The Basic Igneous Rocks of Eastern Otago and their Tectonic Environment.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 72, 1942-43, Page 85

The Basic Igneous Rocks of Eastern Otago and their Tectonic Environment. By W. N. Benson With Chemical Analyses by F. T. Seelye. [Read before the Otago Branch, November 20, 1941: received by the Editor, November 27, 1941; issued separately, June, 1942.] Part II. (i) The Distribution of the Late Tertiary (Pliocene) Eruptive Rocks in Eastern Otago in Relation to Regional Tectonics. (ii) Petrographical Nature and Chemical Composition of the Late Tertiary Igneous Rocks in the Relatively Stable and Moderately Deformed Region around the Dunedin Central Region. (iii) Appendix. Note on the Geology of Kauroo Hill. (i) The Distribution of the Late Tertiary (Pliocene) Eruptive Rocks in Eastern Otago in Relation to Regional Tectonics. In Part I (Benson, 1941a) of this paper, it was shown that there has been a noteworthy continuity in the general tectonic character of the several diverse portions of Eastern Otago since Middle Cretaceous times. The broad depression of the Maniototo Plain, and the narrower Taieri-Waihola and Tokomairiro plains give evidence of repeated “negative” or downward movements throughout this period and the same is true of the narrower coastal zone now covered by thick marine sediments extending from Dunedin to Oamaru. It was further shown that in general the highest of the basalt-covered ridges, thrust up during post-Pliocene crust-movements, had been “positive” areas during the period immediately prior to the eruption of the basalt, since the Cretaceous-Lower Tertiary sediments beneath their basalt-caps, if present at all, are very thin. Mapping within the Dunedin or Central Region, the details of which cannot here be elaborated but are sketched below, shows that intermittent crustal movements occurred during the very diversified Pliocene eruptive activity, and reached their climax at its close. It seems, therefore, permissible to consider, at least tentatively, that the extent of deformation which has occurred in any region since the Late Tertiary† Footnote, added May 20, 1942: The Late Tertiary erosion-surface in Eastern Otago has, for certain inconclusive reasons, been tentatively termed the “Late Miocene Peneplain” in Part I of these papers. While this Part II was in the press, however, there appeared Wellman and Willett's (1942) paper, in which it is held that this erosion-surface was formed during later Pliocene times. If this were so, the greatest age which could be assigned to the igneous rocks discussed herein would be very late Pliocene. It would seem desirable, moreover, to retain for the present the original and not too specific term Late Tertiary for the erosion-surface upon which they rest. peneplanation is some measure of the instability of the crust in that region during the period of Pliocene eruptive activity. Plate 36 of Part 1 of this paper indicates that three major divisions may be recognised in Eastern Otago, which, though not sharply separated from one another, may characterise three successive degrees of mobility of its crust.

The Relatively Stable Region comprises the western portion of the area shown in Plate 36. It has certainly not escaped deformation, but has been arched into the broad warpings of the Rough Ridge, Lammermoor, and Rock and Pillar Ranges, which only in their northern portions are broken by close-spaced faults. The same may be true to a smaller extent of the intervening Upper Taieri syncline, and is markedly true of the long, only gently warped, Tuapeka-Waipori-Strath Taieri depression with its eastward extension into the Barewood Plateau. The Moderately Deformed Region sweeps around the eastern and north-eastern margin of the former region, and is characterised by much narrower fault-folds and more closely-spaced faults. (See Part I, p. 213, and Plate 36.) Lying between the Relatively Stable Region and the coast, it extends from the Clutha River embracing the Tuakitito-Tokomairiro-Waihola-Taieri syncline, and its bounding anticlinal ridges, and sweeping northwards, leaving the Dunedin Central District on the east, it bends north-westward to include the watersheds of the Waikouaiti and Shag Rivers and the northern portions of the Rock and Pillar Range and of the Maniototo Plain with a southward projection into the broken ridges east of the Strath-Taieri Depression. Across the Shag Valley,* The account of the geomorphology of this valley given in Part I of these papers with brief reference to Cotton (1922), should be supplemented by reference to the earlier and fuller discussion by Cotton (1917). the lava-capped Kakanui Range must also be included in this region, which may be held to extend to the Kakanui Valley near Kauroo Hill (Part I, Plate 36, Loc. 100). In view of the deep differential erosion of the varied Lower Tertiary formations extending across the Lower Kakanui Valley to beyond Oamaru, one cannot determine the degree of deformation the Tertiary peneplain has suffered here; but it seems possible that, though the deformation was rather vigorous in Miocene times prior to the Late Tertiary planation, the Oamaru area has been relatively stable since then. The Strongly Deformed Region is that forming the Dunedin or Central Region and comprising the large promontories on either side of Otago Harbour lying east of the Moderately Deformed Region which here is bounded not by the coastline, but by a line almost co-linear with the general trend of the coastline, namely, the seaward (western) slopes of the Silver Stream and South Waikouaiti valleys. Deformation of the earth's crust since Miocene times within this region has been much greater than in the others, its relatively high mobility being marked by strong folding and faulting both on approximately N.E.–S.W. or E.N.E.–W.S.W. axes and on N.N.W.–S.S.E., and by strong regional downwarping and subsidence, all of which movements appear to have been in progress before, during and especially after the volcanic activity, as may be inferred from the data presented on Plate 36, figures 1 and 2, of Part I of these papers and fig. 1 of a previous paper (Benson, 1940). Tectonic Regions Petrographically Characterised. The Relatively Stable Region comprises about fifteen hundred square miles within the area considered. It is almost devoid of

Late Tertiary igneous rocks. A doubtfully recorded dyke at Locality 21, two small flow-residuals at Localities 27a and 28 and possibly a plug or flow at the (unvisited) Trig F, three miles west of Locality 28, are all that are known within Central and Western Otago, and together cover less than a fifth of a square mile. They comprise normal and slightly zeolitic olivine basalts. In the area around Oamaru at the north-eastern extreme of our map, though it contains a great development of probably older Tertiary to Late Oligocene basic tuff, sills, dykes and pillow-lavas and was strongly deformed prior to the Late Tertiary peneplanation, there are no known post-Miocene igneous rocks. The Moderately Deformed Region, as here defined, covers about nineteen hundred square miles. It contains almost all of the rocks dscribed in this series of papers. The eighty-seven localities from which the specimens have been derived contain breccias and tuffs, dykes, plugs and possibly domes of small size, at least one gravitationally differentiated sill (at Locality 17, to be described elsewhere), but chiefly residuals of formerly more extensive but relatively thin flows. There is rarely evidence of the occurrence of more than three or four flows in sequence, or of a total thickness of more than three hundred feet of igneous rock. The largest residual of a single sheet, the Waipiata doleritic basalt (Localities 64–70 and perhaps 71), covers nearly twelve square miles and may have been originally much more extensive. Next in area (six square miles) is the flow-complex on the summit of the Kakanui Range (examined by Brown and Marwick), which, however, contains one originally more extensive flow, outliers of which, at Localities 74 and 95 to the west and east of the main mass, are more than eleven miles apart. Similarly, Service (1934) has shown that several of the flows occurring in the relatively small residual masses in the Goodwood area had formerly a rather wide extent. Most of the remaining flow-residuals cover individually only a small fraction of a square mile. The total area of basaltic rocks now remaining in this region is approximately fortythree square miles, but their former extent may have been several times as great. Except in the cases mentioned, there is little evidence on which to base an estimate of the former extent of a flow indicated by the petrographical similarity of material derived from many isolated residuals; because, as yet, only a rapid reconnaissance-survey has been made over the greater part of this region. The contrast between the rarity of known dykes in this large, hastily-examined region, and their abundance in the more closely-studied Dunedin Central Region, though it will doubtless be lessened by further work, is probably real, and calls for caution against assuming an excessively wide extent of basaltic flows or much fissure-eruption in the Moderately Deformed Region. The close proximity of some sheet-residuals to known faults (not always in the fault-angle depressions) suggests that such faults may have been, or may have been near the channels of effusion (e.g. Localities 29–31, 35–36) and gives further hint of the only moderate maximum extent of the flows. It seems probable, however, that in general the lavas were poured out over surfaces usually of low relief, and that the chief fault-movements occurred after such effusions.

The variety of igneous rocks present in this region. while considerably greater than that among the few occurrences in the Relatively Stable Region, is nevertheless much less than in the Strongly Deformed Region. The following limits for the chief oxides present are shown by nineteen representative analyses by Mr. F. T. Seelye, which are tabulated below (Table V). Table I. Range of Compositions of Igneous Rocks of the Moderately Deformed Area. Per Cent. Per Cent. SiO2 40.01–47.69 Na2O 2.53–6.24 Al2O3 11.78–17.77 K2O 0.64–3.04 Fe2O3 + FeO 11.38–16.40 P2O5 0.17–2.24 MgO 3.15–10.06 TiO2 1.08–3.96 CaO 5.85–13.14 MnO 0.17–0.29 For obvious reasons it is not possible to estimate with any degree of probability the original volume of each type of rock present or the proportion between such volumes. The following table indicates, however, the number of localities out of the above total of 87 wherein the various types of volcanic rocks have been detected. Table II. Normal Basaltic Rocks. Localities Finely granular olivine and/or augite basalts are known at 30 Feldspar augite olivine basalts 8 Feldspar basalts 8. Coasely granular olivine-bearing doleritic basalts 22 Intersertal Dolerite 1 Ankaramite 6 — 75 More Alkaline Basic Rocks. Zeolitic, olivine-bearing doleritic and fine grained basalts 10 Mugearite 2 Atlantites and basanites 14 Olivine theralite 1 Olivine nephelinite 3 Limburgite 3 — 33 Since the normal olivine and olivine-augite basalts are probably present in greater volume proportion than the above figures indicate, and there are no analyses of the richly feldspathic basalt (apart from the mugearite), an average calculated from the analyses now available, taken in the above proportions, affords the nearest approach that can be made at present to the average composition of the igneous rocks in the Moderately Deformed Region, and may be substituted for the average composition of the rocks in this region previously published (Benson, 1941, p. 541, Table Ie) when only five analyses were available. The new average (a) is given in Table III below, together with (b) a crude average composition of the basaltic rocks of the Dunedin Central region.

