Art. XXXVII.—On Avian Remains found under a Lava-flow near Timaru, in Canterbury.
[Read before the Philosophical Institute of Canterbury, 6th November, 1890.]
At the end of last year I received information that some avian bones had been found under a lava-stream near Timaru. On the 3rd February of this year (1890) I paid a visit to the district. Mr. Hogben, head master of the Boys' High School,
had the goodness to meet me at the Timaru Railway-station, and afterwards to introduce me to Mr. Stubbs, the Secretary of the County Council, by whom these bones were first observed and their discovery publicly intimated, and at whose office in town I examined the specimens he had collected. Of these, the largest, nearly 8in. in length, were undoubtedly portions of Dinornis bones of one of the greater forms; but to what species they belong I was unable, on so cursory an examination, to determine; nor were the fragments, indeed, in many cases, in identifiable condition.
These fossils were discovered in the ravine of the Gleniti, an affluent of Saltwater Creek, which is in turn a tributary of the Otopara, in what is known as “Number-Two-Mole Quarry,” because out of it that breakwater of the Timaru Harbour works was being constructed. Mr. Hogben and Mr. Stubbs obligingly accompanied me to the locality, distant from the town a mile or two in a south-south-westerly direction. There I met the manager of the works, to whom my thanks are due for his kindness in explaining to me the operations that had been going on for the past two years, and the character of the rocks and strata he had cut through. Some account of my observations on this occasion and of the bones found ought, I feel, to be put on record in the Institute Transactions.
The quarry formed a crescentic excavation, commencing at the front of the escarpment of the ravine, some 40ft. or 50ft. above the level of the stream, and had in its centre been worked out northwards to the depth of about 150ft. The section exposed at the time of my visit had continued all the way from the quarry-front with little alteration. The accompanying sketch (Pl. XXXVI., fig. 1) shows in a diagrammatic form the succession of the strata, and will render intelligible the description of the beds, which occur as nearly as possible at the same levels on both sides of the ravine. The uppermost layer, A, is composed of tenacious yellow clay, and forms the surface of the ground in this district. It presents all the characters of the clay flanking the Port Hills, near Christ-church, and is, I believe, the formation designated by the name of “loess” by my predecessor, Sir Julius von Haast. This stratum extends to a depth of from 26ft. to 28ft., and is supported on a deep dolerite belt, B, which is the raison d'etre of the quarry. For a few inches at its top and bottom surfaces the dolerite presents a more vesicular and slag like character than the rock composing the bulk of its intervening mass, which is fine-grained and crystalline. Through its centre runs longitudinally a band of still finer grain, more highly crystalline and more brittle than its over- and underlying portions. The quarriers had given it the name of “flint-band,” from its hardness, and because, when blasted, it broke
up into small fragments, which presented sharp fracture-edges, and emitted a “clink” like phonolite, or the clang of flintchips, when struck. The width of this band varies in different parts of the division, B, from 2ft. at its widest gibbosities to a seam of such linear dimensions as to be difficult to trace, till a little further along it widens out, and then tapers off again. Without a closer than the ocular examination which I could alone give it, I could detect little difference in mineral composition in these two portions. In some places pockets of fine clay in thin laminated plates, intersected by cleavage-lines, ran perpendicularly through the dolerite-bed. These pockets exhibited no signs of change from heat, and must therefore be of age subsequent to the dolerite overflow. They occurred in the middle or towards the base of the stratum, and in one place extended down to the laterite. After careful examination I detected cracks in the dolerite leading to its surface, by which it was evident that the loess-bed above, while being deposited, had supplied in the form of liquid mud the packing (subsequently slightly compressed) for the fissures and vacuities, which had occurred in the lava-flow beneath, This dolerite-bed, which is about 42ft. to 45ft. in thickness, rests on the third stratum, C, which must have been at the date of the lava-flow the surface of the ground. The lower surface of A gives no evidence of having suffered from the effects of heat, so that it is apparent that the dolerite band is non-intrusive.
The belt, C, is composed of consolidated clay, presenting many of the characters of the present uppermost bed of loess, and is from 6ft. to 7ft. in depth. The surface-layer contains a few rounded red pebbles, and appears to have suffered a change which looks like fusion for about ¾in. below its upper surface. It does not present any recognizable evidences of having been deposited under water. It exhibits, on the contrary, wherever I have examined it, the characteristics of the Banks Peninsula deposit, with the peculiar tubuli seen in the loess, and, as in it, the surfaces of these tubuli are coated with what to the eye (without chemical examination) is a hardened gelatinous silicate, but what may perhaps be an extremely fine deposit of indurated mud. Besides the small tubuli there are many impressions of a large size, varying from ½in. to ¾in. in diameter, whose sides are incrusted with an extremely thin, black, but non-carbonaceous substance, not improbably an iron-compound. These, from their section and irregular form, are evidently the empty moulds of roots or branches. This layer, C, at its junction with the dolerite, has been converted, for depths varying from 11in. to 14in., into laterite from the heat of the lava-stream in its incandescent state.
