PETROLOGICAL STUDIES
IN GLEN URQUHART,
INVERNESS-SHIRE

G. H. FRANCIS

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

MINERALOGY Vol. 1 No. 5
~LONDON : 1958

Per ROLOGICAL STUDIES IN GLEN URQUHART,
INVERNESS-SHIRE

BY

G. H, FRANCIS

Pp. 121-162 ; Plates 8-10; 10 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
MINERALOGY Vols 2 Ne. $
LONDON : 1958

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted in 1949, is to be
issued in five series, corresponding to the Departments
of the Museum.

Parts will appear at irregular intervals as they be-
come ready. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar ‘year.

This paper is Vol. 1 No. 5 of the Mineralogical
series.

PRINTED BY ORDER OF THE TRUSTEES OF
THE BRITISH MUSEUM

Issued February, 1958 Price Fifteen Shillings

Pe PROLOGICAL STUDIES IN GLEN URQUHART,
INVERNESS-SHIRE

I. LIMESTONES Il. SKARNS

INTRODUCTION

THE area in which lie the rocks to be described in this Bulletin is partly in Glen
Urquhart, partly on the moors immediately north of the glen. In the valley sides it
extends from Polmaily House to Drumnadrochit, and on the moors between Loch
Gorm and Garbeg Farm (Map, Pl. 10 at the end of this Bulletin). The majority of
the ground is underlain by metasedimentary rock comprising (in structural succession)
a bed of siliceous magnesian limestone of which the uppermost 4o ft. are exposed,
an argillaceous bed of perhaps the same thickness and a much thicker succession of
arenaceous beds having no recognized top. This series contains basaltic rock in
streaks and lenses, which by every gradation reach the scale of massive sills 15 ft. thick.

After the emplacement of the basaltic rock the whole assemblage suffered
isoclinal folding and metamorphism under kyanite zone (low amphibolite facies)
conditions. Shortly afterwards an ultrabasic mass intruded the metasediments,
truncating or disturbing their fold structures (Francis, 1956a). Finally, meta-
somatism affected the sediments and the ultrabasic intrusive. Within the sediments
this metasomatism finds its acme in sporadic areas of schists altered to the chemical
composition of granite. The isoclinally folded bedding-plane foliation in the schists
is traceable right through these “ bosses” of granitized rock, and no intrusive
granite is exposed. In and around the “ bosses ”’ the chief material added appears
to be potassium, expressed in newly-formed microcline in the schists and in quartz-
microcline-muscovite—biotite pegmatites. In areas further from the granite—
gneiss “ bosses’ the metasomatism involves sodium rather than potassium. The
uninjected arenaceous rocks are rich in soda and this phase is probably nothing
but a redistribution of the soda of these rocks. Oligoclase-bearing pegmatites free
from microcline, and oligoclase-porphyroblast schists characterize this zone.
In sediments yet further from the bosses the only representatives of the injection
are quartz veins. These have apparently been able freely to absorb wallrock
materials. In traversing the argillaceous rock (kyanite schist) they have absorbed
alumina (in the form of kyanite), and in traversing the limestone they have absorbed
lime (in the form of calcium silicates). At the sedimentary contacts between
kyanite schist and limestone, at the borders of the ultrabasic mass and within it
these lime-silica-alumina-charged veins have deposited skarns. Hydrothermal
bodies of both discrete and gradational character within the ultrabasic mass, equiva-
lent to the skarns in the sediments, have already been described (Francis, 1955,

MIN. I, 5. R

124 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

1956a). Part I of this Bulletin describes the limestones, Part II the various lime-
silica-alumina skarns. The description of the basaltic rocks, now represented by
amphibolites, and of the argillaceous and arenaceous rocks with their alkali meta-
somatism, and a consideration of structural and age problems of the area will be
left to further publications.

PART I

LIMESTONES
Previous Work

The earliest account of the geology of Glen Urquhart to be traced is contained
in The Statistical Account of Inverness-shire (1842) in which the parishes of Urquhart
and Glen Moriston were described by the Rev. Jas. Doune Smith (who wrote his
contribution in the year 1835). This careful summary records all of the chief
rock groups of the glen: “stratified gneiss, serpentine’’, and “ grey and white
primitive granular limestone ’’, together with some of their minerals.

The next account of the geology of the district which has been discovered is con-
tained in M. Forster Heddle’s papers: “‘ Chapters on the Mineralogy of Scotland ”
(1878). In these papers descriptions and analyses are furnished of andesine,
tremolite, and edenite in limestone and of other minerals in the remaining rock
groups. Heddle adds some comments on the structure of the area :

“The serpentine of Polmally is on the north-east wrapped round with an
unusually plicated bed of granular limestone, so convoluted, and fractured, and
cut up by dykes, that it is difficult to determine whether there be not a greater
number of beds than one ’’.

And

“The lime has been quarried here and there where anticlines brought it to the
surface.”

The latter passage agrees with the findings of E. H. Cunningham Craig (1914 : 20),
and the present writer, that the limestone occupies the bottom of the structural
succession in the area. Heddle (1878 : 310) gives a list of minerals found in the
limestone as follows :

tremolite — common pyrrhotite -— common
andesine -— rare sphene — very common
urquhartite apatite -— rare
edenite — common

No reference to urquhartite has been traced in the records or catalogues of this
Museum or elsewhere. Wallace (1886) records from Glen Urquhart the minerals
listed by Heddle, and in addition chondrodite. No confirmation of the existence
of this mineral in the glen was obtained by Heddle (1879), or by the present writer.
The original specimen of this mineral (Greg & Lettsom, 1858 : 223) ‘‘ from Loch
Ness ’’, is still preserved in this Museum. An X-ray powder photograph confirms

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 125

its identification and does not support the contention that it it staurolite var.
“ xantholite ’’’, as suggested by Heddle (1879) and by Read & Double (1935).1

The investigation of Glen Urquhart by the Geological Survey in the first decade
of this century is recorded over the initials of E. H. Cunningham Craig in The Geology
of the Country Round Beauly and Inverness (Memoir No. 83, 1914). Cunningham
Craig’s primary conclusion, that an inlier of the Lewisian Gneiss within the Moine
Series is exposed in Glen Urquhart, is untenable. This will be demonstrated in
the paper on amphibolites and particularly in the study to be published later on the
folded structure of the sediments and their subsequent alkali metasomatism. All
the crystalline rocks of the glen are thought by the writer to be members of the Moine
Assemblage.

The bulk of Cunningham Craig’s field observations are however valid. They
have been verified fully onthe ground. Through the kindness of Dr. T. H. Whitehead
and the Staff of the Geological Survey at Edinburgh the 6 in. to 1 mile manuscript
map of the area (from which a part of the one-inch Sheet 83 is derived) was examined
and copied. The writer’s subsequent mapping (Plate 10) is little different from
the Geological Survey map in respect of the limestone outcrops.

Finally, the limestones have been described in H.M. Geological Survey Special
Reports on Mineral Resources in Great Britain: vol. XXXV, The Limestones of
Scotland (T. Robertson et al., 1949), and vol. XXXVII, idem, Chemical Analyses
and Petrography (A. Muir et al., 1956). This contains a petrographic description
of a slide of Glen Urquhart limestone, together with a chemical analysis and
spectrographic estimation of trace elements in this rock (quoted below).

Field Occurrence
Cunningham Craig (1914) describes the bed as follows :

“The limestone, of which a thickness of 30 or 40 ft. is exposed, is a highly
crystalline white marble, with a considerable development of lime-silicates in
some of the beds. The purest beds have been quarried and burnt for lime and
are still occasionally made use of.”

At its largest development in the area of the quarries just north of Upper Gartally
Farm the lithological variations within the limestone can clearly be seen. Weathering
white or grey the purest bands are always distinguished by a speckling of pale lamellae
of phlogopite. Towards the western end of the largest quarry are striped beds
which plunge beneath the purer beds at the east. These are banded marbles
which break along the mineral banding (which appears to parallel the bedding
planes), to form flat, flagstone-like boulders. The bands are alternately pale,
relatively pure phlogophite marbles and dark, randomly-crystallized actinolite-
diopside rocks, the whole being perhaps a case of metamorphic differentiation.
In the northernmost quarry of this group relatively pure marbles carry thin veneers
of pyrite on bedding planes.

1 Doubt has been cast on the status of Heddle’s “‘ xantholite ’’ (Francis, 1955) but A. Juurinen (1956)

has since shown by chemical analyses of the type material from the Royal Scottish Museum, Edinburgh
that the mineral is a true staurolite somewhat rich in MgO, CaO, and H,O.

126 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

In exposures further north (e.g. on Torr Buidhe) light green crystals belonging
to the epidote group are prominent, whilst strings and blebs of quartz, elongated
with the bedding, are to be seen in all exposures. Actinolitic bands, like those of
the Gartally quarries, are found and, like those, they show a haphazard arrangement
of lustrous, bottle-green amphibole prisms 1 to 2cm. in length. On exposed hilltops
on Torr Buidhe and west of Loch Maolachain weathering has formed “ karst
structures’ or “ grikes’’. “‘ Limestone” has occasionally been mapped in the
present survey in a few areas in which sedimentary impurity or subsequent meta-
somatic activity have so diminished its lime content as to prohibit the crystallization
of abundant calcite. Thus in the sedimentary inclusion within the serpentinite
north-east of Beinn a’Ghairchin a green-white, banded actinolite—diopside—clino-
zoisite—plagioclase rock is mapped with the limestone. The status of some such
rocks must be regarded as uncertain, whilst others may have arisen by the action
of lime metasomatism on a pelitic host.

Mineralogy

Calcite. This mineral occurs in all quantities from the main constituent of a rather
pure marble, through a stage in which it is the groundmass to amphibole, pyroxene,
and mica crystals, to mere interstitial traces. When abundant it forms a simple
mosaic, with polyhedral margins (PI. 8, fig. 1). Crenulate margins are infrequent.
The common development of lath-like (0112) twin lamellae as the long diagonal
of the rhombs defined by the (roI1) cleavages is to be found abundantly in slides of
these marbles. The twin lamellae are usually narrow, suggesting that their formation
(as glide planes) was not induced by very great deformation. Their orientation, as
shown by Tilley (1920), provides a check on the nature of the rhombohedral carbonate.
No glide twins forming the short diagonal of the cleavage rhombohedron (e.g.
(0221)) have been observed, suggesting that dolomite is absent. This inference
is supported by the metamorphic facies of the rocks, the place of dolomite being
taken by actinolite and diopside. The calcite holds minute flecks of graphite,
blebs of quartz and feldspar, and lamellae of phlogopite, even in the purest marbles.
Calcite quite frequently occurs in symplectitic intergrowths with other minerals,
chiefly plagioclase and zoisite.

Quartz is almost universally present in the limestone either in granular inter-
stitial growth or as strings and blebs within and between the calcite crystals elongated
with the bedding, some of it is probably of later introduction into the limestones.

Feldspars. The feldspar of the limestones is, in the great majority of cases, a
member of the plagioclase series. It ranges in composition from a rather pure albite
to andesine (Ans,) according to measurements on extinction in twinned crystals
sectioned normal to a. Crystals up to I mm. diameter occur, but they are
commonly smaller. Albite and pericline twins predominate. Almost all the plagio-
clase of the limestone is decomposed. Sometimes large crystals of the epidote
group replace the bulk of the feldspars, at other times the alteration has formed a
fine-grained saussurite of the same minerals. A fine dust of other minerals such as
quartz, apatite, amphibole, and phlogopite is commonly held as inclusions.
Plagioclase and zoisite form symplectitic intergrowths.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 127

Microcline has been recognized in a few alkali-injected marbles close to granite
pegmatites. The characteristic grid twinning and a yellow stain with potassium
cobaltinitrite after etching with HF confirm this identification.

Zoisite. This mineral is common, particularly in association with plagioclase.
Both the alpha and the beta forms occur, and their optics always agree with those
quoted in the standard texts (e.g. Winchell, 1951 : 446-447). One remarkable crystal
possesses a core of a-zoisite with the optic plane parallel to (010), and 2Vy = 56°.
Dispersion, r>v, is strong, and there is anomalous blue birefringence. An incomplete
rim is developed in which the optic axial plane is normal to (010) and 2Vy = 7°
(both angles measured on the universal stage in white light). The dispersion in
the rim is r<v, distinct, and the birefringence produces greys of the first order
(Text-fig. 1).

o ae Os

Fic. 1. Zoisite crystal containing the « and B forms, from the limestone quarry
north of Beinn a’ Ghairchin.

These groundmass zoisites may increase in amount to become the most abundant
mineral in the rock. Such specimens may be found in the quarries between the
buildings of Upper Gartally Farm. Occasionally prismatic, the zoisites are commonly
xenoblastic and ragged in their outlines towards the quartz and calcite which accom-
pany them. They grow to several centimetres in length and millimetres in width.
Birefringence is low, being predominantly in greys, whilst vivid, anomalous “ ultra-
blues” occur patchily within the crystals. Cleavage parallel to (o10) is well
developed. The crystals are commonly sieved with inclusions of quartz and
amphiboles. Veins of pink f-zoisite in injected limestone probably belong to the
variety thulite.

Some of the zoisite-rich rocks (e.g. around Upper Gartally) may represent impure
limestones, but a further increase in zoisite in the same area leads to massive,
cavernous, monomineralic zoisite rocks which cannot be regarded as metamorphosed
sediments. They must be grouped with the calc-silicate skarns to be described
in Part II of this work.

Clinozoisite and epidote. These minerals, though less abundant than the zoisite,
are quite common in the limestones. They may occur in separate rocks, or
together in the same rock. In the latter case they are usually found in crystals

128 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

zoned from clinozoisite cores to epidote rims. Beautiful examples of zoning may
be found, picked out between crossed nicols by variations in their bright second-order
interference colours. Compositions of these minerals are most readily determined
in sections cut normal to the } axis of individuals twinned on (100), in which the
diagnostic extinction angle a /\c can be measured to the twin plane. No
#-clinozoisite (Johnston, 1949) has been found.

