Art. XXXVIII.—The Younger Rock-series of New Zealand.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 43, 1910, Page 378

Art. XXXVIII.—The Younger Rock-series of New Zealand. By P. Marshall, D.Sc., F.G.S., Professor of Geology, Otago University; R. Speight, M.Sc., F.G.S., Lecturer on Geology, Canterbury College; and C. A. Cotton, M.Sc., Lecturer on Geology, Victoria College. [Read before the Otago Institute, 1st November, 1910.] Plate VIII. [Note.—This paper has been mainly written by the first author. The observations upon which it is based were made conjointly by the three authors in the typical districts of North Canterbury. Other districts referred to have been examined by only two or by one of the authors, but in each case the observations made have been referred in detail to the others, and the authors are in complete agreement in regard to all the critical points set forth in the paper. The critical account of the stratigraphical relations of the Waitemata series has been kindly furnished by Mr. E. de C. Clarke, of the University College, Auckland, and the authors are specially indebted to him for the description of this important locality.] Table OF Contents. I. Introduction. II. Classifications employed by different geologists. (a.) Hochstetter (1864). (b.) Hutton (1885). (c.) Haast (1879). (d.) Hector (1886). (e.) Park (1910). (f.) Comparisons of these opinions. III. Classification proposed by the authors. (a.) Discussion of Waipara Gorge and Weka Pass sections. (b.) Amuri Bluff. (c.) Other “Cretaceous” localities. (d.) Oamaru district. (e.) West coast of the North Island. (f.) North Auckland. (g.) Waitemata. IV. Deposition of rocks of the series. (a.) General nature of palaeontological evidence. (b.) Explanation of apparent rapid change of life-forms. (1.) Isolation of New Zealand coast-line. (2.) An archaic fauna. (3.) Slow rate of deposition. (i.) Conglomerates. (ii.) Coals. (iii.) Greensands. (iv.) Limestones and later beds. V. Correlation of members of the series. VI. Correlation with European horizons. VII. Summary and conclusions.

I. Introduction. The stratigraphical relations of the series of younger rocks of New Zealand has presented a problem which each geological observer who has worked Fig. 1. at them has solved in a different way from all of the others. This obviously implies that elements of difficulty and confusion are involved in a treatment of the question.

The distribution of the series is fairly general in the coastal districts of both Islands and in some portions of the interior parts, but in most instances the outcrops are separated from one another (see fig. 1). In all but a very few of the areas where the series is exposed the stratification is simple. In many cases the strata lie flat; but where they border on an older series of rocks they are generally inclined and folded in some simple manner, though in some places it causes them to lie vertically. It is the belief of the authors that the mere stratigraphy would have offered no difficulties and have caused no confusion if there had not been other matters requiring consideration in connection with the whole question. One of these other matters is the correlation of the strata in the different and sometimes distant areas where they outcrop, but the most important is to be found in the palaeontological succession. The lowest strata of this younger series contain fossils that are generally allowed to be of Cretaceous age, the upper of Miocene or Pliocene age. In general the rocks are marginal. In the North Island they unite in the extreme north portions of the older rocks that crop out here and there, and apparently constitute the greater part of the country as far south as the Upper Waikato. They are extensively developed on the eastern side of the great range from Cape Runaway to Cape Turakirae, and in places near the East Cape form the highest part of the country. In the south part of this coast they are replaced by older series, but in the central part they extend almost from side to side of the Island except for a narrow ridge of old rocks. On the western side they form the whole country between a line through Pirongia, Taumarunui, Waiouru, Waikanae and the coast. In the South Island the formation is more restricted and local, and is nearly always marginal. It extends up many river-valleys—Aorere, Clarence, Waitaki, for example. It is widely extended in North Canterbury, in North Otago, and again in Southland; while it occupies interior basins in the Te Anau and Manapouri districts, as well as at Trelissick. In Westland the development is again larger in the northern part, and especially large areas are found in the Wangapeka, Maruia, and other valleys. Occasionally the rocks are well folded into much older series, as at Nelson, where they occur structurally involved in strata of Trias-jura age. Recognition of this has caused various observers to seek for unconformities in the strata, for different members of the series contain extremely different faunas. Unconformities have been described by every observer and they have been placed in different positions in the series by every observer, and in almost every case the breaks recognized by one have not been admitted by others. In all exposures there is, however, a well-marked lithological series, so far as the observations of the present authors have gone, and this series will be stated in some detail when treating of the different areas. So similar is the succession in the various localities that it is possible without any great error to state a generalized rock-succession that will apply with considerable exactness to the many occurrences. This is as follows :— 7. Loose sands with shells. 6. Mount Brown beds; sands often calcareous. 5 Grey marls. 4. Greensands. 3. Limestone. 2. Greensand. 1. Sands, conglomerates with coal.

This series, of course, varies considerably in different localities. Some members may be quite omitted, while others are relatively thick. In general, it is thought that the following statement represents, without attention to purely local features, the variations that are found in different localities. In North Canterbury practically the whole series is present. In Otago it is unusual to find any beds higher in the series than the limestone. In Westland and at Cape Farewell and Shag Point the basal conglomerate is of great thickness—7,000 ft. in the first case—and it contains important seams of coal. In the North Island, but more particularly in its southern portions, there is an immense development of the grey marls, which in the Wanganui country are perhaps 2,000 ft. thick. The greensands are highly variable in thickness, and are often absent; they have a particularly large development in North Canterbury. The limestone is fairly general, and the most constant member of the series, but landwards always becomes somewhat sandy, and even passes into pure quartz sand where it is marginal to the old land. As will be more fully mentioned later, the fossils contained in the rocks at the base of the series are wholly different from those at the top; in fact, the lowest rocks, in some localities at least, contain Mesozoic forms, and have been referred by all observers to the Cretaceous period. On the other hand, there is a high percentage of Recent species in the highest rocks, and they are classed as Upper Miocene or Pliocene on all hands. It appears to have been the general belief of those geologists who have examined these rocks that a rock-series showing evidence of such great difference in age in its lowest and highest members could not have been deposited in a uniform stratigraphical sequence as a single formation. It has therefore been thought necessary by all authorities to refer different portions of the rock-sequence to different geological systems the members of which are supposed to be separated by unconformities. Unfortunately, while there has been an agreement in principle, there has been a marked divergence in practice, and no two authorities have placed the unconformities in the same places. This is the more remarkable as in nearly all localities where the rocks occur there is a fine exposure of natural sections in which the rock-series are displayed free from all obscurity, and in most cases with nothing but folding of a most simple description. The great variety of opinion in regard to this matter is shown in the following summary of the classifications adopted by those geologists who have published comprehensive works dealing with the geology of the whole or of large portions of the Dominion. II. Classifications Employed. (a.) Hochstetter (“Reise der ‘Novara’: Geologie,” vol. 1, p. 39. Wien, 1864). Kainozoic Series.—(1.) Older Tertiary System: (a) Brown coals, Lower—Coalfields of Drury, Waikato, North Auckland, Nelson, Otago; (b) Marine series, Upper — Waitemata, Kawhia, and Aotea, Motupipi (Nelson), Aorere (Nelson), gold-bearing conglomerate (Aorere), white and yellow calcareous sands and greensands with limestone, Oamaru, Green Island, Shag Valley. (2.) Younger Tertiary: Cape Rodney, Hawke's Bay, Wanganui River, Nelson cliffs, blue clay of Awatere, Waitaki, Moeraki.