The quarry-face does not exhibit the bed, D, next below the clay; but the manager of the works showed me where he had sunk a pit in the floor of the quarry, the debris of which was lying on the surface as a heap of red shingle and gravel, which was met with a few feet below the surface, and continued as far as his digging extended.
The quarry has been worked to obtain rock in large blocks for the breakwater; and to this end blasting has been carried on from long tunnels driven at right angles into the face of the quarry in bed C, underlying the dolerite. During the driving of these tunnels, partly in its softer portion below the laterite, and partly also in the laterite itself, the osteological remains which Mr. Stubbs showed me appear to have been found, and certainly those found later on by Mr. John Miller, the foreman of the blasting operations. This intelligent quarryman was good enough to permit me to remove to Christchurch for some weeks, for examination and identification, the bone-fragments he had exhumed. He showed me also the site of the tunnel in which he had discovered them. I was not myself fortunate enough to find any identifiable bones, or see any of them disinterred; but from amid the excavated clay I picked out a few chips showing undoubted osseous structure resembling that of the bones already found.
Most of the bone specimens were in a perfect state of preservation, and those of them that had been imbedded in the clay below the laterite looked, when cleared from their matrix, no older than the best-preserved moa-bones from caves or from turbaries. A few seemed to have been subjected to a greater or less degree of calcination. The collection made by Mr. Miller contained the following recognizable portions:—
1. The right femur (fig. 2) of a species of Apteryx indistinguishable from Apteryx australis. This specimen is imbedded in the lower part of the laterite layer, and appears to have suffered somewhat from heat. It is the most perfect of the bones recovered yet. It is entire except for the loss of part of the wall at its outer face—whereby its internal texture is revealed—and for a slight deformity due probably to pressure.
2. A portion of the distal end of a tarso-metatarsus of a species of Dinornis, which might have belonged to D. curtus.
3. The proximal end of a tarso-metatarsus of a species of Dinornis, which might have been owned by D. curtus or D. didiformis.
4. A fragment (fig. 3) of the pelvis of a species of Dinornis smaller than those to which 2 and 3 belonged, which fits very well with the corresponding part of D. oweni.
Besides these specimens there was—
5. A portion of bone, ¾in. in length, remarkable for the smallness of its medullary cavity and for the density and thickness of its walls—a fragment which I am in doubt whether to assign to a bird or to a reptile without the opportunity of microscopic examination. If it be a bird's bone it is neither a moa's nor an Apteryx's.
Some time after my visit to Timaru Mr. Hardcastle, of that town, was kind enough to send me a block of the dolerite having imbedded in it a portion of a pseudomorph of a round block of wood, originally measuring, he tells me, 15in. long by 1 ½in. in diameter, which had been entombed in the lava-stream.*
As I have remarked above, the over lying bed, A (which I find it difficult to distinguish by any marked characters from the bed under lying the dolerite), presents all the characters of the clay-deposit on the hills of Banks Peninsula. The origin of this clay in different parts of Canterbury has been ascribed (1) to marine deposit at the mouths (it is suggested)† of “the great rivers;” (2) to the same agencies by which Richthofen has explained the vast deposits of loam—the loess—in China—namely, by subaerial denudation and (chiefly) by the accumulation of the dust carried by the wind across the land, which is retained by the grass, whose roots also, in decaying, assist in raising the level of the ground.
To the marine origin of this formation I am entirely opposed, as it appears to me too extraordinary that a clay capable of preserving the fine tubuli (supposed to be the vacant moulds of decayed roots) should not have preserved a single stray leaf or a fern-frond; and that a current which was dropping down heavy moa-bones here and there should not also have deposited a branch or a fern-stem, or small pebbles in its course; and that there should be so remarkable an absence of unmistakable water stratification. That portion of Professor von Haast's theory which ascribes the accumulation of this loess to subaerial decomposition of the underlying volcanic rocks appears to me to be the sounder theory. But that it owes much of its depth to windborne dust seems to me very improbable, for the loess occurs abundantly on the seaward as well as on the landward side of the hills. In his “Geology of Canterbury and Westland,” page 356, von Haast, speaking of the strata traversed in the construction of the Lyttelton tunnel, observes, “This bed of loess changes gradually before we reach the volcanic rock to a true slope-deposit, consisting of fragments
[Footnote] * This has been submitted to Sir James Hector for analysis at the Colonial Laboratory.
