Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 76, 1946-47
This text is also available in PDF
(783 KB) Opens in new window
– 481 –

Contributions to the Mineralogy of New Zealand. Part 3.

[Read before the Otago Branoh, November 12, 1946; received by the Editor, December 3, 1947; issued separately, August, 1947.]

Summary.

The chemical and optical properties of two octophyllite micas (biotite and phlogopite) and chromian leuchtenbergite are presented herewith. Structural formulae have been deduced on the basis of existing X-ray data, and notes have been made on the modes of occurrence.

Biotite.

The mica studied herein was found near Old Point, Charles Sound, as deep brown to black pseudohexagonal books that averaged 30–35 mm. in diameter,* and 8–10 mm. in thickness. The mineral occurred together with coarse xenoblastic water-clear quartz in narrow lenses up to 20–25 feet in length and these were associated with oligoclase-quartz-biotite gneisses with trondhjemitic affinities, and petrographically comparable to the rocks described by Turner (1939, pp. 580582) from Doubtful Sound further south. The orientation of these lenses appears to conform with the foliation in the gneisses, but in some instances as, for example, on the south side of the head of Foot Arm, Nancy Sound, these “pegmatitic” zones have been observed cutting the foliation of the associated feldspathic gneisses at high angles.

The mica was readily prepared for chemical analysis by careful hand-picking beneath a hand-lens after a preliminary splitting into fine flakes by rubbing in a mortar with a rubber-covered pestle. Finally flotation in bromoform-methylene iodide mixtures served the double purpose of purification and density determination.

Two points are evident from a study of the analysis of this mica (Table I, Anal. A):

(1)

The composition lies well within the known range of composition of rock-forming biotites (Shand, 1927, p. 19), but it is rather poorer in ferrous iron and correspondingly richer in magnesia, though still well removed from the phlogopitic micas, than is normally the case for biotites from acid rocks. It is suggested that this relatively low FeO: MgO ratio can be correlated with the fact that the lenses or schlieren containing the biotite are associated with gneisses decidedly less siliceous than normal granites. This would be in harmony with the views of Chapman and Williams (1935), who found that in passing from the biotite of a gabbro to the biotite of a granite, there is, among other changes, an increase in FeO and Fe2O3 and a decrease in MgO. A similar relationship was observed by Hurlbut (1935) in biotites in a tonalite-gabbro complex.

[Footnote] * Occasional plates up to 90 mm. were observed.

– 482 –
  • 0 74 per cent. of barium oxide has been found in this mica. For purposes of comparison three analyses of biotites from granitic rock types have been selected and set out in Table I. In the search through the literature for analyses comparable to A, it was difficult to find biotites with suitable values for FeO and MgO, and, further, very few biotites appear to contain BaO. It is realised that the apparent rarity of BaO in biotites may to some extent result from failure to estimate this constituent, which, if present in amounts less than one per cent., might not be recognised.

    Although the structure of biotite has as yet not been fully worked out, the work of Pauling (1930), Mauguin (1928), Mauguin and Graber (1928), and others leaves little doubt that the formula for biotite may be expressed as (OH,F)2 W (X,Y)2–3 Z4O10;* it should be noted that the recent work of Hendricks (1939), though drawing attention to the complexity of the structure in micas, does not suggest that any departure from this structural formula is necessary. Four molecules each conaining 12 oxygen atoms make up the unit cell and on this basis the analysis of the Charles Sound biotite has been recalculated; the analysis showing the ionic replacements is set out in Table II. It can be seen that although the analysis agrees well with the structural requirements, the following points require some comment: (1) The value for the X,Y group is 2.75, thereby lying well within the range as postulated by Pauling for these cations. But since there is very limited isomorphism between the heptaphyllite and octophyllite micas, there is little possibility of the X,Y group showing a complete range from 2–3. Recalculation of a number of biotite analyses by the writer fully supports the point made by Schiebold (vide Bragg, 1937) that the number of cations in the Y portion is two in muscovite, and that in biotite (and phlogopite) the number in this position is always in the neighbourhood of three. Thus it appears that the range of the X,Y group cations is much more restricted than Pauling believed and the lower limit for this group is very likely not less than 2.75.

