Some Porphyroblastic Albite-Schists From Waikouaiti River (South Branch), Otago.
[Read before the Wellington Philosophical Society, April 22, 1941; received by the Editor, April 23, 1941; issued separately, December, 1941.]
This paper embodies the results of mineralogical and petrofabric studies of a series of schists collected from the entrenched meanders of Waikouaiti River (South Branch) in the Waikouaiti Survey District, 16 miles north-east of Dunedin (text-fig. 1). These schists are a local development of the Otago Schists (Maniototo or Wanaka “Series” of current literature) and lie within the Chl. 4 subzone of the Chlorite Zone as defined by Hutton and Turner (1936) for quartzo-feldspathic rocks. Large porphyroblasts of albite charged with dusty black iron-ore stand out conspicuously as black knots on weathered surfaces, and impart a decidedly spotted appearance which is otherwise unusual in the schists of eastern Otago. Interbedded locally with the quartzo-feldspathic rocks are minor bands of green schists, probably derived from basic or semibasic tuffs. The albite schists on the other hand are believed to be derivatives of grey-wackes.
The predominating rock-type is quartz-albite-muscovite-schist in which the three minerals mentioned together make up 90% to 95% of the composition. Locally quartz occurs almost to exclusion of albite, while again with the entry of plentiful chlorite, a group of quartz-albite-muscovite-chlorite-schists may be distinguished. With decrease in quartz and mica and increase in chlorite this latter group grades into true green schists.
I. Quartz-albite-muscovite-schists and quartz-albite-muscovite-chlorite-schists.
Eighteen of the twenty-two representative specimens examined belong to the above categories. Prevalence of rocks poor in chlorite and epidote, but relatively rich in muscovite, is noteworthy in that it is otherwise unusual among the schists of eastern Otago. Details of structure and texture are given later under the heading “Structural Petrology,” and the present section is devoted mainly to details of mineralogy.
Apart from structural features there is little to note in connection with the two main constituents, quartz and albite. The grains of quartz are water-clear, irregular in outline (sometimes with interlocking margins, as in No. 4718), and often show some degree of undulose extinction. When the latter is strong the quartz grains are crossed by strings of minute colourless granules 0·005 mm. in diameter, which apparently develop at a stage just prior to cataclastic fracture (cf. Hutton, 1940, p. 61; see also Eskola, 1939, p. 304). Some of the quartz grains are also crossed by tensional
cracks at right angles to the direction of elongation. Albite is usually equally abundant with quartz and occurs mainly in the form of large porphyroblasts usually 2 mm. to 3 mm., rarely as much as 10 mm. in diameter. Careful measurements on a universal stage show that the anorthite content does not exceed 4% in any instance. Simple Carlsbad twinning was noted in a few cases, while
local development of lamellar twinning mainly on the pericline and albite laws is not uncommon. As described in a later section, nearly all the large albites enclose irregular blebs of quartz, and parallel strings of dusty iron-ore together with less plentiful epidote, sphene, and apatite, marking the trend of early s-surfaces (see text-figs. 2B, 3B, 4A, etc.).
Muscovite is relatively coarse in comparison with the mica of other Otago Schists. It is a typical colourless mica in all cases except No. 6247, in which a faint green tint was noted along the γ vibration-direction. The β refractive index, determined in six cases, varied from 1·5930 in No. 6236 to 1·5972 in No. 6252A, and according to Volk's curves (1939, p. 263), indicates a range of composition lying in the region between potassium muscovite [H4K2Al6Si6O24] and phengite [H6K2(FeMg)2Al4Si6O24].
Minerals of the chlorite group are not usually abundant though in some examples they rank as important constituents (viz. Nos. 6237, 6243, 6246, 6247, 6249, 6252A, 4696). Except in a few cases the chlorite mineral is not the usual clear green, coarse, prochlorite [as defined by Orcel (1927, p. 355)] which is found without exception in the schists of western Otago (Hutton, 1940, pp. 17-19), but instead occurs in aggregates of tiny flakes that are colourless or very pale green. This chlorite is commonly interleaved with muscovite or more rarely intimately associated with bars of oxidized or leucoxenized iron-ore. In one case green prochlorite was observed as inclusions in albite porphyroblasts, while the colourless chlorite occurred in the groundmass. The properties of the colourless chlorite are as follows: elongation positive; birefringence low (< 0·005); anomalous interference tints absent; refractive index β = 1·559 (No. 6245) to 1·566 (No. 6247). The value of β corresponds to either antigorite or negative penninite on Winchell's curves (1936, p. 649), but the general petrographic association in the present rocks and the low birefringence favour the latter alternative. It is therefore classed as negative penninite, rich in MgO and low in FeO.
The β refractive index has also been measured for two typical green chlorites separated from quartzo-feldspathic schists: No. 6243, β = 1·6325, weakly positive, ρ<v; No. 4696, β = 1·6315, weakly negative, ρ>v. These data agree well with those typical of the prochlorite group (Wiseman, 1934, pp. 361-365; Hutton, 1940, pp. 17-19).
Tourmaline is generally distributed throughout the schists of the South Waikouaiti area, and, even if inconspicuous or absent in thin sections, is nearly always present in heavy residues separated from crushed specimens. It takes the form of sharply idioblastic prisms which rarely attain a length of 1·0 mm. (No. 6252A). The colour varies from deep blue in some grains to greenish-brown in others; strong zoning was noted in once instance (No. 6252A) as follows:—
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
α = very pale brown
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
γ = deep green
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
γ > α
α = very pale brown
γ = deep greenish-brown
γ > α
Determination of refractive indices (three specimens) and specific gravity (one specimen) gave the following:—
|6252A (Inner Zone)||1.6334||1.6610||0.0276||3.16 ± 0.03|
In the zoned crystal (6252A) the outer zone gave slightly higher values for α and γ. Comparison of these data with the curves given by Winchell (1933, p. 303) indicates, compositions midway between dravite and schorlite, while according to the curves constructed by Quensel and Gabrielson (1939, pp. 80, 81) the content of Al2O3 ranges from 31·8% to 33·3%, and that of FeO from 10·3% to 13%. Tourmaline crystals in the schists described, as is usually the case in schists of eastern Otago, are heavily charged with dusty iron-oxides, while one extremity of the crystal is usually of a different colour from the remainder (cf. Woodland, 1938, p. 446; Hutton, 1940, p. 36). An identical condition holds for tourmaline in schists of western Otago (Turner, 1933, 1935; Hutton, 1939, 1940, p. 36).
