Report

Google

This is a digital copy of a book that was preserved for general ions on library shelves before il was carefully scanned by Google as part of a project

to make the world's books discoverable online.

Il has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books

are our gateways to the past, representing a wealth of history, culture and knowledge that's often diflicult to discover.

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the

publisher to a library and finally to you.

Usage guidelines

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to
prevent abuse by commercial parlies, including placing technical restrictions on automated querying.
We also ask that you:

+ Make non-commercial use of the plus We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a b<x>k is in the public domain for users in the United States, that the work is also in the public domain for users in other

countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means il can be used in any manner
anywhere in the world. Copyright infringement liability can be quite severe.

About Google Book Search

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's hooks while helping authors ami publishers reach new audiences. You can search through I lie lull text of this book on I lie web
at |http : //books . qooqle . com/|

I

,c„.:,Ai. -..C

TRANSVAAL

MINES DEPARTMENT.

REPORT

OF THE

GEOLOGICAL SURVEY

For the Year 1908

presentee to botb Moused ot parliament b» Command of

Wis Excellency tbe Governor.

PRETORIA :
The Government Printing and Stationery Office.

1909.
r.G. 37-'09.

ltt45-2R/«/nj( -imn COPYRIGHT.

Mines Department,

Geological Survey Office,

Pretoria, 29th June, 1909.
Sir,

I have the honour to submit herewith the Annual Report of the Geological Survey for the year
ending 31st December, 1908.

It comprises a general report of the work accomplished, followed by the more detailed and
special reports of the field officers on the results of the geological investigation and mapping of the
different areas assigned to them.

I am, Sir,

Your obedient servant,

H. KYNASTON,

Director, Geological Survey.
The Acting Secretary,

Mines Department, Pretoria.

CONTENTS.

I. — Director's Report

II.— The Geology of the Country North and North-West op Potqietebsrust,
by H. Kynaeton (Director)

III. — On a Portion op the Waterberg District West of Potgietersrust, by
E. T. Mellor (Geologist]

IV. — The Geologv op the Koekbergen. in the Waterbebo District, including
Gatkop, by E. T. Mellor (Geologist)

V. — The Geology of the Country East of Potgietersri'st, including the
Marabastad Goldfield, by A. L. Hall (Geologist)

VI. — On a Portion op the Bushveld bordering the Crocodile River, a
including the Rooibero Tin-field, by W. A. Humphrey (Geologist)

VII. — The Geology op the Country between Zeerust and Zwartruggens, by
A. L. Hall (Geologist)

VIII.— The Geology of the South-Western Portion of the Marico District, by
W. A. Humphrey (Geologist)

IX. — Additions to the Museum and Library

139
161

LIST OF ILLUSTRATIONS.

PLATE

1. — Index Map of the Transvaal, showing area mapped up to end of 1908. (Frontispiece.)

II. — Figs. 1 and 2. Some characteristic forms of granite kopjes in the Older Granite Country,
north-west of Potgietersrust.

III. — A typical Older Granite kopje, north-west of Potgietersrust.

IV.— Fig. 1. A typical valley on the edge of the Waterberg Plateau; Groothoek No. 607
north of Hanglip.
Fig. 2. Hanglip, from the north-east, with the escarpment of the mam plateau in the
distance. The denudation of the alluvium, seen in the foreground, has to some extent
reproduced in miniature similar forms to those of the mountain and plateau.

V.— Fig. 1. Pramkop, an isolated kopje of massive quartzitic sandstones resting upon softer
grits ; Kranskloof No. 608, Sterk River Valley.
Fig. 2. Isolated kopje of Waterberg Sandstones, near south-eastern corner of Schilpadkraal
No. 2253.

VI. — Fig. 1. Wonderkop, near Malips Drift — faulted and highly inclined Timeball Hill
Quartzite.
Fig. 2. View of the Yzerberg (banded ironstones of the Swaziland System), north-east of
Potgietersrust.

VII. — Micro-sections of metamorphosed rocks in the Swaziland System of Mount Mare, near
Pietersburg.

VIII. — Micro-sections of metamorphosed rocks in the Swaziland System near Pietersburg and
in the Pretoria Series of the Rustenburg District.

IX.— Fig. I. Timeball Hill Shales overlooking Zeerust.
Fig. 2. Manganese ground, Doornhoek Mine.

X. — Doornhoek Lead Mine.

MAPS AND SECTIONS AT END OF REPORT.

PLATE
XI. — Fig. 1. Section "ajk" from the Waterberg Plateau near Hanglip, by Paarddrift to the
Springbok Flats near An dries Spruit.
Fig. 2. Section "lms" from the Waterberg Plateau at Jan Boven Jan No. 481, by Riet-

fontein No. 3 to the Springbok Flats near Naboomspruit.
Fig. 3. Section " PQR " from the Waterberg Plateau on H out boach kloof No. 479 to the
Sterk River on Doornhoek No. 896.

XII. — Fig. 1. Section "abcd" from the Waterberg Plateau to the Magalakwin River across

Zaaiplaats No. 2086.
Fig. 2. Section "efgh" from the Waterberg Plateau near Hanglip, by Welgelegen

No. 357 to the Magalakwin River.
Fig. 3. Diagrammatic Section across Waterberg System and Red Granite Area, showing

the most usual relationships of surface contours to Geology.

XIII. — Section across Mount Mare through Eveline Poort near the eastern end of the range.

XIV. — General section from Buffelskloof No. 320 (Waterberg), to Steenkopjes No. 101 (Witwaters-
rand), showing relation between the Bushveld Igneous Complex and sedimentary
rocks of the Transvaal and Waterberg Systems.

PLATE
XV.-Fig.

1. Section across the Timeball Hill beds south of Zeerust.
Fig. 2. Section across the Pretoria Series from near the source of the Great Marico River

to the neighbourhood of Lindley's Poort.
XVI.— Map of portion of the Waterberg District, including the Potgietersrust Tin-fielde.
XVII. — Map of portions of the Zoutpansberg, Waterberg, and Lydenburg Districts.
XVIII. — Map of portions of the Rustenburg and Waterberg Districts.
XIX. — Map of portions of the Marico, Rustenburg, and Lichtenburg Districts.
XX.— Map of portions of the Marico and Lichtenburg Districts.

FIGURES IN THE TEXT.

F
Fig. 1. Section exposed in railway cutting, 3 J miles north of Potgietersrust..

2. Section across ridge near south-western boundary of Rietfontein No. 1669, seven
miles north- north-west of Potgietersrust

3. Section " pq," across Doornhoek No. 896, near boundary of Kromkloof No. 1637.
(Portion of Section "pqr," Plate XI, Fig. 3.)

4. Section "cd" — Semi - diagrammatic section across tin-bearing zone, near and
approximately parallel to south-eastern boundary of Zaaiplaats. (Portion
of Section "abcd," Plate XII, Fig. 1.)

5. Section of ore-body (Groenvlei) ....

6. Section from the Waterberg Plateau, southwards to the Hoekbergen on Rookpoort
No. 1835

7. Section across Peak of Gatkop. .

8. Diagrammatic sections across junction of Transvaal and Waterberg Systems on
Waterval No. 965, showing rocks exposed at surface. .

9. Diagrammatic section across the Transvaal System, east of Cbunies Poort

10. Characteristic shapes of Galena Masses

11. Rough section through Witkop Mine

12. Plan of the Ottoshoop Gold Reefs

13. Section across Magazine Reef, Ottoshoop

70
153
154

I.
DIRECTOR'S REPORT

FOR THE YEAR 190a

CONTENTS.

Distribution of the Field-Work.
General Review of the Field-Work.

Economic Geology.

Summary of the Mapping—

(1) Waterberg District.

(2) Southern Zoutpansberg District.

(3) Rustenburg and Marico Districts.

Area Completed.

Publications —

(a) Official.
(6) Unofficial.

I.— DIRECTOR'S REPORT.

Distribution of the Field- Work.

During the year 1908 the field-work of the Geological Survey was carded on mainly in the
northern and western portions of the Colony. During the previous year, in addition to the work in
the southern Transvaal, the mapping had already made considerable progress towards the Pieters-
bnrg area, on the one hand, and into the western portion of the Rustenburg District, on the other
hand, and it was considered advisable that work in the former area should be continued westward
to the Fietersburg railway line, and that the field-work in the Western Transvaal should be pushed
on as far as the western boundary of the Colony, east of Mafeking.

At the same time, in view of the recent developments in connection with the occurrences of
tin-deposits in the Waterberg, it was evident that a systematic survey of the portion of this district
where such deposits were being mined and prospected, or were likely to be found in the future, would
be of considerable value and importance to a large section of the public. Accordingly the area lying
west of the Pietereburg railway between Warmbaths and Potgietersrust was also included in the
programme of mapping and examinations to be completed during the year.

In accordance with the above scheme, field-work in the Waterberg District was commenced
early in the dry season by Dr. E. T. Mellor in the area between Nylstroom and the Rooiberg, while
at the same time Dr. W. A. Humphrey was occupied in the eastern portion of the Rustenburg District,
mainly in the area west and north of the Rooiberg tin-field. During jutie, the Director visited the
more important occurrences of cassiterite and examined their geological relationships in the area
north-west of Warmbaths, thence proceeding to the Rooiberg and examining, in company with
Dr. Mellor, all the more important mining and prospecting workings for tin and other base metals
in that neighbourhood. During the remainder of the season the Director and Dr. Mellor were engaged
in the mapping of the area lying west of the Pieteraburg railway, which is represented on Plate XVI,
a considerable portion of the time being spent in an investigation of the various tin-deposits in this
area.

In the meanwhile, Mr. A. L. Hall, working westward from Malips River in the southern
Zoutpansberg District, completed the mapping of a broad belt of country as far as the Pietersburg
railway, the western boundary of his area extending from Pietersburg to Potgietersrust, thus joining
up with the work of the Director and Dr. Mellor further west. In the latter part of the season
Mr. Hall proceeded to the Rustenburg District and mapped the upper portion of the drainage basin
of the Great Marico River, while Dr. Humphrey extended the survey of thia portion of the Marico
District and a strip of the northern portion of the Lichtenburg District up to the western boundary
of the Colony.

General Review or the Field-Work.

A general review of the principal results of the field-work during the year shows considerable
progress in the detailed mapping of the country, an area of some 6,000 square miles having been
completed, and it is satisfactory to be able to report that the survey of the western portion of the
Transvaal has now effected a junction with the mapping recently completed by the Geological Com-
mission of the Cape of Good Hope in the neighbourhood of Mafeking. In connection with the work
in this part of the country completed by Mr. Hall and Dr. Humphrey, two sheet maps will shortly
be published separately. Sheet No. 5 (Zeerust) and Sheet No. 6 (Mafeking), each accompanied by
short explanations. On Sheet No. 6 the geology of the country surrounding Mafeking has been
inserted, by permission of the Geological Commission of the Cape of Good Hope, a feature which
not only adds considerably to the completeness and value of the map, but which will serve to bring
the work of the two surveys into closer touch with one another.

In the northern districts the mapping of the Transvaal System, originally commenced by
Mr. Hall along the Delagoa Bay railway, has now been completed by him up to the northern railwav

line between Potgietersrust and Pietersburg, where it has been joined up with the mapping in the
Waterberg District. In the eastern and northern Transvaal, therefore, the Black Reef, Dolomite,
and Pretoria Series have now been followed continuously round the margin of the great central basin
of the Bush veld for a distance of approximately 180 miles, and almost for the same distance westward
from the neighbourhood of Pretoria, the system showing on the whole a remarkable persistency in
its more characteristic features over the whole extent of country traversed. It is thus perhaps all
the more remarkable to find that, when followed westwards from the Pietersburg railway, the whole
system rapidly thins out and finally disappears completely, its place being taken by the intrusive
norite or basic zone of the Busbveld Igneous Complex.

Of considerable importance also, both geologically and in their economic aspect in connection
with the relative age of the cassiterite deposits, are Dr. Mellor's investigations in the area west of
Potgietersrust into the relations of the Red Granite to the two main divisions of the Waterberg
System. The mapping has shown very clearly that there is a distinct unconformity between the
Upper and Lower Waterberg ; and the main mass of the Red Granite, while clearly intrusive in the
felsites of the Lower Waterberg, was apparently not intruded into the sandstones and conglomerates
of the upper portion of the System. The conglomerates contain granite boulders, whose presence can
only be explained by their derivation from some portion of the Red Granite itself. In spite of this
evidence, however, which is fully detailed in Dr. Mellor's report, portions of the Red Granite can also
be shown to be clearly intrusive in the Upper Waterberg beds, as for instance on Gatkop, at the
western side of the Hoekbergen. The explanation of these apparently contradictory observations
doubtless lies in the fact that the general intrusive period of the Red Granite magma extended over a
considerable time, and was probably marked by several phases of activity, The general succession
of events was probably as follows :— Immediately after the outpouring of the felsitic lavas, which
represent the volcanic phase of the Lower Waterberg period, extensive intrusions took place from
the Red Granite magma, which gradually broke through the rocks of the Transvaal System and
spread itself out as a huge sheet, or laccolith, between it and the Lower Waterberg System. In
connection with the earth-movements which followed and the consequent Assuring and fracturing
of the cooler portion of the granite and the overlying strata, the deposits of cassiterite and other
base metals were formed along the contact belt and at varying distances from it. Later, during a
gradual subsidence of the whole area the sandstones and conglomerates of the Upper Waterberg
were laid down unconfonnably upon the lower beds and upon the granite, and finally further manifes-
tations of activity took place from the Red Granite magma, which invaded the Upper Waterberg beds
in the form of intrusions of granite, granophyre, quartz-porphyry, and diabase.

The mapping of the Rooiberg sedimentary series by Dr. Humphrey can scarcely be said to
have finally decided the question of their definite stratigraphical position. The whole question is
discussed in Dr. Humphrey's report on this area, and also in " The Geology of the Waterberg Tin-
fields," and it has been concluded that the field evidence so far collected favours these beds being
regarded as belonging to a low horizon in the Lower Waterberg System rather than to the Pretoria
Series.

Economic Geology.

But perhaps the most important feature of the year has been the work of a more economic
and practical bearing undertaken in connection with the field- mapping. This has included a thorough
examination of nearly all the known occurrences of cassiterite, which together make up the Waterberg
tin-fields, the extensive mining operations in which promise very soon to assume the proportions
of an important industry. A detailed description of all the mine workings and prospecting ventures
visited, together with a full account of the geology of the area in which they occur, will be found in
the special Memoir issued by the Geological Survey, entitled " The Geology of the Waterberg Tin-
fields." The results of a number of recent inspections by the Inspector of Mines for the Pretoria
District have also been included in the Memoir. Only general descriptions therefore of the tin-deposits
are given in the present report, in connection with the reports of the geologists upon those areas in
which such occurrences have been examined.

The cassiterite deposits occupy a very extensive area in the Waterberg District, and tend to
follow with remarkable persistency, a particular belt of country characterised by particular geological
conditions. This belt coincides approximately with the marginal portion of the great synclinal basin
of the Waterberg, the occurrences of cassiterite following more or less closely, or sometimes at a
distance of a few miles, the line of contact between the Red Granite and the felsites, and interbedded
shales of the Lower Waterberg System. The cassiterite is found either in the Red Granite or in

5

fit" 2149

the felsites, and occasionally in the shales. While, however, the mode of origin of these deposits
has been in most cases essentially the same, they present a most remarkable diversity of form and
mode of occurrence, and the cassiterite itself varies from coarse black crystals, sometimes nearly an
inch in diameter, to minute grains that can only be distinguished either on panning or under the
microscope. The stanniferous Tielt of country is now known to extend, with only a few short inter-
ruptions, from the neighbourhood of the junction of the Sterk and Magalakwin Rivers, some thirty
miles north-west of Potgietersrust, to the Rooiberg. In the immediate neighbourhood of Nylstroom,
however, the unconformably overlying sandstones of the Upper Waterberg System extend eastwards
from the main area of the formation towards the Springbok Flats, and appear to conceal the continua-
tion of the belt between the Zwagershoek district and Rhenosterpoort No. 2145. The whole belt
may be thus regarded as divided into a northern and a southern section, the former extending from
Appingadam to Doornhoek and Kroomkloof, and probably also for a certain distance westward
from the latter farm, while the southern section may be followed from near Warmbaths north-
westwards to Rhenosterpoort, thence westwards along the southern slopes of the Hoekb-ergen, ■ and
then south-west to the Rooiberg. On both sections mining and prospecting operations are in pro-
gress with most encouraging results, and already three mines on the northern section, namely
Groenfontein, Zaaiplaats, and Doornhoek, and two on the southern section, Rooiberg and Weynek,
shortly to be followed by Elandsfontein No. 1782, are equipped with their own reduction plants and
turning out a steady supply of concentrates, averaging in the case of the Potgietersrust mines alone
some 200 tons per month.

In addition to their special characters, the general geological relations of the tin-deposits
have been studied along the whole line, and important evidence has been collected bearing upon the
geological age of the period of mineralisation. Not only has it been found that the sandstones and
conglomerates of the Upper Waterberg are unconformable to the felsites and shales of the Lower
Waterberg, and sometimes overlap them, but apparently also the Upper Waterberg sediments were
deposited subsequently to the intrusion and consolidation of a large portion of the Red Granite.
Thus, boulders of the Red Granite are found in several localities in the conglomerates near the base
of the Upper Waterberg, and, in addition to these, boulders of tourmalinised rocks, as first observed
by Merensky, also occur, identical in character to the tourmaline-bearing rocks so characteristic of
the tin-fields. This and other evidence which has been collected appear to indicate that at least the
main period of mineralisation, which was responsible for the formation of the tin, took place before
the deposition of the Upper Waterberg sandstones and conglomerates. So far we have no definite
information of cassiterite deposits occurring actually within any rocks of recognised Upper Water-
berg age, though it is quite possible that they do occur there, and may yet be found, the evidence
so far collected by no means precluding the possibility of the occurrence of tin in other rocks than
the Red Granite and the Lower Waterberg strata.

In addition to the detailed examination of the tin-fields, the lead and zinc deposits of the
Marico District were investigated by Dr. Humphrey. He reports that the ores have been deposited
along certain lines which follow the strike of the Dolomite and are associated with its upper portion,
not far from the base of the Pretoria Series. With perhaps the exception of Witkop, they may be
considered as replacement deposits, the lead occurring as large irregular masses in manganiferous
earth, which can be shown to have taken the place of the Dolomite. Sometimes it occurs in veins
in unaltered dolomite, and masses of unaltered dolomite occur throughout the manganese matrix.
At Witkop the zinc-blende is found on the walls of an inverted funnel-shaped body, consisting mainly
of calcite and dolomite. The outcrop is roughly circular, the circle increasing in diameter in depth.
A large chamber of ore has been found within this funnel, connected by leaders with the ore-body
on tile walls.

The Mahnani Goldfield, which has been for some time in a practically abandoned condition,
was also examined by Dr. Humphrey who gives a description of the various reefs. These occur as
fissure veins running parallel to the strike of the surrounding Dolomite.

In the Pietersburg area the Marabastad Goldfield was visited and surveyed by Mr. Hall. The
greater portion of this area is occupied by the schist belt of the Mount Mare range, which is assigned
to the Swaziland System. It contains several auriferous quartz-reefs, which sometimes occur parallel
to the bedding-planes of the country rock, and sometimes cut across them. Alluvial gold has also
been worked at several localities in the same district.

Summary of the Mapping.
A short summary may now be given of the mapping completed by the staff according to the
different districts over which it was distributed.

(1) Waterberg District.

In the portion of this district lying west and south-west of Potgietersrust, Dr. E. T. Mellor
mapped an area of 1,025 square miles, and traced 660 miles of geological boundary lines, including
also a very intricate and detailed piece of work on Gatkop on the Rustenburg boundary near the
western end of the Hoekbergen range. The mapping comprised mainly the satisfactory interpretation
of the structure of a very large section of the great synclinal basin of the Waterberg, which is illustrated
very clearly in the sections on Plates XI and XII. Special attention was also given to the detailed
and often complicated structure and relations of the outer zones flanking the main plateau, as will
be seen from Dr. Mellor's report on the geology of the Hoekbergen. The more important results,
both, economical and geological, from this area have already been pointed out.

Adjoining Dr. Mellor's ground to the north, and extending to the south-western portion of
the Zoutpansberg District, the Director completed an area of 570 square miles, consisting for the
most part of rocks belonging to the Bushveld Igneous Complex, the Transvaal System, and the Older
Granite. The most interesting feature in this area is the final thinning out and disappearance of the
Transvaal System, north-west of Potgietersrust, and the intense contact metamorphism which
accompanies it.

(2) Southern Zoutpansberg District.

Mr. Hall's work in this district and in small adjoining portions of the Waterberg and Lyden-
burg Districts included the mapping of 1,690 square miles, and the tracing of 581 miles of boundary
lines, the area lying mainly south of Pietersburg between Potgietersrust and the Malips River. The
Transvaal System was mapped in a westerly direction as far as the Pietersburg railway, and the
main stratigraphical horizons were found to be remarkably persistent, though with much variation
in detail, and with a gradual decrease in thickness from east to west. Metamorphic phenomena
were found to be still very pronounced, while near Potgietersrust the norite breaks successively across
the different members of the Pretoria Series until it comes into contact with the Dolomite.

An interesting horizon of interbedded dolomitic marls, exposed immediately east of Potgieters-
rust, was found to be interbedded between the upper and middle quartzites of the Pretoria Series,
while near Chunies Poort a conglomerate occurs close to the base of the series.

The Bushveld Plutonic Complex occupies a large area south of M'Phatlele's Location, and
the remarkable highly acid quartzose facies, first described in the Annual Report for 1904, was found
to be very extensive near the contact with the norite. An interesting feature of the norite is the
occurrence in several localities of banded gneissic varieties, more or less resembling in outward appear-
ance some of the older schists.

The Swaziland System was mapped in a broad belt composing the Mount Mare range near
Marabastad and extending south-westwards as far as the farm Uitloop, north of Potgietersrust. The
detailed stratigraphy of these interesting rocks was worked out by Mr. Hall, and good evidence was
collected of the intrusive relation to them of the so-called Older Granite. Mr. Hall has also made a
detailed study of the ottrelite and andalusite schists of this series.

There is probably no doubt that the Mount Mare series is the equivalent of the Moodies Series
of Barberton. Whether these beds, however, can be eventually correlated with the Lower Witwaters-
rand Series, as was formerly supposed by Molengraaff, must be left to future investigations to decide.
Owing to the very doubtful nature of the evidence at present available, it has been considered advisable
to retain them in the Swaziland System.

(3) Rustenburg and Marico Districts.

In these districts an area of approximately 2,800 square miles was mapped by Mr. Hall and
Dr. Humphrey, the mapping being extended from the south-western corner of the Rustenburg District
to the western border of the Colony, thus completing the Zeerust and Mafeking Sheets.

In the more northern portion of this area, adjoining the Waterberg District, Dr. Humphrey
mapped an exceedingly complicated area of rocks of the Transvaal System. The rocks are much
faulted and folded, show faulted relations to the Red Granite,- and nowhere east of the Crocodile
River do they show the normal relations to the Bushveld Plutonic Complex.

The western portion of the area occupied by the Rooiberg quartzites was also mapped, and
the relations of these rocks investigated to the granite of the Boshofbergen on the one hand, and
to the Felsites of the Elandsberg on the other hand, the former being found to be intrusive and the
latter overlying the quartzites apparently conformably. On the north side of the Elandsberg the
felsites are faulted against a belt of Red Granite which separates them from the Pretoria Series of the
Gatkop range.

In the Marico and south-western Rustenburg District a very large proportion of the area
mapped consists of the Pretoria Series and the Dolomite, the extensive surface development to the
south of Zeerust being mainly due to the very gentle inclination of the beds. The Timeball Hill
shales at the same time increase in thickness towards the west and two additional horizons are
developed below them, consisting of quartzites and banded shales. The Magaliesberg beds, on the
other hand, show a marked decrease in thickness.

Contact metamorphism in the Pretoria Series is again strongly marked, the contact zone
widens from east to west, until near Zeerust it extends almost to the base of the Pretoria Series ;
there is, however, very little evidence of the effects of great pressure. Some interesting information
was collected in connection with the distribution and genetic conditions of the mineral ottrehte,
occurring in the altered shales.

In the south-western comer of the Marico District volcanic rocks belonging to the Venters-
dorp System appear from beneath the unconformably overlying Black Reef quartzites. They consist
mainly of basic amygdaloids, in which occurs a zone of rhyofites, tuffs, and breccias. These rocks
have been satisfactorily connected on the map with the development of the same system, which
occupies the area further west in the neighbourhood of Mafeking.

Area Completed.
The following table shows the total areas completed by the Director and geologists during
the year, together with the number of linear miles of geological boundary lines traced upon the
field-maps :■

Square

Boundary

Mile*

Surrey ed.

Traced.

H. Kynaston (Director)

,. 570

242

E. T. Mellor (Geologist)

.. 1,025

660

, 2,488

1,051

W. A. Humphrey (Geologist)

.. 2,006

1,290

Totals .. .. 6,089 3,243

The total area exceeds that completed during 1907 by 1,174 square miles, and it is of interest
to note that this brings the total area mapped by the Geological Survey, since the commencement
of its present organization in 1903, up to a little over 24,000 square miles.

The number of miles travelled during the year in the course of the mapping and traversing
was as follows : —

By Wagon. Driviiig, and
Walking.

H. Kynaston 520 1,758

E. T. Mellor 531 2,387

A. L. Hall 672 2,018

W. A. Humphrey 480 2,652

Totals .. .. 2,203 8,815

Publication.
(a) Official.

