Art. XLII.—Redcliff Gully, Rakaia River.
[Read before the Philosophical Institute of Canterbury, 4th December, 1912.]
In Redcliff Gully, on the south bank of the Upper Rakaia River, there is a small outlier of pink-coloured limestone, which derives some importance from the fact that Captain Hutton, in his paper on the “Origin of the New Zealand Fauna and Flora,”* cited the occurrence as a proof of his contention that the Rakaia Valley during early Tertiary times was eroded to a deeper level than at present, and that there could have been no plateau to the south of the Rakaia, as demanded by Haast in his explanation of the former extension of our glaciers. McKay also mentions† that the Redcliff limestone was placed on or near the line of one of the great structural faults which he postulated as being responsible for the preservation of several of the small outliers of Tertiary limestone and other beds which lie in the remote recesses of several of our alpine valleys. He suggested that they were remnants of a wide extent of Tertiary beds which had been faulted down, and so escaped to some extent the erosive action of the great Pleistocene glaciers. Seeing that the occurrence was of somewhat varied interest, and that no accurate account of the locality was available, the present author examined
[Footnote] * “Annals and Magazine of Natural History,” 1884 and 1885.
[Footnote] † Repert of the New Zealand Geological Survey for 1890–91,” p. 15.
the beds towards the end of the year 1912, and this paper gives an account of his conclusions arrived at on that occasion. I must take this opportunity to express my obligation to Mr. W. F. Robinson, Lecturer on Surveying at Canterbury College, who accompanied me on the visit, and gave material assistance in determining the relative position of various points of importance, and to whom I am indebted for the sketch of the locality on page 339.
Position and General Features of the Locality. (See map and topographic sketch.)
Redcliff Gully is situated on the south bank of the Rakaia River, about fifteen miles above the Gorge Bridge, and forms part of Mr. Gerard's Double Hill Station. On the opposite side of the Rakaia lies Lake Coleridge, and the pipe-line of the power scheme connected with it reaches the Rakaia about two miles below the mouth of the gully, so that if at any future occasion it is intended to apply this power for the manufacture of calcium carbide or other calcium compound the position is eminently favourable for carrying it to a successful issue.
The level of the bed of the Rakaia is at this spot approximately 1,260 ft. above the sea, while the highest point of the limestone exposure is 3,060 ft. The mountains in the immediate vicinity reach to nearly 7,000 ft., and about five miles to the south-east is Mount Hutt (7,180 ft.), the highest of all the peaks in the range which rises directly from the Canterbury Plains. At the head of the gully is Redcliff Saddle (2,980 ft.), from which an easy track leads south to the head of a branch of the North Ashburton. This pass forms a well-marked depression in the range which bounds the Rakaia on the south, and shows in its flowing contours and U-shaped cross-section unmistakable signs of severe glacial erosion; glacial shelves are common, and morainic matter is scattered over the surface and at times masks with a covering of variable thickness the limestones and beds associated with them. Signs of glaciation are also clearly visible on the mam valley of the Rakaia in the vicinity. Stranded lateral moraines; roches moutonnées rising from the floor of the valley; the silt-filled lake-bed which has been now drained by the erosion of the river; and the imposing wall-like sides of the valley with its steep even slopes rising to 2,000 ft. above its floor; the truncated spurs; and just across the river the great barrier of glacial drift, full of scratched stones, forming the dam across the lower end of the lake, and through which passes the pipe-line for the power scheme—all furnish undoubted proofs of the severity of the glaciation. It is evident from the slope of the glacial shelves in the Redcliff Pass and from the general form of its valley that an overflow of ice from the basin of the Rakaia took place at the height of the glaciation. On the shoulder of the spur from Mount Hutt which juts out into the angle between the tributary valley and the main stream this action is very pronounced. One can easily picture the great ice-stream impinging on this part of the mass of Mount Hutt, and causing pronounced abrasion as the ice was packed at the angle before it flowed away either down the main stream or was diverted towards the Ashburton.* The thickness of ice must have reached 2,000 ft., and it probably exceeded this amount.
The block of reddish limestone from which this locality takes its name lies a little back behind the general line of the valley-walls, and no doubt
[Footnote] * A similar overflow of ice must have occurred higher up the Rakaia Valley, at the junction of the Lake Stream with the main river.
owes its preservation primarily to this position. On both sides of it streams have cut down along the junction of the limestone and the underlying greywacke, so that it is partially isolated, but the level of its upper surface is practically that of the saddle, and there is no marked physical feature isolating it in this direction. The front block, whose thickness exceeds 150 ft., forms a great cap slightly tilted forward, a position perhaps due to creep of the beds towards the Rakaia. The creek on the western side of this block is known as Packer's Creek, named so from a narrow packtrack leading along its side, while the stream on the eastern side, the larger of the two, is known as Cascade Creek. The latter rises from the saddle itself, and on its steep western bank is a slip in which are exposed all the beds of the series associated with the limestone. Between these two creeks the limestone forms a prominent scarp, extending round in a rude semicircle with its convexity facing the Rakaia. The position of the two creeks has been naturally determined by the line of junction of the limestone with the underlying greywacke, but Cascade Creek is placed on the axis of a small anticline of the folded greywackes.
