Tinguaites and Camptonites from the Vicinity of Haast Pass.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 62, 1931-32, Page 215

Tinguaites and Camptonites from the Vicinity of Haast Pass. By F. J. Turner, Otago University. [Read before the Otago Institute, 14th April, 1931; received by Editor, 20th April, 1931; issued separately, 31st March, 1932.] Contents. Introduction. Petrography: (1) Tinguaites. (2) Camptonites. Mutual Relationship of Tinguaites and Camptonites. Course of Crystallisation of the Parent Magma. Correlation and Age. Acknowledgment and Thanks. Literature Cited. Introduction. The rocks described in this paper were collected early in 1930 from boulders in the beds of streams on either side of the Main Divide, in the vicinity of Haast Pass, on the boundary between North-west Otago and South Westland. On the Otago side of the divide, the boulders in question occur in the bed of the Makarora River for some distance below its junction with the Fish, as well as in the latter stream itself. At the point where the Makarora joins the Fish, it emerges from a precipitous, impassable canyon, so that it was not possible to ascertain whether boulders of the same type are to be found in the Makarora, upstream from the confluence. On the Westland side of the pass, igneous rocks are not represented among the boulders in the upper part of the Haast; but they were found to be plentiful in the bed of a small stream which enters the Haast from the west, about one mile below its junction with the Burke. Thus, although these rocks were never actually observed in situ, it is safe to assume that their source lies in the wild and unexplored ranges that stretch northward from the headwaters of the Fish, across the Burke, towards the junction of the Haast and the Lands-borough. The whole of this region is composed of a varied assemblage of mica-, chlorite-, and epidote-schists, which are continuous with the schists of Central Otago further south-east. The igneous rocks in question must therefore occur as dykes invading this schist formation.

Petrography. I.—Tinguaites: Specimens Nos. 1293,* Unless otherwise stated, all numbers refer to specimens and sections in the Otago University collection. 1298, 1299, and 1300 were collected from the stream flowing into the Haast one mile below the mouth of the Burke. Nos. 1294, 1295, 1296, and 1297 were obtained from boulders in the Makarora River, two miles below the confluence with the Fish River. 1294, Cancrinite-tinguaite. In hand specimen, a green, evengrained rock without noticeable phenocrysts. In section it is seen to consist of feldspar 45%, cancrinite 25%, aegirine 15%, zeolite 10%, sodalite 5%, and accessory apatite. (These figures, and all similar percentages in this paper, are approximate estimates only and are not based on micrometric measurements). The feldspar is in subidiomorphic tabular crystals about 2 mm. in length, being considerably larger than crystals of the other minerals present. They are untwinned or simply twinned, with refractive index well below that of Canada balsam, and invariably show pronounced perthitic striping parallel to the direction of elongation. Slender acicular inclusions of aegirine are sometimes present. The optic sign is positive, and the extinction angle in sections perpendicular to the acute bisectrix Z is 14° (measured from × to the basal cleavage). The main mass of each crystal is therefore albite carrying up to perhaps 10% of the anorthite molecule, and the feldspar as a whole may be classed as an albite-rich perthite. Cancrinite is abundant in clear allotriomorphic or subidiomorphic crystals about 0.3 mm. in diameter. It is clearly of primary origin, and is unaltered, except in one or two instances where there is come change to muscovite. Nephelite is absent; but there are numerous sharply bounded hexagonal and rectangular aggregates of finely crystalline zeolitic material, which are almost certainly pseudomorphs after nephelite. They measure up to 1 mm. in length, and are scattered throughout the whole rock. The zeolite is in minute acicular prisms with a more or less radial arrangement in the hexagonal aggregates. The refractive index is considerably less than γ for cancrinite (1.515 to 1.524), but slightly greater than α (1.491 to 1.502). The elongation of the fibres is positive and the extinction is straight or possibly very slightly oblique. The birefringence is distinctly higher than that of the feldspar, and is probably about 0.012 to 0.015. The positive elongation and straight or only slightly inclined extinction exclude all the fibrous zeolites except natrolite, epistilbite, and the rare erionite (Winchell, 1927, p. 389). The last-named may be excluded on account of its low refractive index and its chemical composition. Epistilbite is also unlikely on account of its calcic composition, extinction angle of 10°, and slightly low birefringence. All the properties agree well with natrolite, except that the refractive index of natrolite is 1.473 to 1.493, while the mineral in question undoubtedly has a refractive index as high as 1.500. It is therefore identified