Table III. (a) (b) Norm of (a) Norm of (b) SiO2 46.16 48.25 Orthoclase 7.78 8.90 Al2O3 15.32 16.11 Albite 21.48 28.82 Fe2O3 3.50 6.01 Anorthite 21.13 23.91 FeO 9.67 8.73 Nepheline 5.96 0.28 MgO 7.25 4.07 Augite 18.79 18.83 CaO 9.58 9.96 Olivine 14.80 6.15 Na2O 3.74 3.46 Magnetite 5.10 8.82 K2O 1.34 1.54 Ilmenite 4.56 4.41 TiO2 2.45 2.35 Apatite 1.34 1.01 P2O5 0.65 0.36 Plagioclase = Ab5+ Ab5+ MnO 0.21 0.21* One determination only. It must be emphasised, however, that as all but one of the basalts in the Dunedin region which are considered in the averaging were those from the Otago North Head Series analysed by Marshall (1914) in which the feldspathic basalts are rather more abundant, and the richly olivinic and augite-rich types are proportionally less abundant than is probably true of this region as a whole, the distinction between the average composition of the basaltic rocks of the Central and peripheral regions is probably less than here appears. The Strongly Deformed Central Region. Here, the conditions as to the range of petrographical character, chemical composition, and relative abundance of the igneous rocks is very different from that in the less deformed regions. Of the total area of a hundred and seventy square miles included within the Strongly Deformed Region, one hundred and ten are covered by Pliocene volcanic rocks; and the distribution of scattered outliers or residual boulders suggests that almost the whole of the area was at one time covered by volcanic rocks. The range of chemical composition indicated by the fifty-eight good analyses which are available, (including several by Mr. F. T. Seelye which are as yet unpublished), is shown by the following table. Table IV. Range of Chemical Composition of Igneous Rocks in the Dunedin Central Province. Per Cent. Per Cent. SiO2 44.00–66.04 Na2O 1.74–8.82 Al2O+ 12.92–20.80 K2O 0.46–7.09 Fe2O3 + FeO 0.12–11.00 P2O5 0.11* Lowest determined values, but probably in excess of amounts present in certain incompletely analysed trachytes.–1.68 MgO *0.12–11.18 MnO 0.14*–0.26 CaO 0.96–10.80 As shown in Section D of Fig. 2 in the first part of this series of papers, the basement schists and overlying Upper Cretaceous-Middle Tertiary Sediments in this region were folded into a series of anticlines and synclines with an eastwardly-increasing depth of depression prior to the Late Tertiary peneplanation, which cut a surface obliquely through these sediments and merging westward into approximate coincidence with the Cretaceous peneplain

cut in the schists. The presence of Miocene limestone truncated by the later peneplain under its lava-covering and locally exposed in the Otago Harbour area near Port Chalmers and Hooper's Inlet (see also Benson, 1940, Fig. 1) marks the region of greatest depression of the basement-formations within the Dunedin Central region. It is also the area where the major eruptions commenced, and about which subsequent eruptive activity appears to have been very vigorous as indicated by the concentration of a very varied series of dykes and immense breccia-filled vents in this vicinity, though many outlying centres of explosion and effusion were formed during the prolonged and diversified sequence of eruptive activities the details of which will be given elsewhere. An outline thereof has already appeared (Benson, 1934, 1941), and is here given in a slightly modified form. The earliest eruptive materials now occurring in situ are anorthoclase trachytes. They rose in small amount through several minor vents on the south-western and north-western margin of the Dunedin Central area, breaking through tectonically elevated areas, but were erupted in very large amount in the central originally-depressed Port Chalmers-Hooper's Inlet region, where trachytic agglomerates and tuffs associated with small trachytic flows and cut by very many trachytic dykes built up a low cone nearly two cubic miles in volume. In the centre of this mass, near Portobello, the trachytic agglomerate contains a few fragments of feldspar basalt and trachyandesite (?) the only indication of pre-trachytic effusions. Before the trachyte magma was exhausted, basaltic magma rose chiefly from a centre a few miles south of the main area of trachytic eruptions. Its expulsion commenced with the formation of massive agglomerate followed by finer grained tuffs and more widespread flows of very basic olivine-basalt, and olivine-augite basalt. The earlier members of this series of flows were invaded by the latest trachytic dykes, and among the later basaltic flows of this series is a kaiwekite which the writer interprets as a trachyte-basalt hybrid, in which the trachytic material is usually in excess. Occasional small flows of basanite, trachybasalt and trachyandesite, and rarely of phonolitoid trachyte, the “phonolite” of Flow No. 2 at North Head (Marshall, 1914), indicate a change of magmatic differentiation from a trachytic towards a phonolitic pole. Some crust-movement was in progress and the thinning out of the basaltic agglomerate and tuffs (in so far as it may not be due to its moulding against the trachytic cone) suggests the rise of two low anticlinal ridges running in an E.N.E. and N.N.W. direction respectively and intersecting a little south of Portobello. Here the movement was accentuated at the close of this eruptive phase by the intrusion of laccolites of phonolite and nepheline syenite porphyry, followed by immense explosions from a series of now breccia-filled vents along the axis of the N.N.W. anticlinal fold. The material ejected from these vents was carried far, especially towards the north, west and south-west, and accumulated to form massive mud-flow conglomerates locally more than a hundred

feet thick. Their distribution suggests that a synclinal warping then connected the sites of the present Otago Harbour and Taieri Plain synclines. The second major eruptive phase included a more widespread and diversified series of effusions than the first. The magma may have been thrust into separate reservoirs from which the differentiates, expelled from time to time, produced a complex sequence of interdigitating flows of varied basalts (preceded by basaltic agglomerates in the west and north), of phonolites (including the particularly widespread and voluminous Logan's Point and Waitati flows of what must have been very mobile magma), trachybasalts and trachyandesites, and olivine dolerite together with minor amounts of atlantite, mugearite and kulaite. The later products of this phase are chiefly in the western portion of the Dunedin region, and there are other features suggesting some elevation in the Port Chalmers area during this eruptive phase, as well as indications of a slight arching where is now the Flagstaff-Swampy Ridge. The second phase closed with another explosive eruption of fragmental, largely phonolitic, but also basaltic material from a vent now concealed beneath later flows near the summit of Mount Cargill, which seems to have been then a rising ridge. The greatest thickness of the water-borne detritus conveyed south-westwards from here occurs near the Lower Leith Valley, which seems to have then had a tendency towards synclinal depression. This tendency spread into the middle and upper Leith Valley during the immediately following third eruptive phase. Basaltic eruptions broke out near here, spreading in most directions, but particularly to the south-west over a much more restricted area than the basalts of the second eruptive phase. A little trachybasalt, atlantite and basanite occurs among these flows, which are thickest in the middle Leith Valley, where there is a noteworthy amount of a fine-grained basaltic tuff and “fire-fountain material,” like that of Halemaumau in Hawaii—the comparison is due to Dr. H. T. Stearns (priv. com.). Above these follows an extensive flow of cossyrite phonolite extending over six miles to the north-east from Mount Cargill overlapped by a related flow of olivine-bearing phonolite, rendered hybrid by absorption of basaltic material, and extending over eight miles to the south-west. Crust-movements became much more vigorous and the compressive movements greatly accentuated the anticlinal and synclinal ridges which, modified by erosion and partly drowned, constitute the present hills, valleys and embayments. The latest extrusions of magma accompanied these movements. A small plug of normal olivine basalt rose into a fault-plane breaking the sharp Mount Cargill anticline, and a phonolite-dyke invaded the reverse fault in the steep eastern flank of the asymmetric anticlinal Flagstaff-Swampy Ridge. Distribution of the Pliocene Volcanic Rocks in the Relatively Stable and Moderately Deformed Regions of Eastern Otago with Notes on the Localities from which were obtained the materials studied. Petrographical details will be given elsewhere supplementing the accounts by Marshall and others of the varied rocks, the sequence

of eruption of which within the unstable Dunedin Central Region has been sketched above. The remainder of this paper is concerned with the less varied characteristics of the hitherto little studied Pliocene igneous rocks in the regions peripheral to the Dunedin Central Region, which, on rather inconclusive grounds are thought to be probably coeval with the main (second) eruptive phase of this region. The brief statement of the types of rocks occurring at each specified locality will be followed by a summary of the general features of each petrographic type developed, and a table of chemical compositions thereof. The map (Plate 36 of Part I of these papers) illustrates the distribution of all known masses of Pliocene igneous south and southwest of the Shag River valley and the positions of over 100 localities outside the limits of the Dunedin (Central) Region from which rocks, which have been microscopically examined, were collected by Andrew, Benson, Dunne, Hutton, Marshall, Ongley, Paterson. Service, Turner or Williamson. The name of the collector thereof has been indicated by the initial letter of his surname in the list given below. In the case of slides lent by the Geological Survey for the purposes of this study, the letter “P.” is placed before the registered number of the slides, as is done in the Survey's official register of petrographical slides. Numbers without such prefix are those in the slide-register of the Geological Department of Otago-University. Some rocks are represented by slides in both collections. The gift to the University of the slides studied and described by Dr. Andrew, Paterson and Service is gratefully acknowledged. List of localities from which were derived rocks studied, their collectors, modes of occurrence, petrographic characters, registered slide numbers, and analyses as given in Table V:— 1. Mouth of Taratu Coal Mine, Kaitangata. (O.). Dyke. Intersertal dolerite. (P. 5560.) 2. Head of Waronui Creek. (O.). Dyke. Atlantite. (P. 5596.) 3. Cook's Head (or Cook's Nose), Tokomairiro Beach. (O.). Plug. Fine-grained zeolitic olivine basanite (or atlantite). (P. 4158.) Analysis No. 14. 4. Dunn's Quarry, two miles south-east of Milton. (O.) Plug. Olivine augite basalt. (P. 5561.) 5. End of Waronui Railroad. (O.) Dyke. Ankaramite. (5053, P. 4158.) 6. Head of Noble Creek, Akatore S.D.* S.D. = Survey District. (O.) Plug. Zeolitised basanite or atlantite with deuteric carbonates. (5048, P. 4163.) Analysis No. 16. 7. Head of Narrowdale Creek, five and a-half miles east of Milton. (O.) Dyke. Atlantite. (P. 4143.) 8. Capping of Table Hill, extending two to six miles north-west of Milton. (A.M.) Flow fifty feet thick resting on thin stratum of quartz-conglomerate above schist. Coarse-grained zeolitic dolerite-basanite (fide Marshall, 1918, pp. 63–4) and doleritic basalt. (5682.)

9. One and a-half miles west of Milburn Post Office. (O.) Flow resting on schist and dipping south-east. Doleritic basalt. (5050, P 5167.) 10. Road-cutting half a mile north of Akatore bridge. (B.) Dyke split into three narrow sheets invading sandstone adjacent to strong fault which brings sandstone down against semi-schist. Augite olivine basalt. (5073, 5074.) Note.—The semi-schist on the coast two miles south of this spot contains a lenticle of limestone showing traces of radiolaria and foraminifera (Benson and Chapman, 1938). 11. “Strains,” one mile and a quarter south-east of Milburn railway station. (A.O.B.) Sediments (sandstone, greensand and limestone), overlain by flow of limburgite (5054, 5072, 5693, P. 4139. Analysis No. 22), succeeded by flow of weathered dolerite; all steeply dipping against fault-plane. See Fig. 1. 12. “Kapiti,” a mile and a quarter east of Milburn railway station. (A.B.) Faulted limestone overlain by flow of limburgite (5069, 5694), covered by flow of ankaramite (5702). 13. Milburn Hill and Stony Knob. (A.M.B.) Sediments and limestone overlain on the southern side (above Milburn limestone quarry) by flow of limburgite (5095–6), followed by finely granular zeolitic basalt (5085, 5705, 5707–9, 5712 and ?5703, 5711, 5726), succeeded by zeolitic basanites of medium grainsize (5082–4, 5686, 5710, 5723) and capped by more finely granular iddingsitic zeolitic basalt (5086–7, 5704, 5706). See Fig. 1. 14. Cemetery Hill, half a mile north of Clarendon railway station. (A.M.O.) Zeolitic basalt (5689), resting on schist followed by less finely granular zeolitic basanite (5690–1). Analysis No. 17. See Fig. 1. 15. Summit of Trig. D, three and a quarter miles west-north-west of Clarendon railway station. (B.) Flow-remnant resting on quartz-conglomerate on schist. Olivine basalt (5075). 16. Summit of Trig. L, two miles north of Clarendon railway station. (A.O.) Flow-remnant resting on schist. Olivine basalt (5051, 5692, P. 4162). 17. Promontory of Waihola Hill Trig., one to four miles west of Waihola township. (M.O.B.) Flow of olivine basalt (5052, 5065, 5079, P. 4164, P. 4165), overlapping westward from Abbotsford Mudstone on to schist. Capped in promontory by flow of zeolitised atlantite (5070, 5071), from which it is separated by the westward tapering end of a gravitationally differentiated sill of olivine theralite (5060, 5061, 5064, 5068, 5698–5700) to be described in detail in Part 3 of these papers. Analyses Nos. 18, 19. 18. One hundred yards west of Waihola township. (O.) Plug. Olivine basalt (5791, P. 4149). 19. Near. Ferry Hill Trig., two miles north-east of Waihola township. (O.) Plug. Ankaramite (P. 5597).

