Schists from the Forbes Range and Adjacent Country, Western Otago
[Read before the Otago Institute, November 14, 1933; received by the Editor, November 20, 1933; issued separately, September, 1934.]
For several years the writer has been investigating problems connected with the schists of Central and Western Otago and Southern Westland. During this time it has become increasingly clear that questions relating to the nature, origin, and age of these rocks can be answered definitely only when details of their petrology and structure over wide areas are known. It is therefore proposed to record the results of petrographic examination of schists from different parts of this region, especially the western portion where the relation to the adjacent unaltered rocks may be observed.
During the past three summers, members of the Otago Section of the New Zealand Alpine Club have been engaged in climbing and exploring the high peaks in the neighbourhood of the Dart and Rees Valleys, north of Lake Wakatipu .(1) As this country is difficult of access and geologically unexplored, climbing parties were asked to collect rock specimens in situ where possible, and in this way twenty-three accurately localised specimens have been obtained from various points on the Forbes Range, the northern end of the Richardson Range, and the intervening valley of the Rees. This paper is based upon a petrographic examination of these specimens, and owing to the incomplete data available, is itself admittedly incomplete.
The writer wishes to record his sincere thanks to Messrs O. V. Davies, K. Grinling, A. Jackson, and S. Russell, who collected the material and made it available for description. At the same time, the hope is expressed that further collecting will be carried on in the near future by climbers visiting the adjacent mountain ranges of this inaccessible region.
The rocks classed as phyllites are represented by five specimens from the southern portion of the area under consideration:—
(1a) (1a)No. 1469, Big Devil Creek, altitude 4000 feet.
No. 1487, Summit of Pluto, Forbes Range.
No. 2088, East Peak of Earnslaw, Forbes Range.
No. 2089, Turret Head, Forbes Range.
No. 2090, West Peak of Earnslaw, Forbes Range.
[Footnote] (1) See N.Z. Alpine Journal, vol. 4, pp. 191–205; vol. 5, pp. 62–76, 196–206.
[Footnote] (1a) Numbers refer to specimens and sections in the Geology Department, Otago University.
Map of the Forbes Range (scale 1 inch = 2 miles) and adjacent country showing the distribution of the metamorphic rocks:—Albite-quartz-epidote-chlorite-phyllites are indicated by crosses; Quartz-albite-epidote-chlorite-schists by circles (in outline); Actinolite-bearing albite-epidote-schists by circles (in solid black); Sericitic schists by triangles (in solid black).
In hand-specimen these rocks are fine-grained, highly fissile, grey slaty phyllites in which schistosity is highly developed, though true foliation (2) is completely absent. In some cases (e.g., Nos. 2088, 2090) relatively coarse quartzo-feldspathic veins 1 cm. to 2 cm. in width cut across the schistosity planes.
Microscopically the chief constituents are seen to be albite, quartz, epidote, and chlorite, which together build up a fine-grained aggregate of crystalloblastic grains, averaging 0.01 mm. to 0.03 mm. in diameter. These are usually accompanied by small but never important amounts of finely flaky sericite, while in two specimens (Nos. 1469, 1487) pale green to colourless actinolite is a minor constituent. Owing to the small size of the grains, the properties of the albite, other than the positive sign and distinctly low refractive index (always less than for Canada balsam), could not be determined, so that the exact composition has not been ascertained. The epidote takes the form of minute xenoblastic yellowish grains, usually aggregated into clusters, and is invariably a highly birefringent ferruginous variety, though in one instance (No. 2088) rare grains of clinozoisite are also present. The chlorite is a pale green, very poorly birefringent type with positive elongation and negative optic sign, usually showing the characteristic anomolous blue interference tint when viewed between crossed nicols. Following Winchell's (1927, p. 379) classification it may be classed as negative pennine or delessite.
Colourless sphene in rounded drop-like granules is a constant and sometimes plentiful accessory; iron-ores are scarce or absent and may be partially replaced by sphene or altered to limonite; rarer minor constituents are calcite (No. 1487), apatite (No. 2089), and tourmaline (No. 2089). The latter mineral is a pale variety, pleochroic from colourless to rather pale blue, and occurs in relatively large perfectly idioblastic prisms ranging up to 0.2 mm. in length.