provisionally as natrolite, a conclusion which is, however, strengthened by the well-known tendency for nephelite to alter to natrolite. There are, throughout the section, small but constant amounts of a colourless, clear, isotropic mineral which occurs usually as interstitial grains reaching 0.5 mm. in diameter, but sometimes as irregular patches invading and even enclosing crystals of feldspar. The refractive index is very low—less than that of cancrinite—and there are very faint signs of cleavage. Chemical tests on the crushed rock indicate the presence of abundant chlorine, so that the mineral may be identified as sodalite. The ferromagnesian constituent is abundant aegirine in slender idiomorphic prisms about 0.5 mm. long (Fig. 3), strongly pleochroic, with Z = light brownish yellow, Y = deep green, and × = slightly deeper green. Some of the larger crystals and crystal clusters contain cores of pale, partially chloritised, greenish pyroxene. Iron ores are absent, but there are small prisms of apatite. Duplicate sections of the same rock show several large crystals consisting of brown hornblende, rimmed with aegirine. One of these duplicates (S.W. 66a, in the Auckland University College collection) contains two interesting xenocrysts, which evidently originally consisted of olivine. One is 5 mm. in width, and has a central core of much shattered, slightly altered olivine, surrounded by a border 1 mm. in width, of fibrous talc mixed with a minor amount of pale chlorite. This in turn is bordered by a narrow fringe of fibrous bluish pennine, which passes outwards into an irregular rim consisting of small crystals of aegirine. The second xenocryst is only 2 mm. in diameter, and the central portion consists this time entirely of finely crystalline talc, immediately adjacent to which is a narrow, irregular fringe of colourless monoclinic pyroxene. This passes outward into a wide rim of hornblende, strongly pleochroic (pale reddish-brown to dark brownish-green), and having an extinction angle of 22°. This in turn is bordered with a well-defined band of dark green aegirine, adjacent to which the hornblende assumes a much deeper tint of brown. Bluish pennine is interspersed with the aegirine. 1293, Tinguaite. This is a fine-grained, light green rock with phenocrysts of feldspar reaching 5 mm. in length. It is holocrystalline and porphyritic, with abundant phenocrysts of feldspar and a few of pyroxene, set in a fine groundmass of feldspar, aegirine, and a little sodalite. The large feldspars are untwinned or simply twinned, with a lower refractive index than Canada balsam, and positive interference figure with an optic axial angle approaching 90°. They are therefore identified as albite-oligoclase. They are frequently flecked with calcite, and show marked undulose extinction. The feldspar of the groundmass is mostly of the same type, though a few lath-shaped individuals, showing the simple Carlsbad twin, may be orthoclase. Pale purplish, titaniferous augite, somewhat chloritised, forms rare phenocrysts, about 1 mm. in diameter, rimmed with dark green aegirine, while duplicate sections show occasional phenocrysts of aegirine-augite similarly bordered. Aegirine is itself abundant in

slender prisms 0.2 mm. long, with strong pleochroism (deep green to light brownish yellow), and small extinction angle. Orthopinacoidal twinning is rarely shown. There are a few small patches of finely crystalline (?) natrolite as in section 1294, probably representing original nephelite, while a little colourless sodalite is also present in the groundmass. Magnetite and apatite occur as minor accessories. Secondary calcite is plentiful throughout, for the most part occurring as an alteration product of the feldspar, while there are one or two small grains of secondary epidote. II.—Camptonites: 1295, Camptonite. The rock is holocrystalline, with the abundant ferromagnesian minerals and the panidiomorphic structure characteristic of the lamprophyres. The chief constituents are amphibole 40%, pyroxene 25%, altered olivine 10%, and partially altered feldspar 25%. Only pyroxene and olivine occur as phenocrysts. The pyroxene is pink titaniferous augite, and occurs both as plentiful stout idiomorphic, prismatic phenocrysts, 1 mm. to 2 mm. in length, and to a less extent as smaller, more irregular grains in the groundmass. In a few of the larger crystals, a central core of somewhat granular augite, intermixed with small ragged fragments of brown hornblende, is surrounded by a border of clear pink augite. On the other hand, several phenocrysts show a central, rounded crystal of aegirine-augite rimmed with normal pink pyroxene. The amphibole is barkevikite, and, like the pyroxene, is clear and unaltered. It forms short idiomorphic prisms about 0.2 mm. × 0.15 mm., with a maximum extinction angle of about 15°. It is strongly pleochroic, with × = light yellow, Y = deep reddish brown, and Z = deep brown (absorption × < Y < Z), while the birefringence is about 0.21 and the elongation is positive. The olivine originally present is completely decomposed, and is now represented by pseudomorphs consisting largely of pale green serpentinous material, with subordinate grains and rhombohedra of carbonate. The green mineral is almost uniaxial, negative, with positive elongation and moderate birefringence (about 0.015 to 0.017). It is identified as a ferriferous serpentine close to bowlingite. The feldspar occurs only in the groundmass, either interstitially or as small subidiomorphic crystals ranging up to 0.5 mm. in length. It is much altered to calcite and kaolin, and this, together with strongly undulose extinction prevent determination of optic sign and extinction angle. Traces of lamellar twinning, and a refractive index evidently higher than that of Canada balsam, nevertheless indicate that the feldspar is plagioclase. Granular magnetite and slender prisms of apatite are the only accessory minerals. Section 1295 shows a single irregular inclusion about 1.5 mm. × 1 mm., consisting of plagioclase with two sets of multiple twins, and coarsely crystalline secondary calcite. The negative optic sign of the feldspar, and the extinction angle of -68° (measured from Z to the basal cleavage) in a section perpendicular to the acute bisectrix X, correspond with basic bytownite. The whole inclusion is fringed by a reaction rim consisting of radially arranged prisms of aegirine-augite, with minor granules of pink augite. A duplicate