Text-fig. 2 shows a clinozoisite-epidote individual from an impure limestone
band. It is truncated by a larger clinozoisite-epidote. Johnston (1949) and
Winchell (1951) show that clinozoisite has its a-vibration direction in the obtuse
angle £, whilst epidote has its a-vibration direction in the acute angle (supplementary
tof). Thus the negative extinctions in Text-fig. 2 are in «-clinozoisite, the positive
extinctions in epidote. Further, oscillatory zoning is present. There is a narrow
core of clinozoisite, a wider irregular series of epidote zones and a final wide clino-
zoisite rim. This is the only measured crystal in which a clinozoisite rim has been
found.

Fic. 2. Twinned clinozoisite-epidote crystal, from impure limestone in the block of
schists within the serpentinite 300 m. west of Loch Maolachain. The extinction
angles of the numbered zones are: I, a \ c —2°; 2,a Ac; + 4°; 3,aA ¢ —3°;
4,0 6 —1°%s 5 aN 6 —2°5 6, a \ eo -3°3 7, a Nc 6°: 8a Aso -- to
9,a A ¢ — 2°.

Very similar clinozoisite-epidote crystals are prominent in the skarns to be
described in Part II. Clinozoisite~epidotes in the limestones vary from saussuritic
material up to crystals 1 to 2 mm. in length. They are almost always in close
association with plagioclase.

Amphboles. The amphiboles of the limestones apparently belong to three
transitional types. Actinolite, tremolite, and “ edenite”’. The same three types
recur in the skarns. Heddle (1901, Vol. II) gives three amphibole analyses from
Glen Urquhart, and three more have been carried out for this study. Of these, the
three new analyses and perhaps two of Heddle’s belong to the skarns (he did not
differentiate skarns from limestones in his study of the Gartally localities). Heddle’s
remaining analysis is of a high-silica tremolite (Text-fig. 8) ; it has indeed, more
silica than any in Hallimond’s survey of the calciferous amphiboles (1943), containing

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 129

in fact Si slightly in excess of the 8 atoms required for pure tremolite and lacking
Al in the Z group; the possibility of some impurity or analytical inaccuracy or
both cannot be discounted in an early analysis such as this one. Nevertheless, it is
clearly one of the high-silica calciferous amphiboles and plots in Hallimond’s diagram
along with the amphiboles of metamorphic limestones. The remaining five amphi-
boles from Glen Urquhart plot in the field of “ schists”’ in Hallimond’s diagram,
in which group we may (for this purpose) allow skarns a place (Text-fig. 8).

In the limestones pale to bottle-green prismatic amphiboles of late “ spearing ”’
growth have been distinguished as actinolite, almost colourless prisms as tremolite
and occasional fibrous nests of green, fasciculitic or plumose amphibole as hornblende.
The use of the word “ edenite ’’ for these amphiboles is discussed in Part II. These
are field terms, supported only by the optical data given in Table I. Chemical
analyses of the amphiboles of the limestones proper at Glen Urquhart are still
required.

TABLE I.—Optics of Amphiboles from Limestone

Actinolite Tremolite Hornblende

Pleochroism :

4 : : ; ; : colourless : colourless - colourless

‘ci RR : ‘ , . grassgreen . pale straw 5

Y ; - : : : jade green . pale grass green . pale straw
Extinction :

(uN 6)) : ; : 16°—18° : 18°—20° , 21°-23°
Optic Axial Angle :

(2Vq) : : : . 81° average . 83° average . 84° average
Refr. Index: 8B

in sodium light (+ -002) . 1° 634 : I*623 : I*623

Pyroxenes. The limestone pyroxenes are closely similar to those of the skarns,
which, as shown in Part II, are diopside and salite. Salite predominates in the
limestone. The size may vary widely from small idioblasts of 0-2 0-7 mm. to
large equant crystals 4 mm. across. There is a strong tendency to form crystal
outlines especially towards calcite, although irregular growth may occur.

Micas. Pale phlogopite crystals are common in the limestone. They are usually
small, averaging (0-5 mm. diameter), but may form large plates up to 3 mm. across.
Pleochroic haloes about sphene and zircon grains have been found. The pleochroism
scheme is y and # pale brown, « colourless. The interference figure is normally
uniaxial, sometimes a small optic axial angle may be observed. Cleavage flakes from
parageneses 8, 9 and 11 (Table II) examined in sodium light, gave a uniform refrac-
tive index of 1-587 (+ -002) indicating phlogopite with Mg” : Fe” : go : 10 (Winchell,
IQ51).

Muscovite flakes occur in limestones affected by alkali injection from nearby
granite pegmatites.

Accessories. Sphene is almost universal in the limestones. Graphite is a common

130 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

accessory, and pyrite (with some pyrrhotite) is a frequent associate. Apatite.
zircon, and magnetite also occur.

Petrology. Parageneses recorded in the Glen Urquhart limestones are given in
Table II, below. Limestones which have been affected by injection from neigh-
bouring granite pegmatites are listed separately.

TABLE II.—Limestone Parageneses

ct-qtz-zo-trem-di
ct-qtz-zo-plag(An,,)-act-di
ct-qtz-zo-plag(An,,)-act-di-phlog
ct-qtz-zo-plag-hb
ct-qtz-zo-plag(An,,)-phlog
ct-qtz-zo-phlog

ct-qtz-clz/ep-act

ct-qtz-clz-plag (An,,)-act-di-phlog
ct-qtz-clz-di-phlog

Io ct-qtz-ep-plag (and)-act- phlog

Ir _ ct-qtz-plag (and-hb- phlog

12 ct-qtz-plag(An,,_,)-act-di

13 ct-qtz-plag-act-di-phlog

14 ct-qtz-plag(An;) phlog

15 ct-zo-plag(olig.)-hb-di

16 ct-clz/ep-plag(An,,)-act-phlog

17 ct-hb-phlog

18 qtz-zo-trem

Ig qtz-zo-trem-di

ON AUN PWN H

Ne)

Injected limestone Pavageneses
20. ct-qtz-zo-plag(olig)-mu-phlog
21 ct-qtz-ep-plag(olig)-mu-phlog
22 ct-phlog-qtz-(chl)
23 ct-qtz-mic-mu-phlog-(chl-tourm)
Accessovies ; sphene, graphite, pyrite, pyrrhotite, apatite, zircon, magnetite.
Abbreviations: act, actinolite; and, andesine; an, anorthite; chl, chlorite; clz, clinozoisite;
ct. calcite; di, diopside; ep, epidote; hb, honblende; mic, microline; mu, muscovite; olig, oligo-

clase; phlog, phlogopite; plag, palgioclase; qtz, quartz; tourm, tourmaline; trem, tremolite ;
ZO, zoisite.

Photomicrographs. Four thin sections from the limestones, illustrating parageneses
II, 14, 18, and 19, appear in Pl. 8, with annexed descriptions.

Chemical Composition

An analysis with spectrographic determination of the trace elements (see Table
III), and a petrographical description of a Glen Urquhart limestone specimen are
given in Muir e¢ al. (1956). The analysis and description make it clear that it is
a rock rich in calcite and poor in calc-silicates. This relatively pure limestone is
naturally the rock of interest in an economic study of this kind. The bulk of the
Glen Urquhart limestones are, however, less pure, containing phlogopite “epidote’’?

1 By “ epidote ’’ is here implied a member of the group: zoisite, clinozoisite, and epidote.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 131

diopside and tremolite in some quantity. The two latter minerals, with high mag-
nesia contents, must raise the Mg/Ca ratio of the impure beds above that of the
analysed limestone, and this suggests that the original formation was partly dolomitic.
Alumina, which must also be higher in the impure beds, suggests argillaceous material,
and the sediment, as a whole, was therefore in all probability a somewhat muddy,
siliceous, dolomitic limestone.

TABLE III.—Limestone, Glen Urquhart
(Containing phlogopite and calc-silicates, [Muir et al., 1956))

I 2
SiO, é 9°49 Ga *
Al,O, 3 I-40 Cr 5
FeO P o- 82 Vv =
FeO — Su *
MgO 2°23 Li 3
CaO 47° 86 Ni *
Na,O 0°24 Co *
K,O 0-36 Zr 50
H,O-105° 0°07 AY, *
TiO, 0:09 La *
P.O; 0°03 Sr 300
MnO 0-01 Pb *
CO, 37°36 Ba 20
FeS, 0-08 Rb *

100: 04

sp Gry 2°73

1 Percentage weight of oxides. Analysts, A. Muir and H. G. Hardie.

2 Trace elements, parts per million by weight. Spectroscopist, R. L. Mitchell.

* Present, but below the limit for quantitative determination of this element (see Mitchell 7n Muir
et al., 1956).

Metamorphic facies. The above parageneses suggest crystallization within the
lower part of the amphibolite facies. Ramberg has shown the importance of the
association of plagioclase with the epidote minerals from the viewpoint of the
facies classification (e.g. 1952 : 50-54). He defines the lower limit of the amphibolite
facies by the association ‘“ epidote’’—Angzy in most rocks. Higher anorthite
contents in association with epidote minerals occur at successively higher P-T
conditions (grades).

In the Glen Urquhart marbles andesine (to Ang.) is commonly found in association
with an epidote mineral in rocks showing no evidence of disequilibrium at the time
the pair crystallized. This P-T-sensitive association may possibly be disturbed
in calcite-rich environments by reactions tending to “leach” anorthite out of
plagioclase, even at high temperature, e.g.:

CaCO, + 3CaAl,Si,0, + H,O = 2Ca,Al,Si,0,,0H + CO, . : eet)
anorthite epidote
(Ramberg, 1952 : 150).

132 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

One of the calcite-rich rocks (paragenesis 14) in this study contains an anorthite-
poor plagioclase (An,;), but this does not give evidence of anorthite “ leaching ”
since an epidote mineral is lacking. In the absence of “ epidote”’ the plagioclase
composition at a given grade may vary from the composition crystallizing with an
epidote mineral to pure albite ; the anorthite content of the plagioclase here depends
solely on the original composition of the rock. It seems, indeed, as if this “‘ leaching ”’
of the anorthite content of plagioclases has not occurred in the Glen Urquhart
marbles since plagioclase in association with “‘epidote’’ is not restricted to compositions
close to albite. The association andesine-“ epidote’’, therefore is not ambiguous,
and suggests the lower part of the amphibolite facies for the crystallization conditions
of these rocks.

Ramberg (1952: 150) also states that “ another indication of the amphibolite =
epidote amphibolite facies transition is the lowermost stability border of diopside
in calcite milieu.

“ The following reaction :

epidote amphibolite facies amphibolite facies
2siO, + 3CaCO, + Ca,Mg;Si,0,.(OH), = 5CaMgSi,O, = 3CO, = H,O . a (2)
tremolite diopside

referring to the pure Mg-silicate end-members of the mixed crystals, may be the
proper indicator of the lowermost border line of the amphibolite facies ”’.

Weeks in his recent study of the metamorphism of siliceous carbonate rocks
(1956) has shown that diopside is the third mineral to appear in progressive
“equilibrium ’’ metamorphism of a siliceous dolomite (talc and tremolite are the
first and second, respectively). The temperatures at varying pressures for the cal-
culated curve of the reaction :

CaMg(CO), + 2Si0, = CaMgSi,O, + 2CO, : E : », 13)
dolomite diopside

lie within the lower part of the amphibolite facies (Weeks, 1956, fig. 4, and p. 265,
footnote).

Weeks has used reaction (3) rather than reaction (2) (of the present paper) since
thermochemical data on tremolite were not published at the time he wrote his paper.
It is probable that diopside crystallized by a process of the type of reaction (2)
at Glen Urquhart as dolomite will already have been eliminated in the production
of tremolite in rising metamorphism. There is no textural evidence for reaction (2)
running from left to right. Much of the amphibole in the marbles is in late
“spearing’’ prisms or needles, cutting right through diopside crystals. Such
amphiboles cannot be “ parental’’ to the diopsides. They probably indicate a
retrograde displacement of reaction (2) towards the left. “‘ Parental’’ tremolite
may exist in the less common xenoblastic grains associated with diopside, but clear
textual evidence has not been found.

In some fields diopside and tremolite minerals appear to have crystallized at
equilibrium. Bowen (1940) has shown that in progressive metamorphism of siliceous

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 133

dolomites there is a stage at which this pair may be expected to co-exist at
equilibrium.

To summarize, it seems clear that the Glen Urquhart limestones have crystallized
just within the lower grade limits of the amphibolite facies. The adjacent isograde
kyanite schists have been assigned, partly with reference to diopside in these marbles,
partly on independent evidence, to a similar position in the amphibolite facies
(Francis, 19560).

Low amphbolite facies associations in marbles. Turner (1948) and Ramberg
(1952) have considered problems bearing on calcite—biotite (phlogopite) relationships
in the amphibolite facies. Turner states (p. 84) that the “ occurrence of potash
feldspar as a minor constituent of calc-schists and marbles ... is very general”’
and “‘ the appearance of microcline in an environment of calcite and calc-silicates is
due in the main to the chemical composition of the rock rather than to its physical
properties, for, in all but highly magnesian rocks, potash feldspar will always crystal-
lize in preference to mica if the ratio

CaO + K,O + Na,O
Al,O,
exceeds unity’. Ramberg, on the other hand, remarks (p. 152) that “‘ it is common
to find biotite or, rather, phlogopite as apparently stable mineral grains in amphi-
bolite facies marbles”. He concludes by means of an ACF! + silica tetrahedron
that the association calcite-phlogopite is only stabilized in the absence of excess
silica (quartz) according to the reaction :

6Si0, + 3CaCO, + KMg;AlSi,0,)(0H), = KAISi,0, + 3CaMgSi,O, + 3CO, . (4)
phlogopite potash diopside
feldspar

He comments that “ the placing of calcite or dolomite in ACF diagrams easily causes
misunderstanding and confusion. The equilibrium relationships between the
carbonates and the several principal silicates need at least a tetrahedron diagram
to be visualized’’. This is true but, as will be seen, the best tetrahedron is not
bounded by A, C, F, silica, but by A, C, F, K.

