Art, VII.—The Volcanic Rocks of Oamaru, with Special Reference to their Position in the Stratigraphical Series.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 50, 1918, Page 106

Art, VII.—The Volcanic Rocks of Oamaru, with Special Reference to their Position in the Stratigraphical Series. By G. H. Uttley, M.A., M.Sc., F.G.S., Principal, Scots College, Wellington. [Read before the Wellington Philosophical Society, 16th July, 1916; received by Editors, 31st December, 1917; issued separately, 24th May, 1918.] Contents. I. Introduction. II. Succession of the Rocks of the Oamaru Series. III. Previous Opinions in regard to the Horizon of the Volcanic Rocks. IV. Effects of these Opinions on the Classification of the Oamaru Series. V. Descriptions of the Sections. (1.) Oamaru Lighthouse. (2.) Oamaru Rifle Butts. (3.) Hutchinson Quarry and Neighbourhood. VI. The Pillow-lavas. VII. Chemical and Petrographical Notes. VIII. Conclusion. I. Introduction. Interpretations of the geology of the Oamaru coastal district where the typical Oamaru series is developed have varied greatly in the past, and, although there is now pretty general agreement as to the broad features of the rock-sequence, it is essential that more detailed stratigraphical work should be attempted if the best results are to be obtained from the determinations of the Tertiary fossil Mollusca and Brachiopoda, which are being carried out by Mr. H. Suter and Dr. J. Allan Thomson. Many of these fossils have been collected from the Oamaru district, the exact locality and rock from which they have been gathered are known, and, if the stratigraphical sequence can be determined in greater detail than has been done in the past, correlation of Tertiary rocks in other parts of the Dominion with those developed in the typical locality will possess a sounder basis than it does at present. Misinterpretations of the rock-sequence at Oamaru have undoubtedly been caused by faulty identification of the horizons of the volcanic rocks, and it will assist stratigraphical work if these can be determined more accurately. The position of the lowest volcanic rocks, the Waiarekan, is not in question; it is generally agreed that they lie immediately below the Ototara limestone, and my work in the North Otago district has convinced me that they are invariably associated with deposits of diatomaceous earth in the Oamaru and Papakaio areas. Difference of opinion has, however, been sharply marked when dealing with the volcanic rocks near the Oamaru coast. These have been correlated with the tuffs in the Waiareka area; important unconformities have been introduced into the sequence, on the ground that the volcanic rocks are clear evidence of the existence of a former land surface. An attempt will be made in the present paper to define more clearly the place of the volcanic episodes in the late geological history of the Oamaru district. In 1916 I gave a detailed sequence of the rocks east of the Waiareka Valley, and some of the evidence on which the succession was based was presented in that paper. It is now proposed to bring additional evidence by describing several sections in the neighbourhood

of the town of Oamaru. The pillow-lavas that occur in some localities will be described, and a preliminary reference will be made to the microscopic characters of some of the rocks, a detailed description of which must be reserved for a future paper. I am much indebted to Dr. Thomson and Mr. H. Suter for naming many of the fossils. Their determinations are indicated by an asterisk placed in front of the name of the fossil. The other determinations have been made by myself by comparison with forms determined by these palaeontologists. II. The Succession Of The Rocks Of The Oamaru Series. It will be advisable at the outset to state briefly the detailed succession of the Oamaru series as developed in the typical locality. The classification of the upper rocks is based mainly on evidence to be brought forward in this paper. To the west of Oamaru, in the basin of the Kakanui River, the lowest Tertiary rocks are the grits, clays, and sandstones, sometimes associated with lignite deposits, the whole forming the Ngaparan stage. These are followed by a considerable thickness of greensands, and these in turn are succeeded by a great thickness of breccias and tuffs, occurring in the Waiareka Valley. Towards the top these become fine-grained and tachylytic, and are interbedded with deposits of diatomaceous earth, which, in addition to the vegetable micro-organisms, contains an equal abundance of spongeremains and Radiolaria. Dykes and sills have intruded the tuffs and diatomaceous earth, and the siliceous rock has in many places been completely metamorphosed into a flinty substance. These constitute the Waiarekan stage. The Ototara limestone is the next succeeding rock. In its lower portions it is interstratified with thin beds of marl, and occasional thin layers of rolled volcanic pebbles. In its middle portion it is in some localities intercalated