea level, the signs of a former submersion disappear before the newer fluviatile and lacustrine deposits. During and after the small submergence of its base, this portion of Banks' Peninsula was, of course, subjected to the fury of the waves, wheu in favorable localities caves were formed either by the removal of loose material (tufas) between two harder lava streams or by the enlargement of pre-existing hollows, such as are found as air bubbles, often of gigantic size, in lava streams, running generally parallel to the direction of their flow. In this instance there is no doubt that the Moa Bone Point cave is a pre-exist : ng hollow in a doleritic lava stream, which has been enlarged by the enormous power of the dashing waves of the ocean beating here at one time furiously against the northern foot of the peninsula. In pievious publications (amongst others, Report on the formation of the Canterbury Plains, 1864, page 22, and seq), I have shown how, in postpliocene jtitnes, from the material brought down by the enormous glacier torrents, forming huge shingle fans at the foot of the glaciers, two bars were thrown across the sea ; one to unite the northern or ! Waimaikariri Ashley deposits with the : the northern slopes, another to connect the . southern or Kakaia-Ash burton bed:) of the pame nature with the southern slopes cE

i >,•>.; ■:.■;' Peninsula, belli ml which a huge lako was lurmcii. of which Lake Ellesmerc is tho Jasfc remnant. Of tho northern bar, wo ca i trace the inner or western shores through Kaiapoi to the neighborhood of Wootlend. in" this large freshwater lagoon (occasionally an estuary basin) the Waimakariri, Selwyn, and sometimes the Rakaia, discharged their waters, having an outlet uenr the north-western slopes of Banks' Peninsula, which in going towards Cashmere, the rcsi

dencc of Sir J. Cracroft Wilson, we can easily trace; the lines of dunes and shingle by which the eastern shore of that lake was formed being in the beginning very narrow, and only gradually as more and more material was added assuming a greater breadth. Thus we are able to follow the different lines of these earliest-formed beds from the neighborhood of Kaiapoi, where they are comparatively narrow, along the eastern boundary of Christchurch to the northern foot of the Peninsula, gradually diverging moreandmore. In my former paper, entitled " Moa and Moa Hunters" —"Transactions of the New Zealand Institute," vol iv , page 81)—I have already alluded to the fact that the ovens of the moa hunters were confined to the inner lines of these dunes, and a further close examination of the district between Christchurch and New Brighton has confirmed fully my former more local observations. Thus it is evident that when the former inhabitants of this part of New Zealand existed principally upon the chase of the moa, the sand dunes had scarcely reached the foot of the Peninsula, where now the Ferry road crosses the Heathcote, and consequently that the whole breadth of the sand dunes from opposite that locality to the Sumner bar, where they have now their south-eastern termination have been formed since. There are some Maori ovens and kitchen middens on the northern side of the Heathcote estuary, but they invariably contain only shellbeds. POSITION OF THE CAVE. When the cavity now called the Moa-bone Point cave was enlarged by the waves of the sea, the estuary of the Heathcote Avon in its present form was not yet in existence. Close to this cavity on its western side a hard doleritic lava stream, now passed through by the Sumner road cutt'.ng, reached for some distance into the sea, forming a small headland, against which, principally on its eastern side, the waves of the Pacific Ocean broke with considerable force. Masses of rock were detached by the surf, being taken along in an easterly direction for about a quarter of a mile, forming a ridge, gradually becoming lower, and losing itself amongst the sands The formation of this ridge principally took place when this portion of the peninsula was some 12 to 15ft lower than at present, the upper line of boulders being about 16ft above the present high water mark. When the land rose again the sea was cut off by this boulder-ridge from the entrance of the cave, a huge rock lying here nearly across, protecting it at the same time from being filled by the deposits of driftsanda now forming on the flat close to it.