Turner has foreshadowed this by suggesting (1948 : 81-82) that the ACF diagram
for the amphibolite facies should be interpreted in conjunction with the AKF
diagram. To take up this suggestion the ACFK tetrahedron is necessary, but
there is at first a small alteration to be made. “‘A” of the ACF diagram is not
the same as “A” of the AKF diagram (Turner, 1948). For the purpose of the
ACFK tetrahedron “‘ A ” should be calculated as molecular Al,O, in excess of molecu-
lar Na,O + K,O only. CaO is excluded from the calculation, and the resulting
“A’”’ is similar to that of the ACF diagram.

As we have seen, Turner and Ramberg both imply that marbles will hold either
microcline or biotite (phlogopite) at equilibrium but not both together. Turner
believes that the choice is determined by whether or not the rock is highly magnesian,
and that microcline is the commoner mineral. Ramberg believes that the choice

1 ACF and AKF diagrams, introduced by Eskola, are defined in Turner (1948: 57 and 82).

134 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

is determined by the presence or absence of excess silica (quartz) and that phlogopite
is the commoner mineral.

A study of the ACFK tetrahedron (which is constructed for rocks with excess
silica provides an alternative answer to the problem and shows, incidentally,
that the associations calcite—microcline-phlogopite and calcite-quartz—phlogopite
in marbles and skarns are possible at equilibrium in the lowest part of the
amphibolite facies (see also Table II for observed cases of these associations).

The ACFK tetrahedron can first be visualized in terms of its faces. The ACF
and AKF faces are given, e.g., by Turner (1948, figs. 19, 20)! the remaining two,
the ACK and the FCK are shown in Text-figs. 3 and 4. The ACFK tetrahedron
is shown in Text-fig. 5. In this diagram the field of limestones and skarns, with
the appropriate minerals is picked out.

K K

microcline microcline

phlogopite

ae on ——* =a
A anorthite epidote G C diopside actinolite fF
Fi. 3. Fic. 4.

Fic. 3. The ACK triangle for rocks with excess silica in the
lower part of the amphibolite facies.

Fic. 4. The FCK triangle for rocks with excess silica in the
lower part of the amphibolite facies.

In these diagrams there are two pairs of minerals each of which may be pictured
as a “‘single”’ phase for the purpose of the diagrams, provided their true status
is understood. The pairs are plagioclase and “ epidote’’, diopside and actinolite.
These typical members of impure limestone parageneses involve two components
in excess of the four plottable components of the ACFK tetrahedron, namely Na,O
and H,O. In effect plagioclase, “ epidote ”’ and actinolite all lie outside the ACFK
plot, in the direction of the Na,O or H,O components. From phase rule consider-
ations we may thus have two extra phases (minerals) at equilibrium in the system,

1 Turner (1948: 79, and figs. 18 and 19) has suggested that in ACF diagrams for potash-deficient
rocks in his cordierite-anthophyllite and staurolite-kyanite subfacies the limestone field should be left
as a vacant field, “ since the total potash in all rocks sufficiently rich in lime and poor in alumina to fall
within this area is held in potash feldspar ’’. The theoretical and observed parageneses discussed in this

paper show, however that assemblages free from potash feldspar of this metamorphic grade exist in the
limestone field, which is not, therefore, vacant.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 135

i.e. six plus quartz instead of four plus quartz (cf. parageneses of Table II). The
relationships between plagioclase and “‘ epidote ’’ and between diopside and actino-
lite have been considered above.1_ Muscovite and biotite are also, of course, pro-
jections into the ACFK plot from a system containing water, but they have no
corresponding anhydrous minerals to pair with as have “ epidote’’ (anorthite)
and actinolite (diopside).

microcline

A anorthite epidote calcite
—S—

Fic. 5. The ACFK tetrahedron for rocks with excess silica in
the lower part of the amphibolite facies.

The field of limestones and skarns in the ACF diagram involves the three phases :
ct-““an’”’-‘“‘di’”’. The same field in the ACK diagram (Text-fig. 3) and the FCK
diagram (Text-fig 4) is in each case a quadrilateral, which can, for phase rule purposes,
be split into two triangular fields. Moreover, in each plot there are two possible
ways of dividing the appropriate quadrilateral, indicating that the alternative
parageneses are in crossed relationship. Thus, in the ACK diagram the quadrilateral
is: ct-“‘an’’-mu-mic, which can be divided according to Stability A into {ct-
“an ’’-mu] and [ct-mic-mu] by the two-phase join ct-mu, or according to Stability B
into [ct-‘‘an’’-mic] and [“an’’-mu-mic} by the join “an’’-mic. Similarly, for
the FCK diagram the quadrilateral is ct-‘‘ di’’-phlog-mic, which can be divided

1 For the purpose of representing theoretical assemblages in this discussion “ an ”’ is used to represent

plagioclase or “‘ epidote ”’ or both, “ di’”’ is used to represent diopside or a calciferous amphibole or both.
The remaining minerals are represented by the abbreviations of Table IT.

1336 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

according to Stability A into |ct-phlog-mic] and [‘‘ di’’-phlog-mic] by the join
ct-phlog, or according to Stability B into [ct-“ di’’-mic] and [“ di’’-phlog-mic]
by the join “ di’’-mic.

These considerations applied to the more complex case of the ACFK tetrahedron
show that the complete limestone-skarn field is an irregular triangular prism
ct-“ an ’’-mu-phlog-“ di’’-mic (Text-fig. 5). This may be split initially into a
tetrahedron, and a quadrilateral pyramid by a three-phase plane. The quadri-
lateral pyramid may then be split into two tetrahedra by a second three-phase
plane in one of two possible ways.

It thus appears that by means of two three-phase planes the initial irregular
triangular prism may be split into three tetrahedra.1_ By inspection it can be seen
that there are six of these potential three-phase planes in the irregular triangular
prism, namely,

: ct-mu-phlog
mic-‘‘ di a an )
ct-phlog-“‘ an ”’

: ct-‘ di ’-mu
>: mu- di’’-mic
: phlog-“ an ’’-mic,

AMoOOWS

and that there are, in all, six possible sets, each of three tetrahedra (or six pairs of
three-phase planes).

It may be noted that the three-phase plane A and the three-phase plane B are
in crossed relationship and correspond to the two-phase joins in crossed relationship
in the ACK and FCK triangles.

In the Glen Urquhart limestones there are three four-phase assemblages together
with three three-phase, and two two-phase assemblages. The whole group can be
shown to be at equilibrium under conditions where the limestone-skarn prism
is subdivided by the three-phase planes A and C (and not by any of the others).
The arrangement of the subdivided prism is shown in exploded form in Text-fig. 6
and the theoretical and observed assemblages appear in Table IV below.

TaBLE IV.—Equilibrium Limestone Assemblages

Theoretical (+ qtz) Observed (Table IT)
ct-phlog : 22,

an’ = dil? ‘ 18, 19

ct-“‘ an ’’-phlog : 5, 6, 14

ct-“ di ’’-phlog : 17

ct= san’ *= di ~ : Te ee As 72 el

ct-“ an ’’-“‘ di’’-phlog 351670} TO; 11,13, 10
ct-“ an ’’-mu-phlog ¢ 20, 21
ct-mic-mu-phlog c 23

1 The phase rule allows a maximum of four phases (i.e. a tetrahedral stability field) in this four-
component (ACFK) system. We have already seen that there are three unplottable additional components
playing a less important role in the full system, namely excess SiO, in quartz (which is always present),
Na,O in plagioclase, and H,O in “ epidote ’’, actinolite, atid the micas, thus permitting the crystallization
of three additional phases (minerals) at equilibrium.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 137

Microcline and muscovite are only represented in the limestones affected by
alkali injection near to granite pegmatites. These injected limestones carry some
(? retrograde) chlorite after phlogopite and occasionally some tourmaline. Such
rocks occur at Torr Buidhe, at the northernmost limestone quarry in the area, and
at the quarry above Kilmichael Church. The other parageneses are distributed
evenly through the limestone outcrops.

mic
mu
hlog
mu ct
an ‘ep ct
phloga—~

an ‘ep ct

Fic. 6. Three four-phase fields delimited by the planes ct-mu-phlog and ct-phlog-“‘ an”’
in the ACFK tetrahedron, illustrating the parageneses of the Glen Urquhart lime-
stones.

MIN. I, 5. s

138 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

The biotite of the marbles is highly magnesian, and is stable with diopside (and
amphiboles) and quartz. Ramberg’s reaction (No. (4) of this paper) does not there-
fore seem to have taken place. The high Mg”/Fe” ratio of the mica apparently
reflects a similar ratio in the parent rock (dolomitic limestone).1_ The reasons for
this common condition in the parent rock are chemical factors during its sedimentation
or subsequent diagenic alteration. The origin of the high Mg”/Fe” of many limestone
biotites can therefore be traced back to the geochemistry of some particular sedi-
mentary or diagenic process rather than to reactions within the amphibolite facies.

Systematic absences. Dolomite and talc are absent from the parageneses at Glen
Urquhart, presumably having been resorbed before the appearance of diopside, the
third mineral in the progressive metamorphism of a siliceous dolomite rock.
Wollastonite is absent and quartz plus calcite is the universally stable alternative.
The temperature attained cannot have been sufficient to stabilize wollastonite
under the pressures on the solid and the fluid phases ruling during the metamor-
phism of this limestone. Idocrase and grossular are absent and may well represent
higher grade conditions within the amphibolite facies than those that operated at
Glen Urquhart (Ramberg, 1952). Weeks (1956) has recently calculated that
forsterite should appear at higher grade than diopside. He has also suggested why
enstatite does not occur in metamorphosed siliceous dolomites. These minerals
therefore need not be expected in the Glen Urquhart limestones.

Summary. The metamorphosed siliceous magnesian limestones of Glen Urquhart
were crystallized at equilibrium just within the lower stability limit of the amphi-
bolite facies. All the minerals to be expected in limestones at this metamorphic
grade are present, and some of their characteristics have been described. There
are other minerals whose compositions fall within the same chemical field but whose
absence may be explained on physico-chemical grounds. The assemblages calcite—
microcline—phlogopite and calcite—quartz—phlogopite are both shown to be stable
in theoretical and observed parageneses. It is demonstrated that there is no vacant
space in the limestone field of the ACF diagram for rocks deficient in potash (lacking
potash feldspar) at this metamorphic grade.

PART lia

SKARNS
Previous Work

The calc-silicate skarns mentioned in the Introduction to this Bulletin have not
previously been studied as a separate rock group in Glen Urquhart. It is clear
from the writing of previous workers (particularly Cunningham Craig) that these
rocks were regarded as impurities interbedded with and stratigraphically succeeding
the limestones. The evidence, presented below seems clearly to indicate that the
rocks, at the boundary between the limestone and the kyanite schist, are the products
of metasomatism, rather than of changing sedimentation. Many of the silicate

1i.e. the normally high Mg”/Fe” ratio in impure limestone, or, put another way the more frequent
occurrence of dolomitic rather than ankeritic limestone.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 139

minerals from Glen Urquhart analysed by Heddle can fairly be assumed to come from
these skarns rather than from the limestones proper.

Field Occurrence

The mode of occurrence of the calc-silicate rocks has been mentioned briefly
in the Introduction. Lying between calcareous and aluminous rocks, and accom-
panied by much quartz veining they at once appear to be the products of chemical
mingling between two unlike sedimentary formations in contact. This impression
is confirmed by the chemistry of the skarns which will be shown to be intermediate
between that of the limestone and that of the kyanite schist. It is also confirmed
by the evidence of transported material in the skarn zone, lime, for example, in
quartz veins containing calcite and calc-silicates, and alumina in kyanite-bearing
quartz veins. Metasomatism is demonstrated in mono- and bi-mineralic silicate
rocks, chemically distinct from calcareous or aluminous sediments, often arranged
in convoluted zones, unrelated to the still-recognizable folded bedding-planes of
the adjacent sedimentary rock. Metasomatism is, indeed, most clearly indicated
under the microscope, where replacement of original minerals by others, leading
to enrichment of a kyanite schist host-rock in lime and a limestone host-rock in
alumina (and silica), can clearly be followed in successive stages.

The interaction of two chemically dissimilar rock masses in contact with one
another, under the influence of volatiles and of raised temperature, is to be expected.
This chemical reaction must lead toa diminution of the chemical potential (free
energy) in the rocks at the contact. It is a case of reaction skarn formation in
the extended meaning of the word skarn (Holmes, 1920). Although granite is
not directly concerned at the site of the skarns (as is usually the case) the quartz
veins are clearly linked to granite pegmatites and granite-like rocks exposed at no
more than a few hundred metres from the skarn zones.

The skarns may be considered in three main groups: epidosites, epidote-
hornblende skarns, and plagioclase skarns ; pale-amphibole skarns ; prehnite and
pectolite skarns.

Group I: Epidosites, Epidote-Hornblende Skarns and Plagioclase Skarns

Pure epidote rock is subordinate, but good examples of this rock, epidosite,!
can be found at the Gartally quarries, Torr Buidhe, and elsewhere, associated with
the more abundant epidote-hornblende skarns. Fine specimens of zoisite rock can
be found amongst the Gartally quarries. From one of these a cavernous aggregate
of intergrown zoisite prisms up to 3 cm. X 8 cm. has been obtained. The cavernous
nature of some of these zoisite rocks is in contrast with the compact and strongly
lineated crystallization of some of the other skarns, notably those built of pale
amphiboles. Differences in the physical conditions at various sites and probably
slight differences in age are betrayed by these contrasting crystallizations. On

1 In spite of the objections to the use of this term raised by Flawn (1951), it is here retained. “‘ Epi-
dosite ’”” has always been preferred to the alternative “ epidotite’’ in the literature on the Highlands

(e.g. Flett, 1906; Flett, im Hinxman eé# al., 1913 : 50; Harker, 1939 : 268, Bailey, 1955: 133). Retention
of the firmer term in a work cn Highland rocks appears simpler, and preferable to the use of the variant.

