Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 71, 1942
This text is also available in PDF
(3 MB) Opens in new window
– 169 –

The Rangitata Glacier—The Question of Its Maximum Extension.

[Read before the Royal Society of N.Z., Canterbury Branch, October 2, 1940; received by the Editor, January 7, 1941; issued scparately, December, 1941.]

Table of Contents.




Rangitata Valley—Its Chief Features.


Sub-Glacial Deposits.


River Terraces.


North Bank.


South Bank.


Interpretation of Observations.



A. Introduction.

The distance on to the western margin of the Canterbury Plains that the glaciers extended from the valleys of the Southern Alps, New Zealand, at the height of the Pleistocene Glaciation is still a matter for discussion. Statements have been made by Haast (1879), Hutton (1884), Park (1910), and the present author (1937 and 1938) with reference to the Waimakariri, Rakaia, and Ashburton valleys, but no reference has been made to the Rangitata except brief remarks by Haast (op. cit., p. 212) and by Park (op. cit., p. 238). The locality has recently attracted considerable attention, since the headworks of the diversion race of the great irrigation scheme for the Canterbury Plains are located at the mouth of its gorge, and much heavy excavation has been necessary to negotiate successfully the terraces which attain a maximum development in its vicinity. It is hoped, therefore, that a statement concerning its features may be of some interest, and specially so in a case where the evidence as to the precise extent of glacierisation is indefinite as is frequently the case on the margin of an area affected, and, above all, when the area has been exposed to active erosion by a powerful stream. The fact that the evidence is inconclusive does not appear to me to be an adequate reason why it should not be considered; rather the reverse, since such a consideration may help to throw light on similar phenomena, or the lack of them, elsewhere.

B. Rangitata Valley—Its Chief Features.

Note.—The localities, etc., mentioned in this section will be found on any large scale map of Canterbury.

Before dealing with the special matter under consideration it will be best to give a summary of the salient features of the upper valley of the Rangitata. This river is formed by the coalescence of the Havelock, Clyde, and Lawrence, three streams of sub-equal size, the first two rising in the main divide of the Southern Alps, here about 7,500 feet in height, and the last rising in the high mass of Mount Arrowsmith (9,171 feet) and the country west of it. All three

– 170 –

valleys show evidence of a former glaciation and even now they are all headed by small valley glaciers in which the rivers take their rise. In their lower reaches the Havelock and Clyde flow as braided streams over gravel river-beds, up to two miles in width, but the Lawrence is more confined, its channel being constricted by well-developed moraines which lie at intervals across the valley.

Below the confluence of these streams the main river continues for about twenty miles through a great trough or basin, floored with gravels, and having a width of over four miles in its upper part (Plate 30, Fig. 1). In this portion it receives one considerable tributary from the north, viz., the Potts River, and one from the south, Forest Creek, the latter joining the main river about four miles downstream from the homestead of the Mesopotamia Station, the home of Samuel Butler in the early ‘60's of last century. Near Forest Creek the width of the valley is increased to about six miles, but it narrows sharply immediately below this tributary, high mountains crowding it on either side, the Harper Range on the north and the Ben Macleod Range on the south, both rising to over 6,000 feet. The river still occupies a large part of the floor of the valley, its actual bed being over half a mile wide in places, and it continues thus for ten miles till it passes abruptly into a narrow gorge, five miles in length, from which it emerges to start on its course across the plains.

The basin portion of the valley almost certainly owes its initial formation to tectonic movements analogous to those attending the formation of the other intermonts of Canterbury, since remains of a Tertiary cover occur as coal measures in the valley of the Potts above its rocky gorge, similar beds occur with a cover of limestone near a small stream, called Coal Creek, which joints the Rangitata just above its gorge, and Tertiary sands are to be seen in Boundary Creek, which is parallel to Coal Creek and joins the main stream just above the mouth of the gorge. The dislocation evident in this locality is probably due to pre-glacial faulting or warping oriented on N.N.W.-S.S.E. lines, which is probably responsible for the initial formation of the valley of Puddingstone Creek, a tributary entering the Rangitata just opposite Coal Creek. Also the down country between the Mesopotamia homestead and Forest Creek cannot be credited to glacial modification of an ordinary pre-glacial stream valley, and it probably owes its peculiar features to the foundering of a triangular block of country bounded by faults running (1) approximately N.-S. along the eastern flank of the Sinclair Range, and (2) E.N.E.-W.S.W. parallel to Forest Creek and the northern flank of the Ben Macleod Range. This down-dropped area has been modified by glacial abrasion and deposition.