A second and lower line of boulders was formed in front of the former about sft above present high water mark, with a small terraced space behind it. Since then other deposits, formed in the Avon-Heathcote estuary, have been added as a small belt in front of this last line of boulders brought into its present position by the action of the open sea. In section No 1 I have given the necessary details in illustration of these points. Before giving a description of the cave as I found it before beginning my labors, I may observe that the same was well known from the very beginning of the Canterbury settlement. It was even inhabited by some of the earliest settlers, and for some time afterwards afforded shelter to limeburners, fishermen, and road parties, of whom, as will be seen in the sequel, ample traces were left behind. The entrance of the cave, which is about forty feet from the crown of the Sumner road, having here an altitude of 18*59 feet above high water mark, is situated nearly 5 feet lower, or 13 61 feet above high water, taking the level of the surface for our line. An opening, which is about 30 feet broad by 8 feet high; being, however, much narrowed by a huge rock leads into the cave, of which I found the floor slightly sloping down. The cave itself consists of three compartments, of which the first one possesses by far the greatest dimensions, running nearly due north and south, and being 102 feet long, 72 feet broad towards the middle, and about 21 feet high. From its termination by a small passage, a second cave is reached, which is 18 feet long 14 feet wide, and about 11 feet high ; its direction being North by West to South by East, at its southern end a small passage 3 feet high, by about 2'50 feet broad, leads into a third or inner chamber, which is 22 feet long, with an average width of 1G feet, and about 20 feet high, running again like the principal cave due north and south, its floor being about eight feet above high water mark. My best thanks are due to Mr T. Roberts, the present engineer of the Gladstone and Timaru Board of Works, who, at my request, has taken the necessary levels, and surveyed the cave, the results of his labours being attached to this report. CONTENTS OF CAVE. An examination of the surface beds of the cave showed that the floor of the main cave was in some localities covered with the [remaius of European occupation; in many others, by the excrements of goats and cattle, introduced by the European in 1839 ; but that everywhere below them, when visible, portions"of shells of mollusks were occurring, the same species as still inhabit the estuary close by, and had served as food to the natives of the islands visiting the cave in former times. Towards the end of the main cave, these beds gradually thinned out and were mixed with each other, till near the entrance to the second cave marine sands, the former floor o? the cave, reached the surface. So proceeding with the laborers to the cave, I instructed them to dig two trenches crossing each other at right angles in the centre of the cave, till they [reached what they considered the lowest part of the deposits due to human agency. On September 29th, when arriving early in the morning, the greater portion of that work had been accomplished, the workmen having reached a bed of agglomerate, which they considered the bottom of the cave for our purpose, or at least reaching to the earliest beginning of human occupancy. Digging by my direction through this agglomerate for a considerable distance down into the sands below it, without any proof of human presence being obtained, I also reluctantly, at least, for the present, gave up any further work below it. Sections No 1 and 'A give the details of the i excavations'! then performed, At centre where the two trenches crossed, I noted the following sequence : ft i i 1. Shell bods, consisting of the remains of the following specif" 1 . now still inhabitingthe estuary:— Chicne stutchburyi (cockle) huai or pipi, Mesodesma chemnitzii • (pipi) Amphibola avellona (peiiwinkle) hetiku-tikn, Mytilus aragdinus (mussel) kuku ..= .1 10