In addition to the recrystallised minerals described above, several of the phyllites (e.g., Nos. 1469, 1487) contain large relict grains, ranging between 0.2 mm. and 0.8 mm. in diameter, scattered plentifully through the fine-grained reconstituted base. In No. 1469 these relict constituents include brown and green hornblendes, quartz, plagioclase, and rare ilmenite. The amphiboles show partial conversion to mixed chlorite and actinolite. The quartz is slightly strained, but the porphyroclasts of feldspar have suffered considerable crushing, and have the composition of albite. The larger grains of ilmenite are rimmed with dense granular sphene. In No. 1487 the residual grains make up as much as 25% of the total composition and include both quartz and plagioclase. The former occurs as large clear grains with faintly undulose extinction, which have undergone partial marginal granulation. The plagioclase is now albite containing less than 5% of anorthite; it has usually recrystallised as augen and lenticles of small clear granules, but in some cases the twin lamellae of the original relict grain may still persist. In
[Footnote] (2) The distinction between schistosity and foliation is emphasised by Harker (1932, p. 203).
No. 2089 the only indication of the initial clastic structure is the presence of small lenticles and patches of granular quartz or feldspar which do not exceed 0.2 mm. in diameter. In Nos. 2088 and 2090 relict minerals are absent.
The quartzo-feldspathic veins cutting the schists obliquely to the schistosity are well shown in No. 2088. The vein-material consists mainly of quartz (90%) and albite (10%), together with rare, vermicular, transversely fibrous masses of pale yellowish-green chlorite. The quartz and albite appear in the first place to have been coarsely crystalline; but the grains for the most part have been broken down by subsequent shearing, to a mosaic of irregular granules (averaging 0.02 mm.) which enclose partially sheared porphyroclasts ranging from 0.5 mm. to 1 mm. in diameter.
The rocks of this group include the following specimens:—
No. 1466, Ridge 4 miles South of Mount Clarke, Forbes Range.
No. 1468, Twenty-five-mile Creek, Rees Valley.
No. 1482, Big Devil Creek, altitude 4000 feet.
No. 1486, Summit of Moira, Forbes Range.
No. 2068, Little Lochnagar Peak, N. end Richardson Range.
Macroscopically they are thoroughly schistose, finely foliated, dark greenish-grey schists, consisting of thin laminae, alternately rich in quartz and feldspar on the one hand, and in epidote and chlorite on the other. In section they are seen to be completely reconstituted rocks with an average grain-size of between 0.1 mm. and 0.5 mm. Quartz and albite (Ab98An2 to Ab95An5) together make up nearly 75% of the total composition, the former mineral usually being somewhat in excess of the latter. Chlorite (delessite) and an epidote mineral typically are both sufficiently plentiful to rank as essential constituents, though the proportion of either mineral may in particular instances fall below 10%. Small amounts of sericite are invariably present, while sphene and iron ore are constant accessories. In one specimen (No. 1486) there are scattered small prisms of nearly colourless actinolite, while in two others (Nos. 1468 and 1482) a brown pleochroic mineral, referred to the stilpnomelane group has been formed at the expense of the chlorite.
In mineralogical composition the quartz-albite-epidote-chlorite-schists thus resemble the phyllites described in the previous section. They differ from the latter, however, in coarser grain-size, absence of relict minerals or structures, and development of a perfect foliation, so that they would seem to have undergone metamorphism of a somewhat higher grade than the phyllites. No. 1486 is a relatively fine-grained schist (average grain-size = 0.05 mm. to 0.1 mm.) with indistinct foliation, and should perhaps be regarded as a transition type between the two groups.