20. North side of mouth of Taieri River. Collector unknown. Dyke. Very decomposed trachytoid feldspar basalt (5036). Note.—The fabric of the semi-schist invaded by this dyke has been investigated by Turner (1940, p. 53). 21. Waitahuna Hill. (M.) Pebble, occurrence in situ unknown. (Dyke?) Vesicular feldspar basalt (fide Marshall, 1918, p. 64). 22. Trig. O, four miles north of Waihola township. (O.B.) Stratified mass of coarse basaltic agglomerate (5715) at least 300 feet thick, dipping to the south-east at 30° and resting on schist. 23. A quarter of a mile north-east of the Factory bridge at Henley. (O.B.) Dyke. Invading Kaitangatan sediments. Ankaramite (P. 5598). 24. Trig. O, three-quarters of a mile south-south-west of Otokia railway station. (O). Plug invading Kaitangatan sediments. Ankaramite (P. 5576). 25. Beside railway line half a mile south-west of Allanton station. (O.) Plug? Zeolitic dolerite (P. 5577). 26. Ridge half a mile north-east of Allanton railway station. (O.) Flow dipping north-west and resting on Abbotsford mudstone. Very fine-grained basalt (P. 5584). 27. Summit of Mount Hyde, Trig. X, Mount Hyde S.D. (M.) Flow about 30 feet thick resting on thin layer of quartz-conglomerate on schist. Olivine basalt (5670). Note.—The structure of the schists in this neighbourhood has been described by Turner (1940, pp. 84–93, 185–188). 27a. On hill culminating in Trig. F, Mount Hyde S.D. (B.) Basalt flows in all about 200 feet thick resting on 100 feet of sandstone above schist. (i) At Trig. F, Olivine basalt (5023). (ii) A quarter of a mile west of Trig. F. Zeolitic basalt (5022). (iii) Half a mile south of Trig. F. Olivine basalt (5024). 28. Summit of Mount Stoker, Trig. G, Nenthorn S.D., nine miles east of Middlemarch. (B.) Columnar flow 100 feet thick resting on thin layer of sandstone above schist. Fine-grained olivine basalt (5646). 29. Yellow Hill, Trig. K, Nenthorn S.D. (B.) Flow over 200 feet thick resting locally on vesicular basaltic agglomerate above thin sandstone and conglomerate on schist. Fine-grained olivine basalt (5642). 30. Peat Moss Hill, one and a-half miles south-east of Trig. K, Nenthorn S.D. (B.) Flow about 100 feet thick resting on a few feet thick of sandstone, dipping gently westwards and faulted down against the basalt on Yellow Rock Hill. Finegrained olivine basalt (5657). 31. Hummock, Trig. D, Hummock S.D. (B.) Flow about 150 feet thick on a thin conglomerate on schist. A continuation of the Yellow Hill flow with down-faulted residuals of the Peat Moss Hill flow at its eastern foot. Fine-grained olivine basalt with abundant olivine nodules (5656).

32. Trig. HH, north-west corner of Waikouaiti S.D. (B.) Columnar flow about 80 feet thick dipping west-north-west at 10° and resting on thin ferruginous sandstone. Fine-grained olivine basalt (5669) comparable with that at Locs. 29, 30, 31. 33. Stony Hill, three-quarters of a mile north of Loc. 32. (B.) Flow (on sandstone?) on schist. Continuation of the flow on Trig. HH. Medium-grained olivine basalt (5662). 34. Hill a mile and a half north-east of Trig. HH, Waikouaiti S.D. (B.) At least two flows about 250–200 feet thick in all. Lower flows on the southern spur resting on thin Abbotsford Mudstone above thin sandstone and schist. Medium-grained olivine basalt comparable with that at Loc. 36, but a little richer in augite (5655). 35. The Ram Rock, half a mile S.S.E. of Trig. E, south of Hummockside S.D. (B.) Dyke probably rising along the fault plane running E. of Scratchback Hill and feeding the small dome (?) composing this mass. Nepheline basanite (5659). 36. Scratchback Hill, Trig. E, Hummockside S.D. (B.) Two flows in all about 300 feet thick, on north end of hill, resting on Abbotsford Mudstone above thin white sandstone on schist. Lower flow: Ankaramite containing xenoliths of schist (5660, 5667) covered by:—Upper flow: Coarsely granular olivine augite basalt with a little zeolite in groundmass and vesicles (5664, 5666). 37. Mount Watkin, Trig. F, Hawksbury S.D. (S.) Two flows together about 400 feet thick, resting on Abbotsford Mudstone and thin sandstone on schist. Lower flow trachytic feldspar basalt (2026), overlain by feldspar-augite-olivine basalt (2037–2038). 38. Derdan Hill, Trig. D, Hawksbury S.D. (B.S.) Flows resting on Caversham Sandstone and normal sequence of sediments with schist at base. Lower flow coarsely granular doleritic basalt about 100 feet thick (2014), invaded by dyke feeding a covering flow of trachytic feldspar basalt (2016–2020). 39. Hawksbury Hill, a mile and a quarter E.N.E. of Trig. D, Hawksbury S.D. (S.) Flow about 260 feet thick resting on Caversham Sandstone, etc. Pilotaxitic feldspar-augite-olivine basalt (2021, 2022). 40. Mount Mackenzie, Trig. J, Hawksbury S.D. (S.) Three flows 250–300 feet thick in all, resting on sandstone on schist. Lowest flow. Feldspar augite olivine basalt (2027, 2028), covered by porphyritic dolerite (2035), and capped by pilotaxitic feldspar-augite olivine basalt (2039). 41. Mount Trotter, a mile and a half north of Trig. J, Hawksbury S.D. (S.) Basaltic tuff (?) 100 feet thick resting on Abbotsford Mudstone, etc., and capped by porphyritic dolerite (2034) about 150 feet thick. 42. Middle Mount, two miles W.N.W. of Trig. Q, Hawksbury S.D. (S.) Flow about 200 feet thick on Abbotsford Mudstone, etc. Porphyritic dolerite (2031–2033).

43. Taieri Peak, Trig. R, four miles west of Palmerston. (P.) Plug (?) or flow-residual about 150 feet thick on sandstone on schist. Augite olivine basalt (5671, 5678). Also a dyke of vitric basaltic breccia (5681). 44. Mount Pleasant, Trig. Q, Hawksbury S.D. (S.) Two flows in all about 230 feet thick resting on thin Abbotsford Mudstone on basal sandstone, etc. An olivine augite basalt (2050) covered by flow of ankaramite (2045–9). 45. Mount Royal, Trig. T, Hawksbury S.D. (S.) Caversham Sandstone (with the usual sedimentary series below covered by basalts, etc., about 450 feet thick in all, comprising a basal flow of feldspar basalt (2023–5) succeeded by basaltic tuff, porphyritic dolerite (2029, 2030), basaltic tuff and, as the highest flow, a trachytic feldspar basalt (2040–2). 46. Bobby's Head, Trig. M, Hawksbury S.D. (S.) Goodwood limestone with normal sequence beneath, invaded by dyke of trachytic feldspar-augite-olivine basalt feeding a flow nearly 300 feet thick capping the limestone (2043, 2044). The margin of the dyke, deeply weathered and bleached against the invaded limestone, bears a considerable resemblance to some trachytic tuff in the Dunedin Central Region. The writer must take responsibility for Service's unchecked acceptance of the suggested comparison. Note.—All the above data concerning localities 37–42 and 44–46 are based on Service's (1934) paper. 47a. Puketapu, Trig. O, Moeraki S.D., one mile S.E. of Palmerston. (P.) Feldspathic basaltic breccia nearly 280 feet thick on Caversham Sandstone, etc., invaded by a dyke of slightly zeolitised doleritic augite olivine basalt [5680 = Paterson's (1941, p. 49) “analcite-bearing dolerite”] and overlain by a flow of olivine augite basalt about 100 feet thick, with more pyroxene in the lower (5675) than in the upper (5677) portion. 47b. Little Mountain, one mile south-east of Puketapu. (P.) Dyke of feldspar olivine basalt (5093) invading plug (?) of vitric basaltic breccia. 48. Smyler Peak, Trig. P, a mile and a half west-south-west of Palmerston, Moeraki S.D. (P.) Plug (?) invading Abbotsford Mudstone. A rather coarsely granular feldspar olivine basalt with strongly marked flow structure (5672–4, 5679 = P. 5988). Paterson's (1941, p. 49) “olivine dolerite.” 49. Janet Peak, Trig. V, two miles north of Palmerston, Moeraki S.D. (P.) Plug (?) or sheet-remnant about 100 feet thick rising through or resting on Burnside Mudstone, etc., capped by trachytic basalt tuff 100 feet thick. Medium-grained feldspar (olivine augite) basalt (5676, P. 3824). Note.—The rocks from localities 43, 47, 48 and 49 were collected and described by Paterson (1941). The structure of the underlying schist has been described by Turner (1940, pp. 169–180).

50. Small Conical Hill by Sheepwash Creek, three miles east of Middlemarch, Strath-Taieri S.D. (B.) Dome rising about 150 feet above the surface of the schist. Fine-grained olivine nephelinite (5649). Analysis No. 20. Note.—The structure of the adjacent schist has been described by Turner (op. cit., p. 180). 51. Small Cone half a mile south of Trig. G and five miles east-north-east of Middlemarch, Strath-Taieri S.D. (B.) Volcanic dome about 150 feet high or flow-remnant on schist. Mugearite (5651–2). Analysis No. 1. 52 Smooth Cone, Trig. G, five and a half miles north-east of Middlemarch, Strath-Taieri S.D. (B.) Dome or flow-remnant rising through or on schist. Porphyritic mugearite (5643). Note.—McKay's (1894) map-indication of the presence of sediments beneath the volcanic rocks at the last two localities has not been confirmed. 53. Slip Hill, Trig. J, five and a half miles east of Middlemarch, Strath-Taieri S.D. (B.) Flow dipping west and resting on sandstone above schist. Fine-grained olivine basalt (5648). 54. Hill a mile and a half south-east of Trig. J., Strath-Taieri S.D. (B.) Columnar flow about 50 feet thick resting on very thin sandstone about 100 feet below the local general level of the Late Tertiary peneplain. Carbonated olivine basalt (5645). 55. Bald Hill, Trig. H, five and a half miles east-south-east of Middlemarch, Strath-Taieri S.D. (B.) Flow on schist possibly separated therefrom by a very thin layer of sandstone. Finegrained olivine basalt (5658). 56. County road-metal quarry, Shark Hill, by Moonlight Flat, Budle S.D. (B.) Small extrusive dome rising through sandstone showing transition in quarry face from a lower mediumgrained hypocrystalline nepheline basanite (5653) into an overlying marginal obliquely columnar phase of very finely granular zeolitic augitite (5647). 57. Station Hill, five chains south-east of Trig. S in the S.E. corner of Rock and Pillar S.D. (W.) Dyke (?) or small flow-residual in or on sandstone. Ankaramite (158 = P. 3630). (See Turner in Williamson, 1939, p. 67.) Note.—This mass of igneous rock is not indicated on the official map. (Williamson op. cit., Sheet 7.) 58. Highlay Hill, Trig. GS, Highlay S.D. (W.) Flow on sandstone. Fine-grained zeolitic atlantite (193 = P. 3646). (Described as fine-grained olivine basalt by Turner, loc. cit. supra.) 59. Donaldson's Coal Pit, 60 chains north-west of Trig. G, Rock and Pillar S.D. (W.) Flow resting on sandstone. Zeolitic atlantite described as zeolitic olivine basalt by Turner (loc. cit. supra) (168 = P. 3639). See Analysis No. 12. 60. Hyde, one and a half miles north of railway station and thirty chains east-north-east of Trig. G, Rock and Pillar S.D. (B.) Flow on sandstone 50 ft. thick. Fine-grained olivine (augite) basalt (5665).

