Art. VI.—Structural and Glacial Features of the Hurunui Valley.
[Read before the Philosophical Institute of Canterbury, 5th December, 1917; received by Editors, 31st December, 1917; issued separately, 24th May, 1918.]
The Hurunui Valley is one about which little has been said in geological literature, though it is one of the most interesting of the main river-valleys of Canterbury, not only for its structural pecularities, but also for the glacial features of the country in the vicinity of its headwaters. The comparative neglect is perhaps due to the relative inaccessibility of its higher parts owing to the absence of roads, though before the discovery of Arthur's Pass it was the recognized route from Canterbury to Westland, while the lower portions were, till the opening of the Cheviot Settlement and the completion of the Waipara-Cheviot Railway, quite off the main lines of communication.
In 1865 Haast made a journey up the river across the island, an account of which is given in his Geology of Canterbury and Westland (1879), including a general description of the chief landscape features of the upper part of the basin. In 1871 he visited the middle Hurunui, and furnished a report (1871), in which he referred to the basin of the Mandamus, the main northern tributary of the Hurunui. Hutton (1877, pp. 34, 35; 1889) also gave some account of the locality, and dealt with the origin of the Hurunui Plains (1877, pp. 55, 56). This is practically all that has been written on the features of the main valley, except a brief reference by myself (1915, pp. 347–48) to the formation of the Waiau-Hurunui intermontane basin. Of course, there is abundant reference to the country to the north and south of the river, such as the Pahau Valley, and to the interesting stratigraphical questions connected with the Waikari and Greta Valleys, but a consideration of these is foreign to the scope of this paper. It is intended to give an account of the general geology of the basin only in so far as it is connected with its peculiar structural and glacial features.
(See map, fig. 1.)
The chief stream of the Hurunui rises in the main chain of the Southern Alps, and flows east between bush-clad mountains whose height approximates between 5,000 ft. and 6,000 ft. till after a straight course of some eighteen miles it empties into Lake Sumner. This is a fine lake, seven miles long by one a half wide in its widest part, 1,724 ft. above sea-level. Thence the Hurunui flows south-east for about eight miles, and receives on the south a tributary almost as large as itself, called the South Branch, the main stream being sometimes called the North Branch. In this part of its basin are several small lakes, the most important being Lake Katrine (which is practically an indentation near the head of Lake Sumner), Lakes Taylor (1,914 ft.) and Sheppard (1,916 ft.) in a valley between the two branches, and Lake Mason in a side valley of the South Branch.
Below the junction of the two main streams the valley continues for nearly three miles in a south-easterly direction between somewhat precipitous mountain-sides, and then turns east and passes through a deep, narrow, picturesque gorge, locally known as Maori Gully, and believed to be the scene of an engagement between two Maori tribes in early days.
The river then flows north-east for nearly ten miles through a hilly region in a narrow channel cut in the floors of detached basins and deeply incised in the ridges dividing them, till it reaches the Mandamus River.
Just below the junction with this stream the direction of the main river turns through a right angle and it enters the Hurunui-Waiau basin, flowing for about ten miles through an aggraded flood-plain till it receives the Waitohi on the south. It then makes a sudden turn and runs north-east along the southern edge of the Hurinui Plain, receiving the Pahau River on the north; but after a course of about eight miles it again breaks through a mountain barrier in a south-easterly direction and receives the Waikari River on the south and the Kaiwara Creek on the north, immediately after which it breaks through yet another mountain barrier and debouches into the Greta-Cheviot basin, across which it flows in a broad bed with terraced banks in an easterly direction till it discharges into the sea after cutting a somewhat deep gorge through a rocky bar just at its mouth.
The most striking feature of its course as a whole is the peculiar zigzag direction of the reaches which characterizes the middle part of its basin. These zigzags have alternate north-west and south-east and south-west and north-east arms, and it is their special relation to the grain of the country in some places and their absence of relation in others which is peculiar.
Basement Rocks Of The Area.
The characteristic basement rock of the region is a greywacke such as is typically developed in the mountains of Canterbury farther south. This is usually of the hard grey facies, but slaty greywackes also freely occur, which break down under the weathering agencies into clay and form a covering on the mountain-slopes. On these forest once became thoroughly established, but it has been largely destroyed in the higher parts of the river-valley by the grass fires of settlers. The greywackes have a general north-easterly strike, with local variations. Beds of dark-red slaty shale also occur, as well as occasional outcrops of volcanic rock. Basalts and andesites occur near Lake Sumner on the Crawford Range, and there is in the Canterbury Museum a specimen from the same area marked “eurite” by Hutton. Basalt pebbles occur in the Seaward River, a tributary coming in from the south about three miles below the junction of the two branches, and similar rocks occur in position between it and the Waitohi River.