  • The large size of the barium ion requires that it should be grouped with potassium.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table I.—Analyses of Biotites.
A. B. C. D.
SiO2 36.22 36.34 35.62 34.62
Al2O3 16.39 13.88 15.24 16.82
Fe2O3 3.92 4.49 4.69 8.97
FeO 14.41 15.00 13.67 14.31
TiO2 3.02 3.29 2.61 2.15
CaO nt.fd. 0.28 0.95 0.81
MgO 11.11 11.80 12.70 5.49
BaO 0.74 0.26 0.30
MnO 0.16 0.52 0.74 0.47
V2O3 0.05 0.06
P2O3 trace 0.07
F 0.08
Na2O 0.37 0.57 0.50 1.87

[Footnote] * To avoid confusion, the writer prefers to use the more conventional symbols. It should be remembered that Pauling employed X for cations indicated in this paper as X and Y; he also used Y for cations in four-fold co-ordination, but Z conforms, with standard usage.

– 483 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

K2 8.97 8.80 7.72 8.26
H2O+ 4.00 3.60 4.36 5.88
H2O− 0.75 0.70 0.94
100.19 99.34 100.06 99.95
O for F 0.03
100.16
Sp. Gr. 2.98±0.01

Key to analyses in Table I.

A.

60 chains north-west from Old Point, Charles Sound, New Zealand. Anal.: F. T. Seelye.

B.

Biotite from hornblende-bearing biotite granodiorite, Senmaya, Rikutyu. Anal.: S. Tanaka (Tsuboi, 1935, p. 112 [4]).

C.

Biotite No. 1751 from quartz monzonite, south-west of Blood's Station, Alpine County, Big Tree Quadrangle. Anal.: W. Valentine (Turner, 1899, p. 295).

D.

Biotite from biotite-muscovite granite, Kosista, Eastern Tatra Mountains, Czechoslovakia. Anal.: Z. Weyberg (Weyberg, 1912, p. 399, table 2, anal. 2c).

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table II.—Recalulation of Analysis A (Table I).
Wt. per cent. (O,OH,F.) Metals.
SiO2 36.22 1.206 2.733 4.00
Al2O3 16.39 0.483 1.460 1.267
0.193
TiO2 3.02 0.076 0.172
Fe2O3 3.92 0.075 0.226
FeO 14.41 0.200 0.906 2.75
MgO 11.11 0.275 1.246
MnO 0.16 0.002 0.009
Na2O 0.37 0.006 0.054
K2O 8.97 0.095 0.861 0.94
BaO 0.74 0.005 0.023
H2O 4.00 0.222 2.012
F2 0.08 0.004 0.018 2.03

Formula: (O,OH,F)2 03 (Mg,Fe″,Fe′″,Mn,Ti,Al)2 75 [(Si,Al)4O10] (Ba,K,Na)0.94

(3)

The contribution of aluminium to the sheets of linked tetrahedra is considerable and only the relatively small amount of 0.193 is found to have co-ordination number 6.

(4)

Titanium has been included entirely in the six-co-ordinated group and there is no suggestion that any of the titanium is present in the silicon-oxygen tetrahedra as has been postulated by a number of writers (Dixon and Kennedy, 1933; Barth, 1931; Deer and Wager, 1938) for some titaniferous pyroxenes.

The optical properties have been determined 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=1.586* ± 0.003

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

β = 1.643

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

γ = 1.643

[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 = 0.057

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

2V = 0–8°

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

ρ < v distinct

Optic axial plane is parallel to 010.

Pleochroism:

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

X = yellowish-brown

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Y = brown

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Z = brown

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Z = Y > X

[Footnote] * All values in this paper are for line D.