[Footnote] † Hutton: Trans. N.Z. Inst., vol. xv., p. 413.
of rock more or less rounded, the lines of junction being often impossible to trace owing to the decomposition of the volcanic rocks immediately below the slope-deposits.” This atmospheric decomposition is certainly to a great extent the true, and personally I believe almost the sole, explanation of this formation on Banks Peninsula, and I feel confident that it explains also very largely that occurring at Timaru. In numerous places on Banks Peninsula is to be seen the volcanic rock with its external surface for some distance down absolutely converted into clay in sitû, and requiring only a touch of the fingers to cause it to fall down and mingle with, and be indistinguishable from, the clay, already fallen, covering the surface of the hill. Below this entirely-decomposed layer the rock is seen to be greatly weathered, rotten, and yellow, and only not disintegrated; further down it is less and less changed, till the unaffected rock is reached. This decomposed rock, as clay, becomes under the varying atmospheric changes now a viscid fluid, now a hardened clay, constantly gliding downwards. One has only to traverse or sail round the peninsula to see on every hill-face the most marked evidences of this continual downward movement, in the wave-like terraces, and in the scars or breaks in the surface caused by these landslips, which are constantly increasing mechanically the depth of the deposit on the lower parts of the hills, while the chemical action of the atmosphere is incessantly keeping up the supply not only in the parts exposed to the air, but also in the rocks below the deposit. These incessant landslips, sometimes extensive, but more generally of no great magnitude, are, I feel confident, the cause of the stratification—in my opinion, the false-bed-ding—which has been observed by Professor Hutton in a section near Lyttelton. In a considerable section recently (November, 1890) re-exposed at the Farnley brickfields, at the base of the Port Hills, not far from Christchurch, the stratification-lines, which on close examination reveal none of the certain marks of water-bedding, are most distinctly seen to correspond with the demarcation-lines of different landslips. These landslips, the higher gliding over the lower, would, as is evident, entomb whatever object might be on the surface; and thus can be easily explained the presence in deep layers of moa-bones, &c., and the rootlets of plants.
Both Professor Hutton and Dr. von Haast have remarked on the absence in these Canterbury loess-deposits of the “loess-babies,” or marly concretions, seen in the loess of China. I do not myself see the significance of their presence or absence; but I may mention here that I have gathered many hard marly nodules of curious shapes in the Banks Peninsula deposit since I went to reside on its hills.
In the Gleniti quarry section I could detect no stratification in the uppermost stratum, A, nor in the bone-bearing bed, C; and I believe the upper deposit certainly to be due in chief part to the decomposition of the underlying dolerite going on for a long period. The same explanation should apply to the lower also. I can see no evidence of its being other than a subaerial formation. There is evidence also, I think, strongly pointing to the dolerite outflow having been laid down not, as Sir Julius von Haast has thought, under the sea, but on a land-surface.
I come now to determine the age of this bone-bearing bed. According to Professor Hutton,* “there is no trace of volcanic action having taken place in the South Island during this [the Wanganui] system or later.” As the “Wanganui system” is referred in the same paper to the Pliocene period, it follows that the age of the bones below the lava-sheet at Gleniti must be, according to Professor Hutton, of Miocene age. The bed underlying the bone-bed is, as I have already said, a rough red shingle which belongs, with little doubt, to “the Moutere gravels” that underlie the Canterbury Plains—that is, the alluvial fans. Professor Hutton† considers these gravels to belong to “the upper part of the [Pareora] system.” In the classification, however, adopted by the Geological Survey, the Moutere gravels are assigned to Pliocene age.‡ The clay overlying them may therefore probably be of newer Pliocene, or even Pleistocene age. Volcanic activity would consequently appear to have continued in the South Island, as it has done in the North Island, down to times much later than Miocene.
Mr. Mantell found in 1848 a fragment of a bone, impossible of certain identification, but supposed to be bird's, and from its size a moa's, in a septarium from Moeraki Beach, in Otago; but, as reptilian bones have since been obtained from the same horizon, it was probably of this nature. The bones described in the present paper are, however, the first that can be ascribed without hesitation to Dinornis and Apteryx, and are the oldest certain remains of these interesting birds.
Note.—The report “On the Geological Formation of the Timaru District, in reference to obtaining a Supply of Water,” illustrated by sections, made to the Provincial Council of Canterbury in 1865 by the late Sir Julius von Haast, may be consulted for geological sections of the strata exposed in the neighbourhood of the Gleniti Valley.
[Footnote] * Quart. Jour. Geol. Soc. Lond., p. 217.
[Footnote] † Loc. cit., p. 209.
[Footnote] ‡ Rep. Geol. Surv., 1878-79, p. 3.
Description Of Plate XXXVI.
Fig. 1. Section of strata in Gleniti quarry.
Fig. 2. Femur of Apteryx australis.
Fig. 2A. " " showing internal structure.
Fig. 3. Fragment of pelvis of Dinornis, sp.?