The relative position of the beds occurring in the locality can be best studied in a section that runs in a westerly direction from Cascade Creek, up the slip mentioned previously, over Redcliff Hill, towards Packer's Creek. The underlying beds, of probable Mesozoic age, in Cascade Creek, consist of slaty shales with bands of greywacke having a northerly strike and a nearly vertical dip (see fig. 1). The arrangement here is distinctly anticlinal. The direction of the strike is remarkably variable, for on the southern side of the slip the dip is to the south-west, at an angle of 60°. Similar beds also occur in a roche moutonnée in the bed of the river, with a strike a little south of east and an approximately vertical dip. This variation can be seen at times in a very small section, and small faults and marked contortions are frequently visible, emphasizing the intensity of the earth-movements which produced the folding.
Fig. 1.—Section along Line AB, from Packer's Creek to Cascade Creek.
Length, 1 mile. Direction, N.W. to S.E. 1. Greywackes, &c., varying dip and strike. 2. Greensands (darker towards top), 500 ft. 3. Light-coloured sands, with darker glauconitic layers, 150 ft. 4. Marly and sandy beds, whith concretionary layers, 20 ft. to 30 ft. 5. Limestone, 150 ft., probably more. 6. Sands and shell-beds, over 50 ft. 7. Moraine.
Resting unconformably on these greywackes are the beds with which we are more nearly concerned. The following sequence is exposed in the slip:—
Greywackes and slaty shales.
Greensands, with an estimated thickness of 500 ft. These beds are light-green below, but become darker in the higher parts. They strike
N. 70° E., and dip at an angle of 55° towards the west. The angle of dip gets slightly steeper in the higher levels. The only sign of fossils that we saw were fragments of oyster-shells occurring towards the top. Their base rests on a highly eroded surface of the greywackes and slaty shales.
Light-coloured yellow and white sands with an estimated thickness of 150 ft. These are at times quite incoherent, but they are occasionally cemented together. In their higher levels they are interstratified with argillaceous bands, which become decidedly calcareous towards the top, where there is a thin band of calcareous sandstone, succeeded by a bed 20 ft. in thickness of sandy marl. The actual contact of this with the overlying limestones was not visible, owing to slip-accumulations.
Limestone.—The limestone which follows conformably is pinkish in colour, somewhat crystalline in character, flaggy in the lower parts, but more compact and even and breaking into cuboidal blocks towards the top. It contains numerous fragments of Mollusca and Echinoderms, but it was found impossible to extract them. The thickness of this bed is at least 150 ft. In the slip the strike is N. 70° E., and dip to the west at an angle of from 60° to 70°. The limestone forms the top of Redcliff Hill, but at the actual summit it is capped by a thin veneer of morainic matter.
Calcareous sandstone and shell-beds.—These lie conformably over the limestone, and attain a thickness exceeding 50 ft., the actual amount being uncertain. They become more sandy in their higher parts, and are interstratified with layers of concretionary sandstone. These contain welldefined bands composed of shell-fragments, but it was difficult to get good specimens for determination. The following genera and species were recognized: Glycimeris globosa, G. laticostata, Cuccullaea alta, Chione mesodesma, Crassatellites amplus, Mesodesma australe, Turritella sp.?, Ancilla australis, Voluta sp.?, Polinices gibbosus, Struthiolaria spinosa, Struthiolaria sp. ?, Siphonalia mandarina.
Morainic deposits.—These cover up the surface in many places to a varying thickness. The material is invariably of greywacke, &c., from the Mesozoic rocks, and consists chiefly of boulders and blocks of large size. These beds lie unconformably on both the limestones and the older rocks.
This is the complete series as far as the locality is concerned, and, with the exception of the underlying greywacke and the overlying glacial material, it is conformable throughout. The age of the beds is determined with a certain amount of accuracy by the fossil-content of the sandy layers overlying the limestones. The species identified show that the fauna is of the same character as that occurring in the Pareora beds of North and South Canterbury, which are usually assigned to the Upper Miocene.
No other exposure in the locality shows the series as completely as that in the slip, since the general surface of the rocks in position is masked by material detached from the limestones and by morainic matter. There seems, however, to be no doubt that under the limestone scarp facing the Rakaia the sands of the lower part of the series occur, but they are quite hidden. It is probale, too, that they have formed a slipping surface, on which the great block of limestone which overlooks the bed of the river has been slightly canted forward (see fig. 2).