61. Tiroiti. (a) Half a mile east-south-east of railway station and seventy chains north-west of Trig. F, Rock and Pillar S.D. (B.) Flow on thin sandstone, the lowest and marginal portion of an uninvestigated flow-complex possibly surrounding vent (fide Williamson, 1939, p. 64). Olivine basalt with few phenocrysts (5661). (b) Thirty chains north-west of Trig. F. (W.) Probably overlying (5661) and related to Waipiata doleritic basalts (Locs. 63b–70), and marking their eastward limit. Coarse-grained doleritic olivine basalt (153 = P. 3658) described by Turner (loc. cit., p. 66). (c) Two and a-half miles north-east of railway station on boundary of Rock and Pillar S.D. (B.) Flow on sandstone. Fine-grained olivine basalt (5668) comparable with that at Locs. 60 and 61a. 62. Tiroiti. Three miles east of railway station, Trig. O, Rock and Pillar S.D. (W.) Flow on sandstone. Medium-grained olivine-augite basalt (P. 3651). 63. Kokonga. Two miles east-north-east of station and 72 chains west-south-west of Trig. O, Swinburn S.D. (W.) Flow on sandstone. Analysis No. 15. (See Williamson, 1939, p. 65.) Too decomposed for precise determination, but suggestive of zeolitised atlantite. No specimen available for microscopic study. 64. Kokonga. (a) Half a mile east of railway station. (B.) Massive flow of coarse doleritic olivine basalt (5663). (b) Exact locality uncertain. Large. “Olivine nodule” from basalt collected and presented by Mrs. N. Kingston, M.Sc. (5100), to be described in structural detail elsewhere by Dr. F. J. Turner. 65. Kokonga. On main road two and a-half miles west of the railway station. (B.) Massive flow similar to 64 (a). 66. Kokonga. On main road three and a half miles west of railway station. (T.) Massive flow as at 64 (a) and 65 (5055). 67. Kokonga. (a) Three miles south-west of station, 40 chains north-west of Trig. L, on southern boundary of Maniototo S.D. (W.) Massive flow coarse doleritic olivine basalt (162 = P. 3598). (See Turner in Williamson, 1939, p. 67.) Analysis No. 3. (b) At Flat Cap = Trig. L. (W.) Massive flow on Miocene (?) sand, etc. See Analysis No. 4. Probably the Waipiata doleritic olivine basalt. Specimen not available. 68. Ranfurly. Main road six miles east of railway station. (B.) Northern extreme of the Waipiata doleritic olivine basalt on Miocene (?) sand, etc. (5654). Comparable with rock at Locs. 61 (c), 64–70. 69. Waipiata Road. One mile south-south-west of railway station. (B.) Massive flow as at Loc. 61 (c), 64–70 (5057). 70. Waipiata Road. Three miles south-south-west of railway station. (B.) Massive flow as at Locs. 61 (c), 64–69.

71. Haughton Hill, 50 chains west of Trig. A, Gimmerburn S.D. (B.W.) Massive flow on Miocene (?) sands, etc. Rather coarse-grained almost doleritic olivine basalt probably related to the Waipiata doleritic olivine basalt (308 = P. 3638). See Analysis No. 5. Note.—(a) The tentative suggestion (Benson 1935, fig. 4.B.) that this mass forms part of a flow rising from beneath younger alluvium is now withdrawn. Note.—(b) This is the westernmost mass of Pliocene basalt seen by the writer. McKay's (1884, pp. 65–66) hitherto unconfirmed comments suggest the presence of coeval (or younger?) basalt breaking through Tertiary gravel and lignite in the Upper Manuherikia Valley (near St. Bathans?) about 12 miles further west. Numerous large pebbles of both coarse and finely granular pitted or smooth basalt have been found by J. D. Raeside (priv. com.) during his recent soil-survey of the terrace-gravels on the western side of the Dunstan Range, extending as far south as Cromwell, where Park (1908, pp. 15, 19, 44) has recorded their presence, and along both sides of the Manuherikia Valley as far south as Alexandra and Clyde, where again Park (1906, pp. 46–47) has recorded with petrographic description the presence of olivine basalt pebbles. It would appear probable that flows of basalt, not necessarily of great extent, had recently been stripped off the Dunstan and Raggedy Ranges, unless Park's (1908, p. 34) view be accepted that the pebbles were brought by glaciers from the ranges north and west of Lake Hawea. A small residual patch of basalt separated by a thin layer of sediments from the underlying schist has recently been found by Raeside, near Crawford Hill, on the southern end of the Raggedy Range, six miles east by north of Alexandra. Thus the effects of the Late Tertiary igneous activity in Eastern Otago have extended into Central Otago to a distance of over 70 miles north-west of Dunedin. 72. Swinburn S.D. 100 chains south of Trig. C. (W.) Flow on quartz conglomerate. Olivine basalt (?). No specimen available. See Analysis No. 7. 73. Swinburn S.D. By main road one and a-half miles north-north-east of Trig. I. (T.) Flows on quartz conglomerates. Lower flow coarsely ophitic zeolitised olivine dolerite (5059). Upper flow rather less coarsely granular basanite (5058). 74. Swinburn S.D. Hill west of Pigroot Hut, 94 chains south-east of Trig. L. (W.) Flow on quartz conglomerate. Feldspathic olivine basalt (?). No specimen available. See Analysis No. 8. 75. Swinburn S.D. Beside road immediately west of Round Hill. (H.) Flow (?). Zeolitic basalt (5035). See Fig. 2. 76. Swinburn S.D. The Brothers, near Road. (T.) (a) Flow resting on sandstone forming summit of South Brother. Finely granular basanite with a little glass (5076). (b) Horizontally jointed dyke 100 yards west of road. Rather coarsely granular nepheline basanite (5056, 5081). 77. Highlay S.D. (a) Near Trig. H, in north-east corner of the S.D. (B.) One of the highest of several massive flows resting on quartz conglomerate. Feldspar basalt (5754 = P. 5978). (b) (W.) The erroneously localised specimen noted by Williamson (1939, p. 65) was probably derived from this hilltop, Trig. H. Fine-grained iddingsitic olivine basalt (241 = P. 3636). (See Turner in Williamson, p. 67.) See Analysis No. 9.

78. Green Valley. By road a quarter of a mile north-west of Waihemo School, Waihemo S.D. (T.) Flow resting on greywacke-semi-schist. Medium-grained olivine basalt (5082). 79. Green Valley. On ridge separating upper portion of Green Valley from the Shag Valley, three-quarters of a mile west-south-west of Waihemo School. (B.) Thick flow resting on Duntroonian limestone, Bortonian and basal sandstones on schists. Rather coarse granular olivine feldspar augite basalt (5644). (Cf. 5082 above.) 80. Happy Valley Ridge, three-quarters of a mile north-west of Trig. F, Waihemo S.D., between Happy and Shag Valleys. (B.) Flow resting on limestone, etc., as at Loc. 79. Medium-grained basalt (5650). A little richer in olivine and augite than the otherwise comparable rocks (5082, 5644). ‡ Localities 81–87, 96–99, 101–104; yield older Tertiary volcanic rocks only.88. Kattothyrst. On southern margin of Kakanui S.D. (Br.)* Collector D. A. Brown. Residual mass of flows totalling 300 feet thick resting on semi-schist. Ankaramite (5746 = P. 3823). Hypocrystalline atlantite (5767 = P. 6039). 89. “Crater” on northern boundary of Waihemo S.D. (Mar.)†† Collector Dr J. Marwick. Mass of similar thickness of flows on semi-schist. (a) At Crater, 57 chains at 231° from Trig. C. Zeolitised sanidine bearing nephelinite (5768 = P. 6038). (b) On track a mile and a half to south-south-west of Crater and one mile E. of Trig. B, Maheno S.D. Feldspar basalt (5769 = P. 6040). 90. Siberia Hill. Trig. C, on south margin of Kakanui S.D. (Mar.) Portion of same group of flows. Anorthoclase-rich biotite atlantite (5766 = P. 6036). Analysis No. 11. 91. Siberia Hill. Beside track 75 chains at 112° from Trig. C. (Mar.) Part of same series of flows. Feldspar basalt (5765 = P. 6005). 92. Trig Island. Beside track 50 chains north of Trig. L, on south margin of Kakanui S.D. (Mar.) Portion of same flow series 300 feet thick in all—olivine augite basalt (5764 = P. 6004). 93. Near Mount Difficulty. Beside track 80 chains west of Trig. D, Kauroo S.D. (Mar.) Portion of same flow series resting on semi-schist as at Locs. 90–92. Olivine augite basalt (5763 = P. 6003). Note.—The writer is greatly indebted to Dr Marwick and Mr Robert Gray, of the Dasher Run, for the collection of specimens from Localities 89–93. 94. Mount Difficulty. Trig. D, in south corner of Kauroo S.D. (B.) Basalt flow residual or plug (?) rising 450 feet above quartz sandstone up to 200 feet thick. Feldspar olivine basalt (5762 = P. 6002). Analysis No. 10.