Calcite is a common minor constituent of a number of the quartzo-feldspathic schists, particularly so in No. 6248. It may occur as aggregates of xenoblastic grains or in strings cutting across the rock. The refractive index was determined in two cases and found to be 1·661 (for γ) (Nos. 6243, 6238A); therefore the calcite appears to contain not more than 2–3% of the siderite molecule (Winchell, 1933, p. 70). Lamellar twinning is not uncommon.
In four schists, greenish-brown biotite was recognised, but only in one example (No. 6238A, β = 1·6342) is it an important constituent. Intimate interlamination with chlorite (penninite and prochlorite) and muscovite is the usual mode of occurrence. In some cases aggregates of fine penninite and iron-ore granules adjacent to shreds of biotite suggest the possibility that the penninite is a product of retrogressive metamorphism of biotite and that the latter mineral was formerly much more plentiful than is now the case. The greenish-brown mica is closely comparable in its properties with micas described in schists of the chlorite zone in Western Otago, (Turner, 1933; Hutton, 1940, p. 52) and in other parts of the world (Tilley, 1925, p. 103; Phillips, 1930, p. 244-245).
In eight specimens, subidioblastic to idioblastic grains (occasionally porphyroblastic, No. 6235) of colourless to pale pink garnet were observed. The garnets are usually crowded with dusty black or dark brown inclusions and in some instances these are zonally arranged. Woodland (1939, p. 17) reports similar inclusions which he identifies as haematite, in manganese garnets in manganese ore from Merionethshire. On the other hand, Fermor (1909, p. 177) considered the dark nucleus in the spessartite of the Gondite Series
to be coloured with excess of ferric and manganous oxides over that required for the formation of the garnet. Frequently the garnets show irregular cracks, along which colourless negative chlorite, possibly penninite, is developing. Refractive index determinations for garnets from these specimens gave the following results: No. 6236, N = 1·797 *; No. 6238, N = 1·795; No. 6245, N = 1·790. The specific gravity of the last garnet is 4·01 ±0·03. The optical data and specific gravity are somewhat lower than those recorded by Woodland (1939, pp. 20–21) for a garnet with 88·45% of the spessartite molecule. Therefore the garnets from the schists of the Waikouaiti River area must, according to the data of Winchell (1933, p. 176) and of Woodland, contain somewhat less than 75% of spessartite, together with considerable CaO. According to Winchell's diagram (1933, p. 175), the amount of the grossularite molecule allowable in spessartite must not exceed about 25%.
Apatite typically occurs in moderately coarse xenoblastic to subidioblastic grains, which average 0·5 mm. in diameter. Inclusions of dusty iron-ore are not uncommon and in No. 6236 very thin pale brown non-pleochroic plates of an unknown poorly birefringent mineral with refractive index less than that of apatite were observed lying parallel to the vertical crystal axes of the enclosing apatite. Comparison with inclusions in apatite recorded by Groves and Mourant (1929) is not possible. Optical data for the apatite itself in No. 6236 are: α = 1·6340, γ = 1·6365; γ−α = 0·0025; uniaxial, negative. Determinations on apatite from No. 6238A gave α = 1·632, γ = 1·636. Microchemical tests indicated abundant CaO and P2O5 in both cases. These data are recorded since the writers have previously somewhat doubted the identity of the coarse xenoblastic apatite universally present in the schists of Otago.
Pyrite, often partly replaced by limonite, is a very common minor constituent of the schists, and usually occurs as strings or ragged bands of crystals cutting across the earlier s-planes (text-fig. 2A) or albite porphyroblasts (text-fig. 3A).
Minor accessories noted in these rocks are colourless to pale green actinolite (Nos. 6239, 6245), clinozoisitic epidote, spindle-shaped grains of sphene, magnetite (usually in a dusty form) and zircon. The latter mineral is the most common constituent of heavy residues prepared from these schists: it is usually sharply idioblastic, although rounded grains were also observed; inclusions are often numerous. In one heavy residue (No. 6238A) rare grains of pale purple, pleochroic zircon were noted.
Oxidation of the pyrite has caused considerable staining and alteration of the rock as a whole and of the chlorite and calcite in particular. The apparent alteration of the carbonate to limonite, so common in these schists, at first suggests that the mineral in question is siderite. However, refractive index determinations have shown that the carbonate is calcite, and it is therefore believed that this mineral has reacted with the FeSO4 and H2SO4 derived from the oxidizing pyrite, with resultant local precipitation of limonite.
[Footnote] * All ± 0·005.
II. Green Schists.
Only four rocks belong to this group, and they are characterised in the hand specimen by bright green colour due to abundance of chlorite. They are strongly schistose, foliated, and coarsely crystalline. The schists are not to be considered as a distinct group, unrelated to the previous one, but rather they appear to grade into the quartzo-feldspathic schists with increase of quartz and muscovite.
Albite typically occurs in subidioblastic porphyroblasts, with occasional crystals up to 5·0 mm. in diameter. Simple twinning is uncommon, but patchy lamellar twinning was noted more often. Inclusions of poorly ferriferous epidote, sphene, iron-ore and, in some cases, actinolite (Nos. 6253, 6240) are common; these inclusions are often aggregated into dense clusters of grains and less often in helicitically arranged strings (text-fig. 5A). Blebs of quartz are very common. The composition of the feldspar is, without exception, nearly pure albite.
An epidote mineral is a common constituent. It is typically poorly ferriferous containing not more than 15% of HCa2Fe3Si3O13 molecule and in most cases much less. Zonary structure, with the more ferriferous portions centrally located is typical, while clusters of clinozoisite are frequently enveloped in clouds of magnetite dust formed by the ferriferous epidote → clinozoisite reaction.