with tufaceous bands, but these are probably detrital only. In other localities the deposition of the limestone continued uninterruptedly; and it is free from volcanic material. Towards the end of the limestone period volcanic activity was renewed in localities near the present coastline with the eruption of the breccia at Kakanui and the volcanic rocks in the neighbourhood of Oamaru, the upper pillow-lava of the latter locality being younger than the breccia. After volcanic action had ceased limestone continued to be deposited, but in many places it contains large well-rounded masses of volcanic rocks, and minerals similar to those occurring in the breccia below. It is more than probable that the Kakanui breccia formed small islands or submarine banks, for it is followed by limestone bands and tuffs, the former containing rolled fragments of the breccia. In Oamaru Creek these interstratified tuffs and calcareous bands are invaded by a thick mass of dolerite which has overflowed to the north and formed the upper pillow-lava near Grant's Creek. This in its turn is followed by the limestone containing the large volcanic boulders. The latter disappear towards the top of the limestone, which closes the Ototaran stage. The greensands of the Hutchinsonian stage followed the limestone, and the sequence closed with the mudstones of the Awamoan stage. III. Previous Opinion in regard to the Horizon of the Volcanic Rocks. Hector (1865) considered that the volcanic rocks near Hutchinson Quarry, in the town of Oamaru, were submarine, but he based his conclusion on the

erroneous assumption that the hard limestone bands in that locality had been metamorphosed during deposition by a lava-flow. Hutton (1875, p. 54) said, “No eruptive rocks are found associated with the older or Ototara group of strata …. but at Oamaru Heads we have clear evidence that during the deposition of the Upper or Trelissic group of beds volcanic action was going on.” In 1876 McKay placed the Waiareka tuffs below the Ototara stone, and asserted that a younger series of volcanic rocks occurred at Oamaru. In 1886 Hutton verified McKay's observations as to the position of the Waiareka tuffs, but repudiated his own former statement that volcanic rocks were associated with the upper beds of the series. He recognized but one horizon of volcanic rocks, the Waiarekan, and considered that the volcanic matter in the upper part of the limestone was detrital only. In 1905 Park asserted that volcanic activity commenced at the end of the Waihao greensand period and culminated during the deposition of the Hutchinson Quarry beds. Summarily, according to McKay there were two distinct periods of eruption, the pre-Ototaran and the pre-Hutchinsonian; according to Hutton but one, the pre-Ototaran; while according to Park activity continued during the Waiarekan, Ototaran, and Hutchinsonian periods. IV. Effects of these Opinions on the Classification of the Oamaru Series. The fact that the volcanic rocks are always followed by limestone has undoubtedly caused confusion in classification, and in the absence of distinctive fossils in the limestone the igneous rocks near the coast have been assumed to be Waiarekan. McKay, although recognizing the Tertiary volcanics at Oamaru as distinct from his Cretaceo-Tertiary Waiarekan tuffs, erroneously supposed the breccias at Kakanui to be Waiarekan (1877, p. 56), whereas they are middle Ototaran. Hutton in ascribing the volcanic rocks at Oamaru to the Waiarekan was necessarily compelled to introduce an unconformity above them to account for the non-existence of the Ototara stone. Park's contention that volcanic activity culminated during the Hutchinson Quarry period may be true or not true; it depends entirely on the connotation of the term “Hutchinson Quarry beds.” McKay seems to have been the first geologist to use the term in classification (1882, p. 58). Later in the same report (p. 76) he seems to restrict the term to the greensands alone, correlating the calcareous beds with the Otekaika limestone, and the volcanic rocks below with the Kekenodon beds. In other words, in the typical locality he excludes the limestone bands. This is in substantial agreement with Thomson's use of the term “Hutchinsonian” (1916, p. 34). Hutton was probably correct in considering much of the volcanic matter detrital only, as will be demonstrated in the sequel. Further, McKay considered the volcanic rocks of Oamaru Creek as evidence of a land surface, and he supposed the overlying rocks to be markedly unconformable to them (1877, p. 58). My own opinion, based on the evidence furnished below and on further observations in other parts of the Oamaru district, is that deposition was continuous from the base of the Ototaran to the top of the Awamoan, but interrupted locally by submarine eruptions resulting in the formation of volcanic banks or islands, which, however, suffered rapid denudation, and this minor phase is recorded in the slight unconformities now to be seen in the tufaceous beds.