1. Maps. — The following maps were published during the year on the scale of 2 '35 miles to
1 inch, and illustrated by horizontal sections : —

Portions of the Lydenburg and Zoutpansberg Districts, including the Haenertsburg Gold-
fields (geologically surveyed by A. L. Hall).
Portions of the Potchefstroom, Witwatersrand, and Rustenburg Districts, lying mainly to
the north-west of Krugersdorp (geologically surveyed by A. L. Hall and W. A.
Humphrey).
The following sheet of the Geological Map of the Transvaal was also issued : —
Sheet No. 4 (Rustenburg).
Sheet No. 5 (Zeerust) and Sheet No. 6 (Mafeking) are in preparation.

2, Memoirs. — " The Geology of Rustenburg and the Surrounding District," an Explanation
of Sheet No. 4, by W. A. Humphrey.

(b) Unofficial.

The following unofficial contributions by members of the staff to scientific societies, etc., were
published during tie year : —

A. L. Hall "On Contact Metamorphism in the Pretoria Series of the

Lydenburg and Zoutpansberg Districts." Trans. Geol.
Boa, S.A., Vol. XI, pp. 1-24.
„ . . . . " Note on Contemporaneous Igneous Kocks in the Pretoria

Series of the Lydenburg and Zoutpansberg Districts."
Trans. Geol. Soc., S.A., Vol. XI., pp. 46-47.
A. L. Hall and W. A. Humphrey " On the occurrence of Cbromite Deposits along the southern
and eastern margins of the Bushveld Plutonic Complex."
Trans. Geol. Soc., S.A., Vol. XI, pp. 69-77.

H. KYNASTON,
Pretoria, 28th June, 1909. Director, Geological Survey.

II.

THE GEOLOGY OF THE COUNTRY

NORTH AND NORTH-WEST

OF POTGIETERSRUST.

By H. Kynaston {Director).

CONTENTS.

1. Area.

2. Physical Features and General Geological Structure,

3. The Older Granite.

4. The Transvaal System,

5. Contact Metamorphism in the Dolomite and Pretoria Series.

C. The Bushveld Plutonic Complex —
The Norite.
The Red Granite.
Intrusive Sheets and Dykes.

7. Economic Geology.

II.— THE GEOLOGY OF THE COUNTRY NORTH AND NORTH-WEST
OF POTGIETERSRUST.

By H. Kynaston (Director).

1. Abba.

The area described in the following report lies mainly to the north and north-west of Potgietera-
rust and includes a part of the north-eastern portion of the Waterberg District and a small portion
of the southern Zoutpansberg. On the west it adjoins the area surveyed by Mr. B. T. Mellor, and
is continuous on the east with the country mapped by Mr. A. L. Hall, lying mainly to the east of
Potgietersrust. The area is bounded on the west by the crest line of the granite ranges which separate
the Magalakwin Valley from that of the Sterk River ; on the east by the Pietersburg railway from
Potgietersrust to Luhsklip, and thence by a line running northward through Machachaan's Location ;
and on the north by a line drawn east and west through the southern portion of Matala's Location.

The mapping covers approximately 570 square miles and includes 242 miles of geological
boundary lines. The greater portion of this area is represented in the geological map on Plate
XVI, while a small portion lying north of Potgietersrust is inoluded in Plate XVII.

2. Physical Features and General Geological Structure.

The physical features are of a varied but strongly differentiated character and directly
dependent upon the geological structure ; following which they divide the area roughly into three
main belts with a general north-west and south-easterly trend.

The first of these is formed by the conspicuous range of hills which overlooks the Magalakwin
River from the south-west, and extends from Moorddrift north-westwards to near the junction of
the Sterk and Magalakwin Rivers. It is formed by the Bushveld Granite, which probably here owes
its unusual prominence mainly to the part which it plays in the general structure of this part of the
country. It forms here the outer margin of the great Waterberg basin, and dips at an angle of some
12 to 15 degrees in a general south-westerly direction beneath the felsites of the Lower Waterberg,
which in their turn are covered by the sandstones and conglomerates of the upper portion of the
system, forming the great Palala Plateau.

The general inclination which thus characterises the outer granite belt, combined with
the influence of the overlying felsites in preserving it from denudation, and the more easily dis-
integrated character of the norites which bound it on the north-east have probably been the main
factors in determining its bold escarpment- like form. The highest and most clearly defined portion
of the range rises directly west of the Magalakwin River along the western boundaries of the farms
Lisbon and Blinkwater ; north-westward from here, however, the range becomes more broken into
minor spurs and ridges, and the elevation of the crest line gradually falls, the range finally terminating
as a noteworthy feature among the scattered groups of kopjes about the junction of the Sterk and
Magalakwin Valleys. A characteristic type of rough hilly country still, however, marks the trend
of the Red Granite belt, which now takes a somewhat more northerly direction, crossing the western
portion of Hendrik Maaibi's Location.

In the south-eastern portion of the range the north-eastern face forms a steep, almost abrupt,
descent to the level of the Magalakwin. Further down the valley, however, this character becomes
less marked, the slopes tending rather to form step-like features separated by comparatively level
platforms. This is due partly to modifications in the granite itself and partly to the occurrence of
diabase intrusions, and is well illustrated 1 by the prominent minor ridge, which extends along the
left bank of the Magalakwin from the north corner of Blinkwater to Groenfantein. In addition to
showing these step-like features the northern slopes of the range are cut into prominent spurs
separated by narrow valleys, which on Groenfontein, Mooihoek, and Delagoa tend to broaden out
and form gentle sandy slopes between the spurs, generally ending towards the crest line of the range
in a narrow gap or nek. Good springs frequently rise near the heads of these valleys; the water,
however, from these is usually lost in the sand before it has proceeded very far from ite source.

12

The second belt of country is formed by the norite and constitutes the Magalakwin flats.
These form a somewhat monotonous stretch of country, averaging about six miles in width, which
rises very gently from the level of the Magalakwin River towards the north-east, and closely resembles
the norite belt of country found to the north of Pretoria. It carries the same type of ragged sharp-
crested kopjes in lines and groups, running parallel to the general trend of the igneous series, while
the usual black and red soils cover the more gently sloping areas, which are occupied by a succession
of large and populous native villages extending continuously for some thirty miles north-westward
from Potgietersrust. From the northern portion of Hans Maaibi's Location, the norite belt, like
that of the Red Granite, takes a more northerly direction towards the west corner of Matala's.

The remaining portion of the area is dominated almost entirely by the type of scenery
characteristic of the Older Granite, with the exception of a narrow belt of low ridges, lying between
it and the norite, and extending north-westwards from Uitloop, north of Potgietersrust, to the western
boundary of Vaalkop No. 1439. This narrow belt of low wooded hills is the surviving remnant of
the Transvaal System, which plays such an important part in the physical features of the country
further east. Traced into the present area, however, the system has shrunk and dwindled to a narrow
zone of strata, and forms but an insignificant feature in the general landscape. North-west of
Vaalkop, at which point only the Dolomite is represented, no further trace of the Transvaal System
has been found, the norite zone being in direct contact with the Older Granite.

The Older Granite country forms a marked contrast to that occupied by the norite, and is
perhaps here characterised by a more hilly, or even mountainous, configuration than is usually found
in other parts of the country occupied by it. This is especially noticeable to the north of Uitloop,
where a considerable area is occupied entirely by conspicuous groups of bare lofty hills, separated
by narrow valleys. Further north the country is more open, and constitutes an undulating sandy
area, only broken here and there by conspicuous groups of granite kopjes. In Matala's Location
the country again rises into a group of fine hills, the highest rising to 5,814 feet above sea-level and
serving as a prominent landmark over a large part of the country. The forms of many of these
kopjes are often very characteristic, many consisting simply of bare dome-shaped masses of granite,
recalling the well-known granite hills of the Paarl in Cape Colony. Other forms are less flattened
at the summit and more of the " Spitz-kop " type. The photographs on Plates II and III show
some characteristic forms of these kopjes as seen in the striking group of hills on Zwartfontein No.
1174. A feature, which is more or less marked in the Older Granite country, U a tendency for the
hills and kopjes to be arranged roughly in line, and so to form elevated belts or ridges of granite, with
a general north-east and south-west trend. The diabase dykes which traverse the granite also show
a similar direction. This naturally contrasts with the main features of the area occupied by the
younger rocks further south, whose general trend is north-west and south-east.

Water Supply. — All the small streams in the area drain into the Magalakwin which flows
northward to the Limpopo. In the Older Granite country permanent flowsof water are usually only
found in the higher portions of these streams, the water being usually absorbed by the sandy bed
of the. stream during its comparatively long journey to the Magalakwin. Crossing the Magalakwin
Flats these streams are often quite dry during the winter months, though some of the larger
show a moderate flow along the lower part of their course, the water being brought to the surface by
outcropping bands of norite. Very little water descends the northern slopes of the Red Granite
range, except after heavy rains. The Magalakwin itself shows a fairly good permanent flow, except
perhaps after very dry seasons, below its junction with the Dorps River, which rises in the Dolomite
near Potgietersrust and forms the main supply for that town. Above this point it is mainly a " vlei "
with detached swampy pools.

The Dolomite, although much thinner than in the area further east, has not altogether lost
its reputation as a water-carrying formation, a good supply having been obtained from it on Uitloop
at a depth of 137 feet.

3. The Older Granite.

Very little need be added to what has already been said in previous reports regarding this
formation. On the whole the rock shows very uniform characters over large areas, generally con-
sisting of a greyish acid biotite-granite, occasionally with porphyritic felspars. It is usually very
massive with well marked vertical and horizontal jointing, and foliation is only occasionally seen,
being perhaps more usually observed in the undulating country immediately south of Matala's
Location. Coarse quartz and pegmatite veins are very common.

A Typical Older Granite Kopje, north-west of Potgiotersrust

13

The only well marked modification found in the Older Granite area constitutes a mass of coarse
reddish and pink granite, well developed on the north side of the big kopje, situated on the south-
western boundary of Bloemhof No. 1672, north- north-east of Potgietersrust. This rock extends for some
distance from here towards the north-east, but no sharp line could be found between it and the normal
grey type. It consists of much red felspar, with small patches of chloritic decomposition products,
probably representing biotite, and grains of milky blue quartz. Apart from the somewhat peculiar
appearance of the quartz grains, the rock closely resembles some varieties of the Red Granite of the
Bushveld. It belongs, however, essentially to the Older Granite, though it may possibly represent
a slightly later phase of the general magma.

A strongly foliated variety, constituting a fine-grained gneiss, is exposed near the base of the
Dolomite, on the southern portion of Rietfontein No. 1669. It shows fairly large porphryitic
felspars, sometimes microcline, forming "augen" in a fine-grained well foliated groundmass of quartz
and felspar [A 622]. •

4. The Transvaal System.

The most interesting feature of the Transvaal System in the present area is the rapid thinning
out and final disappearance of its different members as they are followed in a north-westerly direction
from the railway line at Uitloop, about four miles north of Potgietersrust. This can be better
appreciated when one considers the great development of the system, as described by Mr. A. L. Hall,
to the east of Potgietersrust. It is true that a decided decrease in thickness is noticeable as the
different members of the system are followed from the Haenertsburg gold-fields and the Stryd-
poort Range into the area east and south-east of Potgietersrust, but west of the railway line the
process is still more marked in its effects, especially with regard to the Pretoria Series, which is now
reduced to a few insignificant outcrops of quartzite and highly metamorphosed shales. The whole
system forms a belt not more than eleven miles in length, and rarely exceeding a mile in width. The
dips are usually high, frequently up to nearly 60 degrees, and there are frequent local disturbances,
but no faulting or folding on any considerable scale.

The Black Reef Series shows an extremely insignificant development west of the railway line,
After showing a rapid thinning out further east between Weenen and Uitloop, it dies out on the
summit of a small kopje, about a mile east of Uitloop Siding. West of this point its place appears
to be taken by the intrusion of massive pyroxenite, connected with the norite, seen on the west side
of the railway on Uitloop. It appears again, however, in its normal position on the southern portion
of the big granite ridge which forms the north-western boundary of Uitloop. It constitutes, however,
but a small patch of quartzites dipping steeply off the granite and beneath the Dolomite. Immediately
north-west of this ridge the formation is again missing, the Dolomite resting partly on the Older
Granite and partly on norite as far as the southern corner of Biltongfontein, where a narrow zone
of quartzite is again exposed at its base, and extends in a general westerly direction across the
northern portion of Tweefontein. This quartzite closely resembles that of Uitloop and undoubtedly
belongs to the Black Reef. It finally disappears, however, a short distance east of the north-western
boundary of Tweefontein. The thickness of these small patches of Black Reef is quite insignificant,
probably not more than one hundred feet on Uitloop and less than fifty feet on Tweefontein. It is
usually a fairly coarse fehpathic and more or less micaceous quartzite, sometimes with pale
greenish and pinkish bands, and occasionally mottled with darker patches.

The Dolomite is the most persistent member of the system and may be followed continuously
from the Uitloop Lime Works to Vaalkop. Its average dip is probably nearly 45 degrees,
but it apparently becomes somewhat flatter on Tweefontein and Vaalkop. On the south-western
boundary of Vaalkop, while still maintaining its average width of outcrop, it gradually passes into
a mixed zone consisting partly of highly metamorphosed Dolomite and partly of material of igneous
origin belonging to the norite, and finally west of the Sand Sloot, on the eastern portion of Hans
Masibi's Location, its place is entirely taken by the norite, which now stretches uninterruptedly
from the Red Granite on the south-west to the Older Granite on the north-east. Throughout its
entire extent to the west of the railway, the Dolomite shows a high degree of metamorphism of the
contact type, and is in places, as already pointed out, cut by numerous dykes of granite, aplite, gabbro,
and diabase. The general character of the metamorphism will be described below.

Thin bands of altered black shale, about two or three feet in thickness, may be seen inter-
bedded with the Dolomite on Uitloop.

Immediately overlying the Dolomite, and apparently more closely associated with it than
with the Pretoria Series, is a narrow zone of banded magnetite quartzite. This is well exposed in the

SECTION EXPOSED IN RAILWAY CUTTING 3* MILES NORTH

OF POTGIETERSRUST.

s.

. ./ I,

N.

Fig. I.

railway cutting between Potgietersrust and Uitloop, which is cut through a low ridge formed by the
outcrop of the quartzites. These banded rocks here show a thickness of about sixty feet and a dip
of about 50 degrees S.S.W. They consist of alternating bands of quartzite and magnetite, fre-
quently altered into limonite, the individual bands varying from a mere film up to layers five or six
inches in thickness. The magnetite occurs as aggregates of small crystals and grains either filling
the interstices between the quartz grains or altogether replacing them. This band extends a short
distance to the east of the railway and then appears to die out. Probably it may be correlated with
one of the ferruginous bands in the Dolomite, described by Mr. Hall, from the area still further east.
Immediately west of the railway it is considerably broken up and difficult to follow clearly
at the surface. It is well exposed, however, a short distance north of the west corner beacon of
Uitloop, and forms a well marked ridge crossing the southern portion of Rietfontein (see Fig. 2),

SECTION ACROSS RIDGE NEAR SW. BOUNDARY OF RIETFONTEIN No. 1669.
7 MILES NNW. OF POTGIETERSRUST

13

thence trending in a slightly more northerly direction, and again forming a prominent feature at
the east corner beacon of Tweefontein. West of this point it is not well exposed and only appears
again for a short distance near the western boundary of Tweefontein, where it forms a slight feature
overlooking the norite on the south-west. Beyond this point no further trace of any outcrop has
been observed.

The Pretoria Series, considered as a distinct formation, and compared with its more normal
development in other parts of the country, can scarcely be said to be present at all in the present
area to the west of the railway. While in the area further east, the three main quartzite horizons
of the aeries with their associated shales are still distinctly recognisable, although much faulted and
folded, when one crosses the railway line the entire series is only represented by small disconnected
patches and bands of quartzite and bornfels, surrounded by basic and ultra-basic rocks.

Immediately north of Potgietersrust not only has a good deal of faulting taken place, but
the norite encroaches gradually upon the sedimentary rocks, cuts across the Pretoria Series, and
practically occupies the place of nearly the whole of the series to the west of the railway. At the
same time there has probably also been a considerable and rapid thinning out of the various com-
ponent beds.

The surviving remnants of the series found to the west of the railway apparently belong to
the lower horizons of the series, though their broken and isolated condition renders it impossible to
connect them directly with any of the known horizons further east.

The dolomitic marl which lies between the Daspoort and Magaliesberg Quartzites, and which
forms the hill east of Potgietersrust Station, on which the Transvaal Police camp is situated
crosses the railway only for a very short distance, and is then cut off by the norite. Close to Potgietersrust
Station some patches of black compact hornfels occur between it and the norite. These are the
only members of the series on the west side of the railway whose position is definitely known.
A short distance further north, one or two small outcrops of quartzite and altered shale may be seen
close to where the railway crosses the Dorps River ; these, however, are too indefinite in their
character and behaviour to be assigned to any definite position, the quartzite perhaps representing
the Middle or Daspoort Quartzite. Further north, about 2J miles north of the station, a larger mass of
pale somewhat felspathic quartzite has been quarried close to the west side of the railway, whence
it may be traced as a narrow broken band for about a mile in a north-westerly direction. Further
in the same direction, however, it forms a fairly conspicuous kopje, on which is situated the south-
west beacon of Uitloop. It is well exposed in the small spruit to the west of this kopje, but then
thins out rapidly and disappears in the norite. This quartzite is a fairly coarse pale grey or
yellowish rock, and may possibly represent the westerly continuation of the Lower or Timeball
Hill Quartzite. Between it and the banded magnetite-quartzites, which overlie the Dolomite, the
place of the missing lower beds of the Pretoria Series is now taken by pyroxenites and serpentines,
with the exception of a rare but insignificant outcrop of hornfels.

A few miles further to the north-west in the southern corner of Rietfontein a similar quartzite
makes its appearance in narrow broken outcrops for a distance of nearly a mile, with a few small
exposures of hornfels on either side of it (see Fig. 2). This, however, represents the very last appear-
ance of any recognisable portion of the Pretoria Series. Apparently owing to a process of fairly
rapid thinning out and the breaking up and enveloping action on the part of the norite, it has been
reduced to insignificant fragments and has finally disappeared. It has fared worse than the Dolomite,
owing apparently to its being nearer to the intrusive mass of the norite, and to its being composed of
alternations of dissimilar beds, rendering it less resistant to the invasion of the igneous magma and
more easily dislocated and broken. The more massive nature and homogeneous character of the
Dolomite, on the other hand, would render it more yielding and pliant, and so perhaps enable it to
maintain itself longer as a continuous and persistent body, being only finally overwhelmed when
deprived of the protecting influence above and below of the Pretoria Series and Black Reef Quartzites
respectively.

5. Contact Metamorphism op the Dolomite- and Pretoria Series.

Before proceeding to a description of the behaviour and character of the igneous rocks, which
have been intruded into the Transvaal System, it will be convenient to give a short account of
the general nature of the metamorphic phenomena which have been the direct result of their
intrusion. To do justice to this subject, however, a more prolonged and detailed study of the area
would be necessary than it has been possible in the present case to give, and moreover the somewhat
fragmentary nature of the outcrops and the complicated character of the various igneous intrusions

lias rendered the resulting metamorphic phenomena decidedly complex, and it is almost impossible
to assign particular results to the particular intrusions responsible for them. The contact
metamorphism observed has been caused not only by the enormous mass of the norite and its
associated ultra-basic rocks, but also by the numerous dykes of granite and allied rocks which traverse
the sedimentary series.

The metamorphism of the Dolomite is general throughout its extent, but shows more intense
phases locally, depending partly on the varying composition of different beds or sores, and partly
on the proximity to various intrusions. The petrological characters of the formation therefore vary
considerably, almost every degree of alteration being present from an only slightly modified dolomite
to a rock, the original character of which is only recognisable from its stratigraphical relations.
Perhaps the more common type of altered dolomite, representing the purer facies of the formation,
is a fairly coarse holocrystalline rock, closely resembling a marble. Another very characteristic
type ib a grey or brownish crystalline dolomite with numerous minute reddish brown specks, forming
small hollows on the weathered surface. This variety is well seen at the Uitloop Lime Works and at
various localities further north-west, the small specks apparently representing decomposed serpentine
[A 585 b].

Among the more altered varieties the most characteristic minerals are olivine (forsterite),
usually more or less altered into serpentine ; brown mica, probably phlogopite ; pyroxene (diopside),
and garnet. No tremolite, actinolite, or wollastonite has been observed in this area. The brown
mica is sometimes scattered evenly in small flakes throughout the rock, or occurs as coarse plates,
up to half an inch in diameter, frequently crowded into clusters. It is usually associated with the
contact with the Bmall granite dykes. A specimen of a mica-bearing rock from near the contact
with a fine-grained granitic intrusion a short distance east of the common corner beacon of Twee-
fontein, Biltongfontein, and Holmesleigh consists [A 585 a] of crystalline dolomite and calcite, with
numerous rounded grains of olivine (forsterite), more or less altered into serpentine, with fairly large
scattered crystals of mica. The latter show the characteristic sieve structure, enclosing small irregular
grains of dolomite. Varieties, altered by granite dykes, and carrying bunches of coarse mica, serpen-
tine, and garnet, are exposed on the northern portion of Tweefontein close to the base of the Dolomite
[A 492]. The mica in these rocks is usually of a pale bronze-brown colour, and only shows slight
dichroism under the microscope. It is probably associated with certain zones in the Dolomite and
is not always present at the contact with the granite dykes, in which case the altered rocks are more
usually mottled and streaked with pale greyish green, and consist of an intimate mixture of dolomite,
calcite, and serpentine, the latter appearing as if replacing the dolomite [A 490, 586]. Other
varieties show crystalline dolomite, and calcite with numerous small crystals of pyroxene, probably
diopside, and patches of serpentine, and limonite [A 608].

The widespread occurrence of forsterite in these rocks together with serpentine derived from
it, and the presence also of pyroxene recalls the contact metamorphism of the Cambrian Dolomites
of Assynt in the north-west Highlands of Scotland, described by Dr. Teall.* The formation of the
forsterite involves a certain amount of dedolomitisation, and in consequence is usually closely
associated with calcite, while the formation of pyroxene requires the presence of silica in the original
rock. This is abundantly present in the Dolomites here described in the form of the characteristic
chert bands, the pyroxene being formed from the Dolomite by the substitution of silica for carbon
dioxide.

The alteration of the shales, associated with the Dolomite and the surviving remnants of the
Pretoria Quartzites, has also been of an intense nature, but owing to the fragmentary and isolated
nature of the occurrences, it is not possible in the present area to trace the altered rocks directly into
their unaltered representatives.

The principal localities where horn f els or contact-fels, representing highly altered shales of
the Pretoria Series, occurs are a short distance south and south-west of Potgietersrust Station ;
associated with the Dolomite, a short distance west of the railway cutting, 3J miles north of Potgieters-
rust; in the south corner of Rietfontein both above and below the*narrow band of white' quartzite
(see Fig. 2) ; and in the north-west portion of Rietfontein on the west side of the banded magnetite-
quartzite ridge. These rocks often closely resemble fine-grained basic igneous rocks. They are
dark grey or black, very compact, showing little or no sign of original bedding, and often with fairly
conspicuous crystals of secondary minerals scattered throughout the dense matrix, so as to resemble
the porphyritic phenocrysts of an igneous rock. Thus the hornfels which occurs near Potgietersrust

'See "The (ieologi.nl Struct lire of the- N'.irtli-Wpirt Highlands of Scottoptl." Mem. Oeol. Survey, Great Britain, 1907, p. 453,

17

Station from its general appearance might be mistaken for a modification of the norite, with which
it is in contact and by the intrusion of which it has no doubt been metamorphosed from an original
ahaly condition. Its position is apparently immediately above the dolomitic a>ne which overlies
the Daspoort qu&rtzite. Under the microscope [A 630], the real nature of the rock is at once
apparent. It shows some large plates of cordierite and irregular flakes of biotite and grains of
pyroxene, embedded in a fine quartz-felspar mosaic. Both the cordierite and the biotite show the
characteristic sieve structure, enclosing numerous granules of quartz and felspar. The presence of
the pyroxene is somewhat unusual in this class of rock, and may perhaps be explained as having
been derived from the adjacent basic igneous magma.

The hornfels associated with the Dolomite near the railway cutting, north of Potgieterarust,
is of a somewhat similar character to the above and also has a general igneous appearance. It
consists essentially of large plates of cordierite with some irregular flakes of biotite and the usual
crystalline-granular aggregates of quartz and felspar [A 574]. The cordierite, however, is present
in much greater quantity than in the Potgietersrust variety, and almost constitutes a matrix to the
quartz and felspar grains. Both the cordierite and the biotite also show good sieve structure. The
rock may be termed a cordierite-contact-fela.

A specimen from the south corner of Rietfontein, in contact with the basic rocks immediately
above the banded magnetite-quartzite, is a black compact rock with scattered crystals of hornblende
and biotite recognisable by the unaided eye. Under the microscope [A 624], the hornblende and
biotite appear in large irregular patches with well developed sieve structure. Cordierite is also seen
to be present together with a fine-grained quartz-felspar mosaic. The hornblende is of a somewhat
peculiar variety, showing a pleochroism changing from pale brown to a decidedly bluish green,

A closely similar type of contact-fels occurs again in the same position in the more north-
westerly portion of the same farm, and is apparently on approximately the same horizon, though
there is no direct connection between the two, both outcrops being surrounded by the basic igneous
rocks of the norite belt. Here again we find large irregular plates of the same variety of hornblende
[A 607J, surrounded by a mosaic of quartz, felspar, and brown biotite. Cordierite, however, in this
instance is apparently not present. These rocks are probably the highly metamorphosed repre-
sentatives of portions of the shales which normally underlie the lowest or Timeball Hill Qu&rtzite of
the Pretoria Series.