Refractive index determinations again show that the typical green chlorites occurring abundantly in the green schists are members of the prochlorite group (No. 6249, β = 1·6320, weakly negative; No. 6244, β = 1·631, weakly negative; No. 6240, β = 1·6322, weakly negative). In all four schists a minor amount of muscovite
occurs (β = 1·5963, 1·5955) intimately interleaved with the prochlorite.
Very pale green, faintly pleochroic actinolite occurs as swarms of acicular crystals in the prochlorite in No. 6253, but otherwise it is a rare accessory mineral. Occasionally subidioblastic grains of colourless garnet were observed in only one thin slice (text-fig. 5A). Other minor constituents noted in these rocks are calcite, sphene, partly oxidized pyrite, magnetite, apatite, xenoblastic grains of quartz and zircon. In No. 6240 a mineral occurs in clear, colourless grains, with two good cleaveages at 80° to one another; it is optically positive, with 2V about 40°. This is identified as barite, for the properties are very similar to those recorded for barite in a quartz-muscovite-piedmontite schist by one of the writers (Hutton, 1940, p. 43).
The Megascopic Fabric.
The most conspicuous schistosity in the field (S1) is sub-horizontal or dips at angles of 0° to 15° (rarely as much as 25°) across a variable strike. Where the dip locally increases to 45° or more, the strike approaches a north-south trend (170° to 200°), and from petrofabric evidence the tilted condition can be referred to late movements in no way connected with the metamorphism (cf. Turner, 1940, p. 179). Suitably oriented sections and polished specimens usually show a second set of s-surfaces which may be more conspicuous than S1; they intersect the latter at angles of 20°–35° and may dip either eastward (S2) or to the west (S3). As is frequently the case when two acutely intersecting sets of s-surfaces appear to be present, it is sometimes difficult to determine whether this condition actually does exist or whether it is merely simulated by regular microscopic undulation of a single set of s-surfaces. In most of the specimens here considered the separate identity of the two sets may be established by careful microscopic observation, while in several rocks (e.g. No. 6241a) confirmation is afforded by restriction of abundant introduced pyrite and calcite to one series of s-surfaces (S2), the other (S1) being unaffected (text-fig. 2A). This latter would appear to establish a late date for development of S2, since pyritisation and introduction of calcite into many schists from this area took place either late in their meta-morphic history or even after metamorphism had ceased altogether.
Though some schists have a tendency to split parallel to either S2 or S3, the plane of readiest fissility is usually S1, which is therefore selected as the ab plane of the fabric. Within this plane, as in all schists from east and central Otago, there is a lineation (b) the trend of which varies from 170° to 190°, and thus accords with trends previously recorded from this province. In thin sections or on polished surfaces this lineation is seen to be due to a combination of three factors, viz. undulation of mica-rich laminae about b, intersection of S1, S2 and S3 in b, and development of discontinuous pencils of quartz arranged roughly parallel to b. These latter are usually small, about 10 mm. to 20 mm. long by 1 mm. in diameter, but in one locality (No. 6241) take the form of large regular rods 20 cm. to 30 cm. long and lenticular in section (1 cm. × 3 cm.).
In the area under discussion metamorphic differentiation has proceeded along lines somewhat different from those followed elsewhere in the Maniototo schists of eastern and central Otago, and different structural features have therefore evolved. The usually conspicuous lamination resulting from segregation of alternately quartzo-feldspathic and micaceous layers is here almost absent, and in consequence the contorted appearance resulting from subsequent folding and rupture of laminae in many Otago schists is also lacking. In the schists from South Waikouaiti River, the chief effect of metamorphic differentiation has been growth of conspicuous porphyroblasts of albite, 2 mm. to 5 mm. in diameter, which often make up 50% of the total composition of the rock and dominate its structure. They are rounded or irregular in outline and usually contain plentiful small rounded grains of quartz, or subparallel strings of epidote crystals and magnetite dust that allow recognition of internal s-surfaces (si) belonging to a period of deformation preceding growth of the porphyroblasts (text-fig. 2B). Metamorphic differentiation is also responsible for segregation of thin.
Text-fig. 3A—Calcite-limonite (black) strings cutting albite porphyroblasts and continuing across section (No. 6242, ab). × 27.
Text-fig 3B—Crumpling of old s-surfaces within porphyroblast of albite (No. 6242). × 27.
micaceous bands that define the visible s-surfaces S1, S2 and S3, and for development of the discontinuous quartzose pencils already referred to. Growth of albite porphyroblasts, micaceous bands and quartzose pencils appears to have been simultaneous, for none cuts across or otherwise interferes with either of the others. On the other hand, calcite, and to a less extent, pyrite, have often crystallised later along cracks cutting across porphyroblasts and matrix alike, or along certain of the s-surfaces (text-figs. 2A and 3A).
The Muscovite Fabric.
Poles of the (001) cleavage were measured for flakes of muscovite in thin sections cut parallel to the ac and ab planes of the fabric respectively. In sections parallel to ab this could be done only for crystals in which the cleavage made an angle of more than 50° with the plane of the section. For more flatly oriented crystals, therefore, it was necessary to determine the position of the acute bisectrix α [= the pole of (001)] by placing β parallel to the EW rotation axis of the stage and bisecting the optic axial angle so obtained.