2. Ash bed, with some pieces of flax, ft, in. cabbage tree leaves, charred wood, &o 0 8 3. Bed consisting of shells, often very much decomposed, the same species as No. 1 ... ... ... 1 2 4. Ash and dirt bed, with a few pieces of moa bones. ... ... 0 9 5. Agglomeratic bed consisting of pieces of rocks fallen from the roof 0 6 This latter deposit rested upon— 4 11 6. Marine sands in which I had dug down 3ft without results. Between 3 and 4 a sharp line of demarcation was clearly visible which, as the continuation of the excavation showed, was of great importance. European beds do not appear as occurring on the surface at this point, as they had been previously cleared away by the workmen. Near the entrance of the cave the following beds were passed in the longitudinal trench. (See section No 1). ft. in. 1. Beds of European occupation, cowdung. tins, pieces of bottles, &c 0 9 2. Shell beds 2 3 3. Ash beds 0 5 4. Shellbeds 1 4 5. Ash beds, chips of wood, tussocks 0 6 6. Shell beds, often very much decomposed, with small chips of timber, and thin beds of ashes between them, about 3 0 8 3 (Lowest portion of No. 6 not reached.) Owing to the depth of the trench at this spot, the same was not continued. The spot where I noted this section was about 10ft from the entrance of the cave. At the point where it reached the large rock, lying nearly across the entrance of the cave, the sequence was as follows : ft. in. 1. Beds of European origin 0 7 2. Shell beds 2 1 3. Ash beds 0 6 4. Shellbeds 1 4 5. Ash beds 0 9 6. Drift sands 1 0 7. Ash and dirt beds (lower series) 0 7 8. Agglomerate 0 5 9 5 The shells in the beds were exactly of the same description as those given in the foregoing section at the janction of both trenches in the centre of the main cave. The sequence of the beds and this identity of species proved clearly that a native population, living principally upon the mollusks now inhabiting the estuary, had occupied every part of the cave during a very long period—that portion close to the entrance being of course preferred ; this accounts for the greater thickness of the beds in its immediate neighborhood, which, as will be observed, gradually thin out as we advanced towards the termination of the first cave. Advancing to a consideration of the section exhibited in the cross trench, we find that it passes through the following beds on its eastern side (section 2). ft. in. 1. European beds, consisting of wheaten straw, bones of butchers' meat, shells, match boxes, horsedung 2 1 (Here was evidently a favorite spot for the cave dwellers of European origin) 2. Ashbed, tussocks (Maori) ... 0 4 3. Shell beds, similar to those described previously 0 8 Lower Series. 4. Ash and dirt beds 0 5 5. Agglomeratic beds 0 7 4 1 Below 5 the marine sands were examined for about 3ft. down. B.—Western Side. 1. European beds, mostly cattle dung 0 1 2. Shell beds like No 3 in previous enumeration 1 1 3. Ashes, tussocks, &c. 0 10 4. Shell beds often much decomposed 0 9 5. Ashbed 0 4 6. Ditto mixed with shells 0 9 Lower Series. 7. Dirt and Ashbeds 0 3 8. Agglomeratic bed 0 5 4 9 Marine sands proved to exist for about three feet below No 8. Besides the shells, of which the bivalves were, with very few exceptions, found only in single valves, pieces of wood partly charred, portions of wooden implements of Maori Manufacture, plaitings made of phormium tenax, and pieces of two broken polished stone implements were collected, whilst close to the bottom of the trench a few moa bones were obtained, amongst which several species were represented. On the top of the ashbeds immediately above the agglomerate, a small piece of a tibia of meionornis casuarinus, bleached and much decomposed was observed by me, which had been} lying close to a well preserved seal bone, possessing the light brown color the bones generally exhibited when exhumed, thus suggesting that the moa bone must have been brought into the cave from the outside after having become bleached and partly decomposed. In' order to test more fully the general character of the beds above the agglomerate, 1 gave directions to the laborers to work backwards from the cross trench, examining first the south-west corner of the cave, once more cautioning them to use the utmost care, and not to hurry over the examination.

With this work wc continued until October 3rd, when, after having looked carefullyover the specimens obtained, I could not divest myself of the conviction that in and below the agglomeratic beds, remains proving human occupation must be found. Amongst the objects obtained during the last few days, the workmen having turned over deposits covering an area of about 20f t by 150 ft wide, and advancing in a south-west direction, were seen some moa bones, the leg bones usually broken as for the extraction of the marrow, others of them calcined, all of them occurring only in the lowest bed. The overlying shell beds, as I shall ca'l them in future, consisted principally of the usual remnants of shells, together with some seal bones belonging to fur-seal and se: • leopard, portions of the Maori dog, all cv - dently from their Kitchen middens ; bones of fish, without exception belonging to Oligorua gigas, the hapnku ; also bones of small birds, of which the enumeration will le found in the lists attached to this mcmoii ; of the latter, those of Graculus punctatus, the spotted shag, were the most numerous,