No. 1468 may be described in greater detail, as typical of this group of schists. The average grain-size is 0.1 mm. to 0.2 mm. The constituents are albite 45%, (1) quartz 25%, chlorite 10% to 15%, epidote 10%, sericite 2% to 5%, stilpnomelane mineral 3% to 5%, magnetite 1% and accessory sphene. The albite occurs as clear usually untwinned grains, the composition of which, as indicated by the optical propeŕties, is not more calcic than Ab98An2 (in sections perpendicular to Z, X to 001 cleavage = 21°; in sections perpendicular to X, Z to 001 cleavage = 14°; sign +; refractive indices greatly less than for Canada balsam). The quartz is in irregular grains showing undulose extinction. The chlorite is of the type classed by Winchell (1927, p. 379) as delessite; it is distinctly pleochroic from pale yellow (X) to moderately deep green (Y and Z), very poorly birefringent, and shows the anomolous blue interference tint between crossed nicols. The epidote mineral is clinozoisite with a birefringence varying between 0.005 and 0.008, and typically occurs as idioblastic prismatic crystals, often crowded with or closely associated with clouds of magnetite dust. Small flakes of sericite are unevenly distributed throughout the section and occasionally are interlaminated with the chlorite.
A deep golden-brown strongly pleochroic mineral of micaceous habit occurs as irregular patches and well-defined laminae in the larger crystals of chlorite (Pl. 23, Fig. 3), and is obviously a derivative of the latter mineral. The pleochroism follows the scheme:
X = golden brown,
Y = Z = deep golden brown,
X < Y = Z.
It is uniaxial and optically negative (in No. 1482 biaxial with a very small optic axial angle), with positive elongation, and a birefringence of about 0.015 to 0.020 as measured by comparison with quartz and albite. The mineral thus closely resembles biotite, from which it differs, however, in its distinctly lower birefringence (this property being approximately constant in all specimens of this mineral so far examined). Further, it appears in every instance to have formed directly from chlorite and is totally independent of the presence of sericite, though the latter mineral may frequently be observed in direct contact with unaltered chlorite. The optical properties agree well with those of the ferruginous members of the stilpnomelane group. Thus Winchell (1927, p. 371) records jefferisite as having identical properties except that the absorption is not so strong (X = colourless, Y = Z = pale brown). Again, Walker (1924, pp. 39, 40) has described a ferric “chlorite” allied to jefferisite as having a high double refraction and yellowish colour. Stilpnomelane itself has similar properties to those of the present mineral, but the birefringence is normally much higher (Hallimond,
[Footnote] (1) Percentages in all cases are rough estimates based on microscopic inspection.
1924, p. 193); nevertheless, Shannon (1920) records a dark brownish green stilpnomelane with a birefringence of 0.015, so that this property would appear to vary considerably within this series.
(3) Actinolite-bearing Albite-epidote-schists.
The rocks grouped under this heading are confined to the north-western quadrant of the area under consideration. The following specimens are included:—
No. 1477, Ridge south of Clarke, Forbes Range.
No. 1478, Ridge south of Clarke, Forbes Range.
No. 1479, Pyramid Saddle, ½ mile west of Clarke.
No. 1480, Summit of Clarke, Forbes Range.
No. 1484, Hunter Creek, Upper Rees Valley.
No. 1485, Summit of Head, Forbes Range.
In hand-specimen the albite-epidote-schists are fine-grained, grey schistose rocks, distinctly foliated usually on an exceedingly fine scale. Typically neither schistosity nor foliation is so perfectly developed as in the schists of the previous group. Quartzo-feldspathic veins sometimes cut sharply across the direction of schistosity.
Microscopically the rocks consist mainly of albite and an epidote mineral, which together make up between 70% and 90% of the total composition, quartz being unimportant or completely lacking. In such rocks actinolite is constantly present in amounts between 1% and 15%, in contrast with the general absence of this mineral in schists belonging to the other groups described from this region. Chlorite (delessite) and sericite typically occur in minor quantities only, though rarely the percentage of either mineral may reach 10% or more (e.g., chlorite in No. 1484; sericite in No. 1477). Sphene and iron ores are the commonest accessories; the yellowish-brown stilpnomelane mineral noted above is present under similar conditions in two sections (Nos. 1479, 1485), while pale bluish-green and brown tourmaline is locally abundant in No. 1478 (Pl. 23, Fig. 1).