– 484 –

Origin of the Biotite.

The mica was found in a pegmatite-like lens surrounded by quartz-oligoclase-biotite gneiss, the origin of which is doubtful according to Turner (1939, p. 582). However, important in this connection is the occurrence at Henry Saddle, sixteen miles to the north-east, of a barium-bearing muscovite forming large plates in a pegmatitic body (Hutton and Seelye, 1945, pp. 160163) and associated with quartz-oligoclase gneisses that are quite comparable to those of Charles and Doubtful Sounds. The gniesses themselves do not contain significantly more barium than von Engelhardt (1936) would allow as an average for granitic to intermediate rocks, although strontium is rather above the average (Noll, 1934) for that range.

In the Fiordland region pegmatite-like bodies are frequently seen, but in all the instances observed the minerals present are similar to those found in the adjacent gneisses, viz., quartz, oligoclase, biotite, muscovite, and in some examples hornblende; potash feldspars are not common. In few instances fugitive minerals such as orthite were observed, and then these were found in the more clearly igneous pegmatites. It is believed, therefore, that under conditions of high shearing stress and high temperatures the minerals constituting the assemblages in the pegmatitic bodies have been “sweated out” of the gneisses and subsequently segregated into lenses and veins. In all cases these bodies are mineralogically simple, consisting of quartz-biotite, quartz-muscovite, or in some cases quartz-oligoclase. That is, the minerals present are those characteristic of the rocks with which they are associated and from which they have been derived; in this respect then, these pegmatitic rocks have much in common with Holmquist's (1921) venites.

The relatively high percentages of barium in the micas are, it is considered, not coincidental, but they may be explained on the basis of the work of Goldschmidt and his collaborators on the factors controlling the migration of trace elements into a crystal lattice. It is well known that in the last fraction of a differentiating rock magma there is an extreme concentration of the alkali metals, that is of cations of large radius. This is particularly noticeable with the composition of micas formed during successive stages of hydrothermal replacement in pegmatites (Stevens and Schaller, 1942), when such large cations as rubidium and caesium may reach relatively high percentages. Conversely, if a migmatic liquid of comparable composition were to arise by differential anatexis of a series of rocks, then the first elements to be “sweated-out” of the country rocks would be those of large radius, and since barium, is one of the larger cations (r*= 1.43A), a concentration of the first pore-liquid might be expected to contain a much higher proportion of barium than is to be found in the source rocks. Since oligoclase is not often present

[Footnote] * This is the value given by Goldschmidt for the barium ion with co-ordination number 6; however Palache et al (1946) give the figure 1.36A, which presumably has been obtained by averaging the values found by Goldschmidt, Zachariasen (1931), and Neuburger (1936).

– 485 –

with the mica in these pegmatitic zones barium could only enter the mica lattice. Now quite apart from the percentages of potassium and barium in the original gneisses, the relative proportions of barium and potassium in this palingenetic liquid will, it is believed, depend upon two factors:

(a)

The valency or ionic charge of barium and potassium.

(b)

The radius of the barium and potassium ions.

Goldschmidt (1937) has shown that where two ions of similar size but different valency are available, admittance of one or other to a crystal lattice is governed by valency; that is, barium with the higher electrostatic charge would be favoured before potassium. In these circumstances there would be a tendency for potassium to be more completely segregated from the source rocks than for barium. Insufficient data are available, however, to give any opinion thereon in the case of the Fiordland rocks, but it should be pointed out that Wager and Mitchell (1943) are of the opinion that ionic size, and not electrostatic charge, has been the prime factor in regularising the distribution of barium during the diversification of the Kangerdlugssuaq magma. Thus the results of the interplay of these two factors in any particular case cannot be clearly forecast.

Phlogopite.