The surface features of the basin as a whole were substantially affected by the action of glaciers, and I know of no other valley in New Zealand which shows so typically those landscape forms characteristic of a glaciated country—transverse and lateral moraines, perched blocks, roches moutonnées, partially and completely truncated spurs, ice-scoured and glacier-terraced valley sides, boulder clays, etc. Typical moraines occur near Forest Creek, and at the lower end of the basin a well-developed terminal moraine in arcuate

– 171 –

form crosses Coal Creek and the main stream just below the confluence of Puddingstone Creek, while on the upstream side of the junction, in the angle between the creek and the river, a spur of the Harper Range has been partially truncated by ice action. Thus the part of the basin lying furthest downstream shows the undoubted presence of glacier ice, and there is a possibility that the smoothed landscape in the vicinity of Coal and Boundary Creeks, and near the saddle between the latter and Rawle's Creek, over which the road up the valley passes, may be due to glacial abrasion, though no moraine or other glacial deposit confirms this opinion, and the smoothness may be attributed to nivation or even to an Early or Middle Tertiary planation of the surface on which the coal measures in Coal and Boundary Creeks rest.

A large distributary from the main glacier left the valley near the Potts River, passing over the lower course of this river at right angles, and continuing along the northern flank of the Harper Range in the direction of the South Ashburton near Hakatere and Trinity Hill, to leave proofs of its former presence in the roches moutonnées and moraines near the latter, extensive areas covered with recessional moraines between this hill and the Potts, as well as stranded laterals on the lower and intermediate slopes of the hills on either side. It is probable that over one-third of the Rangitata Glacier took this course when at its maximum development.

The cross-section of the gorge of the main river is similar to those of the other Canterbury rivers, but on a smaller scale. It has the form of a U-shaped trough, in the floor of which a deep channel with steep, rocky walls, has been incised after the retreat of the glaciers from the neighbourhood. A typical moraine lies on the northern side of the river just above the outlet of the gorge, and this furnishes definite proof that the glacier reached down nearly to the upper margin of the plains. There may, however, be some difference of opinion as to how much further it extended, and the presentation of the facts related to the problem, as I see them, is the chief reason for this contribution, inconclusive though it may be.

C. Sub-Glacial Deposits.

An important point is the determination, if possible, of the form of the land surface before the onset of the ice-flood. The structure of the substratum of the plains is disclosed to some extent where rivers, such as the Rakaia and Waimakariri, have cut down deeply into the post-Tertiary beds just below the outlet of the gorges through which these rivers issue on to the plains. The lowest gravel beds exposed in these localities are brownish in colour owing to the presence of some amount of iron oxide, and over these lies a less oxidised facies, with a definite break between them, indicating some hiatus or change in conditions; while over this again lies morainic material, indicating that the main formation of the plains antedates at least one phase of the glaciation. No morainic material was observed to be interbedded with these gravels, but in the Rakaia a deposit of varved silt lies right down on the pre-Tertiary basement. It is covered by gravels also, but they perhaps belong to a late terrace deposit, although the weight of evidence suggests that the silt

– 172 –

underlies the oxidised gravels, and therefore indicates the proximity of glacier ice before the gravels were laid down, and this would necessitate the admission that the main part of the gravels forming the plains was deposited during a period of ice advance, or at least a glacial episode.