140 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

Torr Buidhe fine-grained apple-green rocks composed of clinozoisite-epidote have
been collected along with quartz veins enclosing abundant clinozoisite, rimmed with
epidote. These veins grow to a pegmatitic grain size. One vein contains clino-
zoisite, idioblastic to a fine-grained mosaic of quartz, in prisms up to 2 cm. diameter.
More frequently the quartz—epidote mosaic is equigranular, the grains averaging
Imm. diameter. The clinozoisite is often well crystallized, with good terminations,
but these faces proved somewhat too roughened for goniometric work. The crystals
may be apple-green, grey-violet or brown in colour.

Associated with the epidosites, and sometimes by themselves, are rocks containing
dark green prismatic amphiboles, paler green diopside (salite), and sometimes
calcite, quartz, and feldspar. The majority contain “ epidote ’’1 in varying quantity,
some contain none. Again crystallization may be very coarse-grained; a
hornblende—quartz—calcite rock on Torr Buidhe containing amphibole prisms
6 cm. X I cm. and a diopside—hornblende—clinozoisite-prehnite rock on Sgor
Gaoitne with diopsides up to 3 cm. X I-5cm. are known. Torr Buidhe isa prominent
locality for these skarns, and they are also to be found at the following places : Sgor
Gaoithe ; the serpentinite contact due north of the Gartally quarries ; Upper and
Lower Gartally farms; Wester Balnagrantach ; Allt Gartally (at the serpentinite
contact) and at the limestone quarry above Kilmichael Church. The host rock
can sometimes be shown to be the kyanite schist but the limestone is always close
at hand.

The epidosites and the epidote—hornblende skarns are transitional to a group
of metasomatic rocks that form along the margins of the serpentinite mass and of
which the conspicuous feature is the formation of large plagioclase crystals. Some-
times they coalesce to form plagioclase rocks similar in appearance, though not
identical in mode of formation, to the albitites emplaced during the same metasomatic
episode into zoned hydrothermal bodies within the serpentinite (Francis, 1955).
The normal aspect of the plagioclase rocks in the sediments is one of strong
crystalloblastic growth in their host-rock which is kyanite schist (Text-fig. 9).
Along with the big plagioclases the calc-silicates ‘‘ epidote’’, hornblende, and
diopside play an important part in these rocks. These minerals and the clear par-
ticipation of Ca-bearing fluids in their genesis (to be described below) perhaps
allow these rocks to be classed as plagioclase skarns, lying in an extension to the
commonly defined limits of skarn composition. The plagioclase skarns occur at
Sgor Gaoithe and at a few other points north-east of that hill, all lying close to the
serpentinite contact.

Group II : Pale-Amphibole Skarns

These skarns are quite distinct in appearance and in location on the contacts of
limestone with kyanite schist from those of the preceding group. The fact that they
both occur at such contacts and that there are mineral similarities between the two
groups suggest that their mode of formation and age are similar.

The typical rock is almost monomineralic, and on freshly-broken surfaces shows

1 See Footnote, p. 130.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 141

sparkling cleavage faces of the amphibole, of a greenish-grey colour ; occasionally
a waxy-white diopside of slightly earlier formation occurs with the amphibole, and
both minerals will be shown to possess high Mg/Fe” ratios. The pale-amphibole
skarns weather to a pale grey colour and display a furrowed and corded surface.
The latter marks out a strong foliation, with a closely-aligned lineation thereon.
Folds in this foliation parallel those in the adjacent sediments, but it is not clear
whether these folds in the skarn are a palimpsest of the folding of the replaced
sedimentary host-rock or whether they have been generated by expansion during
the metasomatic replacement itself (cf. Poldervaart, 1953). The host-rock of the
pale-amphibole skarns may largely have been limestone. Indeed limestone may
play the host’s part in this group to a greater extent than in any of the other skarn
groups in the Glen (kyanite schist being the resting place of the products of inter-
mingling in most cases). Towards the (structural) top of the pale-amphibole skarn
masses there is microscopical evidence of a certain amount of kyanite schist host-
rock ; anthophyllite has been found as an intermediate product in the formation
of the latter skarns.

Pale-amphibole skarns occur in an anticlinal belt north-west of Loch an Sgor
Gaoithe ; in hillocks between Upper Gartally and the limestone quarries, and in the
valleys of the stream north-west of Lochan an Torra Bhuidhe and of the Gartally
Burn above Milton.

Group III : Prehnite and Pectolite Skarns

This final group forms a very small part of the skarn assemblage. It is closely
linked to the first group of skarns and occurs close to them. Like the plagioclase
skarns, rocks of this group are restricted to the vicinity of the serpentinite contact.
They are found in two places: a few metres west of a number of thin, isoclinally
folded outcrops of marble, near the serpentinite, north of Gartally quarries ; and at
a few centimetres from the serpentinite contact at Sgor Gaoithe. The host rock of
these skarns appears in every case to have been kyanite schist, although limestone
is demonstrably close to them at the former locality and is probably close to them
at the latter. The margin of the relatively impervious serpentinite must have
determined the location of these skarns as much as the sedimentary junction between
kyanite schist and limestone. They are often quite inconspicuous rocks in the field,
having much of the appearance of normal kyanite schist. Pseudomorphs in musco-
vite after kyanite have been found at an early stage in the alterations. They stand
proud from the rock on weathered surfaces. Kyanite relicts within them have
been converted into clinozoisite (this alteration is discussed below). At a slightly
later stage a pale, weathered crust develops and then hydrobiotite, derived from
original biotite becomes conspicuous as pale brown or pearly flakes, giving the rock
a spangled appearance. In one much altered rock xenoblastic garnets up to 6 cm.
diameter occur sporadically. They are intergrown and invested with coarse hydro-
biotite plates and white prehnite. Prehnite and pectolite in white, interstitial
masses and veins (together with vermiculite) characterize the much-altered rocks
quartz and feldspar are eliminated. These rocks bear the imprint of strong lime

142 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

metasomatism. For complexity and interest they are not matched by any of the
other Glen Urquhart skarns.

Mineralogy

Zotsite and thulite. The cavernous growths of large zoisite crystals at the Gartally
quarries have already been mentioned. Here and elsewhere among the epidosites,
epidote—actinolite skarns, and plagioclase skarns both the a- and the #-forms occur,
the f-form being the commoner of the two. Zoisite has been found in the pale-
amphibole skarns and the prehnite and pectolite skarns ; where determined in the
two latter skarn types it has always proved to be f-zoisite. The optical distinctions

ee LT TOP

C2

en er ee
EE EA UE RA
we,

eel

Fic. 7. Minerals of the epidote group: epidote shown black, extinction angles in
Table V. A, Clinozoisite-epidote crystal in epidosite, Torr Buidhe (cf. Pl. 2, fig. 1).
B, C, Clinozoisite—epidote crystals in epidosite, Balnagrantach.

between a- and f-zoisite here employed are those of Termier, given in Winchell
(1951). These distinctions are consistent and no anomalies are apparent amongst
the zoisites studied.

The crystals show zones of anomalous Berlin blue and of grey and buff interference
colours. The boundaries between these colours often parallel a prismatic (100)
parting ; elsewhere they are irregular.

In two outcrops a bright pink zoisite (thulite) has been found, both in the limestone
quarry above Kilmichael Church. In one case the thulite occurs disseminated
in a rock which, although perhaps transitional to a skarn, has been grouped with
alkali-injected limestones (Part I, p. 127). In the second case the thulite occurs
as a centimetre-wide vein in actinolite skarn. The mineral is optically positive

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 143

with a very small optic axial angle (y > v, distinct), the optic axial plane is parallel
to (oor), and straight extinction may be observed in (010) sections ; these data prove
this thulite to be related to f/-zoisite.

Clinozoisite and epidote. These minerals are together about as common in skarns
of Group I as is zoisite. They are usually in crystals with complex zoning, having
clinozoisite cores and narrow, impersistent epidote rims. This apparently indicates
a slight iron enrichment in the metasomatic fluids towards the close of the period
of skarn formation. The epidote is colourless and non-pleochroic. It is, to judge
from its optical properties, iron-poor, having little more than 10% of the
Ca,Fe,Si,0,,OH molecule.

Compositions have been determined by the angles of extinction to the (100)
cleavage. These are best seen where the crystal is twinned on (100) and the section
is normal to the (oor) cleavage, that is to the 0 crystallographic axis (Text-fig. 7
and Pl. 9, fig. 1).

As with the same mineral group in the limestones, no /-clinozoisite has been
found (this has a negative extinction angle exceeding 45°, Johnston, 1949). Extinc-
tion angles determined in these minerals in the skarns range from —6° (clinozoisite
to + 3° (epidote), as shown in Table V.

TaBLe V.—Extinction Angles in Twinned Clinozoisite—Epidote Crystals

(x A c measured | b. The angle is positive in epidote, negative in clinozoisite ;
the crystals A, B and C appear in Text-fig. 7)

A B C
———— (aS SSS See eS ee Ee
LH RH LH RH LH RH
I=+3° 4 = — 6° I= + 3° 4 = — 2° i= — 2" A 77
eae Oa eS a a i ae a 2= + 3° =
3=—6° = 3=-4 coca Sik =

Epidotes, but no clinozoisites, are recorded from the pale-amphibole skarns.
The pectolite and prehnite skarns sometimes contain large zoned clinozoisite—epidote
crystals up to 3 mm. across. They display polarization tints in blues and lemon
yellows, with zoning brought out by these colours.

Amphiboles. These are perhaps the most important minerals of the Glen Urquhart
skarns. The amphibole of the skarns of Group I| is a dark green prismatic type, which
might be termed actinolite in the field. Chemical analysis (Table VI, 1) shows that
it is both too aluminous and too magnesian to fall into the actinolite group as currently
defined (Winchell, 1951) ; the mineral can best be styled a magnesia-rich hornblende.
Trace elements in this amphibole are set down in Table VII, x and its optical
properties in Table VIII, 1. Its chemistry and optical properties are probably
typical of amphiboles in Group I skarns, although extinction angles of 20° and more
are recorded amongst them.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

144

‘oIndy 7eY} JO %I pssoxe you ATqeqord [tM o1n3y 4O* Ey oy} Ur psonporzUt 10110 oy ‘enur10F
ey} UT 00-2 = HO onjea oy} Surumsse Aq pourezqo Aprresziqre useq sey O*H JOJ onyea oy} pue (WoTZTUST UO ssoT ‘a'1) 10}eM J) WOT}VeUTUIIE}Op
Aj[NeF & 0} ONP sv UdYe} USEq o10Jo104} SPY UOT}EUTUINS OY} UT }OYOp oy, ‘a]qeuosveI ore sonTeA 10Y,? Te {anoure reruns e Aq sopus,quioy
JO} OSLIOAL oY} ULY} IOMO] SI (%9F-0) JozeM IOZ OINSY oY} pue ‘Mol %z ynoqe st Sioqspururey Aq woars sisATeue Sty} JO uoTyeurTUINS OUT, 7

uPA + WW + 3W

T cs) =
*(dnoz8 x om urty = ty) “(Sbex) J MET UN + ed ary coz ~ 9
*(Sh6x) yoyouts, Aq uaar8 yojd sfoqrydure ue 107 sozeutp10-07 ooz — (—¥ + eN + 2D) oo1 = g
009 — IS COL = VW

‘pA “WA ‘OD ‘E ‘spjoyoon “yx ‘Ss ‘z pue x ‘SUOT}LUIUIID}Op oUuTIONp YT

‘(siskTeue oyeorIs Jo poyjow prides oy} € 107 Sutsn) urprodeyy “9 “yy ‘€ 10y { uTpredeyD “5 ‘MI pue yTEMujog ‘J ‘q ‘z pue x soy : syshyoup

‘(I ‘HO ‘O)bz 10d suoneo szuasoidox sishyeue YoRa IOF ,, & ,, uuInjo)

1(gSgr) S1eqsjeuueyy “9 Aq siskyeue puv uonydrosep “X'N “09 a8ueIQ ‘aTtAuopy ‘,, yuapy ,, “v
qreynbig wey

‘Aypeyaey toddy ‘Arrenb ysa8req Jo ysvo-yynos ‘wi oS ‘Arrenb [ews ‘ureys aprqrydue eyed ur ‘(,, azuapa ,,) apuajquaoy yotd-visausv py °€
‘yzreynbig uely

‘oyPOeD 1903S ue YOOT JO ySoM-Y}IOU “UT ObI ‘OIG JO d0eI} & SuTUTe}UOO ‘uIeYS ose[OOISe[d—sjoqrydure eyed ul apuarquaoy yoid-visausv yy *z
‘qzreynbig uepy

‘oupmnyd P10] ue ueysoTy Jo yy1ou 4ysnf spisyry ‘ureys oj0[e0-zz1enb-ssepoo1Se[d—syoprde—epusquioy uory ‘apuarquioy YIUA-DiSsaUsv IT “1

6gI ber Ser LEte : @)
oz 6z or gI é ae q
ZI Vor LE Cora : V
* (20.0
6S0.€=q zo -€=,,°q 10 -€=,.°q Go-€=94,°q * W0/"wW3)q
11z8 -66] pl -OOI 99 -66 g6 -66
bz. €1.0 L1-o A =O ssot
gf -OO1 62 .66 CI -oo1