Phlogopite is an important accessory constituent of marbles associated with feldspathic, garnetiferous, hornblendic, and calc gneisses in a number of localities in Fiordland. Chief among these are the outcrops at Deep Cove, Hall's Arm, and Kellard Point (all of Doubtful Sound; Turner, 1939), Anxiety Point (Nancy Sound), and on either shore of Caswell Sound, about a mile and a-half from Styles Island. In the latter locality phlogopite is not readily seen in the white, sugary, closely jointed marble, although in other localities the mica forms very conspicuous flecks and distinctive foliae.

For analysis the mica was separated from the marble as follows:

(1)

The marble was crushed and the coarse powder so obtained was ground finely in a porcelain mortar; grinding and screening (150-mesh) were carried out repeatedly, and since the mica flakes were relatively resistant to the grinding process, a high concentration of phlogopite was soon obtained on the screen.

(2)

The product from (1) was concentrated further by careful manipulation on a cardboard tray, the flaky mica tending to remain behind on the tray, whereas the more equidimensional grains of calcite, diopside, etc., rolled off.

(3)

The mica-rich powder was treated with 20 per cent, acetic acid in the cold, and phlogopite together with about 5–10 per cent, of quartz, diopside, feldspar, and clinozoisite remained as an insoluble residue.

(4)

The pure mica was obtained finally by centrifuging in bromoform-benzene mixtures.

On analysis, the Naney Sound phlogopite has a somewhat lower magnesia percentage than is usually found; phlogopite analyses

– 486 –

with comparable MgO figures normally contain about 8 per cent. FeO or some other six-coordinated ion. Only minor replacement of (OH) by F has taken place, and this substitution is decidedly less than is usual in phlogopite.

As in the case of biotite the phlogopite analysis has been recalculated on the basis of 12 (O,OH,F) atoms to the unit cell, and it will be noted (Table IV) that there is good agreement with the structural formula (OH,F)2 W(X,Y)2–3 Z4O10. Once again the figure for the combined X and Y group cations lies between the limits of 2.75–3; the alkali metal group is slightly deficient.

The optical properties have been determined 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 = 1.546 ± 0.004

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

β = 1.588−1.590

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

γ = 1.590 ± 0.001

[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 = 0.044

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

X = colourless to very pale yellow

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Y = Z = brownish-yellow

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

X < Y = Z

2V: In eleven determinations the range of 0–11°, with eight flakes between 8–11°; a twelfth flake gave 13°.

Dispersion: nil.

Chromian Leuchtenbergite.

Numerous veinlets of chlorite were found in outcrops of meshserpentinites in Gordon's Quarry near Kaukapakapa, North Auckland; the veinlets were completely monomineralic. The mineral has the very pale whitish-green to silvery colour of the leuchtenbergite-variety of clinochlore, although the books are often brown, due to iron-staining. In a few undamaged plates different optical orientations were apparent in different sectors, suggesting twinning on the mica law. The crystals are often large, measuring up to 25 mm. in diameter with a thickness up to 8 mm., but a diameter of approximately 10–15 mm. is most usual. There is an eminent basal cleavage, and the cleavage laminae are flexible but only very slightly elastic; the laminae are very fragile, somewhat curved, and readily tear. There is a very pronounced pearly lustre on surfaces parallel to (001) and the laminae are opaque, or translucent only in very thin flakes.

A careful selection of material was made for analysis, and oxidised or stained flakes were rejected in all cases. It will be seen from the analysis (Table V) that the composition of the mineral is comparable with that of clinochlore in the chlorite group; the silvery green to white colour is clearly the result of a low iron content and hence the classification of the mineral as leuchtenbergite is justified. It will be clear, however, that the composition is not far removed from the field of positive penninites, since, for the latter group, Orcel (1927, pp. 424425) gives the upper limit of alumina at 13.8 per cent. It will be seen shortly that the optical constants are in harmony with the chemical data. The other interesting feature of this analysis is the presence of a noteworthy amount of chromium and also minor nickel.