6. The Bushveld Plutonic Complex,

The norite, as already pointed out, forms a well-marked belt, about six miles in width extend-
ing from Potgieterarust to Hendrik Masibi's Location. Its contact with the Transvaal System is
somewhat complicated, and it usually shows more basic modifications, such as pyroxenites and
serpentines as one approaches the contact. This is the reverse to what is seen along the southern
margin of the main Bushveld mass, where the norite usually becomes more acid, passing into
diabases, in the neighbourhood of the contact with the upper portion of the Pretoria Series, the
ultra-basic phase being more characteristic of the western and presumably shallower portion of the
Bushveld basin. The contact with the Red Granite is fairly sharp where exposed, though it is usually
hidden by the alluvial deposits of the Magalakwin Valley, and frequently also more or less obscured
by later intrusions of diabase. Another feature of the norite of this area, which is not found along
the southern margin of the complex is the occurrence of fairly numerous and widely distributed
dykes of red granite. These will be referred to again below.

The behaviour of the norite in the neighbourhood of Potgieterarust with regard to die Trans-
vaal System is referred to by Mr. Hall in his report on the area further east. It appears to have cut
successively across the different members of the Pretoria Series, and to have broken up the remaining
portions of them into small detached fragments. It has then followed a more definite horizon marked
by a narrow zone of banded ferruginous quartzitee, which immediately overlie the main mass of the
Dolomite. Along this line it consists more usually of pyroxenites and serpentines rather than norite.
Pyroxenites of similar type again make their appearance on Uitloop between the Dolomite and the
Older Granite, where they form the conspicuous kopje on the west side of the railway. There is no
doubt that this more northerly mass belongs to the norite magma. It is identical in petrological
character to the pyroxenites which usually appear in the norite belt as one approaches the Dolomite ;
it is intrusive in the Dolomite, and shows a similar general inclination or dip to that shown by the
main mass of the norite, namely, an inclination of about 12 degrees to the south-west. The
norite magma has therefore probably broken through the Dolomite and now forms an irregular
intrusive mass between it and the Older Granite. It again makes its appearance in a similar position
3

lft

in the western corner of Holmesleigh No. 1349, in the north-western corner of Rietfontein No. 1669,
and on Vaalkop, where the Dolomite terminates. In the latter locality the portion intensive on
the north side of the Dolomite is continuous with the main mass of the norite to the west.

North of Vaalkop the norite is directly in contact with the Older Granite, though the actual
junction of the two rocks is scarcely ever exposed. The line of junction is very irregular, small
tongues and intrusions from the norite protruding for short distances into the granite. Pyroxemtea
are also found near the contact, though they are not so characteristic as at the contact with the
Transvaal System. They are well seen in the spruit on the southern portion of Zwartfontein No.
1174, where they are of a fine-grained character and show a well marked parallel structure of the
nature of fluxion banding, probably due to pressure acting previously to complete consolidation of
the magma. A similar banding is also frequently found in other parts of the mass, producing a
parallel arrangement of the constituents over considerable distances, and thereby frequently rendering
more evident than otherwise the dip which characterises the whole belt.

The contact with the Red Granite along the south-western side of the belt is usually con-
cealed by the alluvial and other superficial deposits of the Magalakwin Valley, the line of junction
generally keeping close to the river. It is partly also obscured by the intrusion, apparently along
the line of contact, of a fairly large mass of diabase, which is exposed for about ten miles along the
left bank of the Magalakwin. On the northern portion of Hans Masibi's Location, however, a well
defined contact with the granite was found on the bank of a small spruit about 300 yards above its
junction with the Magalakwin, and not far from the northern corner of the farm Mooihoek No. 2044.
The appearances indicate that the Red Granite is intrusive into the norite. It becomes finer-grained
towards the contact and shows much pegmatite, developed in bands parallel to the plane of contact,
while the norite remains coarse and of the normal type. A short distance further up the spruit there
are several small dykes of red granite, closely resembling the average type of the main intrusion itself.
In fact the occurrence of similar dykes in the norite is a common and characteristic feature of this
part of the country. They are usually well defined and may sometimes be traced for considerable
distances ; they are fairly numerous about the north-eastern portion of Valtyn Macapaan's Location.
Even under the microscope their contact with the norite is quite sharp, the fine-grained margin of
the dyke contrasting in a marked manner with the coarsely crystalline norite. In one instance,
however, a granitic intrusion of a very complex character was met with in the norite in the more
southerly portion of Hans Masibi's Location not far from the right bank of the river. The norite
was seen to be intersected by numerous small dykes and veins of grey and pinkish granite, varying
from an inch up to twelve feet in width, and often forming a complicated network or plexus of small
intrusions. So numerous, in fact, do these veins and dykes sometimes become that here and there
the outcrop has the appearance of a coarse breccia consisting of large fragments of norite in a granite
matrix.

The occurrence of these granite dykes and the appearance of the contact with the main mass
of the Red Granite are strongly suggestive of the latter being intrusive in the norite and therefore
of later date. The evidence, however, in the more southern portion of the Bushveld is more in favour
of a gradual passage between the two rocks, indicating a general contemporaneity of intrusion.
There is no reason, however, to suppose that the whole mass of either the Red Granite or the norite
was all intruded at exactly the same time, and the difference in the relations between the two rocks
in these widely separated portions of the country may well be accounted for by supposing that the
granite of the more northern area was intruded slightly later than that of the more southern, though
belonging to the same general period of intrusion.

The main mass of the norite is of a perfectly normal character and identical with that which
has already been frequently described from other portions of the area occupied by the Bushveld
Complex. Some of the more basic and ultra-basic modifications, however, deserve a brief descrip-
tion. Their general distribution on the outer portion of the norite belt has already been referred
to, though no sharp line can be drawn between them and the typical norite. Roughly speaking,
it may be said that they tend to occupy the belt which one would expect otherwise to be occupied
by the Pretoria Series, especially to the north of Potgietersrust and along the south-western portion
of Uitloop, where small detached patches and bands of quartzite representing the remnants of that
aeries are seen to be surrounded by pyroxenites, serpentines, and other modifications. This ultra-
basic zone is, however, by no means uniform in its behaviour, sometimes, as above pointed out,
appearing between the Dolomite and the Older Granite, and gradually dying out and becoming
mixed with and finally replaced by normal norite when followed to the north-west. uBeyond the
termination of the Dolomite on Vaalkop it has not been observed beyond the southern portion of

19

Zwartfontein. Its apparent close association therefore with the thinning out and disappearance
of the Transvaal System would appear somewhat significant and might be taken as indicating that
the modified belt owed its origin to the absorption of the sedimentary rocks in the norite magma.
The fact, however, that the ultra-basic rocks take the place of the quartzites and shales of the Pretoria
Series by no means supports this conclusion, as the addition of such to a basic magma would tend
to render it more acid. A certain degree of intermingling certainly seems to have taken place where
the norite finally disappears, but the main mass of the basic rock where it surrounds the disappearing
Dolomite consists mainly of normal norite, rather than of any special modification that could be
explained as a result of magmatic incorporation. Moreover the ultra-basic modifications are of
the same type as those normally associated with the norite in other parts of the country, as for instance
in the Marico District, and on a lesser scale to the east of Moorddrift to the south of the present area.
The less basic phases of these rocks are occasionally found veining the more basic, veins of pyroxenite
occasionally occurring in serpentine and veins of norite in pyroxenite, so that the more basic types
probably represent a slightly earlier phase of the general intrusion. At the same time transitional
phases frequently occur between the more normal norite and the pyroxenite, and peridotite forms
a transition from pyroxenite to serpentine. Less acid types and variations of the normal norite
also sometimes occur, such as gabbro and diorite, a good example of the latter being exposed where
* the railway crosses the Dorps River immediately north of Potgietersrust.

Transitional varieties between the ordinary norite and the pyroxenite occur between the
Uitloop Lime Works and the big conspicuous kopje on the north-east boundary of the same farm,
and again a short distance north of the prominent dolomite kopje on the southern portion of Vaalkop
No. 1439. Specimens under the microscope [A 563, 590, and 602] usually show abundant enstatite,
often with well marked pleoohroism, and in well formed crystals, also a few large crystals of pale
greenish diallage. A small proportion of plngjoclase is always present, either interstitially or in
ophitic plates, and sometimes also a small quantity of biotite. The diallage usually encloses the
smaller enstatite crystals. This rock passes into the pyroxenite type, which consists practically
entirely of enstatite, and constitutes the dominant type of the ultra-basic portion of the norite zone.
It is perhaps best seen on the conspicuous kopje on the north-eastern boundary of Uitloop, the entire
kopje consisting of massive sheets of coarse enstatite rock [A 603], with a gentle inclination to the
south-west and apparently resting upon the Older Granite. Small veins of norite may be seen on
the south-western slopes of the kopje. " >

A similar pyroxenite is again well exposed on the north-east and south-east sides of the small
quartzite kopje, on which stands the south-west corner beacon of Uitloop, The rook is of a brownish
olive-green colour on fracture, with a bronze lustre, weathering rusty red, and consists entirely of
enstatite [A 579], Varieties carrying olivine also occur in the same locality [A 578], and these
graduate into serpentines.

1 ' Pyroxenitea are well exposed for a considerable distance along the bed of the stream which
flows through die southern portion of Zwartfontein No. 1174, and frequently show a distinct foliation,
apparently representing a kind of flow-structure, the enstatite crystals all showing a similar orienta-
tion [A 508]. As one approaches the contact between this rock and the Older Granite it becomes
gradually exceedingly fine-grained and of a pale olive green colour, the fresh rook, consisting almost
entirely of very email enstatite crystals [A 561], having a very characteristic appearance under the
microscope.

A similar tendency to a very fine-grained texture is seen along other portions of the margin
of the ultra-basic zone, where it happens to be exposed, but the composition along this line generally
varies a good deal, possibly owing to a certain amount of admixture with the altered sediments of
the Pretoria Series. North and north-west of Potgietersrust the margin of the basic rocks is
generally in contact with the band of magnetite-quartzites which overlies the Dolomite ;
the actual contact however is scarcely ever exposed. It may be seen in the railway cutting
about three miles north of Potgietersrust, a section along which is shown in Fig. 1. The igneous
rock is here a very dark greenish type, but considerably decomposed and for the most part
hidden by surface limestone. In one part it is seen to be full of small angular fragments of
white quartzite. It is seen to consist under the microscope [A 573] almost entirely of pale
greenish pyroxene with a little green hornblende and a very little interstitial plagioclase. The
quartzite fragments are sharply marked off from the surrounding matrix. A fine-grained rock of
Bimilar general appearance is found in a similar position in Tweefontein on the south side of the
north-west corner beacon of Rietfontein No. 1669, where a tunnel has been driven through it in order
to strike a small copper lode which occurs in the adjoining magnetite-quartzites. The basic rock

20

[A 583] consists here of large plates of green hornblende and pale brown pyroxene, both often showing
sieve structure and enclosing numerous rounded grainB of quartz, giving the rock in places rather
the appearance of a hornfels than of a true igneous rock. In other places near the banded magnetite-
quartzites these rocks appear again to be of a somewhat mixed character, and it is difficult to
distinguish them from included patches of dark hornfels belonging to the Pretoria Series, as for instance
near the southern corner of Rietfontein No. 1669. Here, between the magnetite-quartzite, which
forms a well marked ridge, and a small isolated band of white quartzite, we find norite of the normal
type associated with peculiar fine-grained banded greenish rocks. A specimen of one of the latter
[A 620 a] consists entirely of olivine, in bands slightly varying in texture, with a few scattered grains
of green garnet, while another specimen [A 620 b] shows no olivine, but a fine-grained aggregate of
pyroxene (diopside), and grains of similar green garnet. The origin of these peculiar banded rocks
may perhaps be connected with the incorporation of portions of the Dolomite into the intruding
basic magma.

w ' An extraordinary variety of rocks of mixed character appear on the north-west side of the
Dolomite kopje on Vaalkop, where the Dolomite finally disappears in the norite belt, and it is difficult
to offer any other explanation of them except that they are of mixed origin. Occasionally undoubted
variations of norite or gabbro may be detected ; other outcrops, however, are greenish and greyish com-
pact, patchy, and much decomposed rocks, showing varying proportions of pyroxene, olivine, serpen-
tine, calcite or dolomite, and sometimes a little green hornblende and occasional flakes of brown
mica [A 496, 566, and 5671.

On the whole, the ultra-basic portion of the norite zone is only comparatively rarely exposed
at the surface, being to a great extent concealed by thick deposits of surface limestone which have
been formed as a product of its decomposition. Immediately north of Potgietersrust this surface
deposit extends almost uninterruptedly from the bridge over the Dorps River to the railway cutting,
about half a mile north of the west corner beacon of Uitkyk, thence extending west and north-west
for a considerable distance along the norite belt. It also partially conceals the Dolomite, but is more
usually restricted to the basic igneous rocks. The thickness of the surface limestone, where over-
lying the basic rocks, is often as much as from twenty to thirty feet.

As in other parts of the country so here also the norite sometimes shows a narrow zone of
magnetite-rock. The zone occurs in the upper portion of the norite and usually close to the contact
with the Red Granite, and sometimes only separated from it by a few feet of norite. In the northern
portion of the area a band of magnetite lies at the foot of the eastern slopes of the big diabase ridge,
which runs north and south through the southern portion of Hendrik Masibi's Location, and further
south appears at intervals close to the edge of the granite along the Magalakwin Valley. It is,
however, only rarely well exposed at the surface, and probably has an average thickness of not more
than about six feet. Only one band has been recognised in this part of the country.

The Red Granite. — The main features of the Red Granite of this area have already been
described in the special memoir of the Geological Survey on the Waterberg Tin-fields, and a further
account of them is also given in the report of Dr. E. T. Mellor on the area lying further to the south
and south-west. That portion of the present area, however, which borders the Magalakwin Valley
on the south-west and forms the conspicuous range of granite hills extending from near Potgieters-
rust to near the junction of the Sterk and Magalakwin Rivers, exhibits the main features in the
local behaviour of the Red Granite and the mode of occurrence of its principal modifications in a
striking manner. The granite forms on the whole a huge escarpment facing north-east and sloping
away to the "south-west at an inclination varying approximately from 12 to 15 degrees. The
main mass usually, however, shows several minor escarpments due to the outcrop of different varieties
of the granite, which constitute as it were a succession roughly resembling that of sedimentary
strata. This feature is especially marked in the area occupied by the tin-fields, where the granite
may be divided into the following zones from the contact with the norite upwards:—

(1) Zone of fine-grained granites, aph'tes, and diorit.es.

(2) Zone of coarse massive granite.

(3) Zone of medium-grained granulitic granite. (

(4) Zone of pegmatite.

(5) Zone of red granophyre.

The lowest zone is not always to be found, but is well developed from the drift on the
Magalakwin on the northern portion of Blinkwater No. 707 to the northern corner of Sterkwater.
Beyond here it becomes somewhat obscured by intrusions of diabase, but may be followed in a

northerly direction between the main diabase intrusions and the norite from the eastern portion of
Mooihoek into Hendrik Masibi's Location. It forms a well marked range of low hills and kopjes
facing the Magalakwin Flats from Blinkwater to the northern portion of Sterkwater, thereafter being
mainly exposed on the eastern flanks of the big diabase sheet which strikes northward from the central
portion of Mooihoek, the diabase overlying it and usually forming a more prominent feature. On
Sterkwater and Hoogedooras it is also overlain by diabase, but here the basic rock tends to occupy
a hollow between the aplites and the coarse massive granite above.

The zone consists essentially of different varieties of fine-grained biotite granite and- aphte,
often intermixed in a somewhat peculiar manner with rocks of a more dioritic type'. This mixed
character is well seen in the north-eastern portion of Hoogedoorns No. 706, where acid
biotite-granites and aplites are intimately associated with bands and patches of a more basic
type rich in biotite and hornblende. The two types sometimes form a kind of breccia, the more
acid rock enclosing and flowing round patches and fragments of the more basic [A 569, 570]. A
specimen from the more northern portion of the same farm shows [A 553] felspar (both orthoclase
and plagioclase) with a good proportion of quartz, together with some augite, behaving as if
intergrown with the felspar. A few grains of sphene are also present. Typical aplitej also occur
along this zone and occasionally show a rough banded or foliated structure. The rocks are almost
always fine-grained and of different shades of pink and grey, and occasionally almost white. A
specimen from below the diabase sheet on Mooihoek shows a white rock consisting almost entirely
of orthoclase and quartz, neither mineral showing crystalline outline [A 549].

The zone of aplites is succeeded by a very much broader zone of coarse massive granite, which
constitutes the main mass of the Red Granite and the greater portion of the lofty escarpment over-
looking the Magalakwin River. It weathers into bare rounded hummocks and large boulder-like
masses, which build up the bold north-eastern buttresses of the range, so conspicuous on Roodepoort,
Groenfontein, and Sterkwater. This coarse granite gradually passes upwards into a more medium
grained variety, showing a uniform granulitic or aplitic structure and carrying only a very small
proportion of ferromagnesian constituent. This zone is well represented from Roodepoort to
Solomon's Temple, but thins out gradually to the north-west, being only feebly represented on
Zaaiplaats. It carries the cassiterite deposits of Solomon's Temple, Sterkwater, Groenfontein,
Roodepoort, and the Government portion of Zaaiplaats, those worked by the Zaaiplaats Tin Mining
Company being situated in the coarse massive granite.

Immediately overlying the medium-grained granite is the zone of pegmatite, which is well
developed along the greater portion of the belt of country occupied by the Potgietersrust tin-fields,
though more especially from the north-eastern corner of Zaaiplaats to Solomon's Temple. The
outcrop of this zone is shown on the map on Plate XVI. It consists of one or several bands of
very coarse pegmatite, separated by granite, the pegmatites varying from a few feet to about twelve
feet in thickness. They behave as flat veins or sheets, with a dip of from 12 to 15 degrees
in a general south-westerly direction, in accordance with the general inclination of the associated
zones in the granite. North-west of the north-east corner of Zaaiplaats the pegmatite appears to
die out, at least for a certain distance, but reappears in the same position, in relation to the other
phases of the granite, on Groenvlei and Appingadam. Its course, however, in this part of the area
is not so clearly defined, and its behaviour as a definite band is obscured, especially on Appingadam,
by the occurrence of numerous minor, but very irregular, bodies of pegmatite, which may be very
easily confused with it. Further details regarding this pegmatite zone and its relation to the under-
lying zone of cassiterite- bearing granite will be found in the special memoir, already referred to,
dealing with the geology of the tin-fields.

The pegmatite is immediately succeeded by the zone of Red Granophyre. This occupies all
the higher portions of the main range in the tin-fields area and forms well marked escarpment-like
features, sloping gently away towards the south-west beneath the overlying felsites of the Lower
Waterberg. The rock consists mainly of red or pink felspar and quartz, which show throughout
such a perfect development of the micropegmatite structure that the rock deserves rather the name
of granophyre than that of granite. Both in structure and general appearance it contrasts strongly
with the coarser and more massive zones of granite which underlie the pegmatite. It forms a rough
and broken surface, and shows a well marked platy or pseudo-bedded structure in its lower portions,
which maintains a definite strike and dip over long distances, the upper portions, however, becoming
more massive, of a more greyish colour, and weathering into rounded boulders, as may be well seen
about the north-east corner beacon of Zaaiplaats.

Intrusive Sheets and Dykes. — Closely associated with the Red Granite, both in mode of occurrence

and penological affinities, are various intrusive sheets and dykes of both acid and basic character.
These occur in the granite itself, in the norite, in the sediments of the Transvaal System, and
occasionally also in the Older Granite.

The basic rocks consist mainly of diabase and occur for the most part as sheets, and closely
resemble similar occurrences so frequently found associated with the Red Granite and the Transvaal
System in other parte of the country. Dykes of diabase, gabbro, and diorite, also occasionally occur.
A considerable development of diabase follows the lower zone of the granite along the Magalakwin
Valley, the intrusion for the most part appearing to follow the contact between the granite and the
norite, but occasionally branching into the norite on the one hand, or into higher zones of the granite
on the other hand. A fairly broad belt of diabase is thus seen extending north-westwards from the
northern corner of Blinkwater, through Hoogedoorns, Sterkwater, and Groenfontein, and being
joined on the latter farm by a slightly smaller intrusion traversing the granite immediately above
the aplite zone. In this latter position a similar sheet, about 100 feet in thickness appears again
in the conspicuous ridge on the central portion of Mooihoek and extends through the north-eastern
portion of Mozambique, reappearing in the same position further north, on the southern portion of
Hendrik Masibi's Location.

Dykes of diabase are not common in this area in the Red Granite, a good example, however,
occupying the narrow valley in the north-east portion of Zaaiplaats immediately below the main
workings of the Zaaiplaats Tin M ining Co. This has a general north-east and south-west trend.
Dykes of a similar character and with a similar general direction are, however, fairly common in
the Older Granite area to the north-east. Intrusions of diorite, gabbro, and other varieties allied to
the norite, occasionally cut the Dolomite between Uitloop and Vaalkop.

The majority of the later intrusions in the norite and the Transvaal System, however, more
usually take the form of acid dykes, and in some cases probably sheets also, consisting of different
varieties of granite and aplite. The red granite dykes in the norite have already been referred to
in connection with the norite. They closely resemble the normal type of the Red Granite, have a
general north and south and sometimes north-west and south-east direction, and a width of from
five to about twenty feet. Some small dykes or veins of a slightly different character were also
observed cutting the pyroxenites about half a mile south-east of the quartzite kopje at the south-west
beacon of Uitloop. Specimens of these consisted (a) of a fine-grained grey granite with pale pyroxene
and some grains of sphene, and (b) of a rather coarse aplite containing both plagioclase and orthoclase.

Acid dykes are again very common in the Dolomite between Uitloop and Vaalkop and show
considerable variation. They are usually small, have a general north-west and south-east trend,
and consist of fine-grained granites, ranging from varieties rich in biotite to aplites with scarcely
any ferromagnesian contents, the more acid intrusions sometimes becoming almost felsitic in the
fineness of their texture. More or less granophyric varieties also sometimes occur, and occasionally
a tendency to a foliated structure has been observed. Small proportions of hornblende and augite
may also sometimes be present [A 611, 613-616]. These dykes are especially abundant in the Dolomite
on the northern portion of Tweefontein No. 1033, and a larger and more sheet-like intrusion of
fine-grained biotite-granite occurs between the Dolomite and the magnetite-quartzite on Rietfontein
No. 1669 (see Fig. 2).

7. Economic Geology.

With the exception of cassiterite, the known occurrences of minerals of economic value in the
present area are only small, widely scattered, and unimportant.

Tin. — The principal occurrences of tin-ore, which occur along the higher portions of the Red
Granite country are described in detail in the Special Memoir of the Geological Survey on the Water-
berg Tin-fields. A general account of them is also given in the present volume in Dr. E. T. Mellor's
report on the adjoining area to the west. In addition to those described, occurrences of cassiterite
are also known in the granite on Delagoa and Eckstein, but scarcely sufficient work has as yet been
completed in connection with these to enable a satisfactory description of them to be given.

Copper.— Ores of copper occur on Tweefontein No. 1033 jusff below the summit of the kopje
on which is situated the common beacon of this farm with Biltongfontein, Holmesleigh, and Rietfon-
tein. It occurs as carbonates and peacock-ore (bornite) in a small brecciated almost vertical fault
fissure in the banded magnetite-quartzites which overlie the Dolomite. Thin layers of the car-
bonates are also seen on the joint and bedding planes adjoining the fissure. Some small old workings
exist on the summit of the kopje. A certain amount of prospecting work has also been carried out
here, and a tunnel was driven into the side of the hill in order to strike the ore in depth. The
occurrence, however, appears to be small, the fissure showing no sign of any lateral extent.

Lead. — Galena occuib on the north-western portion of Uitloop on the slopes of the big granite
ridge which marks the north-western boundary of the farm. It forms small scattered patches in|a
large quartz reef, one of a series which traverses the Older Granite and Black Beef Quartzitea of this
locality in a general north and south direction. The quartz body is often brecciated and contains
small fragments of granite. The ore contents are scattered and irregular in their occurrence and
show no well defined body.

A brief description of this occurrence has been given by Mr. J. P. Johnson in the Transactions
of the Geological Society of South Africa.*

Iron. — The occurrence of small bands of magnetite in the norite of the Magalakwin Valley
has already been referred to, the outcropB observed being near the western boundary of the norite
and mainly within the native locations. The bands are probably about six feet in thickness. The
ore is of the titaniferous variety and similar to that which forms so characteristic a feature of the
norite in other parts of the country, where it occurs is much greater quantity than in the present area.

A non-titaniferous magnetite is one of the principal constituents of the narrow zone of banded
magnetite- quartzite which overlies the Dolomite, and extends with slight interruptions from the
railway 3£ miles north of Potgietersrust to the western portion of Tweefontein. Occasionally the
magnetite is concentrated into bands several inches in thickness, but as a rule a very high proportion
of quartz is invariably present either as bands alternating with very thin layers of magnetite or
intimately intermixed with it.

Limestone.— -Certain calcareous zones and patches in the Dolomite are worked for white lime
in the quarries on the east side of the railway at Uitloop Siding, north of Potgietersrust. Similar
bands in the Dolomite to those worked appear to extend further west, but their occurrence is probably
somewhat patchy.

Considerable deposits of nodular surface limestone, of a similar class 'to those worked by the
Daspoort Portland Cement Co. near Pienaars River Station, cover an extensive area immediately
north of Potgietersrust. They overlie the norite and its more basic modifications, and occasionally
attain a thickness of some thirty feet. A good exposure may be seen on the Dorps River a short
distance below the railway bridge.

Building Stone and Road Metal. — The dolomitic marlstone, which forms the kopje behind
Potgietersrust Station, and the norite are both used locally for building material, the former being
also employed to a certain extent in the town as road metal. Excellent road metal in the form of
norite and diabase, etc., may be found in any quantity throughout a large portion of the district,
especially the portion between Potgietersrust and the tin-fields, and it is to be regretted that good
material of this character is not more largely employed for the improvement and maintenance of
the principal transport roads.

-"The Tin, MolyWenuui, and Luail Ocutiiuc

III.