The most interesting occurrence is near the Mandamus. About a mile above its junction with the main stream a massive intrusion of augite syenite occurs in the greywacke. This has a general north-easterly trend, and it appears to have the character of a sill, being approximately parallel to the dip and strike. Its thickness is more than 200 ft., and it extends over a mile in length. Associated with it are trachyte dykes, and flows of augite andesite occur in close proximity. Hutton was of the opinion that the syenite represented the core of a volcano of which the andesite was the effusive representative. But angular fragments of the syenite are found in the andesite, in some places in considerable quantity, so that the intrusion of the syenite was evidently anterior to the andesite. We have therefore, in order of time, (1) greywacke, (2) syenite, and (3) andesites. In this district, too, there are basic volcanic tuffs having in their higher levels a calcareous tufa facies passing into a true limestone; but the volcanic beds are much better developed to the north-east, in the Pahau and Culverden districts, where there are interstratifications of volcanic material between beds of limestone. The occurrence at the Mandamus points to several periods of vulcanicity, the channel opened by the syenite affording a passage for later magmas.
Volcanic rocks have exerted little effect on the area covered by the river basin, in which greywacke is now by far the most dominant member; but at one time the lower parts of the valley were covered with a veneer of Tertiary sediments, remnants of which are still to be found. These later beds have all a north-easterly strike, so that they cut across the river at a average angle of 450, and at present they occupy separate compartments of the valley, cut off from adjacent ones by ridges of greywacke. These isolated areas are as follows: (1) In the Dove River, a tributary of the Mandamus coming in on the east; (2) in the Hurunui-Waiau basin; (3) in the Waikari-Kaiwara basin, or rather trench; (4) in the Greta-Cheviot basin. The special features of these may be taken in turn.
(1.) The Dove River Area.
The Dove River basin is important, not from its size, but because it gives an indication of the origin of the landscape features of a considerable area of hill country forming a kind of platform or terrace at the base of the
higher region to the north-west. The Tertiary beds here consist of the following:—
Limestone, passing down into
Calcareous breccia with volcanic fragments.
Volcanic tuffs and ash-beds.
Sands with concretionary layers.
The lower parts of these probably contain coal, seeing that an adjacent stream is called Coal Creek. (See also Haast, 1871, p. 30.) The limestone is crystalline in texture, but shows traces of bryozoan forms on its weathered surfaces. The strata are bent up into a sharp syncline whose axis runs north-east, the remnant now existing being less than a mile in length and 300 yards in breadth. The underlying beds are naturally existent over a somewhat wider area, and extend across the Mandamus towards the Hurunui, the direction of one of the reaches of this stream corresponding in alignment and direction to that of the axis of the syncline.
The limestone has evidently been squeezed up by folding movements and has occupied the structural basin in which it lies, but the form of the land surface on which the limestone was laid down was not basin-shaped. There are similar basins in the country to the north-west, with parallel orientation, which do not now contain limestone outliers, but their form is so characteristic that their origin is probably similar to that of the Dove. These parallel elements may explain the north-easterly direction of the Hurunui in this part of its course, for after it leaves Maori Gully it apparently follows the line of these basins, with breaks across from one to another.
It may be noted also that the hills in this part of the valley rarely exceed 3,000 ft., but immediately to the north-west mountains rise to between 5,000ft. and 6,000ft., the marked difference in height being perhaps due to the fact that the lower area was faulted down along a line of settlement parallel to those occurring a short distance away in the Hanmer area and still farther away in the Kaikouras.
The indications certainly point to this submontane area having been covered with a veneer of sediments during Tertiary times; that it was raised with some faulting, and certainly with folding, in late Tertiary or in Quaternary times, the folding producing anticlines and synclines of the beds of limestone with a general north-easterly trend; and that these limestones were removed from the basins with the exception of that of the Dove. The drainage which now occurs may be called, as Cotton has suggested (1917, p. 253), “anteconsequent,” in that it was perhaps consequent on the former land surface, but antecedent as far as the present surface is concerned. The determining factors of the original consequent drainage must in this case be highly speculative and almost impossible to determine.