section shows another inclusion consisting of small crystals of indeterminate plagioclase and aegirine-augite, together with secondary calcite. 1296, Camptonite. A fine-grained, dark rock, in which small black prisms of amphibole and yellowish brown grains of olivine are clearly visible with the unaided eye. The chief minerals, as seen in section, are amphibole 30%, pyroxene 25%, olivine 15%, biotite 1%, feldspar 25%, analcite 3%, iron ores 1–2%. The structure is porphyritic, with large phenocrysts of olivine and smaller ones of amphibole and pyroxene, set in a groundmass consisting of all the minerals except olivine. The amphibole is strongly pleochroic, deep brown barkevikite which forms idiomorphic phenocrysts about 2 mm. in length, and also smaller, well-shaped crystals in the groundmass. Many of the phenocrysts have a deep brown, sharply defined central zone, surrounded by a border of lighter colour, but showing the same type of pleochroism. Several large phenocrysts (Fig. 1) have been almost completely converted into granular magnetite and pale greenish pyroxene, with residual barkevikite still showing in scattered fragments throughout. Such resorbed crystals have a very regular border of clear pink augite, and this in turn is irregularly edged with aegirine. The pyroxene is mainly pale pink augite in plentiful idiomorphic prisms averaging 0.5 mm. in length, and in grains of smaller size. Some of the larger phenocrysts have a sharply defined central zone of green aegirine-augite, while in others again the centre is pink augite and the border is colourless augite. A number of the smaller crystals of augite in the groundmass are rimmed with barkevikite, while others again show a narrow broken fringe of deep green aegirine. The latter mineral also occurs as irregular borders to some of the small barkevikites, and also as scattered individual grains in the groundmass. The olivine takes the form of large phenocrysts (reaching up to 5 mm.), which are clear and unaltered, except for occasional slight marginal transition to a pale green bowlingite. It is a normal magnesian olivine, with positive optic sign and fairly wide optic axial angle. The olivines are always bordered with reaction rims consisting of small crystals of barkevikite, biotite and barkevikite, or pink augite. Cracks in the phenocrysts, which have been penetrated by the reacting magma, are also filled with biotite and barkevikite. Biotite is also present in small amount throughout the groundmass as small subidiomorphic flakes with very pronounced pleochroism from light yellow to very deep brown (almost black). The feldspar includes both orthoclase, in slender simply twinned laths, and sodic plagioclase, the latter predominating. The plagio-clase is much altered, but occasional traces of albite twinning and a refractive index less than that of Canada balsam can still be observed. A duplicate section shows a single large rounded xenocryst (2 mm. in diameter) of anorthoclase, with characteristic multiple twinning. The accessory minerals include apatite, pyrite, and small grains of magnetite. There is also a small amount (perhaps 3% or 5% of the whole rock) of clear, colourless isotropic material enwrapping the other minerals. When the crushed rock is boiled in nitric acid, treatment with silver nitrate solution indicates the presence of small

quantities of chlorine, such as might well be accounted for by the apatite content of the rock, and in no way comparable with the abundant chlorine which was indicated in the sodalite-bearing tinguaite No. 1294. It is probable then that sodalite is absent from the present rock, and the isotropic mineral is identified as analcite. This mineral is not sufficiently abundant, however, to warrant classsification as a monchiquite. 1297, altered Camptonite. This is a fine grained, dark grey, somewhat mottled rock, which in section is seen to be a highly decomposed lamprophyre, the original constituents of which are amphibole 15%, pyroxene 20%, olivine 25%, feldspar 35%, magnetite 5%. The amphibole is brown barkevikitic hornblende, in subidiomorphic prisms about 0.5 × 0.1 mm., somewhat paler than the barkevikite of previous sections, but with the same general type of pleochroism. The pyroxene is pale pink augite, developed as small idiomorphic crystals with narrow borders of dusty magnetite. The large phenocrysts of olivine are now completely replaced by pseudomorphous masses of coarse carbonate, fine scales of talc and greenish fibrous serpentine. The feldspar occurs only in the groundmass. It is much altered to calcite and kaolin, and thus rendered indeterminate, though the presence of calcite suggests some variety of plagioclase. Small octahedrons of magnetite are plentiful in the groundmass. 1299, Camptonite. The section shows large phenocrysts of completely altered olivine, smaller ones of pyroxene, and occasional small crystals of hornblende and biotite, in an altered groundmass of pyroxene, feldspar, and iron ore. The large idiomorphic phenocrysts of olivine, constituting 10% of the section, are completely replaced by fibrous talc and small amounts of carbonate. The phenocrysts of pyroxene are, for the most part, idiomorphic crystals of bright green aegirine-augite, which are occasionally bordered with pale pink augite, itself edged with aegirine. There are also a few small phenocrysts of light yellow to almost black, intensely pleochroic biotite, and yellowish brown hornblende fringed with granular magnetite. Feldspar, which makes up about 30% of the section, is confined to the groundmass. It is greatly altered to sericite, kaolin, and calcite, but some crystals with refractive index higher than that of Canada balsam may definitely be determined as plagioclase. Altered pyroxene, some of which is aegirine-augite, also occurs abundantly in the groundmass, together with a minor quantity of biotite and plentiful grains of iron ore. A small percentage of colourless interstitial material, probably analcite, is present in some parts of the section. There are also numerous large prisms of clear apatite, some of which reach as much as 1.5 mm. in length. The only remaining constituent is a zeolite which occurs in clear aggregates, consisting of radially disposed or sometimes reticulated prisms about 0.5 mm. in length. It is colourless, or faintly clouded with alteration products, while the refractive index is lower than that of Canada balsam, but higher than that of analcite which sometimes occurs interstitially between the prisms of zeolite. The birefringence is about equal to that of the feldspar, the optic sign and elongation are negative, while the