€Sz.o gS.o ZEI-O o£ .o 6L1.0 oF.o : c a
Qe +z — — ZQ +I — 60-0 g6.1 — 60.0 "+9 jorr-O°H
00-Z 000.2 [91 -z] _ (£20.24 Z«Z gh .1 09 -I bol. 8g-1 * +5 Jor1+ O°
ghi-o +g .0 040-0 F.o OI0-0 ZI-O Sto.o Sz-o : = OST
OZ -Z< ZOZ-O GL.o 6z-z7< Ibz-o 6-0 O1-z< 9Lo-o 9S-o QI -Z< 002-0 €Z.0 : - OPeON
gbg-1 Zv .zI €96.1 F.€1 QgIO.z oS .€1 gf6.1 98 -zI : * oe
— —_— Z10-0 1-0 100-0 90-0 gIlo-o S1-0 : * OUI
9fg-F LE .€z CSo.4F Z.61 6z6.b zS .0z gIg-€ IZ-QI : * O8IN
€6€. v.€ LzZ.o LE .z 1ZQ-O ZG : : E)
$z.G< 66z-0 98-24 gah ef -9 pe Lee a pr 96 -+ as aes eee
— —- 1ZO-0 z-0 6Z0.0 gz +O $z0.0 ZZ +O ; 2 Hoye

661-0 oof .o gIt.o obz.o , :
eee CL.S ce inal ae oF .g aan ZIG tO1TVv
see a) ey LoniGe ee Chel Wace ame Clio Gal: Upece © 2? \o6Gas 1S .zS . - %01IS

eV v ee € ew z eL I

sajoqiygupouyy) usvys fo sastjoup—]A AAV]

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 145

TaBLeE VII.—Trace Elements in Skarn Clinoamphiboles

(parts per million)

I 2 3 I 2 3
Ga 9 7 = Zr 37 37 45
Cr 47 100 70 ng zt 12 ne)
Vv 95 95 45 La N.D N.D =<
Mo * ‘ ig S) 5 bfe) Io 50
Sn * * * Pb * * *
Li Z 2 3 Ba 30 30 30
Ni 21 21 22 Rb * * *
Co 6 6 5 Cs N.D N.D =
Se ; * : ‘ ; *
F . 4000 . 3000 = O00);

I, 2, 3, as for Table VI. Trace element determinations by Dr. S. R. Nockolds.

* = present, but below the limit for quantitative determination of this element (see Nockolds and
Allen, 1953).

t Determination by C. F. M. Frvd.

TaBLeE VIII.—Optical Constants of Skarn Clinoamphboles

Refractive Optic axial Extinction
No. as in indices: Na angle: Na nee IN Pleochroic
Table VI light (+ 0-002) light (in plane | b) scheme
r a = 1°623 : — 83° : 18° . «colourless.
Gr E634 . greenish straw.
Y = 1:642 . yY pale leaf green.
2 oo — I-6n5 . — 88° : 18° . colourless.
Cisne 627
Yy = 17638
3 % = 1613 : — 84° 20° colourless.
Sy— 1-623
nr £5037

The characteristic amphibole of Group II, the pale-amphibole skarns, was separated
from a rock composed almost wholly of the mineral, together with a small amount
of zoned plagioclase (Ang3, average), and trace quantities of zircon and iron ore.
Its chemical composition is set down in Table VI, 2, its trace elements in Table
VII, 2 and its optical constants in Table VIII, 2. It may be regarded as a magnesia-
rich hornblende, being more magnesian than, but otherwise of similar composition to
the amphibole of the skarns of Group I.

Also to be found amongst the pale-amphibole skarns are the amphiboles called
“edenite’”’ by Heddle (1901, Vol. II). As noted in Part I of this Bulletin some
pale-green fibrous and fasciculitic amphibole corresponding to Heddle’s “ edenite ”
occurs in the limestone but there is more of this material, and also a pale slate-blue
amphibole of similar habit, in the skarns (which were not recognized as a separate
rock group by Heddle). A powdered sample of the latter type, from an almost
monomineralic rock, was cleaned of traces of calcite and plagioclase and analysed.

146 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

Its chemical composition is given in Table VI, 3, its trace elements in Table VII,
3 and its optical constants in Table VIII, 3. The chemical analysis of the “ edenite ”’
was carried out by the rapid spectrophotometric method and results are quoted
to 0-1%, by weight, only. The analysis is, however, likely to be of the same order
of accuracy as the other two, and the three may reasonably be discussed together.

The three amphiboles agree in possessing calcium approximating to 2 atoms per
formula unit, moderately high alumina, a high Mg/Fe” ratio, and alkali atoms
that must be assigned to the vacant space in the amphibole lattice. In addition
their trace element contents are closely comparable. The comparison is most marked
between nos. I and 2, which, along with similarities in major elements, supports
the supposedly similar time and mode of formation of the two skarn types containing
these two amphiboles. No. 2 is slightly further removed from 3 in trace element
content although both belong to pale-amphibole skarns.

The theory that the skarns arose by metasomatic diffusion and commingling of
materials at the boundaries between limestone and kyanite schist is supported in
the analyses of these three amphiboles. Their calciferous and aluminous composition
reflects the contributions of the two parent rocks, whilst their high Mg/Fe” ratios
suggest that the magnesia brought to them from the limestone (which is appreciably
magnesian in bulk composition) must have dominated iron brought in from the
kyanites chists, or any other source. In one case anthophyllite has been observed
forming at an early stage of the transformation of kyanite schist to pale-amphibole
skarn. This probably indicates the transport of magnesia in advance of lime from
the limestone to a kyanite schist host rock.

The alumina content and the moderate filling of the vacant spaces by alkali
atoms remove these three skarn amphiboles from either of the two fields of limestone
amphiboles shown by Hallimond in his graphical plot for calciferous amphiboles
(1943, fig. 3). They fall in a field to which he has given the general heading of
“schists ’’. The optical constants of these amphiboles conform to the curves set
down by Winchell (1945, fig. 7 etc.) ; in particular they all plot within the field
of optically negative amphiboles.

The name “ edenite ” was originally given by Breithaupt (1847) to an amphibole
from ‘‘ Eden” (Edenville, Orange County, New York). The first analysis of an
amphibole designated “‘ edenite ”’ from Edenville is given by Rammelsberg (1858).+
This analysis is plotted in Text-fig. 8, together with the three amphiboles of the present
study, Heddle’s tremolite (see p. 128), and his two “ edenites”. Also plotted are an
amphibole from Edenville named “ edenite ’”’ by Winchell (1931), and an “ edenite ”’
from South India (Subramaniam, 1956).

Rammelsberg’s analysis is of early date and probably incomplete, but it can be
recalculated to a respectable distribution of atoms (Table VI). It may thus be
taken as having priority, and as providing a fair guide to the composition of the type
material. Heddle was correct in naming his slate-blue amphibole “ edenite”’ on
the data available to him. The three amphiboles described in the present study are
also similar to Rammelsberg’s mineral (Table VI and Text-fig. 8). Heddle’s pale-

1 Rammelsberg records the mineral as the “‘ edenite ”’ of Breithaupt, but he prefers to call it hornblende.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 147

green “‘edenite’’ is not close to the original nor are those of Winchell or of
Subramaniam (Text-fig. 8). Winchell’s analysis lies well beyond the probable
limits of analytical error in Rammelsberg’s analysis and cannot be accepted as a
repeat analysis of the type material. Still less can the composition
NaCa,Mg;Al1Si,0,.(OH), be labelled “ edenite ”’.

NaCa,Mg,Al Si,0, (OH),

Na CaMg,Al,Si,0

,91,0,{0H),

2

CEM Si H), Ca,Mg,Al,Si,0,(OH),

Fic. 8. Hallimond plot of the field of calciferous amphiboles. The amphiboles
mentioned in the text are lettered as follows :

A, magnesia-rich hornblende, Glen Urquhart (Table VI, 1) ;

B, magnesia-rich hornblende, Glen Urquhart (Table VI, 2) ;

Cc, magnesia-rich hornblende, Glen Urquhart (Table VI, 3) ;

D, ‘‘ edenite ”’, Edenville, New York (Rammelsberg, 1858, and Table VI, 4 of this
study) ;

E, “ edenite ’’, Glen Urquhart (Heddle, 1901, vol. II) ;

F, ‘‘edenite ’, Glen Urquhart (Heddle, 1go1, vol. II) ;

G, tremolite, Glen Urquhart (Heddle, 1901, vol, ii, and see p. 128-129 of this study) ;

H, ‘‘ edenite ’’, Edenville, New York (Winchell, 1931) ;

J, ‘‘edenite ’’, Sittampundi, Madras (Subramaniam, 1956).

This last usage was introduced by Berman (1937) who employed the term “ edenite-
hornblende”’. The same theoretical formula or “end member”’ has been called
“ edenite ”’ by (e.g.) Winchell (1945, 1951), Sundius (1946), and Subramaniam (1956).
The composition NaCa,Mg;AlSi,O,.(OH), lies in the optically positive field in
Winchell’s diagram (1945) whilst Rammelsberg’s mineral lies in the optically negative
field (as confirmed, e.g., by optical constants of the Glen Urquhart amphiboles ;
Table VIII). The above theoretical composition is merely an unlabelled point in
a variation diagram, in an area unrepresented by any amphibole analysis,! and does
not merit baptism under the name already given to an amphibole chemically and
no doubt optically distinct therefrom.

As for the Breithaupt-Rammelsberg mineral it lies within the field of the common

1 Sundius (1946: 20) himself pointed out that no analyses exist of amphiboles close to NaCa,Mg,
AlSi,O,.(OH)>.

1448 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

hornblendes and it therefore seems advisable to discard the term “ edenite”’
altogether.

Pyroxenes. The clinopyroxene salite is common amongst the skarns of Group I,
but when it is associated with amphiboles the latter predominate. It is leek-green,
and often of coarse prismatic growth. In places it may form a pyroxene rock.
One such rock (from which the analysed sample was separated) contains prisms up
to 3. cm. X 1-5 cm.in size. There is in this rock late growth of calcite, zoisite, and
prehnite together with occasional large sphene crystals. As with the pyroxenes
of the limestones, hornblende needles commonly “spear”’ through the skarn
pyroxenes, whilst in certain of the plagioclase skarns hornblende needles fringe
the pyroxene crystals. The analysis, trace element content, and optical constants
of the salite appear in Table IX, 1. Its optics agree well with the curves for clino-
pyroxenes published by Hess (1949).

TABLE IX.—Skarn Clinopyroxenes

parts per
weight million
per cent cations per 6 [O], (weight)
SSS a aS ee
I 2 I 2 I 2
SiO, 52°96 53°65 BOB 7s a S/T 0Gr a Sie Ss 2°05
Al,O, I°21 I-89 a. 033 2 039 Eat
0-020 0: 042
TiO, 0-10 0-04 0-002 0* OO V L100 467
Fe,O3 0°49 0:58 Or 013": o-o16 | Mo %
FeO 6-00 I°53 0° 185 eke 0*047 aoe
MnO 0°56 0:09 0-018 0-003 Tea Io 8
MgO I4°02 16:23 0: 780 0: 883 Ni 500710
CaO 24°29 25°30 0:967 ; 0-990 |__. Co 50 3
Na,O 0-36 0-19 0° 027 ee 0° 013 Ae Se 40 *
K,O 0:02 alr — Lx 50 gn ae
1s OM PES 0:23 0°47 In * N.D
EO os 0:20 0:07 x 10 0%
F N.D. N.D La * ND
100+ 44 100+ 04 eh asi 5 BEEZ
D(em,/em:4) D7*=3°34 D2—3-28 al Eb * is
+ 0-02 : Ba * ro
————— Rb * *
Cs +2 NED:

1. Salite, in salite-hornblende—zoisite—calcite-prehnite skarn, south flank of Sgor Gaoithe, Glen
Urquhart. Anal. G. H. F. Refractive indices in sodium light (+ -002), « = 1-677, 8 = 1686, y = 1-706 ;
2VY = 58°; » A ¢c = 41°; colour grey, ncn-pleochroic ; atomic percentages, Cayo.3 M&go.7 Fe 4-0.

2. Diopside, in diopside—zoisite-amphibole—calcite skarn, north side of knoll 250 m. north-west of
Lochan an Sgor Gaoithe, Glen Urquhart. Anals. D. 1. Bothwell and K. C. Chaperlin. Refractive indices
in sodium light (-- 002) « = 1:665, 8 = 1-672, y = 1-695; 2Vy=57°; vA ¢ = 39°; cclourless ;
atomic percentages, Ca,,., Mg,;., Fes...

Trace element determinations by Dr. S. R. Nockolds ; *, present, but below the limits for quantitative
determination of this element (see Nockolds & Allen, 1953).

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 149

A waxy-white diopside occurs sporadically in the pale-amphibole skarns. It
is in chunky prisms up to 5 mm. in length, forming an almost monomineralic rock.
The accessory minerals are zoisite, pale amphibole, calcite, sphene, and zircon.
Where it grades into the normal skarn it can be seen under the microscope that the
amphibole spears into and replaces the pyroxene. The analysis, trace element
content and optics of this diopside are set down in Table IX, 2. It appears to be
among the purest natural diopsides analysed. Those of De Kalb, New York
(Zimanyi, 1893), Cascade Canyon and Crestmore, California (Merriam & Laudermilk,
1936), and Organ Mountains, New Mexico (Dunham & Peacock, 1936) are purer,
Cascade Canyon being the closest to CaMg(SiO),. The Organ Mountains diopside,
like the Glen Urquhart mineral, contains some alumina. A large amount of alumina
is not, in fact, to be expected in the latter diopside for Tilley (1938) has shown that
diopsides rich in alumina (“ fassaites’’) occur in spinel-bearing metamorphosed
limestones, and spinel is absent from the Glen Urquhart skarns. Segnit (1953)
has shown, by a study of synthetic specimens, that an increase in alumina diminishes
the birefringence and increases the optic axial angle of diopsides. The Glen Urquhart
diopside is apparently insufficiently aluminous for its birefringence to be measurably
affected and its optic axial angle is in fact only diminished 2° with respect to pure
synthetic CaMg(SiO;), ; the refractive indices are each 0-oor higher than the pure
synthetic material, the extinction angles of the Glen Urquhart, and the synthetic
diopsides are practically identical (cf. Wright & Larsen, 1909). The analysis of
the diopside from the pale-amphibole skarns re-emphasizes the high Mg/Fe” ratio
of the dominant minerals (and therefore the rock) of this skarn group.