The evidence in the case of the Rangitata is not so clear, since there are few natural exposures owing to the completeness of the grass covering, but the excavations for the diversion race disclose to some extent the sub-surface structure. An excellent artificial section occurs on the north bank just above Mr. McIlwraith's farmhouse, where the race follows for nearly a mile round the base of a high terrace. The material occurring here is not morainic, but is chiefly of well-rounded boulders in a matrix of finer grain including sand, the whole exhibiting a rude stratification and occasional crossbedding. Similar sections are exposed on the south bank of the river above and below the inflow of Rawle's Creek. The section on the north bank shows no interstratified moraine, but large blocks lie on the ground above it and in parts of the race further downstream large boulders occur in fluvial material. The largest measured (fide T. G. Beck) was nine feet in diameter, but this was only part of a surface deposit and not buried deep down in gravel. The edges of some of these boulders are rounded, while others are angular and suggest transportation by ice or by some exceptionally powerful stream from a spot in close proximity to where they now lie. The section above Rawle's Creek (Plate 30, Fig. 2) shows a similar structure; it contains no large boulders, but is capped by large angular blocks. It must be noted in this connection that the surface of many terraces, where sections can be seen, is composed of a veneer, either including or composed of angular blocks, both facies capping deposits of finer grade. The experience of engineers carrying out excavations for the race confirms this opinion. The capping of large blocks is especially evident where the top of the terrace—the tread of the terrace stairway—is of limited width; where the tread is wide the blocks show more rarely. No deposit of varved silt is visible on the greywacke rock which forms the threshold of the gorge, so the sequence is not quite analogous to that of the Rakaia.

D. River Terraces (see Map, Plate 29).

Some account of the suite of terraces which occurs below the gorge should be given here, for it has an important bearing on certain aspects of the problem. Since it is most complete on the northern side of the river, and the structure of the beds involved is well exposed in certain places on the sides of the diversion race, this bank will be considered first.

(a) North Bank (Plate 31, Figs. 1 and 2).

1. The uppermost terrace, marked 1 on the map.

This abuts at its north-western end against the greywaekes of Peter Range at a height of about 1470 feet above the sea, and 280 feet above the river at the intake (1191 feet). Its face continues for about seven miles in a south-easterly direction and gradually diminishes in height, for while it is 80 feet high at its upper end, it is 65 feet at Morrow's Road, 30 feet at Moorhouse's Road, and it continues

– 173 –

at a definite bank for another mile and a half, where it ends in a break to be mentioned later. Its upper surface extends eastward to a little beyond the line of the South Hinds River, while its front stretches some five miles in a N.E.-S.W. direction from Trig. U (1139 feet), near the southern end of the Surrey Hills, to a little below Trig. K (1080 feet) on the high bank of the Rangitata River, the line lying in general between the 1050 and 1100 foot contours. (see Plate 29).

The surface is formed of fairly coarse material, much coarser than is usual near other Canterbury rivers under similar conditions, boulders up to three feet in diameter occurring occasionally at ground level. The coarse material is masked in general with material of finer grain, both wind and water borne, and specially so near the foot of the hills. Its surface is channelled with shallow abandoned streambeds, and no marked unevenness occurs till just below the road junction near Montalto, where there is an accumulation of larger angular and sub-angular blocks up to five feet in diameter. In the limited exposures visible on the side of the road, and on the sides of the excavation for the diversion race in close proximity, these are seen to rest on finer-grained, rounded, fluvial material. This accumulation forms a flattish-topped elevation which starts as a definite rise just below the point where the race crosses the Hinds Road, and it continues for nearly half a mile alongside this road past Trig. U and terminates in a fairly steep bank, 45 feet in height, just below the cross-road (Plate 31, Fig. 1). The elevation forms a ridge only a few chains wide at most, and is flanked on its eastern and western slopes with large blocks. Parallel with the western flank lies a former stream channel, which ends downstream in a steeper slope, entirely different from that of the river-channel above or that of the plain below, and it also is veneered in places with large stones. The line of this break is continuous with the face of the bank to the northeast, but in front of the river channel it forms an embayment, and the line advances again to the west in the direction of Coskerries Road, where the break in grade becomes more pronounced. Further west is becomes less defined, and it crosses Moorhouse's Road at an acute angle and continues to the neighbourhood of Perrin's Road, where it becomes more defined again, and takes the form of a blunt nose downstream from Trig. K, and distant a mile and a half or so below the junction of Moorhouse's and Perrin's Roads. This nose is clearly indicated by the contours on the map, and it is apparently formed of rounded gravel as no large blocks show on the surface. Considerable attention has been given to this feature, for it possibly marks the extreme limit of the glacier extension.