The epidote varies in composition from a highly ferruginous yellowish variety with a birefringence of 0.03 to 0.04 (Nos. 1477, 1478, 1479) to less strongly birefringent colourless types approaching clinozoisite (Nos. 1480 and 1484). Indeed, the composition may vary within the limits of a single section. The albite is very poor in anorthite and is never more calcic than Ab95An5. The remaining minerals have the same properties as in the schists described in the previous section.
No. 1479 is typical. It is a completely crystalloblastic rock consisting of albite 40%, epidote 40%, actinolite 10%, chlorite (showing transition to “stilpnomelane”) 5%, sericite 2% to 3%, quartz 2%, and accessory sphene and magnetite. The average grain-size is 0.05 mm. to 0.1 mm. The exact composition of the albite could not be determined owing to lack of suitable sections, but the anorthite content is not greater than 5%. The epidote occurs in
colourless grains and idioblastic prisms with a birefringence of 0.04 (corresponding to 28% of the ferruginous epidote molecule and 13% Fe2O3). There are plentiful parallel slender prisms of colourless actinolite 0.3 mm. or 0.4 mm. in length, with well-marked cross-fracture.
(4) Sericitic Schists.
The following specimens of sericitic schists are recorded:—
No. 1470, 4 miles above Twenty-five-mile Creek, Rees Valley.
No. 1483, Upper Rees Valley (6 miles above Twenty-five-mile Creek).
No. 2066, N. end of Richardson Range.
No. 2067, Spur between Big and Little Devil Creeks, Richardson Range.
No. 2069, 1.½ miles west of Lochnagar, Richardson Range.
No. 2091, Lochnagar Peak, N. end Richardson Range.
These localities all lie within the western half of the map.
Macroscopically these are highly schistose, well-foliated, greyish-green soft schists, with lustrous schistosity planes in the freshly broken specimen. As in the other schists of this area, quartzo-feldspathic veins may sometimes cut across the specimen transversely to the direction of schistosity and foliation.
As seen beneath the microscope the distinctive feature of the sericitic schists is the presence of plentiful sericite which usually makes up between 20% and 30% of the total composition. Quartz and albite together usually amount to about 40%, albite being on the whole the more abundant mineral of the two; whenever the composition of the latter is accurately determinable it lies between Ab97An3. Chlorite (delessite) is usually more plentiful than in the other groups of schists, though not invariably so. Poorly birefringent clinozoisite is always an essential constituent, and rarely is accompanied by small amounts of granular yellow epidote. Iron-ore is a constant accessory, typically in the form of clouds of dust-like particles concentrated in the clinozoisite-rich bands and intimately associated with that mineral. Sphene may sometimes be plentiful (Nos. 2069, 2091), but in other specimens is lacking. Less consistently developed minor constituents are the stilpnomelane mineral (Nos. 1483, 2066), actinolite (No. 1470). and calcite (No. 2091).
The average diameter of crystals and grains of the essential constituents ranges from 0.1 mm. to 1 mm., so that the sericitic schists typically are rather coarser in grain than the rocks of any of the other groups.
No. 2069 is a typical schist of this group, though the percentage of sericite is somewhat lower than usual. It is a relatively coarsegrained rock consisting of albite 30%, quartz 10%, sericite 15%,
chlorite 15%, clinozoisite 30%, iron ore 1%, and accessory sphene. The albite is almost pure, with an anorthite content of not more than 1%. (In sections perpendicular to X, Z to 001 cleavage = 15°, and Z to 010 twinning plane = 74°; in sections perpendicular to Z, X to 001 cleavage = 24°; optically +, refractive indices notably lower than that for Canada balsam.) The mica is sericite with a small optic axial angle; the chlorite is pale green nearly isotropic delessite. The epidote mineral is poorly birefringent (less than 0.01) colourless clinozoisite, occurring as stout idioblastic prismatic crystals, often charged with dusty inclusions of magnetite. Sphene occurs both as large irregular masses and swarms of minute granules. In addition to the clouds of magnetite dust already mentioned, there are scattered larger grains of iron-ore altered to limonite.