(b.) Hutton (Q.J.G.S., 1885). Wanganui System (Pliocene). — Newer Pliocene: Locality names — Kereru, Ormond, Petane, and Putiki series. Older Pliocene: Older glacial deposits—Lignites of Otago, Manukau, &c. Pareora System (Miocene).—Localities: Awatere (Marlborough), Kanieri (Westland), Tawhiti (Poverty Bay), Ahuriri (Hawke's Bay), Waitemata (Auckland), brown coal of Pomahaka (Southland). Oamaru System (Oligocene).—Localities: Mount Brown (North Canterbury), Aotea (south-west Auckland), Ototara (Oamaru), Turanganui (Poverty Bay), coals of Waikato, Kaitangata. Waipara System (Upper Cretaceous).—Localities: Amuri (Marlborough), Awanui (Poverty Bay), Matakea (North Otago), coals of Greymouth, Pakawau (North Nelson). (c.) Haast (“Geology of Canterbury and Westland,” p. 251. Christchurch, 1879). Waipara.—Cretaceo-tertiary. Oamaru.—Upper Eocene or Lower Miocene. Pareora.—Upper Miocene or Lower Pliocene. (d.) Hector (Handbook of N.Z.: Geology. Wellington, 1886: Government Printer). Lower Greensand.—(a) Buller series; (b) Porphyry breccia series; (c) Amuri series. Cretaceo-tertiary.—(a) Grey Marls; (b) Ototara series; (c) Mawhera series; (d) Chalk series; (e) Waireka series; (f) Coal series; (g) Black Grit series; (h) Propylite breccia series; (i) Great Conglomerate series. Upper Eocene.—(a) Mount Brown series; (b) Oamaru series; (c) Waitaki series. Lower Miocene.—(a) Awatere series; (b) Pareora series; (c) Awamoa series; (d) Lignite. Upper Miocene.—(a) Te Aute series; (b) Taueru series. Pliocene.—(a) Dispersed gravels; (b) Napier series; (c) Lignite series; (d) Kereru series. (e.) Park (Geology of N.Z., 1910, p. 25). Amuri System.—Waipara series (Cretaceous): (a) Weka Pass stone and Grey Marls; (b) Amuri limestone; (c) Waipara greensands; (d) Saurian beds; (e) Puke-iwi-tahi conglomerates and sandstone with coal. Karamea System.—Waimangaroa series (Eocene): (a) Brunner coal; (b) Paparoa coal. Oamaru Series (Miocene).—(a) Ototara stone; (b) Pareora beds; (c) Marawhenua beds; (d) Waihao beds; (e) Kaikorai coal-measures. Wanganui System.—(a) Petane series (newer Pliocene); (b) Waitotara series, (c) Awatere series (older Pliocene). (f.) Comparisons of these Opinions. The nature of this divergence of opinion appears more striking when shown diagrammatically (fig. 2), for then it is seen that a total of five unconformities have been described by different observers, and as many as three in the North Canterbury series by a single authority; while if the series is extended to the highest member in the North Island the number of

Fig. 2.

unconformities inserted by Hector is five. Some of these unconformities that are inserted represent great lapses of time. Thus the plane between the Waipara and Karamea systems of Park represents a period of erosion extending throughout the Eocene and Oligocene periods. With Hutton the Eocene is a period of erosion and elevation, but Hector's time-lapses are of brief length, and therefore of relatively little importance. It is, however, obvious that an unconformity between Cretaceous and Miocene rocks in Park's classification should be of such a marked nature as to be a conspicuous feature in all the clear sections in the districts in which, the rocks are developed. Extremely different, too, are the periods to which the main portion of the series, including at least the important coal-measures of the east coast, have been referred. Hector places these in the Cretaceo-tertiary, Hutton in the Oligocene, and Park in the Miocene. III. Classification Proposed by the Authors. (a.) Discussion of Waipara Gorge and Weka Pass Sections. (See fig. 3 and Plate VIII.) The writers of this paper have long been conversant with these opinions, and have seen grave reason to reject them in those portions of the Dominion with which they are familiar. The doubts thus aroused caused them to visit in company the more important of the sections upon the inspection of which the classifications of the authors named were mainly based. Of these localities, it is admitted by every one that the gorge of the Waipara River gives the most complete and satisfactory section, while that of the Weka Pass is one on which Hutton and Park have mainly relied for the demonstration of the unconformities that they have described. The third locality, the Amuri Bluff, is of importance because of the abundance of fossils contained in the lower rock-series. The following quotations appear to us to represent the gist of previously formed opinions on the rocks shown in these sections :— Hutton (N.Z.G.S. Rep., 1873–74, p. 44), at Waipara: “The Weka Pass stone is seen to rest on a water-worn surface of the Amuri limestone.” Also, p. 48: “At Waipara the Pareora formation rests on the Trelissick group without any appearance of unconformity.” Haast (Geology Rep., 1871, p. 8): “The beds [to top of Weka Pass stone] follow in unbroken sequence—that is, they belong to the same series, and mark a well-defined period in the past history of this portion of the globe.” Also, p. 16: “After the deposition of the Weka Pass beds the whole series seems to have risen to such an extent as to come under the full destructive influence of tides and currents, if not even under subaerial influence. However, the newer strata must in some instances have been deposited very soon, as the uppermost beds of the Weka Pass have not only been preserved entirely, so that no sign of denudation is visible, but the Cucullaea beds overlie them in most instances so conformably that it is impossible to detect the least difference in strike and dip.” McKay (Geol. Rep., 1874—76, p. 39): “At all points where the Weka Pass beds are overlaid by any higher beds the marly grey or green beds are the next in succession, and in many places they pass insensibly from one to the other.” Geol. Rep., 1890—1, p. 102: “Although the change from the Amuri limestone to the Weka Pass stone is somewhat sudden and sufficiently clearly marked, there is not in the cross-section exposed

Panorama of Weka Pass from Crest of Mount Donald Range

in the gorge, or along the line of strike for a mile to the north-east or south-west, or elsewhere in the district any sign of uncomformity between these beds…. At the lower end of the limestone gorge the Weka Pass stone is overlaid by the grey marls. The section is clear, and shows perfect conformity between the two…. At the junction between the lower part of the Mount Brown beds and the upper part of the grey marls a stratigraphical uncomformity is here evident enough in the section displayed.” Park (Geol. Rep., 1883, p. 33): “The sequence of the beds just enumerated is the most complete to be found in any part of New Zealand. The stratigraphy is plain and simple, being free from obscurities, and offering few points of possible disagreement.” At p. 35: “At the Waipara, on the other hand, the geological record is complete, the beds following one another in one conformable sequence.” At p. 22: “The Weka Pass stone passes gradually into the grey marls without any sudden change in the character of the deposits such as is seen between the Weka Pass stone and the Amuri limestone.” At p. 28: “The strata at the Waipara, where the complete sequence is exposed, are quite undisturbed, following one another uniformly throughout all parts of the district…. I am strongly of opinion that a complete sequence of beds exists from the base of the Cretaceo-tertiary to the close of the Pareora formation.” Very similar opinions have been expressed by the geologists in regard to the series exposed in the Weka Pass. In other words, Hutton is constantly bringing evidence of unconformity between the Amuri and Weka Pass stone, and, in 1886, between the grey marls and Mount Brown beds. Haast still supports a subdued unconformity between the Weka Pass stone and the grey marls, while McKay and Hector (1890–91, p. 98) place an unconformity at the base of the Mount Brown beds only. Park, however, in 1904 (Trans. N.Z. Inst., vol. 37, p. 542) absolutely changes his position, and places an unconformity between the Weka Pass stone (Cretaceous) and the grey marls (Miocene), but states that this break is nowhere to be seen in section, although shown by the mapping. He also places an unconformity between the Mount Brown and the overlying beds, which he now calls Motunau (Pareora, Hutton). It is certainly remarkable that such a variety of opinion should have been given in connection with a section that is so remarkably clear and free from any obscurity. As the authors are engaged in teaching this subject in the colleges of the New Zealand University, and were much perplexed by these statements and the still more conflicting results when correlations were attempted with the series in other parts of New Zealand, it was agreed to visit the Waipara and Weka Pass in company and try to arrive at a conclusion that might express the actual facts and yet perhaps be unanimous. The result, based on the stratigraphy alone, has been wholly satisfactory, and there is an unreserved agreement. For the sake of clearness, the unconformities described by different observers will be taken in order from below upwards. Relation of Amuri Limestone to Weka Pass Stone. 1. Hutton's Unconformity.—The Amuri limestone is white and rubbly; the Weka Pass stone has much glauconite, and is compact. Hutton describes the proof of unconformity as consisting of pebbles of the Amuri limestone imbedded in the base of the Weka Pass stone. McKay and Park