ON A PORTION OF THE

WATERBERG DISTRICT
WEST OF POTGIETERSRUST.

By E. T. MELLOK (Geologist).

CONTENTS.

1. Area Surveyed.

2. Physical Features.

3. General Geological Structure.

4. The Waterberc System.

5. The Red Granite and other Igneous Rocks Intrusive in the

Waterberq System.

6. The Karroo System.

7. Economic Geology.

Ill— ON A PORTION OF THE WATERBERG DISTRICT WEST OF

POTGIETERSRUST.

By E. T. Mellor (Geoltxjist),

1. Area Surveyed.

The country described in the present report includes an area lying to the north-west of Nyl-
stroom and to the west of Potgietersrust. Its length from near Naboomspruit Station in the south
to near the junction of the Sterk and Magalakwin Rivers in the noith is about forty-five miles, with
an average breadth of about twenty-one miles. The area mapped covers 974 square miles with
624 miles of geological boundary lines, and embraces the country shown in the map on Plate XVI,
with the exception of that portion which lies to the north-west of a line drawn from near Moorddrift
to near the junction of the Sterk and Magalakwin Rivers.

The survey was carried out during the months of June to November, 1908. The topography
is based mainly upon that of Major Jackson's Series (Sheet 4), with additional information supplied
by the Surveyor-General's Office, and from observations in the field. The geological boundary lines
have been inserted, as far as possible, with reference to the farm beacons.

The area is traversed in the south-eastern portion by the Pietersburg line of railway and
includes a considerable portion of the northern tin-fields.

2. Physical Features.

Compared with many other parts of the Transvaal, the district offers considerable variety in
its physical features, and is especially interesting to the geologist, on account of the close connection
which these bear to its geological structure. Moreover, the features presented by this area are
typical of those which characterise a very much larger extent of country to the west, including the
great plateau of Waterberg Sandstones drained by the Paiala and Pongola Rivers, and the outer zone
of high felsite and granite ridges, which runs parallel to the precipitous edge of the plateau from the
neighbourhood of the junction of the Sterk and Magalakwin Rivers southwards and westwards for
a distance of nearly 100 miles to the borders of the Rustenburg District.

Both the topographical and geological features of the area show a more or less concentric
arrangement with regard to the main Waterberg plateau. The district is usually approached from
the low country about the railway ; a much clearer idea, however, of its general topography as well
aa of its geological structure would be obtained by any one proceeding from the main Waterberg
plateau outwards towards the flats, and this course will be pursued in the following description.

The main Waterberg plateau forms a central tract of very elevated country rising abruptly
from 1,000 to 2,000 feet above the encircling valleys, and in the area here specially referred to attain-
ing an approximate average elevation of 5,500 feet above sea-level, its edges rising in places, as at
Hanglip, to a height of about 6,000 feet. Usually borderd by long lines of almost perpendicular
kranzes presenting imposing walls of bare rock to the valleys below, the surface of the plateau itself,
when once attained, presents in a general view no very striking features. It maintains a fairly
uniform level, sloping gently away towards the north, and varied only by occasional rocky hills rising
to two or three hundred feet above the general surface. Streams like the Paiala, which flow north-
wards through the plateau towards the Limpopo, take their rise at an elevation but little below
that of the edge of the outer escarpment. Their valleys are at first broad and open, the streams cutting
more deeply into the plateau as they proceed northwards. On the other hand those streams which,
rising along the margin of the plateau flow outwards to the Sterk River, cut deep and narrow kloofs
in the escarpment, descending rapidly to the level of the valley below. The scenery in and about
these kloofs is often very beautiful. They are usually traversed by a clear stream forming a series
of rocky pools overhung by well-grown trees, above which rise the imposing walls of rock so

characteristic of the Waterberg formation (Plate IV, Fig. 1). Some of the larger kloofs are traversed
by paths which give moderately easy access to the plateau above, as, for instance, that which leads
from Groothoek No. 607 to Doornstock No. 828 to the north of Hanglip, an ascent which is also
possible for vehicles by a more circuitous route, passing round the north corner of Groothoek over
the farm Bachus No. 830. To the south of Hanglip the easiest route to the plateau is by a bridle
path leading up the kloof on BokpoortNo. 376 by the farm Jan BovenJantotheupperpartofthePalala
Valley. The most easterly portion of the margin of the main plateau forms the bold cliffs which he
along the western border of the farm Klipfontein No. 773, beyond which again the huge mountain
mass of Hanglip projects still further into the valley of the Stork River. A good idea of the character
of the country along the edge of the plateau is given by the photographs in Plates IV and V.

The imposing mountain known as Hanglip, apart from its exceedingly picturesque appearance
from every point of view, is worthy of special notice. It forms an exceedingly useful landmark to
any one working in the district, commanding the whole Sterk Valley from end to end, and being
visible from most prominent points in the felsite ranges to the east and south, including the greater
portion of the farms upon which tin has recently been discovered in the district. The characteristic
outline of the mountain makes it easily recognisable even at very great distances. Under favourable
circumstances it can be identified without difficulty from points north of Middelburg distant as much
as ninety miles. The summit of Hanglip lies on the southern boundary of Klipfontein No. 779, a
little east of its middle point. It can be reached with moderate ease by ascending the north side
of the mountain from the south-west corner of Klipfontein. Like its namesake near Louis Trichardt
the mountain derives its name from the overhanging appearance of its precipitous faces as seen from
certain directions. It rises some 2,300 feet above the valley of the Sterk River, or about 5,900 feet
above sea-level.

It probably owes its existence as an outlying portion of the main plateau, whose edges have
elsewhere been carried much further back by denudation, to a local slight synclinal arrangement
of the sandstones, which is related to the sudden change in direction made by nearly all the topo-
graphical and geological lines in the neighbourhood, where they intercept a line drawn a little south
of east from Hanglip to near Moorddrift Station. Further reference will be made to this later. The
base of Hanglip consists of rather soft coarse, purplish red sandstones and grits with many scattered
pebbles mostly of white quartz; the upper portions of the mountain, including the kranzes, are com-
posed of massive beds, from two to six feet or more in thickness, of hard red sandstones and grits,
and with numerous quartz pebbles up to two inches in diameter occurring as isolated individuals
or in thin discontinuous washes throughout the series, which is nearly horizontal, having only a
slight dip of 2 or 3 degrees towards the west.

Opposite to Hangbp the main escarpment of the Waterberg plateau makes a right-angled
bend, the two arms of which run south-west and north-west respectively. This sudden change in
direction is shown by all the other main topographical lines of the district, including the valleys of
the Sterk and Magalakwin Rivers, and the high mountain ranges which form the watershed between
them, so that while these lines maintain a close parallelism to the edge of the plateau, they all show
the same change from a north-easterly to a north-westerly direction about the latitude of Moorddrift.
This sudden change in the direction of the main topographical features is closely dependent upon
the underlying structure, which gives evidence of considerable disturbance, referred to in more detail
later, which accounts for the broken and discontinuous character of the smaller ridges and valleys
on Doomdraai No. 379 and the neighbouring farms.

The Valley of the Sterk River. — At the foot of the precipitous escarpments of the main plateau
just described, there extends a broad belt of low-lying sandy country, some five to ten miles in width,
which everywhere separates it from the outer ranges of felsite hills which run parallel to the margin
of the plateau. This low lying zone corresponds in general with the valley of the Sterk River,
although the river valley proper, frequently forms but an insignificant feature, and it is often extremely
difficult, even from such commanding points of view as Hanglip, to pick out the course taken by the
river itself. This is especially the case from Sterkfontein No. 1531, some twelve miles south of Hanglip,
to the junction with the Magalakwin. The Sterk River, draining an area of some hundreds of square
miles and carrying water throughout the year, is remarkable for the narrowness of its bed and for
the unexpected excursions which it makes from the easily eroded Waterberg sandstones on to the
harder felsites and granite. On Sterkfontein, Paarddrift, and Sterkstroom it runs for the most
part close under long escarpments of Waterberg Sandstone, passing by narrow poorts to successively
lower horizons. On Doomdraai No. 379, where it is dammed, it passes through a narrow gorge cut
in the felsites, and shortly after emerging turns sharply to the north-west to pursue for some distance

29

a more open course across Rooiwal, Rietfontein, and Waterval. Finally, on Klipplatsdrift No. 609,
deserting trie broad open zone formed by the soft strata of the Waterberg Sandstones, the river cute
across the conglomerates at the base of the series, and is hidden from this point northwards to beyond
Appingadam in a narrow valley, which in places can only be traversed on foot. In this portion of
its course the river, after following for a time the strike of the shales which are associated with the
felsites of the Lower Waterberg, enters the granite country on Groenvlei, and continues in that for-
mation as far as ite junction with the Magalakwin.

Although the low lying belt of country followed by the Sterk River, when viewed from the
neighbouring heights of the plateau, or from the outer ranges to the west, appears by comparison a
zone of very low relief, ite surface is nevertheless diversified by numerous minor topographical features,
which assume considerable importance when viewed from the lower levels of the valley itself. Such
are the various small escarpments of sandstones, which rise above the general level in lines parallel
to the edge of the main Waterberg plateau, and which are specially noticeable in the upper portion
of the Sterk Valley. North of Hanglip they axe also present, but are less persistent and confined
to the west of the river. The low lying belt is also broken by various isolated hills which represent
outlying remnants of the plateau, and which frequently assume the most fantastic shapes. They
are especially interesting as showing the progressive stages in the reduction of the plateau to the level
of the valley. The main body of Hanglip may be taken as representative of the stage in which
the full thickness of strata present in the plateau still remains, while the mound-like prolongation at
its eastern extremity may be taken as approximating a reduction to the general level of the valley.
Intermediate stages are shown by the curious hill known locally as " Pramkop " figured in Plate V,
Fig. I, which still retains a considerable thickness of the upper and harder series of sandstone which
forms the kranzee of the main plateau, and by the fantastic kopje shown in Plate V, Fig. 2, in which a
remnant of the harder sandstones still crowns the softer strata of the slopes below.

The Granite and Felsite Ratiges. — The valley of the Sterk River is separated from the still
lower country along the Nyl and Magalakwin Rivers by an almost continuous mountain range, whose
trend is parallel to that of the river valleys on either side and shows the same sudden change in
direction at a point a little west of Moorddrift, from which point the two limbs of the range trend
north-west and south-west respectively. Although interrupted by minor breaks, and presenting
the sharp change in direction already noted, the high ridges between the valley of the Sterk and that
of the Nyl and Magalakwin are best regarded as a single mountain range presenting throughout very
similar topographical and geological features. It will be convenient in describing the range to
proceed from the north southwards.

As a distinct topographical feature the range may be said to begin in a group of granite hills
lying a little to the north of the poort on the farm Eckstein No. 1866, through which the Sterk River
passes to join the Magalakwin. South-eastwards from this poort an almost unbroken range of high
granite and felsite ridges runs for a distance of some twenty-five miles to near Moorddrift. This
portion of the range is well seen from Potgietersrust. It is unfortunate that it has as yet no
distinctive appellation, sharing with the mountains to the north-east of Potgietersrust the general
name of " Makapan's Mountains," a term best known in connection with the latter group, to which
it might conveniently be restricted. From its northern extremity this portion of the range gradually
increases in height south-eastwards and attains its greatest elevation about the common boundary
of the farms Blinkwater No. 707 and Welgelegen No. 357, where it rises some 2,500 feet above the
level of the Magalakwin valley. From Groenfontein to near Moorddrift the range falls but little
below its maximum height and maintains an elevation corresponding very nearly with that of
Hanglip and the Waterberg plateau on the further side of the Sterk Valley.

^t* *■ Along the valley of the Magalakwin the range presents a bold and rocky aspect, and towards
Potgietersrust is capped by a marked escarpment due to the local pseud o- bedded character of the
granite. Towards the valley of the Sterk River the fall is more gradual although the surface is scarcely
less rough. The country in particular from Solomon's Temple southwards, though not presenting
a very formidable appearance from a distance, is accessible only with difficulty, except on foot.
Perhaps the best general view of these high granite and felsite ranges and of the surrounding district
is obtained from the heights about the north portion of Wegelegen No. 357, probably the roughest
farm in the district. The high point, upon which the north beacon of Welgelegen stands, is one of
the highest and most central in this portion of the range, and from it a good general view is obtained
of the granite summits extending north-west and south-east along the watershed. These highest
portions of the granite ridges usually present a somewhat varied and irregular outline. Their slopes
are covered with rounded boulders comparatively small in size and very red in colour, and are dotted

over with a characteristic vegetation of dark green bushes and small trees. On the lower slopes
towards the Magalakwin the coarser and more massive rock of a lower horizon forms a foot range
of typical granite kopjes with broad bare surfaces, supporting characteristic rounded masses of
weathered granite.

On the Sterk River slopes of the range only a comparatively narrow zone along the watershed
presents the varied surface characteristic of the granite, which passes beneath a covering of felsite,
the latter giving rise to a number of very characteristic topographical features. Its upper edge along
its junction with the granite forms a series of eminences which reach an elevation only slightly less
than that of the granite summits of the main watershed, from which they are readily distinguished
by their unbroken contours and apparently smooth slopes. This smoothness is, however, more
apparent than real, the close covering of small crowded angular fragments to which the felsite every-
where gives rise making these ridges more laborious to traverse than the granite itself. Prom Welge-
legen southwards the felsites cover the major portion of the western slopes of the range, while to the
north they have been more extensively denuded, leaving the underlying granite more widely exposed.
The fetaite slopes are everywhere seamed by an approximately parallel system of long and extremely
narrow valleys, or rather kloofs and gorges, deeply incised to a depth of 500 feet or more below the
general level. These valleys are remarkable for the steepness of their rocky sides, which, however,
are rarely actually precipitous. Their courses are frequently almost straight lines, and two neigh-
bouring valleys, running almost parallel for a distance of several miles, may be separated only by a
narrow ridge less than a mile in width. These deeply cut valleys are rarely of any use as a means
of access to the ranges. At the bottom they are often not more than a few yards in width and are
frequently reduced to a few feet, entirely occupied by the bed of the stream, which sometimes takes
the form of a smooth highly polished water-slide. In crossing the zone of the felsites, it is advisable
to keep to the highest portions of the spurs which separate the kloofs, along which such obscure paths
as exist are usually to be found. The lowest portion of the slopes towards the Sterk River Valley
is thickly covered with boulders weathered out from the big conglomerates at the base of the Water-
berg Sandstone. The edge of these conglomerates along their junction with the felsites is usually
marked by a small but conspicuous feature in the form of a low escarpment, thrown into contrast
with the bare and stony slopes of the felsites by its dark fringe of trees and bushes.

On Welgevonden No. 777 and Solomon's Temple the range assumes a double character, the
high granite ridges of Sterkwater and Solomon's Temple being separated by the big valley cut back
into the granite by the Sterkwater Spruit. Beyond Solomon's Temple there is again but a single
range.

From near Moorddrift to the junction of the Sterk and Magalakwin Valleys, the only practical
road over the intervening range U that which traverses the nek, which lies at the heads of the spruits
on Groenfontein No. 871, and Roodepoort No. 813, at an elevation of about 500 feet above the
Magalakwin Valley. At the sharp bend west of Moorddrift a partial break occurs in the granite and
felsite range. From this point the northern portion of the range just described strikes in a north-
westerly direction, while the southern runs south-westwards into the wide stretch of mountainous
country lying to the north of Nylstroom, beyond which it pursues a westerly course. The portion
of the range which lies parallel to the railway between Naboomspruit and Moorddrift, and which
overlooks the Springbok Flats, has many points of resemblance to that already described, being com-
posed of exactly similar geological elements, such differences as exist being mainly due to the higher
dip which prevails as compared with the northern section. The break near Moorddrift occurs between
the high granite ridges of the east and west corners of Waterval No. 720. Here the watershed between
the Sterk and Nyl Valleys falls to a level, but little above that of the Sterk itBelf , there being, however,
a much greater fall to the fow-lying valley of the Nyl. Through this break a good road gives access
over Waterval and Baviaan's Kloof from the Springbok Flats to the valley of the Sterk and the
neighbourhood of Hanglip.

A second and more complete break occurs some ten miles further south on Rietfontein No,
1426, coinciding with the valley of Andries Spruit. Between the two breaks the granite and felsite
rise into a very high and compact group of mountains, which occupy the greater portions of the
farms Nooitverwacht No. 815, Sterkstroom No. 372, and Goedhoop No. 780, the head of the valley
of the principal spruit on the last-named farm forming a deep bay in the hills. South of the break
on Rietfontein No. 1426 the granite and felsite again begin to rise above the general level, the ridges
formed by them gradually increasing in height towards the high point on the northern portion of
BuffelBdoorns No. 1773, which carries a conspicuous beacon. On the western portion of Buffelsdooms
a narrow winding poort in the range, not visible from a distance, gives easy access to the Sterk valley

■ 31

on Rooipoort. South-west of Buffelsdooms the main range continues across the northern portion
of Rietfontein No. 1638 and Welgevonden No. 1772 to the prominent heights on Doornhoek No. 896,
being interrupted only by the narrow poort of the Sterk River on Welgevonden, and by a high nek
crossed by the road leading from Doornhoek over Welgevonden to the Sterk Valley. Ab a rule the
felsite forms the uppermost portions of the main range, but on Rietfontein No. 1638, and Doornhoek
No. 896 the Waterberg conglomerates frequently form a small escarpment on the crest of the ridge.
To the west of Naboomspruit the main felsite ranges do not immediately overlook the Springbok
Flats, but are separated from them by a considerable extent of high ground, which forms part of the
hilly country lying north-east of Nylstroom. This high ground includes the high felsite ridges on
the southern border of Welgevonden No. 1772, which are separated from the main range by a portion
of the upper valley of the Sterk River. It also includes a high plateau of Bushveld Sandstone, which
occupies the greater portions of the farms Rietfontein No. 3, and Buffelskloof No. 704 (see section
on Hate XI, Fig. 2). This plateau, whose top is a nearly level expanse of Bandy country, has
an elevation of about 600 feet above the level of the Springbok Flats, and is bordered by a small
escarpment of horizontal sandstone capping granite slopes. Its edge is deeply incised by narrow
kloofs cut by various streams flowing north-eastwards, while more detached outlying portions of the
original plateau are represented by the flat-topped kopjes, which form conspicuous landmarks west
of Naboomspruit Station. Three similar kopjes, also carrying a capping of horizontal sandstones,
occur also on the north Bide of Tobias Spruit near the eastern extremity of the farm Buff elsdoorns
No. 1773.

The valleys of the Nyl and Magalakwin. — The high ranges of granite and felsite just described
overlook everywhere towards the east , wide stretches of low and comparatively flat country , including
the valley of the Nyl or Magalakwin, which skirts the ranges. This valley lies at a considerably
lower level than the hilly country to the west, with an elevation of about 3,500 feet above the sea.
The rate of fall along the Nyl is extremely small particularly from Naboomspruit to PotgieterBrust,
averaging only some six feet per mile, so that the river, when it flows at all, is not restricted to a
definite channel, but fills a wide black -soil swamp or vlei, half a mile or more in width. Eastwards
from this swampy vlei the extensive plains of the Springbok Flats stretch away for a distance of some
forty miles to the valley of the Olifants River, while westwards the ground, over an average width of
about five miles, rises gently towards the base of the ranges. This portion of the Flats is traversed
by the Pietersburg railway, from which good views can be obtained both of the plain and of the
mountains which overlook it from the west. Ab seen from the railway the area occupied by the
Bushveld Amygdaloid forms broad stretches of black or bright red soil usually dotted with thorn
trees, to the west of which a belt of yellow sandy ground, covered with a denser growth of trees of the
syringa type, marks the outcrop of the Bushveld Sandstone. Further west the granite forms the
lower slopes and the smaller groups of foot-hills and kopjes along the ranges, the main ridge being
formed by the felsite, while the Waterberg conglomerate can only be seen from the Flats in the
wooded crest of the high ridge on Waterval near Moorddrift. Between Moorddrift and Potgieters-
ruBt the flat country about the Nyl is more limited in extent, but further down the river it widens
out again to the north and east.

To summarise, it may be said that the area under review includes four main types of country
characterised by prevailing physical and geological features ; these are (1) the high Waterberg
Plateau, (2) the lower zone traversed by the Sterk River, (3) the outer ranges of high and broken
ridges of granite and felsite, and (4) the low country about the Nyl and Magalakwin. These different
types of country are arranged concentrically about the Waterberg Plateau so that sections outwards
from the latter usually give similar contours. The same types of country and the same relative
arrangement is characteristic alBO of a much more extensive zone of country following the southern
edge of this Waterberg plateau towards the border of the Rustenburg District. It is interesting
also to note that extremely similar physical and geological conditions prevail around the high plateau
of Waterberg rocks lying to the north west of Middelburg, described in the Annual Report of the
Geological Survey for 1906, to which reference should be made, many features of the Middelburg
District having a close counterpart in the present area.

Water-supply. — The name Waterberg suggests an area unusually favoured in the way of
water-supply. Apart, however, from comparison with Borne of the surrounding low country such as
the Springbok Flats and the granite areas to the north and east the suggestion is scarcely justified
with regard to the present area. Here, as elsewhere, the numerous streams which rise in the Band-
stones of the plateau yield a small supply of exceptionally good water for the greater part of the year
and the pools are rarely empty. Many such streams issue from the small valleys and kloofs, which

notch the edges of the main Waterberg plateau, and it is these which enable the Sterk River to
maintain a small but practically perennial stream, dwindling in winter in its lower course to a mere
thread connecting the numerous deep and narrow pools bo characteristic of this nver. Water is a
scarce commodity in the granite and felsite ranges, and the streams which descend from them on
either side to the Magalakwin and Sterk Rivers are entirely dry for a considerable portion of the
year. This is particularly the case with those traversing the felsites, these rocks being especially
characterised by the small amount of water they retain, owing to their exceptionally scanty covering
of surface soil and drift. On the other hand, the narrow kloofs characteristic of the felsites appear
to offer many facilities for the artificial storage of water. The granite in the ranges is much more
favourable to absorption, and the upper parts of the streams on the western slopes of the range
between Moorddrift and Zaaiplaate carry water in numerous small pools throughout the winter,
especially at many points near the margin of the felsites, where the geological conditions tend to bring
the flow to the surface. In winter the river Nyl, or as it is more appropriately termed the Nyl Vlei, is
for the most part dry where it traverses the area, but water is found not far below the surface, and
there is little doubt that very useful supplies could be obtained anywhere along the banks of the
Vlei by shallow bores. This would probably also be the case in the belt of soft sandstones, which
extends between the conglomerates which flank the felsite ranges and the main Waterberg plateau.
Warm springs. — In connection with water-supply, mention may be made of the warm springs,
of which several examples occur in the district, one having a discharge of sufficient volume to serve
to irrigate several acres of ground. This spring is situated near the house of Mr. Van Royen on the
farm Rietfontein No. 1638, to the west of Naboomapruit. Both this and a much smaller spring on
the adjoining farm Welgevonden No. 1772 are situated directly upon the course of the big Welge-
vonden fault, referred to elsewhere, which is no doubt closely connected with the origin of the springs.
The point of issue of the Rietfontein spring lies at an elevation of about 500 feet above the level of
the Springbok Flats on the hanks of the felsite range. It is surrounded by a swampy patch of Boft
black ooze, which supports a rank growth of tall reeds. The temperature of the water in the pools
above the point of issue is 95 degrees Fahr., somewhat lower than that of the Welgevonden spring,
the difference being probably mainly due to the large body of cool water and ooze with which the
' warm water mingles before its temperature can be taken. Occasional bubbles of gas rise to the
surface of the water, which smells slightly of sulphuretted hydrogen. The temperature of the water
is sometimes stated to vary with the time of day and with the season, but this impression, being
dependent on the personal feelings of the observer, is probably due to the degree of contrast with the
varying temperature of the atmosphere at these different periods, and is not borne out by thermo-
meter tests. The spring on Welgevonden is situated within a few feet of the left bank of the Sterk
River, near the old homestead about the middle of the farm, at a slightly lower level than that on
Rietfontein. It issues from a small fissure in the rock and Sows in a tiny stream to the river. The
water at the point of issue has a temperature of 108 degrees Fahr. In the same neighbourhood, but
outside the area mapped, a warm spring occurs on Vischgat No. 2121. Both the spring on Rietfon-
tein and that on Welgevonden occur near the margin of the granite and are associated with a marked
line of faulting, and this applies also to the well known hot spring at Warmbaths, and also
to that on Yischgat.

3. General Geological Structure.

With the exception of the Bushveld Sandstones, which fringe the Springbok Flats and form
the surface of the small plateau shown in the south-western corner of the map (Plate XVI), the whole
of the sedimentary rocks entering into the structure of this area belong to the Waterberg System,
both the Upper and Lower Divisions being represented. The Upper Division includes an extensive
series of red and purple sandstones and conglomerates, occupying the triangular area which has for
its base the western border of the map. This series of sandstones and conglomerates forms, besides
the mass of the main Waterberg plateau, the broad zone of lower lying ground which follows the
valley of the Sterk River, and also the western flanks of the outer ranges of hills which he between
the Sterk and Magalakwin Valleys. The Lower Division of the Waterberg System consists in this
area mainly of volcanic rocks in the form of contemporaneous sheets of felsitic lavas. A well marked
series of shales, from 400 to 600 feet in thickness, is usually found about the middle of the Volcanic
SerieB. At the surface these felsites and shales form almost everywhere an outer border to the Upper
Division of the system, and on account of their resistant character usually give rise to conspicuous
ranges of hills.

Closely associated with the Lower Waterberg rocks is the huge intrusive mass of the Red

Aaitna! I'-fOt-t, Geological Surrey, loos.

A Typical Valley on the edge of the Watcrberg Plateau ; Groothoek No. 607, north of H

Big. 2. Hanglip, from the north-east, with the escarpment of the main plateau in the distance.

