Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 77, 1948-49
This text is also available in PDF
(3 MB) Opens in new window
– 403 –

The Geology of Mandamus-Pahau District,
North Canterbury.

[Received by the Editor, March 15, 1948; issued separately, May, 1949.]


About seventy-five square miles in Mandamus Survey District, North Canterbury, has been geologically mapped on a scale of forty chains to one inch. Tertiary and Quaternary sediments rest on Mesozoic greywackes intruded by syenite and gabbro. The Tertiary and Quaternary sediments are divided into ten stratigraphical units. A brief account is given of topography and structure. Chemical analyses of a mid-Tertiary basalt and Eocene lignite are included.


The Mandamus-Pahau district lies in the northern part of Canterbury Land District. The area mapped comprises the greater part of Mandamus Survey District, and is approximately bounded on the west by the watershed between the Mandamus and the Glenrae rivers, on the north by a line through Mount Jonathan (Trig. J), on the east by a line through Ramatama (Trig. Z), and on the south by the Hurunui River for nine miles upstream from the Christchurch-Waiau railway (Fig. 1). The area thus mapped in detail is about 75 square miles. A rapid reconnaissance was made of a considerable area to the north, east and south.

The nearest centre is Culverden, about six miles to the east. Access from Christchurch is good by way of the main Culverden road as far as the Hurunui River, from which side-roads lead to the junction of the Mandamus and Dove rivers (Tekoa Road) and to the Pahau River at the Grampians Station (Balmoral Road). The greater part of the area consists of hilly country rising to about 3,500 ft.; mostly covered with native grasses, except for patches of scrub and beech forest in the stream valleys. Sheep farming is the principal use to which the land is put, with a certain amount of cropping on the ploughable country, but the area between Tekoa Road and Hurunui River was planted in exotic pines by the State Forest Service in 1920–27 and forms part of the Balmoral State Forest.

My attention was first attracted to the area by the occurrence of intrusive igneous rocks there, and the geological investigation was undertaken for the purpose of (1) studying the field occurrence of these rocks, (2) elucidating the Tertiary sequence in this district, which differs markedly from that south of the Hurunui River, and (3) preparing an adequate geological map of the district. This paper will be largely concerned with the general geology and stratigraphy of the area; it is hoped to describe the petrology of the igneous rocks in detail in a later paper.

– 404 –
Picture icon

Figure 1.

– 405 –

The geological mapping was based on the topographical maps of the Lands and Survey Department Provisional One Mile Series, the area lying in the north-west part of Sheet S61 (Culverden) and the south-west part of S54 (Hanmer). The field mapping was done on a scale of 40 chains to one inch, the topographical maps being photostatically enlarged for that purpose. Further topographical detail was added where required by compass-and-chain, compass-and-pace, and sketch methods. The more important stratigraphic sections were measured in detail. Field work was begun in October, 1944, and continued at irregular intervals till April, 1946.

I should like to take this opportunity of thanking Professor R. S. Allan for the interest he has shown in this work, for his company in the field on two occasions, and for determining the brachiopod faunules; Professor George Jobberns for stimulating discussions, especially on geomorphic problems; Dr. H. J. Finlay for the very great amount of work he has put in on the foraminifera from the various Tertiary horizons, work without which the stratigraphy would have been far less accurately determined; Dr. J. Marwick for visiting the principal Tertiary section and for determining the mollusca collected during the course of the work; and Mr. A. Hampton for having the geological maps and table redrawn for publication. Two field trips were made to this area with my advanced students, and their pleasant company and help was much appreciated. Finally, the co-operation of the local residents was most helpful, and my thanks are especially due to Mr. B. Groome, of the State Forest Service, Mr. A. Shand, Island Hills Station, and Mr. W. Lake, Cascade Station, for their hospitality and their help with transport, and in many other ways.

The cost of the field work was defrayed by a grant from the Hutton Fund, for which I am indebted to the Royal Society of New Zealand.

Previous Geological Work.

Accounts of previous geological work in this area are largely confined to papers by Haast (1871), Hutton (1877) and Speight (1918).

Haast examined this area in 1870 during the course of a survey of the Amuri district for the Geological Survey, and his report was published in 1871. He describes the geology of the Mandamus-Pahau area on pp. 29–32 of this report, and illustrates it by several sections (Sections 17, 19, 27, and 28). His description of the rock types met with is accurate, but his classification and correlations are vitiated by the erroneous determination of the Middle Tertiary limestones and submarine agglomerates and tuffs as Jurassic, apparently on the authority of Professor McCoy, of Melbourne, who examined fossils from these beds sent to him by Haast. It should be noted that in Haast's report the Dove River is called “Dora Creek”; his “Sandford Peak” (Section 19) could not be relocated, but is possibly the hill now known as Charing Cross (Trig. S).

Hutton visited the area in 1872 during the course of a general survey of the “Geology of the North-east Portion of the South Island,” but his report was not published until 1877. His report was illustrated by a geological map on a scale of twelve miles to one inch,

– 406 –

and by numerous sections, of which Nos. II, III, and XV refer to this area. He classified the oldest rocks of this area, and the igneous intrusives, in his Maitai formation, mentioning that he had found “obscure traces of plants in Mandamus River, some of which appeared to belong to a small species of Taeniopteris” … and suggested a Lower Jurassic age for them. The overlying beds, up to and including the submarine agglomerates and limestone, he placed in the Trelissic group of his Oamaru formation, of Upper Eocene age, thus correcting Haast's correlation of these rocks with the Jurassic. The fossiliferous beds overlying the limestone he placed in his Ahuriri formation, of Lower Miocene age. He also mentions an outlier of his Pareora formation (Upper Miocene) in the Mandamus, but without any description, and it is impossible to identify the rocks he may have had in mind. He briefly discussed the nature and mode of origin of the Hurunui-Waiau Basin, and also gave lists of fossils characteristic of his various formations, some of which were collected from this area.

Profesor Speight described some of the geological features of this area in a paper on the structural and glacial features of the Hurunui Valley (1918). He gave a brief account of the basement rocks, making particular mention of the syenite intrusion on the Mandamus River, and also gave a detailed description of the outlier of Tertiary rocks near the Dove River. He discussed briefly the origin of the drainage of this area, and described the salient features of the Hurunui-Waiau Basin, mentioning the isolated mass of Tertiary rocks forming Hurunui Mound… “rising like an island in the sea of gravels…”

Incidental mention of localities in this area occurs in several other papers, without, however, adding significantly to the information contained in the above papers. Hutton (1889) gave microscopic descriptions of the syenite and an andesite occurring near the Mandamus River; McKay made brief reference to the Tertiary rocks in the Culverden district and at Marble Point, on the Waiau River, but does not appear ever to have visited this area itself; and Morgan (1919) referred to “Balmoral” and “Hurunui Mound” in his general statement on the limestone resources of Amuri County. He also gave a list of papers dealing with geology of the county as a whole.

Topography and Structure.

It is not possible to discuss adequately the development of the topography and the structure of a small area such as this by itself, because it is but part of a larger region and its topographical and structural features have been impressed upon it by the reaction of this larger region to geological forces. This larger region which may be discussed as a unit comprises the whole block of country between the Hurunui and the Waiau rivers, and is shown in Fig. 2, which is reproduced from the Lands and Survey four-mile map, Sheet 23. On this map form lines have been drawn at 1,000 ft. intervals in an attempt to decipher the approximate surface upon which the present streams were developed. These form lines suggest a block of country rising with a steep scarp from what is now the valley of the Hanmer, Waiau, and Hope rivers, and from a crest overlooking this valley sloping backwards with a much more gentle slope to the Hurunui Valley. The steep scarp of this block has long been recognised as a fault scarp, and as one of the most important structural features in

– 407 –

this part of the South Island. Judged from the form lines, the vertical displacement along this fault-line increases as it is followed westwards, and there is probably a cross-fault along the line of the Pahau River with a downthrow to the east, as west of this river the summit heights rapidly reach 5,000 ft. and continue thus for many miles, whereas cast of this river the summit heights barely reach 3,000 ft. As a whole, the area of country illustrated in Fig. 1 would appear to be a typical tilted fault block in the sense that term is used by Cotton (1941, pp. 240–42).

Picture icon

Figure 2.