95. Obelisk or Charles Peak. North-west corner of Otepopo S.D. (Mar.) Flow residual or plug rising 75 feet above quartz sandstones 60 feet thick. Olivine augite basalt (5747 = P. 3807). 100. Kauroo Hill. (B.) See Fig. 3. Middle Tertiary dolerite (5719). Late Tertiary doleritic basalt (5738, 5789) and iddingsitic nephelinite (5714–7–8, 5739, 5761, 5770, 5788). Plug of very finely granular atlantite (?) (5737). (ii) Petrographical Nature and Chemical Composition of the Late Tertiary Igneous Rocks in the Relatively Stable and Moderately Deformed Regions around the Dunedin Central Regions. Mugearite:Localities 51 and 52. Analysis No. 1. Table V. This uncommon type of basaltic rock, recently recognised near Dunedin (Benson and Turner, 1940), forms two small, conical hills, rising above a ridge of schist five and a-half miles north-east of Middlemarch. Both phases of mugearite found near Dunedin are represented here. The porphyritic phase (5643, Loc. 52) resulting probably from marginal chilling, is a rather dark grey, finely granular rock without marked fluidal structure, resembles (246), the analysed mugearite from Jeffrey's Hill, Dunedin, and contains rather abundant small (< 0.6 × 0.1 mm.) seriate phenocrysts of andesine (An38) with occasional marginal oligoclase, olivines (< 0.3 mm.) usually elongated parallel to Z and rather decomposed, and octahedra (0.05–0.2 mm.) of magnetite in an extremely fine-grained, feebly trachytic matrix of more or less granular feldspar, prismatic diopside, apatite and dust-like magnetite. The less markedly porphyritic and less finely granular phase (5651, 5652, Loc. 52), an analysis No. 1 of which is here given, is more like the Scottish examples of this rock, especially that from Eigg, both in the more marked flow-structure and the greyish-green colour. (See Harker, 1908, plate vii, fig. 2A.) The seriate phenocrysts (<0.2–0.3 mm.) of feldspar are zoned andesine with a core of An45–52, and rarely a little sanidine may be seen mantling them. Olivine (<0.4 mm.) less altered than the above and similarly elongated, and magnetite octahedra (0.1–0.2 mm.) are also abundant, though the last is not so common in the matrix of this rock as of (5643). Apatite occasionally forms relatively large (0.15 × 0.05 mm.) prisms. The composition of the rock is very similar both to that of the Dunedin mugearite and of the type rock from Skye, the former of which has been given (Table V, No. 2) for comparison. Feldspar Basalts: Under this heading are grouped the more feldspathic of the rocks in the Goodwood area described by Service (1934) as feldspar basalts or trachytic feldspar basalts, together with other flows in the Shag Valley and on the Kakanui Range. They are characterised by the predominance of plagioclase both in phenocrysts and in ground-mass, with which minor amounts of phenocrystic olivine, magnetite and titanaugite may or might not be present, and they possess more or less well-marked flow-structure. The plagioclase tabulae are usually not more than 1 mm. long. Their composition varies—about An62–55 for the bulk of the crystals though a narrow

andesine marginal zone An40 is sometimes present. Rarely a prism of labradorite has grown around a corroded andesine core. The phenocrysts of the coloured minerals are seldom larger than 0.5 mm. (olivine or augite) or 0.2 mm. (magnetite). The ground-mass texture varies between that of the very finely granular with marked flow structure and dust-like magnetite to less finely granular with a more or less pilotaxitic arrangement. A little sanidine may be detected between the plagioclase laths in most rocks of this group and small flakes of biotite are commonly present, usually near the olivine. Among these rocks may be placed those described by Service from Loc. 37 (2026), Loc. 38 (2016–2020) and Loc. 45 (2025, 2040). One (5754 = P. 5978) among the highest flows on Loc. 77 is very finely granular with marked flow-structure and a little flow-brecciation. It is noteworthy for the presence of irregularly ovoid patches less than 5 mm. in diameter containing very pale brown glass with some marginal subradiating tufts of brown microlitic augite (?). A rock (5769 = P. 6040) similar to this (except for the absence of glass and the presence of biotite), occurs at Loc. 89 (b), and forms that portion of the basalt-cap on the Kakanui Range nearest to Loc. 77, including 5765 (= P. 6005) at the adjacent Loc. 91 and 5762 (= P. 6002, Analysis No. 10* The presence of phenocrystic olivine in this rock makes it a type transitional into the next division.) at Loc. 94, a point a mile beyond the eastern end of the long strip of basalts covering this range. Basalts of this character occur among the flows analysed by Marshall (1914) forming the North Head of Otago Harbour, their relative abundance there being partly responsible for the difference between the average compositions of the basalts of the Dunedin region and those of the peripheral districts of Eastern Otago already noted. Basalts containing abundant Phenocrysts of Feldspar, Augite and Olivine: These comprise most of the rocks classed by Service (1934) as feldspar basalts or trachytic basalts, the difference being chiefly in the greater abundance of the coloured phenocrysts, which are, however, never in excess of the plagioclase. As this relative abundance of phenocrysts may depend on the varying factors controlling crystal-sorting prior to eruption, it is not impossible that different portions of the same flow may be grouped in different classes. Among these are rocks from Loc. 37 (2036, 7, 8); Loc. 39 (2021); Loc. 40 (2027–8, 2039); Loc. 45 (2040–1–2); Loc. 46 (2043–4); Loc. 47 (5093); Loc. 48 (5672–4) and (5679 = P. 5988); Loc. 49 (5676 = P. 3824). Probably the material of the so-called “trachytic tuff” of Locs. 46 and 47 is largely composed of weathered and bleached rock-fragments of this character. Normal Olivine or Olivine-Augite Basalt: Under this heading are grouped basalts of the type most abundant in the region discussed, those with a moderately fine grain-size and more or less well-marked flow-structure containing in varying proportions phenocrysts of olivine and/or titanaugite, with or without relatively subordinate phenocrysts of plagioclase, but without such features suggestive of alkalinity as the presence of sanidine, biotite or zeolites. The occurrence of richly olivinic cognate xenoliths is common among some of

these lavas (notably at Locs. 29, 32 and 36), and so also is the presence of more or less digested accidental xenoliths of quartz-schist and quartz showing all stages of reactive absorption. Often the similarity of rocks in several adjacent localities suggests they are portions of a continuous flow. Where this is the case the groups of localities concerned are, in the following list, separated by dashes from those of rocks which were derived from different flows: Loc. 4 (P. 5561); Loc. 10 (5073–4); Loc. 15 (5075); Loc. 16 (5051, 5692, P. 4162); Loc. 17 (5052, 5065, 5079, P. 4165, a flow overlain by an olivine-theralite sill); Loc. 18 (5791 = P. 4149); Loc. 20 (5036); Loc. 24 (P. 5576); Loc. 26 (P. 5578); Loc. 27 (5670); Loc. 27A (5023, 4), — Loc. 28 (5646, with strongly zoned feldspar phenocrysts An62–40 and a little biotite),—Loc. 29 (5642), Loc. 30 (5657); Loc. 31 (5656), Loc. 32 (5669), Loc. 33 (5662), Loc. 34 (5655)—; Loc. 36 (5660), Loc. 37 (2037–8), Loc. 43 (5671, 5678), Loc. 44 (2050); Loc. 45 (2041), Loc. 47 (5675, 5677); Loc. 53 (5648); Loc. 54 (5645); Loc. 55 (5658);—Loc. 60 (5665); Loc. 61 (5661, 5668)—; Loc. 62 (P. 3651), Loc. 77? (P. 3636, Analysis No. 9); Loc. 92 (5674 = P. 6004); Loc. 93 (5673 = P. 6003); Loc. 95 (5747 = P. 3807). The usual minor variations of texture and relative proportions among the phenocrysts and the minerals of the base may be seen among these rocks as well as variations in the manner and degree of alteration. The extent of replacement especially of the olivine by carbonates is noteworthy in some examples. Normal Basalts with few or no Phenocrysts and medium to rather coarse grain-size: These occur at a few places, including Loc. 25 (P. 5584), Loc. 48 (5672-3, -4, -9), the “olivine dolerite” of Paterson (1941), Loc. 49 (5676), and Loc. 62 (P. 3651). Coarsely Granular Doleritic Basalts: These include a group of rocks differing from ophitic dolerites in the possession of a granular to sub ophitic texture. The plagioclase (An50–60) forms large (about 1.0 mm.) tabulae and stout laths. The pyroxene is strongly titaniferous and has a large optic axial angle. The olivine is richly forsteritic (Fa0–22), and the iron ore platey ilmenite rather than magnetite. Acicular apatite is abundant. In some rocks a little zeolite is present though insufficient to carry them into the group of teschenites. Variation occurs in the degree to which a porphyritic texture is developed through the presence of phenocrystic olivine and/or titanaugite. The most extensive single flow in Otago, the Waipiata doleritic basalt, is indistinguishable from the Roslyn dolerite of the Dunedin district either on chemical or mineralogical grounds and, like it, is probably less than a hundred feet thick. It originally covered about 90 square miles, and if, as their petrographic features suggest, rocks at Locs. 62 and 71 are outliers of this flow, it may originally have had three times as great an extent. In the following list occurrences which were probably portions of a single flow are placed between dashes as above. Loc. 8? (Marshall, 1918); Loc. 9 (5050); Loc. 14 (5697); Loc. 38 (2014, possibly an outlier of a flow in the north-east of the Dunedin district); Loc. 36 (5664, 5666);— Loc. 40 (2035); Loc. 41 (2034); Loc. 42 (2031,–2,–3); Loc. 45 (2029, 2030), a porphyritic flow with large phenocrysts of augite and rarely

a little zeolite in the ground-mass and vesicles (e.g. 5666); Loc. 47 (5680);—Loc. 61 (P. 3658); Loc. 64 (5663); Loc. 65 (a weathered hand-specimen); Loc. 66 (5055); Loc. 67 (P. 3598 and Analyses. Nos. 3 and 4); Loc. 68 (5654); Loc. 69 (5057); Loc. 70 (as 65); Loc. 71 (P. 3638, and analysis No. 5);—Loc. 78 (5082); Loc. 79 (5644 = P. 5983); Loc. 80 (5650), three coarsely porphyritic augite-olivine basalts with a matrix of medium grain-size—; Loc. 100 (5789, iddingsitic). The similarity between the Waipiata and Roslyn doleritic rocks suggests that at least part of the basic rocks in the moderately deformed regions were erupted during the second major phase of the igneous activity in the Dunedin district of which the Roslyn dolerite was a product. Intersertal Dolerite: A single instance of this rock-type occurs at Loc. 1 (P. 5560). It contains strongly zoned titanaugite (< 1.5 mm.) and subordinate decomposed olivines, largely decomposed rhomboids of ilmenite and tabulae (< 0.2 × 0.05 mm.) of labradorite fraying out into oligoclase (?) microlites diverging into the matrix of partially decomposed pale brown glass in which are slender prisms (< 0.08 mm. long) of basaltic hornblende arranged in parallel groups and associated with minute plates of ilmenite. Ankaramite:Under this heading are grouped rocks with about 50% of titanaugite in phenocrysts and matrix, together with approximately equal amounts of olivine and labradorite with a little magnetite biotite apatite and deuteric(?)carbonates with or without a little interstitial glass. The more finely granular examples are difficult to distinguish from atlantites in which nepheline occurs in association with plagioclase. Specimens were obtained at Loc. 2 (P. 5596); Loc. 5 (5053, P. 4166); Loc. 14 (5072); Loc. 19 (P. 5597, abnormal in its very minute grain-size, the occurrence of olivine as the only phenocrysts, the presence of picotite and of irregular vesicles filled with aragonite(?) and opal); Loc. 23 (P. 5598); Loc. 24 (P. 5576); Loc. 44 (2045–9); Loc. 57 (P. 3630); and Loc. 88 (5746 = P. 3823). A peculiar variety of ankaramite at Loc. 12 contains seriate phenocrysts of titanaugite (< 1.5 mm.) rarely with a pale greenish core, often grown around and through brown hornblendes (now resorbed with production of magnetite and rhönite (?)), or around iddingsitised olivine, set in a fine-grained, almost panidiomorphic matrix of titanaugite, labradorite, iddingsitic olivine and magnetite. Zeolitic Dolerites and Doleritic Basalts: The most coarsely granular of these rocks is an ophitic dolerite at Loc. 73 (5059), in which the plates of titanaugite may be as much as 4 mm. long. The radiating zeolite either replaces labradorite with augite or occurs interstitially between the feldspar tabulae. Less coarsely granular is a rather analogous rock (5738) at Loc. 100. Paterson's (1941) “analcite dolerite” (5680) from Loc. 47 is still less coarsely granular, but its abundant interstitial zeolite is not analcite but an aggregate of minute weakly birefringent grains, and the same may be the case in regard to the “teschenite” (P. 5577) of Loc. 25, in which the zeolite not only replaces labradorite, but may form distinct veinlets. Dr Andrew's sample (5682) of the flow capping Table Hill