The denudation of lilt alluvium, scon in llic forejirouiKl, lias to sunt- MU'iit reproduced in miniature similar f<

33

Granite, which underlies the felsites and everywhere fringes the Waterberg System, separating it
from older rockB. Southwards of Moorddrift the granite usually forms the slopes below the escarp-
ments of the felsite ridges, its surface gradually sinking to the low level of the valley of the Nyl, but
northwards of Moorddrift and particularly beyond Welgelegen the covering of felsite, to whose
resistance to denudation the formation of the mountainous ridges west of the Magalakwin was
probably originally due, has been to a great extent removed, leaving the granite to form the crests
of the range, as well as the whole of its eastern slopes down to the low country about the Magalakwin.

In its main outlines the geological structure of the district is comparatively simple, but in
many instances it presents locally considerable complexity.

From a general point of view the area may be said to form a portion of an extensive basin
of Waterberg rocks occupying a very large part of the Waterberg District, and having as its central
portion the high plateau drained northwards by the Palala and Pongola Rivera, its outer rim being
formed by the high ranges of hills which form the watershed between the Sterk River and the Nyl
and Magalakwin. ThiB outer rim, however, does not form a simple unbroken curve, but west of
Moorddrift undergoes a sharp bend at right angles, which is accompanied by a considerable variation
from the uniformity of topographical and geological features which elsewhere prevails. Sections
from the plateau outwards all show a great similarity in the general succession and structural
arrangement of the rocks, as will be readily seen from a comparison of the sections in Plates XI
and XII. The sections ABCD and EFGH illustrate the simple type of structure found in the
north, while AJK and LMN afford examples of the folding and duplication frequently
found in the southern portion of the area. In the central plateau the massive sandstones of the
Upper Division of the Waterberg System are nearly horizontal, with a slight centrally directed dip
of 5 degrees or less. The edges of these beds form the long sweep of imposing kranzes which
overlook the valley of the Sterk River from the west, and from which projects the outstanding mass
of Hanglip (Plate IV, Fig. 2. and Section AJK, Plate XI, Fig. 1).

Underlying the harder and more massive grits and sandstones which form the mass of the
plateau is a somewhat softer portion of the series, whose outcrop forms the low lying belt of country
corresponding more or less closely with the Sterk River Valley. The broad sandy slopes of this low
lying belt are occasionally varied by minor lines of escarpment representing some of the harder beds
of the series. From the margin of the plateau where the softer sandstones lie at a very low angle'
corresponding to that in the kranzes above, the dip gradually increases towards the lowest beds of
the series, which are usually massive conglomerates resting with a slight unconformity on the felsites
below. North-westwards from Moorddrift the conglomerate at the base of the sandstone series has
an average dip of about 10 degrees, but south-eastwards from the same point these basal beds are
usually very highly inclined, with a prevailing dip of 45 degrees or more, while .in places they
may be vertical or even overtilted as on the northern portion of Rietfontein No. 1638.

The conglomerates just mentioned which form a very constant feature at, or near, the base
of the Upper Waterberg, rest almost everywhere upon the felsites and shales of the Lower Division
of the system, the latter rocks also dipping consistently away from the encircling granite and towards
the central plateau, at a slightly greater angle than that of the overlying conglomerate and sandstone.
How closely the outcrops of the two divisions of the Waterberg System follow one another is apparent
from a glance at the map. There iB, however, a small degree of unconformity between the two,
which brings about the overlapping of the Lower Division by the Upper, seen about the southern
boundary of Roodepoort No. 813, near Zaaiplaate, and again to the north on Appingadam. The
outer upturned edge of the basement Waterberg rocks formed by the felsites of the Lower Division
Tests everywhere upon the Red Granite which has been intruded below it, the stratified rocks always
showing a dip from the granite towards the central plateau. The granite zone frequently takes a
prominent part in the formation of the outer ranges of hills surrounding the plateau mainly owing
to the protection afforded by the overlying felsites. Along its outer circumference its surface
rapidly falls to the low country, occupied by the norite and other easily weathered basic elements
of the Bushveld Plutonic Series, which form the valley of the Nyl and Magalakwin and portions of
the plains beyond.

In the Bouth-eastern portion of the area this low lying country, originating in the more rapid
weathering of the igneous rocks underlying the Waterberg sediments, is covered by rocks of much
later age belonging to the Karroo System. The Bushveld Sandstones, which crop out along die
borders of the Springbok Flats, are disposed almost horizontally upon the older rocks, with a slight
dip towards the Flats. They are in turn overlain by the broad, sheet of the Bushveld Amygdaloid,
as shown in Sections AJK and LMN, Plate XL

34

Having noted the main structural characteristics of the district we may now consider some
of the more complicated local features, especially those occurring along the margin of the Waterberg
System south-westwards from Moorddrif t towards Doornhoek No. 896. The change in strike of the
whole of the rocks of the Waterberg SyBtem to the west of Moorddrift is associated with a consider-
able degree of folding and faulting, as shown in part on Section AJK, Plate XI. On Kaalleegte
No. 615, Rooiwal No. 317, and adjoining farms the sandstones of the Upper Waterberg are thrown
into a well marked synclinal fold, lying between the high felsite ridges of Geluk No. 897, and Nooit-
verwacht No. 815. It is over the lower ground occupied by the sandstones that the road passes from
Moorddrift to the Sterk Valley. The folding of the sandstones in this area has apparently favoured
tile intrusion of extensive sheets, mainly of diabasic rocks, which are particularly conspicuous on
Rooiwal. Towards Hangup the change of strike in the Waterberg formation is marked by a con-
spicuous fault coursing across the northern portion of Groenfontein No. 195, beyond which the
north-western strike and low dip of the sandstones is maintained with great uniformity to the limits
of the area.

To the south of the syncline on Rooiwal just mentioned a strongly marked anticline occurs,
whose east and west axis runs parallel to and a little north of the common boundary of Sterkstroom
No. 372 and Goedhoop No. 780. Along this axis the felsites form the high mountainous tract of
Sterkstroom and Nooitverwacht, the underlying granite being exposed in the upper portions of the
valleys of two spruits on the south-western corner of the latter farm. The western extremity of
the felsites on Rietbokspruit forms a dome-shaped kopje, almost encircled by the outcrop of the
basal conglomerate of the Upper Waterberg, which is seen dipping away on three sides. The nature
of the Sterkstroom anticline, and its relationship to synclinal folds on either side are well shown in
the section AJK, Plate XI. It will be seen that on the north-west the Waterberg sediments dip
off the felsites at a comparatively low angle, while on the south towards Goedhoop the dip is very
steep, becoming vertical a little east of the line of section. These highly inclined conglomerates on
Goedhoop have a very conspicuous outcrop. When the beds are vertical or nearly bo, they weather
out in the form of walls, from twenty to thirty feet or more in height, which can be recognised from
a considerable distance, following the southern slopes of the felsite ridges.

The structure of the western portion of Goedhoop may be described as a faulted syncline, of
which the steeply dipping conglomerates and sandstones just described form the northern limb,
while the southern consists of the same beds dipping northwards with an average inclination of little
more than 5 degrees, the change in dip occurring abruptly along the fault shown in the section AJK.
Between ika Sterkstroom anticline and the Rooiwal syncline, on the farm Doorndraai No.
379 there is one of the most complicated pieces of country in the district, whose faulted character
is expressed in the broken appearance and varying directions of dip shown by the local ridges. The
faulting in this locality is particulary severe as shown by the erratic outcrop of the Waterberg
conglomerates, and by the degree to which the sedimentary rocks have been invaded by intrusive
diabases. A fault which crosses Doorndraai from east to west brings the felsites to the surface in
two places considerably to the north of their main line of outcrop. The more westerly of these
exposures of felsite is crossed by the Sterk River in a gorge, the narrowness of which has been taken
advantage of in the building of the Doorndraai dam. The low dip of the conglomerate overlying the
felsite causes its outcrop to assume the very sinuous course shown in the map where it crosses the
deep valley of the river at this point. The course of the conglomerate outcrop on Doorndraai may
be compared to that on Goedhoop and Rietbokspruit, the conditions being somewhat similar except
that on Doorndraai the faulting is much more pronounced.

Passing south-westwards from Goedhoop the felsiteB and conglomerates of the ridges are
interrupted by frequent faults, the prevailing trend of which is nearly east and west, cutting the
strike of the rocks in an oblique direction. Of this character is a persistent fault which duplicates
the outcrop of the basal conglomerates over Rietfontein No. 1426 and Koedoesfontein No. 309.
Much faulting also occurs at the same horizon about the northern corner of the farm Rietfontein
No. 1638, where also the dip of the conglomerate is very high. Near the northern beacon of the
farm the beds are vertical, and for a short distance are even overtilted to as much as 60 degrees
beyond the vertical position. Still further to the south-west on Welgevonden No. 1772 and in the
neighbourhood of the tin workings on Doornhoek No. 896, the structure of the country becomes very
complicated. Along the northern boundary of Welgevonden the outcrop of the basal conglomerate
of the Upper Waterberg makes a very sudden bend southwards, resuming its easterly strike again
by a similar bend on Doornhoek. These sudden changes in strike are accompanied by corre-
spondingly rapid variations in dip in the conglomerates, but it is perhaps the felsites and shales of the

35

underlying Lower Waterberg which show the greatest amount of disturbances resulting in re-
duplications of the shale horizon as shown in the map. The section, Plate XI, Fig. 3, is
representative of the conditions prevailing in this locality. As in the case of the disturbed area
about Doorndraai No. 379, the faulting is accompanied by intrusions of igneous rock, which appear
to have followed partly the course of the principal faults and partly the bedding of the sedimentary
strata. Thus the diabase on Welgevonden, where it follows the Stork River, coincides with a very
marked and persistent fault whoBe course is indicated by frequent outcrops of brecciated country
rock, extending in a north-easterly direction across Doornhoek, Welgevonden, and Rietfontein,
while on Gemsbokfontein No. 2123 it behaves more like the interbedded sheets so frequently found
in the Waterberg System. The big fault just alluded to, which may be conveniently termed the
Welgevonden Fault, is remarkable for the occurrence at two points along its course of hot springs,
the high temperature of whose waters indicates a deep-seated origin. These occur on Welgevonden
and on Rietfontoin No. 1638, and have been referred to in more detail in connection with water-supply.

ACROSS DOORNHOEK No. 896, NEAR BOUNDARY OF KROMKLOOF No. 1637.

(.Portion of Section "PQR," Plate XI, Fig. 3.)

Fig. a.

Another well marked fault, which may be termed the Doornhoek Fault, crosses the farm of
that name, intersecting the Welgevonden Fault. It is this fault which brings about the duplication
of the felsites, shales, and conglomerates in the neighbourhood of the Doornhoek Tin Mine, as shown
in Fig. 3 and section PQR, Plate XI. From the latter section it will be seen that an anticline
brings the Red Granite to the surface on the southern part of Doornhoek and on Erstegeluk, south
of which the felsites have a southerly dip.

Entirely unconformable to all the rocks previously mentioned are the Bushveld Sandstones
occurring in tie southern portion of the area. These rocks, which belong to the uppermost division
of the Karroo System, consist of massive beds of fine-grained yellowish or reddish sandstone resting
upon the older rocks either horizontally, or with a scarcely perceptible dip towards the Springbok
Flats. About Rietfontein No. 3 they form the surface of an extensive elevated plateau, resting as
a thin covering directly on the surface of the granite. They also form the capping of several detached
flat-topped kopjes on Naboomspruit No. 2297, and about the south corner of Koedoesfontein
No. 309. Near Naboomspruit Station they are faulted down more nearly to the level of the flats below
the surface of which they pass, being covered in turn by the broad sheet of the Bushveld Amygdaloid
(section LMN, Plate XI).

4. The Waterberg System.

Among the rocks occurring in the district embraced in the present survey the most important
are those of the Waterberg System, with the associated igneous rocks, including the Red Granite.

In the Waterberg District the rocks of this system may be grouped as follows : —
w / h ( Upper Division — Sandstones, grits, and conglomerates.
o r ^ I Lower Division — Felsites and allied volcanic rocks, with interbedded shales.
Oyttem. ^Volcanic Series)

3d

A general account of the characters of the rocks of the Waterberg System in thin area together
with the main outlines of the structure of a portion of the Waterberg District was given by Dr.
Molengraaff in 1898.* An account of the district has also been given by Mr. G. G. Holmes. f

Descriptions of the corresponding rocks in the Middelburg District have been given in the
Annual Reports of the Geological Survey for the years 1903-1906.

Lower Division — Felsites and Shales.

Throughout the area the Lower Division of the Waterberg System consists of a thick series
of volcanic rocks, mainly felsitic in character, with interbedded shales on at least one horizon. The
general arrangement and relationships of the aeries are shown in the sections on Plates XI and XII.
Apart from their purely geological interest these rocks derive additional importance from the frequent
occurrence in them, or about their contact with the Red Granite, of the deposits of tin-ore which
have recently attracted bo much attention to this portion of the Transvaal. The Lower Waterberg
rocks rest everywhere upon the Red Granite, which has been intruded beneath them, and whose
relationships to the felsites will be discussed more fully later. They are everywhere overlain by the
conglomerates and sandstones of the Upper Division of the System. Their outcrop thus usually
occupies a belt of varying width between the Red Granite and the Waterberg Sandstones and follows
closely the margin of the latter. The predominant rocks are the felsites, the shales being comparatively
thin and frequently obscured by the abundant debris arising from the disintegration of the felsites.

The Felsites. — These rocks present a very characteristic appearance which renders them easily
recognisable. They are usually of a purplish red colour when fresh, weathering to a brick red on
their exposed surfaces. In texture they are dense and felsitic, sometimes showing a few small
phenocrysts of reddish or greyish felspar, altered in some cases to greenish pinitoid (pinitoid-porphyry
of Molengraaff). Under the hammer they are hard and brittle, and break with a semi-conchoidal
fracture. In some localities they are slightly coarser and present a finely crystalline appearance to
the eye, and then resemble some of the finer-grained phases -of the Red Granite, to which they are
closely related in composition and from which they are not easily distinguished in hand specimens.
Under the microscope typical specimens show a fine felsitic groundmass of quartz and felspar, slightly
coarser in places, with phenocrysts of orthoclase and scattered flakes of a bright green mica some-
times aggregated into bunches. Iron-ores in minute opaque grains or octahedra are sometimes
present. In the more basic varieties augite is occasionally to be found. In composition the felsites
probably vary considerably, especially in the proportion of silica present, and darker varieties occur
in many places, which would be more correctly termed andesites. The term felsite, however, is in
common use and serves as a convenient field-term of general application.

On account of their resemblance to the finer phases of the Red Granite, which are sometimes
in contact with them, the felsites have frequently been regarded as a marginal modification of the
granite, which, however, is of later origin than the felsites, and intrusive in them, the sharp line of
contact being easily traceable in those localities where the junction of the two rocks is exposed. The
felsites are true volcanic rocks, and in many localities show the characteristic features of lava flows,
including beautiful flow banding, slaggy and amygdaloidal structures, and flow brecciation. Flow-
structures, most clearly shown by weathered surfaces, are well developed in the felsites on Doornhoek
No, 896, a little west of the old store above the drift. Very fine examples are also to be seen in the
valley of the spruit crossing Sterkstroom No. 372, about half a mile below the granito-felsite contact.
Here the felsite exhibits also numerous included angular fragments of banded felsite, in some cases
six inches or more in diameter. Vesicular and amygdaloidal structures are usually only seen near
the upper surface of the felsite sheets. They are well developed immediately below the shale horizon
in the western corner of Appingadam, the felsite containing numerous amygdales from a quarter of
an inch to six inches in diameter, their size increasing as the top of the sheet is approached. The
outer portion of the amygdales consists of prismatic crystals of quartz, the central portions being
frequently composed of tourmaline in radiating needles. Exactly similar conditions are seen where
the same horizon is exposed at the roadside in the north-western corner of Groenvlei No. 610. Just
below the base of the band of shales which traverses the felsites the latter frequently show a well
marked brecciated structure giving the rock the appearance of a fine volcanic breccia or coarse tuff.
Sharply angular fragments are crowded together in a felsitic matrix, and become very conspicuous

"A Geological Sketch of the Waterberg District." Annual Report of the State Geologist of
-"Some Notes on the Geology of the Northern TmnsTftaL" Trans. GeoL Soc^S.A., Vol VII.

37

on weathered surfaces of the rock. The fragments vary from a quarter of an inch to two or three
inches in length. In a specimen from the southwestern portion of Zaaiplaate the included fragments
are composed of partly devitrified glassy material with well marked perlitic structures. In some
cases these fragmental rocks appear to be due to brecciation of the surface of the felsite flows with
inclusion of the resulting fragments in the still fluid magma. In other cases they are more probably
true volcanic tuffs. They are well marked below the shale horizon on Doornhoek No. 896 and
Welgevonden No. 1772, and from Solomon's Temple northwards to Appingadam.

About the same horizon breccias sometimes occur, in which the fragments axe larger and the
matrix quartzose, giving the rock more the appearance of a breccia due to movement or faulting.
This type occurs, together with that just described, near the southern corner of Solomon's Temple.
In many cases the horizon of the breccias shows much tourm&linisation, some of the included frag-
ments being replaced by tourmaline, while others remain either partly or wholly unaltered. An
example of this kind is seen on the western corner of Appingadam just above the amygdaloidal
portion of the felsite already alluded to. The lowest beds of the shales which succeed the felsite
also have in places the characters of a breccia due to the inclusion in them of numerous angular
fragments derived from the felsites below. The felsites just described resemble very closely those
occurring at the same geological horizon in the Middelburg District, described in the Annual Reports
of the Geological Survey for 1905 and 1906. The coarse agglomerates frequent in the Middelburg
District have, however, not been observed in the Waterberg nor do the felsites in the latter locality show
the same degree of differentiation into readily recognisable individual flows of varying composition.
On this account, probably, coupled with their great thickness, the sheet-like form of the felsites and
their maintenance of a regular dip has not been generally recognised. These features come out very
clearly, however, in the mapping of an extensive area like the present, and are at once apparent from
an inspection of the accompanying maps and sections. In the northern portion of the district,
the average thickness of the felsites represented is probably not less than 3,000 feet, of which 1,400
feet lie below, and 1,600 feet above the shale horizon. A peculiar feature of the felsites is the frequent
occurrence in them of small included masses of quartzite from a few feet to a few yards in extent
They are usually rendered conspicuous, when the grass is not too high to conceal them, by the
numerous blocks of whitish quartzite shed over the surrounding surface. Such quartzite inclusions
occur on Roodepoort No. 819, Solomon's Temple, Naauwkloof No. 616, Zandspruit No. 782, and
other farms. They frequently show distinct bedding, but no constant dip, and are veined by the
felsites. They appear to represent rock masses caught up by the felsite flows at the time of their
extrusion. The felsites also show other patches of highly quartzose character and sometimes
resembling quartzitos in appearance. These grade insensibly into the normal felsite, of which they
appear to be a modification comparable to the quartzose patches in some granites.

On account of their resistance to denudation the felsites nearly always form high ranges of
hills distinguishable by their rounded contours and apparently smooth surfaces when seen from a
distance. The felsite ridges are, however, thickly strewn with angular blocks averaging about six
inches in diameter, which render travelling unusually laborious. When denuded of grass the felsite
hills are of a red colour, slightly duller than that of the granite, from which they are further to be
distinguished at a distance by the absence of the small trees and bushes which are usually plentifully
dotted over the latter. Very characteristic also of the felsites, wherever they are found, are the
extraordinarily deep and narrow valleys or kloofs cut by the streams which traverse them, already
described in connection with the physical features of this district.

The outcrop of the felsites attains its greatest width of about five miles, from Zandspruit No.
782, north-west of Moorddrift, to Welgevonden No. 777, forming a broad dip-slope inclined at about
7 degrees on the flanks of the ranges. Further to- the north-west the denudation of the felsite
has left the granite exposed at the surface as seen in the section through Zaaiplaate (Plate XII,
Fig. 1). South-west of Moorddrift the felsites form a group of conspicuous hills on Sterkstroom
No. 372, the great width of outcrop being in this case due to the anticlinal structure, shown in the
section on Plate XI, Fig. 1. Further to the south-west the outcrop is much narrower, owing to the
greater angle of dip, which here averages about 45 degrees. On Welgevonden No. 772, and Doorn-
hoek No. 896 the felsites again occupy a very wide belt of country owing to the lower dip and the
anticlinal structure.

The Shales. — Associated with the felsites of the Lower Division of the Waterberg System is
a well marked series of shales, the outcrop of which is very clearly exposed in the north along the
Sterk Valley from Roodepoort No. 819 across Zaaiplaate and Groenvlei to the southern comer of
Appingadam, and in the south on Welgevonden No. 772 and Doornhoek No. 896, being probably

38

hidden over the greater part of the intervening space by the Upper Waterberg rooks. In the Middel-
burg District many different volcanic flows can be recognised as forming part of the Volcanic Series
of the Waterberg System, and among them sedimentary rocks, mainly shales, with some quartrites
and sandstones, occur at five different horizons at least,* whereas in the Waterberg District, bo far
as it has been investigated, only one shale horizon is met with. The succession seen in the Middelbuig
District is shown in the section given in the Annual Report of the Geological Survey for 1906
(Plate XXIII). In both districts the shales are of a very flinty character, and have frequently a very
similar appearance to the banded felsites which occur above and below them, a resemblance which
has probably given rise to the doubt with which their sedimentary character is often regarded. No
one, however, who has had an opportunity of examining the more favourable exposures, such as
those along the Sterk valley south-west of Zaaiplaats, and has seen the full succession of dark green
or grey normal looking shales alternating with more sandy and flaggy beds, frequently showing very
clear ripple-marking, can doubt the truly sedimentary origin of these beds.

The best and most continuous exposures of the shales occur on the south-western boundaries
of the farms Roodepoort, Zaaiplaats, and Oroenvlei, along the Sterk River, whose course follows
the outcrop of the shales more or less closely for several miles and has been determined by it (see
section ABCD, Plate XII). On account of the association of the shales with the Sterk River Valley
both at the locality just referred to and about Doornhoek and Welgevonden No. 772, the name
" Sterk River Shales " is proposed as an appropriate distinctive term for these beds.

On the north-western corner of Groenvlei the shales pass under the sandstones of the Upper
Waterberg, but reappear again from under the same rocks for a short distance in the north-west
corner of Appingadam. In a similar manner the shales are hidden at the southern extremity of
Roodepoort No. 813, but are exposed as a small inlier in the north corner of Grootrivier No. 2321,
and again half a mile to the south, where the deep kloof carrying the spruit from Solomon's Temple cuts
through the conglomerates. South of this point the outcrop of the shales is probably covered "by
the Upper Waterberg rocks as far as Welgevonden No. 1772 near Doornhoek, although at an inter-
mediate point on Koedoesfontein No. 309, there are indications that the shale horizon is not far
from the present surface. Very good exposures of the Sterk River Shales occur on Welgevonden
No. 772, where they form a small kopje on the north bank of the river about half a mile above the
old homestead. The best sections, however, are seen in the hills immediately to the north of the
same point about the head of a spruit joining the river near the kopje. The same band of shales,
repeated by faulting, crops out about the eastern corner of the farm and on Rietfontein No. 3. On
Doornhoek the same shales are found in several places and on account of the occurrence of tin in
them in the neighbourhood of the mine and elsewhere, they have been extensively prospected.
Thirty miles to the south-west on the other side of Nylstroom, similar shales occur in the felsites on
Witfontein No. 371.

The Sterk River Shales as a rule axe distinguished by somewhat peculiar characters. They
are often extremely hard, with a flinty or felsitic appearance, and break into sharply angular frag-
ments with little tendency to part along the banding. In colour they are light bluish or greenish
grey, or are banded with alternate light and dark grey as on Doornhoek. Certain beds are slightly
coarser in texture, and when much weathered have a fine sandy appearance and reddish colour.
The flinty character of the Sterk River Shales is scarcely surprising, considering that the material of
which they are composed was probably largely of felsitic origin, and that the shales have suffered
considerable induration both by the felsite above and by the granite intrusion below, whose influence
is shown in the frequent occurrence of tourmaline and cassiterite in the shales in localities as wide
apart as Doornhoek and Appingadam. In the Sterk River Valley about Zaaiplaats and Roodepoort
the shales have a much more ordinary appearance than they have further south, showing a
continuous succession of dark greenish or purplish shales, flags, and fine-grained sandy beds, such
as might be met with at various other horizons in the Waterberg System. They also frequently
show clearly defined ripple-marking. In this locality only the lowest and highest beds
nearest the felsites have a flinty or felsitic appearance, which, however, as far as it extends, is
as well marked as in the Doornhoek and Welgevonden types. On Appingadam moderately coarse
yellowish white quartzites are associated with the shales.

The thickness of the shale series in the northern exposures about Zaaiplaats may be estimated
at between 500 and 600 feet, and in the southern occurrences near Doornhoek at about 450 feet.

* 15. T. Mellor.— " The Ueological Structure of the Midrtelbnrg District." Trans. Geo!. Hew. S.A., Vol. X, p. 47, IU07.

Upper Division. — The unconformity at the base of the Upper Division.

The Upper Division, which includes the thick aeries of conglomerates, grits, and sandstones,
which form the most characteristic and easily recognised portion of the Waterberg System, almost
everywhere in the district rests directly upon the felsites and shales of the Lower Division. The
junction, however, is an unconformable one, and is usually marked by the occurrence of beds of
coarse conglomerate which frequently exceed a hundred feet in thickness, and occasionally, taken
together with a few insignificant sandstone bands, attain 200 or 300 feet. This conglomerate horizon
usually forms a conspicuous feature in the local geology and serves as a convenient and easily
recognisable geological landmark. The unconformity between the two divisions of the Waterberg
System is further marked by a difference in dip, the strata of the Upper Division being somewhat
less steeply inclined. As a rule this difference is not very great, and owing to the coarse and massive
character of the basal conglomerates, which only show bedding-planes of the rudest kind and usually
rest upon felsites devoid of bedding-planes altogether, it cannot be demonstrated. Even where
sedimentary beds belonging to the two divisions occur in the same section as at Doornhoek (Fig. 3),
it is very difficult to detect any difference of dip. In the neighbourhood of Zaaiplaats,
however, on the southern portion of Roodepoort No. 819 the sandstones and conglomerates of the
Upper Division rest almost directly upon the shales of the Lower, and the difference of dip can be
observed (Plate XII, Fig. 1).