(2.) Hurunui-Waiau Basin.
The salient features of the Hurunui-Waiau basin have been mentioned before by myself (1915, pp. 347–48). The formation of this mountainringed area is attributable primarily to faulting or folding movements, or a combination of both, for there is ample evidence that both are present. The Tertiaries on the north-west side of the basin lie on the basement beds of greywacke with a general dip to the south-east, but with occasional reversals where they abut against the older rocks. This is specially well seen near the road past Mount Mason into the Virginia country, where the limestones in close proximity to the greywackes experience a sharp fold
backwards as if the beds had been dragged down along a fault-line. Farther north towards the Hurunui Gorge, opposite their junction with the Mandamus, they appear quite normal, but in the Pahau again their structure is obscure, though that may be attributed to the disturbance in the immediate neighbourhood of a volcanic vent. Farther north-east towards Culverden their arrangement is again normal. The floor of the basin is almost completely masked by the gravels of the Hurunui and Pahau Rivers, the only indication of what is underneath being given in the vicinity of Hurunui Mound. Here the Tertiaries rise like an island in the sea of gravels, and they are evidently folded acutely. In the cliffs on the bank of the river near the railway-bridge the structure is anticlinal, but at the Mound itself, about half a mile to the north-east, the beds are also folded, though not on the same line. There is evidence, therefore, of a more complex structure under the plains—that is, they approach a synclinorium.
The southern margin of the plain from the road-bridge eastward is determined by a fault-scarp, along the foot of which the river flows. The settlement of the block of country under the plains appears to be more marked on the south-east side (cf. the Waikari and Greta Valleys, also the fault system of the Kaikouras: Cotton, 1914), and the river has therefore occupied it as the lowest level possible on the plains. The outlet, however, is marked by high-level terraces indicating a former higher level of the river. It is almost certain, therefore, that the deformational movements which caused the basin had not terminated when the river had established its course through the gap at the south-east corner of the plains. Of such recent movements the surrounding districts furnish ample evidence (cf. the fault-scarps near Hanmer, at Glen Wye, and also the recent gorge of the Middle Waipara). The course of the river from the junction of the Mandamus has followed the line of steepest descent to the fault-line, and is therefore approximately at right angles thereto. This explains the necessity of the sudden sharp turn when the line of the fault is reached. Although the high-level terraces at the outlet may be attributed to recent movements in the basin itself, they may be correlated with the uplift which all this region has recently experienced, and therefore are the result of the river accommodating itself to a new and lower base-level. The river-course across the plains is marked by terraces of no great height. It here follows a direction consequent on a surface of its own making, for which the term auto-consequent could be used. Thus the courses of the Rakaia, Rangitata, and other large rivers of Canterbury across the plains are auto-consequent.
(3.) The Waikari-Kaiwara Basin.
After leaving the Culverden Plain the river flows through a gorge cut in greywacke for about six miles till it enters on the Waikari-Kaiwara basin. This extends down the river to the immediate vicinity of the Ethelton railway-station, when the river passes through another gorge cut in greywacke. The basin is therefore completely enclosed by pre-Tertiary rocks. Although the area has a basin-shaped form, its origin is somewhat different from the Waiau-Hurunui intermontane area, and owes its formation entirely to the faulting-down of a strip of Tertiary beds and the subsequent enlargement of the tributary valleys through the rapid erosion of relatively weak beds. These consist chiefly of sands with harder concretionary bands, sandy clays, and marls, with occasional irregular layers of shells, mostly in a fragmentary condition: they are, in fact, the equivalents of the Motunau or Greta beds. Mount Brown beds are existent
as well, but I have found no appearance of limestone, which is so well developed in the Waikari district to the south-west. Limestone does occur in the upper part of the Scargill Valley, in the form of faulted strips, but I have not traced it farther towards the Hurunui. The beds have a general north-east strike, and a dip to the south-east of from 150 to 200 Where the beds cut across the Hurunui, which they do at an angle of about 450, they are disturbed from their proper dip and are pulled up along the line of a fault on the downthrow side till they are nearly vertical; but this disturbance does not extend far from the fault-line, and may be attributed entirely to the movements caused by it. The result is that the beds form a strip running along the north-west side of the Waikari-Kaiwara depression, with slope accordant to that of the underlying surface, and if they were removed a characteristic “stripped surface,” as described by Thomson,* would be disclosed. I do not know what special name has been applied to valleys of this form, except that I think the term “basin range valley” has been applied to somewhat similar valleys in the western United States; but the Waikari Valley is of a somewhat different type, and also the name just cited is an unfortunate conjunction of terms. The name “rift valley” does not apply, because such are determined by faulting running along two sides, whereas these under consideration are attributable to tilting which has accompanied faulting along one line only. I suggest the name tilted strip as being a suitable name in case none has been already applied.