A single specimen (No. 1481, North branch Hunter Creek, Upper Rees Valley) among the material under consideration, falls into none of the four groups of rocks described above, and is therefore best placed in a separate class.
It is a fine-grained, highly schistose, well foliated, lustrous grey schist somewhat resembling the sericitic schists in general appearance, but not so coarse-grained. As seen beneath the microscope it is a completely crystalloblastic rock consisting of quartz 70%, pale greenish sericite 10%, clinozoisite and epidote 15%, pale green chlorite 3% to 5%, and accessory fine dusty magnetite and rare sphene. Though albite was carefully looked for, it appears to be quite absent.
(6) Summary of Petrographic Characters.
The criteria by which the various groups of rocks, just described, have been defined are firstly structural and secondly mineralogical. Under the first heading may be included such characters as coarseness of grain, degree of development of schistosity and foliation, and in a few cases persistence of relict grains of clastic origin. The mineralogical criteria mainly concern variations in the relative proportions of the various crystalloblastic constituents, rather than variations in the nature of the minerals themselves.
The grouping adopted above serves the useful purpose of demonstrating differences in chemical composition and to some extent in metamorphic grade, among the rocks under discussion. It also shows that there is a marked tendency for the development of several distinct assemblages of essential constituents. Nevertheless the boundaries between the various groups defined must not be regarded as sharply drawn, for some intermediate types are already known, and no doubt others will be recorded later when fuller data are available.
Though emphasis has so far been placed upon the petrographic differences that exist between the different groups, it is equally important to note certain constant characteristics that pertain to
all the rocks described, stamping them as members of a single rock-series, with a common tectonic and metamorphic history, and all of similar origin. Some of the most important of these constant characters may be summarised thus:—
(a) Schistosity is always well developed, and except in the finer-grained members there is also perfect foliation.
(b) The great majority of the rocks have been completely reconstituted, relict minerals being absent except in a few of the finer-grained phyllites.
(c) Strain structures, especially undulose extinction in the grains of quartz and albite, are common.
(d) Though there is a great variation in the proportions of the various minerals present, the actual assemblages of minerals, including accessory and minor as well as the essential constituents of the different rocks, is surprisingly constant. Thus albite, quartz, an epidote mineral, chlorite, sericite, sphene, and magnetite are found in almost every specimen. Common but less frequently developed minerals are actinolite and a member of the stilpnomelane group, this latter being formed at the expense of chlorite. Still rarer accessories are tourmaline, calcite, and apatite. The optical properties of the chlorite are constant and conform to those given by Winchell for delessite. The albite is never more calcic than Ab95An5 and usually has a composition of about Ab97An3.
(e) In a number of specimens (belonging to all five of the above groups) veins of quartz and albite, accompanied by minor vermicular chlorite, cut the schistosity at high angles. The constituent minerals of these veins invariably show marked effects of cataclasis.
In a recent account of the schists and other metamorphic rocks from a region whose southern border lies some twenty miles north of that under present consideration, the writer (Turner, 1933) described a series of quartzo-feldspathic schists, the least metamorphosed members of which have many characters in common with those from the Forbes Range and surrounding country, though differing in petrographic detail. Indeed, the schists of the two areas are continuous with each other. The conclusion previously reached (Turner, 1933, pp. 214, 215, 220), viz., that the schists are the metamorphosed equivalents of greywackes, applies also in the present instance and need not be discussed further.
The Course of Metamorphism.
(1) Nature and Grade of Metamorphism.