(Loc. 8), though not containing recognisable nepheline, may be a zeolitic representative of the “doleritic basanite” which Marshall (1918) reports here. Figure I. Geological Sketch Map of the Milburn-Clarendon District, showing by numbers the localities of petrographically studied rocks. Zeolitic Basalts: In this group are placed those basaltic rocks in which, though there is no indication of the former presence of nepheline, there is a considerable amount of scattered zeolite, apparently a primary magmatic constituent and not merely the result of deuteric alterations of feldspars or the filling of vesicles. It is convenient to distinguish between intersertal zeolite basalts in which the characteristic mineral occurs in apparently primary granules between the feldspars and those basalts in which there are irregular veinlets or segregations of zeolite associated with or including crystals of augite larger and more idiomorphic in form than those in the

surrounding ground-mass as if crystallising in a micropegmatoid (see fig. 2)—an association which the writer described in a basalt from New South Wales (Benson, 1915, p. 618). Both modes of occurrence are found in Eastern Otago, though the former is more abundant, especially in the Milburn-Clarendon district, of which a geological sketch map, based on a partial revision of that of Andrew (1906), is hereto supplied. (See fig. 1.) Only general comments may be given here as detailed studies of these zeolite-bearing rocks are being made by Dr Marshall. The basal lava (except where the limburgite occurs), is a finely granular olivine basalt usually without phenocrystic augite and containing but little or no flaky biotite. Acicular apatite is noteworthy, and there is usually present clear intergranular often isotropic zeolite (analcite?). Slides 5085, 5705, 5707–9, 5712 illustrate this type of rock, which may be traced all round Milburn Figure 2. Margin of a zeolitic segregation containing aggregated and scattered crystals of titanaugite and magnetite larger than those in the groundmass of the enclosing basalt. Slide 5035. Rock from roadside, west of Round Hill, Upper Shag Valley. Hill. Occasionally (5703, 5711) zeolites cannot be recognised with certainty. Further, in place of the minutely tabular plagioclase seen in these rocks there may be poikilitic grains of feldspar in which, however, the augite and magnetite grains are as small as those in the ground-mass of the rock (5684,–5,–7,–8,–9), and in a comparable but less zeolitic rock of Cemetery Hill (P. 4160), a little nepheline seems also to be present. Thus with increasing grain-size and content of biotite these rocks become difficult to separate from the overlying zeolitic basanite. The higher parts of Milburn Hill, above the basanite are composed of a fine-grained zeolitic iddingsitic basalt (5086, 5704, 5706).* Note—Andrew (1906) found such evidence of transition between the several types of igneous rocks on Milburn Hill that he mapped and described the whole complex as a single unit. During the revision no exposure of weathered interflow surfaces or tuff-beds could be found, but there seems sufficient petrographical evidence to warrant, at least tentatively, the division here suggested, apart from the improbability that so great a thickness (over 300 feet) of lava should consist of a single flow. The basal limburgite is clearly distinct from the zeolitic basalt which overlaps it. The latter shows fairly uniform petrographic characters in the lowest portion of the complex all round Milburn Hill, with the possible exception of the eastern extremity. Doubt arises in the case of other rocks and especially the basal volcanic rock on Cemetery Hill, which show features transitional to the more coarsely granular median zeolitic basanite, and these may actually be the rather finely granular marginal portion of the latter, if the basanite here overlaps the basalt to lie directly on the schists below Cemetery Hill. The petrographical similarity between the median basanite and the iddingsitic zeolitic basalt which caps Milburn Hill is no less marked, but the occurrence on the eastern slopes of the hill of a very finely granular flow-brecciated basaltic rock (5087) affords some indication of the presence of an interflow surface separating these two into more or less independent flow-units. A slightly zeolitic basalt

occurs at Waihola Hill, Loc. 17 (P. 4164) and at Loc. 27A (5022). That at Loc. 75 (5035) is particularly interesting in the presence in an otherwise normal olivine augite basalt, of large, irregular, ovoid patches of finely crystalline indeterminate zeolite, enclosing abundant small, aggregated prisms and small phenocrysts of titanaugite, octohedra (< 0.2 mm.) of magnetite, a few zoned tabulae of labradorite, and prisms of apatite. Figure 2 illustrates portion of the margin of one of these zeolitic patches about 5 × 3 mm. in area. Atlantite or Basanite (sensu lato): Lacroix (1927, p. 20) in 1916 coined the term ankaramite to designate strongly melanocratic basalts in which augite predominated over olivine (in contrast with oceanites in which the reverse relation holds), and remarked that the term ankaramite-basanite might denote the melanocratic rocks in which nepheline was associated with the plagioclase. Lehmann (1924) had, however, introduced the term atlantite for just this purpose, and as it has been found useful in the Dunedin Central area, it will be employed here. The following are brief notes on rocks of this character in the localities listed herein:— Loc. 2 (P. 5596). A rock on the boundary between normal basanite and atlantite. The phenocrysts of olivine (< 0·20 mm.) are commonly changed to carbonates; those of titanaugite (< 1·0 mm.) may have corroded green cores. The matrix of minutely prismatic (< 0·1 mm.) titanaugite, tabular (< 0·1 mm.) or poikilitic (< 0·4 mm.) labradorite (Ab50), and very subordinate nepheline contains also magnetite, acicular apatite, a little biotite and much secondary carbonate. The rock (P. 4158) at Loc. 3 is less markedly porphyritic, has a less purely crystalline matrix and is much fresher. (See Analysis No. 14.) That of Loc. 6 (P. 4163, Analysis No. 16) is comparable with it, as also is (P. 4143) from Loc. 7, though the last is richer in nepheline showing incipient zeolitisation, and the olivine is almost completely replaced by bowlingite. A more coarsely granular rock at Loc. 8 has been described by Marshall (1918, pp. 63–4) as a doleritic basanite. Loc. 13 (see Figure 1). A considerable extent of rock of medium grain-size with frequent development of poikilitic labradorite and nepheline, and more or less abundant zeolite forms the middle portion of Milburn Hill, and the mass at Cemetery Hill, Clarendon, described

by Marshall (1912). It is less finely granular than the zeolitic basalt, has abundant small (± 0.5 mm.) phenocrysts of titanaugite as well as of olivine (0·5 mm.) and magnetite (< 0 1 mm.) set in a matrix composed of tabular labradorite-andesine, subidiomorphic prisms or irregular grains of nepheline, augite and magnetite with a little flaky biotite and apatite. There are many irregular veinlets or patches of a zeolite with optical properties and twinning resembling those of phillipsite. Examples of this rock are 5082–3–4, 5686, 5710, 5723, and more coarsely granular upper portion of the igneous capping on Cemetery Hill (Loc. 14, 5690–1). The samples partially analysed by Andrew (1906, Nos. 1, 2 and 3* Andrew noted that the powder of these rocks gelatinised on treatment with acid. Marshall noted the presence of nepheline and of the zeolites, the nature of which he is investigating.) and by Marshall (1912, B) and that from Cemetery Hill completely analysed by Seelye (Analysis No. 17 herewith, cited through generous permission of Dr. Marshall) may be considered representative of this rock-mass. At Loc. 17 (5070–1) the flow above the olivine theralite sill is more finely granular and otherwise different from these. The dark minerals are unevenly distributed; a little analcite and thompsonite (?) occurs interstitially, and in irregular micropegmatoidal segregations of granular or poikilitic nepheline and labradorite, in which are minute idiomorphic crystals of titanaugite, apatite and magnetite, with little segregations of sanidine (or anorthoclase) and biotite-flakes. At Loc. 56 a quarry has been cut in what appears to have been a low dome which displays two very distinct phases of basanitic rock. The lower apparently medium-grained rock (5653) proves under the microscope to be a hyalobasanite in which phenocrysts (< 1.0 mm.) of olivine and of titanaugite (< 0·3 mm.) with prisms (1·0 × 0·5 mm.) of nepheline and plagioclase (An48) containing innumerable small crystals of titanaugite and magnetite are embedded in a matrix of dark tawny brown glass with similar small grains of the last two minerals locally grouped into a streaked arrangement continuing alike through the glass and the colourless minerals. Under high illumination the glass shows abundant rod-like deep brown crystallites in parallel bundles sometimes radiating from the augite-grains. Carbonates and a little chlorite occur as secondary minerals. This rock passes up into an aphanitic dark grey slightly vesicular rock (5647) with a columnar structure developed on a small scale obliquely to the present land surface. It has some of the features of augitite. The phenocrystic olivine is partly changed to iddingsite, the titanaugite is as before, but no colourless minerals are present save for a little zeolite lining small openings. But because of the greater development of the rod-like brown crystallites, the intervening glass is almost colourless, though patches of dark brown glass remain free of such microlites. A stage in crystallisation following on that reached by (5653) is displayed by (5659) from Loc. 35, a rock the magma of which rose through the fault-plane east of Scratchback Hill and flowed eastwards over the thin layer of Abbotsford Mudstone and sandstone capping the schist. The rock was collected from

the western end of the fault where it forms small oblique to horizontal columns. It is almost holocrystalline with abundant seriate phenocrysts of zoned titanaugite and olivine (0·5 mm.) and magnetite octahedra, the smaller crystals being included in poikilitic more or less idiomorphic crystals or grains of nepheline (0.8 mm. long) are grouped with microlites of labradorite with a little interstitial zeolite. A very little brown glass remains, notably in the centre of a globular aggregate (2·0 mm. in diameter) of small (0·2 mm.) idiomorphic augite prisms surrounded by an outer shell of close-packed augite-granules. The zeolite in the vesicles has some of the features of phillipsite. At Loc. 58 a rock (193 = P. 3946) previously termed (Turner, in Williamson, 1939, p. 67) a zeolitic olivine basalt has since been found to contain abundant small idiomorphic prisms (0·05 × 0·03 mm.) of nepheline. A small amount of poikilitic nepheline discovered in rock at Loc. 59 (168 = P. 3936), also once described as a basalt, relates it to the ankaramitic basanites. Seelye (in Williamson, 1939, p. 66) had pointed out that its chemical composition (Analysis No. 12) is very similar to that of the Clarendon zeolitic basanite (Analysis No. 17). At Loc. 73 the zeolitic dolerite (5059) is covered by a more altered and completely crystalline basanite (5058), in which the bowlingitised olivine seriate phenocrysts of titanaugite and magnetite are wrapped or enclosed by a matrix of coarsely granular (0·5–1·5 mm.) allotriomorphic plagioclase and nepheline in about equal proportions. The very fresh rock (5076) capping The Brothers (Loc. 76) is very like (5659) from Loc. 35, but the horizontally columnar dyke feeding this flow (5056, 5081) is doleritic in its grain-size and in the ophitic form of its deep lilac titanaugite. The tabular plagioclase (0·8 × 0·2 mm.) is a calcic labradorite (An68), and the subidiomorphic nepheline prisms (< 1·0 mm.) contain minute zonally arranged inclusions. Between the colourless minerals is an aggregate of microlites of sanidine (and oligoclase ?) together with chlorite and dusty magnetite as in the olivine theralites of Waihola (Loc. 17). At Loc. 90 on the Kakanui Range (5766 Analysis No. 11) is a rather special member of this group, noteworthy for the abundance of potassic feldspar, probably anorthoclase rather than sanidine. The olivine phenocrysts (<1.0 mm.) are richly forsteritic (Fa8–28), the smaller crystals in the ground-mass are more ferruginous (Fa36–42) with marginal zoning. The titanaugites (< 0.4 mm.) have 2V = 52° to 56° (+) in the phenocrysts, and = 40° (+) in the tiny ground-mass prisms (0.02 mm.), which are associated with magnetite, and rather abundant flakes of strongly pleochroic biotite occasionally 0.2 mm. wide. These are set in a matrix of poikilitic colourless minerals which tend to form irregularly bounded segregations and comprise nepheline, “anorthoclase” with simple Carlsbad or Baveno twinning and 2V=65° to 68° (-), axial plane perpendicular to (010), and labradorite-andesine An53–40. These optical measurements are due to Dr. F. J. Turner.