The most remarkable feature of the course of the Hurunui is its continuance across this depression without any apparent effect on its course. Although the earth-movements accompanying the faulting must have been of fairly recent date, the river has maintained its original course. It is interesting to compare this case with that of the Clarence Valley, farther north, where a similar valley caused by faulting on a much larger scale has dominated the course of the river. In the case of the Hurunui the movement must have been slow, and some cause must have been present which enabled the river to reach a lower base-level almost as fast as the downward movement occurred in the beds in this portion of its course. This cause will be evident from a consideration of the features of the next compartment into which the river-valley has been divided.
(4.) The Greta-Cheviot Basin.
The greywacke gorge of the river continues for about three miles below the Ethelton Station, when the valley opens out and the river has a wide shingly bed with flanking terraces cut in the marls of the Motunau or Greta series. Soon after its emergence from the gorge it receives the small Greta Creek, which occupies a valley similar in form and origin to that of the Waikari and Kaiwara. The beds let down by the fault which determines this valley develop northward into those of the true Cheviot basin, which is some five miles across, and extends past the Cheviot township across the Waiau as far as the Conway River. The structure of this basin is dominantly synclinal. The beds exposed on the floor of the basin are clays, sandy clays, sands, & c., of Mio-Pliocene age, passing down conformably into calcareous greensands (— Weka Pass stone) and hard limestone (— Amuri limestone). The limestone is exposed in places along the western edge of the greywacke ridge which separates the basin from the sea, and through which the Jed has cut its gorge. On the seaward side of this grey-
[Footnote] * J. A. Thomson, Coal Prospects of the Waimate District, South Canterbury, N.Z. Geol. Survey, 8th Ann. Rep., p. 160, 1914.
wacke barrier the limestone also occurs, with reversed dip, and under the limestones are exposed sands and greensands with saurian bones, and thin beds of impure coal. On following the beds across the strike a synclinal arrangement is found, and the limestone forming the south-eastern limb appears as a reef at Port Robinson, striking out to sea just as the limestone reefs do at Amuri Bluff. This syncline is well seen in the cliffs at Gore Bay, and it no doubt extends south-west as far as the Hurunui, and appears where the rocks dip up-stream just above the lowest bridge across the river. The upturned beds of the south-easterly wing of this syncline rest at Port Robinson on greywackes, and at the Hurunui Bridge on the same rocks. In the last-mentioned locality there is evidence that the Tertiaries are bent over this core of greywacke in mild anticlinal arrangement. The river has cut a gorge through the greywacke, which has been used as a solid basis for the abutments of the bridge.
Up-stream from this, the traces of the anticlines and synclines which occur between Port Robinson and Cheviot can be seen occasionally where the Motunau beds are exposed in the river-terraces, but no limestone is visible; the general arrangement is, however, synclinal.