extinction is inclined at about 5° to 8° to the direction of elongation. The elongation and oblique extinction indicate either scolecite or stilbite among the common zeolites. The extinction angle is, however, too low for scolecite (15° to 18°) and it is therefore identified as stilbite. 1298, Olivine-rich Camptonite. Megascopically this is a porphyritic rock, with very numerous light coloured phenocrysts which often reach 5 mm. in diameter, and constitute between 30% and 40% of the whole rock. In section, these are seen to be idiomorphic pseudomorphs after olivine, which is now completely replaced by finely granular carbonate, together with minor amounts of talc, serpentine, and secondary quartz. The groundmass is of much finer texture, contrasting sharply with the coarseness of the phenocrysts (Fig. 2), and consists, for the most part, of augite, hornblende, and feldspar, with minor iron ores. The augite occurs in pinkish or almost colourless idiomorphic crystals, which range from 0.1 mm. to 0.4 mm. in length, and usually show a narrow reaction border of reddish brown hornblende. Barkevikitic hornblende also occurs abundantly as small idiomorphic prisms. These are frequently terminated or fringed with a narrow rim of blue amphibole, which may also occur as scattered individual prisms and grains, and which is strongly pleochroic, with × = deep indigo blue, Y = lavender blue, Z = very pale lavender blue (absorption × > Y > Z). The elongation is negative, the extinction angle small (approximately 5°), and the dispersion strong (red > violet). This last factor, combined with the natural colour of the mineral, renders the exact determination of the birefringence difficult, but it is certainly low, and probably about 0.006. The mineral is evidently a sodic amphibole very close to riebeckite, from which it differs only in its slightly high birefringence (riebeckite has 0.004) and paler absorption tints. The remaining minerals are feldspar (probably plagioclase), which is much altered to calcite, and numerous ragged grains of iron ore, best referred to ilmenite. The proportions of the two main ferromagnesian constituents of the groundmass vary widely, even within the limits of a single section. Feldspar always makes up about 30% of the groundmass, but whereas in some parts brown hornblende is the dominant mineral, in others it is almost absent, while the proportion of augite is correspondingly increased. While this rock is essentially similar to those just described, it differs in the great abundance of large phenocrysts of olivine. S.W. 61 (Auckland University College collection), Olivine-rich Camptonite. This very closely resembles the previous rock (1298), and was collected from the same locality. At least 50% of the rock consists of large idiomorphic pseudomorphs after olivine, sometimes attaining 10 mm. in diameter. In one or two crystals, where a central core of unaltered olivine still persists, the lines along which alteration has proceeded may be observed. First of all fine strings of magnetite are thrown out along the cracks, which become bordered with finely granular carbonate, thus developing a pronounced mesh