This being so, the remarkable parallelism of the main rivers crossing the area described in detail in this paper—the Mandamus, Glencoe, Dove, and Pahau rivers—may be plausibly ascribed to their originating as consequents on the back slope of this tilted fault block. If this back slope was originally covered with a veneer of Tertiary strata, then these rivers are now superposed consequents, superposed on the greywackes and argillites which make up the mass of this block.

Another explanation, which has been suggested to me by Professor Jobberns, is that some at least of these rivers may be antecedent to the faulting, and have been beheaded by the rapid rising of the fault scarp. The fault would thus have dislocated an originally north-south system of parallel-flowing rivers, and diverted that part of their courses north of the fault line along the base of the scarp. A study of maps of this district, especially Sheet S54 of the One Mile Provisional Series, indicates a number of features which may be significant for establishing this hypothesis. Thus both the Mandamus and the Pahau rivers have low saddles at their heads leading into tributaries of the Waiau River, that at the head of the Mandamus being only 2,500ft., and that at the head of the Pahau only 2,100ft. above sea-level, i.e. only about 1,000ft. above the Waiau Valley at these points, whereas the hills on either side rise to more than 5,000 ft.

– 408 –

above sea-level. Again, the Mandamus and the Pahau rivers are aligned in continuation with the Upper Waiau and Grantham rivers respectively, southerly-flowing streams joining the main Waiau River opposite the saddles from which the Mandamus and the Pahau rivers take their rise. Of course, these features may be purely coincidental, or they may be due to other causes, such as alignment along important north-south fractures, and the above hypothesis can only be satisfactorily tested by a physiographic examination of the Waiau Valley, something which lies outside the scope of the present paper.

The parallel pattern of the Mandamus, Glencoe, Dove, and Pahau rivers has been modified to some extent by other geological factors. The Dove River, after flowing in a southerly direction for the first five miles of its course, turns sharply to the west and flows for three miles in this direction to join the Mandamus. This latter part of its course is apparently determined by the strike of the inlier of Tertiary rocks in this valley. Similarly the considerable stream flowing westwards parallel to the lower course of the Dove from below Hurunui Peak to the Mandamus River near Tekoa Station, was probably developed originally along the strike of the Tertiary rocks. which presumably extended over the Hurunui Peak ridge, and with the removal of the Tertiary rocks by erosion has been let down into the massive syenite through which it now flows for the greater part of its course.

Glencoe River apparently presents an example of capture. From the Hurunui Peak ridge the north-south valley of the Glencoe is seen to be continued by a broad and low (1,700ft.) saddle into the valley of the Dove. It is probable that the Glencoe originally continued its southerly course until it met the Dove, but that it has since been tapped off by a small stream working headwards from the Mandamus, the sharp bend in the Glencoe where it turns to join the Mandamus then being an elbow of capture (Plate 43). This suggested small stream responsible for the capture may have been established on non-resistant Tertiary strata, now eroded away, forming a line parallel to the outlier along the Dove.

The structure of the Tertiary rocks in this area is interesting in that it does not conform with the comparatively simple pattern characteristic of much of North Canterbury, viz. the arrangement as moderately dipping tilted strips preserved along fault-angle depressions, the present structure having been accomplished apparently by faulting without much folding. In the Mandamus-Pahau area, on the other hand, the uplift of the Tertiary rocks has been accompanied by strong folding. In the strip of Tertiary rocks along the Pahau Valley most dips are in the vicinity of 90°, and the structure appears to be due to isoclinal folding, considerably complicated by local faulting. The syncline of Tertiary rocks in the Dove and Mandamus valleys (Plate 43, Fig. 2) probably formed the northern extension of an anticlinal structure represented by the Hurunui Peak ridge, from which the Tertiary rocks have since been removed. save for isolated fragments along its southern base—the small mass of limestone near Flaxdown homestead, and the Lower Tertiary rocks poorly exposed in the banks of the Hurunui River below the Mandamus junction. The axis of this anticlinal fold would cross the Hurunui River near

– 409 –

the Mandamus junction, and the steep (ca. 50°) dip slopes of limestone visible on the south side of the Hurunui from this point probably are part of the southern limb of this anticline, which has been practically removed by erosion on the north side of the river.

The plains in this Mandamus-Pahau area are part of the large continuous extent of flat-lying gravels forming the floor of the Hurunui-Waiau Basin (Plate 44, Fig. 3). Here, again, it would be unwise to attempt to draw far-reaching conclusions as to the nature and mode of origin of this great depression from a study of the features of this small area. However, certain observations may here be recorded. First, there is evidence for the presence of at least two sets of flatlying beds flooring this basin. Along the north bank of the Pahau River between Ramatama Creek and Black Birch Creek these two sets of beds are clearly exposed. The lower set consists of well-stratified clays, sandy clays, and gravels, in bands generally from 4 to 8 ft. thick. The gravels appear to be made up entirely of greywacke pebbles, no fragments of Tertiary rocks having been seen. A feature of the clay and gravel bands in places is the presence of numerous granules of bright blue vivianite, up to about half an inch in diameter. This lower set of beds is disconformably overlain by coarse greywacke gravels, similar to those in the present riverbed, the upper surface of which forms the present surface of Hurunui-Waiau Basin in this part. From their lithology the lower set of beds has obviously been deposited under entirely different conditions from those prevailing when the upper set was laid down, and under entirely different conditions from those ruling in the present-day river beds. The presence of this lower set of flat-lying beds underlying the floor of the Hurunui-Waiau Basin implies a very different topography of the surrounding country at the time they were laid down.

This lower set of beds has not been observed anywhere else within the area examined. No sections are exposed in the high banks along the Hurunui River; the banks are all grassed or obscured by talus from the coarse gravels forming the surface of the plain. However, a search over a wider area in the Hurunui-Waiau Basin, coupled with a scrutiny of logs of wells put down in the basin, may reveal further information as to nature and distribution of this lower set of beds.

The floor of the Hurunui-Waiau depression is evidently somewhat irregular. Within the Mandamus-Pahau area a small hill of Tertiary rocks projects through the gravels—Hurunui Mound—on the north bank of the Hurunui where the railway crosses the river. This outcrop of solid rock, as contrasted with gravels, is the only one along the north bank of the river for nearly twenty miles from the Mandamus junction to the mouth of the gorge below the confluence with the Pahau, and has been utilised to provide a solid abutment for the railway bridge at this point. Other inliers of Tertiary rocks projecting through the gravels of the Hurunui-Waiau basin are the small hill (locally known as The Mound), upon which the Culverden cemetery is sited, and Mount St. Leonards, a hill rising 200 ft. above the plain four miles east of Culverden township. “The Mound” appears to have a cap of gravels covering most of its surface, but limestone has been struck at no great depth during excavations, and is at present exposed at the rifle butts on its northern side.

– 410 –

The cap of gravels on the surface of “The Mound” is a minor link in a chain of evidence indicating that the present surface of the Hurunui-Waiau Basin is the result of re-excavation, the basin having at one time been filled with gravels to a much greater depth. The presence of a lower set of flat-lying beds flooring the basin at least along the Pahau River suggests this possibility, which is further strengthened by the presence of high-level terraces around the margins of the basin. (During the course of the present work a prominent terrace remnant about 750 ft. above the present river level was noted on the south bank of the Hurunui River some distance upstream from the Mandamus junction, and high-level terraces bordering the mouth of the gorge through which the Hurunui leaves the basin are clearly visible from the main road through Culverden.) Along the southern part of the basin, in the Hawarden district, there is a group of rounded hills (Trigs. A, C, and N, Sheet S61), apparently largely made up of gravels (outcrops are practically absent), all rising to a common altitude of about 1300 ft., about 500 ft. above the present, surface of the plains; these hills probably represent residuals of an old surface preserved from total removal by their position along the watershed between the Waipara, Waitohi (a tributary of the Hurunui) and Waikari rivers. The evidence is there; but the working out of the hypothesis is a research project in itself, unconnected with the purpose of this paper.


For reasons similar to those advanced by Gage and Wellman (1944) the twelve distinct lithologic units recognised in the geological mapping of this area were marked by symbols numbering from M1 to M12. This was done on account of the difficulty of applying New Zealand Tertiary stage names or regional formation names directly to the lithologic units recognised in this area, and to avoid burdening the nomenclature with a multiplicity of stratigraphical terms, which may have no more than provisional significance. A further reason for this course is that the Amuri Subdivision, which adjoins this area on the east, has been partly mapped by the New Zealand Geological Survey; as most of the lithologic units recognised in the Mandamus-Pahau area are present also in the adjoining part of the Amuri Subdivision, when the survey of the latter district is completed and published it will be possible to correlate the lithologic units designated M1 to M12 with the formations recognised in the standard sequence for the Amuri Subdivision.