An important point as regards the history of the river is the comparatively recent elevation of the coast-line. This has amounted to as much as 600 ft., judging from the shore-platforms extending to that height at Port Robinson, between the mouths of the Hurunui and the Blyth Rivers (three miles to the south), and just south of the Blyth River on the summit of the Napenape Cliffs. This elevation has been noted previously by Haast, Hutton, and McKay. In the last-named locality there are sea planed limestone surfaces 600ft. above sea-level covered in places with marine gravels. In the country just south of the Hurunui this plain of marine denudation extends back from the present coast-line for some five miles to the base of the greywacke hills, and exhibits a peculiarity in that the wave-cut surface is higher near the coast than farther inland. This suggests that a slight warping has taken place since the plain was cut; but the peculiarity may perhaps be explained by the more ready erosion of the softer beds farther inland than the harder limestone exposed near the coast where it forms the floor of the high platforms. The first explanation is, however, the more reasonable, and if it is correct the axis of warping would approximate to that of the line of the greywacke bar near the river mouth. Apart from the effect on the river in this vicinity, probably apparent in the gorge of the river incised in a somewhat wide flood-plain, an elevation of the land totalling some 600 ft. would exert considerable influence on a river which had reached approximate base-level, as it is reasonably certain that the Hurunui had, before the coastal elevation took place. The power of vertical corrasion would be greatly increased over a considerable part of the course of the river. At the present time a considerable portion of the Culverden Plains are under 600 ft. above the sea, and unless some compensations in level have taken place inside the coastal belt the level of the riverbed should have been greatly affected as far as the junction with the Mandamus at least, where the solid bars of rock would delay adjustment for a long period after it had taken place in the relatively weaker beds farther down-stream. There is, however, evidence of a lowering of the inland portion of the river-basin relative to the coastal portion as a result of the faulting which took place on the Greta line, on the Waikari-Kaiwara line, and again in the deformational movements of the Hurunui-Waiau basin. The effect of this would be to make this portion of the stream an aggrading one if the lowering were in excess of the coastal elevation. This has certainly been the case, for the aggregate throw of the faults must total
considerably over this. The effect of this has no doubt been to make the river an aggrading stream in that part of its course which lies in the Culverden Plain, and to neutralize the effect of the elevation perhaps as far down as Ethelton, but to rejuvenate the part between the Greta and the sea. Even this part is near a temporary base-level, judging by the great amount of shingle in its bed and the very low terraces of some parts of its course. This rejuvenation enabled the river to maintain its course in its lower portion across the grain of the country, to cut deep gorges through greywacke rocks, and to do this in spite of movements which would tend to turn it from its original course. As the present valley of the river is situated, there are several easier routes than that which it actually follows, such as that past Hawarden down the Waikari Valley, or past Hawarden into the valley of the Waipara and thus into the sea near Amberley. But it appears that under certain conditions, when the course of a river is once definitely established it will maintain that position in spite of influences which should divert it from its original path.
Development Of The Course Of The Hurunui River.
The géological features of this region which have primarily determined the course of the river are briefly stated as follows: On a greywacke surface, incompletely base-levelled, a series of beds was deposited chiefly in middle and late Tertiary times. These consist of sands sometimes with coal, greensands, limestones, marls, sandy shell-beds, and sandy marls passing up into conglomerates, the higher members being of Pliocene age. The general character of the strata indicates deposition on a sinking sea bottom in the early part of the period, followed by deposition on a rising bottom at the end, the whole sequence being laid down without a physical break. It is probable that there was some differential elevation towards the close of the time, so that some of the earlier beds were eroded in places while continuous deposition was going on elsewhere. The sea in which deposition took place gradually extended over a wider area with the sinking of the land, since the higher members overlap the lower and cover a more extensive area. Thus it is that limestone is very thin or entirely absent in certain localities—for example, the Greta and Waikari Valleys—the land occupied by those localities being the last to be invaded by the sea during submergence, and having an entirely different form from that which it now has. No doubt a slightly elevated area occupied the site of those valleys in early Tertiary times.
The covering beds extended far to the eastward, but have been cut back by marine erosion, which is at present making marked inroads on the sea-cliffs composed of loose sands and marls; while to the westward the Tertiaries extended beyond the Mandamus River, probably to the vicinity of Maori Gully, but fragments of the greywackes rose like islands in the cover of more recent beds, though not in the position of the high lands existing at present.
On this surface of covering beds as it emerged from the sea a consequent drainage was established, consisting of subparallel streams running seaward in an easterly direction. Although it cannot be stated with certainty, it is probable that the first elevation of the land took place with comparatively little deformation, and the river-courses were well established before the dislocations became pronounced. After the rivers had been completely established, folding and faulting took place on lines cutting the direction of the main streams at an angle of approximately 450, and these lines have determined the courses of the principal tributaries, most of which enter the main valley along fold and fault lines. The recency of the movements is
emphasized by the marked dependence of the landscape forms on the features resulting immediately from these movements. In some cases time has not been sufficient for the weak covering beds to be removed from the higher elevations, though in general these are more perfectly preserved in the folded and faulted intermonts. The movements producing these must have been slow, although they have been comparatively recent, since the main stream has preserved its original direction with but slight modifications, in spite of the opportunities presented for departing from it as a result of these movements, while farther north in the Kaikoura region the movements were on such a scale that the stream-systems are almost entirely dependent on them for their direction. The Hurunui region thus illustrates the condition that a powerful stream may at times maintain its original direction in spite of strong forces tending to deflect it. Dr. Cotton has drawn my attention to a paragraph in a paper by Professor W. M. Davis, entitled “An Excursion in Bosnia, Hercegovina, and Dalmatia,” * which seems appropriate in this connection. It reads as follows:” It is evident that this hypothesis [warping] accounts simply enough for the occurrence of irregularly alternating basins and uplands; and that the basins thus produced might be connected by gorges eroded through the uplands by the master rivers; the gorges marking either the paths of antecedent streams that had maintained their course in spite of the warping, or paths selected by the drainage consequent on some early stage of warping and antecedent to the rest.” This idea of anteconsequent streams has been elaborated by Cotton (1917, p. 253), and it seems entirely applicable to the case of the Hurunui, except that faulting has ensued as a result of the strains set up in the warping movements.