Plagioclase. This feldspar is a stable phase in the Group I and Group II skarns
(attaining importance in the plagioclase skarns). In these Groups the mineral
develops from the groundmass as porphyroblasts, and in symplectites. Zoning
and patchy replacement are usual in these feldspars. They range in composition
from An, to An,,. The plagioclase of the Group III skarns is a relict mineral sur-
viving from the kyanite schist host (composition: Ans) 3s, zoned). In highly
metasomatized rocks of this Group it is replaced by a fine-grained aggregate of
prehnite. Stages in this replacement may be studied in thin section.

Chlorite. Allthreeskarn Groups contain chlorite. For the most part it is optically
positive and shows grey polarization tints, low in the first order of birefringence.
Amongst the Group III skarns the chlorites may reach a diameter of 1:5 mm. In
some cases they can be seen in thin section to be interleaved with hydrobiotite,
recalling the submicroscopic mixture of the two minerals described by Barshad
(1948) in a mixed-layer mineral from Lenni, Pennsylvania.

An unusual rock associated with the epidosites of Group I on Torr Buidhe is made
up of areas of fine-grained, dark-green chlorite interspersed with pale pink blotches
composed of irregularly intergrown calcite, quartz, albite, and muscovite, with
abundant apatite. The chlorite is in mossy, almost spherulitic growth. The green
colour masks the true interference colour but the birefringence appears to lie between
0-004 and o-or0. Cleavage flakes have a refractive index f# of 1-593 (in sodium
light : + 0-002) and show 2Vy near 0°. These figures suggest that the mineral
may be sheridanite, variety grochauite (Hey, 1954).

150 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

Hydrobtotite. In the Group III skarns a transition can be followed, under the
microscope, from normal biotite of kyanite schist to a biotite-like mineral in pecto-
lite and prehnite skarns, having sinuous foliae with stringy, or rat-tail edges and much
prehnite growing between the cleavages. The sinuous mica is closely comparable
in thin section to the investigated hydrobiotite in nearby hydrothermal bodies of
the same age (Francis, 1955). It shows limited exfoliation on sudden heating, and
is presumably also hydrobiotite.

Prehnite. This mineral, which is abundant in the skarns of Group III, has been
identified chiefly by means of X-ray powder photographs. In much-altered rocks
it occurs as a polycrystalline aggregate in the groundmass, and in white veins.
For the most part it is biaxial positive, with a moderate optic axial angle and bire-
fringence up to 0-020 in bright second-order colours. A mineral of very similar
microscopic appearance, containing, like the prehnite, abundant dusty inclusions,
was found to be sensibly uniaxial (negative). A crystal (from a white vein in the
skarn) detached from a thin section gave: «= 1-612, 8 = y = 1:652 (sodium
light), 2Va~o°. This fragment yielded the X-ray powder pattern of prehnite ;
the vein from which it came is white and apparently monomineralic, and several
of its crystals are seen to be uniaxial negative under the microscope. This vein
can be distinguished in the hand specimen, and a sample from it also gave the X-ray
powder pattern of prehnite.

Winchell (1951 : 360) records the frequent occurrence of a variety of optical
anomalies in prehnite (including oscillation of the optic plane, apparent reduction
of the optic axial angle, etc.). It is an open point as to whether these anomalies
are sufficient to explain the optically negative character and high birefringence of
the present material. Prehnite occurs in the interstices between salite crystals
in the pyroxene-rich skarn which yielded the analysed salite on Sgor Gaoithe
(PI) 9; fig. 2).

Pectolite. Identification of this mineral has also been by X-ray powder photo-
graphy. It is a white crystalline aggregate in the groundmass and in veins in the
Group III skarns. Although very similar in appearance and in birefringence
(middle second-order) to prehnite it has a more markedly fibrous habit. The typical
habit is in sprays, rosettes and sheaves not usually exceeding 1-0 mm. in length.
The fibres have positive elongation. Pectolite is never associated with relict
kyanite which is always associated with prehnite in these skarns. A pectolite
vein has been identified in a salite-rich skarn of Group I, close to the locality of the
analysed salite.

Xonothte. Fine, brownish, opaque fibres, filming pre-existing minerals, are
common in the skarns. In one locality they form veins, white in hand specimen,
reaching 6 cm. in width, cutting across diopside—hornblende skarn of Group I.
The veins incorporate small blocks of skarn without apparent reaction. They are
exposed in the Gartally burn where the skarns are adjacent to the margin of the
serpentinite mass, above the village of Milton. The veins are made up of sheaves
of radiating fibres, milky-white and clear at their centres, brownish and opaque
at their extremities. This relates to their state of hydration. The centres of the
sheaves are xonotlite, the margins may be close to tobermorite (Dr. J. D. C.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 151

McConnell, personal communication). The fresh fibres have positive elongation
and a birefringence of about 0-014, with a refractive index for light vibrating parallel
to their length of 1-587 (sodium light; -+ 0-002); they are optically positive.
The type material from Tetela de Xonotla, Puebla, Mexico, and the Isle Royale,
Michigan xonotlite have higher refractive indices (1-593, 1-594 respectively). Spacings
of the lines in X-ray powder photographs of the type mineral and the Glen Urquhart
mineral are given in Table IX. The Isle Royale xonotlite has similar spacings except
that the one at 8-59 A. is not recorded.

Heddle (1901, vol. Il) has recorded wollastonite from outcrops nearby, recently
described as zoned hydrothermal bodies in serpentinite (Francis, 1955), close to the
Free Church of Milton. His two analyses report too much water to be considered
as pure wollastonites, and they are accompanied by specific gravities (each 2-175),
closer to xonotlite (2-17) than to wollastonite (2-915). It appears that Heddle’s
materials were both xonotlite.

TABLE X.—Xonollite: X-ray Powder Photograph Spacings and Intensities

A B A B

Line —— a Line i ec oe
No. d(A.) I LS No. d(A.) i aA.) It,
I 8°59 vw 8-59 30 17 1-86 Vw — —

2 7+03 S 7+ 06 70 18 1°84 vw = —
3 4°27 ¢ 4°24 go 19 1°83 Vw 1°83 70
4 es zs 3+ 89 90 zo I°75 a 1°74 30
gute wen = 21 E7T —_ I*70 70
22 I-68 — 1-68 Io

6 3°51 w 3°44 10 23 65° = ae pa
7 S25 P 3°23 = 4 24 1°64 — 1°63 40
8 3°09 m 3:06 100 25 I-60 as. 1°59 40
9 2:89 m 2°82 go 26 I-56 vw Tes 40
Io 2°70 s 2°69 70 7h OMe aly — I-51 70
LL 2°64 w 2°62 10 230.5. — = I*42 60
12 2°51 m 2°50 70 2) He) | E530 — 1-38 60
13 2°33 m ONE) 30 30 : — — I+ 36 ae)
I4 2°25 m 2°25 40 3I- ae E33 — I*35 Io
ne — — 2°03 go 32: we eae — W732 30
16 I°95 fal Gy TEC COY.| fete) ao7 Ps seen — I*30 30

A. Xonotlite, Gartally Burn, Milton, Glen Urquhart.
B. Xonotlite, Tetela de Xonotla, Puebla, Mexico. (A.S.T.M. Index 2-0598 (ICI-N).)

Other minerals (including accessories). Calcite and quartz are recorded in allthree
skarn Groups, as are accessory sphene, zircon, and magnetite. Biotite, some of it
phlogopitic, is common in Group I. It is partly relict from the schist host, but is
often in apparently stable growth. In Group II small amounts of phlogopite are
stable, whilst the biotite in Group III, mentioned above, is in early stages a relict
from the schists, later to be transformed into hydrobiotite. Muscovite is stable
in Group I, but is a relict mineral, appearing only in the early stages, in Group III.

Garnet, staurolite, and kyanite are also disequilibrium minerals in Group III,
although they persist, in part, into the most altered rocks. The refractive index

152 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

of the garnet is 1-786 (sodium light : + 0-003), a slightly lower value than the average
for kyanite schist garnets in Glen Urquhart (1-796). Garnets in Group III are deep-
red, xenoblastic and reach 6 cm. in diameter.

Of the remaining accessory minerals pyrite and apatite occur in Groups I and II,
rutile in Groups I and III (abundantly and in large crystals in the latter), graphite
in Group II, and serpentine in Group III.

Petrology

Group I: Epidosites, epidote-hornblende skarns and plagioclase skarns. The
parageneses observed in this Group are set down in Table XI. The great variability
of the Group is evident. There are seldom more than two representatives of a
paragenesis amongst the slides examined. The appearance and nature of the
epidosites and epidote—hornblende skarns has been sufficiently covered in preceding
pages and it remains to describe in greater detail the plagioclase skarns and transi-
tions between them and the first two types. These rocks, of insignificant areal
extent, are found close to the serpentinite margin. The host rock is always kyanite
schist, and early stages of transformation are always marked by the crystalloblastic
growth of large plagioclases (Text-fig. 9). At a later stage the whole rock may
be composed of plagioclase (oligoclase). It is always poorly crystallized, showing
patchy extinction, ragged intergrowth, irregular boundaries and other signs of
replacement.

oT Pr

Fic. 9. Polished face of a hand specimen of mica schist showing oligoclase porphyro-
blasts that have grown into a distinct mineral zone and have thrust themselves out
into the groundmass of the host-rock.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 153

On the south flank of the hill Sgor Gaoithe, close to the salite rock and to the
serpentinite margin, is a mass of plagioclase rock outcropping over an area of
several square metres. Its junction with kyanite schist is well exposed. It is a
sharp boundary with mineral zoning. The schist near the boundary, although
mapped with the kyanite schist, has not been observed to contain any kyanite or
the normal accessory rutile of this schist. It is otherwise mineralogically and
texturally similar to the kyanite schists. The first new mineral to appear towards
the boundary is sphene in crystals up to I mm. diameter. Immediately thereafter
appear porphyroblasts of oligoclase up to 2 mm. across. They have ragged edges
towards the quartzo-feldspathic ground, and wavy extinction. At this stage,
also, coarse clinozoisite comes in. The biotite of this zone is split and interleaved
with prehnite. The next stage involves the resorbtion of biotite, plagioclase,
and quartz and the formation of a pale-green, prismatic amphibole (y /\ c = 18° ;
pleochroism: «, colourless; /, pale grass-green ; y, grass-green). The amphibole
grows in quantity (at the expense of the remaining minerals of the schist) until it
forms, with occasional sphene crystals, a distinct mineral zone. (PI. 9, fig. 3).
At this stage the amphibole has slightly different optics (y /\ c = 22°; pleochroism
a, 6, colourless; y pale grass-green) suggesting an increased Mg/Fe” ratio. The
inner margin of this zone probably represents the primary junction of the schist
with the plagioclase rock. The amphibole zone is separated from the plagioclase
rock by a narrow rim of “ mossy ”’ clinozoisite. Some “islands’”’ of plagioclase,
which lie within the amphibole zone, are also fringed with mossy clinozoisite. The
plagioclase within the body of plagioclase rock has the composition Any , that close
to the mineral zones is more sodic (Ang). A number of thin calcite veins can be
seen in the plagioclase rock. The feldspar is considerably altered by an overgrowth
of calcium-silicate-hydrate material, especially near to the mineral zones.

As suggested, the first-formed boundary apparently lay between plagioclase
rock and schist. Alterations in the schist, with added material (possibly coming
from the direction of the plagioclase rock) may then have led to the formation
of the amphibole zone, adjacent to the boundary surface. Lastly the clinozoisite
would develop between the plagioclase rock and the amphibole zone. This order of
growth is convenient for picturing the changes but cannot be demonstrated. The
zones may have arisen in a different order or simultaneously. Chemical reactions
for the changes cannot profitably be set down in the absence of analyses of the
participating phases (feldspar, biotite, amphibole, and clinozoisite). The influence
of lime is, however, clear (and receives support from the veining and overgrowths of
lime minerals in the plagioclase rock). Kyanite and rutile, usually present in the
schists, seem, with the addition of lime, to have contributed anorthite (in the
plagioclase) and sphene to the marginal part of the schist. Biotite, with absorption
of lime and loss of potash, has altered to amphibole. Plagioclase rock near the margin
with the schists has been de-calcified and the liberated anorthite has probably
united with excess lime to produce clinozoisite. The clinozoisite, needing a small
amount of ferric iron, is localized at the margin of the amphibole zone, from which it
presumably withdrew (and oxidized) the necessary iron.