– 487 –
Table III.—Analyses of Philogopites.
1 2 3
SiO2 40.95 40.69 39.66
Al2O3 17.28 16.62 17.00
Fe2O3 0.43 0.27
FeO 2.38 0.03 0.20
TiO2 0.82 0.51 0.56
CaO nt.fd.
MgO 22.95 26.28 26.49
BaO 0.03 0.62
MnO trace 0.02
V2O3 nt.dtd.
F 0.62 nil 2.24
Na2O 0.16 1.08 0.60
K2O 9.80 10.18 9.97
H2O+ 4.23 4.40 2.99
H2O− 0.48
100.13 99.81 100.60
O for F 0.26 0.94
99.87 99.66
Sp.Gr. 2.78±0.01.

Key to Analyses in Table III.

1.

From marble, 40 chs. S.E. of Anxiety Point, Nancy Sound. Anal.: F. T. Seelye.

2.

From marble, Ascona, Tessin; analysis includes 0.29 per cent. Ti2O3. Anal.: J. Jakob (Jakob, 1937).

3.

Burgess, Ontario. Anal.: F. W. Clarke and E. A. Schneider (Clarke and Schneider, 1890, p. 411).

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table IV.
Wt. per cent. (O,OH,F.) Metals.
SiO2 40.95 1.364 2.862 4.00
Al2O3 17.28 .507 1.419 1.138.281
Fe2O3 0.43 .008 .017
FeO 2.38 .033 .138 2.87
TiO2 0.82 .020 .042
MgO 22.95 .569 2.388
Na2O 0.16 .002 .017 .89
K2O 9.80 .104 .873
H2O+ 4.23 .235 1.973
F2 0.62 .033 .139 2.11

Formula: (O,OH,F)2 11 (K,Na) .89 (Mg,Fe″,Fe′″,Ti,Al)2 57 (Si,Al)4O10

Pauling's (1930) formula deduced from X-ray studies is not entirely satisfactory, since it does not take into account a most important characteristic of the chlorite family, that aluminium and trivalent iron enter both the X and Z groups in equal amounts, that is, iron in addition to aluminium occupies some of the positions in the tetrahedral network. Incidentally, it is possible that this unusual condition in the tetrahedral network has effected the slight curvature of the plates of the chlorite, a condition which Berman (1937, p. 381) suggests results in the distortion found in cronstedtite; in this latter case, since no aluminium is available, Berman believes the vacant positions in the tetrahedral network can be occupied by iron only.

Berman's formula for the chlorite family,

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

[Xn Z1 O10 (OH) 2(n-2).SH2O,

or when fully expanded,

– 488 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

(Mg,Fe″,Al,Fe′″)n (Al,Fe′″,Si)4O10(OH) 2(n-2).SH2O]

takes this distribution of Fe” into account and is, therefore, much more satisfactory.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table V.—Analyses of Clinochlore.
1 2 3 4 5
SiO2 31.87 30.90 32.98 30.62 32.42
Al2O3 14.51 17.18 15.94 17.51 16.95
Cr2O3 1.10 nt.fd. 0.85
Fe2O3 1.86 0.02 0.64 1.60 0.65
TiO2 0.17 str.tr.
FeO 3.57 9.06 0.73 3.62 1.05
MgO 32.76 29.83 36.43 32.35 35.88
NiO 0.28 nt.dt.
MnO trace 0.14 0.35
CaO nt.fd. nt.fd. nt.fd.
Na2O 0.04 nt.fd. trace
K2 0.02 nt.fd.
H2O+ 13.05 12.60 13.00 12.62 12.79
H2O− 0.90 0.35 0.78 1.54
100.13 100.08 100.50 100.21 100.56
Sp.Gr. = 2.60±0.01

Key to Analyses in Table V.

1.

Clinochlore from Gordon's Quarry, Kaukapakapa. Anal.: F. T. Seelye.

2.