V. Descriptions of the Sections. (1.) Oamaru Lighthouse. In the sea-cliff below the lighthouse near the Oamaru Breakwater a good section is exposed (see fig. 1). The tufaceous beds (a) are interstratified with limestone bands, which often contain large subangular pieces of vesicular basalt. The bands themselves vary up to 1 ft. in thickness, the lowest being 170 ft. below the base of the lower pillow-lava. There is a marked discordance in the dip of the tufaceous rocks below (a). These are not shown in the figure. They dip 40° N. by E., while the dip of (a) is only 20° N. by E. It is probably at this point that McKay introduces his unconformity between his Cretaceo-Tertiary and Upper Eocene beds (1877, p. 58). From the calcareous bands in the tuffs I collected the following forms: *Emarginula wannonensis Harris, Siphonalia sp., Dentalium solidum Hutt., *Pecten hutchinsoni Hutt., Siphonium planatum Suter, Liothyrella oamarutica (Boehm), L. boehmi Thomson, Terebratulina suessi (Hutt.), Aetheia gualteri (Morris), and Hemithyris sp. The overlying pillow-lava is at a minimum estimate 100 ft. thick. The interspaces are filled with fossiliferous limestone, much hardened in places by a secondary deposit of calcite, and the fossils are difficult to extract A detailed description of this peculiar rock and others similar to it will be given later. The following fossils were obtained from the interstitial limestone: *Trochus sp., *Turritella sp., *Polinices huttoni von Ihering, *Lima bullata Born, *Lima lima (L.), Ostrea sp., and Hemithyris sp. Fig. 1.—Section near lighthouse, Oamaru. (a) Tuffs with limestone bands; (b) lower pillow-lava; (c) fine tuffs (current-bedded); (d) tuffbed (very calcareous); (e) limestone band with rounded and subangular pieces of volcanic rock: (f) blue tufaceous clay; (g) limestone; (h) tufaceous limestone; (j) raised beach; (k) broken pillow-lava. The tuff-beds (c) are very fine and current-bedded, but unfossiliferous. The overlying tuffs (d) are very calcareous and coarser in texture. I collected the following forms: Epitonium lyratum (Zitt.), Lima jeffreysiana Tate, Ostrea sp., Venericardia purpurata (Desh.), and Diplodonta zelandica Gray. The limestone band (e) is crowded with subangular pieces of volcanic rocks, while small pieces of augite were also identified. Bed (g) is a much purer limestone than bed (h), which is very tufaceous. In the limestone (g) the following forms occurred: Epitonium lyratum (Zitt.), *Pecten hutchinsoni Hutt., *P. delicatulus Hutt., Terebratulina suessi (Hutt.), *Liothyrella boehmi Thomson (?), and *L. oamarutica Boehm. The overlying tufaceous limestone (h) is also fossiliferous, and the following species were identified: *Limopsis aurita Brocchi, Pecten delicatulus Hutt., *Lima jeffreysiana Tate, Ostrea sp., *Venericardia purpurata (Desh.), V. zelandica (Desh.), Protocardia pulchella (Gray), and Terebratulina suessi (Hutt.).

This bed passes insensibly into a completely tufaceous bed with occasional pillows scattered through it. The tufaceous matter, however, rapidly diminishes, and the rock becomes a pillow-lava, which will be discussed in the sequel. (2.) Oamaru Rifle Butts. Near the Oamaru Rifle Butts, on the south-west side of Oamaru Cape, a clear section is exposed on the beach. It is interesting, as the pillow-lavas are absent and a bed of limestone nearly 50 ft. thick is followed almost immediately by the Hutchinsonian greensands. A fault occurs immediately north of this section, and just beyond the fault there is a marked strati-graphical break in the tuff-beds, exactly similar to the unconformity described in the tuffs near the breakwater. The section extends from a point immediately north of the target sheds to the fault which cuts the tuffs just past the first headland. Fig. 2.—Section northwards from Rifle Butts, distance about 160 yards. (a) and (a1) tuffs; (b) calcareous fossiliferous tuffs, 1 ft.; (c) limestone agglomerate, 2 ½ ft.; (d) (7 ft.) and (d1) (40 ft.) limestone; (e) fine blue tuffs, 8 ft.; (f) greenish calcareous tuffs, 11 ft.; (g) indurated nodular limestone, 4 ft. 4 in.; (h) brachiopod greensand; (i) shell-bed, 2 ft.; (j) blue clav 100 ft.; (k) raised beach. The tuffs (a) and (a1) are calcareous throughout. From the band (b) I collected the following forms: Turbo sp., Turritella sp., Siphonalia conoidea Zitt., Venericardia purpurata (Desh.), Diplodonta zelandica Gray, Chione mesodesma Reeve (?), Dosinia caerulea Reeve, Mesodesma subtriangulatum (Gray), Siphonium planatum Suter, and Liothyrella oamarutica (Boehm). Band (c) is a limestone crowded with subangular pieces of volcanic rock, with occasionally inclusions of a coarsely holocrystalline basic igneous rock. Masses of tuff, broken minerals, and pieces of rounded vesicular basalt occur, one of those being 1 ft. in diameter. The bed (e) is a fine brownish tuff weathering blue, and containing several limestone bands. I obtained the following fossils from one of these bands: *Pyrula sp., *Lima lima (L.), *Pecten sp., and Penacrinus sp. Mr. Henry Suter writes in regard to the genus Pyrula, “It is an unexpected addition to our fauna, and indicates a much warmer sea, the genus living now only in tropical latitudes. It is not found Recent in Australasia, but a species was described in 1888 by Pritchard from the Eocene of Table Cape, Tasmania.” The thicker bed of limestone (d) resembles the building-stone of the Oamaru district; it is poor in Mollusca and Brachiopoda, but I collected the following: Siphonium planatum Suter, *Pecten hutchinsoni Hutt., Aetheia gualteri (Morris), and Hemithyris sp. Overlying the limestone is a fine light-greenish calcareous tufaceous mud, which is very fossiliferous. The species identified were: *Siphonalia conoidea (Zitt.), *Limopsis zitteli von Ihering, *Pecten delicatulus Hutt. *Lima angulata Sow., *L. bullata Born, *L. colorata Hutt., *Venericardia purpurata (Desh.), and *Mesodesma australe (Gmel.).