‘

MIN. I, 5. T

154 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

TABLE XI.—Skarn parageneses
Group I Skarns

zo-qtz-plag-hb clz-qtz-plag-hb clz/ep-plag(An,,)-phlog
zo-qtz-plag-mu-hb clz/ep-qtz-plag(An,.)-bi-hb clz/ep-hb
zo-plag-hb-di-ct clz/ep-qtz-plag(An,9)-mu-bi-hb-di clz/ep-hb-di
zo-plag-hb-di-ct-xon clz/ep-qtz-plag (An, )-bi-hb-di-ct clz/ep-bi-hb-di-ct
zo-chl-hb-di clz/ep-qtz-di ep-qtz-plag-hb-di
zo-hb-di clz/ep-qtz-hb-di ep-qtz-plag-hb-ct
zo-hb-xon clz/ep-qtz-hb-ct ep-qtz-plag-bi-hb-xon
zo-hb-di-xon clz/ep-qtz-hb-ct-pect ep-qtz-hb
zo-di clz/ep-plag(An,;)-bi-hb ep-bi-hb-di-ct
zo-hb-di-preh clz/ep-plag-chl-ct-xon qtz-plag(An,)-mu-chl-ct
clz/ep-qtz
Group II Skarns

trem-zo-di-ct trem-qtz-ct-phlog

trem-ep-qtz-plag (Ang, _47)-di trem-qtz-plag (An,,)-phlog

trem-ep-qtz-plag-chl trem-plag (An,5)

trem-qtz-ct anth-qtz-plag (An,,)-mu-bi

Group III Skayrns
preh-bi-mu-zo-plag(An,,)-qtz-chl
preh-pect-hbi-chl-ct
preh(?)-pect-hbi-mu(?)-plag(An,,)-qtz-chl
pect-clz/ep-plag(An,,)-chl-ct-xon
preh-pect-hbi-mu-zo-dsp-gar-ky-st-chl-ct-xon
Accessories: allanite, apatite, graphite, magnetite, pyrite, rutile, serpentine, sphene, zircon.

Abbreviations in Table XI: as in Table II, p. 130, and also: anth, anthophyllite; dsp, diaspore,
hbi, hydrobiotite,; ky, kyanite; pect, pectolite; preh, prehnite; st, staurolite; xon, xonotlite.

Similar relationships to the above have been found in several rocks near to the
serpentinite contact. Sometimes the fringes of clinozoisite are larger than those
on Sgor Gaoithe, and are seen to be delicate plagioclase—clinozoisite symplectites.
The host is plagioclase whilst the dactylites are clinozoisite. Diopside may be
present in these rocks. It is usually fringed with fine-grained amphibole needles,
in parallel growth. This late-stage amphibole is not fringed with clinozoisite at
its boundaries with plagioclase. Thin prisms of late-stage amphibole “ spear”
through other minerals in all the Group I skarns, as they do in the limestones.

Group II: Pale-amphibole skarns. Only a restricted number of rock types has
been found in this skarn Group, (Table XI). The pale-coloured magnesian horn-
blende which dominates these skarns, and the subordinate diopside and phlogopite
which accompany it (see above) impose a high Mg/Fe” ratio on the rock. As
suggested, magnesia has probably been contributed in much greater amount from
the limestone (originally of magnesian type) than has ferrous oxide from the kyanite
schist or any other source.

Some specimens reveal transitions from kyanite schist to skarn. In one case
needles of anthophyllite are collected in clusters of late, ““ spearing’”’ growth. They
are particularly associated with biotite of the schist, and seem to replace it. Some
extra magnesia would probably be needed for this transformation, whilst potash

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 155

is carried away. The anthophyllite is soon converted to the normal clinoamphibole
of the skarn, with the arrival of lime. The latter seems, in this case, to lag
behind transported magnesia. The potash liberated in the alteration of the biotite
probably contributes to pseudomorphs of “ shimmer-aggregate ’’ white mica after
kyanite which characterize an early stage in the replacements. This white mica
apparently does not survive into the final skarn parageneses. It may unite with
introduced magnesia to form phlogopite. Plagioclase becomes very turbid during
the metasomatism and both it and quartz are diminished, or even eliminated during
the changes.

Group III: Prehnite and pectolite skarns. Mineral assemblages of these skarns
are set down in Table XI. Amongst the rocks studied the most thoroughly meta-
somatized specimen was collected by Dr. H. I. Drever, and kindly loaned to the
author for study (paragenesis 5). At the locality close to the limestone isoclines
(the northern locality) rocks illustrating the earliest stages of alteration are not
difficult to find. Dr. Drever’s specimen is also from this area, but it is apparently
a rarity as no similar rocks have been found. At Sgor Gaoithe (the southern locality)
these skarns are only known in a state of alteration approaching that of Dr. Drever’s
specimen.

At the northern locality unaltered kyanite schist with fresh kyanite porphyro-
blasts can be seen passing into a biotiteschist, impoverished in muscovite and con-
taining pseudomorphs of “ shimmer-aggregate ’’ white mica after kyanite which
stand proud on weathered surfaces. Identical pseudomorphs occur in alkali-
injected kyanite schists in Glen Urquhart, as will be described in a further study.
The latter seem clearly to be due to the addition of potassium to the system, but the
skarns of Group III lie outside the area affected by potash metasomatism. An
interesting early stage in the alterations can be studied in a slide in which the
muscovitization of kyanite has been only half completed before this type
of alteration has yielded to lime-metasomatism. A relict core of kyanite in
“ shimmer-aggregate’’ has been converted to twinned and zoned prismatic
clinozoisite, whilst the “ shimmer-aggregate’’ appears to have taken up some
prehnite (Pl. 9, fig. 4). Plagioclase in this rock is, in general, fresh and strongly
twinned on the albite and pericline laws. It has the composition Any. In patches
and in threads along cleavages are fine-grained, brightly-polarizing minerals intro-
duced into the feldspar. These appear to be prehnite and muscovite.

At the later stages of alteration represented at both localities quartz and plagioclase
are eliminated, biotite has become hydrobiotite, and prehnite and pectolite have
greatly increased in quantity. Diaspore occurs in some slides. Epidote minerals
are common. Late veins of calcite and overgrowths of xonotlite can be found in
most of the specimens. Kyanite crystals, with inclusions of staurolite, persists
as unstable relicts in Dr. Drever’s rock. The significance of staurolite inclusions
in kyanite in Glen Urquhart has been discussed elsewhere (Francis, 19560 : 356).
The relict crystals lie in pools of prehnite aggregate. Garnets in the same rock reach
a size unknown in the kyanite schists that are not affected by metasomatism ;
there appears to have been some solution and redeposition of these crystals, accom-
' panied by a small compositional change, at an early stage in the metasomatism.

156 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

TABLE XII.—Chemistry of Skarn Formation

I 2 Ia Difference 2a

SiO, 61-86 43°20 Si : 57°89 —20°77 37°12
Al,O, 23°78 II*30 Al ; 26: 32 —14°98 11-34
TiO, I°O1 I-00 Ti - O-72 0°07 0°65
Fe,O; 0°83 0:80 Fe’”’ 0°59 —o:o1 0:58
FeO 6-31 4°70 Fe” 4°98 —I-60 3°38
MnO 0:09 0:06 Mn 0°07 —0:03 0°04
MgO 1°87 6-80 Mg 2°63 +6:07 8-70
CaO 0°56 26°50 Ca 0°55 +23:84 24°39
Na,O o°71 0:30 Na 1°30 —o:80 0°50
K,O 1°85 it Tatfo) K 2°23 —I-03 I*20
P.O; 0°27 0:05 1p o°21 —Oo'17 0°04
H,O+ 0:68 3°70 OH 2°41 +9°51 II-g2
H,O- 0°05 0°25

Motalan 99° 87 99:76

Sp. Gr. 3°06 2°74
Ga : 20 15 16 —5 Il
Cr . 65 7o 53 3 50
Vv : 90 I0o 71 +1 72
Mo 9 ‘ “4 — — _—
Sn - - * — — —
ei : Io 85 83 +549 632
Ni . 35 45 35 +4 39
Co 20 22 19 fo) 19
Sc 20 “3 24 —24 —
Zr 550 600 342 me 339
In © N.D. — — —
Y4 65 * 41 —4I —
La 120 - 49 —49 —
Sr 150 45 IOI —75 26
Pb * * — Sos ie
Ba 350 150 145 —89 50
Rb 85 50 55 25 30
Tl * N.D. —_ — —
Cs * * = pos =

1. Kyanite-garnet schist: Gartally Burn, Glen Urquhart. Major constituents (percentage weight,
oxides). Analysis by classical method, G. H. Francis. Trace constituents (parts per million by weight).
Spectrography, S. R. Nockolds.

2. Prehnite-pectolite skavn : Sgor Gaoithe, Glen Uruhart. Major constituents (percentage weight,
oxides). Analysis by rapid method, D. I. Bothwell. Trace constituents (parts per million by weight).
Sfectyvography, S. R. Nockolds.

tat Analysis 1 recalculated to cation percentage. Major elements as parts per million x 10-4, trace
elements as parts per million.

2at As with ta.

* Present, but below the limit for quantitative determination of this element (see Nockolds & Allen,
1953)-

+ To calculate major and trace constituents to cations (p.p.in.) the follcwing steps were adopted :

Foy Analysis 1 (classical method) :
(1) Ga, Cr, V, Sc, Zr, Y, La were converted to oxides and the total was subtracted from Al1,O, ;
(2) Sr was converted to oxide and subtracted from CaO ;

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 157

In their present state the garnets are cut through and enveloped by prehnite aggregate
from the groundmass.

The replacement of quartz, plagioclase, and kyanite of the original schist by
pectolite, prehnite, and, to a smaller extent, diaspore, may be represented by the
addition of only lime and water to the original assemblage, e.g. :

Al,SiO; + 2CaO + 2Si0, + H,O = Ca,Al,Si,0,9(OH).
kyanite quartz prehnite

NaAlSi,0, + 2CaO + 3H,O = 2NaCa,Si,0,(OH) + HAIO,
albite pectolite diaspore

CaAl,Si,0O, + CaO + SiO, + H,O = Ca,Al,Si,0, (OH),
anorthite quartz prehnite

The absence of amphiboles and pyroxenes from these skarns is noteworthy. The
bulk composition of the skarns, as will be shown, might be expected to favour the
growth of one or both of these minerals. Some undiscovered chemical bar to their
formation, perhaps related to the very high water content of the skarns, must
have operated.

The scope of the metasomatism may be studied in Table XII where the chemical
analyses, with trace elements, for un-injected material’ and the skarn product
are set down. Major and trace elements have been recalculated to atomic
percentages (Eskola, 1954), and gains and losses are indicated. Although there
is only one analysis of kyanite schist from Glen Urquhart available, and some
variation in the chemistry of that rock type definitely occurs, the possible departure
of the analysis from that of the skarn’s host rock must be slight in comparison with
the metasomatic changes here recorded.

There has clearly been a massive accession of lime to the skarn, together with
a small but definite rise in magnesia. The mobility of magnesia in the skarns
has already been noted. Hydroxyl has notably increased, suggesting the hydro-
thermal nature of the metasomatism, whilst among the trace elements there has been
a large gain in lithium. The remaining elements remain almost steady, or have
decreased in percentage. The decreases are probably related solely to the increase
in Ca, Mg, and OH. No field evidence has been observed for removal of any of

1 The initial material, kyanite-garnet—muscovite—biotite—plagioclase—quartz schist, will be described
fully in the forthcoming publication on alkali metasomatism in Glen Urquhart.

(3) Ni and Co were converted to Ni,P,0,, Co,P,0, and subtracted from Mg,P,O,, the difference was
converted to MgO ;

(4) Rb was converted to Rb,PtCl, and subtracted from K,PtCl, in the alkali estimation ;

(5) Li was converted to LiCl and subtracted from NaCl in the alkali calculations ;

(6) The weights (p.p.m.) of constituents thus revised were converted to cations and recalculated to
parts per million.

For Analysis 2 (rapid method, alumina by difference from total R,O,) :
(1) Ga, Cr, V, Zr were converted to oxides and subtracted from Al1,O, ;
(2) As for (6) in Analysis 1.

158 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

the chemical constituents of the rock with fixation elsewhere. This one-way meta-
somatism therefore implies expansion of the rock during injection. Strontium and
barium, which are scarce in the limestone (Table III) have not, apparently, followed
calcium in the metasomatism. As with some of the other elements the decreases
in Sr and Ba may be relative, rather than absolute, in this case

Summary. A metasomatic episode has occurred after the folding and meta-
morphism of the sediments of the Moine Series and the intrusion of the serpentinite
mass in Glen Urquhart. Some of the metasomatic rocks show clearly the replacement
of a regionally metamorphosed mineral assemblage and fabric by metasomatic
mineral assemblages. The metasomatism is linked with neighbouring alkali
injection, pegmatitic invasion, and feldspathization. In the area here described,
remote from the centres of alkali metasomatism, only veins bearing silica and
volatiles have penetrated. They appear to have picked up alumina from a kyanite
schist horizon (as shown by quartz—kyanite veins in this bed) and lime from the
(structurally) underlying metamorphosed limestone (as shown by quartz veins
cith calc-silicates in and near this bed). Under the influence of metasomatic fluids
these two constituents have mingled to form calc-alumino-silicate rocks at the
junctions of the two beds. These rocks may be regarded as reaction skarns.
Volume-for-volume replacements cannot generally be demonstrated. Some of
the skarns possess tightly folded and Jineated fabrics which may be palimpsests?
after the fabric of the regional metamorphism, but may equally reflect volume
increases during metasomatism, as suggested by Poldervaart (1953). Other skarns
have cavernous crystallization, reflecting different formation conditions and perhaps
slightly different age.

The mineralogy of the skarns is on the whole simple. Minerals of the epidote
group, plagioclases, phlogopite, amphiboles, and pyroxenes are dominant. The
amphiboles are closely comparable to the type “ edenite ”’ of Edenville, New York
(Rammelsberg, 1858). The term “ edenite’”’ is examined and it is concluded that
it is an unnecessary name for amphiboles close to the original material in composition,
which are simply common hornblendes. Still less desirable is its use for the
theoretical amphibole formula NaCa,Mg;AJ1Si,0,.(OH),., remote from the composition
of the original material, and unrepresented by any amphibole analysis.

A small group of skarns, differing from the rest, is characterized by the less
common minerals prehnite, pectolite, diaspore, and xonotlite.

Mineralogical and chemical evidence suggests that magnesia, in lesser amount,
accompanied the lime during the skarn formation. The limestone is of a somewhat
magnesian composition. Iron oxide (e.g. from the pelitic rock) has not played such
a positive rdle in the metasomatism. Some skarns have a moderate iron content,
others little or none. In the analysed examples strontium and barium have not
followed calcium in the metasomatism. Lithium is definitely increased and there
is a large accession of water.

