The Basic Igneous Rocks of Eastern Otago and Their Tectonic Environment, Part III. The Olivine Theralite of Waihola, East Otago, a Gravitationally-Differentiated Sill, with Notes on Related Rocks.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 72, 1942-43, Page 160

The Basic Igneous Rocks of Eastern Otago and Their Tectonic Environment, Part III. The Olivine Theralite of Waihola, East Otago, a Gravitationally-Differentiated Sill, with Notes on Related Rocks. By W. N. Benson. Including Appendices by F. J. Turner and C. O. Hutton and Chemical Analyses by F. T. Seelye. [Read before the Otago Branch, November 20, 1941; received by the Editor, November 27, 1942; issued separately, September, 1942.] General Geology and Petrology. In 1911 J. P. Smith discovered on the western shore of Lake Waihola, twenty miles south-west of Dunedin, a basic nepheline-rich rock, comparable with one labelled “Diorite, Lake Waihola,” which had been placed in the Otago Museum by Professor F. W. Hutton some time before 1879. Subsequently, in company with Mr. Smith, Dr. Marshall examined it in the field, and wrote an account thereof, and of other nepheline-rich rocks in the outskirts of the Dunedin district, and of one found near Auckland (Marshall, 1912). The mode of occurrence of the Waihola rock was left uncertain, and is described in the present paper. A number of allied rocks have now been found, and their varied petrographical features have been studied, the investigation being greatly helped by universal stage measurements by Dr. F. J. Turner, recorded in an appendix hereto. Dr. C. O. Hutton has made centrifuge-separations of a titanaugite and a zeolite from one of the rocks, has determined their optical properties and densities, and has discussed their compositions in a second appendix. Through the courtesy of the Director of the Geological Survey and the Dominion Analyst, analyses of these two minerals together with two complete rock-analyses have been made by Mr. F. T. Seelye. To these gentlemen the writer's most cordial thanks are due. The boundary of the igneous rocks west of Lake Waihola, as sketched in the maps of Marshall (op. cit.) and of Ongley (1939), needs some modification, and this, together with the boundaries of their subdivisions, are shown on Fig. 1. The basement rock is the prevalent quartz-albite-muscovite-chlorite-epidote-schist of about the Chl. 3 stage as recognised by Dr. Turner (1940). On a peneplain cut in this in Cretaceous times there rests a westward-tapering wedge of sediments comprising a few feet of Upper Cretaceous coal-measure sandstone followed by 60–100 feet of very Late Cretaceous glauconitic Abbotsford mudstone, both of which formations are closely similar to their equivalents in the Dunedin district, wherein evidence of their age has been obtained (Ongley, 1939). The glauconitic mudstone is

Fig. 1 best exposed in a steep cliff rising from the lake-shore beneath the southern margin of the igneous rocks, and is here capped by a formation without known analogy in Eastern Otago, a yellow-buff chert of small extent and about three feet thick lying immediately below the lava. Though apparently homogeneous in hand-specimen it shows in microscopical section remains of the stem of a large brown fucoid like Durvillea* For this comparison we are indebted to Dr. J. E. Holloway. and numerous indeterminate foraminifera. In Late Tertiary times the area here described was tilted eastward, and a second peneplain was cut, passing westward obliquely from the chert and mudstone on to the schist. The igneous rocks which rest on this erosion-surface were erupted probably during the late Pliocene times. Three groups of rocks may be recognised among them. The earliest and most widespread is an olivine basalt, represented by slides P.4164–5, cut from material collected for the Geological

Survey by Ongley, and by 5052, 5065, and 5079 in the Otago University Geological Museum. The positions of these are shown in the accompanying map (Fig. 1), the last two digits only being given for the University slides. The flow was erupted probably from beneath Waihola Hill, where it is about 300 ft. thick, but east and west thereof it extends as a sheet less than 100 ft. thick. The rock, which has a specific gravity of 3·00, is an almost normal fine-grained basalt with a minor proportion of small olivine phenocrysts, though on the summit of Waihola Hill, a small amount of zeolite is present, and the rock begins to show something of the characters of the earliest zeolitic basalt in the Clarendon-Milburn district four miles further south, described in detail elsewhere (Marshall, 1912; Benson, 1942), though it is not to be correlated therewith. The overlying flow of basanite (atlantite) rests directly on the basalt on the landward side of the promontory, but further east the flows are separated by the intervening theralite sill. The higher flow (5070, 5071) is very distinctive both in hand-specimen and under the microscope. It is dense (sp. gr. 3·06), and contains abundant phenocrysts of olivine and subordinate titanaugite, and seriate phenocrysts of labradorite, set in a fine-grained basaltic matrix. In this matrix there is a rather uneven distribution of the dark minerals (minutely prismatic and granular titanaugite and titaniferous magnetite), while a little analcite and an indeterminate, birefringent zeolite occur both inter-stitially and in small, irregular, micropegmatoid patches associated with rare small patches of irregularly granular or poikilitic nepheline and labradorite containing idiomorphic titanaugite, and also with little segregations of sanidine and minute flakes of biotite. Xenoliths of quartzose schist show the replacement of chlorite and epidote by small augite prisms, and many xenocrysts of quartz illustrate all stages in the development of a finely granular, pyroxenic reaction-rim up to complete replacement of the quartz by more or less ovoid aggregates of minute augite granules. Rarely xenocrystic picotite is present. Appearing from beneath this upper flow and extending eastward to the point of the promontory, is the wedge-like sheet of more or less coarsely granular rocks containing those which Marshall (1912) described. Exposures are poor, but there can be little doubt that this westward tapering sheet is intrusive and dips eastward at about 5°, reaches a maximum thickness of rather more than fifty feet, covers in all about a hundred acres above lake-level, and extends for some distance eastward beneath the lake. Its lower portion is best studied along its north-eastern margin, where, immediately above the basalt, is a ledge in which there occur, both in situ and as loose blocks, many exposed masses of a rather fine- to medium-grained melanocratic rock with a specific gravity of about 3·00. Such rocks, and the top of the underlying basalt, may be traced down to the water's edge at the end of the promontory. Immediately above them the loose rocks have a rather coarser grain-size, are lighter in colour, and less dense (sp. gr. about 2·9). What must be near the top of the sill is exposed, in deeply weathered form, in a low cliff, and in several immense isolated blocks about 600 yards south-west of the point. This is a very coarse and meso-leucocratic rock with a specific gravity of 2·8–2·75,

the lower figure being given by the most strongly zeolitic and slightly drusy specimens. The rock may be traced by loose blocks about fifty yards westward up the grassy slope, until blocks of the covering basanite are encountered. No evidence was obtained of the presence of an intervening upper melanocratic layer of the coarsely granular rock. The southern margin of the intrusion is more obscured by soil-drift, though dark coarse-grained loose blocks occur here and there. The western development at the head of the small valley is wholly under ploughed land, and the loose blocks therefrom have been collected to form stone fences. The general features of the rocks in this sill are illustrated in Fig. 2, with the proviso that as each micro-drawing is of a field selected to show the structural relations of as many minerals as possible, it is not also representative of their relative abundance. Thus 5061 of Fig. 2 is structurally typical of those rocks of the upper portion of the sill which contain relatively little zeolite, though the field of view chosen for illustration is rather unusually rich in dark-coloured constituents. Except for the scarcity of zeolites 5067 is very typical of the bulk of the moderately coarse and not strongly melanocratic rocks in the lower portion of the sill, while 5064 represents the least coarsely granular and most basic rock occurring in small amounts near the base of the sill in its north-eastern exposure. The source of 5699 is not known. The specimen bears only the locality “Waihola” in Smith's handwriting. The petrographic features, however, suggest it also was obtained from near the base of the sill. The locality from which Marshall collected the meso-leucocratic and richly zeolitic specimen 5698 was almost certainly near the top of the sill. The rocks of intermediate character are represented by the analysed and figured specimen 5067 (sp. gr. 2·910). It contains more or less idiomorphic or occasionally sub-ophitic crystals of titanaugite up to 3·0 mm. across. They are associated with iron ores and apatite, and rarely show a tendency to be moulded against the plagioclase or to fray out into sub-graphic intergrowth with nepheline. They have a zoned or hour-glass structure. The inner portion, usually displaying a lighter colour than the margin, has a larger optic axial angle (2V=(+)64°–47°) than that of the broad outer zone (2V=(+)54°–42°), from which it is usually separated by a sharply marked but irregular line. This angle then increases quickly to 2V=(+)72° in a thin greenish outermost zone. The extinction-angle (Z∧c=46°–53°), in the absence of salite structure or twinning, cannot be read without a possible error of about 3° ±. It is rather large in all parts of the crystal. From the diagram given by Deer and Wager (1938, p. 22) it may be inferred that the high values of Z∧c for crystals having 2V=(+)45°–50° betoken a general richness in iron, and that the cores of the crystals of titanaugite in 5067 have compositions lying on the boundary between diopside and hedenbergite. The outer zone, though less calcic, is still rich in FeSiO3, and the change of colour and increase in optic axial angle in the outermost portion accompany the entry of the acmite molecule into the composition of the pyroxene. In these greenish mantles finely granular magnetite is abundant, though but little occurs in the purplish pyroxene.