The composite section published herewith (Fig. 3) gives the main features of the stratigraphical succession in this area, and indicates probable correlations with the formations recognised in the Amuri Subdivision (according to the provisional scheme of Fyfe, 1931), and with the classic Middle Waipara and Weka Pass area (Thompson, 1920). The important features of each of the lithologic units recognised will now be discussed in detail.

M1: Greywacke and Argillite.

The greywacke and argillite in this area, along with the intrusive igneous rocks, form the basement on which the Tertiary rocks lie. They form the north-west part of the area, and extend north and west to the Waiau Valley and beyond. In nature and appearance the greywacke and argillite are similar to the rocks which make up the most of the mountain country of Canterbury and Marlborough. In this area the main mode of occurrence is as alternate bands of greywacke and argillite, although massive greywackes are also common. In thin section the greywacke is seen to consist almost entirely of rounded and subangular grains of quartz and decomposed feldspar in variable amounts, with small amounts of chlorite and other accessories. As judged by the boulders brought down by the rivers, the massif of Mounts Tekoa and Te Kooti is largely made up of conglomerate, presumably interbedded with the greywacke; the pebbles in this conglomerate are well rounded, generally up to about 1 in. in diameter, and consist largely of greywacke also, but notable quantities of quartz and jasper are present, and rare pebbles of granite were observed.

– 411 –
Picture icon

Figure 3.

– 412 –

A band of coarse conglomerate interbedded with massive greywacke is exposed in Mandamus River just below the junction with the Dove, and along the road on the west bank. The pebbles in this conglomerate are up to 2 ft. in length, rounded and subangular; they consist mainly of greywacke, but pebbles of porphyritic igneous rocks are not uncommon.

Observations on the strike and dip of the greywacke and argillite indicate an average strike a little west of north. The great majority of strikes recorded ranged between 320° and 25° (true). Dips are uniformly high, on either side of 90°. The rocks are much folded, faulted and jointed.

The only fossils found in these rocks were indeterminable plant remains—carbonised stem and leaf fragments—which are not uncommon in the greywackes of the Dove River. In this connection it is interesting to note that Hutton (1877) collected what he believed to be a small species of Taeniopteris in the Mandamus River. Even if this identification can be relied upon, it is of little value for exact dating, but it indicates a Triassic or Jurassic age for these rocks.

M2: Intrusive rocks.

The intrusive rocks of this area are of considerable variety. The largest mass is the syenite extending from Mandamus River along the Hurunui Peak ridge for some miles. A small body of coarse olivine gabbro is present in Hut Creek, a considerable tributary of Dove River draining the western slopes of Charing Cross (Plate 44, Fig. 4). Numerous smaller intrusions are common in the greywackes adjoining the syenite and gabbro, and diminish in abundance away from these rocks. These smaller intrusions are especially well exposed along the lower course of Dove River. There are numerous dykes in the upper part of Pahau River, beyond the area described in this paper, but they are independent of the intrusions described here, being probably feeding channels for the agglomerates of M4. The smaller intrusions appear to be mainly sills, being generally accordant to the bedding in the greywacke and argillite. They are generally not very thick, few being more than 10 ft. In composition they are mainly trachytic, but a few lamprophyric types occur (Plate 44, Fig. 5).

The petrography of these rocks will form the subject of a separate paper. They are younger than M1, being intrusive into the greywackes and argillites of that formation, but are older than the Tertiary rocks,

– 413 –

as none of the intrusions penetrate these rocks, and pebbles of syenite and other intrusive rocks occur with greywacke in the basal Tertiary conglomerates in Coal Creek. Lithologically they show a close resemblance to the intrusive rocks of the inland Kaikoura Mountains, which were collected during a trip through the Middle Clarence Valley in February, 1946. Thomson (1919, p. 308) inclined to the belief that these intrusive rocks were of Clarentian age, being influenced towards that belief by the occurrence of masses of Clarentian lavas on the flanks of the Inland Kaikoura Mountains. There is no evidence of Clarentian lavas in the Mandamus-Pahau area, but the similarity of the intrusives to those of the Inland Kaikoura Mountains suggests a correlation between the two. The intrusives of the Mandamus-Pahau area must be pre-Tertiary, and their age may well be Clarentian, i.e. Lower Cretaceous.

M3: Glauconitic quartz sandstone, etc.

This formation is the basal part of the Tertiary succession in this area. It includes all those beds up to the incoming of the volcanic material that marks the base of the succeeding formation, M4. The most complete and best exposed sections are those in the Mandamus River and Coal Creek. In Coal Creek, however, the succeeding formation has been removed by erosion, and some of the upper part of M3 is thereby missing, and in the Mandamus River the upper part of M3, including the contact with the overlying formation, is well exposed, but the lower part is somewhat obscured. There are some isolated outcrops of this formation in the banks of Hurunui River about 800 yards downstream from the Mandamus junction. Along Pahau River and its tributaries it is exposed at a number of places. A complete sequence is probably present in the isoclinal fold exposed in the strong bend of Pahau River below Hugh Gully, but this section is very much slumped. Isolated outcrops occur along this fold at intervals as it is followed southwards. A fairly complete section is present in Awatui Stream, but here the basal beds may be missing on account of faulting. There are a number of exposures in Ramatama Stream, but here the beds are very much disturbed and the sequence is obscure.

The material making up M3 is principally a glauconitic quartz sandstone, varying in composition from practically pure glauconite to a quartz sandstone with only a little glauconite, the latter type predominating. Glauconite is most abundant towards the base of the formation, where it is sometimes concentrated into bands of nearly pure greensand. The quartz grains are generally quite angular, except in the coarser sandstones, which contain rounded quartz grains 1–2 mm. in diameter. The degree of induration is very slight, most sandstone being easily rubbed down between the hands; occasional cemented bands are found, however, the cementing medium being calcite. In some places the beds are riddled with borings, presumably by marine organisms.

Although glauconitic sandstone is the predominant material in this formation, there are some bands of glauconitic mudstone in the lower part. In Coal Creek, where the base of this formation is clearly exposed, the sequence is as follows:—

200 ft. ± Glauconitic mudstone, overlain by glauconitic sandstone,

– 414 –

100 ft. ± Greensand becoming less glauconitic in the upper part. 1 ft. 6 in. Puggy clay, possibly due to movement.

20 ft. Coarse conglomerate, with rounded and subangular boulders, up to 2 ft. 6 in. in diameter, the boulders becoming smaller in the upper part. The boulders are mainly of greywacke, but syenite and other igneous rocks are represented.

2 ft. Lignite, containing much iron sulphide (see appendix for analysis).

5 ft. Conglomerate, made up of rounded pebbles, up to 3 in. in diameter.


In other localities where the contact between M3 and the underlying greywacke is actually exposed, viz., in Pahau River below Hugh Gully, and Awatui Stream, the conglomerates and coal are absent. The contact in Awatui Stream may be a fault, but in Pahau River shattered greywacke is seen in the core of the isoclinal fold, and is immediately overlain by glauconitic mudstone. It appears that the conglomerate and the coal occur only in the western part of the area.

The thickness of M3 is probably about 300 ft. in Mandamus River and Coal Creek, although an exact measurement is not possible, and about 150–200 ft. in Pahau River. It is separated by a great unconformity from the underlying greywacke and argillite.

The only macrofossils which have so far been found in this formation are fragments of a large Ostrea and some shark's teeth, of little value for correlation or dating. A number of samples were submitted to Dr. H. J. Finlay for foraminiferal examination. Some were barren, and the remainder in general had poor faunas.

In the Coal Creek section a sample from the puggy clay immediately overlying the coarse conglomerate gave a small fauna indicating Mangaorapan or Heretaungan age, a sample 80 ft. higher in the section gave a small fauna indicating Heretaungan age, and a third sample (F.6416) gave an indeterminate assemblage of a few battered foraminifera, not higher than Duntroonian and not lower than Kaiatan. A sample from the base of the sequence in Pahau River (F.6449) gave a small but typical assemblage which is either Heretaungan or just below this stage. A sample from this formation where it is exposed in Ramatama Stream (unfortunately it cannot be localised to a definite position within the formation) gave a poor fauna indicating a Bortonian age.