Fig. 3.—Panorama of Weka Pass from Crest of Mount Donald Range. refer to these as phosphatic nodules. Our opinion is that the change from pure (Amuri) to glauconitic (Weka Pass) limestone does not take place throughout the mass of the rock, but interlaminations of glauconitic matter arise and separate pieces of limestone. As the conditions that control the depositions become more changed, the interlaminations of glauconitic matter become larger, and the pure limestone is reduced to nodules which appear like rolled pebbles. A similar lamination is distinct at Milburn and Clarendon. In addition to this, it is evident that the elevation of a foraminiferal limestone and its subsequent erosion followed by depression and deposition of glauconitic limestone in perfect accordance without any intervening strata would necessitate extraordinary conditions; in fact, it is almost impossible to suggest a series of conditions that would satisfactorily account for such a sequence. In this matter we agree with Hector, McKay, Haast, and Park; but differ from Hutton. 2. Relations of Weka Pass Stone to Grey Marls.—Very slight unconformity of Haast, but break between Cretaceous and Miocene of Park. No junction is described by either of these observers. We found one below the limestone gorge of the Waipara, as described by McKay, and found a perfect conformity. This is accentuated at the Amuri Bluff, where the Weka Pass stone is absent, and in an absolutely clear sea-cliff the Amuri limestone graduates into the grey marls to such an extent that it is impossible to say where one begins and the other ends. There is another clear section of a similar nature at the mouth of the Jed. The idea of this unconformity was based on the mapping of the district (Trans. N.Z. Inst., vol. 37, 1904, pl. 48; but compare with Geol. Rep., 1888, sketch-map opp. p. 30). We believe that the former map is totally misleading, though the latter almost represents the actual outcrops. The statement that “the Weka Pass and Amuri limestone are thrown into folds in which the Tertiaries take no part whatever” we believe to be erroneous. The actual structure is as shown in the accompanying map and diagrams (figs. 4, 5; also sections 2, 3, fig. 6), an anticline pitching sharply to the east and bringing the limestone outcrop on the east of the anticline to the level of the railway. To the north this is succeeded by a syncline pitching east and by another sharp anticline, but east of the railway-line the pitching anticline and syncline radiate from a point near the viaduct, where they die away; so that further east the north limb of the

Fig. 4.

Fig. 5.

syncline or the south limb of the anticline beyond alone remain. Not only does this simple structure explain the somewhat irregular nature of the outcrops, but also the sudden bend of the Weka Pass Creek into the hard limestone, for this course we believe it followed in the upper softer beds above before it corroded its bed as low as the surface of the harder rock. In addition to this observation, there is further evidence in the parallel nature of the outcrops of the Mount Brown and Weka Pass stone between the Weka Pass and the Waipara Gorge. Standing on the summit of Mount Donald and looking southward, this is a particularly marked character. The two beds are hard, and dip approximately S.E. at an angle of from 15° to 20°. Sharp escarpments have naturally been formed, and are the dominating features of the country. They can be seen and followed with the utmost distinctness over miles of country without any bend in the strike of either being noticed. This statement is in perfect accord with the observations of McKay quoted earlier. 3. Relation of Grey Marl to Mount Brown Beds. — Hector, McKay, and Hutton agree on an unconformity here; Haast and Park have failed to observe one. Hector and McKay rely mainly on the junction of these beds exposed in the Waipara Gorge. This junction is now well exposed, owing to the formation of a road-cutting. We examined this with great care on the 6th February, 1910, and unreservedly agreed that in this clear section there is no unconformity whatever, but a gradual change from the one formation to the other. Hutton relied mainly on a railway-cutting in the Weka Pass a little south of the 44th mile-post. Here there is a clear break in the stratification. However, the beds on the two sides are of the same nature (Mount Brown beds), and the plane of union is most regular, and inclined 45′. This we believe to be a small fault, but the throw cannot be measured. Park and Hector do not mention this section. 4. The Junction between the Mount Brown and Greta, Motunau, or Awatere Beds of Different Classifications.—Park alone has described an unconformity here. The section on which he relies is near the 43rd milestone. The coralline sandstone (of his Mount Brown or Miocene series) that he describes as having an isolated outcrop at the south-east we found was continued without break above the sands that he refers to the Motunau or Pliocene series. At the north-west end there is no indication of uncomformity on the south-west side of the cutting to which he refers. The section here, however, is so much obscured by vegetation of many years' growth that details were not distinct. On the north-east side of the section the details are perfectly distinct, and there is a complete conformity in the series. In regard to this matter we are in agreement with all the other geologists who have examined this country. Careful and detailed examination of all these described unconformities, made with descriptions and diagrams of previous observers in our hands, has convinced us that there is no unconformity in the Waipara and Weka Pass sections. (b.) The Amuri Bluff. Here McKay has described a complete conformity, but Hutton states that on the south side of the Bluff the Amuri limestone is unconformable to the Grey Marls which here overlie it to the exclusion of the Weka Pass stone. This unconformity is based on the change in dip of the two series. Our observations showed that the Grey Marls dip more and more steeply as they are followed to the west, and that the Amuri limestone does not

dip more steeply than the lower strata of the grey marl which are in contact with it. At the entrance to the Okariki Stream there is, however, a strong suggestion of an unconformity which apparently was not noticed by Hutton. Here the Amuri limestone is locally puckered in an extraordinary manner, while the Grey Marls are stratified simply. A phenomenon of a precisely similar nature is to be seen at Kaikoura, and, judging by Haast and Hutton's descriptions, in the valley of the Conway as well. At Kaikoura Hutton describes this as an unconformity.* Hutton: Q.J.G.S., 1885, p. 273. McKay, whose opinion is indorsed by Hector, is strongly in favour of conformity.† Geol. Rep., 1886–87, p. 74; also Hector, loc. cit., p. x. (See fig. 6, section 1.) Boehm, who has also visited this locality, refuses to give an opinion on this matter: “Ich gestehe, dass ich nicht daruber in's Klare gekommen bin, ob der grey marl concordant oder discordant uüber dem Amuri limestone liegt.” ‡ Boehm: Zeit. d. Deutsch. Geolg. Gesellschaft Jahrg., 1900, p. 173. Our opinion is that the two series are perfectly conformable, but during the folding movements to which they have been subject the plasticity of the grey marl allowed this rock to yield to the force without its whole mass becoming deformed. The hard and unyielding nature of the Amuri limestone, however, prevented anything of this nature taking place, and finally adjustment by folding was necessary. It is only locally in the most restricted sense that the Amuri limestone has been affected in this way, as is clearly seen at the Amuri Bluff, where at the north head this rock and the grey marls are displayed in perfect clearness, and are absolutely conformable. The section at the Bluff is also of extreme importance, because there the Black Grit is well developed, and this is the lowest bed of Hector's Cretaceo-tertiary. There is here certainly no unconformity separating this bed from the rocks accepted as Cretaceous by all the authors cited; and this is admitted by Hector, who states that “the further work of the Survey tended to show that the Lower Greensand formation (Cretaceous) was almost universally followed conformably by the Cretaceo-tertiary series, and that, too, in a manner which but little supported the idea of any unconformity separating the lower beds and higher beds of the two formations.” § Hector: Rep. Geol. Surv., 1890–91, p. li. In other respects the section at the Amuri Bluff calls for no further mention at the present time. We only wish to emphasize our opinion that here, as at the Waipara and the Weka Pass, there is a perfect conformity from the youngest to the oldest of those members of this younger series of rocks which are present. We have also been unable to find any unconformity between any members of the series of younger rocks at the mouth of the Jed River. (c.) Other “Cretaceous” Localities. There are many other localities where Cretaceous rocks have been described by one or more authors, and the existence of unconformities between them and the Cainozoic rocks has been asserted. These localities we have not visited in company, and but brief mention of them will be made here, as they have not been taken as typical. Castle Hill (Trelissick) Basin.—This has been described by Hector, Hutton, and McKay, and unconformities have been noted by all three.

Fig. 6.