Billings and Sharp (1937, pp. 283–289) have drawn attention to the influence of crystal habit upon the pattern of fabric diagrams. In the case of minerals of tabular habit, such as mica, crystals flattened parallel to the plane of section have less chance of being included in a rock section than crystals that are steeply inclined to the section. Actually the influence of crystal habit is relatively unimportant in mica diagrams based on measurements in rock sections cut parallel to the ac plane of the fabric, provided either the b axis or the ab plane of the mica fabric is strongly defined, since in such cases all the crystals are steeply inclined to the plane of section. However, important effects may be shown if both b and ab are weakly defined, as in many of the rocks discussed in this paper, or if the measured section is cut parallel to b. In interpreting the mica fabric of the schists from South Waikouaiti River the writers have therefore considered the ac and ab diagrams in conjunction with each other, and have reached the following conclusions:—
(1) The predominant element in the muscovite fabric is a completely developed girdle the b (= B) axis of which approximately coincides with the megascopic lineation. In ac diagrams such as pl. 37, figs. 1–3 * the girdle is to a slight extent artificially sharpened as a result of the influence of the tabular habit of the measured crystals. For example, the girdle of pl. 37, fig. 5, based on measurements in an ab section, is somewhat broader than the girdles of ac diagrams; in fig. 2 the angular width of the girdle at the point of emergence of a is 40°, while in a diagram based upon measurements in the ab section of the same rock the width at a is 60°. Comparison of ab diagrams shows that the b (= B) axis of the mica girdle is less sharply defined in the porphyroblastic albite-schists discussed in this paper than in the typical non-porphyroblastic schists of eastern Otago previously described (Turner, 1940).
(2) Figs. 3 and 4 (pl. 37) depict the fabric of thin rather coarse undeformed tables of muscovite (0·3 mm. in diameter), enclosed in regular rods of quartz 1 cm. to 3 cm. in diameter that lie parallel to b in specimen No. 6241. It is obvious from comparison of fig. 4, with fig. 5 that the b axis of this mica-in-quartz fabric is much more sharply defined than that of the mica fabric in rocks (such as No. 6240) where the mineral is concentrated in thin layers that undulate around porphyroblasts of albite and small pencils of
[Footnote] * The figures in this paper are representative examples selected from fabric diagrams prepared from sections of fifteen oriented hand-specimens.
quartz. Pl. 37, fig. 4, also serves to illustrate that the position of b may be more precisely determined in ab than in ac diagrams. In this particular case (No. 6241) the trend of b oscillates slightly about c, and in pl. 37, fig. 4, the mean direction of b (b′) of the mica fabric within the measured section may be compared with the mean direction of the lineation (megascopic b) as determined in the hand-specimen.
(3) In pl. 37, figs. 1 to 3, the maxima at c correspond with the main schistosity S1 and are accompanied by maxima at K or N, corresponding to S2 or S3 respectively, if either of these is present. On the other hand, maxima, in ab diagrams such as pl. 37, figs. 4 and 5, are almost unrelated to the fabric and merely reflect the tabular habit of the mineral in question; nevertheless the marked tendency for concentration of mica poles in the ac quadrant of pl. 37, fig. 3, is also obvious in the corresponding quadrant (ac) of pl. 37, fig. 4.
(4) To ascertain the relative importance of the internal structure of the mica lattice and the external form of the mica crystal in bringing about the preferred orientation of crystals with (001) parallel to the principal s-surfaces, the position of the β vibration-axis was determined in flakes of mica inclined to ab at angles of less than 30°–40° *. This was done for all available crystals (usually about 40) in ab sections, and the degree of preferred orientation was demonstrated by plotting and counting the number of poles of β falling within 1% of the total area of projection, and drawing contours at densities of 4, 3 and 1 poles per 1% area (pls. 37 and 38, figs. 6–9). If the preferred orientation in S1, etc., were determined by external form alone, then the poles should be evenly distributed around the periphery of the projection upon ab; and if we assume that the poles in question are 36 in number and lie in a peripheral zone, of width equal to one-fifth the radius of the projection, then the density of poles should be 1 per 1% total area of projection. Actually, concentrations of four or five times this density are developed in pls. 37 and 38, figs. 6–9, and it is therefore concluded that lattice orientation plays an important part in the mica fabric of the rocks in question. The diagrams show that there is a strong tendency for β to lie either perpendicular to b (maximum X) or parallel to b (maximum Y), or both (figs. 6, 8). Sander (1930, pp. 225, 326, D137) has described instances of orientation of muscovite with the a crystal axis (= the β vibration-axis) as glide-line parallel to the a fabric axis and hence perpendicular to b of the fabric. Maximum X in the present diagrams accords with this rule of orientation, and the presence of the second maximum Y might well indicate, on the same assumption, that there has been some gliding of mica crystals parallel to b of the megascopic fabric. It should be borne in mind, however, that the available crystallographic data for muscovite have been obtained largely from coarse crystals in granites and pegmatites; and it is therefore just possible that muscovites of the low-grade
[Footnote] * In each case the position of β was determined carefully by standard universal stage procedure. It is not sufficient merely to test the relative velocities of the two vibration-directions unless (001) is inclined to the plane of the section at angles smaller than 15°–20°.
crystalline schists include two optically distinct varieties, one with β parallel to the a crystal axis and the other with β transverse to a as in micas of the biotite series. Whatever may be the correct explanation, the presence of maxima X and Y at the points of emergence of the a and b fabric axes appears to be characteristic of this type of fabric diagram for all schists from eastern Otago so far investigated (e.g. pl. 38, fig. 8; specimen No. 4701, 3ml. E. of Middle-march, Central Otago).
The Albite Fabric.
Measurement of α, β and γ in 110 porphyroblasts of albite in No 6238 failed to reveal any obvious preferred orientation for grains of this mineral. The same holds true for albite of non-porphyroblastic schists from elsewhere in eastern Otago, in so far as these have till now been investigated (e.g. No. 4708, south of Brighton; No. 4723, between Outram and Hindon). Actually the apparent lack of preferred orientation may possibly be due to operation of several different laws of orientation controlled by several sets of s-planes within the rock.
In many sections of the South Waikouaiti schists, porphyroblasts of albite enclose strings of epidote and dusty iron-ore that mark the trend of older s-surfaces (si) that existed prior to development of the porphyroblasts. Complex microfolding and rupture of these s-surfaces within the limits of a single porphyroblast (text-fig. 3B) have been noted in several instances, showing that strong deformation had already been accomplished before the porphyroblasts began to develop. The si in such cases must be a truly relict structure and its trend cannot be related to the space-lattice of the albite as in the examples recently described by Ingerson (1938). Growth of the large albites seems to have been in the main a static process, for in no case were observed the S-shaped trend-lines that characterise the si fabric in paratectonic porphyroblasts of albite from schists of
Text-fig. 4A—Old s-surfaces continuous and undeformed between adjacent porphyroblasts of albite (No. 6252a parallel to ac). × 53.