Numerous large angular and sub-angular boulders show on the face of this terrace (No. 1), the largest measured being five feet in diameter, and they continue to a little beyond the end of Morrow's [ unclear: ] Road and the vicinity of Trig. O (1268 feet), some four miles southeast of the intake of the race.

2. (Marked 2 on the map.)

In the vicinity of the intake there are only two main terraces. The highest has just been referred to, and the lower one with a height

– 174 –

of about 200 feet above river-level presents certain interesting features. Upstream it ends in the moraine mentioned earlier which lies just above the mouth of the gorge. One feature of this should be recorded, viz., the blocks are of greywacke with very occasional argillite, but with no granite-conglomerate. This last forms such an important element in the collection of blocks encountered elsewhere that its absence in this case is noteworthy. It necessitates serious consideration of the possibility that the presumed moraine may be really a rock-fall, but after a close scrutiny of all the circumstances I have come to the conclusion that it is really moraine, and if so, it implies that a late phase of glaciation did not result in the transport of material from the known Jurassic granite-bearing conglomerate beds in Puddingstone valley, so it is extremely likely that when the moraine was formed no ice came from this locality. But it is quite possible that the conglomerate may occur unlocated on the slopes of the Harper Range facing the Rangitata and then this explanation ceases to be valid. The valley walls upstream from this moraine give clear evidence of modification by ice.

In the gorge this terrace is flanked on the river frontage by a lower shelf, from which the slope falls steeply to water-level, the stream-bed being very narrow for a river the size of the Rangitata. Near the intake this lower shelf is absent, and the steep slope leading down directly to water level is faced sporadically with large angular masses. Between the two main terraces in this locality lie two intermediate shelves, one lying at the foot of the upper main terrace, and the other at a slightly lower level terminated upstream in an accumulation over a quarter of a mile in length of very large masses (Plate 31, Fig. 2), one of which measured 15 feet by 10 feet by +5 feet, with many others of considerable size. This collection is oval in form, with its long axis parallel to the river, and it lies separated from its counterpart at the base of the top terrace by an old stream channel. When traced downstream the part of the accumulation at the base of the top terrace divides again into two parts at slightly different levels, but both are veneered on their top and flank with large blocks (Plate 32, Fig. 1), which become specially apparent where the shelves are broken into by an embayment formed by lateral erosion of the river when swinging northward at a lower level later in its history. Owing to this lateral erosion the terrace as a whole has been much narrowed, but it widens again downstream and continues past the end of Morrow's Road, where it is 180 feet above river level, and on for a mile and a half downstream from Moorhouse's Road, where it merges into the general surface of the plains with an accordant level, below the break referred to earlier in connection with the upper main terrace.

In all parts this second main terrace is veneered with large blocks on the riser face, and in places the tread is similarly covered, and specially so just below Moorhouse's Road, where in a ploughed field large angular and sub-angular blocks appear, one of granite-conglomerate measuring seven feet in length.

In some places, specially near the intake and also near the end of Moorhouse's Road this terrace falls directly to water level, but in other places it is flanked by a suite of smaller terraces, often mere

– 175 –

shelves, varying in number in different localities, and even when the number is not variable showing discordance in level within short distances. Some have been eliminated by the lateral erosion of the river at lower levels. About a mile from the intake there are five such shelves, while near Mr. McIlwraith's there are six, and a little further on again six with some discordancy. Their treads and risers are veneered with large stones, one near Mr. McIlwraith's measuring ten feet in length. Apart from this feature they have little relation to the question under consideration and nothing further need be said concerning them.

(b) South Bank.