Thus Mangaorapan, Heretaungan, Bortonian, and higher stages are represented within M3. The upper limit can probably be taken as Whaingaroan, as the succeeding formation is Duntroonian. It thus includes a long time interval, the greater part of the Eocene and the base of the Oligocene. Either deposition continued slowly but continuously throughout this period, or disconformities are present within the formation, caused by non-deposition or possibly by erosion. The comparatively thin succession militates against the first hypothesis, and the second is more probable.

Similar problems are present in the probable correlatives of this formation in other parts of North Canterbury. Probable correlatives are:—

– 415 –

Beds correlated with the Amuri Limestone in the adjacent part of the Amuri Subdivision. These include all rocks above the Cretaceous and below the Cookson beds. Fyfe (1931, p. 6) wrote: “Only in the neighbourhood of Monkey Face and the northern end of Whale Back are the beds correlative with the Amuri Stone typical of that rock. Elsewhere it is represented by calcareous sandstone or argillaceous sandstone, all more or less glauconitic, the more glauconitic facies presenting a mottled appearance due to the presence of numerous fine worm borings with a non-glauconitic filling.” This latter type of rock evidently corresponds closely to the typical material of M3.


The beds in the Middle Waipara and Weka Pass district classified under the heading of “Amuri Limestone” by Thomson (1920). These included all rocks between the top of the Waipara greensands and the base of the Weka Pass greensand. The Amuri limestone forms the top of this sequence, and passes down into marls (“Chalk Marls”), glauconitic mudstones, and glauconitic sands. Dr. Finlay states (private communication) that he has recognised Mangaorapan, Heretaungan, Bortonian, Kaiatan, Runangan, and Whaingaroan horizons in samples from this sequence.


The Amuri limestone, Karetu sandstone, and Ashley mudstone of the Mount Grey district (Mason, 1941). The Ashley mudstone yielded a good foram fauna indicating a Mangaorapan age (sample F.5328), a sample 150ft. above the base of the Karetu Series (F.5449) was upper Bortonian, a sample from the glauconitic base of the Amuri limestone (F.5448) was Runangan and a sample from the middle of the limestone (F.5575) was Whaingaroan. Samples from the critical part of the Karetu sandstones between F.5449 and F.5448 have not so far yielded foram faunas.

The absence of the Amuri limestone from the Mandamus-Pahau area and from the adjacent part of Amuri Subdivision and its apparent replacement by glauconitic sandstone poses an interesting question, in view of the widespread and uniform nature of this formation throughout North Canterbury and Marlborough. I would suggest that it may be due to this area lying on the margin of the trough or basin in which the Tertiary rocks of North Canterbury and Marlborough were laid down, and that while the Amuri limestone was accumulating in the centre of the basin, sands were being deposited along the margin. This suggestion requires an examination of the whole of the North Canterbury Tertiary for its working-out, but there are some significant facts which may be stated. In North Canterbury and Southern Marlborough the Amuri limestone appears to be thickest approximately along the present coastline (Montserrat 200–300 ft.; Motunau 300 ft.; Napenape Cliffs 300 ft. ±; Amuri Bluff 300 ft.) and to thin when traced towards the north-west. The earliest post-Hokonui beds in the Mandamus-Pahau area are lower Eocene, whereas to the south-east, in the Weka Pass district, they are Cretaceous, suggesting a progressive overlap in a north-westerly direction. The occurrence of middle Tertiary volcanism in the Mandamus-Pahau area, continuing north-east into the Amuri Subdivision, suggests a zone of instability which may be linked with its marginal position to this suggested basin of deposition. And the Tertiary rocks of the

– 416 –

Mandamus-Pahau area are farthest to the north-west of all the Tertiary rocks in North Canterbury, which again may indicate that the margin of the basin extended very little further in this direction.

M4: Agglomerates and tuffs.

This formation overlies M3 and underlies conformably M5, a semi-crystalline shell limestone. The best section through it is in the east bank of Mandamus River opposite Tekoa homestead, as here it is continuously exposed from the top of M3 to the base of M5. There are good sections all along Pahau River and in its side streams, but on account of the disturbed nature of the beds here a clear and continuous section from the base to the top of this is difficult to find; the most promising is that from the base of M5 in Cascade Creek down to its junction with the Pahau, and thence along the left bank (going downstream) to the point on the map where the line of M3 meets the river, about 200 yards below the mouth of Cascade Creek. A strip of this formation occurs along the south-eastern base of Charing Cross, being well exposed in the upper part of Cascade Creek and extending thence in a northerly direction nearly to Awatui Stream. An isolated occurrence of calcareous tuff associated with the glauconitic sandstones and mudstones of M3 on the south bank of Hurunui River below the Mandamus junction probably belongs to this formation also.

Along Pahau River and its tributaries the rocks of the formation are mainly coarse black agglomerates with a tuffaceous matrix; the beds at the base of the series are comparatively fine tuffs. Occasionally the agglomerates show flow and pillow structures, and in the bend of the Pahau below Hugh Gully the rock of this formation is apparently a solid lava flow with a little interbedded ash.

Through the good offices of the Director of the New Zealand Geological Survey a typical specimen of these submarine basalts was analysed in the Dominion Laboratory with the following results:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

SiO2 46.55
Al2O3 12.88
Fe2O1 6.61
FeO 5.96
TiO2 3.15
MgO 5.99
CaO 9.96
Na2O 2.90
K2O 1.26
H2O+ 1.72
H2O 2.32
CO2 0.27
P2O5 0.52
S 0.02
MnO 0.26
Cr2O3 0.04
ZrO2 Nt. fd.
BaO 0.04
Cl trace

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

C.I.P.W. Norm.
Q 2.5
Or 7.46
Ab 24.54
An 18.42
wo 10.80
di en 8.98 20.25
fs 0.47
en 5.93
hy 6.23
fs 0.30
mt 9.58
il 5.98
ap 1.25
(c.c.) (0.61)

Symbol: III. 5. 3. 4. Camptonose

Basalt, Pahau River, North Canterbury. Analyst, T. A. Rafter.

Picture icon

Fig. 1—View north from Hurunui Peak. The valley in the foreground is that of the Dove River flowing west. The long straight valley in the centre of the picture is that of the Glencoe River which originally flowed into the Dove River, but has been captured by a small tributary of the Mandamus River.

Picture icon

Fig. 2—Synclme of Tertiary rocks seen from near Tekoa Stream. The Mandamus River flows from left to right across the foreground. The base of the limestone, M5, is outlined with white dots.

Picture icon

Fig. 3—View east from Hurunui Peaks. The Balmoral State Forest is in the middle distance on the right. The downs projecting through the plains in the left middle distance are of Tertiary rocks. rising to a height of 1,625 ft. in Green Hill Trig.

Picture icon

Fig. 4—View west from Charing Cross down Hut Creek. Gabbro outerops in creek bed at A; B is junction of Hut Creek and Dove River.

– 417 –

The analysed specimen is a holocrystalline medium-grained olivine basalt containing 5–10% of olivine in idiomorphic to subidiomorphic phenocrysts up to 1 mm. long. The olivine is largely altered to greenish-yellow chlorite. The feldspar is labradorite in lath-shaped crystals averaging 0°2 mm. in length. A few small phenocrysts of colourless augite are also present, and the groundmass consists of a fine-grained aggregate of skeletal crystals of ilmenite and magnetite and granules of augite. Thin sections of other specimens of these rocks show similar mineralogical composition, although some are hypocrystalline, with crystals of olivine, augite, and feldspar set in a matrix of brown glass, and others are amygdaloidal, with the amygdules completely or partly filled with calcite, with which analcite also occurs occasionally.

The centre of eruption of these agglomerates seems to have been near Pahau River, because in the outlier of Tertiary rocks in the Dove-Mandamus area they are represented by fine-grained red-brown calcareous tuffs. As igneous material of any kind is unknown in the Tertiary rocks of the Waikari-Weka Pass district, the country on the south side of the Hurunui River was rapidly examined during the course of this survey in order to determine how far this tuffaceous horizon could be followed to the south. The tuff content gradually decreases, and about a mile beyond the river, behind the Waitohi Downs homestead, the formation passes into a calcareous sandstone lithologically resembling the Weka Pass Stone.