Text-fig. 4B—Post-crystalline movement of porphyroblasts (No. 6252a parallel to ac). × 53.
Text-fig. 4C—S-shaped inclusions in paratectonic porphyroblast, Kawarau, Gorge, Western Otago (No. 2655, parallel to schistosity). × 53.
western Otago (text-fig. 4C) or other areas of regionally metamor-phosed rocks (e.g. Bailey, 1923; Turner, 1933, p. 230). This static-process of growth is sometimes further illustrated by uninterrupted passage of si, without alteration of trend, through several adjacent porphyroblasts (text-fig. 4A). Careful examination of polished specimens and sections cut perpendicularly to b shows, however, that most rocks have been affected by a later deformation, for at the boundaries of adjacent porphyroblasts the trend of si may change abruptly through angles of as much as 90° (text-fig. 4B) as a result of external rotation of the porphyroblasts subsequently to their crystallisation.
Taking all these facts into account, we may therefore recognise three phases in the development of the albite fabric:—
(1) Strong deformation, with evolution of the s-surfaces now represented by si enclosed in albite.
(2) Growth of porphyroblasts under almost static conditions.
(3) Late deformation involving rotation of porphyroblasts and enclosed si.
The Quartz Fabric.
(a) Typical Quartz-in-quartz Fabric. Quartz for the most part is concentrated in narrow pencils trending roughly parallel to b. The majority of the component grains range from 0·1 mm. to 0·4 mm. in diameter, and in some cases slightly undulose extinction may be observed between crossed nicols; smaller grains (0·005 mm.) may also be present. The fabric developed in quartz of this type will be described as the quartz-in-quartz fabric, as distinct from the quartz-in-albite fabric which concerns the small grains of quartz enclosed in the large porphyroblasts of albite.
The degree of preferred orientation exhibited by the quartz-in-quartz fabric of most of the strongly porphyroblastic schists is illustrated by pl. 38, fig. 10, based upon 450 grains in two ac sections of No. 6238. The b axis of the megascopic fabric is also the b (= B) axis of a quartz girdle with many poorly defined maxima that tend to spread inwards from the periphery of the diagram; points L and M mark the vaguely defined centres about which concentration of the quartz axes tends to be strongest, but even here the values recorded do not exceed 2·5% per 1% area of projection. The noticeably weak definition of b in comparison with that shown by diagrams representing the quartz fabric of schists from other parts of eastern Otago (cf. Turner, 1940) is perhaps connected with the higher proportion of albite present in the rocks from the South Waikouaiti. A comparable inversely proportional relationship between strength of definition of the quartz fabric and amount of mica present has been recorded in rocks from other parts of the world (cf. Phillips, 1937, p. 600). In the schists from Otago, size of the porphyroblasts of albite apparently has no connection with the degree of definition of the quartz fabric, for in pl. 38, fig. 11 (No. 6247) representing 400 grains of quartz in a schist with small albites b is just as poorly defined in fig. 10.
Weak though the maxima are in typical diagrams for the quartz-in-quartz fabric (pl. 38, figs. 10–12), they appear, nevertheless, to be significant in some respects. Two principal centres of concentration
of quartz axes, L and M, can be recognised in approximately the same positions in all three diagrams, even though there are notable differences in the mica fabrics of the same three rocks, Nos. 6238, 6247, 6241 a (e.g. contrast pl. 37, figs. 1 and 2 *). Furthermore, partial diagrams prepared for No. 6238 were found to be almost identical, showing that the quartz-in-quartz fabric is relatively homogeneous. Though the maxima in pl. 38, figs. 10–12 are too weakly defined to warrant speculation as to the mechanism controlling preferred orientation of quartz in these rocks, it may be stated that, no matter which of the current “orientation rules” for quartz may be assumed, no relation can be traced between the quartz maxima and either the visible s-surfaces or the maxima of the mica diagrams. These conditions accord with the writer's earlier conclusions that in the schists of eastern Otago the quartz and mica fabrics are largely independent and that the quartz fabric was the later of the two (e.g. Turner, 1940, p. 81).
(b) Quartz-in-albite Fabric. The quartz-in-albite fabric was investigated in two rocks, Nos. 6238 (ac section) and 6240 (ab section). Single porphyroblasts of albite in these rocks may completely enclose from two to ten grains of quartz, sometimes rounded, sometimes sub-angular in outline, and seldom exceeding 0·1 mm. in diameter. No consistent relationship could be detected between the orientation of the quartz and that of the enclosing albite, nor does there appear to be any preferred orientation of grains within any one porphyroblast of albite. The quartz-in-albite fabric is illustrated by pl. 39, fig. 13 (No. 6238), in which the optic axes of 180 grains were plotted on the ac plane of the megascopic fabric. Preferred orientation of the quartz is much weaker than in the quartz-in-quartz diagram for the same rock (cf. pl. 38, fig. 10). The presence of a b axis coinciding with b of the quartz-in-quartz fabric is suggested by a tendency for the maxima to lie close to the boundary of the projection, but the usual minimum at b is lacking. The maxima around L and M in pl. 39, fig. 13, show some degree of correspondence, possibly fortuitous, with concentrations similarly lettered in pl. 38, fig. 10, but in the unpublished diagrams for No. 6240 such resemblances are almost lacking.