Glacial Features of the Hurunui Valley.
(See map, fig. 2.)
Although there are no present-day glaciers in the valley of the Hurunui, the mountains not being sufficiently high in that part of the alpine region to intercept sufficient snow to feed them, the upper part of its basin was subjected to the severe glaciation which affected a large part of the South Island of New Zealand in Pleistocene times and perhaps later. Haast has indicated (1879, plate 11) that the Hurunui Glacier at its greatest extension came down below the junction of the Waitohi River with the main stream—that is, well on to the Culverden Plains; but on what evidence he bases this great extension is not clear, and in my opinion there is no reason to demand it. The absence of morainic and other glacial deposits, as well as the form of the river-valleys in the middle course of the Hurunui, render his supposition very improbable. Especially is the latter evidence strong in the case of Maori Gully, the striking gorge which the river has cut in the edge of the high-mountain country before it runs through the foothills of the Alps. This is so deep and narrow that it is almost impossible that ice could have come through it and left it in its present condition. It seems to me extremely probable that the ice did not extend below the junction of the two main branches of the river, if, indeed, it came so far, since there is no proof of its former presence even at this point except the somewhat indefinite evidence based on the form of the river-valley, which may be attributable to ice action or may be the result of ordinary stream erosion. In the absence of other proof this solitary line of evidence must be viewed with considerable reserve.
[Footnote] * Bull. Geog. Soc. Philadelphia, vol. 3, No. 2, pp. 21–50, 1901.
Above the junction of the two branches the evidence is undoubted, especially from the vicinity of the Lakes Station and Lake Sumner towards the head of the river. Old moraines, smoothed and rounded surfaces, and the form of the cross-section of the valleys furnish indubitable evidence of the former presence of ice. In the upper part of the course of the North Branch the even alignment of the valley-walls, their steep lower slopes and gentler upper ones, the truncated and semitruncated spurs, and the roches moutonnées on the valley-floor are as characteristic of the results of glaciation as anything in the valleys of the Southern Alps farther south. On the north side of the river the mountain-tops have a very flat plateau-like form, and this feature continues as far as the valley of the Waiau, if not farther, so that the streams run in deep trenches incised in the tableland. To the south of the river the mountain-tops are more like those characteristic of middle and southern Canterbury, with rugged summits and wide expanses of moving debris dislodged from solid rocks by the action of frost. To the east of the plateau region the mountains take on this form even to the north of the river.
Specially interesting features of the river-basin have resulted from the action of these ancient glaciers on a mature valley-system which had become established in pre-glacial times. These features are so strongly reminiscent of those of the valleys farther south, especially of the Waimakariri, that they must be attributable to a common cause. The only difference in the two cases is that the features of the Hurunui are not so strongly marked, which is no doubt due to the more moderate intensity of the glaciation in the northern river. Thus there are distinct traces of the original directions of the streams, which are wanting farther south, but which may give some clue to the origin of the characteristic features of the valleys.
Parallel Valley System of Upper Part of Basin.
The most striking landscape form of the upper basin of the Hurunui is the series of subparallel valleys flanking the North Branch on its southern side. These are quite analogous to those of the Waimakariri, also on its southern side, and to those in the vicinity of Lake Coleridge in the valley
of the Rakaia. The arrangement is as follows, taking the valleys in their turn, commencing from the north:—
The main valley of the North Branch leading into Lake Sumner. This has a general east-and-west trend, with wall-like sides in good alignment, but broken by tributary valleys, especially on its northern side. Lake Sumner occupies a continuation of this valley, but about half-way along the lake the trend assumes a north-west and south-east direction more in agreement with that of the others. This corresponds in direction and salient features with the main valley of the Waimakariri, but the landscape characters are on a smaller scale.