The nodular limestone is of a similar character to the concretionary bands described by the present writer at Kakanui (1916, p. 23), but in the present occurrence there is very little glauconitic sandy material present, and the band is extremely hard throughout, the nodules being set in a calcareous matrix. The nodules vary in size up to the size of a cricket-ball; they show a concentric structure, with occasionally a central nucleus, while sometimes the centre is hollow. The brachiopod band is a calcareous glauconitic sand, crowded with the typical Hutchinsonian fossil Pachymagas parki (Hutt.). Other species that occur are Pecten delicatulus Hutt. and Pecten (Pseudamusium) huttoni Park. The shell-bed (i) is 2 ft. thick, and consists of a mass of shells embedded in fine grey sands. The fossils are much broken and very friable, and it was difficult to obtain specimens. The bed, however, is similar to the shell-beds at Target Gully and Ardgowan described by Marshall and myself (1913). One fossil obtained here was Aetheia gualteri (Morris), which has not hitherto been obtained above the characteristic Pachymagas parki greensands. Overlying the shell-bed is a blue mudstone with well-preserved fossils which clearly indicate that the bed is Awamoan. A list of fossils from this rock has been given by Marshall (1915, p. 384). The two sections described above are exposed on the extreme north and extreme south respectively of a main anticlinal fold, the arch of which has been thrown into minor undulations and faulted in several places. Between the two exposures tuffs are exposed everywhere along the foreshore, but they have not yet proved fossiliferous. Although, as mentioned above, there are stratigraphical breaks in these tuffs and breccias, they are apparently of minor significance, as the rocks above and below are lithologically similar. Former observers have considered this underlying mass of volcanic rocks to be Waiarekan, but there is not the slightest positive evidence to support this contention. If they are Waiarekan, the breaks in the sequence mentioned above assume a much greater importance, for they will represent the time during which the greater part of the Ototara stone was being deposited. (3.) Hutchinson Quarry and Neighbourhoods. Near the abandoned quarry at Eden Street the typical Hutchinson Quarry beds occur. Although the exposures in this locality are small and disconnected, the succession is clear, and will be best represented by a diagrammatic vertical section. The lowest beds are fine calcareous tufaceous rocks (a) weathering greenish-brown, containing in places large fragments of decomposed vesicular basalt, and ramified throughout by calcareous veins. The bed is 20 ft. thick in the section, but the base is not visible. This passes gradually into the overlying bed (b), which is 6 ft. in thickness. It is a confused mass of glauconitic sands, hardened limestone, and tufaceous matter. In places, however, the limestone shows lime in bands from 1 in. to 1 ft. in thickness. Fragments of fossils are visible but none are recognizable. Overlying is a mass Fig. 3—Section at Hutchinson Quarry. (a) Tufaceous rock, 20 ft.; (b) stone bands with tufaceous rock, 6 ft.; (c) limestone conglomerate, 11 ft.; (d) greensand, 15+ft.