The coarse-grained schists of eastern Otago are believed to be “blastophyllonites” which have developed from the parent grey-wackes by intense mechanical granulation accompanied in the later stages by mineralogical reconstitution under the influence of slowly rising temperature (Turner, 1941, p. 11). On this assumption numerous small granules of quartz, a few of which have been preserved as relicts within the albite porphyroblasts, have recrystallised in the later stages of metamorphism to give an aggregate of much coarser and correspondingly less numerous grains that now make up the quartz-in-quartz fabric. The observed differences between the two types of quartz fabric (e.g. pl. 38, fig. 10, and pl. 39, fig. 13) might therefore appear at first sight to be due to simple intensification of preferred orientation by some process of selective solution during this paratectonic recrystallisation. It will be recalled, however,
[Footnote] * The mica diagram for No. 6247 (not reproduced here) is almost identical with pl. 37, fig. 2.
that some of the albite porphyroblasts have been strongly displaced by rotation subsequent to their growth, so that the quartz-in-albite fabric as recorded in pl. 39, fig. 13, must be regarded as a relict fabric that has been distorted to an undetermined extent by a post-crystalline deformation.
(c) Fabric of Large Quartzose Rods. In rocks from one locality (No. 6241) the quartz is concentrated in large regular rods 20 cm. to 30 cm. in length trending parallel to the lineation of the rock (N.10°E.). In section these rods are lenticular in outline (3 cm. × 1 cm.) with the longer diameter of the lens dipping westward at about 30° and thus cutting obliquely across the subhorizontal schistosity S1 (cf. inset in pl. 37, fig. 3). Coarse slightly undulose grains of quartz (0·4 mm. in diameter) together with less numerous smaller grains of the same mineral (0·05 mm. to 0·1 mm.) make up 90% of these rods. The remainder consists of muscovite in sharply defined thin flakes 0·3 mm. in diameter, a little albite and calcite and rare coarse apatite.
The quartz fabrics of two rods from the same hand-specimen were investigated in some detail, and the results, depicted in pl. 39, figs. 14–17 are summarised as follows:—
(1) For an ac section of both rods two selective elemental diagrams (Knopf and Ingerson, 1938, p. 251) were prepared from measurements of coarser and finer grains respectively. In each case these were almost identical, so that the deductions that the fabric of each section is approximately homogeneous and that the type and degree of preferred orientation of the grains are independent of their size appear to be warranted. The possibility, suggested by the
appearance of the sections between crossed nicols, that the coarser grains are in process of being broken down mechanically to smaller grains, is therefore eliminated. Nor could the larger crystals be growing at the expense of the smaller unless under static conditions and without any alteration of the pre-existing preferred orientation. To the writers it seems most probable that grains of both types have developed simultaneously during the same stage of deformation of the rock.
(2) Pl. 39, figs. 14 and 15, are the collective diagrams including measurements of both large and small grains of quartz in two adjacent rods. Their general similarity indicates substantial homogeneity of the quartz fabric within the field of the hand-specimen. To test this conclusion further, a diagram (pl. 39, fig. 16) was prepared from measurements of 300 grains of all sizes in a section of one rod cut parallel to ab. The principal features of pl. 39, fig. 16, after rotation through 90° into the ac plane, are shown in pl. 39, fig. 17, which should be compared with pl. 39, fig. 15, representing the fabric of the same rod as deduced from the ac section. Substantial but not absolute homogeneity of the fabric is indicated by the comparison.
(3) The dominant feature of the quartz fabric is a well defined b axis which in pl. 39, figs. 14 and 15, departs somewhat from the megascopic b. This departure is much more obvious in pl. 39, figs. 16 and 17, * and has been indicated in the former by drawing a” parallel to the intersection of the quartz girdle and the plane of projection. Comparison of elemental diagrams prepared for different parts of the same section and ultimately combined into pl. 39, fig. 16, shows that the b axis of the quartz fabric maintains a constant inclination of 25° to the megascopic b throughout the whole section. From pl. 37, fig. 4, it will be seen that within the same thin section the b axis of the mica fabric (b′) is inclined to the lineation b at a mean angle of 10°; but, unlike the axis of the quartz girdle, the trend of b′ for the mica fabric swings regularly from being parallel to the lineation b at one end of the thin section to an inclination of 20° at the other. The writer therefore concludes that the b axis of the quartz axis does indeed depart slightly from both the megascopic lineation b and the axis of the mica girdle, and that the quartz fabric must have been the last to develop. It is not unlikely that the observed slight undulation of b of the mica fabric within the ab plane may have been caused by movements accompanying the imprint of the quartz fabric.
(4) In pl. 39, figs. 14–17, there is a strong split maximum XM in which the concentration of the quartz axes is always higher (at X) than elsewhere in the diagrams. This cannot be related to S1 or to the maxima of the mica diagram (pl. 37, fig. 3). no matter which of the various current orientation rules for quartz is invoked. It should be noted, however, that maximum X does indicate a tendency for the optic axis in quartz to lie parallel to the long axis of the cross section
[Footnote] * For quartz as well as for mica diagrams the position of the girdle in relation to the b axis of the megascopic fabric seems to be more precisely given in ab than in ac diagrams.
of the quartzose rods. Less clearly developed but still strong maxima occur in the top right quadrant between L and the point of emergence of the c axis of the fabric. Maximum L in the quartz diagrams corresponds with the similarly lettered submaximum in the mica diagram (pl. 37, fig. 3), and might well indicate a hidden s-surface parallel to which the optic axes of quartz grains and the (001) faces of mica flakes both tend to be oriented.
(5) The maxima in pl. 39, figs. 14–16 are stronger and the minimum at b is more sharply defined than in diagrams such as pl. 38, figs. 10 and 11, which represent the “normal” quartz-in-quartz fabric of typical South Waikouaiti schists in which coarse quartzose rods are lacking. The general arrangement of the maxima, however, is much the same (e.g. compare pls. 38 and 39, figs. 10 and 15), except that in the fabric of the rods maximum X is much more prominent. This maximum is well shown, however, in certain previously published diagrams for the normal quartz-in-quartz fabric of typical porphyroblastic schists from a nearby locality (Turner, 1940, Figs. 70, 72). These general similarities in fabric indicate that the coarse rods of No. 6241 and the slender pencils of quartz in the more typical schists of this district are products of essentially similar processes operating at the same stage of deformation.