The olivine may show a zoning marked by variation in birefringence, of extinction-angle in oblique sections, and of optical axial angle, indicating an outward increase in the proportion of fayalite as described by Tomkieff (1·39). A strikingly zoned crystal in 5067 (see Fig. 2) shows a variation of axial angle indicating that a central portion of composition Fa34 has an outer zone of Fa56 beyond which, but still in optical continuity with it, are outlying remnants of an almost completely resorbed mantling zone of Fa69. The particular grain figured does not, however, lie in such a position as to display a zonal change of extinction-direction or birefringence. The feldspar is in more or less idiomorphic tabulae, approximately 1·5×0·2 mm., flattened parallel to 010, elongated parallel to c and partially replaced by zeolite, which has formed in cracks traversing the feldspar. It includes both potassic feldspar and plagioclase (An38). The former has 2V=(-)45°, the optic axial plane being perpendicular to 010, and according to Spencer's (1937) data, might be either orthoclase or more probably anorthoclase rich in soda. This uncertainty, together with the impossibility of determining the composition of many of the grains of pyroxene or of feldspar, and the difficulties introduced by the development of zeolite, discourages the use of Rosiwal's method for estimating quantitatively anything more than the general proportion existing between several rather heterogeneous groups of minerals (cf. Smith and Chubb, 1927). Nepheline was the last anhydrous mineral to crystallise and formed a matrix of poikilitic grains often over 10 mm. in diameter. Like the nepheline in the nephelinite of Lake Kivu, Africa (Shand, 1939), that in the Waihola complex is unzoned. Its refractive indices are only very slightly less than that of Canada balsam. The analysis of the rock as a whole suggests that it is probably a potassic type. It is partially replaced by zeolite, the fibres of which have positive elongation (but see p. 181). This occurs in irregular marginal or interstitial patches or in the cracks, which traverse the mineral in many directions. Iron ores more or less idiomorphic and including both titaniferous* Confirmed by magnetic separation and chemical test for TiO2. magnetite, and ilmenite and also apatite are abundant accessories. The last mineral forms prisms up to 2·0 × 0·15 mm. long and thick, embedded in all the other constituents, but though it was thus the first mineral to commence crystallising, it seems to have continued to form during most of the period of rock-consolidation, for it often projects from the coarse-grained material into the interstitial patches of finely crystalline material in which it is often more abundant than elsewhere, just as was noted by Tyrrell (1928) among analogous rocks in Ayrshire (cf. also Elsden, 1908, pp. 289–290). In this rock, as in most of the other Waihola theralites here described, the apatites contain inclusions, often filled with liquid. These are not swarming minute objects elongated parallel to the vertical axis of their host, such as darken the larger chiefly xenocrystic apatite-prisms in many Dunedin rocks: usually a single tube only occurs about 0·01 (rarely 0·02 mm.) in width extending in the above direction for the greater part of the length of the crystal. In this tube is mineral-matter seemingly in part of a composition related to

Fig. 2

that surrounding the apatite, some of which may have been originally included in the crystal or has been introduced into it by diffusion. Edwards' (1938, p. 303) comment concerning the frequent association of apatites containing abundant inclusions with the breaking down of basaltic hornblende is not applicable to the present case. Here, where the apatite is in feldspar, the central tube may contain a little cloudy to dark brown limonite, or merely colourless fluid; when in magnetite the filling may be quite black over the whole or part of the length of the tube, and among the patches of finely crystalline microlitic interstitial material, the filling of the central tubes in the late-formed but idiomorphic apatite is often either dark green, like the more or less chloritised sodic pyroxene (or glass?), or pale green chlorite, with very minute scales of (?) ilmenite, as in the adjacent completely altered matrix (cf: Baker, 1941). The irregular finely granular patches consist of minute (up to 0·12×0·015 mm.) arcuate microlites of anorthoclase with interstitial aegirine, dust-like magnetite or minute plates of ilmenite set in a zeolitic matrix. Of the more basic rocks 5068 (sp. gr. 2·970) differs from the above chiefly in the greater proportion of dark constituents. In the latter, however, the pyroxene is sometimes unzoned (2V=(+)49°) or may show a purplish core (2V=(+)48°) surrounded by a paler outer zone (2V=(+)50°–58°), the extinction angle Z∧c varying from 45°–53°. Here the unusual content of hedenbergite is seen once more, but in the latter crystal there seems to have been the theoretically normal increase in the amount of diopside in the pyroxene molecule as the crystal grew, rather than the increased clinoenstatite seen in 5067, in most of the phenocrysts of titanaugite in the Dunedin basalts (Benson and Turner, 1940), and in general in the basalts of the Pacific region (Barth, 1931). Differing from this is 5064 (sp. gr. 3·004) (see Fig. 2), characterised by its smaller grain-size and by the occurrence of the abundant calcic titanaugite in smaller grains 0·1–0·3 mm. in diameter, rarely idiomorphic and showing but little zoning; 2V=(+)54°–56° in the centre to (+) 58° in the thin marginal zone. Olivine is subordinate and slightly yellowish: investigated grains include one which has a core of composition Fa37–46 and a mantle of Fa54 as well as an unzoned grain of Fa33. Large and small prisms of apatite together with the iron ores and coloured silicates are enclosed in a matrix of poikilitic nepheline forming grains up to 10 mm. in diameter and, in approximately equal amount, poikilitic plates of andesine up to 7 mm. in diameter. A very little zeolite and calcite and partial replacement of the olivine by bowlingite have resulted from deuteric processes. The tendency here displayed for the development of comparatively coarsely granular poikilitic nepheline and plagioclase after the separation of abundant relatively small grains of the coloured minerals from a rather aqueous basic alkaline magma, is characteristic of many of the more finely granular more or less zeolitised lavas in the peripheral regions of the Dunedin province, as has been shown elsewhere (Benson, 1942). It is exhibited, for example, by the basanite which overlies the Waihola sill.

Another basic rock (5700, sp. gr. 3·062), which occurs near the lowest portion of the sill at its north-eastern exposure, differs from the above in the greater size of the coloured crystals, the presence of a small amount of the microlitic product of a residual melt, and the more advanced zeolitisation of the abundant nepheline, with which the purplish titanaugite makes irregular intergrowths. Zoned olivines in this rock show a range in composition from Fa37 to Fa49 in one case and Fa18 to Fa38 in another. In the latter the section was cut nearly perpendicular to an optic axis, and showed also a variation of nearly 15° between the extinction directions in the centre and in the very thin outer zone of the grain. A strikingly different rock 5699 (sp. gr. 3·002) comes from an unknown locality probably near the base of the sill, if such inference may be drawn from its composition. It is characterised by the presence of a variable but smaller amount of nepheline than occurs in the foregoing, though not greatly subordinate to the amount of plagioclase, also by the basicity and marked zoning of the plagioclase, and the ophitic character of both the pyroxene and olivine. The feldspar is twinned on the albite, Carlsbad and occasionally pericline laws. Where it was free to develop, its composition ranges from a bytownite core (An85) to a thin mantle of andesine (An45) and there are a few unzoned laths (or prism-edges) of andesine only (An44). The feldspar enclosed within the ophitic pyroxene at an early stage of the crystallisation is bytownite only—An82 or An79 in two cases; that in the olivine may be still more calcic—An85 or An90, one of two examples of the last having a very thin outer mantle of An72. But, contrary to Tomkieff's (1939, pp. 241–2) experience with ophitic olivine dolerite, the olivine here is not of the late formed fayalitic type, but the investigated crystal has in the main the composition Fa35 with remnants of a very thin peripheral zone of Fa47. It is therefore inferred that the feldspar in these rocks began to crystallise at a very early stage in the rock-consolidation, and subsequently, after inclusion in augite or olivine, had little opportunity for much reaction with the magma in its latter more alkaline residual stages. The purple titanaugite has sometimes an outer zone with 2V=(+)46° sharply separated from a paler and more diopsidic core with 2V=(+)54°: and in these Z∧c=45°±3°. The nepheline varies considerably in its distribution. In the slide illustrated in Fig. 2 it is in abnormally small amount, occurring interstitially among the feldspars and constituting but 5% of the rock. In another slide it occurs in large poikilitic masses and is nearly as abundant as the feldspar. In both it is partially zeolitized. Linking the melanocratic rocks with those in which the amount of light and dark minerals are more nearly equal (mesocratic), there are rocks such as one with sp. gr. 2·837, while others, in which the lighter minerals tend to dominate—(meso-leucocratic), 5060 (sp. gr. 2·751), 5061 (sp. gr. 2·78), and 5698 (sp. gr. 2·77), are still less dense; of these, 5060 is slightly drusy. In these rocks the olivine is present in but small amount, though its grain-size may still be large. It is usually more ferruginous than in the denser rocks. One grain, not noticeably zoned, is over 5 mm. long, and has the composition Fa51. Some golden-brown iddingsite, as well as a little greenish bowlingite,