In some cases these are founded on an incomplete examination of the district. The sequence is slightly different from that which occurs in other parts of the country. The lower part (the Waipara system of Haast) corresponds almost exactly with that at the Waipara Gorge, but the upper limestone—the equivalent of the Weka Pass stone of Hutton and the Mount Brown limestone of the Survey—is said by both Hutton and McKay to rest on the lower beds unconformably. The interpretation of the sections is, however, rendered very difficult owing to the disturbance caused by the volcanoes which broke out at the period when the limestones were being laid down, so that this unconformity may be deceptive. In any case, in those parts of the basin where the sequence is little disturbed no undoubted unconformity can be seen. Another unconformity above the limestone has been urged by the same two observers, but there seems to be no necessity for this as well. It cannot be said definitely, however, that none exists, as the beds are much disturbed by faulting, folding, and by volcanic action, and the district can hardly be regarded as a critical one for the elucidation of the Tertiary sequence. It should perhaps be noted that the section given by Hutton of the Hog's Back, which shows an undoubted unconformity between the Pareora series and the Amuri limestone, is not correct, and is apparently based on unreliable information. Curiosity Shop Beds.—This well-known occurrence of the Tertiary series has been thoroughly dealt with by Hector and Hutton. The former, basing his conclusions on the observations and report of McKay, places an unconformity at the base of the equivalents of the Mount Brown series and over a limestone correlated with the Ototara limestone of the Survey. Hutton, however, maintained that the series was conformable from top to bottom, and after a careful examination of the locality we can see no reason whatsoever for the unconformity and agree with Hutton in his interpretation of the section. Motunau Creek.—This locality is an important one, as it gives a complete sequence from the base of the series, with coal-beds and saurian remains, up to marls which have been classed as Upper Miocene by McKay and Lower Pliocene by Park. A careful examination of the sections so clearly exposed in the Motunau Creek reveals no unconformity, the dip and strike of the beds being constant from top to bottom, or, if any undetected variations do occur, they can only be very slight, and cannot affect the general conclusion that the sequence is conformable. The only appearance of an unconformity is suggested by the thinning-out of the Amuri limestone (the Weka Pass stone is absent in the section exposed in the creek) on going north, where it is apparently replaced by a calcareous sandstone, which is a natural transition on approaching a shore-line. West Coast of South Island.—The coal-series of the west coast of the South Island was tentatively classed in the Cretaceous by Hector, and definitely so by Hutton. There is practically no evidence of the age of these beds, for that given by Haast has never been confirmed.* Haast: Report of a Topograph. and Geolog. Explor. of Western Nelson, p. 106. Nelson, 1861. It appears that, despite the many excellent sections that are exposed, no definite evidence has yet been found of an unconformity. Lately Morgan has laid emphasis on the occurrence of pebbles of coal in a grit.† Morgan: Third Annual Rep. Geol. Surv., 1909, pp. 12, 13. This he regards as evidence of the elevation and erosion of the coal-series before the upper

rocks were deposited. He has, however, up to the present been unable to find any stratigraphical break, and the occurrence of the coal pebbles may, in the meantime at least, not unreasonably be ascribed to contemporaneous erosion, or even in many cases to the inclusion of wood-fragments in the sediment. The coal-measures of Pakawau have been placed by Hutton in the Cretaceous. This is apparently based on the occurrence of fossil plants and on the nature of the coal. One of us has carefully examined the ground, and has formed the opinion that in that locality no unconformity separates the coal-measures from the overlying rocks of admitted Cainozoic age. This is in accord with the opinion of Park, who in 1890 classed no beds but the Cretaceo-tertiary as lying between the glacial matter and Palaeozoic strata.* Park: Rep. Geol. Surv., 1890, p. 229. Cox, in 1883, had described a Lower Greensand formation as lying below the Cretaceo-tertiary series, but makes no mention of a break.† Cox: Rep. Geol. Surv., 1883, p. 71. Poverty Bay.—In the Poverty Bay district a complete series of Cretaceous and Tertiary rocks has been described by McKay.‡ McKay: Rep. Geol. Surv., 1886, p. 192, map; also Geol. Explor., 1900–1, p. 23. This district has, however, lately been worked in detail by Adams,§ J. H. Adams: N.Z. Geol. Surv. Bull. No. 9 (n.s.), p. 12. who classed all the rocks of Upper Miocene age, stating that no marked unconformity was observed between any of its members. One of us has examined the sections in this area in great detail, and is positive that no unconformity was shown in the many miles of continuous sections displayed. Shag Point.—Here Hutton,‖ Hutton: Geol. of Otago, 1875, p. 46. Haast,¶ Haast: Geol. Rep. 1873–74, p. 24. and Park ** Park: Geol. of New Zealand, 1910, p. 116. have described an unconformity, but in different localities. McKay,†† McKay: Geol. Rep., 1886, p. 22. in 1886, showed clearly that Hutton's and Haast's break was really due to a fault, as stated previously by Cox. The unconformity referred to by Park is a matter of inference, and is placed in the middle of the Shag Estuary, where no rocks crop out at the surface. One of us visited this district, but was not able to go over the ground in any detail. No evidence of an unconformity was found as a result of the examination made on this occasion. The rocks are here somewhat more folded than usual, and it is natural to hesitate to allow the presence of an unconformity until the effects of faulting and folding have been fully set out. This was done by McKay, and, we believe, with considerable success in showing that such structural features satisfactorily accounted for the position of the various outcrops. (d.) Oamaru District. The variety of opinion in regard to the relation of the rock-series in the very clear sections of North Canterbury becomes still more confusing when attempts are made to correlate with the Canterbury series the younger rock-outcrops occurring elsewhere (see fig. 7). This is particularly true of Oamaru, where there is a clear series with quartz gravels at the base and marly shell beds at the top. The different members of the series are shown in the accompanying diagrams. In this series Hector inserted two

unconformities, above and below the Hutchinson Quarry beds, which rest on the well-known Oamaru stone. The Hutchinson Quarry beds are green-sands, with many brachiopods and other fossils. They are generally taken as equivalent to the Mount Brown beds. The Awamoa beds are more marly, and undoubtedly rest on the Hutchinson Quarry beds. They contain an abundance of forms of littoral Mollusca. The series is particularly well exposed at the rifle-butts south Fig. 7. of Oamaru, and there the conformable nature of the series is distinct, as noticed by Park. Another junction between the Awamoa beds and the thin development of Hutchinson Quarry beds is exposed at the north end of All Day Bay, and here there is a conformity, as was admitted by McKay;* Geol. Rep., 1886, p. 236. who in the same report† Loc cit., p. 235. states that the Miocene beds (Awamoa) pass gradually into the Hutchinson Quarry beds, and thus he admits the conformable nature of the whole series.

One of us has spent much time in this district, and has made a careful inspection of these and other sections (see fig. 8), and is positive that the different members of the series are perfectly conformable in all the sections that he has visited. Even if, as is reasonable, the series be recognized as completely conformable, the correlation of the strata allows room for a difference of opinion when they are referred to the typical occurrences in North Canterbury. How great these differences of opinion are may be seen from the accompanying table (fig. 7). In our opinion, it is perfectly reasonable and satisfactory to correlate the groups that are lithologically similar throughout. Thus the Oamaru stone and the Amuri limestone are similar, and should be correlated. So with the concretionary greensands and the Moeraki boulder-beds, the quartz grits and the basal conglomerate, the Weka Pass stone and the Hutchinson Quarry beds. The full reasons for this suggested correlation will be explained later. It is of importance to note that Boehm is unable to accept the separation of the Oamaru stone as Cretaceo - tertiary from the Hutchinson Quarry as Eocene. He says that the only distinction he can discover is that one formation is richer in fossils than the other, and that at Oamaru he can find no evidence in favour of placing the Oamaru stone in a Cretaceo-tertiary class.* Boehm: Zeit. d. Deutsch. Geolg. Gesellschaft Jahrg., 1900, p. 175. Park has made much of a suggested distinction between two limestone series at Oamaru.† Park: Trans. N.Z. Inst., 1904, vol. 37, p. 504 et seq. The only place where these two limestones are said to occur in the same section is at Kakanui North Head, and there they are separated by a volcanic rock only. Since the breccia at Oamaru and numerous sections in its neighbourhood show clearly that volcanic action was prevalent during the deposition of the limestone, this section is not convincing. McKay in his last report on the district omits all mention of two limestones, and one of us who has closely examined the district has quite failed to find any evidence of the presence of more than one limestone. This limestone is commonly known as the Oamaru stone, but in geological reports it is called in almost every case the Ototara series. While too much space would be taken in stating the stratigraphy in detail, the writers are of opinion that there is but one limestone formation in the district, and that its character becomes in general less pure and gradually more arenaceous as the old shore, usually to the west, is approached. Fig. 8. This point of view is strongly supported by Boehm, who specifically states that the rock at the Devil's Bridge, the upper limestone at Kakanui,