(6) The quartz fabric of the rods in No. 6241 may be compared with that of a quartz-rich “mullion” in the Moine series recently described by F. C. Phillips (1937, pp. 595–597; pl. xxxiv, D.11–14). There is a general similarity in the arrangement of the maxima in relation to the axis of the mullion or rod, and in the absence of any obvious relationship between the quartz maxima and the visible s-surfaces, which in both cases are unrelated to the outline of the quartzose body. The writers cannot see in the present case, however, any evidence for the operation of two systems of folding with mutually perpendicular axes, such as has been postulated by Read (1926, p. 121) and by Phillips (loc. cit.) to account of the mullions and rods of the Moine schists.
(d) Relation of Quartz Fabrics to Dip of Schistosity. At most localities within the area here considered the principal schistosity S1 is horizontal or dips at low angles, usually less than 15°. Locally, however, the dip may be much steeper and the question arises as to whether the steep inclination of S1 is to be correlated with deformation during metamorphism or with subsequent tilting movements unrelated to metamorphism (Turner, 1940, pp. 82, 83; p. 185). Pl 38, fig. 11, shows the quartz fabric of a rock (No. 6247) in which the schistosity S1 dips westward at 45° across a strike of 170° which coincides with the lineation. When compared with pl. 38, fig. 10 (typifying the quartz fabric of schists with subhorizontal S1) there is a strong correspondence between the two only when S1 (=ab) is used as the datum for comparison. On the same basis of comparison there is a strong resemblance between the diagrams for the quartz rods of No. 6241 (pl. 39, figs. 14 and 15) and those in a previous paper (Turner, 1940, figs. 70, 72, p. 182) representing rocks about three miles distant in which S1 locally dips 45° eastward across a strike of 20°. It is therefore concluded that in these cases the inclined
schistosity is the result of late local tilting of the originally horizontal S1, without sufficient penetrative movement occurring to disturb the mica or quartz fabrics of the rocks.
Conclusions as to Tectonic History.
At the close of the paper on structural petrology of east Otago schists, a provisional summary of the tectonic history of the area based upon petrofabric evidence was appended (Turner, 1940, pp. 188–190). One of the main purposes of the present paper is to determine to what extent this previously deduced regional tectonic history is borne out by the fabric of a group of porphyroblastic schists, which differ in some degree both structurally and mineralogically from the laminated non-porphyroblastic schists of the surrounding region. Certain differences in detail of fabric, as would be expected have been brought out in the course of this study, but the complex deformation which has moulded the fabric of the porphyroblastic schists of the South Waikouaiti River appears to have involved a sequence of stages comparable with those previously deduced from the fabric of the more generally distributed laminated schists of eastern Otago.
Evidence of strong deformation, probably with considerable lateral transport, is supplied by traces of relict s-surfaces, often strongly folded and transposed by flexure-slip and even rupture of folds on a microscopic scale, still preserved within the albite porphyroblasts of many rocks. At the close of this stage of deformation the rock is pictured as reduced to a fine-grained phyllonitic condition with subhorizontal or gently dipping s-surfaces of slip; the small granules of quartz and other minerals enclosed within many of the large albites give some indication of the prevalent grain-size of the constituent minerals at this time. In the section dealing with the albite fabric evidence was cited to show that growth of the large albites (with simultaneous great increase in grain-size of the other constituent minerals) occurred under almost static conditions after movement along the slip-surfaces had been greatly reduced. If this conclusion is correct, the gently dipping or horizontal s-surfaces that now control the megascopic schistosity must already have been in existence when the albite porphyroblasts and pencils of quartz grains began to develop. The present preferred orientation of the mica crystals is related to these same s-surfaces, and is therefore thought to be partly inherited from previously oriented “seed-crystals,” and partly a product of mimetic crystallisation with the (001) crystal face parallel to s-surfaces already in existence. The prevalent lattice orientation of the mica crystals (figs. 6–9) favours the former of these two mechanisms as being the more important.
The main features of the quartz fabric, especially in so far as these now disagree with the mica fabric, must be referred to a still later stage of deformation subsequent to growth of the albite porphyroblasts. The latter themselves have been affected by mechanical rotation and displacement in many rocks. The tectonic axis of the deformation in its earlier stages, as evidenced by the trend of the megascopic lineation and the axis of the mica girdle, varied 10° on either side of N. But during the later stages of metamorphism this trend had shifted about 20° to the west, as shown, for example, when
the axis of the quartz girdle of pl. 39, fig. 17, is compared with the lineation of the same rock.
Finally, long after metamorphism of the east Otago schists was complete, the schistosity locally has been tilted to angles of 45° about a more or less northerly strike, without disturbance of the internal fabric of the rocks in question.
Bailey, E. B., 1923. The Metamorphism of the South-west Highlands, Geol. Mag., vol. 60, pp. 317–333.
Billings, M. P., and Sharp, R. P., 1937. Petrofabric Study of a Fossiliferous Schist, Mt. Clough, New Hampshire, A.J.S., vol. 34, pp. 277–292.
Eskola, P., 1939. In Die Entstehung der Gesteine (T. F. W. Barth, C. W. Correns, P. Eskola), Berlin, J. Springer.
Fermor, L. L., 1909. The Manganese-ore Deposits of India, Mem. Geol. Surv. Ind., vol. 37, pt. 1, pp. 3–231.
Groves, A. W., and Mourant, A. E., 1929. Inclusions in the Apatites of some Igneous Rocks, Min. Mag., vol. 22, no. 125, pp. 92–99.
Hutton, C. O., 1939. The Significance of Tourmaline in the Otago Schists, Trans. Roy. Soc. N.Z., vol. 68, pp. 559–602.
—— 1940. Metamorphism in the Lake Wakatipu Region, Western Otago, New Zealand, Mem. Dept. Sci. Ind. Res., No. 5, Wellington.
Hutton, C. O., and Turner, F. J., 1936. Metamorphic Zones in North-west Otago, Trans. Roy. Soc. N.Z., vol. 65, pp. 405–406.