It has already been shown (Turner, 1933, pp. 251–3) that the schists of western Otago and southern Westland owe their present condition to progressive dynamothermal metamorphism caused by
folding, combined with the contact effects of invading granitic batholiths in the deeper zones of intense alteration. The constant assemblage albite-quartz-epidote-chlorite-sericite-sphene which is so characteristic of the schists of the present area, places them all within the chlorite zone. In this zone, representing the lowest of the three grades of metamorphism so far recognised in the Otago-Westland region, and remote from the granites of the deeper zones, the metamorphism is almost purely of the dynamothermal regional type. Yet it should be noted that even here, as in the chlorite zone of Westland fifty miles north, the occasional presence of tourmaline bears witness to the passage of boric fluids emanating from the subjacent granite mass.
All the rocks described in this paper belong to Eskola's green-schist facies; furthermore, all belong to that sub-facies which Tilley (1924) recognised as characterised by chlorite-epidote-albite, in contrast with the hornblende-acid oligoclase-epidote sub-facies which corresponds to a slightly higher grade of metamorphism. (1) Nevertheless, slightly different grades of metamorphism may be recognisable even within tthe limits of a single zone, among rocks of the same facies, and the rocks under consideration afford a clear instance of this. Though it has not yet been found possible to subdivide the chlorite zone, it is nevertheless clear that the grade of metamorphism increases from south-west to north-east within the limits of the map. Thus the non-foliated, often incompletely reconstituted phyllites with a single exception are confined to the southern end of the Forbes Range, in the south-western corner of the map. While allowance must be made for initial differences in texture and composition, the phyllites undoubtedly represent a slightly lower metamorphic grade than do the coarser, well-foliated, completely reconstituted schists to the north and east. The presence of even less completely altered semischists and sheared greywackes still further to the southwest beyond the limits of the map (on the western side of the Dart Valley), lends further support to this view.
(2) Mineralogical Changes.
The mineralogical changes involved in the metamorphism may be summarised briefly as follows:—
(a) Clastic plagioclase breaks down to almost pure albite, with the liberation of lime and alumina which enter into the epidote mineral.
(b) Hornblende (in the one instance where relict grains of this mineral still persist) is transformed to a mixture of actinolite and chlorite, a reaction involving liberation of lime, alumina, and ferric iron, all of which enter into the epidote.
[Footnote] (1) To avoid confusion it should be noted that the writer, in a paper which has appeared since the present paper was submitted for publication, has suggested a slight modification of Tilley's scheme, though this is immaterial to the present discussion. (F. J. Turner, The Genesis of Oligoclase in Certain Schists, Geol. Mag., vol. lxx, pp. 529–541, 1933.)
The actinolite-bearing schists are believed to owe their content of actinolite to similar replacement of now completely destroyed hornblende and probably pyroxene also. This lends further support to the suggestion recently advanced by the writer (1) that under some conditions actinolite may form from hornblende or pyroxene, even at the lowest grades of metamorphism within the chlorite zone.
(c) Though the transition was not actually observed, it is obvious that in some cases the original amphiboles and pyroxenes of the parent greywackes have broken down still further to a mixture of chlorite and epidote. As pointed out by Vogt (1927, p. 410) this involves the removal of a considerable amount of lime, some of which has been fixed in the present rocks as sphene and occasionally as calcite.
(d) Clastic ilmenite is replaced by sphene.
(e) In the fine-grained phyllites the epidote is a yellow, highly ferruginous variety occurring as xenoblastic grains and granular clusters. In the more coarsely crystalline schists of slightly higher grade the mineral has recrystallised, usually completely, as idioblastic prisms of colourless clinozoisite or more rarely as poorly ferriferous epidote. This may account for the clouds of magnetite dust that are frequently associated with or even enclosed by the clinozoisite. Progressive decrease in the iron-content of epidote with increasing intensity of metamorphism has been noted by various observers (e.g., Tilley, 1923, p. 185; Sugi, 1931; Turner, 1933, pp. 240, 243), usually, however, at higher grades of metamorphism than those with which we are at present concerned.
(f) Crystallisation of a brown pleochroic stilpnomelane mineral appears from the available literature to be a sufficiently unusual feature to deserve fuller comment.