The highest terrace (1a) on the south bank of the river starts from just below the junction with Boundary Creek, and follows along the base of the foot-hills of the Mount Peel Range in a south-easterly direction across Rawle's and Chapman's Creeks to just above Lynn Creek and the vicinity of the Mount Peel homestead. It is over a mile wide between Rawle's and Chapman's Creeks, but north and south of them it has experienced the destructive action of streams coming in from the south as well as the inswing of the river at various levels, so that near Boundary Creek it shows only as small detached fragments. This terrace is higher than the highest terrace north of the river, and in the vicinity of Rawles Creek its surface is some 425 feet above river level (aneroid reading), and its face is 225 feet high. There are small remnants of a shelf forty feet higher still, but these have been modified by wash from the neighbouring hills so that a determination of their true height is illusory. The structure of the main terrace is well displayed on the banks of the tributaries from the south which have incised it deeply and have steep banks, and, in my opinion, the material belongs to an older very late Tertiary or early Pleistocene gravel series, for the material is smaller than that of the undoubted river terraces, is less rounded, more consolidated, contains much clay in places, and is also at times well stratified. No glacial material appears to be associated with it.

Mention should be made of two isolated shelves below Lynn Creek whose upper surface is slightly higher than that of the main terrace upstream. They are composed of the same gravels, etc., as the main terrace, as can be seen on the high banks of the Lynn and of a sub-parallel creek to the south-east, and the apparent increased height, which amounts to thirty feet at most, may be attributable to the raising of the surface of the terrace by the fans of two tributaries when they entered the main river at a higher level, a modification described by Cotton (1940, p. 28) in connection with terraces in the Esk Valley of North Canterbury. This modification naturally increases the height of the terrace most in close proximity to the point where the tributary leaves the steep wall of the valley to debouch on to its floor, and when the terrace is cut back subsequently by the lateral erosion of the river it leaves a remnant out of accord with the general level of the terrace.

Below the terrace just referred to, and near its Boundary Creek end, there are short terrace remnants, and then some 225 feet below it, that is, 200 feet above river-level, lies the best-developed terrace (2) south of the river. This is broad and accordant with the second

– 176 –

main terrace north of the river. It extends downstream from Boundary Creek past Chapmans Creek to the vicinity of Mount Peel homestead, where it has been cut off by the inswinging of the river at a subsequent date when it destroyed the upper terrace as well and left the small shelves near Lynn Creek. The two parts now separated by the river form the most striking physiographic feature of the locality. Opposite the ravines where Rawle's and Chapman's Creeks and other tributaries issue from the base of the upper terrace the surface has been modified by former high-level fans of these tributaries according to Cotton's suggestion (loc. cit.). Measurements made with an abney show that the surface may be raised as much as fifty feet, an amount which diminishes progressively in the direction of the margin of the terrace. This gives some idea of the extent to which the level of a terrace may be raised consequent on lateral fan formation. Subsequently the terrace has been trenched by these tributaries as the main river has lowered its base-level, and their high banks show that the terrace has been cut to some extent, if not altogether, in the same material as composes the highest terrace.

It resembles its northern counterpart in the occurrence and distribution of large blocks. While they show only occasionally on the tread of the terrace considerable accumulations lie on the edge and down the riser slope facing the river, and they continue as far downstream as the terrace is preserved, a noteworthy occurrence being about two miles from the Mount Peel homestead. Included in this concentration is a large block of granite-conglomerate seven feet in diameter. This lies over two miles upstream from the site of a similar collection near the end of Moorhouse's Road on the opposite bank of the river. Wherever clear-cut sections can be seen the large blocks rest on rounded gravels and are not embedded in them (Plate 30, Fig. 2).

Lower level terraces are not developed as fully on the south side as on the north side of the river. They occur upstream from the Mount Peel homestead, above and below the infall of Rawle's Creek, while near the intake of the race is an interesting suite of smaller terraces, occasionally mere shelves, five in number in some places and seven in others (Plate 30, Fig. 2), the discrepancy being due to the removal of portions of terraces already formed by the inswinging of the river at lower levels. All these terraces near the intake are veneered with large blocks, while at the intake itself the bed of the river is floored with large blocks, and I am assured by the Resident Engineer that they did not change in position to the slightest extent during the record flood in February of this year (1940). If the blocks now lying in the somewhat confined bed of the present river have not been moved by a record flood, it seems unlikely that those at a higher level on the terraces, where the streams would have been less confined and not so deep, though perhaps larger in volume, could have been carried an appreciable distance to their present positions by water alone.