has formed in the cracks of this olivine. The crystals of pyroxene, which may be over 4 mm. long, have usually a dark purple zone with 2V=(+)44°–48° and slight axial dispersion surrounding a paler purplish brown more diopsidic core with 2V=(+)50°–54° and strong axial dispersion, though in a few crystals paler material may occur outside the darker. In either case a greenish border with 2V=(+)58° expresses the entry of the acmite molecule in notable amount into the construction of the pyroxene. Such material may fray out into fine greyish to bright green intergrowths with feldspar or be rarely moulded thereon, or even on the nepheline in rocks such as 5061 and 5698 in which this mineral is very abundant and idiomorphic. (See Fig. 2.) The feldspar occurs in large irregular prismoid grains, occasionally over 6 mm: long. As Marshall (1912) noted, “it is much twinned on the albite and pericline laws, and has in many places the appearance of microcline. The extinction-angle, however, proves it to be andesine.” Two universal stage measurements give An39. In some cases it is surrounded by a thick outer zone of anorthoclase sharing the (010) and pericline twinning-planes in common with the plagioclase. Nepheline forms large more or less irregular to subidiomorphic grains in 5060 and 5061, but in 5698, in which it is the dominant mineral, it crystallised before the feldspar forming short idiomorphic prisms over 1·0 mm. thick against which the feldspar and sodic pyroxene may be moulded. Rarely, as in 5061, subidiomorphic nepheline prisms are enclosed in purple titanaugite. Complete replacement of idiomorphic nepheline has given rise to pseudomorphs in which the deuteric zeolite is commonly present as irregularly dispersed and bounded aggregates of radiating fibres, which, when viewed in a direction perpendicular to the basal plane of the original nepheline, may be seen in places to be grouped into several sets of thin prisms (possibly perpendicular respectively to the prism faces of the nepheline), the several sets being arranged in stellate fashion and occupying definitely limited sectors of the pseudomorph. (See Fig. 2). The irregular patches of residual material are generally less finely granular than in the melanocratic rocks. They contain apatite prisms 1·2×0·08 mm. with the characteristics already described. Anorthoclase ranges in form from nearly rectangular laths up to 0·6×0·05 mm. to less sharply bounded, narrow laths at most 0·5 mm. long and thence down to minute more or less curved microlites. Between these is a mixture consisting of pale to dark green or greenish-brown chloritic material, and little flakes of deep brown biotite, apparently developed by reaction between the alkaline residual liquid and the iron ores, about which they may be wrapped, though they also occur separately and are occasionally moulded on the ends of the feldspar microlites. Between these is a matrix of zeolite, either clear or slightly stained by chlorite or limonite, or turbid, perhaps through partial dehydration during the making of the rock-section. In some specimens (e.g., 5061) the zeolite forms rather large patches (4×2 mm.) from which the central portion has been removed, so that the rock appears slightly drusy. In the rock-sections, the thin prisms of zeolite usually show a positive elongation, a birefringence slightly greater than that of the adjacent plagioclase, and an optic axial angle as measured in a number of crystals approximately 2V=60°±5°. Occasionally,

however, a zeolite, not otherwise distinguishable from these, shows 2V=30°. In rare cases different sectors of a fibrous radiating zeolite may show either a positive or negative elongation (but see p. 181), indicating that it is probably a thomsonite. In order to determine in more detail the composition of the zeolite in a typical and abundant richly sialic rock, Dr. C. O. Hutton, at the writer's request, very kindly separated the zeolite out of 5698, and his observations recorded in Appendix II, confirm the thomsonitic nature of the zeolite, but indicate that it is a very abnormal, if not hitherto quite unrecorded, markedly potassic type. The opportunity was taken while making this separtion to obtain and determine the composition and other properties of the titaniferous augite in this rock, and the refractive indices and density of the apatite which are also recorded in Appendix II. The deuteric processes in these rocks are:— (a) The more or less advanced zeolitisation of feldspar and especially of the nepheline. The process works inward from crevices and may result in the complete and sometimes pseudomorphous replacement of the nepheline with occasionally (as above) partial removal of its substance, leaving a small cavity with resultant lowering of the density. (b) The formation of scraps of biotite by the action of the residual alkaline liquid on the iron ores. (c) Incipient chloritisation of the late-formed pyroxene. (d) Formation of golden-brown iddingsite, the alteration product of the olivine when acted upon by the residual liquid, sometimes associated with a little bowlingite, which may be a still later product. The whole association of rocks affords in miniature a vivid picture of the processes attending the consolidation of an aqueous basic alkaline magma, differing in some respects from the rather similar though not identical rock-associations in Utah, Ayrshire, and Shiant described by Gilluly (1927), Tyrrell (1928) and Walker (1930) respectively. Apatite was the first mineral to commence crystallisation. The formation of the iron ores followed, shortly to be succeeded by the separation of very calcic plagioclase (anorthite-bytownite). Then forsteritic olivine (about Fa33) commenced to separate and, though more magnesian than the olivine which formed later, its rather high content of fayalite reflects the abnormally ferruginous character of the parent-magma, one of the distinctive features of the basaltic rocks of the Dunedin district (see Benson and Turner, 1940, p. 58). That some concentration of iron had occurred in the initial segregation of the theralite magma from the parent reservoir is suggested by the generally smaller content of fayalite (averaging Fa21) in the early formed olivines in the basalts of the Dunedin district (Benson and Turner, 1940, p. 67). Separation of titanaugite, again rather unusually ferruginous, commenced after that of olivine and both of these mafic silicates as they grew included ophitically tabulae of the still very calcic plagioclase, and began to sink into the lower parts of the intrusive sheet of magma, where from the residual magma there formed less calcic outer zones about such feldspar crystals as were free to react with it, together

with a subordinate amount of nepheline which was rather uneven in its distribution (e.g., 5699). This early separation left a melt rather impoverished in lime and magnesia. Of the crystals suspended therein, the olivine received outer zones which were increasingly ferruginous, up to Fa65 or more, or grains of like composition formed about new nuclei, and both were in their turn more or less resorbed probably during the later stages of pyroxene-formation. A slightly less calcic but still rather ferruginous titanaugite formed about the earlier crystallised more calcic-hedenbergitic grains, and ultimately merged outwards into greyish to bright green aegirine-augite, forming a mantling zone containing abundant finely granular magnetite as the residual magma became enriched in both alkalies and iron. At this time also some resorption of the fayalitic outer zones of the olivines may have occurred and the plagioclase was made over into a probably rather potassic andesine, and continued to grow with that composition. It was at this time no longer idiomorphic but was moulded against the olivine and pyroxene. In contrast with the sequence of mineral formations in the analogous complexes in Scotland and Utah, no plagioclase more sodic than andesine has been found in the Waihola complex. Nepheline, probably more potassic than before, and still the last of the principal colourless minerals to form, made irregular poikilitic grains extending in optical continuity for several millimetres. As crystallisation advanced towards its close, the increasing concentration of potash in the residual magma was expressed in two ways. Microlites of anorthoclase (together with finely granular interstitial aegirine-augite, apatite and iron ores) crystallised in irregular patches among the larger crystals in ever-increasing amount and grain-size. In addition, the tendency for potash-feldspar to remain in strained isomorphism with the plagioclase crystals gave way with decreasing temperature to the separation of anorthoclase forming thick zones about the plagioclase grains. With its formation and consequent discharge of accumulated potash from the magma there may have been a renewed concentration of soda in the melt, so that potassic (?) nepheline became the most abundant anhydrous product, and where this is the case its period of crystallisation was advanced until it largely preceded that of the feldspar. On the other hand, in alkaline melts which had been developed in the upper portion of the sill the period of separation of pyroxene was extended by the entry of the acmite molecule into its constitution, and the latest crystallisation of pyroxene either accompanied or followed both that of the nepheline and of the plagioclase. The texture and increased grain-size of such rocks approach that characteristic of pegmatite, though they must have consolidated under a cover of basalt at most only a few hundred feet thick. The relatively rapid cooling, however, prevented the coarsely granular texture from extending throughout the whole of the complex. In every part of it (except perhaps the lowest portion composed in large part of gravitationally accumulated crystals), a residual more or less aqueous magma consolidated as aggregates of microlitic anorthoclase with apatite prisms, interstitial aegirine-augite, and minutely granular iron ores. Meanwhile the effect of the increasing concentration of water in the melt was seen

in the conversion of some olivine into iddingsite as the concentration of iron increased and the temperature fell, a deduction in full accord with Edwards' (1938, p. 486, 1938a, p. 280) view that “it is essential for the formation of iddingsite that the magma should be not only rich in water vapour, but that it should have differentiated in such a way as to give rise to an iron-rich fluid.” The formation of bowlingite rather than iddingsite from the residual olivine was probably a still later deuteric process. The partial chloritisation of the minutely granular pyroxene in the interstitial patches may be assigned to the same cause, and so, also, because of the alkalinity of the residual melt, the formation of the rare scraps of deuteric biotite in the most alkaline rocks. That small amounts of carbonates (calcite, etc.) occur in certain of the Waihola theralites, not in direct association with these residual patches or with the masses of zeolitic material, leaves in doubt the decision as to whether their formation preceded or followed the more explicitly deuteric processes. The features of the associated volcanic rocks seem, however, to be consistent with the view that all these processes may have been more or less contemporaneous. The deuteric formation of zeolite was most advanced in the more highly alkaline upper portions of the complex, but all of them are affected by the process. None of the rocks so far investigated appears to afford evidence of the formation of any analcite, such as occurs in the rocks of the other complexes already cited, nor of any other zeolite by primary crystallisation, i.e., during or before the crystallisation of any of the anhydrous rock-forming minerals. On the contrary the zeolite everywhere replaces nepheline, or to a smaller extent feldspar. The reason for this contrast is not apparent at present. It is not possible to detect any difference in composition between the zeolites replacing nepheline and those replacing the feldspar. The average composition of the secondary zeolite has been shown to approximate to that of a potassic thomsonite. (See Appendix II.) Discussing the origin of a series of rocks comparable to some extent with those herein described and the others to which comparative references have been made, Drescher and Krueger (1928) concluded that after the continuance of formation of anorthoclase down to an assumed temperature of 400° C., zeolitisation supervened, commencing at about 350° C. and continuing as the temperature fell to about 90°. They pointed out that had a concentration of potash preceded that of soda, myrmekite and checker-board albite might have replaced the anorthoclase, but that with such a mineral sequence as occurs in our rocks, namely, CaNa→K→Na, the temperature at the time of the second concentration of soda had been lowered to such an extent that this form of replacement, though very common in pegmatites, could not here take place. It is, therefore, uncertain to what extent the greater volatility of compounds of potassium than those of sodium, to which Bowen (1933) refers, may be responsible for the increasingly sodic character of the residual magma at this late stage. It is interesting to recall the experiments cited by Morey and Ingerson (1937) which bear on the conditions of origin of the late-