Ingerson, E., 1938. Albite Trends in Some Rocks of the Piedmont, A.J.S., vol. 35-A, pp. 127–141.
Knopf, E. B., and Ingerson, E., 1938. Structural Petrology, Mem. Geol. Soc. Amer., No. 6.
Obcel, J., 1927. Recherches sur la Composition Chimique des Chlorites, Bull. Soc. Franç. Min., vol. 50, pp. 75–456.
Phillips. F. C, 1930. Some Mineralogical and Chemical Changes induced by Progressive Metamorphism in the Green Bed Group of the Scottish Dalradian, Min. Mag., vol. 22, no. 129, pp. 230–256.
—— 1937. A Fabric Study of Some Moine Schists and Associated Rocks, Q.J.G.S., vol. 93, pp. 581–620.
Quensel, P., and Gabrielson, O., 1939. Minerals of the Varutrask pegmatite. XIV. The Tourmaline Group, Geol. Forens., vol. 61, pt. 1, pp. 63–90.
Read, H. H., 1926. The Geology of Strath Oykell and Lower Loch Shin, Mem. Geol. Surv. Scotland.
Sander, B., 1930. Gefügekunde der Gesteine, Vienna, J. Springer.
Tilley, C. E., 1925. Metamorphic Zones in the Southern Highlands of Scotland, Q.J.G.S., vol. 81, pt. 1, pp. 100–112.
Turner, F. J., 1933. The Metamorphic and Intrusive Rocks of Southern Westland, Trans. N.Z. Inst., vol. 63, pp. 178–284.
—— 1935. Metamorphism of the Te Anau Series in the Region North-west of Lake Wakatipu. Trans. Roy. Soc. N.Z., vol. 65, pp. 329–349.
—— 1940. Structural Petrology of the Schists of Eastern Otago, New Zealand, A.J.S., vol. 238, pp. 73–106, 153–191.
—— 1941. The Development of Pseudo-stratification by Metamorphic Differentiation in the Schists of Otago, New Zealand, A.J.S., vol. 239, pp. 1–16.
Volk, G. W., 1939. Optical and Chemical Studies of Muscovite, Amer. Min., vol. 24. no. 4, pp. 255–266.
Winchell, A. N., 1933. Elements of Optical Mineralogy, Pt. II., John Wiley and Sons, New York.
—— 1936. A Third Study of Chlorite, Amer. Min., vol. 21, no. 10, pp. 642–651.
Wiseman, J. D. H., 1934. The Central and South-west Highland Epidiorites: A Study in Progressive Metamorphism, Q.J.G.S., vol. 90, pp. 354–417.
Woodland, A. W., 1938. The Harlech Grit Series; II, Petrography and Petrology of some Grits, Geol. Mag., vol. 75, no. 892, pp. 440–454.
—— 1939. The Petrography and Petrology of the Lower Cambrian Manganese Ore of West Merionethshire, Q.J.G.S., vol. 95, pt. 1, no. 377, pp. 1–35.
Fig. 1—No. 6241 a, Muscovite, poles of (001) in 150 flakes. Contours, 10, 8, 4, 0.7%; maximum concentration 13%.
Fig. 2—No. 6238, Muscovite, poles of (001) in 100 flakes; rotated through 180° about a. Contours 10, 8, 4, 1%; maximum concentration 12%.
Fig. 3—No. 6241, Muscovite, poles of (001) in 160 flakes in two quartzose rods. Contours 10, 7, 4, 0.6%; maximum concentration 15%. Quartz rods in which muscovite was measured are shown stippled in the inset circle (diameter 6 cm.).
Fig. 4—No. 6241, Muscovite, poles of (001) in 60 flakes measured in section of rod No. 2 cut parallel to ab; b = b-axis of megascopic fabric; b′= mean direction of b-axis of mica fabric within the field of the measured section.
Fig. 5—No. 6240, Muscovite, poles of (001) in 100 flakes measured in section parallel to S3; rotated 180° about b.
Fig. 6—No. 6240, Muscovite, β in 41 flakes having (001) subparallel to S1 and S3; measured in section parallel to S3 and rotated 180° about b. Contours 4, 3, 1 poles per 1% area.
Fig. 7—No. 6247, Muscovite, β in 43 flakes having (001) subparallel to ab; measured in section parallel to ab. Contours 4, 3, 1 poles per 1% area.
Fig. 8—No. 4701, 3 miles east of Middlemarch, Central Otago; Muscovite, β in 40 flakes having (001) subparallel to ab; measured in section parallel to ab. Contours 4, 3, 1 poles per 1% area.
Fig. 9—No. 6238, Muscovite, β in 31 flakes having (001) subparallel to ab: measured in section parallel to ab. Contours 4, 3, 1 poles per 1% area.
Fig. 10—No. 6238, Quartz, 450 grains in quartzose pencils; rotated through 180° about a. Contours 2.5, 2, 1, 0.2%.
Fig. 11—No. 6247, Quartz, 400 grains in quartzose pencils; rotated through 180° about a. Contours 3, 2, 1, 0.25%.
Fig. 12—No. 6241 a, Quartz, 150 grains in quartzose pencils. Contours 4, 2.7, 0.7%.
Fig. 13—No. 6238, Quartz, 180 grains enclosed in porphyroblasts of albite; rotated through 180°, about a. Contours 3, 1.5, 0.5%.
Fig. 14—No. 6241, Quartz, 290 grains in rod. No. 1. Contours 4, 3, 2, 1, 0.4%; maximum concentration, 7%.
Fig. 15—No. 6241, Quartz, 400 grains in rod No. 2, Contours 4, 3, 2, 1, 0.25%; maximum concentration, 5%.
Fig. 16—No. 6241, Quartz, 300 grains in rod No. 2, measured in ab section. Contours 4, 3, 2, 1, 0.3%; maximum concentration, 6%. Mean position of a axis of quartz fabric is indicated by a”.
Fig. 17—No. 6241, Quartz. Fig. 16, rotated 90° about a to bring the projection into the ac plane. Contours 4, 3, 0.3%.