of limestone (c) thickly crowded with decomposed rolled volcanic rocks up to 1 ft. in diameter, while the limestone itself contains small fragments of augite and olivine. The junction of this bed with the underlying tufaceous bed is unconformable in the section, but is probably due to contemporaneous erosion. Towards the top the limestone is free from the conspicuous volcanic boulders, and becomes very hard. This limestone has been proved phosphatic. It is overlain by glauconitic greensand—the typical Hutchinson Quarry greensand—crowded with brachiopods belonging chiefly to the two species *Pachymagas parki (Hutt.) and *Rhizothyris rhizoida (Hutt.). As the gully is followed towards the Oamaru reservoir, outcrops are conspicuous on the hillside, but the sequence is similar to that just described. Past the farmhouse, however, the underlying massive volcanic rock crops out, and this proves to be the same rock that occurs at Chelmers Street quarry as a thick dyke which has overflowed to the north, and developed pillow structure in the bed of Oamaru Creek near Grant's Creek. At the latter locality it clearly underlies a limestone “conglomerate” and limestone capped by the greensands, exactly as at Hutchinson Quarry. The section at Grant's Stream is described in another paper in this volume (p. 121). The brecciated pillow-lava is undoubtedly the same as the upper pillow-lava that occurs above the fossiliferous beds near Oamaru Breakwater (see p. 109). Park, McKay, and Hutton, in discussing the section at Oamaru Breakwater, referred these fossiliferous beds to the Hutchinson Quarry horizon; but whatever interpretation is given to this term, these beds at the breakwater are certainly below the volcanic rock which, at Hutchinson Quarry and the locality near Grant's Creek, is overlain in ascending order by tufaceous beds, from 10 ft. to 20 ft. of limestone, and the typical Hutchinson Quarry greensand. The well-rounded appearance of the volcanic boulders in the limestone in the neigh bourhood of Oamaru Creek, and the confused intermingling of limestone, greensand, and rolled igneous rocks above the brecciated pillow-lava furnish evidence of denudation of the underlying volcanic rocks, which had formed small islands in the Tertiary sea. Formation of islands by submarine eruption is not an uncommon occurrence even in quite recent times, and has been noted by competent observers, who have always recorded the rapidity of their disappearance. This is probably the explanation of the various stratigraphical breaks that have been already referred to. Lister (1891, pp. 596–606), in a paper on the geology of the Tonga Islands, arranges the islands in three main divisions—(a) purely volcanic islands; (b) islands formed of volcanic materials laid out beneath the sea, since elevated, with or without a covering of reef-limestones; (c) islands formed entirely of reef-limestone. Some of the islands of class (b) exhibit suggestive resemblances to the upper Ototaran rocks. These resemblances may be summarized under the following heads:— (1.) They are built up of layers of tuff capped by calcareous rocks; (2.) The fragments of some of the breccias are cemented by a calcareous matrix; (3.) Rounded boulders of volcanic rocks occur in layers embedded in a calcareous matrix; (4.) Some of the tuffs are penetrated by volcanic dykes which do not penetrate the overlying limestone; (5.) Fragments of basic plutonic rock occur on one of the islands;

(6.) Some of the limestones are not true coral-reef limestones, but, according to Sir John Murray, are “chiefly made up of calcareous organisms, fragments of molluscs, echinoderms, Polyzoa, and calcareous algae, together with a large number of Foraminifera.” The plutonic rock mentioned above is a gabbro. Garnet also occurs, as it does in the Oamaru rocks. The intrusive rocks, however, are augite and hypersthene andesites. In view of Mr. Suter's remarks in regard to the genus Pyrula, mentioned above, it is interesting to find that this genus occurs as a fossil in the tufaceous rocks of the island of Mango, one of the Tonga Islands. VI. The Pillow-lavas. Park (1905, p. 513) was the first geologist to recognize the lower rock near the breakwater as a pillow-lava. Since then the present writer has discovered the same peculiar rock in other localities of the district—in the basin of Oamaru Creek, and in the Awamoa Creek near Deborah. At the breakwater, also, the upper so-called “tachylyte breccia” is undoubtedly a pillow-lava which in parts has become brecciated, probably through local explosive action when coming into contact with the sea-water. The lower pillow-lava (see fig. 1) consists mainly of spheroidal masses of lava with the interspaces filled with fossiliferous limestone. The junction with the tuffs below is quite even, and the surface of the tuffs shows no irregularity or indication of having been baked. The dip of these beds is N. 20° E. at an angle of 16°, and the tuff-beds overlying the lava have a similar dip. Each pillow has a tachylytic selvage about 1 in. in thickness, but the rock is holocrystalline at the centre. The pillows at the base of the flow are somewhat irregular in shape, but higher in the mass they become