and the limestone at Totara are all exactly the same series.* Boehm: Zeit. d. Deutsch. Geolg. Gesellschaft Jahrg., 1900, p. 174. McKay also correlates the limestone at the Kakanui with that at Totara.† McKay: Geol. Rep., 1883–84, p. 63. (e.) West Coast of the North Island. The series of Cainozoic rocks on the western side of the North Island from Kawhia to Palmeraton is lying almost horizontally. There appears to be no unconformity throughout, and there is the same order of deposits as found in North Canterbury, Nelson, and elsewhere. The lower members of the series are, however, exposed in the north only as far south as Taumarunui. They have been divided into different series from the Cretaceous to the Pliocene,‡ Park: Geol. Rep., 1886–87, p. 180. but on general principles only. Bell§ Bell: Annual Rep. Geol. Surv., 1910, p. 6. regards them as Tertiary, and mainly Miocene; Park‖ Park: N.Z. Geology. as mainly Pliocene, though the lower members are classed as Miocene. The most noticeable point about the development of the rocks in this area is the great thickness of the upper members of the series: the grey marls are probably not less than 2,000 ft. thick. (f.) North Auckland. Recently a careful survey has been made by Clarke of the Whangaroa district.¶ Bell and Clarke: N.Z. Geol. Surv. Bull. No. 8 (n.s.), 1909, p. 47. The Kaeo series as there described is formed of the younger rocks. It consists of conglomerates, tuffs, concretionary shales, massive limestones, greensands, and calcareous sandstones. It is, however, doubtful whether this is a conformable series. If it is so, it strongly resembles the series developed in North Canterbury. (g.) Waitemata. [Contributed by E. De C. Clarke] As elsewhere in New Zealand, the succession and relationships of the later sedimentary rocks immediately to the south and east of Auckland have been interpreted in more than one way. Most of the interpretations have probably attracted little notice. In his widely known “Summary of the Geology of New Zealand,” ** Q.J.G.S., 1885, vol. 41, pp. 209, 210. however, Hutton gives a section illustrating his conception of the relation between the representatives of his Pareora and Oamaru systems between the Wairoa River and Howick. He gives no detailed account of the section beyond saying that it “is hard to understand but is quite clear. The Pareora system has been shown by Mr. Cox and myself to lie quite unconformably on the Oamaru system in the Auckland Province.” In an earlier paper†† Trans. N.Z. Inst., vol. 3 (1870), p. 247. he says, “At Turanga Creek we find the water-worn surface [of the Papakura series] covered by a series of yellow clays and sandstones which form part of the Waitemata series of Professor Hochstetter.” He also gives a section (loc. cit., pl. xxvii, sec. iv) with which the section given in Q.J.G.S. agrees in essentials. General.—After a complete examination of the district, and specially of critical localities, the author has come to the conclusion that the following

account gives the most correct interpretation of its stratigraphical relations. The old rocks of the Maitai series—possibly of early Mesozoic age—form the core of the Maraetai Hills and reach the shores of the Waitemata (Auckland) Harbour near Maraetai. Skirting these hills, and lying unconformably on the older rocks, are various sedimentaries—conglomerates, Fig. 9. sandstones, clays, and limestones—which are typically exposed in Slippery Creek, near Papakura, and to which the name of Papakura series was given by Hochstetter. The characteristic beds of the Papakura series have since Hochstetter's time been found to extend along the flanks of the hills in a northerly direction at least as far as Waikopua Creek, where limestone resembling that in Slippery Creek was found by Park.

The beds of sandstone and clay with interbedded volcanic grits exposed round the shores of the Waitemata Harbour have been known since the days of Hochstetter as the Waitemata series. The Waitemata series is unconformably overlain by the lava-flows and fragmental volcanic material from the volcanic vents of the Auckland Isthmus, and at Panmure by later sedimentary deposits. All observers appear to agree that between Maraetai and Turanga Creek rocks belonging to the Papakura series are represented, while the sedimentaries between Turanga Creek and Auckland are unanimously called Waitematas. Near Turanga Creek, therefore, the contact between the two series, if they are distinct, must occur. Cox, in a paper “On certain Points Connected with the Geology of the Auckland District,” * Rep. Geol. Surv., during 1881. gives a section from Maraetai to Tamaki West, in which he shows the “Maitai slates” overlain by the following sequence of rocks: “(b) Calcareous sandstone, (c) clay marls, (d) concretionary tufaceous sandstone, (e) clay marls, (f) concretionary tufaceous sandstone, (g) bedded sandstones and clay marls, (h) plastic clays and sands.” He says that either a direct sequence occurs from (b) to the Orakei Bay beds, with a possible unconformity between (d) or (f) and the Howick beds—presumably (e), (g), or (h)—or else that the beds forming Howick Peninsula are unconformably younger than both (d) or (e) and than the beds found between Tamaki West Head and Orakei Bay. While he regarded the latter view as the more probable, he considered that stratigraphical evidence favoured the former view. Hutton's views have already been summarized. Park (with whom McKay† Rep. Geol. Surv., during 1887–88, p. 40. agrees in this matter) in two papers‡ Rep. Geol. Surv., during 1885, p. 136, and Trans. N.Z. Inst., vol. 22, p. 391. affirms the conformity of the Waitemata and Papakura series east of Howick, considering that the irregular overlap taken as evidence of unconformity by Hutton and Cox was due to unequal erosion of hard and soft beds. In his paper on “The Volcanic Beds of the Waitemata Series” Fox§ Trans. N.Z. Inst., vol. 34, p. 485. remarks (p. 485) that from the Papakura limestone to the highest Waitemata sandstones the series apparently has no break. The evidence for unconformity between the Waitematas and the Papakuras appears to depend on (1) the relation between the series as seen near Turanga Creek; (2) the relation between the series as seen in the cliffs between Howick and Maraetai. As regards (1), the writer has followed the sequence from the so-called Turanga greensands (which are volcanic grits)‖ Fox: Trans. N.Z. Inst., vol. 34, p. 485. which Hutton regards as part of the Papakura series to the undoubted Waitematas at Howick, and after mapping all the observed strikes and dips¶ Too numerous to place on the rough sketch-map accompanying this paper. can find no evidence of unconformity. As regards (2), passing along the shore-line from Motu Karaka towards Maraetai, alternating sandstones and clays, in places dipping steeply, in others almost horizontal, are seen to be quite unconformably overlain near the top of the low cliffs by a soft clay showing little or no bedding. In the upper part of this clay is a well-defined horizontal seam, about 6 in.