Clinochlore from Red Mountain, South Westland. Anal.: C. Osborne Hutton (Hutton, 1936).

3.

Leuchtenbergite from metamorphosed dolomite, Syunkôdô, North Korea. Anal.: M. Teshima (Satô, 1933).

4.

Clinochlore from serpentinite, Southern Serbia. Anal.: M. Tajder (Tajder, 1938).

5.

Chromiferous clinochlore associated with chromite, Patevi, 20 km. S.S.E. of Atakpamè. Anal.: M. J. Orcel (Orcel, 1925).

The analysis has been recalculated on the basis of 18(O,OH) atoms to the unit cell and the grouping of ions is set out in Table VI.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Table VI.
Wt. per cent. (O,OH.) Metals.
SiO2 31.87 1.062 3.042
Al2O3 14.51 .426 1.627 3.96
TiO2 0.17 .004 .011 .922
Fe2O3 1.86 .033 .126 1.844
Cr2O3 1.10 .021 .080 .922
FeO 3.57 .049 .280 5.91
MgO 32.76 .819 4.692 4.989
NiO 0.28 .003 .017
H2O 13.05 .725 8.308 8.31

Formula: (OH8.3 (Mg,Fe″,Al,Fe′″,Ti,Cr)5.91 [(Al,Fe′″,Ti,Cr) 92SiaO10] Mg/Fe″ = 16.8; Al/Fe′″=13; p = 0.87.

In making the distribution of Al and Fe′″ to the X and Z groups, Ti and Cr have been treated in the same manner, and the amounts allotted thereto are equal in each case. As a result of this grouping the values for X and Z are very close to those required structurally, except that hydroxyl is in slight excess. The value of n is almost 6, thus differentiating this chlorite group of the chlorite family from the leptochlorite group when n = 5 5 (approximately).

The following optical properties have been found for leuchtenbergite:—

– 489 –

Refractive indices:

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

α = β = 1.581 ± 0.002

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

γ = 1.586

[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 = 0.005

Optic axial angle:

Usually 0–10°, but values up to 20° were recorded.

Optically positive and elongation of plates negative.

Dispersion: Distinct, ρ < v, cross-sections of plates giving deep brown tints between crossed nicols.

Colourless.

When these data are plotted on Winchell's diagram (1936, p. 649) the chlorite is found to lie within the field of clinochlore but very close to the field of positive penninites. This is a good instance of the ease with which the composition of chlorites may be accurately ascertained merely by careful determination of the optical constants. The amount of chromium is apparently insufficient in this case to cause the development of the violet or lavender tints so characteristic of kämmererite.* The only effect then of the introduction of chromium will be to cause a slight rise in the refractive index, but apparently the amount is also insufficient to bring about a change of the optic sign to negative; Winchell (1936, p. 646) believes that this is the chief effect of this cation on the optical properties of chlorites.

Origin of the Leuchtenbergite.

Clinochlore is the usual variety of the chlorite family found crystallizing in veins in serpentinized ultrabasic rocks, but the composition of clinochlore is distinct in one particular point from that of the host rock in that the former is relatively aluminous. The intimate and constant association of this mineral with the ultrabasic rocks themselves suggests, however, that it is genetically connected therewith. If, however, the total amount of alumina in these veins were distributed evenly throughout the main mass of serpentinite, the percentage of alumina in the serpentinite itself would not amount to more than a fraction of one per cent.; but since most dunites and ultrabasic rocks contain anything from 0.75–2 per cent. of alumina, it does not seem necessary to have to look to some external source for this alumina as Durrell and Macdonald (1939) and Macdonald (1941) believe. These writers suggest that the alumina necessary for the crystallization of clinochlore must have been introduced from acidic plutonic rocks, although Macdonald is willing to allow that where clinochlore is rare or uncommon, the alumina contained therein could be derived from the serpentinization of monoclinic pyroxenes in lherzolitic rocks. However, the conditions under which Macdonald finds chlorites crystallizing in serpentinites are rather different from those at Gordon's Quarry, since there is evidence that other highly aluminous minerals such as spinel and tourmaline are present in the former locality.