more spheroidal. One of the pillows near the base has a diameter of 30 ft. Higher in the section they are smaller, decreasing in diameter to about 2 ft., but towards the surface they again increase in size. Some of the pillows are much elongated; sometimes they show an indented periphery; occasionally the indentation has penetrated to the centre of the mass, and the upper half appears as if it had fallen over toward the lower half while still in a viscid state. Vesicles are by no means prominent in the rock. Occasionally large scattered ones occur, but on the tachylytic margin they are small and rounded. At the bottom of the flow the vesicles are small, and occur chiefly towards the exterior of the spheroid. Near the top of the lava the pillows are almost free from vesicles. By infiltration of calcareous solutions the rock in places becomes amygdaloidal. The fossiliferous limestone which separates the pillows is indurated, but there is no indication of alteration by heat. The broken-up pillow-lava, which is separated from the rock just described by a thickness of 50 ft. of calcareous tuffs and interstratified limestone, has always been referred to as an agglomerate or breccia; but it is clearly a pillow-lava that has been locally broken up during flow. The pillows vary much: some are similar to those already described, others are much elongated and almost scoriaceous, the vesicles being abundant and much drawn out. The rock throughout is much more vesicular than the lower lava. One pillow was noticed in which a large central cavity was coated with tachylyte, as well as the periphery. There is great variation in the size of the masses, the smallest having a diameter

of 6 in., while the largest are as much as 10 ft. across. The material that separates the pillows consists of fragments of pillows that have become broken up; some of the pieces are tachylyte, others have a selvage of tachylyte, while others are free from tachylyte; and this interstitial material is cemented by crystalline calcite. Only in one place does fossiliferous limestone fill the interspaces, and that is in the neighbourhood of a “limestone dyke,” seven or eight of which have penetrated the vertical fissures in the lava. As limestone occurs above the rock, it is more than probable that in this case at least the separating material has come from above. Boulton (1904, p. 158), in describing British pillow-lavas similar in many respects to the present rock, was of the opinion that the limestone came from below. In places the fragmentary matter makes up most of the rock, with a scattered pillow here and there; elsewhere pillows are massed together and the group is isolated in a matrix of the finer tufaceous-looking material, while the greater portion of the pillows are grouped in a fashion similar to the lower pillow-lava. The occurrence of fossiliferous limestone between the pillows at Oamaru Cape, together with the fossiliferous tuffs and limestones above and below the lower lava, undoubtedly point to eruption under submarine conditions. The glassy selvage which is everywhere present indicates rapid cooling of the masses, and this would take place in contact with water. The differences in the structure of the two rock-masses just described evidently point to some difference in the mode of eruption. Reid (1907, p. 51) says, “It is still a moot point whether pillow-lavas are true outflows or are intruded sills.” Tempest Anderson (1910, p. 632) witnessed the formation of the pillows as the lava entered the water at Savaii. Geikie (1897) and Teall (1899) ascribe the structure to intrusion into loosely compacted sediments. Benson (1915, p. 125) recognized intrusive contacts with the surrounding sediments in the Nundle district, New South Wales. In the present case the lower lava may be intrusive, but there is no positive evidence to support this view, except the fact that there is no sign of explosive action in the mass, as there is in the upper lava. The indications in the field strongly suggested that the latter rock after coming into contact with the water, and after the invidualization of the pillows, underwent disintegration through local explosions in the mass of the flow, the resulting fragments then settling down through the water and becoming incorporated in the main mass as it flowed over the sea-bottom. After cooling, the rock became fissured, and the subsequently deposited calcareous mud penetrated and filled the cracks, forming dyke-like masses. Pillow-lavas, although frequently occurring as deep-sea lava-flows, are not restricted to conditions of great depth, for Tempest Anderson observed the lava flowing into the sea at Savaii. The nature of the sediments above and below the upper lava at Oamaru clearly indicate shallow-water conditions. In fact, there is reason to believe that the rocks of the Oamaru series were all deposited in comparatively shallow water. VII. Chemical and Petrographical Notes. Some preliminary work has been done on the microscopical characters of these pillow-lavas, and chemical analyses made of one of the freshest types from Awamoa Creek near Deborah. More detailed field work is necessary before a full account can be given of these rocks.