formed minerals in the Waihola complex. Thus Thugutt (1894–5)* See references cited by Morey and Ingerson (1937). obtained from nepheline attacked by K2CO3 solution at 200° C. a potassic natrolite readily converted to the sodic mineral by heating with Na2CO3 on a steam bath. Doelter (1906)* found that analcite would crystallise at temperatures between 400° C. and 190° C. from a variety of solutions of analcite and natrolite; also that natrolite would form from similar solutions at temperatures between 190° and 90°. Koenigsberger and Müller (1906)*, moreover, obtained anorthoclase as a result of the attack of water on alkaline glass at 360°. Less significant for our problem was the work of Friedel (1912)*, who obtained a little natrolite from the attack of aqueous solutions of KOH on mica at temperatures as high as 510°–600°, and of Gruner (1929)*, who found natrolite to crystallise freely when muscovite or paragonite were attacked by aqueous solutions of KOH and NaOH especially at 400° C., but also at lower temperatures. The above descriptions of the rocks in the Waihola complex raise the question of nomenclature. The term nephelinite originally applied to them seems inappropriate in view of the abundance of feldspar in all of the members of the complex, which would bring them much closer to the scope of Rosenbusch–Osann's (1922) latest definition of theralite as hypautomorphic granular rocks consisting of abundant or predominating pyroxene together with lime-soda plagioclase, nepheline and possibly sodalite with accidental or accessory biotite, hornblende, olivine, iron-ores and apatite. Lacroix (1920) added that orthoclase in these rocks should be rare or concealed. Its presence at all, Johannsen (1938, p. 222) holds, would exclude the rocks from the strictly Rosenbuschian theralite, though Rosenbusch himself (1907, p. 428) notes its occurrence forming a shell around the plagioclase crystals in the material from Duppau, Bohemia, which he describes most carefully as a type of the rock-species. The plagioclase should be decidedly basic and dominant over the nepheline. Lacroix (1920) recognised two distinct varieties of theralite, berondites characterised by the presence of brown hornblende, and luscladites in which olivine accompanies the pyroxene. His typical luscladite had a hyperitic structure and only a small amount of nepheline (normatively amounting to 5%, fide Tröger, 1935, p. 229), between the plagioclase tabulae which were mantled by orthoclase, and he remarked on the transition toward gabbro which resulted from a decrease on the amount of nepheline (see Analysis No. 7, p. 179). Later, however, Lacroix (1927, pp. 10, 35–7) accepted and extended Smith and Chubb's (1927, pp. 318–322) modification of the definition of luscladite so as to include mesocratic medium to coarsely granular rocks with titanaugite, olivine and biotite as the chief dark constituents, plagioclase with or without potassic feldspar as an independent mineral, and nepheline (sometimes accompanied or replaced by sodalite), which may be as abundant as the coloured minerals and may enter into graphic intergrowth with the pyroxene. If the name be applied to all our rocks, its definition must be further extended to cover meso-leucocratic rocks in which nepheline is the dominant though not the sole sialic mineral, a complete departure from its original significance. We therefore follow Johannsen (1938, p. 197)

in his view that the term luscladite has become too vague to be useful, and believe that there is little to be gained by its adoption in place of the term olivine theralite which we have employed. Comment may here be added concerning certain other rocks originally classed as nephelinites by Marshall (1912). The rock first noted by J. P. Smith on the beach at Omimi, fourteen miles northeast of Dunedin, has been found by the writer to be enclosed in a dyke of relatively fine-grained olivine nephelinite. The coarsely granular rocks form irregularly bunched segregations 1–3 feet wide, and may be drawn out into schlieren-streaks sometimes only 1·1–0·2 inches thick, passing by transitions into the surrounding nephelinite. The dyke probably fed a sheet of olivine nephelinite about 100 feet thick (possibly a sill rather than a flow), which extends for two miles under Porteous Hill, W.S.W. and W.N.W. of the exposure of the dyke on Omimi beach. Like the dyke, it varies in grain-size and texture, and also in its small content of feldspar, and encloses streaks of relatively coarse-grained material. We are not here dealing with independent hypabyssal coarsely-granular rock-masses as at Waihola, but with pegmatoid segregations such as Lacroix (1928) has described. Marshall's (1912, p. 306) clear account of the more coarsely granular rock together with his analysis thereof (No. 4) may be quoted, supplemented, and illustrated here. (See Fig. 3, 0.) “The olivine is” (often) “in extremely small needles, sometimes 1 cm. long but only 0·08 mm. wide. The direction of neighbouring crystals is remarkably parallel in longitudinal as well as transverse areas. They are similarly oriented over a considerable area.” It is a strongly magnesian type, ranging in composition from Fa18–Fa35 such as is normal in the atlantites. In some specimens the olivine has been more or less replaced by carbonates. “The phenocrysts of augite have the pleochroism and hour-glass structure as in the Waihola type.” There is, however, usually little variation in the optic axial angle (2V=(+)50°–58°) which indicates a fully calcic character in all but one case of a zoned crystal with a thin mantle of less calcic composition with 2V=(+)36°. “An appearance of lattice-structure in the feldspar (similar to that in the Waihola rocks) is very noticeable.” The bulk of the feldspar in the coarsely granular segregations is anorthoclase with transverse optical plane and 2V=(-)52°–68°, but andesine An45–An39 is present as well. “The nepheline is wanting in crystallographic boundaries and is usually intergrown in complete micrographic fashion with augite. In some instances this augite is in optical continuity with the large crystals. This micrographic growth is sometimes formed in the ground-mass in a very minute scale and constitutes its dominant feature.” (Fig. 3, 0, illustrates intergrowths of intermediate grain-size.) “The augite is sometimes slightly green in its smaller members. There are minute crystals of feldspar and apatite in the ground-mass. The larger crystals of apatite and ilmenite are the same as in the Waihola rocks.” In slides of rocks other than those studied and kindly lent to the writer by Marshall, notably in 1048—a small pegmatoid segregation—there are occasional sub-radial clusters of rhönite showing its characteristic pleochroism (X=deep brown, Z=greenish brown), oblique extinction up to 10° in sections nearly perpendicular to (010), and up to 30° in sections

Fig. 3

parallel thereto; refractive index about that of hornblende and birefringence near to 0·014. It is apparently a primary mineral in these rocks. In (1045), one of the more finely granular of these rocks differing but little from the enclosing nephelinite, a pegmatoid vein 4 mm. or more wide contains large grains of olivine (Fa0–Fa18) much more forsteritic than in the more coarsely granular rocks, and unzoned calcic titanaugite with large crystals of apatite and iron ores in a coarsely granular matrix of anorthoclase and nepheline, the latter in part graphically intergrown with pyroxene. A single small, almost rectangular plate of andesine (An45) is all the plagioclase seen in this veinlet. Zeolitisation has occurred to greater or less extent in many of these segregations. The only (and possibly rather inadequate) analysis of these rocks that is available suggests that little or no feldspar was present in the specimen investigated. The sheet of olivine nephelinite fed by this dyke, though doleritic in appearance (982, 983, 1069, 1082, 1100, 1121), is rarely as coarsely granular as the material of the large segregations in the Omimi dyke. It contains smaller idiomorphic crystals of titanaugite, partially bowlingitized olivine, occasionally (983) determinable as pure unzoned forsterite, magnetite and apatite set in a matrix of poikilitic nepheline more or less replaced by thomsonite (?) (and analcite?), and of coarsely granular pericline-twinned plagioclase of composition An54–51 in one rock (983) in which it is associated with dominant anorthoclase, or An45–38 in another (1100) without any separate potassic feldspar. The Domain Cricket Ground at Auckland, 650 miles to the N.N.E. from Dunedin, is situated in a crater in which large blocks of coarsely granular rock were found lying scattered on the land surface, their mode of occurrence being indeterminable, though probably they formed segregations in the associated basaltoid lava in which Marshall stated (1907, p. 366, 1912, p. 307) a little nepheline occurred. This, however, has not been confirmed by Bartrum* Private communication. The writer is indebted to Professor Bartrum for details concerning these rocks and the loan of representative slides. after long search. The rock contains large crystals of olivine (Fa8–16 with narrow marginal zones of Fa24–30), and is so richly pyroxenic that it (e.g. 7095) might be compared with ankaramite. Campbell Smith,* however, remarked in a letter (16, XII, 1929) to Professor Bartrum: “I should compare (this) with some of the limburgites with colour-less glass in the interstices. There is very little feldspar present; I do not recognise any nepheline, but the glass would probably gelatinise with HCl, and if so might be taken for nepheline. Limburgites of this kind must be very nearly related to the basanites or nepheline-basalts.” It seems probable that a chemical analysis of this rock would show the presence of normative nepheline as occurs in approximately coeval basaltic rocks in the district immediately south of Auckland. (Cf. Henderson, 1926.) If so, the term basanitoid, as redefined by Lacroix (1919) could be assigned to it. Marshall's description of the coarsely granular material (7085, 7091, 7101, 7107) may be illustrated by Fig. 3A and supplemented thus: It is somewhat similar to that of Omimi. In most specimens the augite in large grains has an ophitic form, but is less calcic than that of the Omimi rocks (2V=(+)42°±3°). It is strongly titaniferous