The difference of 10 degrees here shown 1b the maximum amount hitherto noticed in this
area. The portion of the section just referred to is one of the most instructive in the district and
is easily accessible from the road which passes from Groenfontein across Roodepoort towards the
Sterk Valley. Moreover, as shown in the mapping, the basal conglomerate of the Upper Division
is here Been to pass from the upper to the lower felsite horizon across the outcrop of the shales, thus
affording additional evidence of the unconformity. Further north the base of the Upper Division
again crosses the outcrops of the felsites and shales and comes to lie upon the Red Granite. In this
case there is also probably involved a certain degree of overlap in the beds of the Upper Division.
It will be seen, however, from an inspection of the map that although this unconformity exists, the
margins of the two divisions maintain an approximate parallelism, and this is the case in the Water-
berg District not only over a much wider area than that included in the map, hut is also seen in the
Middelburg District. It appears, therefore, that although a break exists between the two divisions
of the Waterberg SyBtem, it is not usually Bhown by any great difference of dip. From the evidence
afforded by the conglomerates, however, it appears that although intrusions of the Red Granite
occurred subsequently to the deposition of a portion at least of the Upper Waterberg strata, earlier
intrusions involving the introduction of cassiterite and its associated minerals had taken place before
the Upper Division of the Waterberg System was laid down.

The Conglomerates. — Conglomerates form a very constant feature at or close above the base
of the Upper Division and although sometimes very thin are only wanting at that horizon in a few
instances and at widely separated localities. The broad outcrops of conglomerates shown in the
Roodepoort section (Plate XII, Fig. 1), diminish in breadth northwards and on Appingadam
purplish red false-bedded sandstones rest directly on the Lower Waterberg rocks. Ab already stated
it appears probable, however, that there is here a certain degree of overlap in the members of the
Upper Division. In many sections the conglomerates, even when very strongly developed, do not
form the actual base of the series, but are separated from tthe underlying rocks, usually felsites,
by varying thicknesses of sandstone, or of purple or deep brick red hard flaggy shales.
The latter are usually only a few feet in thickness, forming lenticular masses between
the conglomerates and felsites, which come again into contact as the shales thin out and
disappear. Similar flaggy shales frequently occur among the beds immediately succeeding
the basal conglomerate, alternating with sandstones and pebbly grits. One of the largest
developments of such shales is seen in a section approximately on the north-west boundary of Riet-
fontein No. 1638, and about a mile from the western corner of the farm, where it is crossed by a kaffir
path leading over the ridge to Rooipoort. Here about six feet of deep purple conglomerate, composed
of pebbles averaging over one and a half inches in thickness and resting directly on the felsite, is
succeeded by as much as 500 feet of hard reddish purple flaggy shales, whose ragged edges project
in long lines above the surface. Less than a mile to the south-west in the line followed by the
Bection LMN, Plate XI, this exceptionally large body of shales is represented by sandstones and
thinly bedded flaggy quartzites, with only occasional shaly bands a few feet in thickness, the
conglomerates at the base being more strongly represented.

In composition as well as in thickness, the Waterberg conglomerates show considerable

40

variation . The pebbles are usually crowded in s scanty dark purple matrix of a coarse sandy character,
which, however, may increase in proportion till the rock becomes a coarse sandstone with scattered
pebbles. The pebbles in the basal beds usually range from one to biz inches, but boulders up to a
foot or more in diameter are frequent. Near the base the pebbles consist mainly of red and purplish
felsitic rocks similar to those upon which the conglomerate rests and derived from them, and this
type frequently includes almost the whole of the pebbles. Bed, purple, and white quartzites are
also abundant, especially in certain localities, particularly in the northern portion of the area. West
of Hoorddrift on Baviaanskloof , pebbles are frequent which can be recognised as having been derived
from rocks of the Pretoria Series, similar to those now exposed in the ranges about ten miles to the
eastwards. These include some dark purple quartzites and a pretty conglomerate composed of
Bmall rather angular lightly coloured fragments of cherty and jaspery rocks in a white quartzitic
matrix. The latter type are noticeable in the conglomerates for a long distance to the south-west
and are fairly numerous in the conglomerate at Doornhoek No. 896. Next to the felsite pebbles bo
abundant in the lowest beds, pebbles of banded magnetite- jasper rocks, usually flat iu shape and
small in size, are the most numerous, and form the predominant type in many conglomerate beds,
especially those occurring a little distance above the base of the aeries. In some places beds of this
description alternate with others composed mainly of felsite pebbles, as if the different beds had
been supplied successively from distinct areas in which the materials of the respective type of pebbles
predominated, while in other cases the same bed contains both these types in approximately equal
proportion.

Besides those occurring bo constantly near the base of the Upper Division of the Waterberg
System conglomerates are found at various higher horizons. As a rule in these higher conglomerates
the pebbles are smaller in size, averaging two or three inches in diameter, and felsite pebbles are
comparatively rare or absent, the predominating types being composed of banded jaspery rocks and
quartzites. The distribution of these higher conglomerate horizons is shown in the accompanying
map and sections. The beds forming the mass of Hangup are sometimes spoken of as conglomerates.
They are, however, grits and sandstones containing pebbles, many of white quartz, scattered very
widely through the rock or distributed in thin washes.

Perhaps the most interesting of the pebbles and boulders in the conglomerates are those com-
posed of granite and of various tourmalin iaed rocks, because of their bearing on the relative ages
of the granite and sedimentary rocks. Granite pebbles are only occasionally met with in the Water-
berg conglomerates. It might be expected that such pebbles would be numerous where the base of
the sedimentary rocks rests upon granite, as on Waterval west of Moorddrift. A search for granite
pebbles in this locality, however, was not successful. In the basal conglomerates the only ones met
with occurred near Oatkop on the eastern boundary of the Rustenburg District. In the present area
granite pebbles were found in the conglomerate beds on Klipspruit No. 267, on the south side and
near the foot of Hanglip, and again further south in the same conglomerate on the banks of the Groot
Sterk River, near the western corner of Rooipoort No. 15. The granite of these pebbles is a dark
purplish variety of irregular or rather close-grained texture, resembling some of the marginal modifi-
cations of the granite near its contact with the felsite, and not very typical of the more normal Red
Granite. Two specimens found at the locality mentioned near Hanglip are, however, of more usual
type.

The presence of granite pebbles in the conglomerates points to the granite being in part at
least older than the Upper Division of the Waterberg System. Although not very typical, there
seems no reason to doubt the derivation of the pebbles from the Red Granite, especially as the
evidence afforded by them is supplemented by the occurrence of tourmaline-bearing rocks, which
are both more numerous and more easily found than the granite pebbles. In the course of the
present survey boulders and pebbles of tourmaline-bearing rocks have been obtained at intervals
along the base of the Upper Waterberg from Appingadam to Gatkop near the border of the Rustenburg
District north of the Booiberg. Their occurrence on Appingadam was noted by Mr. H. Merensky*
in his description of the rocks associated with the tin-deposits. About Appingadam well marked
conglomerates, such as usually form the lowest beds of the Upper Waterberg, are not developed,
although they appear at this horizon a few miles further south, and form a very prominent feature
in the geology of the district. On Appingadam the beds resting directly on the felsites of the Lower
Waterberg are rather soft purplish red Bandstones marked by unusually pronounced false-bedding,

e Complex, etc." Trans. OeoL Soc. S.A.,

41

giving local dips in aiiy direction up to 20 degrees, but having a general dip towards the west
of about 3 degrees and continuous in that direction with the main body of the Waterberg Sand-
stone. The lowest beds contain numerous fragments derived from the older rocks on which they
rest. Among these are fragments of felsite, some of which include irregular masses of radiating
tourmaline needles and prisms. In other cases the fragments consist entirely of such tourmaline
masses, which may be two or three inches in diameter. All the fragments are rather angular and
show few signs of being water-worn, and are probably therefore not far from their origin. Both the
felsites, which underlie the sandstones, and the associated shales are much tourmalinised in this
locality.

On Geluk No. 897 to the west of Moorddrift, a large boulder of yellowish white quartzite,
penetrated by coarse needles and small prisms of tourmaline, was found in the conglomerate at the
base of the Upper Waterberg. A few miles further south on Sterkstroom No. 372, near the middle
of the east boundary, tourmaline occurs in the felsites in radiating needles filling small veins.
Pebbles of a similar rock occur in the conglomerate which rests upon the felsites. On the northern
portion of Rietfontein No. 1638 large boulders of tourmaline-bearing rocks occur at the base of the
conglomerates, along their junction with the felsites, and include some specimens of the dark grey
banded rocks, often tourmalinised, associated with the Sterk River Shales, which occur in situ not
far away on Welgevonden and Doornhoek.

Beyond the limits of the present area a pebble of felsite with rosettes of tourmaline was found
among those weathered out from the conglomerates on Roodepoort No. 2148, to the west of Warm-
baths, and several pebbles of a coarse tourmaline-quartz-rock were met with in the conglomerates
on Waterval No. 965 near Gatkop, beyond the Rustenburg boundary. The extensive distribution
of these tourmaline-bearing boulders indicates a widespread mineralisation from the Red Granite
prior to the deposition of the Upper Division of the Waterberg System.

Sandstones. — The conglomerates of the lower portions of the Upper Division of the Waterberg
System are succeeded by a thick series of sandstones and grits, which, together with the conglomerates,
form the better known portions of the ByBtem.

They have been so frequently described that only a brief reference is necessary here. They
show the colouration characteristic of the Waterberg Formation, brown and red being the prevailing
colours, almost invariably tinged with purple, which sometimes becomes the predominating colour.
Thin-bedded flaggy and shaly types occur at rare intervals, but as a rule individual beds average
two or three feet in thickness, and the coarser sandstones and grits, which form a large proportion of
the total thickness, are frequently very massive. This is particularly the case in the upper part
of the series, which forms the precipitous kranzes flanking the main Waterberg plateau. These
thick-bedded grite and sandstones usually contain numerous scattered pebbles up to two or three
inches in diameter as well as more conspicuous pebble washes and irregular conglomerate bands.
Beneath these harder grits and sandstones there is very generally developed a zone of much softer
beds, often very coarse in texture, which give rise to a belt of country, largely covered with deep
surface sand, which usually intervenes between the edge of the Waterberg plateau and the ridges
formed by the outcrops of the conglomerates and sandstones near the base of the series. The latter
are usually hard and quartzitic and very dark in colour, chocolate browns and brownish purples
prevailing.

5. The Red Granite and other Igneous Rocks intrusive in the Waterberg System.

The Red Granite. — Closely associated with the Waterberg System throughout the area is the
Red Granite, a member of the Bushveld Plutonic Complex. The general characters of this rock
are now so well known and have been so frequently described that its distribution and such features
only iw are of interest in relation to this particular area need be dealt with here.

The general distribution of the Red Granite will be best understood by reference to the map
on Plate XVI. It occupeis a belt of country, varying from three to seven miles in width, between
the outcrop of the felsites of the Lower Waterberg and the more basic modifications of the Bushveld
Plutonic rocks, such as the norite. In the southern portion of the area the granite usually forms
the margin of the flats surrounding the mountainous tract, rising into the foot-hills of the ranges and
forming the lower slopes below the felsites. From Naboomfontein No. 718 northwards, however,
it extends from the flats to the crests of the ranges, the felsites occupying what may be regarded as
broad dip-slopes descending towards the Sterk River Valley. Even where this is the case it is
evident that the granite owes much of its present prominence to the protection from denudation
formerly afforded by the felsites, after the removal of which the granite tends to weather somewhat

42

rapidly. To the west of Potgietersrust, from Lisbon to Blinkwater, where the granite ranges attain
their greatest elevation, the edge of the felsites lies not far from the crest of the range, while farther
north the felsites hare been denuded to a much greater extent and the continuation of the granite
range diminishes very considerably in height.

As a rule the granite and felsite boundary follows a very even coarse, except where it is inter-
cepted by deep kloofs in that portion of the ranges lying west of Potgietersrust. South-west of
Moorddrift, denudation along the axis of an anticlinal structure in the felsites and overlying Water-
berg strata produces a bay in the felsite margin on Noottverwacht No. 815, and exposes the granite
as an inlier on the south-western corner of the farm. In the south the full extent of the granite is
partly hidden beneath the sandstones and amygdaloid of the Springbok Flats, the sandstones also
forming a thin covering over the surface of the small plateau, which extends from near Doornhoek and
Welgevonden towards Naboomspruit.

The granite shows many variations in texture, some of which are of purely local, while others
are of a more general, character, such as the frequent differentiation into zones distinguished by
special features. This feature is particularly well shown in the northern portion of the area especially
along Zaaiplaats, Groenfontein, and Solomon's Temple. The deeper portions of the granite are
coarse in texture and very massive, weathering into bare rounded humps and huge boulder-like
masses. Above this coarser type is found a finer-grained, harder variety, usually of a granophyric
character, which weathers into a much more broken surface strewn with small boulders and sub-
angular blocks. Along a narrow zone between the coarser and finer types, the course of which has
been indicated on the map, a zone of very coarse pegmatites is of frequent occurrence, above which
for some distance the finer granite displays a strong tendency to a platy structure, which maintains
a definite dip and strike over long distances and gives rise to the well marked escarpments which
occur along the crests of the granite range.

North of Zaaiplaats the distinction between the coarse and finer types of granite is not so
sharply marked. The line appears to follow the face of the range to near the north beacon of
Groenvlei No. €10 when it turns sharply to the south-east, owing to the rapid fall in the surface
contours. Leaving Groenvlei near the point where the north-western boundary crosses the Sterk
River, the line appears to continue nearly due north across the middle of Eckstein to near the centre
of Leyden, where it probably passes under the Waterberg Sandstone.

Besides the special zone already referred to pegmatites frequently occur in the body of the
granite and especially near the contact with the felsites, good examples of the latter being seen along
the granite and felsite junction in the deep valley below the police post on Rietfontein No. 3, west
of Naboomspruit. Fine-grained veins are also frequent in many localities traversing the mass of
the granite.

The chief interest of die Red Granite lies in the association with it of the extensive tin-deposits,
to which much attention has recently been given throughout this area and in other parts of the
Transvaal. The special features of the granite associated with the tin occurrences are described
in a special Memoir* dealing with the tin-deposits.

Relationship of the Red Granite to the felsites. — The fact that the Red Granite usually shows
a tendency to finer grain in its upper portions and that these are directly succeeded by the felsites,
which, though more usually dense, with small phenocrysts, sometimes show a more distinctly
crystalline structure, has led to the felsites being frequently regarded as a marginal modification of
the granite mass. The intrusive relationship of the granite to the felsites found in the survey of the
Middelburg District! and other areas and described in previous Annual Reports of the Geological
Survey is however fully borne out by observations in the present area. The transition from medium
or fine-grained granite to a close-grained felsite is frequently perfectly sharp as far as can be distin-
guished by the eye and can be followed over long distances in the field. When the rocks are
unweathered, specimens showing the junction are easily obtained. Under the microscope it is
difficult to draw an absolutely sharp line between the two rocks, there being apparently a tendency
in the felsite towards a coarser crystallisation near the junction, which possibly in some localities
extends to some distance from it. It would, however, be remarkable, in the case of the intrusion of
so large a mass of granite into felsites of a very similar composition, if the junction were found to
be defined with absolute sharpness. It would be still more remarkable if the felsites, as a marginal
portion of the granite, showed no variation from a micro-crystalline texture with a few porphyritic

• H. Kyunston and K. T. Mellor.—'- The Geology of the Waterbei?; Tin -Fields." Memoir No. i, Geological Surrey,
Pretoria, 1909.

t E. T. Mellor.— " The Geology of the Neighbourhood of Rlieuoster Kop." Ann. Rep. Geol. Surrey, 1904, p. 63.

43

felspars for a thickness of several hundred feet, and changed to a medium-grained granite within
a sixteenth of an inch. The granite also frequently veins the felsites, or includes large masses of them,
whose edges are sharply and clearly defined. This is especially the case about the sharp bend in the
strike of all the stratified rocks west of Moorddrift. Good examples of the veining and inclusions
of the felsite by the granite can be seen near the common boundary of Lisbon and Zandspruit about
a mile from its southern end, on the rough slopes to the north of the footpath leading from Pot-
gietersrust to Geluk No. 897. Here many portions of the felsite, from a few inches to many square
yards in extent and with sharply defined margins, are included in the granite, which is a
moderately coarse variety becoming somewhat finer-grained near the contact with the felsite.
Similar veining and inclusion can be seen a little further south by following up the larger arm of
the spruit on Waterval No. 720 to within a hundred yards of the margin of the sedimentary rocks
which cover the western portion of that farm. Here in the bed of the spruit a small mass of felsite,
a few square yards in extent, is enclosed in the granite mass and also veined by it. In this instance,
however, the granite shows very distinct marginal modification, which comes on imperceptibly as
the granite is followed up the spruit, the extreme type being a hard felsitic rock with a purplish colour
and an indefinite spherulitic structure, entirely different from the felsites associated with the Lower
Waterberg and invaded by the granite. Near the margin of the felsites on Zandspruit No. 782,
about half-way across the farm, the felsites are traversed by narrow veins of medium-grained granite,
and in a similar position in the course of the spruit near the north-western boundary a wide tongue
of granite separates a broad band of felsites from the main body.

The relationships of the Red Oranite to the Waterberg System.— Many new data with regard to
the Red Granite and the Waterberg System have been accumulated in the course of the present
survey. They tend to show that the relationship between the two is not of a very simple nature
and point to the probable occurrence of several intrusions originating from the same or connected
sources extending over a very long period of time. So far, the felsites and shales of the Lower
Division of the system appear to be everywhere of earlier age than that of any of the granite
intrusions, as shown over extensive areas by the intrusive relationship of the granite to the felsites
and the extensive mineralisation that both the felsites and the sedimentary rocks associated with
them have undergone. With regard to the upper division of the Waterberg System, the frequent
occurrence of boulders of tourmalinised rocks in the conglomerates, referred to in more detail in
connection with those rocks, and of granite boulders in some localities, points to at least some of
the granite intrusions, including that which brought about the mineralisation of the Lower Division,
being earlier than the sedimentary rocks in question ; while on the other hand the clearly intrusive
character in the Upper Waterberg sediments of the granite of Gatkop, at the western end of the
Hoekbergen, described in a later portion of this Report, and in parte of the Middelburg District,
referred to in previous Annual Reports of the Geological Survey, shows that intrusions also occurred
subsequently to the deposition of much of the Upper Division of the System. Further, the character
shown by the Rooiwal intrusion, described in connection with the diabases of the district, indicates
that late in the Waterberg period or even subsequently to it, intrusions took place from a magma of
the same character as that which produced the Bushveld Plutonic Complex and which was capable
of supplying the materials for intrusions very similar in character to the Red Granite.

Diabosic intrusions. — Besides the Red Granite intrusion, which is mainly confined to the
felsites of the Lower Waterberg, igneous intrusions occur extensively throughout the Waterberg
System. In nearly all cases these are diabases or closely allied rocks, similar to those in the Water-
berg rocks in other districts which have been described in previous Survey Reports. They are nearly
always moderately coarse greenish grey to dark green rocks, composed mainly of plagioclase and
augite with some iron-ores. In the great majority of cases they take the form of intrusive sheets,
true dykes being less frequently found. This is the case even where the diabases are intruded into
the Red Granite and the felsites, a feature strikingly shown by the extensive diabases shown on
various farms between Solomon's Temple and Moorddrift. Of these perhaps the most conspicuous
is that on Solomon's Temple. This intrusion takes the form of a sheet dipping to the south-west
at about 5 degrees. Its sheet-like form is probably due to its having followed the platy structure
shown by the granite at this horizon. When the big ridge on Solomon's Temple is viewed in profile
its northern slopes present a number of step-like features. The diabase outcrop occupies the slopes
of the second of these from the top, the platy granite the edge of the steps above and below. Owing
to its nearly horizontal position the diabase sheet crops out on both sides of the ridge, the outcrop
being continuous round the northern end. At this point the diabase is about 100 feet in thickness,
but towards Welgevonden it is more nearly 300, After crossing the bridle path which leads from

44

Groenfontein to Welgevonden, the diabase passes up a small valley to the east and bo near the north
beacon of Welgelegen into the (ace of the granite ranges below the upper escarpments, and continues
across Blinkwater, possibly joining the intrusion shown on the southern portion of Lisbon. The
rock is typically a dark grey medium grained diabase. It is composed mainly of plagioclase and
abundant pale brown augite with a little quartz and some iron-ore. An interesting feature is the
occurrence of frequent patches of interstitial quartz and felspar in granophyric intergrowth. A
tendency to such growths, so characteristic of the Red Granite and related rocks, is shared by many
of tile intrusions in the Waterberg System, and is very common among the extensive diabasic
intrusions occurring in the neighbourhood of the railway line between Pretoria and Middelburg.
The feature is still more conspicuously shown by the intrusion on Rooiwal No. 317 described below,
and makes it extremely probable that the two are of the same age, that is, later than Upper Waterberg,
a point of some interest since the big sheet on Solomon's Temple is said to " cut out the tin." In
view of the probable age of the diabases, however, this appears unlikely and any diminution in tile
frequency of tin occurrences in the neighbourhood of the diabase must probably be attributed to
other causes. The sheet-like form of the diabase intrusion further south on Zandspruit No. 78U
is very well shown by its behaviour at the bead of the valley of the main spruit and by the way
in which, in the prominent kopje which stands between two branches of the spruit, a small cap of
granite lies on the diabase sheet whose edges form the slopes below.

The diabase, whose outcrop ascends from the flats south-west of Potgietersrust by the north
beacon of the farm Moord drift to the level of the top of the ranges, traverses in turn granite, felsite,
and Waterberg sandstone, and is continued as a sheet among the sedimentary rocks.

By far the most interesting of the diabasic intrusions in this neighbourhood is that which
covers large areas on Rooiwal No. 317 and Rietfontein No. 375. The local sandstones have a shallow
synclinal arrangement and the diabase is intruded between the beds, for the most part following
the bedding-planes for considerable distances. As, however, the bedding in the coarse sandstone
of this series is often of a lenticular character with false-bedding also conspicuously shown, the thick-
ness of the diabase appears to vary considerably, and its outcrop does not show the regularity which
might be expected if the intrusion followed a better defined horizon. The sandstones project into
the diabase as tongues or are exposed as inliers below it. A conspicuous little flat-topped kopje to
the west of the spruit, in the south-west comer of Rooiwal, carries a small outlier of sandstone
forming a thin cap lying horizontally on the diabase. The interesting feature of the intrusion,
however, is the extreme degree of differentiation shown by various portions, ranging from a thoroughly
typical dark grey coarse diabase to bright red granophyre and other granitic phases, which can
scarcely be distinguished in hand specimens from portions of the mass of the Red Granite itself.
The acid phases of this intrusion are probably the basts of the statement sometimes made by local
prospectors that the Red Granite is exposed on Rooiwal.

The general distribution of the different phases is shown on the map. The basic portions
form the lower lying parts of the area occupied by the intrusion, the more acid form low ridges, as
in the centre of Rooiwal, or considerable kopjes rising seventy or eighty feet above the general level,
as on the southern portion of Rietfontein No. 375. The most conspicuous mass is tile one lying
about a mile and a half east of the Sterk River. In this case, as the acid phase shown in the kopje
is separated from the diabase of the flat by a zone of deep red soil, tile connection between the two
is not apparent, but in the example further west, at a spot close to the bank of the Sterk River, the
transition from basic to acid types can be clearly seen, and examples of all the intermediate phases
obtained. The change from a greenish grey basic looking rock to a red granophyre takes place in a
few feet. The more basic phases of the Rooiwal intrusion are dark grey medium grained diabases,
consisting mainly of lath-shaped plagioclases and pale brownish augite. Occasional interstitial
patches consist of a granophyric intergrowth of quartz and felspar, and a few flakes of brown biotite
and grains of iron-ores also occur. A slide cut from a specimen in the transition zone is very similar,
with the exception that the ferro-magnesian constituent is an abundant dark green pleochroic
amphibole, and quartz is present in noticeable quantity in addition to that occurring in granophyric
intergrowth. A specimen of the more acid portion of the intrusion from the kopje is a red medium
grained rock with both red and white felspar and numerous very slender crystals of amphibole up to
one-third of an inch in length. In thin section it shows numerous lath-shaped porphyritic orthoclase
crystals surrounded by fine micrographic quartz and orthoclase, with the larger interstices filled
with a coarser granophyric intergrowth of the same minerals. There is a small proportion of a
green amphibole, much decomposed, showing occasional intergrowth with the felspar and also »
little greenish biotite.

45

Other portions of the acid phase closely resemble the granophyric portions of the Red Granite,
both the felspars and quartz being coarser and the latter often standing out on the surface in con-
spicuous grains. The ferro-magnesian mineral in this variety occurs in irregular patches instead of
in scattered slender individuals.

A further interesting feature of the intrusion is the occurrence in it of included well rounded
pebbles of fine-grained yellowish white quartzite. Several specimens ranging from one to four inches
in diameter were found in the largest kopje. These would appear to have been derived by the intrusive
rock from the conglomerates below the sandstones in which the intrusions occur, and may be com-
pared to those occurring more abundantly in the dyke referred to below on the farm Jan Boven Jan.