structure. The olivine remaining between the meshes is now converted into talc, so that the mesh structure is preserved in the completed pseudomorph, while sometimes both talc and carbonate ultimately become more coarsely crystalline. Many of these altered olivines are crossed by narrow veinlets of transversely fibrous bluishgreen chlorite, which seems to be allied to bowlingite as defined by Winchell (1928, p. 382). Though the outline of the olivine pheno-crysts is usually strikingly idiomorphic, there is nevertheless evidence of considerable reaction between the phenocrysts and the liquid magma which is now represented by the groundmass. Frequently the liquid has forced its way along cracks, so that small angular fragments have been broken off the large crystals and strewed through the groundmass in their immediate neighbourhood. In yet other instances, the liquid has eaten its way along cracks and ultimately solidified as “inclusions” within the phenocryst, connected with the groundmass by narrow veinlets now largely filled with chloritic material. These “inclusions,” representing the solidified reaction product of olivine and liquid, always contain abundant prisms of golden-brown hornblende, which are enclosed in either the bluish-green chlorite mentioned above, or else in a colourless isotropic base which is probably analcite. A little feldspar is sometimes presen. In addition to the large olivines there are also a few much smaller phenocrysts of partly chloritised titaniferous augite, about 0.5 mm. to 1.5 mm. in length. The remainder of the rock is a fine groundmass, consisting of tiny prisms of golden-brown hornblende, a somewhat less amount of pale pink augite, small crystals of altered indeterminate plagioclase, and plentiful magnetite. Secondary flakes of kaolin, talc, and calcite are abundant. As in the previous section, the distribution of ferromagnesian minerals is irregular, some portions of the section consisting almost entirely of hornblende and feldspar. There is a single xenolith about 6 mm. × 3 mm., consisting of irregular grains of quartz, the whole being bordered by a narrow zone of serpentine, talc, and magnetite. 1300, Heterogeneous augite-camptonite. Under the microscope this rock presents a curious heterogeneous appearance since the mineral composition of the rock is not uniform, but is of two distinct types without intermediate gradation (Fig. 4). In the two sections examined, about 60% of the whole rock is a holocrystalline mass, of which the chief constituents are augite 60%, olivine 10%, hornblende 10%, feldspar 10%, and biotite and magnetite about 5% each. The idiomorphic phenocrysts of olivine (1 mm. to 3 mm. in diameter) are now entirely replaced by talc and a little granular carbonate, which in some cases outlines the original cracks, thus imparting a rough mesh structure to the pseudomorph. Pale violet augite occurs both in rather small clear idiomorphic phenocrysts about 1 mm. in length, and also as very numerous smaller prisms and grains which occasionally are bordered with hornblende. In one phenocryst, a central core of aegirine-augite was noted. The hornblende is pale yellowish brown to deep golden brown, in small idiomorphic prisms, frequently showing the orthopinacoidal twin, and having a maximum extinction angle of about 15°. The biotite is also in small flakes with

intense pleochroism from very pale yellow to deep reddish brown. It is somewhat rare in section 1300, but much more plentiful in a duplicate section (S.W. 62, in the Auckland University College collection). The feldspar is almost completely altered to a highly birefringent aggregate of kaolin, sericite, and calcite. Granular magnetite is abundant, and there are a few small prisms of apatite. The second type of mineral association is developed in numerous, often sharply defined areas, about 2 mm. to 5 mm. in diameter, enclosed by the crystalline mass just described. It consists mainly of light coloured, semi-opaque highly birefringent material which appears to be kaolin mixed with smaller amounts of calcite and sericite, and thus probably represents original feldspar, occasional fragments of which may still be observed. The ferromagnesian constituents are brown hornblende and biotite either singly, or associated in widely varying proportions. They form slender crystals which sometimes exhibit a tendency towards radial arrangement, and are often fringed with borders of dusty magnetite. Augite and primary magnetite are quite absent, though small fragments of altered olivine are present in one or two places. These leucocratic patches no doubt represent the solidified reaction product of a residual magmatic liquid, which has occupied the interspaces of a loosely aggregated mass of accumulated crystals rich in olivine, augite, and magnetite, and which has solidified there in reaction with the surrounding crystals. Flett (1911, p. 90; pl. 6, fig. 3) describes and figures “ocelli” of similar nature in the analcite-bearing camptonites of the Ross of Mull. These “ocelli” are described as rounded or oval spots, up to half an inch in diameter, consisting of radiate feldspars, long brown hornblendes, and less numerous crystals of augite, while, as in the section described above, there is also a small amount of interstitial analcite. The same writer (Flett, 1908, p. 125) refers to similar “ocellar” patches in the Tertiary camptonites of Oban. Mutual Relationship of Tinguaites and Camptonites. Various lines of reasoning indicate that the tinguaites and camptonites of the Haast and Makarora Valleys are intimately related rocks, representing different stages of the same differentiation series. In the first place, all are found in a restricted area, many miles from any other centre of igneous activity, the nearest igneous rocks being the granite-pegmatites which occur at the mouth of the Haast River, some twenty miles westward. In the second place there appears to be regular gradation between the extreme types. Thus, while large phenocrysts of altered olivine are a constant and conspicuous feature of the camptonites, so also the highly alkaline cancrinite-tinguaite No. 1924 occasionally contains large xenocrysts of olivine similarly altered. Again there seems to be a gradation between the tinguaites with abundant aegirine, through camptonites such as No. 1299, containing plentiful phenocrysts of aegirine-augite, to more typical camptonites in which the pyroxene is mostly augite, aegirine and aegirine-augite being present in very small proportions only. On the