[Footnote] * Two kämmererites from Yugoslavia (Tucàn, 1924) for example, were found to contain 10.19 and 11.72 per cent. of Cr2O3; both minerals showed strong violet tints.

– 490 –

It is believed therefore that the leuchtenbergite veins have crystallized at a very late stage from the deuteric solutions that brought about the serpentinization of the peridotite, and that ample alumina would be available following the breakdown of the pyroxenes and minor plagioclase in the ultrabasic rocks themselves. In any case, it is not possible to appeal to post-peridotite acid intrusives to supply alumina, since these are not known in the Kaukapakapa area (Bartrum, 1920). In conclusion, it should be noted that Turner (1933, p. 273) attributes the crystallization of clinochlore in serpentinites in Red Mountain, Western Otago, to deposition at a late stage directly from magmatic waters.

Literature Cited.

Barth, T. W. F., 1931. Pyroxen von Hiva Oa, Marquesas-Iseln und dic Formel titanhaltiger Augite, Neues Jahrb. fur Min., Abt. A, Beil.-Bd. 64, pp. 217224.

Bartrum, J. A., 1920. The Conglomerate at Albany, Lucas Creek, Waitemata Harbour. Trans. N.Z. Inst., vol. 52, pp. 422430.

Berman, Harry, 1937. Constitution and Classification of Natural Silicates Amer. Mineral., vol. 22, no. 5, pp. 342408.

Bragg, W. L., 1937. Atomic Structure of Minerals, Oxford University Press, London.

Chapman, R. W., and Williams, C. R., 1935. Evolution of the White Mountain Magma Series. Amer. Mineral., vol. 20, no. 7, pp. 502530.

Clarke, F. W., and Schneider, E. A., 1890. Experiments upon the Constitution of the Natural Silicates. Amer. Jour. Sci., vol. 40, no. 239, pp. 405415.

Deer W. A., and Wager, L. R., 1938. Two New Pyroxenes included in the System Clinoenstatite, Clinoferrosilite, Diopside, and Hedenbergite. Mineral Mag., vol. 25, no. 160, pp. 1522.

Dixon, B. E., and Kennedy, W. Q. Optically Uniaxial Titanaugite from Aberdeenshire. Zeits. Krist., vol. 86, pp. 112120.

Durrell, C., and Macdonald, G. A., 1939. Chlorite Veins in Serpentine nean Kings River, California. Amer. Mineral., vol. 24, no. 7, pp. 452456.

Engelhardt, W. Von, 1936. Die Geochemie des Bariums. Chemic der Erde, vol. 10, pp. 187246.

Goldschmidt, V. M., 1937. The Principles of Distribution of Chemical Elements in. Minerals and Rocks. J.C.S., pp. 655–673.

Hendricks, S. B., 1939. Polymorphism of the Micas, with Optical Measurements by M. E. Jefferson. Amer. Mineral., vol. 24, no. 12, pt. 1, pp. 729–771.

Holmquist, P. J., 1921. Typen und Nomenklatur der Adergesteine. Geol. Foren. i Stockholm Forhand., vol. 43, pp. 612–631.

Hurlbut J. R., C. S., 1935. Dark Inclusions in a Tonalite of Southern California. Amer. Mineral., vol. 20, no. 9, pp. 609–630.

Hutton, C. O., 1936. Mineralogical Notes from the University of Otago. Trans. Roy. Soc. N.Z., vol. 66, pp. 3537.

—— and Seelye, F. T., 1945. Contributions to the Mineralogy of New Zealand—Pt. 1. Trans. Roy. Soc. N.Z., vol. 75, pt. 2, pp. 160168.

Jakob, J., 1937. Uber das Auftreten von Dreiwertigen Titan in Biotiten. Schweiz. Min. Petr. Mitt., vol. 17, pp. 149153.