Works of human industry were not wanting, as we obtained pieces of timber, evidently worked and planed down by polished stone implements, and upon one of which a coating of red color was still risible. Amongst the other objects made of wood hitherto exhumed, were several pieces of " Toa," a thin and long wooden spear, made of tawa (Mesodaphne tawa), a tree, which grows only in the northern part of the Northern Island. This spear is used by the Maories for shooting birds. For this purpose they form, as it were, a short tube around it with the one hand, through which, after taking proper aim, they jerk the thin spear up suddenly with the other. 2. A " Patu aruhe," or fern-root beater, made of maire (Santalum Cunninghamii) another strictly Northern Island tree. 3. The greatest portion of a whaka-kai, a wooden dish made of puketea (Atherosperma Novaj-Zealandise), used for placing fat birds in order not to lose the oil, or for the preparation of the juice of the tupakihi (Coriaria ruscifolia). 4. Several large pawa shells (Haliotis iris), in which the holes near the exterior border are filled with the fibres of flax or ti leaves, thus forming a vessel for the preservation of oil and other liquids. 5. A fish hook (matoa), used for catching hapuku, made from the wood of the kaikai-atua (Rhabduthamus solandri), another Northern Island tree. 6. A long, slender switch, of which part was broken off, and having at the other side a notch for tying. This is called a tokai, made of aka, one of the metrosideros or rata species. It is used to keep the entrance of a fishing net open. 7. Anotherpiece recognised as a taka ore kaka, a parrot perch made of puketea. 8. Several pieces belonging to a canoe, such as the puru (two specimens), made of manuka (Leptospermum scoparium), used to stop the holes in a large canoe for letting the water out, and a square piece of wood made of totara (Podocarpus totara) called takatiki whaka, used to fasten the sides of a canoe. 9. Also, portions of a matiha or fighting spear made of manuka and several others. I owe a great deal of this information to Hone Taahu and Tamati Ngakahu, two skilful Maori artificers, of the Ngatiporou tribe Poverty Bay> Northern Island, who for some months past have been occupied at the Canterbury Mnsewn ivith the necessary preparations foi the erection of a Maori house, carved and painted in! the original style of ancient Maori art, now test dying out. As they came from a part of New Zealand where the ancient native customs have been retained longest, their information may be considered very reliable. However, besides the few pieces of moa bones, which might accidentally have been brought into the cave from the ouside, their . was nothing which could prove that there had been a regular occupation by the moa hunters. I therefore set the laborers to work to go through the agglomeratic bed once more and I was delighted to find very soon that this time my expectations were not doomed to disappointment. After having passed through that bed, which I found to be here Sin thick, another ashbed of a thickness of 3 in was reached, in which I obtained several moa bones, some of them calcined, others in a splendid state of preservation, belonging to Euryapteryx rheides and Meionornis didiformis, as well as some pieces of charred wood. Proceeding with the utmost care, several large stones were reached, covered with a few inches of sand, some of them blackened or split by the action of fire, and placed in such a position as to show that evidently an oven had here been excavated in the sands ; these stones, like the remains of the meal taken here, having been trampled repeatedly over, and before the ash and dirt beds had been deposited above them. In digging round this spot I obtained the upper mandible of Aptornis difossor in a fine state of preservation, and a quantity of moa bones, also two wooden sticks made of pukatea (Atherosperma Novre-Zealandije) for producing fire. This simple apparatus, the only one found in the cave, has the peculiarity that fire instead of being obtained by friction lengthwise was procured by giving the upper stick a turning motion. However, I may add that this was not the only mode by which the moa-hunting population obtained fire, as in the same lower beds firesticks of the other kind were also found, resembling ia this respect those belong to the upper or mollusk-eating population, which are used at the present time by the Maoris, and are called kauwaki by them. About 4ft from this oven we came across some large pieces of eggshells, of which many had still the lining membrane attached, proving by their form of curvature that they are portions of a dinornis egg of very large'size. Towards the western side of the cave, partly buried in the sands, partly in the ash bed below the agglomerate, a well preserved skull of a fur seal, probably Arctocephalus lobatus, was obtained. Having 3aeen so far successful, I had the sea sands examined over a considerable space, and to a depth of 7ft, when water was reached. Since then I have been boring near the same spot, and found that the sea sands continued for another sft before the rock on the bottom of the cave was reached, thus showing that there is here a total thickness of 12ft of marine sands in the cave. The following shells were obtained in these sands, without doubt brought with them into the cave by the waves of the sea, viz : —Maetra discors, M. donaciformis, Mesodesma cuneata, Artemis subrosea, Turitella rosea, and fragments of some others, but no estuary shells. Section No 3 gives the details of this important point. On the surface we found—ft. in. 1. European deposits, dung of cows, goats, &c, wheaten straw, ashes ... 0 6 2. Shell bed (Maori) 0 9 3. Tussock and ash beds 0 4 4. Shell beds 1 4 5. Ash beds 0 2 6. Ditto, mixed greatly with shells (often very much decomposed)... 0 10 7. Ash and dirt beds (Lower Series) 0 2 8. Agglomeratic bed 0 6 9. Ashbed 0 3 10. Marine sands to \vater 7 0 10 10 Amongst the shell beds, occasionally blocks of rock?, often of large size, were met with, evidently fallen down from the roof, showing that since the formation of the agglomerate bed the cave continued to be still insecure. There was thus conclusive evidence of the moa hunters having used the cave occasionally as a, cooking plr.ee, whilst the absence of any shells proved, as I shall also show when "speaking of the numerous moa ovens amongst the small hillocks of drift sand near the entrance of the cave, that the population who exterminated our huge birds did not look with a favorable eye upon tho