other hand, barkevikite, which is such an abundant constituent of the camptonites, is also represented as occasional clusters and xenocrysts in the tinguaites. Finally an extensive series of similar rocks described by Smith (1908) from northern Westland is said to show a perfect sequence from tinguaites to lamprohyres (Smith, 1908, p. 126). It is thus evident that tinguaites and camptonites alike have originated from a common source, and any theory of origin must account for the presence of both types of rock. Course of Crystallisation of Parent Magma. Writing with reference to the origin of lamprophyres, Bowen (1928, p. 258) states that Niggli and Beger (1923, pp. 571–574) “conclude that these rocks are accounted for by local accumulation of early crystals, which have then remelted or redissolved, and given a liquid of lamprophyric composition.” In criticising the above suggestion, Bowen (1928, p. 258) then points out that the porphyritic and panidiomorphic structures so characteristic of the lamprophyres, are themselves strong evidence against the existence of lamprophyric liquid. The camptonites of the Haast and Makarora Valleys all contain plentiful large phenocrysts of olivine, which in some specimens are so numerous as to constitute at least half the total composition of the rock. Now Bowen (1928, pp. 159–164) has conclusively demonstrated that magmas never contain more than some 12% to 15% of normative olivine. It therefore follows that liquids having the total composition of these camptonites could never have existed, and the high olivine content must be due to crystal accumulation. As will be shown later, there is abundant additional evidence that crystal sinking and accumulation have played an important part in the evolution of the camptonite-tinguaite series. Bowen (1928, pp. 258–268) has outlined a theory according to which the olivine-bearing lamprophyres have originated as a result of reaction between an alkaline magmatic liquid and accumulated crystals of femic minerals, notably olivine and augite. In a subsequent chapter (1928, ch. 14), he suggests that this association of alkaline liquid and femic minerals is by no means fortuitous, but “is believed to be due to the fractional resorption of complex minerals, notably hornblende,” which he pictures as sinking into a relatively hot basaltic magma from higher levels. In a later section, it will be shown that the facts at present available suggest that the camptonites and tinguaites of Westland are considerably younger than the great series of peridotites, diorites, and granites which are so plentifully developed in that province. There are therefore not sufficient data to warrant speculation as to the ultimate source of the parent camptonite-tinguaite magma, nor as to whether fractional resorption of hornblende has been an important factor in its production. There does seem, however, to be

sufficient evidence to indicate that the camptonites and tinguaites have arisen as a result of differentiation of a highly sodic magma rich in ferro-magnesian constituents, from which abundant crystals of olivine had already separated. In the subsequent paragraphs an attempt will be made to outline this differentiation process, as deduced from the petrographic characters observed in the rocks in question. During the initial stages the main feature was continued separation of olivine crystals, the large size of which suggests that this process continued for a considerable time. As a result of continuous settling of the first-formed crystals, the lower portion of the magma became much enriched in olivine, while the upper levels were impoverished in this component and correspondingly enriched in the increasingly alkaline residual liquid. With falling temperature, a stage was at last reached when reaction between olivine and magma set in, with resultant precipitation of colourless or pale pink augite, reaction rims of which sometimes border the olivine phenocrysts. Precipitation of augite continued further until a temperature was reached when the augite and magma began to react to produce amphibole, which, as a result of the high soda-content of the reacting magma, took the form of the sodarich variety barkevikite. Olivine, augite, and barkevikite are all represented as phenocrysts in the rocks under consideration, but later products of the femic reaction series are represented only as reaction rims bordering the phenocrysts, or as constituents of the groundmass. These later members were therefore apparently formed as a result of reaction during rapid cooling following upon injection of the magma as dykes. The general reaction sequence appears to be olivine → augite → barkevikite → aegirine or biotite. A certain amount of variation is introduced by differences in composition of the liquid throughout the mass. Thus aegirine-augite may appear in certain of the more alkaline rocks (e.g., as phenocrysts in No. 1299), but whether or not it precedes the formation of barkevikite is uncertain. It certainly appears after normal augite, and before aegirine, but so also does barkevikite (e.g., in section S.W. 66a, a duplicate of the tinguaite No. 1294, one of the large xenocrysts of altered olivine shows perfectly the reaction series olivine → augite → barkevikite → aegirine). In another rock, section 1298, sodic amphibole very close to riebeckite was formed at a late stage by reaction between barkevikite and the liquid. The last ferromagnesian silicates to be formed are aegirine, biotite, or both. These are never abundant except in the tinguaites, where aegirine is the predominant femic constituent. No doubt the factor determining their crystallisation was relative abundance of soda and potash in the residual liquid when crystallisation was nearly complete. As a result of different rates of cooling and variation in total composition throughout the magma mass, it is obvious that at a given moment different stages in the above reaction sequence would be reached in different parts of the mass. Thus augite might still be