At Loc. 88, adjacent to the last described rock, is a more finely granular basanite (5767 = P. 6039) in which prismatic nepheline and tabular labradorite are idiomorphic against the small amount of interstitial brown glass in which are irregular patches of analcite. Besides the normal phenocrysts of olivine and titanaugite there are aggregates of diopside and magnetite surrounded by titanaugite which may be derived by the resorption of brown amphibole. Finally, it may be noted that what is possibly an exceedingly fine-grained atlantite (5737) forms a plug a mile west of Kauroo Hill (Loc. 100). It contains in vesicles, and irregular plagioclase (and nepheline?) segregations, microlites of alkaline feldspar with a little pyroxene and chlorite recalling the aggregates in the olivine theralites of Waihola. But the rock is too finely granular and altered for exact determination. Olivine Theralites, the more or less coarsely granular intrusive product of a basanitic magma are well exemplified by the sill at Loc. 17 (5060,–1, 5064, 5067–8, 5698–9, 5700) which will be described elsewhere. The sill appears to exhibit a marked gravitational differentiation ranging from a richly melanocratic basal portion to an upper portion enriched in feldspars and nepheline which has been more or less replaced by a potassic thompsonite. The variation in composition of the plagioclase from the bytownite-labradorite of the lower portion to the andesite of the upper, the increase of potassic feldspar in the higher parts of the sill, the change in average composition of the zoned olivine, strongly forsteritic in the lower portion increasingly ferruginous not only in the outer zones of any one crystal but in the average composition of the crystals in the higher portions of the sill, and the variation of the composition of the pyrozene during its growth, features made apparent by Dr. F. J. Turner's universal stage measurements, will be discussed in detail. Olivine Nephelinite: This name approved by the British Nomenclatural Committee and by Tröger (1935) applies to rocks confusingly termed nepheline basalt by Zirkel and Rosenbusch. A good rather basic example forms a small conical hill at Loc. 50 (5649). Analysis No. 20. The rock is perfectly fresh and contains scanty phenocrysts of olivine (< 0.5 mm.). The matrix consists of faintly pleochroic yellowish to purplish-brown prisms (0.05–0.2 mm. of titanaugite with fluxional arrangement, subidiomorphic grains of nepheline (0.05–0.10 mm.), magnetite (0.005–0.05 mm.), and short needles of a brownish apatite in some of the nepheline grains. No feldspar could be detected in the slide. Very little or no feldspar could be detected in the olivine nephel-inites of Kauroo Hill (Loc. 100). Their peculiar mode of occurrence is described in an appended note. (See also Fig. 3.) The rock (5714, 5717, 5718, 5739, 5761, 5770, 5788) occurs with either fresh or slightly bowlingitic olivines or showing extensive change to iddingsite. The olivine forms the dominant phenocrysts more or less corroded and in all sizes up to 2 mm. in length. Rarely a little picotite is present, apparently xenocrystic and derived from a scattered and partially resorbed olivine nodule. Phenocrysts of titanaugite are less abundant and smaller (<0.5 mm.) but rarely

larger (< 3 mm.) crystals occur containing numerous small inclusions of olivine and large magnetites. The ground-mass consists essentially of short (± 0.03 mm.) prisms of titanaugite, with small octahedra of magnetite and rather less abundant olivine grains, set in interstitial minutely granular or poikilitic nepheline which may extend in optical continuity for a millimetre, including many small augites, etc. There is often a partial replacement of nepheline by analcite or a birefringent zeolite. Sometimes a zonal extinction in such a zeolite may simulate or be simulated by a small fragment of untwinned plagioclase, but only very rarely can thin idiomorphic tabulae of lamellar plagioclase be recognised in association with the nepheline. Very small, thin apatite needles are abundant in the nepheline. Rarely a little finely flaky biotite is recognisable. Deuteric carbonates are present in some slides. What may be a zeolitised and otherwise abnormal variety of olivine nephelinite (5768 = P. 6038) occurs at Loc. 89 on the Kakanui Range. Its olivine phenocrysts (1.0 mm.) and sparse, elongated prisms (< 0.3 mm.) of titanaugite are enclosed in a ground-mass of minutely granular titanaugite and magnetite, with a very little biotite inset in a colourless matrix containing some sanidine and some almost uniaxial optically negative grains, probably of the same mineral, though possibly nepheline where the refractive index is not noticeably different from 1.540. There is much indeterminable, probably zeolitic, material less refractive than sanidine which may replace nepheline. The grain-size is too minute to permit exact microscopic determination. Limburgite in Eastern Otago is so far known only in the Milburn District, where it was collected by A. R. Andrew, though not described in his paper of 1906. The rock lies directly on the limestone or overlaps from it on to the schist. In hand-specimen it is obviously vitreous and breaks easily into many fragments with a sub-conchoidal fracture. Its small vesicles contain zeolite. The rock consists chiefly of a glass which is pale brown in very thin section, with a refractive index determined by Dr Hutton as R.1. (5054) = 1.550–1, R.1. (5760) = 1.549, which corresponds with a silica content of about 53% according to George's (1924) investigations, a figure suggesting that, by the separation of the mafic silicates and magnetite, the residual glass has become richly felspathic. In the glass are innumerable idiomorphic seriate prisms of almost colourless augite up to 0.2 mm. long, octahedra of magnetite rarely reaching 0.1 mm. and olivine subordinate in amount to the pyroxene though larger in grain-size (< 1.0 mm.) either fresh or replaced by iddingsite. Cloudy dark brown aggregates of crystallites are common in the glass especially where the rock shows a rather more advanced crystallisation, though patches of it appear to have been changed to a paler yellow palagonite (?) with minute spherulitic structures. The distribution of the rock is illustrated in Fig. 1. The samples microscopically examined were Loc. 11 (P. 4139, see Analysis No. 22, 5054, 5972, 5693); Loc. 12 (5069, 5694), Loc. 13 (5095–6). Fragmental Rocks: Comparatively little petrographical study has been made of the basaltic breccias and agglomerates. Coarse basaltic agglomerate and tuff making the two small hills at Loc. 22

(5715) probably formed part of a lenticular mass about a mile in length and nearly 300 feet thick, roughly stratified and now dipping S.E. at 30°. It contains many large or small angular blocks of schist as well as small fragments of a variety of basalts set in a matrix of rather fine-grained basaltic tuff. Agglomerate composed of fragments of vesicular basalts are exposed beneath the massive basalts on the eastern side of Yellow Rock Hill (Loc. 29), and Mount Trotter (Loc. 41), interbasaltic agglomerates have been described (Service, 1934) in Mt. Royal (Loc. 45), and the material in the adjacent Locs. 46 and 47 described as trachytic is better considered as derived from a richly feldspathic basalt magma. A dyke of vitric basaltic breccia (5681) occurs at Loc. 43. Very massive and extensive bedded tuffs and agglomerates associated with the basaltic flows near the head of the Shag Valley between Locs. 74 and 75 may indicate the proximity of an important explosive vent. List of Analyses (all by F. T. Seelye, except where otherwise indicated). 1.—Mugearite.—Slightly porphyritic (5652). Hill half mile south of Smooth Cone, 5 miles N.E. of Middlemarch), Strath Taieri S.D. 2.—Mugearite.—Porphyritic type (246). Quarter mile N.W. of Jeffrey's Hill Trig. C, Dunedin and E. Taieri S.D. (In Benson and Turner, 1940, p. 193.) 3.—Doleritic Olivine Basalt.* No specimen of the analysed rock available for microscopic examination.—(Or otherwise dolerite.) North slope of Flat Cap Trig. L, Section I, Block 16, Maniototo S.D. (In Ferrar, 1929, p. 27.) 4.—Doleritic Olivine Basalt.—Flat Cap Trig L* (? P. 3598). Maniototo-Rock and Pillar S.D. (In Williamson, 1939, p. 65.) 5.—Doleritic Olivine Basalt.—(Probably P. 3598.*) 50 chains W. of Trig. A on Haughton Hill, Gimmerburn S.D. (In Williamson loc. cit.) 6.—Doleritic Olivine Basalt.—(42) Farley Street, Dunedin. 7.—Olivine Basalt.*—100 chains S. of Trig. C, Swinburn S.D. 8.—Feldspathic Olivine Basalt.* (?)—Hill W. of Pigroot Hut, 94 chains S.E. of Trig. L, Swinburn S.D. (In Williamson, 1939, loc. cit.) 9.—Feldspathic Iddingsitic Basalt.—Transitional towards basanite (?) (241 = P. 3636). Probably near Trig. H, Highlay S.D. (not Swinburn S.D., as in Williamson, loc. cit.) 10.—Feldspathic Olivine Basalt.*—(Probably 5672 = P. 6002), Mt. Difficulty Trig. D, Kauroo S.D. 11.—Biotite Atlantite.—Rich in Anorthoclase (5766), 15 chains S of Siberia Hill, Trig. C, Kakanui S.D. 12.—Atlantite.—(168 = P. 3936) Donaldson's Coal Pit, 60 chains N.W. Trig. G, Rock and Pillar S.D. (In Williamson, loc. cit.) 13.—Atlantite.—Rungwe, N. Nyassaland, E. Africa. Anal. ? In Lehmann, 1924, cited by Tröger, 1935, p. 240. 14.—Zeolitised Atlantite.—(P. 4158) Cook's Head, Tokomariro Beach, Akatore S.D. 15.—Zeolitised Atlantite.*—72 chains W.S.W. of Trig. O, Swinburn S.D. (In Williamson, loc. cit.) 16.—Zeolitised Atlantite.—With abundant carbonates (5048 = P. 4163). Head of Noble Creek, quarter mile S. Trig. V, Akatore S.D. 17.—Zeolitised Basanite.*—(Probably comparable with 5689), Cemetery Hill, Clarendon, Waihola S.D. 18.—Olivine Theralite.—(5601) Upper portion of Sill, Lake Waihola. 19.—Olivine Theralite.—(5067) Lower Middle portion of Sill, Lake Waihola. 20.—Olivine Nephelinite.—(5649) Small Conical Hill, 3 miles E. of Middlemarch, Strath Taieri S.D. 21.Olivine Nephelinite.—Adenau, Hohe Eifel. L. Koch. anal. (In Tröger, 1935, p. 257.) 22.—Limburgite.—(P. 4139) On limestone a mile south-east of Milburn Railway Station.