The position of the skarns as intermediate reaction products between kyanite-
schist and limestone can be represented on the ACF diagram (Text-fig. 10). Skarn

1 Palimpsest stvuctuve: A structure of metamorphic rocks, due to the presence of remnants of the
original texture of the rock. [Dictionary of Geol. Terms.]|

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 159

A

kyanite

PELITE K°
x . FIELD

calcite
pectolite

xonotlite

diopside amphiboles

Fic. 10. ACF triangle, for rocks with excess silica, showing the plots of rocks and
minerals of the skarns and related rocks at Glen Urquhart. Arrows indicate the
chemical contributions made by the two sedimentary rock types to the reaction skarns
which have formed between them .

K, kyanite-garnet schist (Table XIV, 1) ;
s, prehnite—pectolite skarn (Table XIV, 2) ;
L, limestone (Muir e# al., 1956, see Table III of this study) ;
I, magnesia-rich hornblende (Table VI, 1) ;
2, magnesia-rich hornblende (Table VI, 2) ;
3, magnesia-rich hornblende (Table VI, 3) ;
4, tremolite (Heddle, rgo1, vol. II) ;

5, ‘‘ edenite ’’ (Heddle, 1901, vol. II) ;

6, “ edenite ’’ (Heddle, 1901, vol. II) ;

7, salite (Table IX, 1) ;

8, diopside (Table IX, 2) ;

9, 10, 11, zoisites (Heddle, 1901, vol. II).

160 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS obi

minerals analysed in the present study, and by Heddle (1901), together with analyses
of kyanite schist, prehnite—pectolite skarn, and limestone, are plotted on this diagram.
The varied mineral assemblages and the varying proportions ‘of minerals within
each assemblage amongst the skarns suggest that a large number of rock analyses
would be necessary in order adequately to define the chemical field of these skarns.

It is simpler, and almost as accurate, to represent them in a field in the ACF
diagram (Text-fig. 10) extending between anorthite, “ epidote’’, tremolite, and
diopside (all of which are found, in some cases, as almost monomineralic rocks!) and
extending from this area some distance towards the C corner and the F corner,
respectively, where further skarn minerals are plotted. It can be seen that the plot
point of the analysed rock (Text-fig. 10) lies close to the fields of pyroxene and
amphibole, yet contains neither mineral.

”

Acknowledgments. The early part of this work was carried out in the field at
Glen Urquhart and at the Department of Mineralogy and Petrology, Cambridge
University, and forms part of a Ph.D. Thesis. I am indebted to Prof. C. E. Tilley
for advice during the course of this work, also to Dr. S. R. Nockolds for trace element
determinations, and to Mr. K. Rickson for taking photomicrographs and X-ray
powder photographs. I have also to thank Dr. H. I. Drever of St. Andrews University
for the loan of a skarn specimen from Glen Urquhart.

Maintenance grants from the Ministry of Education and the Department of Scien-
tific and Industrial Research are gratefully acknowledged.

The work has been continued in the Department of Mineralogy, British Museum
(Natural History). I am indebted to Dr. M. H. Hey for reading the manuscript
and for his helpful criticism ; also to Messrs. D. I. Bothwell and K. C. Chaperlin
for chemical analyses, and Mr. C. F. M. Fryd, of the Government Chemical Laboratory
for’a fluorine determination. Mr. R. T. W. Atkins prepared the diagrams and
Mr. D. I. Wiliams took a photomicrograph. My thanks are due to them both.

REFERENCES

A.S.T.M. INDEX, 1950. American Society for Testing Materials, Philadelphia, Pa.

BarLey, E. B. 1955. Moine tectonics and metamorphism in Skye. Tvans. Edinburgh Geol.
Soc. 16 : 93-166.

BarsHaD, I. 1948. Vermiculite and its relation to biotite as revealed by base exchange
reactions, X-ray analyses, differential thermal curves and water content. Amer. Mineral.

33 : 655-678.
Berman, H. 1937. Constitution and classification of the natural silicates. Ibid. 22 : 342—
404.

Bowen, N. L. 1940. Progressive metamorphism of silicious limestone and dolomite. Journ.
Geol. 48 : 225-274.

BreiTHAvupPT, A. 1847. Volist. Handb. Min. 3 : 558-559.

CUNNINGHAM CraiG, E. H. t914. In The Geology of the country around Beauly and Inver-
ness. Geol. Surv. Scotland Mem. No. 83.

1 Monomineralic rocks or rocks with only traces of minerals other than their dominant constituent
are a feature of the Glen Urquhart area. The following have been collected as rocks (asterisks denotes
those that are strictly monomineralic) : antigorite*, biotite, calcite, diopside, gedrite, kyanite, oligoclase*,
pectolite*, quartz*, serpentine, tremolite, hornblende cf. “‘ edenite ’’, vermiculite, zoisite.

PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE 161

Dunuam, K. C. & Preacocx, M. A. 1936. Xenoliths in the Organ batholith, New Mexico.
Amer. Mineral. 21 : 312-320.

Esxora, P., 1954. A proposal for the presentation of rock analyses in ionic percentage. Ann.
Acad. Sci. Fenn., Ser. A III, 38, 1-15.

Frawn, P. T. 1951. Nomenclature of epidote rocks. Amer. Journ. Sci. 249 : 769-777.

Fett, J.S. 1906. On the petrographical characters of the inliers of Lewisian Gneiss among
the Moine Gneisses of the North of Scotland. Jn Summary of Progress for 1905. Mem.
Geol. Surv. Gt. Britain : 155-167.

Francis, G.H. 1955. Zoned hydrothermal bodies in the serpentinite mass of Glen Urquhart.

(Inverness-shire). Geol. Mag. 92 : 433-447.

1956a. The serpentinite mass in Glen Urquhart, Inverness-shire, Scotland. Amer.
Journ. Sci. 254 : 201-226.

—— 1956b. Facies boundaries in pelites at the middle grades of regional metamorphism.
Geol. Mag. 93 : 353-368.

Grec, R. P. & Letrsom, W.G. 1858. Manual of Mineralogy of Great Britain and Ireland.
Van Voorst, London.

Hatiimonp, A. F. 1943. On the graphical representation of the calciferous amphiboles.
Amer. Mineral. 28 : 65-89.

HarKER, A. 1939. Metamorphism. Methuen, London.

HEDDLE, M. F. 1878. Chapters on the mineralogy of Scotland. Tvans. Roy. Soc. Edinburgh,
28 : 246, 310, 507, 531.

—— 1879. Preliminary notice of substances which may prove to be new minerals. Min.
Mag. 3: 59-61.

1go1. The Mineralogy of Scotland. 2 vols. David Douglas, Edinburgh.

Hess, H. H. 1949. Chemical composition and optical properties of common clinopyroxenes,
part I. Amer. Mineral. 34 : 621-666.

Hey, M.H. 1954. A new review of the chlorites. Min. Mag. 30: 277-292.

Hinxman, L. W. et al. 1913. The geology of Central Ross-shire. Geol. Surv. Scotland Mem.
No. 82.

Hormes, A. 1920. The Nomenclature of Petrology. Murby, London.

Jounston, R. W. 1949. Clinozoisites from Camaderry Mountain, Co. Wicklow. Min. Mag.,
38 : 505-515.

JuurINnEN, A. 1956. Composition and properties of staurolite. Ann. Acad. Sci. Fenn.,
Ser. A., No. 47.

MErrRIAM, R. & LAUDERMILK, J. D. 1936. Two diopsides from southern California. Amer.
Mineral, 21 : 715-718.

Moir, A. et al. 1956. The limestones of Scotland, chemical analyses and petrography. Geol.
Surv., Gt. Britian. Special Reports Min. Res. 37.

Nockotps, S. R. & ALLEN, R. 1953. The Geochemistry of some igneous rock series. Geo-
chim. et Cosmochim. Acta, 4 : 105-142.

PoLDERVAART, A. 1953. Petrological calculations in metasomatic processes. Amer. Journ.
Sci. 251 : 481-504.

RAMBERG, H. 1952. The Origin of Metamorphic and Metasomatic Rocks. Chicago University
Press.

RAMMELSBERG, C. 1858. Ueber die krystallographischen und chemischen Beziehungen von
Augit und Hornblende, sowie von verwandten Mineralien. Ann. Physik u. Chemie, 103 :
435-465.

Reap, H. H. & DouBLe, I.S. 1935. On the occurrence of chondrodite in the Glenelg lime-
stone of Inverness-shire. Min. Mag. 24 : 84-80.

Rogpertson, T. et al. 1949. The limestones of Scotland. Geol. Surv. Gt. Britain Special
Reports Min. Res. 35.

Srenit, E. R. 1953. Some data on synthetic aluminous and other pyroxenes. Min. Mag.
30 : 218-226.

SmitH, J.D. 1842. In The Statistical Account of Inverness-shire. Edinburgh.

MIN, I, 5. U

162 PETROLOGICAL STUDIES IN GLEN URQUHART, INVERNESS-SHIRE

SUBRAMANIAM, A. P. 1956. Mineralogy and petrology of the Sittampundi complex, Salem
district, Madras State, India. Bull. Geol. Soc. Amer. 67 : 317-390.

Sunpius, N. 1946. The classification of the hornblendes and the solid-solution relations in
the amphibole group. Sveriges Geol. Undersék. Ser. c, 40, No. 480.

TittEy, C. E. 1920. The metamorphism of the Pre-Cambrian dolomites of southern Eyre

Peninsula, South Australia. Geol. Mag. 57 : 449-462.

1938. Aluminous pyroxenes in metamophosed limestones. Geol. Mag. 75: 81-86.

TurRNER, F. J. 1948. Mineralogical and structural evolution of the metamorphic rocks.
Geol. Soc. Amer. Mem., No. 30.

WALLACE, T. 1886. On kyanite localities in the north. Min. Mag. 6: 106-107.

WEEKS, W. F. 1956. A thermochemical study of equilibrium relations during metamorphism
of siliceous carbonate rocks. Journ. Geol. 64 : 245-270.

WriGcut, F. E. & Larsen, E.S. 1909. Diopside and its relations to calcium and magnesium
metasilicates, II. Amer. Journ. Sci. 27: 28-47.

WINCHELL, A. N. 1931. Further studies in the amphibole group. Amer. Mineral. 16 : 250—

260.
1945. Wariations in composition and properties of the calciferous amphiboles. Jbid.
30 : 27-50.

1951. Elements of Optical Mineralogy. Pt. II, 4th ed. John Wiley, New York.
ZIMANYI, K. 1893. Die Hauptbrechungsexponenten der wichtigeren gesteinsbildenen
Mineralien bei Na-Licht. Zezts. Krist. 22 : 321-358.

Pie Ay Bs

Fic. 1. Marble (paragenesis No. 14, Table II), from the largest limestone quarry, Upper
Gartally. A very pure variety with minor amounts of quartz, plagioclase, phlogopite, and
sphene set in calcite. The calcite has simple polygonal boundaries, (o1¥1) cleavages and
(o1T2) glide lamellae. Magnification 25 x, crossed nicols.

Fic. 2. Amphibolic marble (paragenesis No. 11, Table Il), from the above quarry. Fasci-
culitic and plumose groups of amphibole (called “ edenite ’ by Heddle) in limestone ; much
sieving of the amphibole by blebs of calcite, larger crystals at the top of the field. Magnification
25 X, crossed nicols.

Fic. 3. Zoisite-quartz rock associated with the limestones (paragenesis No. 18, Table I1),
Upper Gartally Farm House. Low relief, quartz; high relief zoisite. Actinolite needle of
late growth in transverse section, at the centre of the field. Magnification 25 x, ordinary
light.

Fic. 4. Zoisite-quartz—tremolite—diopside rock associated with marble (paragenesis No, 19)
Upper Gartally Farm House. Abundant late spearing growth of actinolite transecting the
ground of «-ziosite, diopside (note pyroxene cleavage), and quartz. Magnification 25 x,
ordinary light.

8

PLATE

Bull. B.M. (N.H.) Min. 1, 5

IAC INANID, 6)

Fic. 1. Large twinned clinozoisite-epidote crystal in epidosite, Torr Buidhe. The crystal
is cut normal to b, and shows crystal faces, the (100) twin plane, and the (oor) cleavage. The
distribution of zones is shown diagrammatically in Text-fig. 7a. Magnification 25 x, crossed
nicols.

Fic. 2. Prehnite in diopside rock, Sg5r Gaoithe. Note fine needles of actinolite at lower
left. Magnification 95 x, crossed nicols.

Fic. 3. Zones in metasomatic rock, Sgor Gaoithe. At the right of the field are matted
needles of actinolite, with a large crystal of sphene. To the left is a dark, ““ mossy’ zone of
clinozoisite, and to its left are large, turbid crystals of plagioclase. Magnification 20 x, plane
light.

Fic. 4. Twinned clinozoisite which has replaced a relict core of kyanite within a mantle of
“ shimmer-aggregate ’’ white mica. Elsewhere in the field are somewhat altered biotite,
muscovite, and plagioclase altering to prehnite. Magnification 12 x, crossed nicols.

E 9

ATE

Pp

5

I,

Bull. B.M. (N.H.) Min.

Pap ATE ao:

Geological map of the limestones and skarns, and their contiguous rocks at Glen Urquhart.

Bull. B.M. (N.H.) Min. 1, 5 PLATE

RE GHE NN 2)

on Psamrmnitic gneiss

Kyanite schist

= Limestone
Serpentinite
Alluvium

x Limestone guarry

ES
< LOCH MAOLACHA/N

GHAIRCHIN

Ne aes
CNS Rear

SS, a Skarns of Group J

i, eo IGE
A eee

. oN
V.SGOR GAO/THE
SCALE
1000 FT.
ed
° JOO M

To

ADLARD AND SON, LIMITED
_BARTHOLOMEW PRESS, DORKING

tr