and shows hour-glass structure. The olivine crystals, again, have a great length, with elongation parallel to the a axis. The unzoned crystals in the most coarsely granular rocks have compositions between Fa37 and Fa49 and the zoned crystals Fa8–54, while in a rock of medium grain-size an unzoned crystal of Fa61 and zoned crystals of Fa46–61 were noted. The intergrowth of augite and nepheline is very complete, but is not carried to the extent of excessive fineness that is found in the Omimi type. The ground-mass is rather more plentiful and contains aegirine, apatite and feldspar. The chief difference between the pegmatoid rocks of Omimi and those of Auckland lies in the dominance of feldspar in the latter, 53·2% of normative Or29Ab33An38 being indicated in the rock analysed by Marshall (1912, No. 5 below), and 43% of normative Or29Ab25An48 in that collected by Iddings and analysed by Washington (1917, p. 571, No. 6 below). Universal stage measurements indicating the modal character of the feldspars are recorded in Table II herewith. It is not quite clear what is the most appropriate name for the coarsely granular rocks from Omimi and Auckland herein considered. As Marshall (1912) and Campbell Smith (loc. cit. supra) have noted, they have a general resemblance to the “nepheline dolerite” of Löbauer Berg in Saxony (Analyses No. 9 and 10)† For a detailed estimation of the (modal) mineral composition of this rock see Tscherwinsky (1929). The abundant apatite appears to be normal fluor-apatite as analysis of a less melanocratic specimen than those recorded in Table II herein, (sp. gr. 2.888), gives P2O5 1.65%, CaCl2 0.04%, CaF, 0.27%. which Rosenbusch (1908, vol. II, p. 1431) retained under the nephelinites. This rock, as shown by Stock (1888) and Siegert (1894), “forms schlieren in nepheline basalt of which the boundaries are not sharp, but denticulated and embayed” (Rosenbusch, 1907, p. 1432), as was clearly visible in the rock-exposures at that famous locality studied by the writer (Benson) in 1913. The New Zealand rocks, however, differ from those of Löbau, and also from the otherwise similar rocks of Fernando Noronha (Williams, 1889; Campbell Smith and Burri, 1933), also cited as analogues by Campbell Smith, in the fact that they contain abundant olivine and plagioclase. Campbell Smith therefore concludes (loc. cit. supra) “that the name nephelinite should not be applied to any of these coarse-grained ‘nepheline dolerites.’ They are not true lavas, but occur as small bodies, schlieren or inclusions and show many variations. There is a “close resemblance between the Auckland Cricket Ground boulders and the luscladite (Analysis No. 9) described by me (Smith and Chubb, 1927) from Rapa in the Austral Islands” (see below), “and I do not see any reason for not using that name unless it be that one ought to avoid using the name of a true intrusive rock for these small bodies and schlieren.” Lacroix (1928) has called such bodies pegmatoids, and has used this word in conjunction with the petrographic names of volcanic rocks of corresponding mineral-composition. He (1928, p. 325) instanced the rock at Omimi as an example of ankaratrite pegmatoid. Johannsen (1938, p. 366) and apparently Tröger (1935, p. 255), following Lacroix's (1916, p. 256) original description of ankaratrites, have made the presence of 10–8 per cent, of biotite the chief feature separating them from more normal melanocratic

olivine nephelinites, the tannbuschite of Johannsen (1938, p. 364). But as Holmes (1920, p. 32) and (tacitly) Lacroix (1927, p. 22, 1928, p. 325) do not recognise this as an essential feature of ankaratrite, there seems little to be gained by the use of either Lacroix's or Johannsen's term in place of the longer but more descriptive name, which may be applied to a rock containing a very subordinate amount of feldspar. Johannsen's (1938, pp. 232, 301) term nephelinite-basanite may well be useful in denoting rocks in which the feldspar is not greatly subordinate to the nepheline, while in nepheline basanites plagioclase is in excess. On these grounds the coarse-grained rocks of Omimi are termed olivine nephelinite pegmatoid, those of Auckland nepheline basanite pegmatoid. Brief mention may also be made of other rocks in New Zealand more or less closely related to those here discussed. Bartrum (1925, p. 10) has described a teschenite rich in analcite, and green-mantled titanaugite invading Cretaceous sediments at Mangapai on Whangarei Harbour about seventy miles north-north-west of Auckland.* For analysis see Ferrar, 1934. Sollas and McKay (1906, 11, pp. 155–7) have described with very striking microphotographs several teschenites invading Lower Cretaceous strata, but possibly represented by tuffs and dykes in later Cretaceous beds, about sixty miles east of Wellington. They are being described in geological and petrographic detail by Brown and Hutton. In the Middle Waipara Valley in North Canterbury about thirty miles north of Christchurch, Dr. R. S. Allan and the writer found a number of pebbles of hitherto undescribed rocks of doleritic appearance which were probably derived from intrusions into Lower Mesozoic sediments, and though not exactly like the East Wellington rocks, may have been co-magmatic with them. They also are free from hornblende. The affinity is most marked in the case of 5777, an originally olivine-rich “natrolite”† This term is used in a general sense for radiating zeolite not clearly distinguishable from natrolite in a micro-slice. The precise nature of the Waipara zeolite has not been determined.-teschenite. The tabular (0·8 mm.) labradorite (An53) is associated with a minor amount of orthoclase and is partly replaced by natrolite, though generally sharply idiomorphic against this locally abundant zeolite. Tyrrell's (1917) comment regarding analcite when paraphrased appears applicable to this rock. The zeolite was produced during “that late period in the history of the magma when the rock was stewing in a hot alkaline solution … Thus the (natrolite) was not derived from the feldspar but the alteration of the feldspar is due to the (natrolite).” Faintly tinted augite (0·7 mm.) is the dominant ferric mineral, but almost as abundant are chlorite-rimmed, slightly greenish-yellow bowlingite pseudo-morphs after olivine, with accessory magnetite, ilmenite, apatite, and rare minute flakes of biotite. An otherwise closely-comparable rock (5778) differs from this in the presence of but little zeolite, mainly analcite. A more coarsely granular natrolite†-teschenite (5772) also containing but little zeolite, contains large flakes of biotite almost as abundant as the pyroxene and much of a scattered reddish-brown strongly birefringent chloritic mineral, which when aggregated is sometimes suggestive of iddingsite-pseudomorphs after olivine. There

are also numerous porphyritic rocks with phenocrysts of two or all of the group labradorite (An55), titanaugite and olivine, set in a medium-grained basaltic ground-mass in which accessory biotite flakes abound. Zeolitisation of the phenocrystic and ground-mass feldspar has occurred to a small but varying extent, and there are occasional zeolite-filled vesicles (Nos.: 5771–2, 5774–6). The pebbles of olivine-biotite-dolerite and olivine-dolerite found by Thomson (1913) in the bed of the River Dee are possibly co-magmatic with these. They were derived from the varied complex of dykes invading the pre-Cretaceous argillites of Mount Tapuaenuka, the highest point of the Kaikoura Ranges, about a hundred miles north-east of the Waipara River. Though they were considered of pre-Cretaceous age by McKay, Thomson suggests that they may be co-eval with the as yet undescribed basaltic flows in the Mid-Cretaceous sediments of the adjacent Awatere Valley. Rocks from other Pacific lands more closely analogous to the Waihola, Omimi and Auckland theralites and basanites include the already mentioned luscladite of Rapa in the Austral Islands (Smith and Chubb, 1927, pp. 318–9. See Table 1, Analysis 11), which makes a plug 150 yards in diameter, and a number of other rocks occurring along the Main Dividing Range of Eastern Australia. Attention was first called by the writer (Benson, 1911) to the occurrence of nepheline-bearing basic igneous rocks at various points between 50 and 150 miles north of Sydney and later (1913) he noted their extension into south-eastern Queensland. Browne (1928, 1933) greatly increased the knowledge of these rocks and traced their extension into the south-eastern corner of New South Wales. Though the structural resemblance between some of these rocks and those of Löbauer Berg was noted, the presence of plagioclase caused the writer to term them olivine theralites or basanites. Some of them occur perhaps as basanite pegmatoids in more finely-grained lava (e.g. at Mount Warrawalong?), but others form independent intrusive masses either as sills or plugs (fide Browne, 1933, pp. 63–67), associated with Tertiary olivine basalts, basanites and other more or less alkaline basic lavas. (See Table I, Analysis 12.)