On the south bank of the river near the intake is an interesting occurrence not obviously connected with glaciation, viz., a discordance in level of about four feet which affects the lowest two terraces of the suite (Plate 32, Fig. 2). Had this affected only one terrace

Picture icon

Map of the country in the vicinity of the outlet of the Gorge of the Rangitata River. The part north of the river is based on the survey for the Irrigation Diversion Race, with slight alterations. That south of the river has been sketched in, but the position of the two main terraces is reasonably correct; these are numbered 1 for the top main terrace and 2 for the lower main terrace; other terraces are largely diagrammatic.

Picture icon

Upper part of the Rangilata [ unclear: ] intermont view looking north-west from Ross's Cutting near Potts River, the bed of which shows in the foreground.

Picture icon

Gravel bank on south side of the river between the intake and Rawle's Creek. On the extreme left on the top of the bank lies a collection of large blocks; these do not occur interstratified with the gravels.

Picture icon

Edge of the break in grade near Trig. U. just south of the Surrey Ilills, view looking north, sleep slope 45 feet high facing the plains.

Picture icon

Large blocks on the top of the second main terrace on the north side of the river, view looking downstream. This is not the largest block; others can be seen behind the figure.

Picture icon

Morainic blocks on edge of intermediate terrace. 15 feet above No 2, north side of the river, view looking downstream.

Picture icon

The outlet of the Rangitata Gorge looking north-west; Boundary Creek in the middle, the river coming in on its right; the site of the intake is in the middle at the bottom of the picture; the face of the lower main terrace (No. 2) on the north bank lying on the right, and the whole suite of terraces on the south bank on the lett. The trace of the fault-scarp can be seen leading up from the river-bank on the lett.

– 177 –

it might have been attributed to some accident in river erosion, but the fact that it affects two terraees, and on the same line, discounts this explanation, so it must be attributed to faulting. Some difficulties remain, however, for the dislocation affects higher terraces of the suite to a less and less extent, and the uppermost are not affected at all, nor does the break show on the north side of the river where it might be expected, unless a slight waviness in the surface of the second main terrace can be attributed to this cause. The greywacke basement is not exposed on the line of the break so the effect on the solid rock cannot be seen. The dislocation is clearly posterior to the formation of the lowest terrace, so it must date from comparatively recent, perhaps even from historical times, and it furnishes another indication that tectonic movements have continued in the Southern Alps down to the immediate past, and in all probability have not ceased yet. Its occurrence here extends the area where recent faulting has been recognised. Due weight was given to this proof of crustal instability in the locality when the mode of formation and the accumulation of angular masses in the gorge, i.e., the moraine, was under consideration.

E. Interpretation of Observations.

The foregoing is a statement of the facts as I see them relevant to the question of the extent of glacierisation of the locality, and the following is a summary of the conclusions to be drawn therefrom.

The first point to be considered is the distribution of the large angular and sub-angular blocks of the highest terrace north of the river. The concentration off the southern end of the Surrey Hills near Montalto, rising as it does above the general surface of this terrace, appears to be abnormal if regarded entirely as an alluvial deposit. Its form, notably the bank flanking it on all sides, and specially facing the plains, is against this contention (Plate 31, Fig. 1). It is reminiscent of the low mounds composed of rounded gravels, which margin morainic deposits in other places, such as between The Point and Glenroy in the case of the Rakaia, and its flat top indicates that this has been planed by stream action, but if the blocks which veneer its flanks be regarded as having been derived from a morainic deposit not completely dispersed and buried by streams, then the whole occurrence, and specially the scarp facing the plains, can quite well be accounted for.