Macdonald, G. A., 1941. Geology of Western Sierra Nevada, between the Kings and San Joaquin Rivers, California. University of California Publications. Bull. Dept. of Geol. Sciences, vol. 26, no. 2, pp. 215286.

Mauguin, C., 1928. Etude des Micas (non-fluorés) au Moyen des Rayons X Compt. Rend. Acad Sci. Paris, vol. 186, pp. 879–881.

– 491 –

Mauguin and Graber, L., 1928. Étude des Mieas Fluorés au Moyen des Rayons X. Compt. Rend. Acad. Sci. Paris, vol. 186, pp. 1131–1133.

Noll, W., 1934. Geochemie des Strontiums mit Bemerkungen zur Geochemie des Bariums. Chemie der Erde, vol. 8, pp. 507–620.

Neuberger, M. C., 1936. Gitterkonstanten für das Jahr 1936. Zeits. Krist., vol. 93, pp. 136.

Orcel, M. J., 1925. Sur Deux Clinochlores Chromifères du Togo. Compt. Rend. Acad. Sci. Paris., vol. 180, pp. 836–838; analysis on p. 837.

—— 1927. Recherches sur la Composition Chimique des Chlorites. Bull. Soc. Franç. Mineral., vol. 50, pp. 75456.

Palache, C., Berman, H., and Frondel, C., 1946. The System of Mineralogy of Dana, vol. 1, second printing. John Wiley and Sons, Inc., New York.

Pauling, L., 1930. The Structure of Micas and Related Minerals. Proc. Nat. Acad., Sci., vol. 16, pp. 123129.

Satô, S., 1933. Alteration of Talc and Antigorite to Leuchtenbergite in the Metamorphosed Dolomite of the Mateurei System, North Korea. Jouru. Shanghai Sci. Inst., sect. 2, vol. 1, pp. 1724.

Shand, S. J., 1943. Eruptive Rocks, 2nd Ed. John Wiley and Sons, New York.

Stevens, R. E., and Schaller, W. T., 1942. The Rare Alkalies in Micas. Amer. Mineral., vol. 27, pp. 525537.

Tajder, M., 1938. Klinochlor von Dobro polje. Bull. Inst. Geol. Yougoslavie, vol. 6, pp. 235238.

Tsuboi, S., 1935. Petrological Notes (1)–(10). Jap. Jour. Geol. Geogr., vol. 12, nos. 3–4, pp. 109 (1)–113 (5). Anal. on p. 112 (4).

Tucan. F., 1924. La Kämmererite des Chromites de Yougoslavie. Compt. Rend. Acad. Sci. Paris, vol. 178, pp. 1911–1913.

Turner, F. J., 1933. The Metamorphic and Intrusive Rocks of Southern Westland. Trans. N.Z. Inst., vol. 63, pp. 178284.

—— 1939. Hornblende-Gneisses, Marbles and Associated Rocks from Doubtful Sound, Fiordland, New Zealand. Trans. Roy. Soc. N.Z., vol. 68, pt. 4, pp. 570598.

Turner, H. W., 1899. Some Rock-forming Biotites and Amphiboles. Amer. Jour. Sci., 4th Ser., vol. 7, no. 40, pp. 294298.

Wager, L. R., and Mitchell, R. L., 1943. Preliminary Observations on the Distribution of Trace Elements in the Rocks of the Skaergaard Intrusion, Greenland, Mineral Mag., vol. 26, no. 180, pp. 283296.

Weyberg, D. Z., 1909. Materialien zur Kenntnis der Chemischen Zusammen-setzung der gesteinsbildenden Glimmer. Warschauer Universitats-Nachrichten, 1909.

Zachariasen, W. H., 1931. A Set of Empirical Crystal radii for ions with Inert Gas Configuration. Zeits. Krist., vol. 80, pp. 137153.