Interbedded with the agglomerates in the Pahau River area are lenses of limestone, up to 60 ft. thick in places, but thinning rapidly both in the direction of the dip and of the strike. These limestones are lithologically very similar to the limestone M5 overlying this series, and the number of limestone bands in this very disturbed area caused some difficulty in interpretation during the early part of the field work. They were evidently deposited during lulls in the volcanic activity. Another interesting rock in this series is a curious conglomerate with pebbles of agglomerate set in a limestone matrix.

In places these agglomerates show cavities filled with well-crystallised zeolites; natrolite, apophyllite, and analcite have so far been identified (Mason. 1946) and thorough collecting may yield further species.

In the Pahau area it is difficult to be sure of the thickness of these agglomerates, as they are repeated by folding and possibly also by faulting, but it is at least 150 ft. and is probably about 200–300 ft.; in the Dove-Mandamus area, where tuffs alone occur, the thickness is some 60 ft.

These rocks undoubtedly originated from the submarine eruption of basaltic material, as shown by the nature of the rocks themselves and the occurrence of marine fossils and limestone lenses within them. This formation is apparently conformable with the rocks above and below; the contact with the rocks of M3 is a sharp one, but there is no discrepancy in strike and dip, nor is there any sign of erosion between them, the incoming of volcanic matter marking the junction of the two formations.

Macrofossils are rare in these beds, as might be expected from the nature of the rocks, Fossils have been found at several places in the

– 418 –

Pahau River area, but the best locality for collecting is in the agglomerates and the interbedded limestone lens exposed as a high cliff across Pahau River from the woolshed on the Grampians Station. In the finer beds near the base of the formation at this place numerous specimens of the brachiopod Liothyrella cf. boehmi (Thomson) were collected, and the same brachiopod appears to be abundant in the limestone lenses, but cannot be extracted for examination. Other fossils collected include Mesopeplum cf. burnetti (Zittel), Chlamys sp., Serripecten sp., a coral, echinoid spines, and (at the base of these beds in Ramatama Stream) what appears to be the calyx and portions of the stem of a crinoid.

Only one sample from this formation gave a foram fauna on washing, and of this Dr. Finlay writes:—

“F.6390, 50ft. above base of tuffs and agglomerates, Pahau River, opposite Grampians woolshed. A very poor fauna, not older than Duntroonian, but internal evidence is lacking to fix the upper limit definitely. It is not Hutchinsonian, and unlikely to be Waitakian.”

On the fossil evidence, combined with that from the stratigraphical succession as established from the field work, the age of these beds is taken to be Duntroonian. This is also the accepted age for the Cookson beds of the Amuri Subdivision, which are lithologically identical with the rocks of M4, and occupy a similar stratigraphical position. M4 together with the overlying limestone M5 may be correlated with the Weka Pass Stone of other parts of North Canterbury, and indeed, when traced south from the Hurunui River away from the Mandamus-Pahau area it grades into a calcareous sandstone similar to the typical Weka Pass Stone.

M5: Limestone.

Wherever the top of the agglomerates and tuffs of M4 can be recognised with certainty (and this is not always possible in the strongly folded area along Pahau River), it is found to be overlain by a hard, semi-crystalline shell-limestone, similar in appearance to the limestone lenses interbedded with the agglomerates themselves. A survey of a wider area is necessary before it can be stated that the limestone of M5 is a constant and persistent horizon. Along Mason River, north of Waiau, the Cookson beds (the equivalent of M4) are overlain directly by the Sugarloaf beds without the interposition of limestone.

Along Pahau River this limestone has been certainly identified at only one place, in the bed of Cascade Stream just below the road crossing. Here the limestone is about 30 ft. thick, is practically vertical, and stands out as a resistant band over which the stream tumbles in a waterfall about 100 ft. high. On the downstream side the limestone is underlain by the agglomerates and tuffs of M4, and on the upstream side it is overlain by the glauconitic sandstone of M6. The limestone here cannot be traced along the strike in either direction, the ground surface along the line of strike being a level terrace underlain by gravel.

This limestone has not been certainly identified elsewhere in the Pahau Valley, as nowhere else is there a section showing the three formations M4, M5, and M6 in stratigraphical succession. An isolated

– 419 –

patch of limestone at the base of the Hurunui Peak ridge, near Flaxdown homestead, may belong to this formation, but neither the underlying nor the overlying formation is here exposed. The limestone of M5 is, however, well exposed in the Dove-Mandamus outlier, where it is about 60 ft. thick and caps the syncline of M3 and M4, its resistant nature protecting the softer underlying rocks from erosion. The limestone is the highest bed of the Tertiary succession in this outlier, succeeding formations having evidently been removed by erosion.

The limestone of M5 is made up almost entirely of comminuted shells of brachiopods and molluses, which have been cemented by calcite to form a hard and resistant rock. The main part of the limestone is white to cream-coloured, but towards the base the presence of small amounts of tuffaceous material gives it a handsome pink to red tint. Although it is a shell limestone it is not possible to extract fossils, except from some softer tuffaceous bands near the base in the Dove-Mandamus area, where a few indeterminate brachiopods, in a poor state of preservation, were collected.

Deposition of the limestone apparently followed closely upon that of the agglomerates and tuffs, as the latter pass rapidly into the limestone as the amount of volcanic material falls off. It is perhaps questionable whether the limestone should be considered a distinct formation from the agglomerates and tuffs; the presence of limestone as lenses in the agglomerates and tuffs suggests that the deposition of the two rock types was closely interlinked, temporary or permanent cessation of volcanic activity resulting in the deposition of limestone.

The limestone of M5 is too hard to yield a foram fauna by the usual technique, and the contained fossils are too poorly preserved to be of use in the accurate dating of this formation. Stratigraphical evidence, however, indicates that the age of this formation is probably Duntroonian; the overlying formation, M6, is Waitakian, the underlying agglomerates and tuffs are probably Duntroonian, and as the limestone belongs lithologically with the underlying beds rather than with the overlying glauconitic sandstone, from which it is separated by a bored surface, M5 is placed along with M4 in the Duntroonian stage.

The limestone of M5 is identical in lithology and stratigraphical position with the Isolated Hill limestone of the Amuri Subdivision, and is therefore correlated with that formation. A similar limestone is that of the Tertiary outlier at Marble Point, in the gorge of the Waiau River between Culverden and Hanmer, from which the well-known building stone, “Hanmer Marble,” is extracted. The correlative of M5 in the rest of North Canterbury is the Weka Pass Stone, into which it appears to grade when followed south beyond this district.

M6: Glauconitic sandstone.

This formation is founded on a single exposure, in Cascade Stream, where it overlies the limestone of M5 and underlies the grey muddy sandstone of M7. It may be merely a basal phase of M7, but it is lithologically distinct, and in this section is clearly marked off both from M5 and M7. Its total thickness in this section is only some five feet; the top of the limestone upon which it rests is bored to a depth of 12 in. to 18 in. and the borings are filled with the glauconitic sandstone, suggesting an interval of non-deposition between

– 420 –

the two formations; the upper surface of the glauconitic sandstone is a hard concretionary layer about a foot thick, above which lies the grey muddy sandstone of M7, which is lithologically distinct in being free from glauconite.

One other exposure of a lithologically similar glauconitic sandstone is tentatively correlated with M6. It crops out on the north bank of Pahau River, one mile below the mouth of Cascade Stream, where it appears along the edge of the riverbed for about 50 yards. The section here is not well exposed, but the glauconitic sandstone appears to be faulted against agglomerates belonging to M4 on the downstream side, and to be overlain by grey muddy sandstones of M7 on the upstream side. Although the lithology is similar to that of the rock overlying the limestone of M5 in Cascade Stream, the exposure in Pahau River is distinctive in being fossiliferous, whereas no trace of macrofossils has yet been seen in the glauconitic sandstone in Cascade Stream. However, the exposure in Cascade Stream covers only a few square feet, so the absence of macrofossils here may be more apparent than real.