Further south-west and more in a line with the glacial streams as they issued from the gorge, any such deposit would feel the full force of their destructive action, and one would expect it to be scattered, or destroyed, or buried in the gravel of the river fan, the cross-section of which rises to the very edge of the highest terrace facing the river, as can be seen from the contour lines on the map. But in spite of this exposure to destruction the scarp which commences near Trig. U persists as a definite feature. This persistence appears inexplicable if the form and especially the grade of the plains is attributed to river deposition, but it is what might be expected if the glacier front and its moraine were in close proximity and were exposed to the wash of water. The large blocks on the tread of the second main terrace near Moorhouse's Road are close to the

– 178 –

extension of the line of this scarp, and may have some relation to it. It should be noted that no block of greywacke or associated rock from the back country was found on the southern end of the Surrey Hills and such blocks should be easily detected if they lay on the surface of the volcanics forming these hills, so it is reasonable to conclude that the ice did not reach as far as their lower slopes.

It would be a simple solution of the problem to regard the veneer of boulders on the face of the top terrace as having been dropped initially by a glacier retreating from a more advanced position and included subsequently in the material forming the top layers of the terrace, the larger blocks sinking gradually in the loosely compacted gravel as happens at the present time in front of many glaciers in New Zealand, and also in the case of concrete blocks washed from the protective works on the banks of gravel-bearing rivers. The scouring of the finer material either on the downstream or the upstream side of the block is largely responsible for its settlement and ultimate disappearance. If this is correct, one must explain why large blocks lie thickly on the treads of the lower terraces, and also why a veneer of coarse material masks a finer-grained alluvial basement. I can only suggest that the underlying gravels, as they form part of the lower strata composing the plains, have had an opportunity to become more compacted, and therefore more capable of supporting larger blocks on a water-worn surface; such a compacted surface will be less liable to scour. The upper layers of the top terrace have not had this chance of becoming consolidated, and the larger blocks have sunk into them when washed by water, only to appear on the edge of the terrace when the river has cut down to a lower base-level and has removed the finer material forming a matrix in which the larger blocks are included. Of course, if the glacial river be considered competent to move such blocks for a considerable distance from the ice-front and to concentrate them into definite accumulations, then it must be conceded that the terrace affords no positive proof of the former presence or proximity of ice, but the break near Montalto facing the plains has still to be explained, and its somewhat irregular line, with embayments facing upstream, cannot be accounted for by subsequent erosion by a river flowing along its front.

When considering this aspect of the matter, it must be remembered that at times glaciers do not form terminal moraines, and the conditions that determine their absence are not thoroughly understood. In this case a paucity of moraine may be due to a glacier issuing from a somewhat confined valley and deploying on to a plain where it would tend to spread, thin out, and develop an indefinite front, and the material it carried through the gorge would be scattered as a thin veneer over a wide area. The paucity of evidence in this case may be accounted for in some such way.

Another problem in this connection is the origin and subsequent history of the blocks that veneer the treads and risers of the lower terraces (Plates 31 and 32). They may have been derived from material included in the top terrace when its surface extended right across the present site of the river, the large blocks contained therein merely sinking to a lower and lower level as the stream eroded its bed deeper and deeper into the pre-existing cover. But in all probability

– 179 –

this does not represent the entire case. As noted earlier, the lower main terrace ends upstream in a moraine, which is at a much lower level than the uppermost terrace on the north side of the river, so the river must have cut down to the level of the second main terrace before the moraine was deposited, the vertical erosion taking place during an interglacial interval. It must be noted, however, that the cutting down of a terrace is often associated with glacier advance in areas outside those where glaciers operate, but this case is within the glacial area and the cutting down has been interglacial. The formation of terraces during a glacier maximum is frequently credited to the lowering of the sea-level and consequent stream base-level when the water has been taken from the ocean to provide the ice-cover of the land, but the terrace formation in the case of the Canterbury rivers in their course across the plains can hardly be attributed to eustatic changes, for the terraces are all lowest near the sea and rise steadily when followed inland. If they were due entirely to change in sea-level this should not be the case unless the change were accompanied by a tilting movement along an axis parallel to the western edge of the plains, and of this there is no evidence at present available, although the removal of the ice-cover and the consequent crustal adjustment has been mentioned as a possible explanation.