separating in the lower portion, at a time when at higher levels, where olivine has disappeared, aegirine-augite or barkevikite would be the stable solid components in equilibrium with the corresponding liquids. Consequently, to complicate matters further, sinking crystals of aegirine-augite and barkevikite would find their way into liquid where normal augite was still separating out. Reversal of the normal reaction sequence would then result as pictured by Bowen (1928, pp. 275–276). The results of such a process would be for the most part confined to the larger phenocrysts, while the crystals of the groundmass would be expected to indicate the normal reaction sequence. Section No. 1296 is a good example of a rock in which the normal and reversed reaction sequences have both been developed. The normal reactions observed are olivine → augite, olivine → barkevikite, and olivine → biotite, shown by the large phenocrysts of olivine; augite → aegirine, shown by phenocrysts and small crystals of augite; augite → barkevikite → aegirine shown by small crystals in the groundmass. Reversed reactions, plainly shown by large, partially resorbed crystals of barkevikite and others of aegirineaugite, are barkevikite → augite and aegirine-augite → augite. These large partly resorbed barkevikites also show the results of the normal reaction augite → aegirine, which expresses itself in an outermost fringe of aegirine, formed at a late stage in crystallisation, simultaneously with the separation of aegirine in the groundmass. The reversed reactions described above are also illustrated by a few of the phenocrysts in Section No. 1295. In Section No. 1299, the large phenocrysts of aegirine-augite are bordered with a narrow rim of colourless augite, representing reversal of the normal reaction, and this in turn is surrounded by an irregular outer rim of aegirine indicating the normal reaction augite → aegirine at a late stage in crystallisation. Injection of the partially crystallised magma as dykes took place when the parent mass was in the heterogeneous condition described above. Dykes drawn from the higher levels would be of tinguaitic composition, while those emanating from the lower portions of the mass would be olivine-rich camptonites. In the tinguaites, aegirine, though abundant in the groundmass, is not developed as phenocrysts; on the other hand the presence of large phenocrysts of albite-perthite indicates that in the upper levels from which the tinguaites were drawn, the magma had already attained, prior to intrusion, a composition such that alkali-feldspar had commenced to crystallise. In the camptonites, on the other hand, feldspar is confined to the groundmass. Camptonites such as Nos. 1296 and 1299, in which partially resorbed phenocrysts of barkevikite and aegirine-augite from the higher levels still persist, in addition to the usual abundant olivine and augite, represent magma drawn from intermediate levels. Those camptonites in which olivine is extremely abundant (e.g., No. 1298), and the augite-rich variety, No. 1300, doubtless represent the lower portion of the magma where

Fig. 1—camptonite (1296) showing pheno crysts of barkevikite set in a groundmass of augite, barkevikite, and altered feldspar. The large dark crystal is a resorbed barkevikite immedimmediate with clean titan-augite Magnification, 40 diams. Fig. 2—Olivine-rich Camptonite (1298) Large altered crystals of olivine are enclosed in an extremely fine crystalline groundmass (black). Magnification, 15 diams. Fig. 3.—Tinguaite (1294) showing dark prismatic crystals of aegirine enclosed by light-coloured Camptonite and feldspar. Magnification, 40 diams Fig. 4—Augite-rich Camptonite (1300) An “ocellan” patch of altered feldspan and analcite with seattered crystals of barkevikite and biotite (lower left), is partly surrounded by a granular aggregate consisting largely of augite and magnetite which constitutes the normal groundmass of the lock. Magnification, 40 diams.

accumulation of early-formed crystals has been effective. It is of interest to note that even here, the interstitial liquid was sufficiently alkaline for biotite, riebeckite, and analcite to be formed among the final products of crystallisation. An interesting feature of some of the more basic olivine-rich and augite-rich camptonites is the heterogeneous nature of the groundmass already referred to in a previous section. This is most striking in No. 1300, where “ocellar” patches are developed (Fig. 4), comparable with similar structures observed by Flett (1908, p. 125; 1911, p. 90) in some of the Tertiary camptonites of Scotland. The injected material, in the case of No. 1300, must have consisted of crystalline augite and olivine, with about an equal quantity of liquid scattered as globules throughout the interstices of the mass. Rapid cooling following on injection, brought about complete solidification of the liquid, before complete reaction with the olivine and augite had been affected. The liquid is now represented in the resultant rock by conspicuous “ocellar” aggregates of feldspar, radially disposed hornblende and biotite, and interstitial analcite, surrounded by a mass consisting almost entirely of augite, olivine, and magnetite, the products of early crystallisation. Bowen (1928, pp. 270, 271) stresses the importance of filtration and squeezing out of liquid from crystals during injection, as a means of producing a nepheline-rich liquid from such a magma as we have been considering. In the case of the Haast-Makarora rocks, it is probable that this process was of minor importance only. It has already been shown that crystal-sinking prior to injection of the dykes had a profound influence on the differentiation of the series. On the other hand, the presence in the tinguaites of large phenocrysts of feldspar, obviously the result of crystallisation before injection, and the presence of “ocellar” patches in the more basic camptonites indicating that abundant liquid was still interspersed throughout a spongelike mass of crystals after injection, are both facts which suggest that the squeezing-out process was not an important factor in determining the present constitution of the rocks under consideration. Correlation and Age. Many varieties of lamprophyres and related basic hypabyssal rocks have been recorded from localities in West Nelson and North Westland, ranging from Reefton in the north to Hokitika district in the south (Bell and Fraser, 1906, pp. 82, 83; Smith, 1908; Morgan, 1908, pp. 138, 139; Morgan, 1911, pp. 80, 81; Bartrum, 1914, pp. 267, 268; Henderson, 1917, pp. 107, 108). The prevailing types are hornblende- and augite-camptonites, which grade into porphyrites and diabases, or with the development of analcite pass into monchiquites. Other varieties are also present, especially in the Reefton district, where Henderson (1917) has described minettes, kersantites, vogesites, spessartites, camptonites, and diabases.