True dykes are not very common in the northern part of the area. One likely to be seen by
most visitors to the district is that which has been to some extent exposed in a deep excavation made
on Messrs. Maggs and Munroe's property on Zaaiplaate for the purposes of a dam. This appears
to continue south-westwards along the course of a very straight valley, leading to the Sterk River,
but outcrops are difficult to find.

In the area about Doornhoek No. 896 and Welgevonden No. 1772 and northwards to the
plateau, dykes appear to he of much more frequent occurrence. The big diabasic intrusion on
Welgevonden and Gemsbokfontein has on the former farm the characters of a dyke, which appears
to follow a very strong line of faulting, already referred to in connection with the structure of the
district. The whole intrusion has probably followed some of the main lines of fracture connected
with the disturbed nature of the country in this locality. The long straight kloof which gives access
by a bridle path from the Sterk Valley to the Palala River over Jan Boven Jan No. 481 follows the
course of a considerable dyke. At the head of the kloof the stream is crossed by a smaller dyke,
which stands up in wall-like fashion above the surface. ' It consists of a rather fine-grained dark
grey rock of diabasic type crowded with quartzite pebbles similar to those in the conglomerates
associated with the sandstones traversed by the dyke, and no doubt derived from them.

6. Thb Karroo System.

The Karroo System is represented only by the Bushveld Sandstones and the associated Buahveld
Amygdaloid, which cover the low lying tract crossed by the railway on the edge of the Springbok
Flats. The sandstones also form a thin covering to the small plateau, which lies west of the main
road near Naboomspruit Station.

The Bushveld Sandstones. — The Bushveld Sandstones have here precisely the same characters
as in other parts of the Springbok Flats.* They consist of massive beds of yellowish or reddish
sandstone, of uniformly fine texture, distinguished by their peculiarly sharp feel, by the rarity and
indefinite character of the bedding planes, and by their tendency to weather into peculiar hummocky
surfaces, often covered with small pits and fantastically shaped projections.

Owing to their comparatively soft nature and to the heavy covering of yellow sand which
arises from their disintegration, outcrops, especially on the more level tracts, are rare, the surface
of the sandstones being occasionally visible above the sandy covering, but rarely rising more than
two or three inches above the general level. As a rule the sandstones can only be traced by the heavy
covering of fine yellow sand they produce, which carries a characteristic growth of trees of the syringa
type of larger growth than those found in neighbouring localities. The best outcrops are to be seen
west of Naboomspruit Station along the courses of the spruits draining Vlakfontein No. 2187, and
again further north along the spruit on the eastern portion of Rietvlei No. 1348. Still further north
in the northern comer of Klipfontein No. 116, the sandstones also rise to the level of the surface and
have been quarried to a small extent for use in the railway culverts. About a mile to the north of
this point on either side of the main road many loose blocks of coarse grit lie about the surface.
These grits are very similar in character to those of the coal-measures, which in other places fre-
quently underlie the Bushveld Sandstone and may indicate that the latter rocks are locally underlain
by a small thickness of such grits. As, however, the lowest beds of the Bushveld Sandstone, when
they rest directly on older rocks, are occasionally coarse or even conglomeratic, as in the case referred
to below, the grits in this case may only represent such basement beds, hut no outcrop showing their
exact position or thickness was met with.

The Bushveld Sandstones of the Springbok Flats are frequently terminated by a line of
faulting, west of which the sandstones again occur at a somewhat higher level. Such an elevated

E. T. Mellor.— "The Sandstones oi BuiFkop and the

portion of the sandstones covers the plateau on Rietfontein No. 3 and Buffelskloof No. 704, while
detached remnants form outliers crowning granite kopjes on the western edge 'of Naboomspruit
No. 2297 and about the southern corner of Koedoesfontein No. 309.

The western edge of the sandstone plateau, where it is deeply cut into by the head of Tobias
Spruit on Rietfontein No. 3, affords many excellent sections of the base of the sandstones showing
their junction with the Red Granite below. An interesting feature in the sections is the presence of
a conglomerate of variable thickness, but sometimes as much as fifteen feet, at the base of the
sandstones. In its dark purplish colour and in the character of its pebbles, this conglomerate very
closely resembles the Waterberg conglomerates, from which it is hardly to be distinguished in hand
specimens. It is, however, not so compact, fractures passing round the pebbles instead of through
them. Many of the pebbles are sub-angular and nearly all are of dark purplish felsitic rocks ; they
average about one inch in diameter, rarely exceeding two.

Immediately above the conglomerate in this locality there is about ten feet of brick-red sand-
stone in beds one to three feet in thickness. Where it lies directly on the granite, owing to the
thinning out of the conglomerate, this sandstone is frequently so thoroughly silicified as to form a
brittle glassy pink quartzite. The silicification has taken place from the joints, the portions remote
from these being comparatively soft. Above this bedded sandstone comes the usual fine-grained
yellowish or reddish sandstone with little or no trace of bedding, which covers the plateau.

A narrow strip of the Bushveld Sandstones forms a small escarpment at the constabulary
post on Rietfontein No. 3, and continues for some distance along the northern side of the road, which
is here cut in shales belonging to the Lower Waterberg. The very interesting section seen along
the road at this point is shown in the section LMN, Plate XI. The low dip shown by the shales
in places and the coincidence of their strike with that of the edge of the sandstone escarpment suggests
that both belong to the same series of beds. Their true relationship, however, is that shown in the
section referred to, the Bushveld Sandstones resting unconformably on the Waterberg shales and
felsites. About half a mile further north on the descent towards Rietfontein No. 1638 the road
passes close to a small outlier of the Bushveld Sandstones, in which the conglomerate at the base is
well displayed.

The Bushveld Amygdaloid. — From Naboomspruit north-eastwards along the Railway line the
Bushveld Sandstones are covered by the Bushveld Amygdaloid, the rock having here the same
appearance and characteristics as those shown by it in other portions of the Flats, and described
in the Annual Report of the Geological Survey for 1904.

Here, as elsewhere, outcrops are exceedingly rare, but specimens may be obtained from the
wells sunk along the railway at Naboomspruit Station and Tobias Spruit, and in the north-west corner
of Olifanteklip No. 1325. The Amygdaloid has almost everywhere a deep covering of deep red or
black soil, often containing numerous calcareous nodules. Along its margin there is frequently a
broad belt of laterite or " ouklip," which covers its junction with the underlying sandstones.

7. Ecohomic Geology.

Tin. — During the past three or four years considerable developments have taken place in
the Waterberg District in connection with the occurrences of cassiterite, which have been found in
that area, mainly distributed along the zone occupied by the Red Granite and the felsites and
associated shales of the Lower Division of the Waterberg System.

What it is hoped will subsequently prove a very important industry can scarcely et be said to
have passed much beyond the stage of prospecting and development. The disappointing results which
followed the sanguine expectations in connection with some of the earlier finds of tin in the Bushveld
have at least had the beneficial effect of preventing unduly optimistic statements being made con-
cerning them in the Waterberg District. Already, however, up to the end of 1908 the producing
mines on Zaaiplaats, Groenfontein, and neighbouring farms have exported tin to the value of
£146,849, and several other enterprises promise soon to give satisfactory results. In a country where
tin-mining had previously received little attention much had to be learnt, and the cassiterite deposits
already found show bo much diversity in mode of occurrence that a good deal of local experience
is a necessary preliminary to successful work. The experience once gained, however, will probably
be of value not only in this district, but also in other tin-bearing localities, which, from the extensive
occurrence of the Red Granite and associated rocks, may reasonably be expected' to be found in the
future. Occurrences of cassiterite, which have either been proved already to be of importance or
which justify perseverance in their investigation, have been found over a large extent of country
from Zaaiplaats to the Rooiberg. The distribution and mode of occurrence of these deposits have

47

already been dealt with in detail in a special Memoir of the Geological Survey,* to which reference
should be made. In the present report a brief summary only will be given of the more important
features of the cassiterite deposits occurring in the area dealt with.

The most productive up to the present time, and in many respects also the most interesting
of the tin-deposits, are these worked on Zaaiplaate, Groenfontein, and the adjoining farms. The
cassiterite occurs in the Red Granite which here shows certain characteristic features, the most
striking of which is its differentiation into mote or less well defined zones, whose special characteristics
appear to be related to their depth below the upper limit of the granite intrusion (see
Fig. 4). The granite, in which the tin occurs most abundantly, is characterised by its massive

TION "cd,"-SEMI-DIAGRAMMATIC SECTION ACROSS TIN-BEARING ZONE NEAR AND APPROXIMATELY
PARALLEL TO SOUTH-EASTERN BOUNDARY OF ZAAIPLAATS.
(Portion of Section "ABCD," Plate XII, Fig. 1.)

ssw

NNE

%•* Brr*

.

t- t- +■ + + + + + + + + tsttZ^X?
+ + + + + + + + + + +^*^ , +

&*% + + -T- + 4- + + + + + + -T- + + -

+ + + + + + + + + + + + + + + +

l cmm on*. r. •-r*K"-«Mi a. w.n
I a— *•> aw «. -9m*i —few:

'■'

T"...t*. .? "!""' "«?• "V *t» 1"

ft . . 1^,1 R -Jfc- 1 -kf

Fig. 4.

nature and tendency to weather into rounded bare domes and humps, strewn with large boulder-like
masses. This type is succeeded by a finer-grained variety usually granophyric, showing on a large
scale a pseudo-bedded structure, causing it to weather out into terraces and small escarpments and
to produce more or less angular blocks of debris. At a still higher horizon this bedded structure
is not noticeable and the granite weathers more like the massive type, but with all its features on a
much smaller scale. The horizon which includes the lower portion of the bedded type of granite, is
marked by the very constant occurrence of coarse pegmatite, which frequently shows the pseudo-
bedded structure in a marked degree, and also a tendency for the large quartz crystals to orient
themselves at right angles to the divisional planes. In these pegmatites cassiterite in very coarse
crystals is of frequent but erratic occurrence and consequently unprofitable to mine. It was the
coarse cassiterite of the pegmatites which first attracted attention to the district.

The massive granite below the pegmatite horizon carries tin in peculiarly shaped bodies of
ore, known as " pipes " or " chimneys." These consist of bodies of highly altered and mineralised
granite, having in general a roughly cylindrical form but showing frequent changes of direction
sometimes at very sharp angles. These bodies also frequently branch, and in one instance two pipes
of somewhat different types intersect one another without losing their individuality. The diameter
of the pipes varies from about two to ten feet, and in some instances they have been followed to a
depth of over 300 feet. Their general appearance and relationships to the granite will be Been in
the section, Fig. 4, which also shows the relationship of the various modifications of the granite.
This section should be compared with that on Plate XII, Fig. 1, of which it forms a portion.

In cross-section the tin-bearing pipes often show a zonal arrangement, due to progressive altera-
tion of the granite from the centre outwards. Such sections exposed at the surface are known as
"eyes" when opened up; in some cases such "eyes" are found to terminate at a depth of a few
feet or to be portions of bodies containing no cassiterite, but in most cases they indicate the position
of a pipe. Examined from the circumference inwards the granite about the pipe shows successive
degrees of alteration, similar to those shown in Fig. 5, which, however, is not drawn from a typical

4*

pipe, but from a body of ore, the deposition of which has probably been much influenced by the
pronounced fissure shown.

The most conspicuous feature in many of the pipe sections is a border of varying breadth,
consisting almost entirely of nearly black tourmaline, which is composed of rosette-like groups, about

SECTION OF ORE-BODY (GROENVLEI).

+ hmwy + + +

^ '"*-/+ -f ' 4 4

•'4 4 4 4

+ PF + + + + 4

+ 4 4 + 4

4/444444

+ 4 + 4

1. Unahred medium -grained red granite.

2 r%rtially altered granite with a little ftuortte tending to part
from unaltered rock along dotted lint

3. Z ont of light greenish altered granite with Jencitijed felspar

4. Zone almost entirely composed of while quarts 16 inch wide with
toumia/int m lower wider portions

5. Central portion nearly alt of tourmaline with quart} fcassiter/te

6. Irregular patches of greenish chert/ silica

7. Continuation of fissure in unaltered granite

half an inch in diameter, of radially arranged slender prismatic or needle-like crystals. In some
cases the tourmaline occupies practically the whole space, in others the spaces between the rosettes
is rilled with white secondary quartz or the rosettes may be sparsely scattered through a quartz
groundmass. The central portion of the pipe section may be occupied by a compact mass of reddish

49

secondary silica, with some sericite, and only a very small proportion of tourmaline, the whole dotted
over with crystals of cassiterite, or it may be composed almost entirely of white quartz with an
abundance of cassiterite crystals, the mass showing numerous small drusy cavities, into which the
crystals of the various minerals project. Calcite, associated with a varying proportion of iron-
carbonate, is also frequently present. In some pipes these drusy cavities are more pronounced and
may become extensive " vughs," which show a great variety of accessory minerals of comparatively
late development ; fraorite, galena, zinc-blende, and arsenical and copper pyrites being especially
conspicuous. From pipes of the latter class some of the richest ore is produced.

Among the accessory minerals which occur in the tin area, although not necessarily moat
abundant where the richest tin-deposits have bo far been found, may be mentioned wolframite,
molybdenite, scheelite, orthite, and topaz. In some cases fissures are seen to be connected with the
pipes and the mineralisation then extends along the fissure. Throughout the greater part of their
length, however, no connection is traceable between the pipes and any Assuring of the granite, and
the occasional association of the two appears to be accidental. Although similar alteration of the
granite has in many cases taken place along clearly defined fissures, such as that shown in Fig. 5,
it is very difficult to establish such a connection in the case of the most typical pipes, whose origin
must probably be explained in some other way. The subject has been more fully discussed in the
Memoir referred to.

Occurrences of cassiterite similar in their main characters to those of Zaaiplaats.and Groen-
fontein are found southwards as far as Solomon's Temple. Northwards on Groenvlei, Appingadam,
and the adjoining farms, the examples hitherto found of typical pipe-like bodies are more rare,
cassiterite occurring occasionally in connection with irregular bodies of pegmatitic character in the
granite, which usually do not persist in depth. In some of these molybdenite is very conspicuous.
Of a much more promising nature are certain fissures or groups of parallel fissures traversing the
granite and usually striking a little north of west. Along these the granite has been altered to a
variable width from one inch to two feet or more in a somewhat similar manner to that shown in the
pipes. The tin-bearing rock consists mainly of tourmaline, arranged in radial groups about a
quarter of an inch in diameter, and quartz, or more frequently finely crystalline tourmaline, without
any clearly marked radial arrangement, carrying cassiterite crystals from Vsr of An mc& down-
wards. Very rich patches occasionally occur with a large proportion of more coarsely crystalline
tourmaline, and these are frequently found as outcrops on the surface or as " float," appearing to
resist weathering more effectively than the rest of the material occurring in the fissures.

On Welgevonden No. 777 and Welgelegen No. 357 prospecting, which at the time of this survey
was not far advanced, has shown the existence of occurrences of cassiterite, which are probably
closely associated with the contact plane of the felsite and intrusive granite. Along this horizon both
the granite and felsite are either brecciated or traversed by numerous narrow fissures usually not
exceeding a quarter of an inch in width, filled mainly with fine fibrous tourmaline associated with
cassiterite, which also occurs in small crystals scattered through the granite near the contact. Small
flat bodies of ore containing rich patches of coarse cassiterite also occur, which appear to have been
developed along joint planes in the granite and felsite.

In the extreme south of the area much work has been done on the farms Doornhoek No. 896
and Welgevonden No. 1772, on both of which cassiterite is found to occur in small veins and in more
definite lodes traversing the hard shale associated with the felsite (see Fig. 3). The lode
first discovered on Doornhoek differs from many tin occurrences in being very clearly denned,
a feature which should greatly facilitate economical working. Over a distance of 1,250 feet the shales,
which are extensively impregnated with fine tourmaline, are traversed by a well defined fissure
averaging about 1£ feet in width, dipping northwards at from 50 to 60 degrees, and cutting
obliquely at an acute angle across the betiding planes of the shales, which dip northwards at an angle
of 45 degrees. Near its eastern end the lode cuts more abruptly across the bedding planes.
The filling of the fissure, in addition to many brecciated fragments of the country rock, consists
mainly of fine tourmaline and chloritic and sericitic material, traversed in places by small veins of
quartz. In many cases the lode matter is a dense greenish black mass, in which no individual con-
stituents can be picked out by the eye. In places this type of material is rich in very finely divided
cassiterite. The tin contents of the lode as a whole, however, appear to be very variable in quantity
at different points, and the well-defined character of the fissure itself does not appear to exempt it
from the irregularity in the distribution of the cassiterite shown by other tin-deposits, rendering an
estimation of their value only possible after considerable development. A second less definite line
of fracture in the shale series more recently discovered and known as the " New Strike," follows
5

50

very closely the strike of certain beds, and also agrees with them in dip, so fax as was shown by the
woikingB to die depth of ten or twelve feet opened up at the time of the writer's visit. Along this
line these beds are fissured and broken and show in places deposits of coarse cassiterite yielding very
rich sampler

In the felsites underlying the shale series on Doornhoek cassiterite also occurs distributed in
small crystalline masses or " nests," averaging about half an inch in diameter, associated with
modified ovoid patches in the felsite, a few inches in diameter and with a central portion darker from
the presence of tourmaline. The impregnation of the felsite with tin extends over an area many feet
in width ; the indefinite nature of such deposits, however, will probably render them difficult of
exploitation, compared with the more easily followed fissures.

Gold. — Before the discovery of tin in the district, prospecting had been carried on to a con-
siderable extent for gold, and somewhat extensive workings exist on Welgevonden No. 1772. In
this neighbourhood the felsites particularly are traversed by numerous reef-like bodies of brecciated
felsite and quartz, which appear to follow lines of faulting. When the country rock is bare of grass,
the outcrops of these reefs are very conspicuous on account of the quantity of quartz shed over the
surface. The reefs were described in bis description of the Waterberg District by Dr. Molengraaff,*
who states that besides brecciated felsite they contain quartz, nuorite, and specular iron, and gold
amounting in some places to not less than 2 ozs. per ton, but very variable. Since the early work
on these occurrences many years ago no further developments have resulted.

The conglomerates at or near the base of the Waterberg Sandstones have also been prospected
for gold, which is sometimes found to be present in small quantities, and may become of importance
where locally concentrated by denuding agencies as on Grootrivier No. 2321.

A very common feature of nearly all the quartz veins found in the rocks of the Waterberg
System is the presence of abundant specular iron. The small quartz veins containing specularite
have often been worked by local Kaffirs, the powdered mineral being valued by the women for pur-
poses of personal adornment. Such workings bear a close resemblance to those which in some
localities, like the Rooiberg, have been such good indicators of tin occurrences. The workings for
specularite, however, can usually be identified as such by the numerous fragments of quartz and
specular iron found among the refuse.

' Annunl Report of the State (Jeolnpist, 1898, pp. 23-2

IV.

THE GEOLOGY of the HOEKBERGEN

IN THE WATERBERG DISTRICT

INCLUDING GATKOP.

By B. T. Mellor (Geologist).

CONTENTS.

1. Introductory.

2. Topography.

3. Geology.

4. The Structure of Gatkop and the Western End op the Hobkbergen.

5. Summary.

IV.— THE GEOLOGY OF THE HOEKBERGEN IN THE WATERBERG
DISTRICT, INCLUDING GATKOP.

By E. T. Mellor (Geologist).

1. Introductory.

The name Hoekbergen is given to a well marked and conspicuous range of hills which traverses
part of the Waterberg District, lying to the west and north of Nylstioom. Although the
neighbourhood of the Hoekbergen has not yet been mapped in detail by the Geological Survey,
various traverses have been made in connection with the study of the Waterberg System to
the north-east. The results of these traverses are interesting as showing the continuation westwards
from Nylstroom of a geological structure closely similar to that prevailing to the north of the same
town described in another portion of this Report, and in bringing out certain noteworthy structural
features developed in the neighbourhood of Gatkop and Sunday River Poort, near the Rustenburg
Border, which throw some light on the geological conditions prevailing along the margin of the
Bushveld.

SECTION FROM THE WATERBERG PLATEAU SOUTHWARDS TO

THE HOEKBERQEN

"■ ON ROOKPOORT 1836-

•■

«'.'"«"■

— ^S^, t

fett^—

—r"

^^j"-^r^s* r ^ i ^ :r ^^ J ^~" tk

^"^ ^^ = rz^ r -^ r ~~r^ ^ ^^-''^'-^^^^'■^^^'''//^

Sl'-YffiHAA L*1* 1*

•*•***

f.Saa/sfam. 1 1 „,„.___-

»

2. Topography.

The Hoekbergen form a range of hills rising some 1,200 feet above the adjacent valleys, and
extending from the valley of the Zand River, north of Nylstroom, westwards into the Rustenburg
District. The range is really a continuation of that already described in the preceding Report on
the area west of FotgieterBrtutt, as running parallel to the edge of the main Waterberg plateau from
near the junction of the Magalakwin and Sterk Rivers, by Potgietersrust to the neighbourhood north
of Nylstroom, where its persistence as an individual topographical feature overlooking the low country
of the Springbok Flats, is perhaps not so evident, owing to the existence of the outstanding tract of
hilly country about Nylstroom, which is due to a local synclinal arrangement of the Waterberg rocks.

The correspondence between the- Hoekbergen on the south of the Waterberg plateau and
the ranges lying east of it between the Sterk and Magalakwin Valleys holds good both topo-
graphically and from the point of view of geological structure, as will be readily seen by a comparison
of the section in Fig. 6 with those in Pkfcea XI and XII.

It will be noted in the Rookpoort section, which is very clearly exposed in the narrow gorge by
which the spruit from Buffelskloof cuts through the Hoekbergen, and in which almost every bed of the
lower portion of the section is visible, that the lowest beds, consisting of conglomerates and hard red

54

quartzites, are over-tilted to 70 degrees as in Gatkop, the dip, after becoming vertical, gradually
decreasing to a very low angle in the flat country to the north and in the escarpments of the plateau.

Like the more northerly ranges, the Hoekbergen consist of the volcanic rocks of the Lower
Division of the Waterberg System, with the overlying conglomerates and sandstones. Like the
northern ranges abo they overlook on one side the low country of the Bushveld, while on the other
they are separated by an intervening belt of low sandy country from the margin of the main Water-
berg plateau, which rises into the long line of precipitous escarpments known as the Zandrivier-
bergen, and which is practically continuous from this locality into the more northern area. Near
the boundary of the Rustenburg District, on Donkerpoort No. 945, the Hoekbergen range rises into
the exceptionally fine peak of Gatkop, which attains an elevation of 2,800 feet above the base of the
mountain.

Opposite the central portion of the Hoekbergen the lower slopes of the range pass almost
uninterruptedly into the low lying country of the Bushveld, but at its eastern extremity the hills
running north-westwards from Warmbaths approach it, while on the west several smaller ranges,
including the Rooiberg and Elandsberg, intervene between the Hoekbergen and the flat Bushveld
country.

3. Geology.

The existence of the Hoekbergen, as is the case with the corresponding ranges further north-
has resulted from the resistance to denudation offered by the hard felsitea of the Lower Waterberg,
which form as a rule the southern slopes as well as the crest of the range. Along the foot of the
southern slopes the Red Granite, intruded below the felsitea, extends most probably without inter-
ruption from Rhenosterpoort No. 2145, on the east, to Zandspruit No. 944 on the west. At either
extremity of the range the outcrop of the granite becomes comparatively narrow, but opposite the
central portion it is continuous with, a broad stretch of granite country interrupted only by occasional
isolated masses of sedimentary rocks very variable both in strike and dip.

On the northern slopes of the Hoekbergen the felsites are overlain by the conglomerates and
sandstones at the base of the Upper Division of the Waterberg System, which occasionally rise to
the crest of the range and form its uppermost escarpments, as on Rietpoort No. 913 and Rietfontein
No. 2267.- A softer series of very thick-bedded coarse grits and sandstones, which yield readily to
weathering agencies, forms the low-lying belt north of the range, while a harder Beries of somewhat
qunrtzitic sandstones with conglomerates forms the escarpments of the main plateau. With regard
to the succession of beds shown by the Waterberg System, there is little to add to the description already
given in previous pages for the corresponding rocks in the area west of Potgietersrust. The felsites
are quite similar in character, the conglomerates show similar variations in thickness, and it is difficult
to detect any important difference in the succeeding series of Bandstones, except perhaps that the
beds, which form the low ground between the outer ranges and the plateau, tend to become coarser,
more massive, and more felspathic in composition, and suggest more Btrongly a derivation from an
area of rapidly denuded granitic rocks. Near the eastern end of the range an anticline, whose axis
runs east and west across Rhenosterpoort No. 2145, results in the felsite and conglomerate ridges
being repeated to the south in the hills which extend towards Warmbaths, the conglomerate outcrop
passing from Rhenosterpoort almost due southwards to the valley of the Buffelspruit which it follows
as far as Roodepoort No. 2148 near Warmbaths, where it turns more directly eastwards. At the
western end of the Hoekbergen, however, disturbances of a much more pronounced character have
taken place, resulting in faulting on such a scale that the strata of the Transvaal System now overlie
the upturned edges of the Waterberg rocks. These disturbances appear to culminate in the neigh-
bourhood of Gatkop and their geological interest is such as to warrant a more detailed description.

The neighbourhood of Gatkop has been described by E. Jorissen* in a paper communicated
to the Geological Society of South Africa. The writer, however, finds himself in disagreement on
many fundamental points with the description there given of the conditions presented on Gatkop
and their bearing on the geology of the Hoekbergen.