food used almost exclusively by their suecGssors. However, 1 may here observe that near the oven in question a few valves of our common fresh water mussel (Naio Aucklandicus) were obtained, which must have been brought by the moa hunters into the cave. For the next few days we continued to excavate towards the end of the main cave, where, near the entrance to the small middle chamber, the marine sands sometimes reached the surface, European, Maori, and moa-hunter remains being here occasionally mixed with each other, trodden down into the sands by man or cattle. In a few more protected spots, ash and dirt beds to a thickness of several inches remained undisturbed above these sands. Advancing from the entrance to the middle chamber towards the big fragment of rock B , fallen from the roof which is 6ft broad, by 12ffc long, and 10ft high, and forms a remarkable feature in the cave, the artificial deposits soon became more considerable and full of interest. Close to the rock on its southern side they reached a thickness of nearly 3ft, consisting of : . ft. in, 1. Beds of European occupation, cowdung ... ° * 2. Shell beds (Maori) 0 10 3. Dirt and ash beds, with tussocks flax 0 * 4. Shell beds 0 9 5. Lower series dirt and ash beds ... 0 f> 6. Agglomerate beds, altering gradually again to ash beds upon the sands • 0 3 7. Marine sanas as far as excavated 3 0 5 11 In the lowest beds partly imbedded in the sands we obtained a great number of moa bones, belonging at least to six specimens, of which four were well represented, namely, three specimßns of Meionornis didiformis, of which two were immature birds, and one specimen of Euryapteryx rheides, also not yet full grown. Advancing towards the huge rock previously alluded to, I observed that one portion of its unequal southern undersurface stood above the sands, thus leaving a space below from which we took a number of things, amongst them a fine and perfect pelvis, and several leg bones of an immature specimen of Meionornis didiformis, some bones of the Maori dog, like the former, partly calcined and broken in having been used for food, as well as portions of skeletons of kakas, penguins, and some other birds. When examining the shell beds we had repeatedly found amongst them match boxes, small bones of sheep, and other remnants of European life, evidently brought into their present position by means of numerous ratholes, passing through these upper beds ; also here to thi3 small spot not filled up by the sea sands a few small European remains had found their way, which, if the mode of their transport had not been clear to me, might have been a great puzzle. Section No. 4 gives the details of the arrangement of the beds abutting against the rock. Also a considerable amount of drift timber was lying here, without doubt mostly brought so far back by human agency ; a great deal of it being charred or partly burnt, and all the evidence before me went to show that this spot, hidden as it was from the entrance by the huge rock in front of *t, had been a favorite camping and eating ground, both of the moa-hunting and afterwards of the shell-fish-eating populations.

For another week I continued to occupy the workmen in the south-en stern portion of the cave, but gradually advancing towards the western side of the cross trench, of which section No. 2 gives the details: Before reaching the trench at the spot marked O in the ground plan of the cave we came across a stand having been used for the stabling of a horse, which had been dug into the shell bed to a depth of several feet, in some spots reaching actually down to the marine sands.