The correlation of these two exposures is confirmed to some extent by the microfaunas. Samples from both exposures were examined by Dr. Finlay (F.6389, Cascade Creek; F.6410, Pahau River). Neither gave absolutely distinctive faunas, but Dr. Finlay is inclined to place both in the Waitakian stage. He suggests, however, that F.6410 may be somewhat higher in the sequence than F.6389. This may mean that the sample from Pahau River comes from the upper part of M6, or that it may in fact represent a glauconitic facies of some horizon near the base of M7. The first hypothesis is accepted for the present, but the second cannot be excluded.

The macrofossils from the glauconitic sandstone on the north bank of Pahau River comprise numerous brachiopods, lamellibranchs, and irregular echinoids, but the matrix is tough and it is difficult to extract good specimens. The following is a list of identified specimens:—

Lentipecten hochstetteri (Zittel) Notocallista (Fossacallista) sp.
Lima cf. colorata (Hutton) Baryspira sp.
Serripecten cf. hutchinsoni (Hutton) Aetheia gualteri (Morris)
Waiparia cf. elliptica (Thomson)
Serripecten cf. beethami (Hutton) Pachymagas n.sp.
Limopsis cf. zitteli von Ihering Flabellum sp.
Spissatella n.sp. Dentalium cf. solidum Hutton

The stratigraphical and palaeontological evidence indicates that the age of M6 is Waitakian. There is thus some evidence for a minor time break between M5 and M6, probably a period of non-deposition during which M5 formed the sea floor and its upper surface was bored by marine organisms.

M7: Muddy sandstone.

The best section for this formation (and for all from M5 to M10 inclusive) is in Cascade Stream. Along this stream it is exposed practically continuously from its base just below the road crossing to the contact against M8, about half a mile upstream. It is practically vertical throughout, and the total thickness (assuming no repetition

– 421 –

by faulting, for which there is little evidence) is at least 1,700 ft. This formation apparently forms the eastern face of the downlands extending south from the valley of Cascade Stream nearly to the Tekoa road, but is poorly exposed throughout, there being a few outcrops in the lower part of Balmoral Stream, and occasional boulders of the harder shelly bands on the surface of the ground in places. Incomplete sections of this formation, interposed with the limestones and agglomerates of M4, are present in the banks of Pahau River below the mouth of Cascade Creek. The outlier of Hurunui Mound consists of hard shell beds of this formation, which also form the north bank of Hurunui River below the Mound for about 500 yards.

The typical rock of this formation is a grey muddy sandstone occasionally slightly glauconitic, comparatively soft, except when cemented by calcium carbonate into concretionary bands. These concretionary bands are often rich in molluscan fossils, and mollusca occur sporadically throughout.

The mollusca from this formation have been identified by Dr. Marwick, and the following fauna is recorded:—

Hurunui Mound:
Glycymeris (Glycymerita) sp. Tropicolpus albolapis (Finlay)
Anomia sp. Polinices sp.
Eumarcia enysi (Hutton) Euspinacassis sp.
Hedecardium cantuariense (Laws)

Many casts of smaller venerids, perhaps juveniles of E. enysi, Dentalium cf. solidum (Hutton).

The following collections come from a number of horizons exposed along Cascade Creek and are enumerated from the highest downwards:—

  • 20 ft. below top of M7:

  • Glycymeris af. huttoni Marwick

  • Eucrassatella ampla (Zittel)

  • Polinices cf. huttoni von Ihering

  • Baryspira cf. robusta (Marwick)

  • Baryspira hebera (Hutton)

  • 140 ft. below top of M7:

  • Polinices huttoni von Ihering

  • 200 ft. below top of M7:

  • Tropicolpus albolapis (Finlay)

  • Polinices aff. huttoni von Ihering

  • Polinella lobata (Marwick)

  • Alcithoe cf. reflexa Marwick

  • Baryspira cf. hebera (Hutton)

  • Zeacuminia cf. pareoraensis (Suter)

  • 240 ft. below top of M7:

  • Modiolus sp.

  • Gari cf. lineolata (Gray)

  • Baryspira robusta (Marwick)

  • Comitas sp.

  • 550 ft. below top of M7:

  • Glycymeris aff. huttoni Marwick

  • Eucrassatella ampla (Zittel)

  • Venericardia awamoaensis (Harris)

  • Bartrumia tenuiplicata (Bartrum)

  • Zenatia sp.

  • Panope cf. worthingtoni Hutton

  • Zeacolpus n.sp.

  • Tropicolpus albolapis (Finlay)

  • Polinices huttoni von Ihering

  • Struthiolaria (Callusaria) cf. spinosa Hector

  • Galeodea apodemetes Marwick

  • Austrosipho cf. marwicki (Finlay)

  • Baryspira spinigera (Marshall)

  • Baryspira (Alocospira) sp.

  • Conospirus trailli (Hutton)

  • Austrotoma sp.

  • Rugobela canaliculata (Suter)

  • 800 ft. below top of M7:

  • Maoricolpus cf. cavershamensis (Harris)

– 422 –
  • Polinices aff. huttoni von Ihering

  • Zelandiella subnodosa (Hutton)

  • Zeacuminia pareoraensis (Suter)

  • 850 ft. below top of M7:

  • Paphirus aff. largillierti (Phillippi)

  • Eumarcia (Atamarcia) cf. sulcifera Marwick

  • Hedecardium cf. cantuariense (Laws)

  • Maoricolpus cf. cavershamensis (Harris)

A number of samples from M7 were submitted to Dr. H. J. Finlay, and the microfaunas show that this formation, although a lithologic unit, includes a number of distinct stages. A sample from the base of the formation in Cascade Stream (F.6388), just above the hard concretionary band at the top of M6, gave an indecisive fauna, reported on as Waitakian or Otaian; a sample from the top of the formation, just below M8 (F.6413) gave a Lillburnian fauna. Intermediate samples from the Cascade Stream section were disappointing, having either no fauna or only a few battered foraminifera insufficient to date them. A sample from the inlier of this formation on the north bank of the Hurunui River below Hurunui Mound (F.6412) gave a probable Hutchinsonian fauna, as did F.6411 (north bank of Pahau River. 105 chains below Cascade Stream), and a sample from the south bank of Pahau River, 30 chains below Cascade Stream. The molluscan faunas have a generalised Awamoan appearance, that from 550 ft. below the top indicating definite Awamoan.

This formation thus covers a considerable time interval, from Waitakian or Otaian at the base to Lillburnian at the top. Lithologically it is remarkably uniform throughout, and no evidence of stratigraphic breaks within it was noted during the present survey. In this respect it is similar to the Mount Brown Series of the Weka Pass area (up to the Hinnites shell bed), with which it may be correlated.

M8: Calcareous sandstone.

This horizon is characterised by the presence of a large species of the brachiopod genus Stethothyris, which can readily be recognised in the field even as fragments. The rock type is a yellow-brown calcareous sandstone some 30 ft. thick, moderately soft on the whole, but irregularly cemented by calcium carbonate into resistant lenses and bands. It is well exposed in the two branches of Cascade Creek, and is occasionally visible in the bed of Balmoral Stream when floods have removed the debris which generally obscures it here. It is occasionally in evidence along the strike as hard boulders or isolated outcrops, but is generally obscured by a mantle of soil. Fossils, mainly brachiopods, are common, but are often difficult to extract; the best locality for collecting is the hill slope above the south (smaller) branch of Cascade Stream, which is most easily reached from directly behind Cascade homestead.

This band is stratigraphically conformable with the beds above and below. Its upper surface is markedly bored, and the borings are filled with the glauconitic sandy mudstone of the overlying beds. As judged from the palaeontological evidence, however, this bored surface does not seem to represent an important time interval. Samples from this band itself did not yield foram faunas, but a sample from the top of the underlying formation (F.6413) gave a Lillburnian

– 423 –

fauna, and one from the base of the overlying formation gave a characteristic Waiauan fauna. Therefore the age of M8 is either Lillburnian or Waiauan, or intermediate between the two. This is borne out by the macrofossils, of which the following have been identified:

Echinodermata: Large species of irregular echinoids, not uncommon.

Mollusca: Lentipecten hochstetteri (Zittel).

Brachiopoda: Dr. Allan has kindly supplied the following note upon the brachiopod fauna and its significance:—

Stethothyris epsilon Allan—common and typical.

Pachymagas aff. finlayi Allan—one specimen, but not so elongate and less strongly folded.

Rhizothyris fortis Thomson—common, fairly typical.

Neothyrisnovara von Ihering—common.