Captain Hutton states that there is no disturbance of the beds, a conclusion based, I believe, on an incomplete examination. The steep inclination of the series as seen in the slip facing Cascade Creek is continued towards the west, and on following the outcrop of the limestone towards the south-west it is found to swing round and become steeper still. There is a gap then in the line of outcrop, but the limestones appear again on the east side of the upper part of Packer's Creek, and here they are in all probability almost vertical, if not slightly overturned. This observation is, however, somewhat doubtful, as the planes of stratification could not be determined with certainty. The beds appear to form a basin-shaped syncline, although I could see no sign of the sandy beds which overlie the limestone in the position that they should occupy on its western wing, owing to the morainic and other detrital accumulations on the surface. This appears to me to be the most satisfactory explanation of the structure of the beds in this part of the area. On the north side of this syncline there is a block of limestone which also shows synclinal structure with the axis running nearly north. On the western flank the beds have a slight dip to the south-east, and are distinctly overlain by the shell-beds which usually cap the limestone. They are thus in the position required by the presence of the syncline. The limestone on the western wing of this syncline forms a high escarpment fronting the middle part of Packer's Creek. Its southern termination is 500 yards distant from the block in the upper part of Packer's Creek, and is 300 ft. above it. It seems, therefore, that a fault is present, the effect of which has been to raise the middle block of limestone relatively to the southern block, or to depress the southern block relatively to the middle block (see fig. 3). This fault is in the nature of a flaw or blatt cutting the main direction of folding nearly at right angles. It is unfortunate that no exposure of rock allows the structure of the syncline to the south to be definitely determined, but this appears to be the only explanation, especially when taken in conjunction with the structure of the most northerly block—viz., that overlooking the Rakaia directly. This also appears to be separated from the middle block by a flaw or blatt, the effect of which is to cause an apparent downthrow of the block to the north. This fault does not pass right through the series to the east, but its effect is noticeable in the sagging downward of the limestone scarp on the east from its greatest height in the slip facing Cascade Creek as it is followed north to towards the Rakaia. On tracing
this, scarp round, it is found that the dip gradually flattens out till the cliff is reached at the northern end, where the beds are practically horizontal. On following the scarp towards Packer's Creek the dip changes gradually, and the inclination is to the east at very low angles, and this gradual swing continues till the fault is reached. This portion of the area thus exhibits a synclinal arrangement.
Fig. 3—Geological Sketch-Map Of Limestone, Etc., Beds In Redcliff Gully, Rakaia Valley.
Surface deposits not indicated. Scale, 1½ in. to the mile (approximately). 1. Greywaeke, slaty shale. 2. Greensands. 3. Sands. 4. Marl. 5. Limestone. 6. Calcareous sands and shell-beds.
It seems, therefore, that the beds have been subject to folding agencies which have operated in a roughly east-and-west direction, and that under their influences the middle block has been raised and cut off on its western margin from the neighbouring blocks whose edges preserve their proper alignment. This is the only sign that I observed of the beds having been affected by faulting. They show just that intensity of dislocation
which can be observed in other parts of Canterbury, notably in the Trelissick basin at Castle Hill, distant sixteen miles to the north-east. Their elevation—fully 2,000 ft. above similar beds at the Curiosity Shop, three miles below the Gorge Bridge—implies that a considerable amount of differential elevation has taken place in the central parts of the Southern Alps, a fact which is also confirmed by a consideration of the Trelissick basin. This elevation, taken in conjunction with the slight folding which the beds exhibit, shows that the movements which formed these mountains had not ceased in late Tertiary times, although, judging from the absence of earthquakes associated with the chain, they have apparently ceased by now.
One statement of Captain Hutton's should not be passed over without comment—viz., that the mound in the bed of the Rakaia known as Castle Rock is composed of limestone. Although a casual examination of it from a distance, especially as it is coloured slightly red in places, suggests that it may be a fragment of the Redcliff limestone, a close examination of it shows that it is entirely composed of greywacke and slaty shale, with a strike a little south of east and a vertical dip. The rock is only 10 chains long, and from 3 to 4 chains wide, and about 25 ft. raised above the riverbed, so that there can be little doubt that a mistake has been made. No other rocks could be seen in the river-bed to which Captain Hutton's description could apply.
His argument as to the erosion of our valleys in pre-Tertiary times, as based upon this evidence, must therefore be taken with a considerable amount of reservation. At the same time, it must be admitted that the surface of the Alps had suffered erosion before the end of the Cretaceous period, and that during a subsequent time of depression the sea invaded the inlets thus formed, and portions of the marine beds then laid down were, in specially favourable localities, preserved from the erosion of glaciers and other denuding agents. Whether faulting has contributed to the preservation of these beds is still an open question as far as Canterbury is concerned. It must be admitted that a strong case can be made out in regard to the Trelissick basin, but the Redcliff outlier furnishes none, as far as my own observations go.
Further, there is no sign in the conjunction of beds of dissimilar lithological character, or in the landscape features, that any major line of faulting occurs connecting the Trelissick basin and Redcliff Gully. The former is a basin the result of folding, or prehaps a senkungsfeld, but it is quite isolated, and surrounded on the western and south-western sides by the uniformly high (between 6,000 ft. and 7,000 ft.) Craigieburn Mountains, in which there is no notch or other landscape peculiarity demanded by such an earth-movement as one accounting for the simultaneous formation of both of these limestone outliers would necessarily require.