The set of subparallel valleys leading from the vicinity of the head of Lake Sumner, near Lake Katrine, and running south-east. At the head of the system there is only one main valley, but it breaks up within a short distance into distributaries consisting of—(i) A valley immediately to the south of Lake Sumner and divided therefrom by a ridge of which the peaks known as The Brothers (4,563 ft. and 4,507 ft.) are the highest points; (ii) a valley in which lies Lake Sheppard, divided from (i) by a discontinuous ridge ending in The Sisters (3,281 ft.); (iii) a valley in which lies Lake Taylor, divided from the former by Conical Hill (2,783 ft.) and from the valley of the South Branch by the Oronoko Range. These valleys are quite analogous to those in the Waimakariri basin, which may be called (i) the Lake Black-water Valley, (ii) the Lake Sarah – Sloven's Creek Valley, and (iii) the Lake Grassmere – Lake Pearson – Winding Creek Valley. They also resemble the still more remarkable and perfect system to the east of the Rakaia basin in the vicinity of Lake Coleridge.
The ridges which divide these valleys are analogous in their physical characters. They are very steep-sided, with somewhat narrowed crosssection, so that when viewed end-on they appear conical; hence the frequent occurrence of such names as “Sugarloaf” and “Conical Hill” in North Canterbury. But when viewed from the side they form long ridges cut into well-defined saddles (cf. Mitre Peak). When this saddle is low and little elevated above the floor of the valley in the vicinity the ridges become isolated hills, which are often in pairs or linear series, and give rise to such names as “The Brothers” or “The Sisters.”
The valleys indicated above junction with the main valley of the North Branch after it leaves Lake Sumner and takes the first decided bend of the river to the south-east. It soon afterwards receives the South Branch. In its upper part this valley has an east-west direction, but it soon takes on the characteristic north-west and south-east orientation, and finally turns and joins the other branch nearly at right angles. The dividing wall between it and the Lake Taylor Valley to the north, called the Oronoko Range, has been partially broken down in several places. The most important of these lies just opposite the head of Lake Sumner, where a low pass leads from the northern to the southern valley of the Hurunui. On the southern side of the pass is Lake Mason, tucked away in a tributary valley of the South Branch. The country in this neighbourhood has been highly glaciated, roches moutonnées and smoothed surfaces forming characteristic features of the landscape. Opposite the Lakes Station there is another saddle—some-what high, it is true—and the ridge has also been lowered in a line with the Lake Taylor Valley leading directly to the North Branch south of Dog Hill, indicating an overflow in that direction.
The partial dismemberment of this ridge affords a clue to the conditions which obtained before dissection and isolation overtook the ridges to the north-east. By noting its features it is possible to restore with reasonable
certainty the general direction of the streams that flowed through this tract of country anterior to the glaciation. In addition to the two main branches of the river a large stream rising near the head of Lake Sumner followed the course of the Lake Taylor Valley, parallel with the South Branch; this entered the North Branch about half-way between Lake Sumner and the junction of the two main branches. A small tributary entered this valley on the north side, rising near the head of Lake Taylor and following the course of Lake Sheppard. Another small stream rose near Lake Katrine and joined the North Branch below the outlet of Lake Sumner. In pre-glacial times the ridges dividing these valleys would be more or less entire, though they might have saddles at their heads. It is impossible to reconstruct such features exactly, but the description just given affords a fairly accurate view of the stream conditions which obtained in this tract of country before it was modified by glaciation.
Whatever was the prime cause which promoted glacier extension, it is reasonable to assume that it was gradual in its incidence. Snow would slowly accumulate, glaciers would be formed at higher altitudes and slowly extend down the valleys. Thus the heads of the small valleys would probably be filled with corrie glaciers, while the glaciers of the first order would be extending down the main valleys. These would help to lower the divides in the way suggested by Matthes.* As the ice-flow increased in volume the main streams would be filled, and in time overflows would take place over the lowest part of the divides, which would be lowered at the same time by active ice abrasion. It is significant that the greatest amount of lowering has taken place near the head of Lake Sumner. This would be due to the marked overflow of ice from the main Hurunui Valley, no doubt due to the narrowing of the cross-section of the valley at Lake Sumner, which caused the ice to overcrowd into the headwaters of the neighbouring streams, as it has done in several of the valleys of the Canterbury rivers. The full force of this would be felt at the head of the Lake Taylor Valley, and thus its divide has been completely reduced. The headwaters of the intermediate tributary valleys were also invaded and the saddles at their heads reduced. Thus a clear passage for the ice was opened down these valleys past the site of the Lakes Station in the direction of the south-easterly reach of the North Branch below Lake Sumner, while the main stream of ice followed down the valley now occupied by this lake.