Pachymagas cf. coxi Thomson—one specimen somewhat larger than the holotype.

Pachymagas n.sp.—two specimens.

“The association of the three common species, viz. Stethothyris epsilon, “Neothyrisnovara, and Rhizothyris fortis suggests correlation with the Hinnites shell bed and the Uppermost Mount Brown Limestone (E) of the Weka Pass District.”

The evidence from both the foraminifera and the brachiopods indicates that M8 may be correlated with the Hinnites shell bed and the overlying limestone of the Mount Brown Series at the type locality in the Weka Pass. By interpolation from the foraminifera, M8 can be correlated with the shell bed at the top of the Tokama Series in the two branches of the Grey River (Mason, 1941, pp. 120–1).

M9: Muddy Sandstone.

The typical rock of M9, a grey muddy sandstone, is so similar to the grey muddy sandstone of M7 that it is not possible to distinguish them in the field except where the Stethothyris epsilon band (M8) is present. The best section through M9 is along the main branch of Cascade Stream, where there are more or less continuous exposures from the top of M8 to the base of M10. A similar section is present in the south branch of Cascade Stream, but the bed of this smaller stream is so overgrown with blackberry and scrub as to make examination most unpleasant and difficult. An intermittent section is also visible along Balmoral Stream. Elsewhere beds of this formation are obscured by vegetation and soil, although they probably cover a fair area of country between Cascade Stream and Balmoral Stream and beyond until the Tertiary rocks disappear beneath the gravels of the plain.

As pointed out in the previous paragraph, the rocks of M9 are muddy sandstones, sometimes glauconitic, similar to those of M7. They also contain molluscan fossils sparingly throughout, occasionally concentrated in shell beds and concretionary bands. The base of this formation rests on a bored surface of calcareous sandstone of M8. The top of this formation, where it is exposed in Cascade Stream, shows a zone, several feet in thickness, of a leached yellow sandstone, on which rests the basal conglomerate of M10. This leached zone probably represents the weathered and possibly eroded surface of

– 424 –

M9, suggesting a disconformity which is evidenced on other grounds also. The total thickness of M9 in the main branch of Cascade Stream is about 700 ft.

Samples were taken at intervals throughout this formation where it is exposed in the main branch of Cascade Stream and submitted to Dr. H. J. Finlay, who reports that they all give a typical Waiauan fauna, characterised by the presence of the species Loxostomum truncatum (Finlay). A molluscan fauna collected 20 ft. below the top of M9 contained the following species (identified by Dr. J. Marwick):—

Limopsis n.sp. Zelandiella aff. subnodosa (Hutton)
Modiolus altijugatus Marwick
Sectipecten n.sp. Acominia sp.
Musculus cf. impactus (Herman) Baryspira subhebera Marwick
Moaricrypta wilckensi (Finlay) Aoteadrillia n.sp.aff. callimorpha (Suter)
Struthiolaria praenuntia Marwick

Dr. Marwick considers this fauna to have Waiauan and Tongaporutuan affinities. As these mollusca are associated with foraminifera, including the characteristic Waiauan species Loxostomum truncatum, M9 is here classed as Waiauan.

M10: Gravels, sands, sandy clays.

The rocks of this formation underlie the greater part of country in the downlands between Balmoral and Tekoa roads, south of Cascade Stream. These downlands rise to a height of 1626 ft. in Green Hill (Trig. T), that is, about 600 ft. above the surrounding plain. In general the surface of these downlands is completely covered with soil and vegetation, and outcrops of this formation are practically confined to the stream channels. The best section is exposed along Cascade Stream, where the contact with the underlying M9 can be seen, and the beds can be followed upwards until they are terminated against a fault which brings up agglomerates of M4 to the west. Less complete sections are exposed in the headwaters of Balmoral Stream, and isolated outcrops are found in the small gullies draining the western side of the downlands.

The actual contact between M9 and M10 is not well exposed in Cascade Stream, but sufficient can be seen to make out its general characters. The top of M9 appears to be weathered and leached to a yellow-brown sandstone to a depth of about 20 ft. The outcrops then fail for a short distance, and the next exposure shows a conglomerate of small greywacke pebbles up to an inch in diameter set in a yellow sandy matrix. This conglomerate cannot be more than 6 ft. stratigraphically above the nearest outcrop of the leached yellow-brown sandstone. The boundary between M9 and M10 is placed between these two outcrops. The conglomerate is not well exposed, but about 25 ft. higher in the sequence it changes in character rather suddenly by the incoming of large, poorly rounded pebbles of black basaltic rock, evidently derived from the agglomerates of M4. This type of conglomerate continues for some distance before the incoming of limestone pebbles derived either from lenses in M4 or the limestone of M5. As the formation is followed upstream there is an irregular alternation of layers of conglomerate, soft gritty sandstone, and sandy clay, with the coarser beds predominating. As the formation is

– 425 –

followed upwards, the amount of basalt and limestone in the conglomerate decreases, and is replaced largely by greywacke. This does not take place continuously, however, there being an alternation of bands of conglomerate consisting almost entirely of pebbles of limestone and/or basalt with bands of greywacke conglomerate becoming predominant in the upper part of the formation. In these conglomerates the pebbles are usually of moderate size, up to 6i in. in diameter, but occasional boulders of greywacke up to 18 in. across are not uncommon. The dominant materials in the conglomerates, as indicated above, are greywacke, basalt, and limestone, but occasional pebbles of igneous rocks similar to the sills intruding the local greywackes, and of Tertiary sandstone and mudstones, are not uncommon.

At the base of M10 in Cascade Stream the dip of these beds is the same as that of the underlying formations, practically vertical but when followed upstream, the dip gradually decreases till in the upper part it is about 35° west. The thickness of M10 in Cascade Stream between the base of the formation and the fault which terminates it in the upper part of the stream, calculated from the breadth of the outcrop and the observed dips, is about 1500 ft., which is therefore less than the total thickness of the formation. Unfortunately, this formation is not well enough exposed elsewhere to enable estimates of total thickness to be made, but as judged from observed dips and the area covered by this formation, its true thickness is probably considerably greater than the 1500 ft. measured in Cascade Stream.

The nature of the pebbles indicates that the conglomerates of M10 probably had a local origin. The alternation of bands consisting practically entirely of greywacke pebbles with bands of limestone and basalt pebbles suggest that the movements responsible for the deposition of these conglomerates were intermittent, repeatedly bringing fresh areas of Tertiary rocks above base level. The alternation of sandstones and sandy clays took place in still or sluggish water. No fossils have been found in them, and samples of the finer clays, examined by Mr. C. R. Russell for diatoms, were barren of organic remains. There is thus no evidence to decide whether deposition was terrestrial or marine.

The absence of fossils means that this formation must be dated by its stratigraphical position and its lithological similarity with beds of known age in adjacent areas. The stratigraphical position shows that it must be younger than M9, and that orogenic movements intervened which exposed both the undermass and the lower Tertiary rocks to subaerial erosion. The fact that it has a regional dip indicates that it is older than the latest orogenic movements in this area; which means that it is probably pre-Pleistocene. Thus the stratigraphical evidence is that M10 is probably Pliocene in age.

In the adjacent Amuri Subdivision the corresponding formation comprises the Bourne and Highfield beds (in the latest report on the Amuri Subdivision, Healy (1939, p. 4) suggests that the Bourne congolmerates are merely the basal phase of the Highfield beds). Lithologically the beds of M10 are closely comparable with the Bourne and Highfield beds, and their stratigraphical position is the same, so that they may be correlated with some confidence. The evidence in the Amuri Subdivision is for a lower Pliocene age, possibly Waitotaran, for the Bourne and Highfield beds.

– 426 –

Throughout North Canterbury there is evidence of an important stratigraphical break at about the base of the Pliocene. A long period of more or less continuous sedimentation uninterruped by orogenic movements throughout Lower and Middle Tertiary times was apparently brought to an end by earth movements in the late Miocene or early Pliocene, and Pliocene beds rest uncomformably or discomformably upon the older rocks. The Pliocene beds are also characterised by the presence—in many cases the predominance—of coarse conglomerates containing pebbles of the underlying Tertiary and pre-Tertiary rocks. Where fossiliferous, these beds are apparently of Waitotaran or Nukumaruan age. They have received different names in different localities: The Greta or Motunau beds of the Middle Waipara and Weka Pass district (Thomson, 1920), the Kowai (or Kowhai) Series to the south of Mount Grey (Speight, 1919; Mason. 1941), the Hundalee conglomerate of the Conway and Okarahia valleys (Jobberns, 1928). The beds of M10 are the equivalent of these in the Mandamus-Pahau area, and the geological conditions giving rise to them were apparently prevalent throughout North Canterbury at that time.