Table I.—Analyses of N.Z. Theralites and Pegmatoid Basanites and of Rocks Allied Thereto. 1 2 3 4 5 6 7 8 9 10 11 12 SiO2 36.00 40.50 43.14 45.30 46.60 46.66 45.61 44.46 39.88 39.43 41.56 41.86 Al2O3 14.51 14.89 17.77 16.44 16.79 18.57 14.35 13.14 15.37 10.36 11.16 14.40 Fe2O3 7.19 3.14 2.38 1.82 3.87 2.14 6.17 3.31 8.67 13.19 1.84 3.91 FeO 10.28 8.24 7.90 8.82 7.58 7.48 4.03 9.05 2.91 3.98 10.36 7.29 MgO 4.02 5.27 3.52 2.73 2.88 2.93 6.05 9.48 7.16 5.52 14.26 8.76 CaO 12.95 13.14 9.51 7.85 7.85 7.93 9.49 11.80 13.83 15.50 11.34 13.74 Na2O 3.61 3.96 6.24 8.60 5.18 6.49 5.12 3.20 4.73 4.23 2.84 3.21 K2O 3.04 2.04 2.10 4.05 3.31 2.79 3.69 1.45 2.01 2.24 1.49 0.91 H2O- 4.40 2.95 2.89 2.96 3.04 0.70 2.60 0.84 2.17 0.81 1.01 2.23 H2O+ 0.90 0.88 0.20 0.29 0.79 CO2 — tr. tr. — — — — — — — — 0.02 TiO2 2.50 2.52 2.04 0.71 1.76 3.00 1.96 2.84 1.04 2.27 3.25 2.30 P2O5 1.56 2.24 1.22 1.68 1.76 0.40 0.74 0.48 2.29 2.76 0.49 0.67 ZrO2 — nt. fd. nt. fd. — — — — — — — — — S — 0.04 0.07 — — — — — — — — — MnO — 0.16 0.15 — — 0.19 — — — — 0.14 0.21 BaO — 0.08 0.09 — — — — — — — — 0.04 SrO — 0.05 0.04 — — — — — — — — Abs. Cr2O3 — nt. fd. nt. fd. — — — — — — — — tr. V2O3 — 0.05 0.03 — — — — — — — — 0.01 NiO — 0.01 tr. — — — — — — — — tr. Cl tr. tr. — — — tr. F 0.15 0.10 — — — — Total 100.06 100.33† Less O for F. 100.25 and 100.02 respectively. Fluorine estimated by Willard and Winter's method, slightly modified. 100.07† 100.96 100.62 99.48 99.81 100.04 100.06 100.29 100.03 100.35 Sp. gr. — 2.910 2.777 — — — — — 2.918 3.058 3.089 3.023 1. Coarse “Nephelinte” (Olivine Theralite). Lake Walhola. P. Marshall (1912) Anal. 2. Olivine Theralite (5067). Lower middle portion of Sill, Lake Waihola. F. T. Seelye Anal. 3. Olivine Theralite (5061). Upper portion of Sill, Lake Walhola. F. T. Seelye Anal. 4. “Nephelinite” (Olivine nephelinite pegmatoid with micrographic structure). Dyke, Omimi, P. Marshall (1912) Anal. 5. “Nephelinite” (Nepheline basanite pegmatod). Isolated block, Auckland Domain. P. Marshall (1912) Anal. 6. “Nepheline Basanite” (Nepheline basanite pegmatoid). Auckland Domain. H. S. Washington (1917, p. 571) Anal. 7. Theralite. Average of six analyses. Daly (1933, p. 22). 8. Luscladite. Average of seven analysis. Lacroix (1920, p. 25). 9. Nepheline Dolerite. Löbauer Berg. Stock (1888) Anal. 10. Nepheline Dolerite. Löbauer Berg. Stock (1888) Anal. 11. “Luscladite” (Olivine Theralite). Rapa, Austral Islands. E. D. Mountain (in Smith and Chubb, 1927) Anal. 12. Nepheline Analcite Dolerte. Wharton's Mill, on road to Barrington Tops, New South Wales. W. G. Stone (in Browne, 1993, p. 69) Anal.

Appendix I. Optical Measurements with the Universal Stage. (F. J. Turner.) Optical measurements were made with a Universal stage, for representative grains of feldspar, olivine, zeolite, and augite in typical thin sections. These determinations were used as a check upon the ordinary petrographic examination covering a greater range of material. The procedure for feldspars was the same as that described in a recent paper (Benson and Turner, 1940, pp. 194–196). A note on the method of deducing the value of the true optic axial angle (2V) from universal stage measurements upon olivines (and augites), is appended in view of Tomkieff's recent discussion of this question (Tomkieff, 1939, pp. 234–236). The writer has followed Nikitin (1936, pp. 31, 32, pl. 1) in applying the correction for difference in refractive indices of olivine (β=1·72–1·78) and the hemispheres of the stage (μ=1·648), directly to all recorded tilts on E.W. or N.S. axes of the stage; at the same time crystals were selected in which tilting through low angles only was necessary, so that the corrections involved were always small. On the other hand, Tomkieff (op. cit., p. 235) gives a table for conversion of apparent to true optic axial angle in olivine, taking into account (a) the magnitude of apparent axial angle and (b) the corresponding mean refractive index β for olivine of all compositions. However, the appropriate correction to be applied to measurements made upon grains of constant true axial angle is not constant, but depends upon the orientation of the grain in relation to the plane of section; if low tilts upon the universal stage are involved, the correction to be applied is less than in cases where higher tilts are necessary. An actual example of a zoned olivine measured by the writer in section 5067 is cited to illustrate this discrepancy. Stage Readings. On inner circle. On NS Axis. On EW Axis. Z parallel to EW 10 ½° 14° Y parallel to E.W. 279° 10° Optic axis vertical, with Y parallel to EW axis of stage Inuer zone 29° Middle zone 24° Outer zone 20° From Nikitin (plate 1) the recorded tilts are corrected as follows: Recorded. Corrected. 10° 9 ¾° 14° 13 ½° 20 18 ½° 24° 22 ½° 29° 27 ¼° When the corrected tilts are plotted on a stereographic projection, the values of 2V obtained for the three zones (from within outward) are respectively 81 ½° (Fa34), 72° (Fa54) and 64° (Fa69). On the other hand, if the procedure given by Tomkieff is followed, the corresponding corrected values for 2V are 82° (Fa35), 70° (Fa58) and 61 ½° (Fa74) respectively. The discrepancy is noticeable only in cases where uniformly low tilts on the stage are recorded for olivines of high refractive index (i.e. rich in the fayalite molecule).

Table II.—Universal Stage Measurements Made by Dr. F. J. Turner for this Investigation. Slide. Olivine. Pyroxene. Feldspar. Zeolite Inner Zone. 2V Outer Zone. 2V Remarks. Inner Zone. 2V Outer Zone. 2V Z∧c Remarks. Determination and remarks. 5060 Waihola 54°–50° (+) strong ρ<p 44°(+) slight ρ<p Outer zone purple, inner brownish with sharp intervening boundary. 1. Plagioclase An40 in tabulae 2 mm. square. 2. Anorthoclase microlites 2V=48°(−) optic axial plane transverse to 010. 5061 Waihola — — — — — — — 1. Plagioclase An40 surrounded by Anorthoclase 2V=66°(−) with optic axial plane perpen-dicular to 010. The two feld-spars show both pericline and albite twinning and are in parallel crystallographic position. 5064 Waihola 76°(−) 80°(−) 82°(−) 72°(−) unzoned No marked elongation 54°(+) 56°(+) 58°(+) 58°(+) Purple, with outer zone very narrow. 1. Plagioclase An41. 5067 Waihola 81 ½°(−) 73 ½°(−) 65°(−) All three measurements on 1 crystal. See Fig. 2 64°(+) *62°(+) 56°(+) 52°(+) 47°(+) 44°(+) 54°(+) 42°(+) 46°(+) — — — — 46° 51° ± 3° 52° ± 3° *Outer zone rather broad, with narrow greenish rim in which 2V=72°(+). Central zone sharply limited and all zones purple. 1. Anorthoclase laths 1.5 × 0.2 mm. 2V=45°(−). Optic axial plane perpendicular to (010). Partly replaced by zeolite. 2. Similar laths of plagioclase An38. Zeolite 2V=50°–60°(+) in several measurements elongated parallel to Z. 5068 Waihola 74°(−) 73°(−) 70°(−) 67°(−) Very narrow outer zone 49°(+) 48°(+) 48°(+) —58°(+) 50°(+) 45° ± 3° 53° ± 3° 40° 43° 48° Purplish core with very pale outer zone. 1. Anorthoclase in tabulae—less than 2×1 mm. 2V=45°(−). Optic axial plane often transverse to (010). 2. Small laths of probably similar composition. 5698 Waihola 73°(−) Not appreciably zoned. Partial alteration to iddingsite 48°(+) ±02°(+) 63° (+) 50°(+) Inner zone deep purple. Outer pale purple followed by greenish mantle with 2V=58°(+). Purplish unzoned. 1. Plagioclase An39 in tabulae 2×2 mm. 2. Anorthoclase microlites with 2V=48°(−) and optic axial plane transverse to (010). Zeolite, elongated parallel to Z. D.R. = 0.010 approx. Slightly greater than that of feld-spar. ·2V=60° ± 5°(+) in several measurements. 30°(+) in one only (Thomsonite). 5699 Waihola 81°(−) 76°(−) Ophitic with very narrow outer zone 54°(+) 46°(+) 45° ± 3° Crystals ophitic with very sharply distinct zones. Plagioclase tabulae with Carlsbad, albite, and sometimes pericline twinning. 1. In olivine (a) An37 (b) An90 (c) An90 with narrow outer zone of An72. 2. In augite (a) An82 (b) An79 3. Free zoned tabulae, centre An85, narrow border An45. 4. Free unzoned laths rare An44. 5700 Waihola 88°(−) °80°(−) 80°(−) ±74°(−) Nearly perpen-dicular to optic axis. Variation of extinction between inner and outer portion. See Fig. 2 983 Porteous Hill (Sill?) 88°(+) Unzoned 1. Anorthoclase dominant. Carlsbad and pericline twinning. 2V = 52°–68° (−). Axial plane transverse to (010). 2. Plagioclase An54–51. 1045 Omimi dyke. (a) Main mass 88°(−) ±81°(−) 82°(−) Narrow outer zone. Unzoned 58°(+) 54°(+) 50°(+) 56°(+) — 36°(+) — 40° 40° Unzoned. Good determination of Z ∧ c. 1. Anorthoclase dominant with 2V=66°(−). 2. Plagioclase subordinate An54–51. (b) Narrow pegmatoid veinlet 86°(+) 88°(−) Large grains — — — Almost wholly anorthoclase with 2V=55°(−) to 70°(−). One crystal has a small inclusion of plagioclase An45. 1100 Porteous Hill 1. Plagioclase An45–38. No anorthoclase determinable. 7085 Auckland Domain Nepheline basanite pegmatoid 80°(−) DR. = 0.041 assuming DR. of Ab42 = 0.008 42° ± 3°(+) Not markedly zoned. 1. Plagioclase An65–54 unzoned. abundant and coarsely granular. 2. One small crystal of An54 sur-rounded by pericline-twinned anorthoclase 2V=62°(−), followed by incomplete mantle of K feldspar with 2V=40°(−). 3. Anorthoclase (or sanidine) microlites. 43F Auckland Domain Coarse basanite pegmatoid 86°(−) 74°(−) 80°(−) 72°(−) Narrow outer zone. Unzoned crystals elongated ‖ to a. Medium-grained basanite pegmatoid 72°(−) 76(−) 68°(−) 68°(−) 72°(−) Crystals elongated ‖ to a. Square cross-section ⊥ to a. Slides in Collection of Geological Department, Auckland University College. 22 Fine-grained basanitoid? adjacent to coarse-grained rock 89°(−) 88°(+) 88° (+) 83°(−) 86°(−) 86°(−) Very large grains with narrow outer zones.