It is therefore probable that the river had cut down its channel to the level of the second main terrace during an interglacial interval and that there was a subsequent advance of the ice, certainly as far as the moraine inside the gorge, and perhaps much further. The accumulation of large angular masses on the surface and flanks of this terrace certainly suggests the proximity of ice. It seems hardly possible that they could not have been moved to their present positions from an ice-front near the mouth of the gorge by water alone. It is admitted that they have been subject to strong water action, but this could have taken place as the ice retreated from a more forward position. The size of the masses suggests that the ice passed the site of Mr. McIlwraith's farm, and there is also the accumulation of blocks up to seven feet in diameter just beyond Moorhouse's Road. Could a stream of the size of the Rangitata or perhaps a little bigger have moved and concentrated such blocks when it was entirely unconfined and free to wander over the surface of a plain with an inclination of about 50 feet per mile? It is possible, but I hardly think it probable. I therefore conclude that the glacier extended about six miles from the mouth of the gorge.

I should like to refer very briefly to the evidence as to the maximum extent furnished by the other glaciers that deployed on to the western margin of the Canterbury Plains during the Pleistocene period. This is most complete in the case of the Rakaia, since undoubted moraine, travelled blocks, boulder clay with scratched stones, and rounded hills composed of gravels, occur some six miles out on to the plains from the outlet of the gorge of the river. This glacier was larger than the others since it was fed by a greater length of the snow-clad main divide, and also the outlet of the valley was wider than the others and this allowed freer egress for the ice-stream.

– 180 –

The Ashburton Glacier had its origin in Mount Arrowsmith (9171 feet), a peak lying some 10 miles east of the main divide, and it was reinforced by the Cameron Glacier, also rising in Mount Arrowsmith, as well as by an overflow from the Rakaia coming along the line of Lake Heron and the Clent Hills, and by a distributary from the Rangitata mentioned earlier in this account (page). This enlarged glacier reached the vicinity of the Mount Somers-Springburn Railway, for accumulations of large stones like those in the Rangitata occur near the township of Mount Somers, and rounded hillocks similar to those marking the front of the Rakaia Glacier lie just beyond the line of railway. The analogy of the evidence to that of the Rangitata is perhaps due to the fact that after coming through a somewhat narrow gorge and over the hills in its vicinity it spread out on the lower country and thus lost concentration.

The conditions obtaining in connection with the Waimakariri were still more closely analogous to those of the Rangitata, for the glacier came through a narrow valley after leaving the great intermontane basin whose floor it occupied, and no terminal moraine, washed or unwashed by the glacial stream, is to be found either in the gorge or on the plain near its outlet. When the glacier reached the plains, if, indeed, it really did so, it must have deployed on to a wide area, and thus the proof of its former presence has become as indefinite as that of the Rangitata.

In conclusion, I should like to express my indebtedness for valuable assistance received from Messrs. F. Langbein and T. G. Beck, District Engineers of the Public Works Department, Christchurch, and specially to the latter for permission to use a map prepared in connection with the survey for the Rangitata Diversion Race. I have also to thank Mr. G. Stokell for substantial help in connection with the examination of the area under consideration.

F. References.

Cotton, C. A., 1940. Classification and Correlation of River Terraces, Journal of Geomorphology, vol. 111, no. 1.

Haast, J. von, 1879. Geology of Canterbury and Westland, Christchurch.

Hutton, F. W., 1884. On the Lower Gorge of the Waimakariri, Trans. N.Z. Inst., vol. 16, pp. 449–54.

Park, J., 1910. The Geology of New Zealand, Whitcombe and Tombs, Ltd., Dunedin.

Speight, R., 1937. The Geology of Mount Somers, N.Z. Geol. Survey Memoir, no. 3.

—— 1938. Morainic Deposits of the Waimakariri Valley, Trans. Roy. Soc. N.Z., vol. 68, pp. 143–60.