This strange place for a stable was now mostly filled with horse dung and European kitchen middens well trampled down, and above them with a layer of the excrements of cattle.

Altogether in this part of the cave the beds had been much disturbed by the cave dwellers of European origin, so that in some instances moa bones were actually mixed up with bones of batcher's meat, lying now together in the disturbed shell beds. When advancing towards the point where the two main trenches crossed each other, the workmen observed standing vertically in the sands the remains of two much decomposed piles, having a diameter of about Sin, and which evidently had been deprived of their bark by means of a smooth stone implement, before having been placed in their present position. Evidently they Jhad been burned to the ground before the lowest dirt bed had been deposited, their charred ends standing scarcely above the level of the marine sands. Of these piles, the first was observed 15ft from the eastern wall of the cave and Cft behind the cross trench, the second opposite to the first on its south-western side, and at a distance of 12ft.

They were found during my absence, and the men not thinking their occurrence of sufficient interest, simply took them out—but noting their position, instead of leaving them standing until I had come down.— They reported that they had reached about 16 inches down into the sands.

During my presence I caused new excavations to be made around the spot where these piles were reported to have stood, but I could not get any other object except portions of one of tbe piles, which on examination proved to b • si mi (D 'crydium cupressinum ) In the agglomeratic bed in this southeastern portion of the cave, we obtained a number of moa bones, of which portions of a skeleton of Euryapteryx rheides were the most conspicuous. With the latter'also, the two rami of the lower mandible were found, but not the least portion of the skull ; in fact, the absence of any but very small fragments of skulls in all kitchen middens, shows that the brain of the moa was considered a great delicacy. Ht-re wc got again a few small pieces of obsidian and some chips and cores of flint together with similar rough and primitive tools, made of a hard and compact doloritc found in situ close to the cave. In the lower beds also senl bones, a few phalanges from the flipper of a small whale, fc-mes of birds still at present inhabiting New Zealand, were collected, amongst the latter those of the spotted shag and small blue penguin were most numerous. In the dirt and ash bed above the agglomerate we obtained a number of bones belonging both to our extinct and living vertebrate

fauua, amongst them the greater portion of the skeleton of a fur seal. In the shell beds above numerous Maori remains were found, amongst them a few fern-root beaters made of wood, some canoe pins, flax plaitings, all of which will be enumerated in the appendix 0. When examining first the two main trial trenches crossing each other at right angles in the centre of the cave, the absence of the agglomeratic beds was here noted by me, but I then thought that it might have been caused by the roof having in that part of the cave accidentally possessed a greater solidity. In this surmise I was still more confirmed by finding that in those spots the dirt and a«h bed was much thicker, lying hvro directly upon the sands, so that the former had a nearly uniform upper surface. However, when continuing the excavations across the cross ditch, towards the entrance of the cave, to the description of wlrch I shall devote another portion of this memoir, we found in the longitudinal trench a third trench a third pile, and observed that in the space between these three points and another point (where, however, no remains of a pile were existing), forming an oblong square, 36ft x 12ft wide ; the agglomerate bed was entirely missing and the influence was therefore natural that at one time a human dwelling of some kind had been standing here. My first impression was that the lake dwellers, in order to protect themselves from the pieces of rock becoming loosened at intervals from the ceiling, had built a strong roof resting upon four corner piles, which after the principal fall of rocks ceased had accidentally been burned to the ground ; but on closer examination it becama clear to me that the time during which the agglomeratic beds were formed was of such long duration, that it is impossible to assume such a frail construction having lasted so long. Moreover, one scarcely believe that a primitive race—and which evidently only at intervals inhabited the cave—before the agglomerate bed was deposited upon the marine sands, should act with such forethought and care. There remains only one other explanation, which I advance with some diffidence, namely, that the builder of the dwelling, whoever he may have been, excavated, not only the four holes for fixing the corner poles into the agglomerate, but actually lifted the same in the space between them ; against this however, it may be observed, that if such as we may presume unnecessary work was performed, the agglomerate bed ought not only to end abruptly round the former dwelling, but that the removed material having been thrown outside the thickness of the bed in question ought to be here much more considerable.