Appendix II. Descriptions of Zeolite, Titanaugite and Apatite in an Olivine Theralite at Lake Waihola. By C. O. Hutton. The zeolite separated from theralite 5698 described above is apparently homogeneous. The specific gravity at 19° C. is 2·25–2·26, and the refractive indices repeatedly determined using a variety of oils to avoid any effects of base exchange, give consistently the values α=1·5073, β=1·5095, γ=1·5160 (each ±0·0003) with γ–α 0·0087. These show that the mineral is not natrolite, but is more nearly allied to the thomsonite group. The optic plane and acute positive bisectrix are parallel to the length of the fibre in nearly all cases in material mounted in Canada balsam. Rarely, however, the fibres are twinned, one segment being positively elongated, the other negatively. This also holds true when the mineral has been mounted in oil, after being warmed on the hot plate just sufficiently to permit mounting in Canada balsam. But if the mineral be mounted in oil without this brief heating, nearly all the fibres are negatively elongated, and but few positively. The same is true of the heated mineral which has been subsequently cooled and mounted in a drop of water. It would seem as if the heating brought about sufficient dehydration to change the optical properties significantly, but the original condition is regained when water is returned into the crystalline mesh. The optical orientation makes it difficult to measure accurately the value of 2V. That calculated from the above refractive indices is about 50°(+), that measured by Dr. Turner on the heated mineral mounted in balsam is generally nearer 60° ± 5°(+) but rarely 30°(+). The composition of the zeolite as determined by Mr. F. T. Seelye is peculiar in its content of K2O, which is abnormally high for a member of the thomsonite group of zeolites. Table III—Analysis of Zeolite. Waihola Zeolite. Mol. Prop. Atomic ratio 80 atoms of 0 in unit cell. Hey's Thomsonite No. 1. SiO2 42.10 0.701 22.91 40.3 Al2O3 26.87 0.263 17.19 28.5 Fe2O3 0.21 FeO 0.36 MgO 0.35 CaO 5.16 0.092 3.00 11.2 Na2O 8.50 0.137 8.95 5.7 K2O 2.63 0.028 1.83 nil H2O+ 11.35 H2O− 2.57 0.773 25.26 14.1* Assuming Hey is using total Water. CO2 nt. fd. TiO2 0.03 P2O5 0.01 MnO 0.03 BaO nt. fd. SrO 0.04 Cl nt. fd. 100.21 99.9

After considerable search it has not been possible to find an analysis completely comparable with that of the Waihola zeolite. The nearest approach is one listed by Hey (1932, p. 54, table 1) which is given above. It would appear from this that the Waihola zeolite is a potash-bearing member of the thomsonite group near its extreme end containing thomsonite rich in SiO2 and poor in Al2O3. The refractive indices of the Waihola material are also in accord with the generally decreasing value of the indices towards this end of the group. Further, its analysis, recalculated on the basis of 80 atoms of oxygen per unit cell, would tend to confirm its position in that group, though K2O is more abundant in it than in any thomsonite described by Hey. Its formula would appear to be Ca3Na8.95K1.83 Al17.19S22.91O80 25·26 H2O with Al+Si=40·10 and Ca+Na+K = 13·78. These figures agree fairly well with Hey's data, and the latter would seem to show that the Na⇌K⇌Ca substitution is of importance. No name is suggested here for this zeolite. Certainly Hey's (1933) work on ashcroftine leaves the term kalithomsonite available, but it seems undesirable to revive it by its application to the Waihola zeolite. The titanaugite (separated from the thin greenish mantle enriched in aegirine-augite) has the following properties:—Sp. gr. at 16° C. = 3.425±0·005. Refractive indices α = 1·698, β = 1·704, γ = 1·724 (each ±0·001) and γ–α = 0·026. The value of 2V calculated from these figures is about 60°(+), that measured by Dr. Turner on the purplish augites in this rock varies from 48°(+) to 63°(+). In the absence of twinning it was not possible to measure the angle Z ∧ c. The composition according to Seelye's analysis is as follows:— Table IV—Analysis of Augite. Waihola T. augite. Waihola Mol. Prop. Atomic Ratio to 6 (0.0H). (1) (11) SiO2 45.28 0.754 1.724 2.00 45.56 44.71 Al2O3 7.60 0.075 0.343 0.276 0.067 8.15 7.85 Fe2O3 2.45 0.036 0.068 2.46 4.46 FeO 7.48 0.104 0.238 5.46 4.23 MgO 10.31 0.257 0.587 11.88 11.74 CaO 22.45 0.401 0.917 22.84 22.37 Na2O 0.60 0.010 0.023 1.02 0.90 K2O 0.07 — — 0.62 0.09 H2O+ 0.26 — — — 0.26 H2O− 0.12 — — 1.98 — 0.09 CO2 nt. fd. — — — — TiO2 2.88 0.036 0.082 1.87 2.92 P2O5 0.40 — — — — 0.12 V2O3 0.047 — — — — — S 0.02 — — — — — MnO 0.18 0.002 0.004 0.42 0.10 NiO tr. BaO nt. fd. — — SrO< 0.01 Cl nt. fd. 100.16 100.28 99.84* Min. Abstracts, vol. 24, p. 212, gives 100.34.

Analysis (I) by A. Hueber (in L. Jugovics. Constitution of the Basalt Plateau of Mount Medves and its Crystal Tuff. Mat. Természettud, Ertesitó, Budapest. 1934. Vol. 51, pp. 443–470) is of an augite with sp. gr.=3·31, 2V=57°2′, and Z∧c=45°2′ to 47°5′. A better comparison with the Waihola titanaugite is, however, afforded by (II) a titanaugite from a monzonitic teschenite (W. Wawryk. Sur l'augite commune et titanifère des teschénites en Pologne, Arch. Min. Tow. Nauk. Warszaw. 1935. Vol II, pp. 175–181). On this augite 2V=51°. Z∧c=50°, and α=1·721, β=1·725, γ=1·746; sp. gr.=3·401. The main difference between this and the Waihola titanaugite lies in the higher ratio of Fe2O3 to FeO, though the total iron content remains about the same. It is significant that the higher oxidation of the iron in (II) appears to have caused considerable increase in the refractive indices. The presence of P2O5 in the Waihola augite is attributed to the occurrence of fine needles of apatite within the pyroxene grains, and not to the presence of apatite grains associated externally with the purified (centrifuged) powder. Allowance has been made for the lime associated with the P2O5 in calculating the formula for this augite. On the basis of 6 (O.OH) atoms to the unit cell, the analysis agrees well with the structural formula XY (SiAl)2(O+OH, F)6 suggested by Machatschki (1929) yielding, as shown above, the formula (Mg, Fe”, Fe”′, Al, Ti, Ca, Mn, Na)1.98 [(Si, Al2)O6]. Here the Al is split between the Si and XY groups to satisfy the silicon chains of the pyroxene structure. Although the augite is fairly titaniferous, none of the Si is replaced by it, as apparently can take place in augites high in Ti but low in Al. If the Fe2O3 be calculated as 2FeO and the Al2O3 and TiO2 be ignored, there is not quite sufficient silica to give the formula Wo48En36Fs22 which would correspond with positions on Deer and Wager's (1938) triangular diagrams which otherwise accord fairly well with the optical properties observed. Apatite separated from the same rock has sp. gr.=3·16±0·01 and ω=1·6370 and ω=1·6339, both ±0·0002. These figures are within the ranges given by Hausen (1929) for the normal apatites. The composition calculated from Seelye's analyses of the Waihola theralites is that of a nearly pure fluor-apatite. Bibliography. Baker, G., 1941. Apatite Crystals with Coloured Cores in Victorian Granitic Rocks, Amer. Min., vol. 26, pp. 382–390. Barth, T. F. W., 1931. Mineralogical Petrography of Pacific Lavas, Amer. Journ. Sci., vol. (5), 21, pp. 377–405, esp. 389. Bartrum, J. A., 1925. The Igneous Rocks of North Auckland, New Zealand, Gedenboek Verbeek. Verhandl. v.h., Geol-Minjb. Genootschap voor Ned. en Kolonien, Geol. Serie viii, pp. 1–16, s'Gravenhage. Benson, W. N., 1911. Preliminary Note on the Nepheline-bearing Rocks of the Liverpool and Mount Royal Ranges, Proc. Roy. Soc. N.S.W., vol. 45, pp. 176–168. —– 1913. The Geology and Petrology of the Great Serpentine Belt of New South Wales, Part iii, Proc. Linn. Soc. N.S.W., vol. 38, p. 703. —– 1941. Mugearites in the Dunedin District, Ibid., vol. 70, pp. 188–189. Benson, W. N., and Turner, F. J., 1940. Mineralogical Notes from the University of Otago, No. 2, Trans. Roy. Soc. N.Z., vol. 69, pp. 56–72.