In addition to the overflow toward Lake Taylor a powerful stream passed over into the tributary which runs into the South Branch from the north. The saddle at the head of this stream was thus reduced, but not so much as its neighbour, which was more in the line where the ice-stream would impinge on the valley-wall. If, however, glacier action had continued this saddle would have been reduced and the mountain ridge to the north of the South Branch would have been completely isolated. It is possible that ice also overflowed into the valley of this stream near the Lakes Station, and as at the height of the glaciation the country in its vicinity would have the form of an intermontane basin, and would be an efficient gathering-ground, overflows from it took place along several lines from the front of the ice-sheet in the direction of the valley of the North Branch. These would produce the breaks in the valley-wall between the South Branch and the country to the north which occur immediately up-stream from the junction of the two branches.
[Footnote] * F. E. Matthes, Glaciation of the Big Horn Mountains, U.S. Geol. Surv. 21st Ann. Rep., 1899–1900.
The dismemberment of the ridges would no doubt be promoted by the sapping-back of the valley-walls and their complete reduction in places where the ice-stream impinged more powerfully. If we compare the results of the glaciation in other valleys we see that in their cases the dismemberment has been more complete, the dissection of the dividing ridges carried to a further stage, and the straightening of their sides more thoroughly carried out, because they experienced more complete glaciation. If we were furnished in these cases with more clear indication of the intermediate condition of the direction of drainage it would be possible to reconstruct the original stream-system.
There is one point, however, which has not been considered fully—viz., the agreement in the direction of the tributary valleys with those of the Waimakariri and Rakaia. Is this agreement in orientation the result of accident, or is it based on some structural condition which has influenced the country in the basin of the Hurunui as well as the country farther south?
I have shown (1916, pp. 142–43) that there exists in the mountain region of Canterbury a well-marked series of fractures or lines of folding which lie in a north-west and south-east direction. Cotton (1917, p. 273) draws special attention to the importance of the north-west system of earth-fractures in Otago as compared with the other parts of this Island, but they certainly occur in Canterbury in conjunction with the Kaikoura system, and it is possible that in the upper Hurunui, as in the Waimakariri, they are co-existent. Since writing the article referred to above I have noted additional lines with the north-west orientation, especially in the upper valley of the Waipara and in South Canterbury in the country between Fairlie and Cave. In both these cases there are undoubted, well-marked lines of fault. It is possible, therefore, that the general direction of the valleys at the head of the Hurunui were determined initially by lines of structural weakness. It is remarkable also that some of the valleys on the northern side of the river have a characteristic east-north-east orientation, and these are parallel to other valleys farther north, such as the Hope, in the basm of the Waiau. On the line of one of these upper valleys is the hot spring which forms a notewerthy physical feature of the Upper Hurunui, and I am informed that other springs occur in the valley which stretches north-east from this locality. This certainly points to the presence of an earth-fracture with east-north-east orientation.
Another feature of this district should be noted—viz., the north-and-south trend of the upper valleys of the Waiau and Clarence, a direction which is parallel to the twin valleys of the Mandamus, and to that of the Glenrae, lying to the west of these. The arrangement may be only a coincidence, but it is certainly a striking one.
Cotton, C. A., 1914. The Physiography of the Middle Clarence Valley, New Zealand, Geog. Journ., vol. 42, pp. 225–46.
—— 1917. Block Mountains in New Zealand, Am. Journ. Sct., vol. 44, pp. 249–93.
Haast, J. von, 1971. Rep. Geol. Explor. dur. 1870–71, pp. 30, 48, sections.
—— 1879, Geology of Canterbury and Westland, pp. 69–78, 216–17.
Hutton, F. W., 1877. Rep. Geol. Explor. dur. 1873–74, pp. 34–35, 55–56.
—— 1889. The Eruptive Rocks of New Zealand, Trans. Roy. Soc. N.S.W., vol. 23, pp. 125, 156.
Speight, R., 1915. The Intermontane Basins of Canterbury, Trans. N.Z. Inst., vol. 47, pp. 347–48.
—— 1916. The Orientation of the River-valleys of Canterbury, Trans. N.Z. Inst., vol. 48, pp. 137–44.