However, from the sections made during the process of the excavations it does not appear that the agglomerate bed was generally thicker outside this oblong square, or that it ended abruptly ; on the contrary, the same was found to thin out close to the intersecting lines, the ash and dirt beds becoming gradually thicker. The same was the case in some of the other porti ms of the cave, where the agglomerate was also occasionally missing, and I can ouly regret that when that portion of the cave towards the entrance was excavated, where a great thickness of the overlying shell beds had first to be removed, my official work at the Museum would not allow me to go so often to the ground as I should have wished. This question has therefore to remain an open one. Having reached, Saturday,the 19th October, the cross trench on the eastern side of the cave, and thus examined the whole southeastern portion, I began to continue with the excavations on the south-western side towards the termination of the cave in that direction.

Hitherto we had not been successful eifher in obtaining human bones, nor Maori objects of any value, which I had hoped might have been placed in a cache similar to those found in carefully excavated hiding-places in the moa hunter (and afterwards Maori) encampments at Rakaia. However, that evening we came a few feet from the south-western wall upon disturbed ground, and carefully taking off the material the skeleton of a Maori was reached, who, as section No 5 shows, had been buried a considerable time.

The aborigines who had placed the body there, had dug through the shell bed for about 8 inches, then 2 inches through the dirt and ash b«d belonging to the older series, and 4 inches through the agglomeratic deposit. They had then excavated the marine sands for several feet, and had placed thfl corpse in a sitting position in the grave thus formed, tied together with (lax, the face towards ihc wall of rock, covering it with part of the sands thrown out, the rest being thrown with the shells excavated around the spot. However, it was clearly visible that the ground had aftervvands been levelled, as it were, under the feet of human occupants, and about G inches of newly formed shell bed, being continuous, and level with the more distant bed, had been deposited over the grave, the whole being capped with 3 inches of European accumulations.

It is thus evident that the burial had not only taken place long before the Europeans came to the cave, but that the Maories continued for a considerable number of years to frequent the cave, and to take their meals there after that event had happened. This fact naturally leads me to assume that the cave was not constantly, or even regularly, visited by the Maories, and that its occupation occured only occasionally, and by different tribes of natives, because, judging from the character and superstitious of the Aborigines of the present time, we can safely infer that after the burial of one of them the cave would have become strictly tapu to all of those having any knowledge of the fact, at least, as far as the taking of meals is concerned.

This opinion is also shared by the Rev J. Buller, whom I consulted on this question, and who, having been living for many years amongst them in the Northern Island, is perfectly acquainted with all their customs. From this fact alone, and the conclusions therefrom, if admitted, we are obliged to assume that an enormous space of time was necessary to form this shell bed alone. The body, as before observed, had been tied together with flax, the knees being placed below the chin. Owing to the antiseptic properties of the sand, there were still some ligaments and skin upon the bones, and some hair upon the skull. The skeleton, which has been articulated by Mr P. Fuller, and now stands in the Canterbury Museum, belongs to a man of a height of nearly Oft, passed manhood. The ulna of the left arm is broken, and was only partially healed when the man died.

We are so accustomed to observe natives possessing a fine set of teeth, that it is rather striking to see that this aborigine must have

suffered very much from bad and distorted teeth. Thus we find that most of the premolars and molars are missing in the lower jaw, the alveoles being already quite absorbed. In the upper jaw the first molar on the right side, and the first molar on the left, are distorted inwards, their anterior surfaces being adherent to the alveoles which are developed into a slight bony outgrowth. Owing to a very remarkable distortion of the left molar, mastication was performed with its inner surface, which was worn. Examining the two smaller caves, we obtained also some moa aud other birds' bones, lying close to the surface of the sands, mixed up with ashes and other sicrns of human occupation : so that it is evident that casual visitors penetrated to these inner caves, probably to hide themselves from their enemies, and cooked their meals, or at least lighted fires. As these two smaller caves for years past have been visited by Europeans, a number of moa bones have, as I understand, been carried away, having been observed amongst the sands. To he continued.