Bowen, N. L., 1933. The Broader Story of Magmatic Differentiation briefly told, (In). Ore Deposits of the Western States. Amer. Inst. Min. and Metal. Engineers, pp. 108–128, esp. 123, New York. Browne, W. R., 1928. Petrological Notes on Some New South Wales Alkaline Basic Rocks, Proc. Roy. Soc. N.S.W., vol. 61, pp. 371–382. —– 1933. An Account of the Post-Palaeozoic Igneous Activity in New South Wales, Ibid., vol. 67, pp. 9–94. Daly, R. A., 1933. Igneous Rocks and the Depths of the Earth, McGraw-Hill, p. 22, New York. Deer, W. A., and Wager, L. R., 1938 Two new Pyroxenes included in the system Clino-enstatite, Clino-ferrosilite, Diopside and Hedenbergite, Min. Mag., vol. 25, pp. 15–22. Drescher, F. K., and Krueger, H. K. E., 1927. Der Peridotit von Kaersut (Grönland), und sein Gangefolge als Beispiel einer Sekretions-differentiation, Neu. Jahrb. für Min., Beil. Bd. 57, Abt. A, pp. 569–616, esp. 609–610. Edwards, A. B., 1938. The Tertiary Volcanic Rocks of Central Victoria, Quart. Journ. Geol. Soc., vol. 94, pp. 243–320, esp. 299–301, 303. —– 1938a. The Formation of Iddingsite, Amer. Mineralogist, vol. 23, pp. 277–281. Elsden, J. V., 1908. The St. David's Head “Rock Series” (Pembrokeshire), Quart. Journ. Geol. Soc., vol. 64, pp. 273–293, esp. 289–290. Ferrar, H. T., 1934. The Geology of the Dargaville Subdivision, N.Z. Geol. Surv. Bull., no. 34, pp. 55–6, 62, Analysis 21. Gilluly, J., 1927. Analcite Diabase and related Alkaline Syenite from Utah, Amer. Journ. Sci., vol. 214, pp. 199–211, esp. 205–207. Hausen, H., 1929. Die Apatite, deren chemische zusammensetzung, und ihrer Verhältnis zu physikalischen und morphologischen Eigenschaften, Meddel. frän Abo Akademis Geol.-Min. Institut, no. 9. Henderson, J., 1926. The Geology of the Huntly-Kawhia Subdivision, N.Z. Geol. Surv Bull., no. 28, pp. 70–1, Analyses 15–18. Hey, M. H., 1932. Studies on the Zeolites, Part ii. Thomsonite (including faroelite and gonnardite), Min. Mag., vol. 23, pp. 51–125. —– 1933. Studies on the Zeolites, Part iv. Ashcroftine (Kalithomsonite of S. G. Gordon), Ibid., pp. 305–8. Holmes, A., 1920. The Nomenclature of Petrology, Murby and Co., London. Johannsen, A., 1938. A Descriptive Petrography of Igneous Rocks, vol. 4, Univ. of Chicago Press. Lacroix, A., 1916. La Constitution des roches volcaniques de l'extreme nord de Madagascar et de Nosy bé; les ankaratrites de Madagascar en général, Comptes Rendus, vol. 163, pp. 256–8. —– 1919. La Constitution mineralogique et chimique des lavas des vulcans du Tibesti, Ibid., vol. 140, pp. 401–7, esp. 402. —– 1920. La Systematique des Roches grenues à plagioclase et feld-spathoides, Comptes Rendus, vol. 170, pp. 21–26. —– 1927. La Constitution lithologique des Iles volcaniques de la Polynesia Australes, Mem. de la Acad. des Sciences, vol. 59, pp. 1–82, esp. 10, 35–6. —– 1928. Les Pegmatoides des Roches volcaniques à facies basaltiques, Comptes Rendus, vol. 187, pp. 321–6. Machatschki, F., 1929. Uber die Formel der Monoklinen Amphibolen und Pyroxenen, Zeits. Krist., vol. 71, pp. 219–236. Marshall, P., 1907. Distribution of Igneous Rocks of New Zealand, Trans. Aust. Assoc. Adv. Sci., vol. 11, p. 366. —– 1912. Nephelinite Rocks of New Zealand, Trans. N.Z. Inst., vol. 44, pp. 304–7. Morey, E. W., and Ingerson, E., 1937. The Pneumatolytic and Hydrothermal Alteration and Synthesis of Silicates, Econ. Geol., vol. 32, Suppl., pp. 607–761. Nitikin, W., 1936. Die Federox Methode, Gebrüder Borntraeger, Berlin.

Ongley, M., 1939. The Geology of the Kaitangata-Green Island Subdivision, N.Z. Geol. Surv. Bull, no. 38, Govt. Printer, Wellington. Rosenbusch, H., 1907. Mikroskopische Physiographie der Mineralien und Gesteine, Bd. ii, Erste Hälfte, Schweizerbart, Stuttgart. —– 1908. Ibid., Bd. ii, Zweite Hälfte. Rosenbusch, H., and Osann, A., 1922. Elemente der Gesteinslehre, pp. 223–229, Schweizerbart, Stuttgart. Seigert, Th., 1894. Erlauterung zu Sektion Löbau-Herrenhut der Geol. Spezielkarte von Sachsen, Leipzig. Shand, S. J., 1939. On the Staining of Feldspathoids, and on Zonal Structure in Nepheline, Amer. Min., vol. 24, pp. 508–513. Smith, W. C., and Chubb, L. J., 1927. The Petrography of the Austral Islands and the Geology of Maiao, Quart. Journ. Geol. Soc., vol. 83, pp. 317–413, esp. 318–322. Smith, W. C., and Burri, C., 1933. The Igneous Rocks of Fernando Noronha, Bull. Suisse de Min. et Petr., vol. 13, pp. 405–43. Sollas, W. J., and McKay, A., 1906. Rocks of the Cape Colville Peninsula, vol. 2, pp. 155–7. Govt. Printer, Wellington. Spencer, E., 1937. The Potash Soda Feldspar I. Thermal Stability, Min. Mag., vol. 24, pp. 454–494. —– 1938. Ibid., II. Some Applications to Petrogeneis, Ibid., vol. 25, pp. 87–118. Stock, J., 1888. Die Basaltgesteine des Löbauer Bergs, Tschermak Min. Petr. Mitt., vol. 9, p. 429. Thomson, J. A., 1913. On the Igneous Intrusions of Mount Tapuaenuka, Marlborough, Trans. N.Z. Inst., vol. 45, pp. 308–315. Tomkieff, S. I., 1939. Zoned Olivines and their Petrogentic Significance, Min. Mag., vol. 25, pp. 229–251. Tröger, E., 1935. Spezielle Petrographie der Eruptivgesteine, Deutsch. Min. Ges., Berlin. Tschirwinsky, P., 1929. Quantitative mineralogische und chemische Zusammensetzung des Nephelinits vom Löbauer Berg, Sachsen, Centbl. für Min. Geol. Pal., Abt. A, pp. 207–11. Turner, F. J., 1940. Structural Petrology of the Schists of Eastern Otago, New Zealand, Amer. Journ. Sci., vol. 238, pp. 73–106, 158–191. Tyrrell, G. W., 1917. The Picrite-Teschenite Sill of Lugar, Quart. Journ. Geol. Soc., vol. 72, pp. 84–131. —– 1928. Some Dolerite-Sills containing Analcite-Syenite in Central Ayrshire, Ibid., vol. 84, p. 540. Walker, F., 1930. The Geology of the Shiant Isles, Ibid., vol. 86, pp. 355–395. Washington, H. S., 1917. Chemical Analyses of Igneous Rocks, U.S. Geol. Surv. Prof. Paper, no. 99, p. 571. Williams, G. H., 1889. Geology of Fernando de Noronha, Part II, Petrography, Amer. Journ. Sci., vol. (4), 37, pp. 178–189. Corrigenda. Dr. C. O. Hutton has kindly indicated that certain rocks in the collections of the New Zealand Geological Survey and described by the writer in Basic Igneous Rocks of Eastern Otago and their Tectonic Environment, Part II, Trans. Roy. Soc. N.Z., vol. 72, pp. 85–110, were not cited by numbers corresponding with those they bear in the Survey's official catalogue. The following corrections should be made:— Page 92—Locality 4. For P.5561 read P.4129. Page 93—Locality 9. For P.5167 read P.4167. Locality 19. For P.5597 read P.5594. Page 94—Locality 23. For P.5598 read P.5595. Locality 24. For P.5576 read P.5570. Locality 25. For P.5577 read P.5574. Locality 26. For P.5584 read P.5578.