The mineral was at first mistaken for biotite, which it closely resembles but for minor differences in the optical properties, especially a distinctly lower birefringence. However, both in the area under discussion and throughout an extensive region further south-west, the mineral in question appears in rocks representing an extremely low grade of metamorphism, whereas brown biotite is always developed only when a much more advanced grade is reached (e.g., Barrow, 1912, p. 2; Harker, 1932, p. 214; Turner, 1933, pp. 239, 240). (2) Finally, the characteristic mode of occurrence of the mineral in question, as a direct derivative of chlorite in no way dependent upon the presence or proximity of sericite, is itself strong evidence against identification as biotite.
[Footnote] (1) Op. cit., Geol. Mag., 1933, p. 534.
[Footnote] (2) The mineral is quite distinct from the green biotite that often makes an early appearance in green schists within the chlorite zone.
Hallimond (1924) has suggested that stilpnomelane may play an important part in schists originating by metamorphism of sediments containing abundant ferruginous chlorite but little sericite. According to Winchell (1927, p. 379), delessite such as that of the schists under discussion is a relatively ferruginous variety of chlorite. Comparison of the compositions of delessite and stilpnomelane as given by Winchell shows that the latter may well have originated by a change involving complete oxidation of the ferrous iron present in the chlorite. Possibly in the schists described above a further supply of available iron in the magnetite has also been utilised.
In conclusion, it may be noted that the presence of leverrierite, an aluminous member of the stilpnomelane group, as a crystalloblastic mineral in sericite-schists has been demonstrated clearly by C. S. Corbett (1925).
(3) Structural Changes.
Four successive stages of structural evolution may be recognised, characterised respectively by the following processes: Cataclasis of the original mineral grains, crystalloblastic growth of the reconstituted minerals, formation of transgressive quartzo-feldspathic veins, and finally a further cataclasis of the reconstituted minerals.
Judging from the persistence of strained, partially altered relict grains of clastic origin only in the fine-grained phyllites (the least metamorphosed rocks), reduction of grain-size by cataclastic degradation of the constituents of the original sedimentary rock is the main feature of the earliest stage in metamorphism. This purely mechanical effect is accompanied by complete chemical reconstitution of the rock, and simultaneous development of schistosity. Thus the initial coarse-grained greywacke is transformed into a phyllite consisting of an assemblage of minerals in mutual equilibrium under intense stress. It is of course probable that some of the phyllitic rocks owe their fine grain to correspondingly fine grain in the parent sediment, but in the majority of the cases examined, cataclasis appears to be the responsible factor.
The second stage (not shown in rocks from the south-western corner of the map) involves the crystalloblastic growth of the previously crushed, largely reconstituted mineral grains. By this process relict grains disappear as the new assemblage attains equilibrium, mineral grains increase in size, the grains of those minerals which are placed high in the crystalloblastic series (sphene, epidote, clinozoisite, actinolite) assume idioblastic outlines, and foliation is developed parallel to the schistosity by a process of metamorphic diffusion. The resultant rocks are schists as distinguished from the phyllites of lower grade. Sporadic introduction of tourmaline may be assigned to this stage. The orientation of the tourmaline prisms with their long axes in the plane of schistosity may be due to growth either under stress (compare Harker, 1932, p. 194) or according to Sander's principle of mimetic crystallisation after cessation of deformation (compare Knopf, 1933, pp. 461, 462).
Following upon the cessation of the above processes, which, taken in conjunction, constitute the main phase of metamorphism, comes a minor phase characterised by the formation of quartz-albite veins cutting transversely across the schistosity and foliation. The occurrence of similar veins is a familiar feature of a number of regionally metamorphosed areas. They are usually attributed (e.g., Tilley, 1923, pp. 194, 195, 204) to segregation of quartz and albite derived from the surrounding schist, under the influence of circulating waters after the cessation of metamorphism. This explanation accords well with the observed facts in the area under discussion. The fact that the veins run obliquely to the schistosity suggests that they were formed long subsequently to the main metamorphism, when a new system of stresses influenced the rock mass.
A final phase of relatively mild, purely dynamic metamorphism is clearly indicated by the marked cataclastic effects shown by the vein material. The undulose extinction so frequently observed in the constituent grains of the schists themselves should probably be correlated with this final movement.