Probably the most interesting series of dyke rocks from this North Westland region is that described by Smith (1908) from stream boulders collected from New River in the vicinity of Greymouth. According to Smith, there is complete gradation between tinguaites with abundant cancrinite, nephelite, and aegirine, and camptonites and vogesites in which, though feldspathoids are absent, barkevikite and aegirine indicate alkaline affinities. From the same area Smith also describes gabbros, diabases, and theralites, while undescribed specimens subsequently presented by the same writer to the Geological Museum of the Otago University include a coarsely crystalline ditroite in which blue sodalite is clearly visible in the hand specimen. The Haast-Makarora area lies about 150 miles south of the district studied by Smith. Nevertheless there is a striking similarity between the tinguaites and normal and aegirine-bearing camptonites of these two widely separated localities. Park (1909) noted hornblende-camptonites and monchiquites from the Shotover and Kawarau Valleys in the Lake Wakatipu district, some 50 miles south-west of the present area. While strongly alkaline types are not included among these rocks, they are nevertheless closely similar to the normal camptonites of Westland. An augite-hornblende-monchiquite from the Shotover in the Otago University collection (La. 13) resembles Section No. 1300 described above, in that there is a tendency towards heterogeneous composition, resulting from concentration of augite or hornblende in different parts of the section. Some difference of opinion at present exists as to the exact age of the basic dyke rocks of Nelson and Westland. Bell and Fraser (1906, p. 82) give the age of rocks of this type in the Hokitika district as probably early or mid-Tertiary. Morgan and Bartrum (1915, p. 104) state that dykes of camptonite and monchiquite not only invade the ancient greywackes, schists, and intrusive granites, but even penetrate the lower beds of the unconformably overlying Eocene coal-measures. They suggest an early Eocene date of intrusion. Henderson (1917), while recognising that the camptonites and some of the dolerites of the Reefton district are early Tertiary, nevertheless states that the diorites and vogesites of that area followed immediately upon the granite intrusions of the (?) Early Cretaceous, while some of the dolerites even antedate the period of granite intrusion. The evidence available indicates that many of the basic dyke-rocks of the West Coast region, including camptonites, monchiquites, and tinguaites, were injected probably in Early Tertiary times. The rocks of this group are thus considerably younger than the great intrusions of peridotite, diorite, and granite, which, according to most authorities (e.g., Henderson, 1917, p. 104) accompanied the profound mountain-building movements of the Early Cretaceous. It is therefore probable that the petrographically similar tinguaites and camptonites of the Haast-Makarora area, which have not been appreciably affected by stress, post-date the Lower Cretaceous orogeny.

Acknowledgments and Thanks. The rocks which form the subject of this paper were collected during February, 1930, in the course of an expedition into the extreme south of Westland by way of the Makarora and Haast Valleys and Haast Pass. A large part of the expense incurred was defrayed by a grant from the Otago University Council, to which body I am much indebted. My thanks are also extended to my companions, Professor J. A. Bartrum, of Auckland, and Messrs G. Simpson, and J. S. Thomson, of Dunedin. I am also much obliged to Professor Bartrum for the loan of duplicate sections cut by him from rocks collected in this locality; to the Cawthron Institute for loan of sections for comparison; and to Dr B. Dodds, of the Otago University Dental School, for use of photomicrographic apparatus. Literature Cited. Bartrum, J. A., 1914. Some Intrusive Igneous Rocks from the Westport District. Trans. N.Z. Inst., vol. 46, pp. 262–269. Bell, J. M. and Fraser, C., 1906. The Geology of the Hokitika Sheet, North Westland Quadrangle. N.Z. Geol. Surv. Bull., No. 1 (n. s.). Bowen, N. L., 1928. The Evolution of Igneous Rocks. Princeton University Press. Flett, J. S., 1908. In The Geology of Oban and Dalmally. Mem. Geol. Surv. Scotland, Expl. Sheet 45. —– 1911. In The Geology of Colonsay and Oronsay with Part of the Ross of Mull. Mem. Geol. Surv. Scotland, Expl. Sheets 35, 27. Henderson, J., 1917. The Geology of the Reefton Subdivision. N.Z. Geol. Surv. Bull., No. 18 (n. s.). Morgan, P. G., 1908. The Geology of the Mikonui Subdivision. N.Z. Geol. Surv. Bull., No. 6 (n. s.). —– 1911. The Geology of the Greymouth Subdivision. N.Z. Geol. Surv. Bull., No. 13 (n. s.). Morgan, P. G., and Bartrum, J. A., 1915. The Geology of the Buller-Mokihinui Subdivision. N.Z. Geol. Surv. Bull., No. 17 (n. s.). Niggli, P. and Beger, 1923. Gesteins und Mineralprovinzen, I. Park, J., 1909. The Geology of the Queenstown Subdivision. N.Z. Geol. Surv. Bull., No. 7 (n. s.). Smith, J. P., 1908. Some Alkaline and Nepheline Rocks from Westland. Trans. N.Z. Inst., vol. 40, pp. 122–137. Winchell, A. N., 1927. Elements of Optical Mineralogy, Part II. New York, John Wiley and Sons.