Table V. Analysis of Late Tertiary Basic Igneous Rocks of East Otago and Comparable Rocks. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 SiO2 47.09 48.34 46.92 45.85 45.83 45.77 45.83 46.75 44.64 44.38 4303 41.29 42.96 43.21 42.96 40.83 41.26 43.14 40.50 40.95 40.01 40.86 Al2O3 17.74 16.14 14.03 13.27 13.86 12.98 14.35 16.17 15.98 14.90 13.82 13.39 14.95 14.32 14.01 11.59 13.73 17.77 14.89 14.62 14.13 13.20 Fe2O3 4.74 3.43 6.80 5.06 1.95 3.17 1.71 2.50 2.99 2.82 3.70 4.21 6.24 2.88 8.58 3.19 4.20 2.38 3.14 4.87 4.57 6.24 FeO 8.47 9.80 5.91 7.18 9.76 8.66 10.76 9.26 9.47 8.85 8.95 9.73 7.58 8.88 5.14 7.63 10.04 7.90 8.24 8.58 6.04 9.18 MgO 3.15 3.16 7.73 9.19 8.55 11.00 9.47 5.25 6.53 8.32 8.59 5.60 7.68 8.35 4.36 10.06 6.78 3.52 5.27 8.51 11.18 5.80 CaO 5.85 6.54 10.22 11.04 10.33 9.67 9.29 8.51 9.12 10.69 9.23 9.47 11.77 10.06 9.48 10.63 9.16 9.51 13.14 11.60 12.07 7.69 Na2O 5.31 5.24 2.93 2.53 2.76 2.33 2.55 3.91 3.42 3.48 3.91 5.42 3.65 3.64 3.82 3.18 5.48 6.24 3.96 4.10 3.27 2.77 K2O 2.12 2.01 0.81 0.78 0.80 0.83 0.68 1.50 1.40 1.26 2.14 2.10 1.43 1.63 1.31 1.50 1.91 2.10 2.04 1.55 1.11 2.00 H2O+ 1.82 0.77 0.74 1.02 0.42 0.98 0.98 10.58 0.77 0.77 2.37 0.95 0.57 1.83 2.98 1.98 1.41 2.89 2.95 1.34 2.96 4.34 H2O- 0.34 0.30 0.80 0.88 0.64 1.40 2.46 0.48 1.67 0.70 0.78 0.90 0.28 1.17 2.30 1.33 0.53 0.88 0.90 0.66 0.52 1.92 CO2 0.03 0.17 0.58 0.78 2.87 0.10 0.04 0.36 0.03 0.44 tr. 2.43 — 0.08 1.12 5.00 abs. tr. tr. 0.04 0.66 tr. TiO2 1.08 2.20 2.11 1.94 1.58 2.05 1.62 2.56 3.04 2.54 2.66 2.73 2.10 3.06 2.95 1.88 3.72 2.04 2.52 1.42 2.18 3.96 P3O5 1.19 1.50 0.33 0.36 0.30 0.40 0.17 0.65 0.57 0.60 0.73 1.05 1.00 0.61 0.85 0.93 1.41 1.22 2.24 1.26 0.66 1.56 ZrO2 nt.fd 0.01 — — — nt.fd. — — — nt.fd. t.fd. — — 8bs. — — — nt.fd. nt.fd. nt.fd. — — MnO 0.19 0.23 0.19 0.18 0.17 0.20 0.18 0.20 0.20 0.19 0.21 0.29 — 0.19 0.22 0.20 0.21 0.15 0.16 0.21 0.19 0.19 NiO nt.fd. nt.fd. — — 0.04 0.04 — — 0.02 0.03 0.01 0.03 — 0.02 — 0.03 — tr. 0.01 0.02 — 0.01 BaO 0.08 0.06 0.06 0.04 0.03 0.02 0.02 0.06 0.04 0.03 0.06 0.10 — 0.03 0.06 0.05 0.07 0.09 0.08 0.09 — 0.08 SrO 0.015 0.06 — 0.02 0.03 0.04 0.02 0.02 0.03 0.11 0.08 0.04 — 0.05 0.03 0.09 — 0.04 0.05 0.03 — 0.07 Cr2O3 nt.fd. nt.fd. 0.05 0.07 0.05 0.07 0.06 0.02 0.02 0.05 0.04 0.03 — 0.03 0.03 0.06 abs. nt.fd. nt.fd. 0.03 — tr. V2O3 tr. tr. — — — 0.03 — — — 0.04 tr. — — — — — — 0.03 0.05 0.03 — 0.025 Cl 0.02 0.11 — — tr 0.01 abs. tr. tr. 0.02 tr. 0.12 — — — — 0.11 tr. tr. 0.09 — 0.12 F — 0.05 — — — — — — — — — — — — — — — 0.10 0.15 — — — S 0.02 0.04 0.03 tr. 0.04 0.02 0.05 0.05 0.02 0.05 0.04 0.12 — 0.06 tr. 0.08 0.04 0.07 0.04 0.05 0.28* 0.04 — — — — — — — — — — — — — — — — — — — — — — — 99.85 100.16 100.24 100.19 100.01 99.77 100.24 100.20 99.96 100.27 100.35 100.00 100.21 100.09 100.20 100.24 100.12 100.07 100.33 100.05 99.83 100.06 — — — — — — — — — — — — — — — — — — — — — — — 99.85 100.16 100.24 100.19 100.01 99.77 100.24 100.20 99.96 100.27 100.35 99.97 100.21 100.07 100.20 100.21 100.12 100.00 100.25 100.03 99.83 100.03 *SO3

Literature Cited. Andrew; A. R., 1906. On the Geology of the Clarendon Phosphate Deposits, Otago, N.Z., Trans. N.Z. Inst., vol. 38., pp. 447–482. Benson, W. N., 1915. The Geology of the Tamworth District, Proc. Linn. Soc. N.S.W., vol. 40, pp. 541–624, esp. 618–9 —– 1940. Landslides and Allied Features in the Dunedin District, etc., Trans. Roy. Soc. N.Z., vol. 70, pp. 249–263, fig. 1 —– 1934. The Geology of the Dunedin District, New Zealand. Abstract Proc. Geol. Soc. London. Dec. 5th, 1934 —– 1941. Cainozoic Petrographic Provinces in New Zealand and their Residual Magmas, Amer. Journ. Science, vol. 239, pp. 537–552 —– 1941 (a) The Basic Igneous Rocks of Eastern Otago and their Tectonic Environment, Part I, Trans. Roy. Soc. N.Z., vol. 71, pp. 208–222.

Benson, W. N., and Chapman, F., 1938. Occurrence of Radiolarian Limestone among the older rocks of South-Eastern Otago, Ibid., vol. 67, pp. 373–4. Benson, W. N. and Turner, F. J., 1940. Mugearites in the Dunedin District, Ibid., vol. 70, pp. 188–199. Brown, D. A., 1938. Moeraki Subdivision, Rept. Geol. Survey Branch, Dept. Sci. Ind. Res., 1937–38, pp. 9–12. Cotton, C. A., 1917. The Fossil Plains of North Otago, Trans. N.Z. Inst., vol. 49, pp. 429–32 —– 1922. Geomorphology of New Zealand. Govt. Printer, Wellington. George, W. O., 1924. The Relation of Physical Properties of Natural Glasses to their Chemical Composition, Journ. Geol., vol. 32, pp. 353–372. Haast, J von, 1877. Notes accompanying a Geological Map and Sections of the Shag Point District, Province of Otago, Report Geol. Explor., 1873–4, pp. 19–26. Harker, A., 1908. The Geology of the Small Isles of Inverness, Mem. Geol. Surv. Scotland. Sheet 60. Lacroix, A., 1927. La Constitution lithologique des Isles Volcaniques de la Polynesie Australe, Mem. de l'Acad. de Science, tom, 70, pp. 1–82, esp. pp. 19–21. Lehmann, E., 1924. Das Vulkangebiet am Nordrand des Nyassa als Magmatische Provinz, Zeits für Vulkanologie, Erganzungs Bd. 4, pp. 76–85, 118–124, 174–5. McKay, A., 1884. On the North-Eastern District of Otago, Report Geol. Expl., 1883–4, pp. 46–66, esp. 65–66 —– 1887. On the Young Secondary and Tertiary Formations of Eastern Otago—Moeraki to Waikouaiti, Report Geol. Expl., 1886–7, pp. 1–23. Marshall, P., 1906. The Geology of Dunedin (New Zealand), Quart. Journ. Geol. Soc., vol. 62, pp. 381–422 —– 1912. Nephelinite Rocks in New Zealand, Trans. N.Z. Inst., vol. 44, pp. 304–7 —– 1914. The Sequence of Lavas at the North Head, Otago Harbour, Quart. Journ. Geol. Soc., vol. 70, pp. 382–408 —– 1918. Geology of the Tuapeka District, N.Z. Geol. Surv. Bull., No. 19, pp. 63–4, Govt. Printer, Wellington —– 1925. The Igneous Rocks of New Zealand. Gedenboek Verbeek, Verhandl, v.h. Geol. Mijnbouwkundig Genostschap v. Nederlanden Kolonien, Geol. Serie VIII, s' Gravenhage, pp. 357–367, esp. 364. Park, J., 1906. The Geology of the Area covered by the Alexandra Sheet, Central Otago Division, N.Z. Geol. Survey, Bull., No. 2 —– 1908. The Geology of the Cromwell Subdivision, Western Otago Division, Ibid., Bull. No. 5. Paterson, O. D., 1941: The Geology of the Lower Shag Valley, N.E. Otago. Trans. Roy. Soc. N.Z., vol. 71, pp. 32–58. Service, H., 1934. The Geology of the Goodwood District, N.E. Otago, N.Z. Journ. Sci. Tech., vol. 15, pp. 273–9. Tröger, E., 1935. Spezielle Petrographie der Eruptivgesteine, Deutsch Min. Ges., Berlin. Turner, F. J., 1939. Petrographical Notes on the Igneous Rocks of the Naseby Subdivision in Williamson, 1939, pp. 66–7 —– 1940. Structural Petrology of the Schists of Eastern Otago, Amer. Journ. Soi., vol. 238, pp. 73–106, 153–191. Wellman, H. W. and Willett, R. W., 1942. The Geology of the West Coast from Abut Head to Milford Sound, Part I. Trans. Roy. Soc. N.Z., vol. 71, pp. 282–306. Williamson, J. H., 1939. The Geology of the Naseby Subdivision, Central Otago, N.Z. Geol. Surv. Bull., No. 39. Govt. Printer, Wellington. Corrigendum for Part I hereof (Benson, 1941a). On Plate 36, for Milburn read Clarendon. (Milburn lies two miles south-west of the point indicated.)