thick, which may be traced for a considerable distance along the cliffs. This band appears to be a very impure pumiceous earth. According to Hutton, the lowest beds seen in the section described above are to be classed with the Papakuras, the upper beds with the Waitematas. But the lower beds show a very close resemblance to the typical Waitematas as seen near Howick, at the other side of the Turanga Creek, and may be traced at intervals round the shores of Turanga Creek until they are found to pass into the upper rocks which Hutton describes as Waitematas, and which he says unconformably overlie the greensands (belonging also to the Papakura series) at Turanga Creek. On the other hand, the uppermost beds seen in the cliffs between Motu Karaka and Maraetai are quite distinct lithologically from the Waitematas as developed elsewhere. Stratigraphically and lithologically, therefore, there seems no reason to regard the lower beds between Motu Karaka and Maraetai as other than Waitematas, and lithologically there is good reason to regard the upper beds as distinct from the Waitematas, and younger. It may be of interest to note that at Slippery Creek, while the bed of the creek is occupied by the fossiliferous limestone and conglomerate of the Papakura series, the upper parts of the hills bordering the stream are composed of alternating sandstones and clays showing a close resemblance to the typical Waitemata beds. No evidence of unconformity between the upper and lower beds just described could be found by the writer. Stratigraphical evidence too detailed to be given in this paper has been collected by the writer which shows that there is little or no vertical difference between the lowest beds of the Waitematas and the lowest beds of the Papakuras. In conclusion, therefore, it seems highly probable that the Waitemata and Papakura series are part of one conformable series The differences between the typical beds of the two series would, of course, be due to differences in conditions of deposition. IV. Deposition of Rocks of the Series. (a.) General. A careful consideration of the stratification of these younger rocks, a brief summary of which has been given in the preceding pages, has impressed the writers with the belief that there is a single stratigraphical series of younger rocks in New Zealand, and that this series is of very general development and has a remarkably similar lithological succession in all the localities. There are, however, some problems of considerable difficulty that require solution if this statement be adopted. Briefly stated, the main problem is of this nature. At the Waipara Gorge, Amuri Bluff, and the Malvern Hills, fossils that have decidedly Cretaceous affinities are found in the lowest rocks. The most important of these are Belemnites australis Phillips, Conchothyra parasitica McCoy, Trigonia costata Hector, Trigonia sulcata Hector, Inoceramus haasti Hochstetter, as well as a considerable number of species of pythonomorphs and sauropterygians. In the middle beds (Mount Brown) the Mollusca have a definitely Cainozoic appearance, and 20 per cent. belong to Recent species. In the highest beds 60 per cent. (Hutton) or 71 per cent. (Park) belong to Recent species. Though it is not here intended to lay great emphasis on the percentage

of Recent species, these figures are quoted to show the great palaeontological difference between the lowest and the highest beds. So great is this difference that it is natural to look for breaks in the stratigraphical sequence, and the temptation is naturally great to lay emphasis on any slight local irregularity in the stratigraphy and to magnify it into an unconformity. A consideration of this difference perhaps causes one to demand from an observer who holds to the idea of a conformable sequence a special explanation of the rapid faunal change, since the whole series as developed in North Canterbury is not more than 3,000 ft. thick. (b.) Explanation of Rapid Change of Life-forms. We believe that this explanation is to be found in these three considerations: (1) A possible isolation of New Zealand during the late Mesozoic; (2) a possible lingering of archaic types; (3) the very slow rate at which the deposits accumulated. (1.) Isolation. Evidence as to the previous isolation of the eastern coast-line of New Zealand is somewhat conjectural. It is, however, certain that in the middle Mesozoic great rock-movements were in progress, and the early Mesozoic rocks of New Zealand were folded and elevated into mountainranges of great size. So important was this movement in New Zealand that it is reasonable to inquire whether it extended beyond the boundaries of the present land. It is then found that to the north the rocks of New Caledonia of early Mesozoic age are also folded. In the south the quartz gravels of the lowest members of the series of younger rocks in Campbell Island were evidently derived from folded and metamorphic rocks similar to those of Otago. There is no definite evidence of further extension, for the Beacon sandstone of South Victoria Land can hardly be cited in this connection. If the New Zealand shore-line at that time (the close of the Mesozoic) extended from very low to high latitudes a measure of isolation would be thus obtained. It is noteworthy that in the folded and highly eroded rocks of early and middle Mesozoic age marine reptiles were numerous. Trigonia, Inoceramus, and Belemnites were well represented; so that in regard to these organisms at least there is a suggestion that during the great interval throughout which immense rock foldings and erosion were in progress the fauna of the coast-line suffered but little from the competition of new and more vigorous organisms. (2.) An Archaic Fauna. It is well known that at the present day many forms long extinct in Europe and America linger and maintain an existence in the south-west Pacific. The Trigonia of Australia, Ceratodus, Sphenodon, and such genera as Struthiolaria are, of course, examples, and it is possible that owing to causes apart from isolation many Mesozoic forms had survived in New Zealand after they had become extinct elsewhere. (3.) Slow Rate of Deposition. (i.) Conglomerates.—We are inclined to ascribe much of the advance in the fauna to the mere lapse of time. It is true that the total thickness of the strata is not more than 3,000–3,500 ft. at Waipara, where all the strata are present; but it is also known that nearly every member of the series has a thicker development elsewhere—to take a single example, the bottom hundred feet of sands and gravels is represented by 7,000 ft. of

gravels in Westland. It is also true that the greater part of the series was deposited on a slowly subsiding sea-floor, and was derived from a slowly subsiding land on which all the stream-grades were thus being gradually reduced, and most of the material was doubtless utilized in filling up previously eroded valleys. At the same time the area from which sediment could be derived was constantly being reduced. A more detailed description of the different strata is, however, necessary in order that the slow rate of deposition should be fully realized. The basal gravels are, as previously stated, 7,000 ft. thick in Westland. They are certainly more than 1,000 ft. thick near Cape Farewell, and perhaps 3,000 ft. thick at Shag Point. This thickness points to a considerable lapse of time In most localities where they are relatively thin the component pebbles show evidence of great and prolonged attrition. This is most marked in Otago, where the gravels are composed almost wholly of quartz pebbles, and they are unquestionably derived from rocks of mica-schist. The rivers that flow from the schist at the present time carry down schist pebbles amongst which are a few of quartz. A great amount of attrition of the schist and a gradual survival of the hard quartz pebbles would be necessary before the change from the schist pebbles to a quartz gravel would result. This is equivalent to the statement that the rivers supplied material more slowly, and that it was subject to most prolonged attrition on the beaches before it was screened by a new supply. The lowest stratum, therefore, by its nature or by its thickness indicates that much time elapsed during its deposition. (ii.) The Coals.—Interstratified with these conglomerates there are in many places beds of coal which are sometimes of great thickness. Those certainly indicate a slow rate of deposition, and imply long periods of nearly stationary conditions separating the periods of more rapid depression. As the depression proceeded the area of land must have been greatly decreased, and this satisfactorily explains why in some localities an arenaceous limestone rests directly on the eroded surface of the older rocks. This is the case at Kawhia, and on the Gouland Downs in the west of the Nelson Province. The search for geological indication of the presence of coal-seams has perhaps more than any other reason been the assignable cause for the close examination to which the New Zealand Cainozoic rocks have been submitted. In all places where coal has been found in these rocks it has been near the base of the series, interstratified with the conglomerates. It is, however, of local extent, of very variable thickness, of variable composition, and occurs at very different levels. The fossil Mollusca associated with the coal in different places are, however, distinctly different. This has been the reason for assigning the seams to very different ages. Hector classed some in the Cretaceous, but the greater part in the Cretaceo-tertiary at the horizon of the black grit. Hutton considered several coals of Cretaceous age, others Oligocene, and some Miocene, in each case forming the base of the younger series of rocks in the districts where they occur. Not only have they been classed in various ages by different observers, but the same observer has placed them in very different ages in different publications, and the whole subject of the coal-seams has thus become of a complex and confusing nature. To illustrate this it is only necessary to refer to the following works of Park in reference to New Zealand coals :— “The Extent and Duration of Workable Coal in New Zealand” (Park, Trans. N.Z. Inst., vol. 21, p. 327, 1889): “The workable coals of New