M11: Greywacke gravels and clays.

This formation is described on the basis of exposures along the north bank of Pahau River between Ramatama Stream and Black Birch Stream, just beyond the area mapped in detail for this report, but is discussed here because it probably underlies the plains between the foot of the hills and the Hurunui River, and because it has an important bearing on the history of the Hurunui-Waiau depression. It consists of well-stratified flat-lying clays, sandy clays, and gravels, in bands generally from 4 ft. to 8 ft. thick. The gravels appear to be made up entirely of greywacke pebbles, no fragments of Tertiary rocks having been seen. These gravels are well sorted and are considerably finer than the terrace gravels which discomformably overlie them, and the present river gravels. A feature of the clay and gravel in places is the presence of numerous granules of bright blue vivianite, up to half an inch in diameter. These beds have not been seen in other places, but are probably widespread beneath the veneer of coarse greywacke gravel which forms the present surface of the Hurunui-Waiau basin; where they might be expected to crop out, along the high banks of the Hurunui River, talus from the coarse terrace gravels obscures the sections.

No section has been seen which shows their relationship to the beds of M10, but they are presumably unconformable and considerably younger, not having been involved in the movements which have given the beds of M10 a regional dip. The beds of M11 are disconformably overlain by coarse greywacke gravels (M12), and the distinctive lithology of M11 suggests that this formation was laid down under totally different conditions. Whereas M12 has clearly been deposited by the present rivers when they were flowing at a higher level, M11 has probably been deposited in still or sluggish flowing water, as judged from the presence of thick bands of fine clay.

It was first thought that these beds might represent lacustrine deposits, and, with this in mind, samples were submitted to Mr. C. R. Russell, of Christchurch, who kindly examined them for the possible presence of diatoms. Mr. Russell reports as follows:—

– 427 –

“In both samples of clay no trace of vegetable or other organic matter was found, and both were free from diatoms when boiled with H2SO4 and KOH… I am not prepared to make a definite statement on the subject, but from previous experience I am inclined to think that the deposits have been laid down by slowly flowing water rather than in a lake. From almost every lake water I have examined the presence of diatomacea has been marked. but in neither sample could I find a single specimen.”

From this work it seems unlikely that these beds were laid down in a lake. but if they were deposited by the ancestors of the present rivers, then these rivers must have been much more gently flowing than they are at the present day. The presence of this set of flatlying beds beneath the floor of the Hurunui-Waiau basin implies a very different topography of the surrounding country at the time they were laid down.

As mentioned above, no internal evidence of their age is present in these beds. They are tentatively classed as Pleistocene, and it is believed that they were deposited at a time when the topography was less diversified than now, i.e., prior to the latest movements along the active fault lines which divide up this region into individual blocks. The Lydon beds of the adjacent part of Amuri Subdivision may be contemporaneous, but they are tilted, at least in places, although less so than the underlying Bourne and Highfield beds.

M12: Terrace gravels.

Coarse greywacke gravels cover the present surface of the Hurunui-Waiau basin, and extend as discontinuous terrace remnants along the rivers draining into the basin, the Pahau, Dove, Glencoe, Mandamus, and the Hurunui itself. Where Hurunui River flows out into the basin just below the junction with the Mandamus, the present river is entrenched about 100 ft. below this surface, the amount of entrenchment decreasing gradually downstream. Remnants of other and higher terraces are preserved at several places around the margin of the basin; from the Hurunui Peak ridge a terrace remnant about 750 ft. above the present river level shows up prominently on the south bank of the Hurunui some distance upstream from the Mandamus junction, and there may be another about 100 ft. higher. The upper surface of the downlands between Cascade Stream and Green Hill is littered with rounded and subangular boulders of greywacke and Tertiary limestone up to 18 in. in diameter; no exposures showing the relationship of these boulders to the gravels and sandstones of M10 which make up the greater part of these downlands were seen, but the evidence is in favour of these boulders belonging to a veneer of flat-lying gravels which probably formed the upper surface of these downs when the base level of erosion was higher than it is now. There is evidence from flat-lying gravels on its flanks that Hurunui Mound was once buried by gravels and has been re-exposed by subsequent erosion. A study of terrace remnants around the Hurunui-Waiau basin would probably support the view that it was filled to a considerably greater depth in Pleistocene times, and that the present surface is the result of re-excavation.


A sample of coal from the 2 ft. seam in Coal Creek was submitted

– 428 –

to the Coal Survey and analysed in their laboratories, with the following results:—

Laboratory Number C.S. 1031.
Proximate analysis (on air-dried coal).
Moisture 8.7
Volatile matter 21.2
Fixed carbon 21.8
Ash 48.3
Sulphur 1.2

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Calorific value, 4,820 B.Th.U./lb. Non-coking.
Ultimate Analysis (calculated to dry, ash-free coal).
Carbon 86.7
Hydrogen 5.1
Nitrogen 0.8
Sulphur 2.7
Oxygen 24.7

Remarks: The sample has a high ash content which clouds the interpretation of the above analysis. We give below the proximate analysis calculated to an ash-free basis.

Proximate Analysis (calculated to air-dry ash-free basis).

Moisture 16.8
Volatile matter 41.0
Fixed carbon 42.2
Sulphur 2.3
Calorific value, 9,320 B.Th.U./lb.

I am indebted to the Secretary, Coal Survey Committee, Wellington, for the analyses of this coal and for the remarks accompanying the analyses.


Cotton, C. A., 1941. Landscape as Developed by the Processes of Normal Erosion. Cambridge University Press.

Fyfe, H. E., 1931. Amuri Subdivision. N.Z. Geol. Surv., 25th Ann. Rept., pp. 5–6.

Gage. M., and Wellman, H. W., 1944. The Geology of Koiterangi Hill, Westland. Trans. Roy. Soc. N.Z., vol. 73, pp. 351–64.

Haast, J. von, 1871. On the Geology of the Amuri District, in the Provinces of Nelson and Marlborough. Rept. Geol. Expl. during 1870–1, pp. 25–46.

Healy, J., 1939. Amuri Subdivision. N.Z. Geol. Surv., 33rd Ann. Rept., pp. 2–4.

Hutton, F. W., 1877. Report on the Geology of the North-east Portion of the South Island, from Cook Straits to the Rakaia. Rept. Geol. Expl. during 1873–4, pp. 27–58.

—— 1889. The Eruptive Rocks of New Zealand. Jour. Roy. Soc. N.S.W., vol. 23, pp. 102–156.

Jobberns, G., 1928. The Raised Beaches of the North-east Coast of the South Island. Trans. N.Z. Inst., vol. 59, pp. 508–570.

McKay, A., 1890. On the Geology of Marlborough and the Amuri District of Nelson. Rept. Geol. Expl. during 1888–89, pp. 85–185.

Mason, B. H., 1941. The Geology of the Mount Grey District, North Canterbury. Trans. Roy. Soc. N.Z., vol. 71, pp. 103–127.

—— 1946. Apophyllite, Analcite and Natrolite from the Pahau River, North Canterbury, New Zealand. N.Z. Journ. Sci. and Tech., vol. 28, pp. 53–54.

Morgan, P. G., 1919. The Limestone and Phosphate Resources of New Zealand. Part I—Limestone. N.Z. Geol. Surv. Bull. 22.

Speight, R., 1918. Structural and Glacial Features of the Hurunui Valley. Trans. Roy. Soc. & N.Z., vol. 50, pp. 93–105.

—— 1919. The Older Gravels of North Canterbury. Trans. N.Z. Inst., vol. 51, pp. 269–281.

Thomson, J. A., 1919. The Geology of the Middle Clarence and Ure Valleys, East Marlborough, New Zealand. Trans. N.Z. Inst., vol. 51, pp. 289–349.

—— 1920. The Notocene Geology of the Middle Waipara and Weka Pass District, North Canterbury, New Zealand. Trans. N.Z. Inst., vol. 52, pp. 322–415.