Analyses of Pillow-lava, Awamoa Creek. Inferior of Pillow. Tachylytic Selvage. SiO2 49.7 50.6 Al2O3 18.6 17.55 Fe2O3 3.2 1.11 FeO 8.15 11.3 MgO 8.62 7.92 CaO 8.2 8.4 Na2O 1.24 1.13 K2O 0.61 0.55 H2O+ 2.25 1.23 100.57 99.79 H2O- 0.79 0.52 The specific gravity of the tachylytic selvage was found to be 2.74, while that of the interior of the pillow was higher, 2.83. This is in accordance with Hutton's results for the similar rock at the lighthouse, his figures being 2.72 and 2.80 respectively (1887, p. 416). The pillow-lavas at Awamoa Creek (near Deborah) and at the lighthouse near the breakwater show much the same character. In the field there is noticeable on each pillow a distinct black resinous-looking selvage about 1 in. in thickness, while the central portion has all the characteristics of a basalt. A number of sections have been made from these rocks, and as far as examined they present several features of some interest: it is possible to trace the gradual changes from a basaltic glass with a few phenocrysts, undoubtedly of intratelluric origin, to a holocrystalline dolerite, in which the phenocrysts display a similar attitude towards a completely crystalline ground-mass. The extreme edge of the tachylyte selvage consists of light brownish-yellow glass in which small porphyritic labradorite crystals and larger crystals of olivine occur as phenocrysts. Skeleton crystals of a basic feldspar also occur, and minute granules of magnetite are scattered sparingly through the glass, which is irregularly fissured. The olivine is almost invariably corroded by the ground-mass, and some of the crystals are penetrated deeply by the glassy base, this being particularly noticeable in the case of the larger ones. Flow-structure is indicated by the parallel alignment of the feldspars. In a section cut farther from the edge the feldspars are bordered by a dark-brown fringe, giving a shadowy extinction between crossed nicols. Under a high power this resolves itself into minute spicules arranged radially around the feldspars. Where the growth has occurred around a minute feldspar there is an approach to a spherulitic arrangement, and an indistinct black cross is seen when the nicols are crossed. It is noticeable that the glass is not bleached on the periphery of this aggregation of crystallites, as is frequently the case in spherulitic tachylytes. Near the centre of the pillow the glass has completely disappeared, and the ground-mass consists of this fibrous brown material shot through with innumerable skeleton feldspars in all stages of growth, the whole enclosing the same minerals that were developed porphyritically in the tachylitic variety of the rock. Still nearer the centre of the spheroid the ground-mass becomes lighter in colour, and rods of magnetite are plainly distinguishable in a faintly

polarizing base. Another slide from the centre of one of the larger pillows possessed similar characteristics in parts, but augite in bladed ragged forms occupied the same relative position in regard to the feldspars that the spicular growth did in the other sections. The last two varieties of the rock can be paralleled exactly in sections cut from the intrusive rock in Oamaru Creek, west of Hutchinson Quarry. This rock is found to vary in different parts of its mass, however, and a specimen taken from the lower quarry was holocrystalline, and the augite enclosed the feldspars ophitically. The rock shows another variation, in that larger feldspars and olivines are embedded in a mass of granular augite and smaller feldspars. This specimen is much coarser in texture than the other varieties. This granulitic variety of the dolerite is probably due to movement towards the end of the process of consolidation, while the ophitic type indicates that the cooling took place under quiescent conditions. Pillow-lavas invariably belong to the group of intermediate or basic igneous rocks. Among the Palaeozoic rocks well-developed pillow-lavas are of frequent occurrence in Great Britain and are known as “spilites.” These rocks are characterized by their richness in soda and poverty in potash, and mineralogically by the abundance of a soda-feldspar. Albite is the principal constituent, next in importance is augite, and olivine occasionally occurs. There is frequently a glassy base, occasionally the feldspars are microporphyritic, and often those of the ground-mass have pointed or acicular forms. Sometimes they consist almost wholly of feldspar laths with a fluidal arrangement. In some types the augite may occur as irregular masses enclosing the ends of feldspar rods, producing a subophitic structure. Diabases, representing the intrusive magma, invariably occur with these rocks. Flett (1911, p. 246) considers that the characteristic feldspar of these rocks, albite, is due to the action of pneumatolytic emanations containing water with soda and silica in solution upon the basic feldspars soon after the rocks had solidified. Further, he states that the spilitic suite of rocks are essentially rocks of districts that have undergone a long-continued and gentle subsidence. The Tertiary pillow-lavas at Oamaru contain no albite, and this is confirmed by chemical analysis, which shows that the present rocks are remarkably poor in soda, but, like the spilites, remarkable for their poverty in potash. Their structure and mineralogical composition can be paralleled in the spilites, except, of course, for the absence of albite. The rocks were erupted under submarine conditions, but the water was shallow. The association with the intrusive olivine dolerite is analagous to the association of variolitic pillow-lavas with diabases in Anglesey and Cornwall. Yet the poverty of the rock in soda precludes its classification as a spilite. VIII. Conclusion. 1. There are three horizons of igneous rocks—the Waiareka tuffs, Kakanui breccia, and the upper pillow-lava—although the latter two were almost contemporaneous, the breccia being the earlier. 2. The stratigraphical unconformities and “limestone conglomerate” may be explained on the assumption that volcanic islands were rapidly formed and rapidly destroyed. 3. The unconformities introduced into the sequence by former observers are merely local, and of no significance in classification. 4. The Oamaru series is otherwise conformable throughout.