(4) Evidence of Repeated Metamorphism.
The writer has recently discussed the available evidence as to the period at which metamorphism of the schists of Otago and Westland took place, and has advanced the tentative view that the main metamorphism occurred in Palaeozoic times, and was followed by relatively mild epi-zone changes accompanying the Hokonui orogeny of the Early Cretaceous (Turner, 1933, pp. 255, 256). On this assumption, while the main metamorphism of the rocks described in this paper may be assigned to the Palaeozoic Era, the writer sees in the later phases of quartz-albite veining and minor cataclasis, the much less pronounced changes accompanying the Hokonui deformation.
It must be borne in mind that the effects of polymetamorphism of this type, where both phases of metamorphism have been of relatively low grade, are difficult to detect with certainty, for the second metamorphism is unaccompanied by retrogressive mineralogical changes, and the evidence is purely structural. In emphasising the difficulty of interpreting evidence of this type Eleanora Knopf (1933, p. 462) writes as follows: “… An area of phyllonites may show more than one phase of deformation, yet these different stress conditions may have belonged to one and the same period of progressive metamorphism. But, on the other hand, the regional study may show up in some rocks relics of a higher rank metamorphism than the present, which would indicate that the recorded changes in stress are associated with polymetamorphism.” Thus while the structural evidence recorded above indicates pronounced changes in stress conditions during the metamorphic history of the present rocks, this taken alone does not necessarily imply the existence of more than one metamorphism. But taken in conjunction with the
marked evidences of retrogressive metamorphism observed in southern Westland (Turner, 1933, pp. 249–251), it does afford strong corroboration to the view that there has been widespread repeated metamorphism in the Otago-Westland region.
Babbow, G., 1912. On the Geology of the Lower Deeside and the Southern Highland Border, Proc Goel, Assn., pp. 1–17.
Corbett, C. S., 1925. Leverrierite as a Schist-forming Mineral, Am. Jour. sci., ser. 5, vol. 10, pp. 247–268.
Hallimond, A. F., 1924. On Stilpnomelane from North Wales. Min. Mag., vol. 20, no. 104, pp. 193–197.
Harrer, A., 1932. Metamorphism, London, Methuen and Co.
Knopf, Eleanora B., 1933. Petrotectonics, Am. Jour. Sci., ser. 5, vol. 25, pp. 433–470.
Shannon, E. V., 1920. Diabantite, Stilpnomelane, and Chalcodite of the Trap Quarries of Westfield, Massachusetts, proc. U.S. Nat. Mus., vol. 57, pp. 397–403.
Sugi, I<., 1931. On the Metamorphic Facies of the Misaka Series in the Vicinity of Nakagawa, Prov. Sagami, Jay. JOUI. Geor. and Geogr., vol. 9. no. 1–2, pp. 87–142.
TITLLY, C. E. 1923. The Petrology of the Metamorphosed Rocks of the Start area (South Devon), Q.J.G.S., vol. 79, pp. 172–204.
— 1924. The Facies Classification of Metamorphic Rocks, Geol. Mag., vol. 61, pp. 167–171
Turner F. J., 1933. The Metamorphic and intrusive Rocks of Southern Westland, Trans N.Z. Inst., vol. 63, pp. 167–171.
Voot, T., 1927. Bultelmefelteta Geologi og Petrografi, Norgen. Geol. Undersokelse, No. 121.
Walker, T. L. 1924. Cheical and Microscopic Examination of Ferrc and Ferrous Vein meterials and of Chart from the Kelly Mine, Univ. Toronto studies Geol. Ser. no. 17, pp. 38–41.
Winchell, A. N. 1927. Elements of Optical Mineralog, Part 2, New York, John Wiley and Bons
Fig 1—Large idioblastic prisms of tourmaline in actmolite bearing albite-epidote-sel [ unclear: ] (No. 1478)
Fig2—Aereculau prisms of actmolite and granular epidote in actinolite-leaning albite-epidote sel [ unclear: ] (No 1480).