Zealand are all of the same age.” “At the base of the group with a Tertiary facies” (p. 325). “Notes on the Coalfields of New Zealand” (Proc. Inst. Min. and Met., vol. 8, 1899, p. 148): “All the workable seams of coal are found in measuues of Lower Eocene age.” “Coal never has been, and never will be, found below Amuri limestone” (p. 150). “Age and Relations of New Zealand Coalfields” (Trans. N.Z. Inst., vol. 36, pp. 411, 418, 1904): Cretaceous Coals (below Amuri limestone): Waipara Series—Pakawau, Mokihinui, Westport, Greymouth, Malvern Hills, Shag Point. Miocene: Oamaru Series—All the other coal-seams, of which seventeen are enumerated. Proceedings N.Z. Inst., 1909, p. 59: The Green Island Coalfield is classed as Mesozoic. “Geology of New Zealand,” 1910, p. 293: Upper Cretaceous: Waipara—Shag Point, Kaitangata, Malvern Hills, Kawakawa, Hikurangi, Ngunguru. Waimangaroa Series: Eocene — Grey, Paparoa, Westport, Mokihinui, Pakawau. Oamaru Series: Miocene—Taupiri, Waipa, Mokau, West Wanganui, Inangahua, Mount Somers, Kakahu, Waihao, Ngapara, Waikouaiti, Green Island, Forest Hill, Nightcaps. Thus, to take a single example, he has classed the Kaitangata coal as Lower Tertiary in 1888, Lower Eocene in 1899, Miocene in 1903, and Cretaceous in 1910. Our observations have led us to the opinion that the coal is always at the base of that development of the series of younger rocks that happens to be present in any district. Since, owing to the great overlapping of the higher beds of this series, the base in any locality may be of any age between the Cretaceous and the Miocene, it is evident that the coal in particular places may be in any part of the Eocene, Oligocene, and perhaps occasionally Miocene system. It is therefore, in general, useless to classify the fossils and assign the shell-bearing strata to any definite period as a preliminary to a search for coal. It is only necessary to deter mine the depth of the base of the series, for it is there that coal may be found, whatever the age of the base of the series in that locality may be. It has been recognized by all authorities that individual coal-seams are not widely extended, and the frequent presence of quartz or other pebbles in them suggests detrital origin. It is therefore evident that general principles cannot here be used in predicting the occurrence of possible coalseams. Prospecting in each separate locality is necessary, for detailed stratigraphical methods are in the main useless. The opinion that is here expressed is strongly supported by the acknowledged absence of seams of coal in any horizon above the basal conglomerate in any section. If coals had been formed in unconformable series, it is not unreasonable to suppose that seams would be found in some localities in two different formations at different levels. This has not yet been found. (iii.) Greensands.—The greensands at the Waipara and Amuri Bluff were apparently deposited under somewhat exceptional conditions. Though containing little or no pyrite, they liberate large quantities of sulphuretted hydrogen from the natural exposures in cliffs, and in many places there is an effloresence of sulphur formed on the surface of the sandstone in some quantity. The associated grey and white sands are also almost entirely formed of quartz, and the same remarks apply to them. The greensands are of

great thickness in the Waipara, and at Amuri Bluff 600 ft. We are not yet fully acquainted with the conditions under which greensand is formed, but the latest information shows that it is deposited near a steep coast where the water is particularly clear and perhaps 500 fathoms deep.* Murray and Lee. Such conditions evidently favour slow accumulation, and this idea is supported by the abundance of sharks' teeth in the greensand in many localities. (iv.) Limestones.—The limestones vary much in thickness and in nature. The limestone at the Amuri Bluff is formed mainly of isolated chambers of Globigerina. It is here little more than a chalk. At other places the larger species that live on the sea-floor are more prominent. This is very noticeable in the more sandy varieties at Dunedin, and still more conspicuously at the Mokau, where the limestone contains also an abundance of the remains of the calcareous alga Lithothamnium. At Oamaru the building-stone consists mainly of plates of echinoderms, with Polyzoa and Foraminifera. Near Cape Farewell coral is the chief organism in it. The deposition of the limestone must represent a considerable lapse of time, for at Amuri Bluff it is 650 ft. thick, and its nature there shows that it formed with extreme slowness. In all places it is uncontaminated with sediment, except quartz-grains. Its wide occurrence and penetration into many mountain-gorges shows how much of the present land-area was submerged at this time. Above, as below, the limestone passes in many places into a greensand, but this upper stratum is relatively thin when compared with the lower one, and this, of course, suggests that the upward movement was more rapid than the downward movements. The grey marls are very thick in some places, though it is quite often the case that they have been eroded off the surface of the limestone. Their absence in the southern part of the South Island is perhaps to be explained in this way, though it is quite possible that they were never deposited in that part of the country. The grey marl consists largely of minute scales of mica, though these are mixed with much calcareous matter. The Mount Brown and Pareora beds that lie on the grey marls are relatively coarse detrital formations, and may have been deposited much more rapidly than the lower beds of the series. Taking the series as a whole, we think that the time required for its deposition was sufficiently great to allow of considerable faunal change to take place, and specially when a possible previous isolation is borne in mind. We regard it as almost sufficient in itself to account for the important differences between the fauna of the lowest and highest members of this conformable sequence. V. Correlation of Members of the Series. As previously stated, we believe that there is complete evidence of stratigraphical conformity of the rock-series here described. It is, however, a fact that the lowest member of the series in some localities, such as Waipara and Amuri Bluff, is much older than that of Kawhia and of Oamaru (Livingstone). This has always appeared to justify the division of the series into different geological systems, which was based on the difference in faunal characters of the various members of the series at Waipara, and supported by the stratigraphical breaks described in the series by different

geologists. We believe that this occurrence of a younger fauna in the basement beds in some districts than in others is due simply to overlapping. This, however, is difficult to prove, for earth-movements that have occurred since the deposition of these rocks have greatly altered the relative levels of different portions of the country, so that it is now difficult, if not impossible, to restore the early Cainozoic relief. The disposition of the Cainozoic series shows quite clearly that the relief was then highly varied, and, if that was the case, overlapping must have taken place to a great extent in a series which in some places was 3,000 ft., in others perhaps 10,000 ft. thick, and deposited during a movement of depression that was more rapid than the rate of deposition. This creates great difficulties in the way of all attempts to correlate the members of the rock-series as developed in different districts, for the relief of the land before depression had been so great, and the movement of depression was so much more rapid than deposition, that limestones and conglomerates were in the middle of the period being deposited within a comparatively short distance from one another. This difficulty is particularly marked because the lowest rocks are usually unfossiliferous. While great difficulties arise for these reasons in all attempts to correlate the basal conglomerates, there is not the same trouble in correlating other members of the series. This is particularly true of the limestones. The nature of this member of the series proves that it was deposited at the time of maximum depression, when some of the areas at least were covered by deep water, and the area of the land-surface was so decreased that little sediment was derived from it, and in many places calcareous conglomerates were deposited on the very shore-line. The differences in the limestone in different localities have already been described, but it is necessary to state that the variations are found in detail only. The fact that where a complete series is developed the limestone always occupies the same position strongly supports the correlation of the limestones throughout the series.* See T. C. Chamberlain: “Diastrophism as the Ultimate Basis of Correlation” (“Journal of Geology,” vol. 17, 1909, p. 685). This is the correlation adopted by Hector, and almost by Hutton, except that he placed the Amuri limestone in a lower unconformable series. It is wholly opposed to the classification of Park, who correlates the great limestone formation throughout the country with the calcareous knobbly conglomerate of Mount Brown. From this we wholly differ, for the palaeontological evidence upon which it is based is far from complete, and can be interpreted in very different ways. The correlation is mainly based upon the resemblance between the fossils collected by him at Mount Donald (Mount Brown beds) and those of the Black Point beds, which are lower than the limestone in this locality; and, again, the similarity of fossils in these beds to those in the Awamoa beds, which lie over the Oamaru stone, appears to be the reason for suggesting that a second bed of limestone—Waitaki stone—should rest on the Awamoa beds. It is also asserted that McKay and Hector always agreed that the Pareora fauna lay below the Waitaki stone.† Trans. N.Z. Inst., vol. 37, p. 504, 1905. This appears to be an error, for in Hector's Handbook the Pareora and Awamoa beds are placed in the Miocene, the Hutchinson Quarry beds in the Eocene, and the Ototara series (Oamaru and Waitaki stone of Park) in the Cretaceo-tertiary. When this similarity of fossil