5. The Oamaru pillow-lavas, like similar rocks in other parts of the world, owe their peculiar structure to eruption and solidification under submarine conditions. 6. These rocks, though similar in some respects to British Palaeozoic pillow-lavas, are clearly differentiated from them by their poverty in soda. 7. The structural and lithological resemblances of the upper pillow-lava at the lighthouse and the pillow-lava in Oamaru Creek are sufficiently strong to justify the assertion that they are at the same horizon. The succession of rocks at the latter place and at Hutchinson Quarry is similar from the lava to the top of the Hutchinson Quarry greensand. At the breakwater we get a clear succession of the beds below the lava down to the tuffs. At the Rifle Butts we get a sequence from the Awamoa mudstone down to the tufaceous rocks. Piecing the evidence together, it would seem that the top of the thicker limestone at the Rifle Butts represents the horizon of the upper pillow-lava, and it is probable that the rocks shown below these in figs. 1 and 2 will be found equivalent. The beds are fossiliferous throughout, and careful and more exhaustive collecting should enable this point to be decided. 8. The fossiliferous beds below the upper pillow-lava at the lighthouse are not the equivalent of the fossiliferous beds at Hutchinson Quarry, as asserted by former observers, but are separated from the latter by a considerable thickness of lava, tufaceous beds, limestone, and “limestone conglomerate.” Bibliography. Anderson, T., 1910. The Volcanoes of Matavanu in Savaii, Quart. Journ. Geol. Soc., vol. 66, pp. 621–39. Benson, W. N., 1915. The Geology and Petrology of the Great Serpentine Belt of New South Wales, Part iv, Proc. Linn. Soc. N.S.W., vol. 40, pp. 121–73. Boulton, W. S., 1904. On the Igneous Rocks at Spring Cove, near Weston-super-Mare, Quart Journ. Geol. Soc., vol. 60, pp. 158–69. Dewey, H., and Flett, J. S., 1911. On some British Pillow-lavas and the Rocks associated with them, Geol. Mag., dec. 5, vol. 8, pp. 241–48. Geikie, A., 1897. Ancient Volcanoes of Great Britain (ref. to p. 26). Hector, J., 1866. Reports and Awards of the Jurors, Appendix A, New Zealand Exhibition, 1865, pp. 371–416. Hutton, F. W., 1887. On the Geology of the Country between Oamaru and Moeraki, Trans. N.Z. Inst., vol. 19, pp. 415–30. Hutton, F. W., and Ulrich, G. H. F., 1875. Report on the Geology and Goldfields of Otago. Lister, J. J., 1891. On the Geology of the Tonga Islands, Quart. Journ. Geol. Soc., vol. 47, pp. 590–617. McKay, A., 1877. Oamaru and Waitaki Districts, Rep. Geol. Explor. dur. 1876–77, pp. 41–66. —— 1882. Waitaki, Vincent, and Lake Counties, Rep. Geol. Explor. dur. 1881, pp. 56–76. Marshall, P., 1915. Cainozoic Fossils from Oamaru, Trans. N.Z. Inst., vol. 47, pp. 377–87. Marshall, P., and Uttley, G. H., 1913. Some Localities for Fossils at Oamaru, Trans. N.Z. Inst., vol. 45, pp. 297–307. Park, J., 1905. On the Marine Tertiaries of Otago and Canterbury, with Special Reference to the Relations existing between the Pareora and Oamaru Series, Trans. N.Z. Inst., vol. 37, pp. 489–551. Reid, C., 1907. The Geology of the Country around Megavissey, Mem. Geol. Suru. England and Wales, pp. 1–73. Teall, J. J. H., 1899. Silurian Rocks of Britain, Mem. Geol. Surv. Great Britain (ref. to pp. 420–31). Thomson, J. A., 1916. On Stage Names applicable to the Divisions of the Tertiary in New Zealand, Trans. N.Z. Inst., vol. 48, pp. 28–40. Uttley, G. H., 1916. The Geology of the Neighbourhood of Kakanui, Trans. N.Z. Inst., vol. 48, pp. 19–27.