fauna is closely questioned doubts are at once raised as to the correctness of inference drawn from it. Thus, of the twenty-eight species of Awamoa molluscs listed by Park * Trans. N.Z. Inst., vol. 37, p. 512, 1905. only fourteen occur in the list of thirty-six species recorded in the Mount Brown beds, † Loc. cit., p. 540. and eleven of his twenty species at Pareora. Again, eight of these Pareora ‡ Loc. cit., p. 531. species are listed in the twenty-eight of the Awamoa, and eleven of the twenty occur in the Black Point beds, while thirteen of the twenty-eight Awamoa fossils in the Black Point and eleven of the twenty-three Hampden Beach fossils are found in the Black Point beds. So that when reduced to actual figures these lists do not show any special resemblances. In nearly every case about 50 per cent, of the species in one list are found in other lists. It is certainly true that far more complete collections are required before palaeontology can be used as a basis for forming an opinion as to the exact stratigraphical position of any of these beds. When applied to the Waitaki and Oamaru limestone of Park, palaeontology is still less satisfactory. Thus, of four species stated § Loc. cit., p. 494. as distinctive of the Waitaki stone three are quoted as found in the Oamaru stone. Again, the Oamaru stone near the old Miocene shore-line is said to “gradually merge into a yellowish-brown calcareous sandstone containing the scattered remains of huge echinoderms.” ∥ Loc. cit., p. 496. The Waitaki stone as it approaches the old shorelines is “represented by the band of yellowish-brown calcareous sandstone with Meoma crawfordi.“∥ Since, as before stated, Boehm did not distinguish between the two limestones, and since in no place can two limestones be seen in the same section except at Kakanui, where volcanic matter is intercalated, even in this district, where limestone is very generally developed and nearly always seen in exposed sections, it appears unreasonable to attempt to maintain the presence of two limestone beds. The fauna throughout the various shell-beds, as already stated, maintains a general resemblance whether beneath or above the limestone, which is seldom more than 50 ft. thick. This merely shows that in this area, which was marginal during the deepest depression, little change took place in the littoral fauna during the deposition of the limestone. Greensands occur below the limestone at Maerewhenua, and above it at the Oamaru rifle-butts, Hutchinson's Quarry, and Deborah Railway-station. In the three last named the brachiopods are almost solely large Magellanias; in the former, smaller species and Bouchardias. It is not desired to insist upon this difference, for it may be due to a difference in station, but it does at least suggest hesitation in correlating these greensands, as is done by Park. We are, however, in complete agreement with Park in his belief that there are no unconformities other than those of an interformational nature associated with volcanic outbursts in the Oamaru district. The unconformities supposed to separate the Miocene, Eocene, and Cretaceo-tertiary of Hector, and the Oamaru and Pareora systems of Hutton, were found not to exist. On the other hand, clear interformational unconformities are to be seen at Hutchinson's Quarry and in many places at Oamaru Cape, and in every case they are associated with the presence of fragmental volcanic matter. The general idea of a single stratigraphical series being present is, we believe, very strongly supported by the fact that there is no place in New

Zealand where the members of one series rest on another with a development in any way approaching their normal character. This statement is true even if the unconformities described by different authors be accepted. It is asserted by several geologists that one series has its typical development in the Oamaru area, and that another series is typically developed at the Waipara. There is no locality yet described where a series resembling that at Oamaru even in a general way can be found resting upon the lower or Cretaceous portion of the series at Waipara or Amuri. The same is true in a lesser degree of the other systems described. When the numerous excellent sections in different parts of the country are considered, this remarkable fact appears to the authors to strongly support the view that is here adopted—namely, that there is a single series, of which the upper members overlap the lower, and hence in many places the series is restricted and the upper horizons only are represented. It is noticeable that in all the bulletins issued by the reorganized Geological Survey under Dr. Bell it has not been found practicable to divide the younger rock-series into two or more systems. VI. Correlation With European Horizons. Any correlation of this nature must be purely tentative, except in regard to the lowest rocks at Amuri. The different correlations that have hitherto been made have been based on the percentage of Recent Mollusca among the fossil forms in different members of the series. This is wholly unsatisfactory, because (1) we do not even know with any exactness the present molluscan fauna, though this difficulty will disappear when Mr. Suter's “Manual of New Zealand Mollusca” appears; and (2) many members of the series, especially the greensands and limestones, were deposited in deep water, and our knowledge of the Recent fauna off the coast of New Zealand is, at most, fragmentary. The only comparisons that can be of any value are those between the littoral fauna of different strata. It is not certain that we have at present a description of the fauna of the basement beds throughout the overlapping series, and it is therefore impossible to trace the changes that took place as time went on. Some time must elapse before this can be remedied, for in the great majority of localities these beds do not contain fossils. Where fossils have been found they are of a very different nature in the various places. Thus, at Amuri Bluff and Waipara the lowest beds appear to contain no Recent species. The same is true of the Malvern Hills and Brighton, though here there are very few fossil species. At Black Point, in the Waitaki Valley, however, the number of Recent species is considerable. As previously stated, we believe that the series was deposited during a great lapse of time. The lowest beds are, at the latest, Eocene, and perhaps Upper Cretaceous; the highest of the conformable series are Upper Miocene or Pliocene. If, as is here suggested, the limestone is of the same age throughout the country, it should be possible to come to a definite conclusion as to what that age is. The fauna that it contains are mainly echinoderms, Polyzoa, corals, Foraminifera, and sponge-spicules. Collections of each of these groups have been examined by specialists, and the following statements have been made by them :— Tate said of the echinoderms, “There is no doubt that the Oamaru formation is correlative with the Lower Murravian of Australia.” This,

he says, is comparable with the European Eocene, with a slightly more Cretaceous complexion.* N.Z.G.S., 1892–93, p. 121. Tenison-Woods says of the corals, “I have no doubt that from the fossil corals the formation at Oamaru and that at Mount Gambier (of Australia) were contemporaneous. The stage accepted by him for the latter is a stage later than the Upper Eocene.† “Palaeontology of New Zealand,” pt. 4, 1880, p. 4. The same author examined the Bryozoa and compared them closely with the Mount Gambier formation of Australia. Stache examined the Foraminifera obtained by the “Novara” expedition from Raglan and Whaingaroa. He classes them as the same age as those of the Vienna basin or of the Upper Oligocene of north Germany.‡ “Reise der ‘Novara’: Palaeontology,” p. 299. Hinde and Holmes classified sponge-spicules from Oamaru. They class the deposit as belonging to the Upper Eocene or Oligocene.§ Journ. Linn. Soc., Zool., vol. 24, 1892, p. 178. In addition to these, the sharks' teeth of the greensand have been examined by Davis, who, however, makes no suggestion as to the age of the beds in which they occur. It is evident that there is a general consensus in favour of the limestone from which all these groups of fossils have been obtained being classed with the early Tertiary, between the late Eocene and the late Oligocene. At present we do not intend any further correlation than this, and we believe that the late Cretaceous or early Eocene may be taken as the age of the oldest bed, the Oligocene as the age of the limestone, and the late Miocene or Pliocene as that of the youngest bed in the conformable series at Waipara. A skeleton classification is thus formed which can satisfactorily be filled up by the other members of the series. It is the intention of the authors to state fully the palaeontological side of the question in future papers. VII. Summary And Conclusions. 1. In all the sections of this series of younger rocks that we have had opportunities of examining in different parts of the country we have been unable to find any evidence of a stratigraphical unconformity in a single instance, though our observations have extended over those sections that have been regarded as crucial by different observers. 2. Though every geologist who has written about these sections previously has at one time or other insisted upon the existence of stratigraphical breaks, each observer has placed these in a different position in the series from the others. 3. The palaeontological evidence shows that the fauna which existed when the upper beds were deposited was very different from that which existed when the lower beds were deposited. It is also suggested that this rapidity of faunal change is more apparent than real, and that the strata of this series were deposited at a very slow rate. 4. Correlation has been confused because the overlapping nature of the upper members of the series has not been fully recognized. There is reason for supposing that it is correct to correlate all the conspicuous limestones of this younger series as contemporaneous. 5. So far as correlation with the European geological system is concerned, the limestone appears to be of early Oligocene age. The lowest beds of the series are perhaps Cretaceous, and the upper perhaps of Pliocene age.