4. The Structure of Gatkop and the Western End of the Hoekberoen.
The mountain known as Gatkop (" Cave Mountain ") derives its name from a very extensive
cavern in the Dolomite at its base, which served as a refuge to the natives in early Kaffir wars.

j the north-eastern part of the Ruatenbnrg

55

Gatkop forma one of the most conspicuous and easily recognised landmarks in the district,
rising to a height of 2,800 feet above the valleys at its base. From most parts of view, and especially
from the south-east the mountain presents a very striking and picturesque appearance, its graceful
and varied outlines, which terminate in a double peak, being in marked contrast with the somewhat
monotonous ridges or continuous kranzes which characterise most of the adjacent ranges. The
summit of Gatkop lies about half a mile south-east of the middle point on the western boundary of
the farm Donkerpoort No. 945. The uppermost portion for about 600 feet below the summit consists
of precipitous kranzes, from the foot of which very steep slopes descend abruptly to a narrow poort,
by which the spruit flowing from Buffelskloof to join the Sand River passes through the Hoekbergen.
South-westwards from the summit the mountain is prolonged into a long ridge of diminishing height
for a distance of about four and a half miles to the Sunday River Poort, by which the Sunday or
Sterk River passes southwards through the range, the latter being continued beyond the poort in a
westerly direction towards the Crocodile River.

On the south side the slopes of Gatkop everywhere descend to the level of the Sand River
Valley, but on the north, a broad shoulder bridges the wide gap which usually separates the
Hoekbergen and corresponding ranges from the Waterberg plateau.

In the description of a portion of the Waterberg District lying west of Potgietersruat, given
in the preceding report, reference has frequently been made to the persistence of this feature, which
takes the form of a broad depression like a continuous valley (see Plate XII, Fig. 3). This
depression affords an easy route for many of the roads which skirt the plateau, and when its con-
tinuance without any notable interruption has been observed over a distance of nearly 100 miles
from near Zaaiplaats to Gatkop, and it is found again west of the mountain, the interruption opposite
Gatkop at once attracts attention. More especially is this the case, since being practically impassable
except on foot, the presence of this abnormal feature prevents travellers from Zwagershoek passing
directly round the margin of the Waterberg plateau to the head of the River Matlabas, and so joining
the main road to the north. A long detour to the south of Gatkop, passing twice through the
Hoekbergen, is thus necessitated.

As the depression referred to follows everywhere approximately the same geological horizon,
its sudden interruption at Gatkop suggests at once the presence of some unusual factor in the local
geological structure. As will be seen from the description which follows, such a factor does exist,
the series of soft sandstones, to the erosion of which the depression is due, having been locally pro-
tected from denudation by a covering of harder rocks which still in part remains.

From the summit of Gatkop, which commands magnificent and extensive views in every
direction, the strongly marked outcrops of the Waterberg System can be followed by the eye for a
distance of many miles. Especially conspicuous are the long lines of sandstone escarpments which
form the margin of the plateau, separated by the low Bandy zone from the almost equally con-
spicuous ridge of the Hoekbergen formed by the felsites of the Lower Waterberg and the overlying
conglomerates. The main escarpments of the plateau, trending almost due east and west, are
continued without interruption to the north of Gatkop, the low sandy zone on their south side,
passing into the base of the northern portion of the mountain, while the felsite and conglomerate
ridge of the Hoekbergen, interrupted only by the narrow poort on Donkerhoek, passes into the central
portion of the mountain, bo that the strata of which the Hoekbergen ridge is composed appear to
traverse the mass of Gatkop and pass beneath its summit. The summit itself, however, consists,
not of Waterberg rocks, but of the characteristic banded magnetite-quartzites referred to the Pretoria
Series, which thus come to overlie the younger Waterberg rocks. That this is actually the case is
fortunately conclusively demonstrated from the remarkably clear section given by the eastern end
of the mountain, shown in Fig. 7, all the details of which are exposed in the deeply dissected slopes
above the poort on Donkerpoort. Even from below it is possible to follow the main structural lines
shown in the section, after the details have been followed on the ground, a process
unusually laborious owing to many of the important parts of the section lying at elevations of
approximately 2,000 feet above the base of the mountain.

It will be seen from the section given that the Waterberg Beds which. run into the base of
Gatkop have suffered a considerable degree of folding, resulting in very high dips at the base of the
sandstones gradually diminishing in the upper portion of the series, a structure very commonly met
with at the margin of the Waterberg System. The basal conglomerate and the beds immediately
succeeding it are here in fact overtilted in places to an angle of 70 degrees, a condition well seen
about the foot of the main gully which descends the eastern face and joins the spruit at the poort.
This gully offers one of the best traverses of the eastern end of the mountain. The sharp flexure

56

in the sandstones resulting in some places in a number of small faults, accompanied by i
small quartz veins, is well exposed in the spur which forms the northern side of the gully referred to.
Towards the top of the mountain the upturned edges of the Waterberg strata are covered by
the mass of banded magnetite-quartzites which form its crest. Beneath the northern peak at an
elevation of about 2,000 feet above the base, the banded quartzites of the Transvaal System rest
directly upon massive beds of Waterberg grits, while below the south peak the rocks of the two
systems are separated from one another by an intrusion of granite in the manner shown in the section
in Fig. 7. The eastern end of Gatkop thus shows with exceptional clearness a strongly marked over-
thrust fault, along which the banded quartzites of the Pretoria Series have overridden the edges of
the Waterberg strata. Along the fault plane has been intruded a mass of Red Granite which now
partially separates the rocks of the two systems. This granite mass forms the whole of the spurs
which descend from just below the southern peak to the poort, and is continuous with the broad
stretch of granite which occupies the valley to the south-east of the mountain on Donkerpoort No.
945 and Zandspruit No. 953. The crest of the mountain is everywhere of banded quartzites. In
the gully descending to the poort, already alluded to, the contact of the granite with the Waterberg
conglomerates and sandstones is well exposed and can be closely followed almost from bottom to
top, and single hand-specimens which include both igneous and sedimentary rocks can be obtained.
The well marked marginal phenomena shown by the granite leave no doubt as to its intrusive relation-
ship to the sediments, and its transgression from lower to higher horizons of the nearly vertical beds
is clearly seen. Passing round the north side of the mountain at an elevation of 1,500 to 2,000 feet,
the banded quartzites, with the exception of one locality where a small mass of dolomite a few feet
thick intervenes, continue to rest directly upon the younger Waterberg rocks, as far as the head of

Fig. 7.
the most southerly of the three chief spruits draining the north-western slopes of the mountain.
Here the dolomite appears below the banded quartzites, separating them from the Waterberg rocks.
For a short distance the dolomite is in contact with the latter rocks along the plane of the overthrust,
but a little beyond this point the two are separated by a thin tongue of granite intruded along the
fault plane, which can be traced easily as far as the Sunday River, where it appears to die out. The
Waterberg rocks, originally faulted against the Dolomite along this line, are the big conglomerates
at the base of the Upper Division, which, disappearing at the eastern end of the mountain under the
south peak, emerge again near the point where the Dolomite also appears. Parallel to the thin sheet
of granite, intruded along the fault between the Dolomite and the conglomerates, a second and thicker
sheet having conglomerate on either side occurs. This intrusion has an outcrop of considerable
breadth, and is well exposed where it descends to the hat on the north side of the Gatkop ridge, near
the middle of the south Bide of BufTelshoek No. 799. If followed eastwards from this point up the
slopes of the mountain along the course of a gully which follows for some distance the junction of
the two rocks, the surface of the granite can be clearly seen dipping south below the conglomerate
at an angle of about 30 degrees. The granite appears to have been intruded along a fault plane
parallel to that between the Dolomite and the Waterberg conglomerates, only a thin band of the
latter intervening. This intrusion can be traced to within a short distance of the road on the east
side of the Sunday River. About the river and for some distance beyond, it is hidden by surface
deposits, but a band of granite of the same character and occupying an exactly similar position is seen
about three-quarters of a mile west of the river, on the northern flanks of the big ridge which crosses the
middle of Waterval No. 965. At this place the outcrop of granite has still conglomerate on either

side, the beds on the north dipping northwards at an angle of 15 degrees, while those on the south,
overtilted against the Dolomite dip southwards at 80 degrees, as shown in the sections in Pig. 8.
Finally the granite breaks across the latter beds, and, after following the fault plane between
the conglomerate and Dolomite for a short distance, thins out, leaving the two again in contact.
The same granite intrusion is shown both at the eastern and western ends of the mountain by
Mr. G. G. Holmes* in the map illustrating his paper on the geology of a part of the Rustenburg
District, dealing mainly with the area west of Gatkop. From Mr. Holmes' description it appears
probable that the thinning out of the granite referred to above is only local. Mr. Holmes showB that
it continues much further west.

Throughout the whole length of the Gatkop ridge the faulted character of the junction between
the rocks of the Transvaal System and the Waterberg strata is very clearly shown, and where the
two come into direct contact without the interposition of the intruded granite, shearing is common.
The overthruBt character of the fault is also shown with exceptional clearness, and can be followed
foot by foot in a traverse from either end of the mountain to the other. It is rare to find the details
of faulting of this character exposed with the unusual clearness shown by the section afforded by

DIAGRAMMATIC SECTIONS ACROSS JUNCTION OF TRANSVAAL AND
WATERBERG SYSTEMS ON WATERVAL 966.

KIDDLE OF N.C MM OF W.TEF.VAL 91,5

"_ .^^ ^^^

-™£Srm*- s ^S««;.;.'.;.-.:.\?ra.- •'

NUH MNTM Of W.TEBVAL lit

,=s -SsfelEi— - ! ?-7-r-?0W^*^' '

t. Dolomitt 2 Wattrberg Coitglomirur§s. 3. Granite

Fig. S.

the steep slopes of Gatkop. The existence of the mountain itself is a result of the protection afforded
by the comparatively thin covering of hard banded quartettes to the underlying mass of more easily
eroded Waterberg sandstones. That the overthrust quartzites once extended still further northwards
is shown by the existence on this side of the mountain, far beyond the present range of the debris
arising from their disintegration, of several large accumulations of blocks and fragments representing
the remains of former screes. The existence also of the elevated shoulder previously mentioned,
which connects the main body of Gatkop to the Waterberg plateau, must also be attributed to a
former northerly extension of the overthrust rocks. The extent of the overthrust shown in the
section on Fig. 6, so far as it can be actually seen, is not less than one mile, and when the felsites,'
which probably underlie the conglomerates, are taken into consideration, the distance must be very
much greater. Overthrusts of this type are not very common in Transvaal geology as far as it is
at present known. A tendency towards such a structure appears, however, very common along the
margin of the Waterberg System, both in the Waterberg and Middelburg Districts-! The statement

i.ii. i-ii

58

is frequently made that low dips or a nearly horizontal disposition are characteristic of the
Water berg System. On the contrary, especially in the lower portions, its strata are frequently found
vertical and not uncommonly overtilted.

The geology of Gatkop presents so much interesting detail that a full account of it would
necessitate a very lengthy description. In addition to those already dealt with, the main features
only, in bo far as they are of general interest, are given below.

The banded magnetite-quartzites which form the rugged ridges of the upper portions of Gatkop
show the results of much disturbance, partly the result of folding consequent on the extensive move-
ments in which they have been involved, and also probably due in part to the removal by solution
of the dolomite from below them in the manner so characteristic of that formation. The broad mass
of quartzites which crowns the main mountain may be said to be made up of two portions dipping
towards one another. The edges of these masses form two approximately parallel ridges running
eastward to the twin peaks of the summit. Here the beds of the northern ridge curving southwards
in horse-shoe fashion come into connection with those of the southern ridge, the whole mass thus
forming a faulted syncline with one end closed. The beds of the southern limb dip to the north-west
at from 30 to 40 degrees, those of the northern to the south-east at 50 degrees. This limb
is the more persistent and continues almost uninterruptedly from the summit south-westwards for
nearly five miles to Sunday River Poort, forming the main ridge of the mountain. The dip becomes
much steeper towards this poort, at which point the beds are almost vertical. In the highest parts
of the mountain the banded quartzites rest either upon the Waterberg Conglomerates and Sandstones
or upon the granite. In the long ridge by which Gatkop is continued to Sunday River Poort they
are underlain by Dolomite, whose outcrop continually increases in width towards the Sunday River.
On the south of the banded quartzite the Dolomite forms the lower slopes of the mountain, and
occupies the broad valley of the Sand River, which is again bounded southwards by a further ridge
of banded quartzites.

Throughout the whole range of the Hoekbergen east of Gatkop, the Waterberg conglomerates
show the usual association with the felsites of the Lower Division of the System. They are strongly
developed on the ridge to the east of Gatkop. After crossing the boundary of Donkerpoort, however,
their outcrop gradually narrows and just about the poort they are cut out by the granite, which
passing over the felsite comes into contact with the highly tilted beds of the conglomerates,
immediately west of the poort. The felsites probably continue to underlie, the conglomerates to the
westward, but do not appear at the surface near Gatkop.

The conglomerates are well exposed at the base of the eastern flanks of the mountain in the
kloof already referred to, which joins the main spruit on the poort. They can be followed up this
kloof to near its head. For the greater part of the distance they form the lower portion of the steep
sides on the south of the kloof, the upper precipitous portions being formed by the bare faces of the
granite. The contact of the conglomerates and granite is clearly exposed as well as the manner in
which the latter rock over-rides the edges of the conglomerate beds, passing on to successively higher
horizons in the series, and resting at the head of the kloof at an elevation of about 1,500 feet above
the poort on the pebbly grits which succeed the conglomerates. The sharp fold in the conglomerates
and sandstones, shown in the section on Fig. 7, can be clearly seen throughout the spurs forming the
northern side of the kloof, particularly near the head. The vertical portions strike directly under the
south peak of the mountain, the nearly horizontal portions form the escarpments which run out
from the northern flanks of the mountain towards the main Waterberg plateau. These nearly
horizontal beds, some of which are exceedingly massive, consist of very coarse felspathic grits with
occasional boulders, and weather into broad bare humps. They form the greater portion of the
northern slopes of Gatkop, below the banded quartzites of the crest, and continue westwards into
Buffelshoek.

The big conglomerates near the base of the Waterberg Sandstones emerge from below the
super-imposed Dolomite and banded quartzites of Gatkop near the boundary of Donkerpoort and,
crossing the south-eastern corner of Buffelshoek obliquely, pass across Waterval. Here a wonder-
fully complete section of the conglomerates is exposed in the narrow bed of the Sterk River com-
mencing a little north of the police post. At this point the conglomerates show very distinct lines
of stratification, the bedding planes being vertical or overtilted a few degrees. Passing northwards
the section can be continuously followed, when the river is low, for a distance of more than a mile.
From the base onwards there is a rapid decrease in the angle of dip, similar to that shown in the
mountain sections, and the beds soon become nearly horizontal. Looking east and west from tie
river near the police post, somewhat conspicuous kopjes composed of the less steeply dipping beds

59

of the conglomerate are seen rising from the flat. They give the impression on a general view that
the conglomerates are resting directly on the older rocks of the range upon which they have been
laid down unconformably, as shown in the section accompanying the paper by Mr. E. Joriseen,
previously referred to, and the relationships of the various rocks appear to be so clearly shown on
the ground, in such a general view, as to render further examination unnecessary. The difficulty,
however, of reconciling the conditions, which are so evident in the section at the east end of Gatkop,
with such an interpretation suggests a closer inspection of the western sections, and in every case
it is found that the low-dipping Waterberg beds are never in contact with the older rocks, but are
always separated from them by steeply-dipping or vertical beds of the conglomerates, associated
with the granite intrusions already described (Fig. 8). Moreover, in one place at least, situated
nearly in the centre of Waterval, the junction of the conglomerates and the Dolomite can be seen,
and it is clearly a faulted one, both rocks being considerably sheared along the junction line.

On the low-lying portions of Waterval crossed by the outcrop of the vertical or highly inclined
beds of Waterberg conglomerate, the latter are covered by a bed of later ite or "on klip" a few feet
in thickness, consisting of a matrix of the usual "ouklip" character enclosing abundant fragments
of banded quartzites. Where this recent surface deposit is cut through by various small spruits
draining to the Sunday River the conglomerates are exposed' below.

Along the Sunday River, northward of the section near the police post where the con-
glomerates are so admirably exposed, a good section can be obtained extending to the high kranzes
which about seven miles to the north rise on Groothoek No. 946 and Haartebeestfontein No. 948,
to the level of the main Waterberg plateau. On Groothoek there is a broad platform lying between
the base of the escarpments and the top of those of the lower rank, which lie along the northern border
of Buffelshoek and overlook that farm. Along the edge of this platform a thick series of con-
glomerates forms a very broad outcrop owing to the low dip of the beds. The pebbles in these
conglomerates are rather smaller and more angular than usual and contain a large proportion of
cherty and jaspery types. Near the foot of the same escarpments some beds of hard bluish flaggy
shales are met with, and somewhat shaly beds of bright purple-red colour are also found above the
conglomerates in the foot slopes of the kranzes above Groothoek and Haartebeestfontein (see
Fig. 6).

In the paper by E. Jorissen referred to, the view is taken that the lowest beds of the Water-
berg System in Gatkop rest unconformably at low angles on the highly inclined strata of the Transvaal
System, and that the beds have Buffered comparatively little displacement since the Waterberg
strata were deposited. This view, however, depends upon no account being taken of the
great thickness of highly inclined or vertical Waterberg beds which lie in the line of section indicated
by the author of the paper, and which can be shown to extend for more than a mile on either side
of it by numerous good exposures. Further, in the map and sections the stretch of Dolomite,
a mile in width, which on the line indicated intervenes between the Waterberg conglomerates and
the banded-quartzites, and which crops out conspicuously in the kopjes crossed by the section fine
about half a mile east of the Sunday River, nearly opposite the police post, is entirely overlooked.

The unconformity between the Waterberg and Transvaal Systems is well established from other
evidence, but has little to do with the striking contrast in the dips of the two as shown in Gatkop. The
difference in dip between the higher and lower beds in the conglomerates themselves is in fact even more
striking than that between the former and the banded quartzites of the Transvaal System, although not
so conspicuously exposed. The interpretation of the Gatkop section and the recognition of the part
played by Waterberg rocks in the building of the mountain have an important bearing on the structure
of the surrounding areas and particularly of that towards the east. Dr. Jorissen* regards the Dolomite and
banded quartzites about Gatkop as forming an anticline, probably continued eastwards for more than
forty miles to the neighbourhood of Wannbaths. It is very difficult, however, to apply this theory any-
where to the eastward of Gatkop. As already pointed out there is a stretch of Dolomite on Waterval, a
mile in width and showing high dips, lying entirely beyond the more northerly of the limbs of the
anticline, shown by the author in his section. This stretch of Dolomite, very nearly as broad as
that which is regarded as lying between the two limbs of the anticline, is entirely overlooked in the
section. An item of such significance can, however, scarcely be disregarded in framing any explana-
tion of the local structure.

The Dolomite is always a difficult formation to follow structurally, and in the neighbourhood
of Gatkop presents many variations in dip and in width of outcrop. With the exception of the

* Lm-.eit,, p. 31.

SO

neighbourhood of the caves, the dips seen are almost invariably at very high angles frequently
approaching the vertical. The conditions shown about Gatkop itself, both by the limestones and
by the banded quartzites, appear to point to a much more complicated structure than that of a single
anticline. Eastwards of the mountain die absence of an anticlinal axis, such as that indicated by the
map accompanying the paper referred to, is clear. The banded quartzites of Gatkop terminate
abruptly at the summit, the range being continued eastward by the Waterberg felsites, conglomerates,
and sandstones of the Hoekbergen. South of these is a broad belt of granite, which crops out freely
and conspicuously over many square miles on Donkerpoort, Sandspruit, and Rookpoort, in the
position assigned in the map to the Dolomite along the axis of the anticline. Although the explana-
tion of the ranges running eastwards from Gatkop as the limbs of an anticline is suggested by the
topography and by the well marked dips shown by the strata of which they are composed, the
explanation cannot be applied when it is known that while the ridges to the south of the
Hoekbergen taken as one limb of the anticline, are composed of rocks of the Transvaal System, the
Hoekbergen themselves, which are taken as the other limb, consist entirely of Waterberg rocks, while
the intervening low ground shows numerous outcrops of granite. Beyond Nooitgedacht No. 301
the granite widens rapidly and cropB out over extensive areas on Rhenosterhoekspruit No. 1008,
Vingerkraal No. 313, Earreefontein No. 2147, and the farms between these and the Hoekbergen.

5. Summary.

It may be useful to summarise briefly the main points in the structure of the Hoekbergen and
Gatkop.

The Hoekbergen correspond closely in structure and composition with the corresponding
felsite and conglomerate ranges which skirt the main mass of the Waterberg formation further east
and north. The range passes westwards into the mountain mass of Gatkop, the strata of which it
is composed forming the bulk of the eastern end of that mountain.

The beds of the lower portion of the Waterberg conglomerates are highly inclined or vertical,
and are overlain by the banded quartzites which form the highest part of Gatkop. The relative
position of the rocks of the two systems is to be explained by a well marked overthrust fault, which
can be clearly traced in the eastern sections of the mountain.

Along the plane of the overthrust a considerable body of granite has intruded, which forms
the northern slopes of the eastern end of Gatkop. At the western end of the mountain intrusions
of granite occur as narrow bands both in the Waterberg conglomerates and along the course of the
fault between these and the Dolomite.

Faulted and possibly overthrust relationships between the Transvaal and Waterberg Systems
probably prevail along the margin of the Bushveld eastwards from Gatkop, with extensive intrusions
of granite along the faulted zone.

V.

THE GEOLOGY OF THE COUNTRY
EAST OF POTGIETERSRUST.

INCLUDING THE MARABASTAD GOLDFIELDS.

By A. L. Hall (Geologist).

CONTENTS.

1. Introduction.

2. Physical Features.

(A) The area east and north of the Strydpoort Range.

(B) The mountainous belt of the Strydpoort Range.

(C) The country between the Strydpoort Range and the Olifants River.

(D) The hilly country east of Potgietersrust.

(E) Water-supply.

3. Structural Geology.

4. Descriptive Geology.

(A) General Remarks.

(B) The Older Granite with the associated basic schists and Utter intrusive rocks.

(C) The Swaziland System.

(D) The Black Reef Series.

(E) The Dolomite.

{F) The Pretoria Series.

(G) Metamorphic Rocks of the Pretoria Series and the Dolomite.

(H) The Waterberg System.

(/) The Karroo System — Buskveld Sandstone Series.

(J) Rocks intrusive in the Transvaal System.

(K) Rocks belonging to the main body of the Buskveld Plutonic Complex.

(L) The Buskveld Amygdaloid.

{M) Superficial Deposits.

5. Economic Geology.

-THE GEOLOGY OF THE COUNTRY EAST OF POTGIETERSRUST,
INCLUDING THE MARABASTAD GOLDFIELDS.

By A. L. Hall {Geologist).

1. Introduction.

The following report deals with the westerly continuation ot field-work completed during the
winter months of 1907.

During that year mapping was carried on from the junction of the Steelport and Olifants
Rivers in a north-westerly direction as far as the Malips River ; in the present year the work was
extended towards the west and north-west as far as the Pretoria — Pietersburg railway ; on the
northern side mapping was carried up to a straight line drawn from near Pietersburg to Haenerteburg,
while on the south the work was limited by a line drawn roughly from near Moorddrift Station to
Adrianskop, a few miles north of the junction of the Zebedelas with the Olifanta River. The area
examined formB a Btrip of country roughly thirty-five by forty-five miles in extent and comprises
1,690 square miles, over which 581 miles of boundary lines were traced.

Whereas the formations met with while mapping the eastern portions during the previous
year all belonged to the Transvaal System and its associated igneouB rocks, the area at present dealt
with contains representatives of all the geological horizons present in the Colony, with the exception
of the Witwatersrand and Ventersdorp Systems. The examination of the extensive region of granitic
rocks between Chuniespoort and Pietersburg included the well known and important belt of schists
running from the Yzerberg north-eastwards through Mount Mare, which contains several gold-bearing
deposits.

The Transvaal System, with its persistent horizons of quartzite, is well developed and again
remarkable for the great variety of metsmorphic rocks associated with it ; while of special interest
is the occurrence in the Pretoria Series of a considerable thickness of calcareous beds intensely
metamorphosed by the intrusion of the main body of the Bushveld Plutonic Complex. Attention
was drawn in previous reports to the remarkable persistence of certain horizons in the Pretoria Series,
and the same holds good in the present case. Thus, for example, the Daspoort Quartzite has been
traced, without any permanent break, from Waterval Boven Station to the railway line at Pot-
gietersrust, a distance of about 190 miles, except that near Zebedelas River it is overlain for some
thirty miles by rocks of the Karroo SyBtem.

Very interesting observations were made in connection with the Transvaal System round
Potgietersrust and its relationship to the main body of the BuBhveld Plutonic Complex. While
in the eastern and central Transvaal the margin of the norite always coincides more or less accurately
with the strike of the Magaliesberg Quartzite, it was found that along the Pietersburg line north
of Potgietersrust, the basic margin of the Bushveld Complex comes to lie against every member of
the Transvaal System in turn, until north of the Uitloop Lime Works the norite has an intrusive
relationship with the older granite ; probably therefore the entire thickness of the System has been
broken across during the intrusion of the igneous complex.

2. Physical Features.

Considering the great variety of formations, which go to make up the geology of the area,
it is not surprising to find a somewhat complex topography, for which reason it is more convenient
for descriptive purposes to divide the area into portions, more or less naturally characterised by a
prevalent set of land forms.

(A) The area east and north of the Strydpoort Range.

Enclosed on the south and west by the conspicuous wall-like escarpment of the Black Reef
Quartzite lies an extensive tract of gently rolling country composed almost entirely of Older Granite ;

64

its average elevation is somewhere over 4,000 feet, but from north to south there is a distinct fall,
well shown along the main road leading from Pietersburg to Chuniespoort. The more elevated parte
round the former locality consist of nearly flat to slightly rolling country recalling high veld scenery,
almost devoid of vegetation and presenting a monotonous topography, only broken by isolated low
kopjes composed of large rounded granitic blocks. Similar scenery continues along the Chuniespoort
road for about twenty miles, till some five miles from that poort the aspect of the country rapidly
changes. There is a considerable drop in the general level and the surface relief becomes more varied
owing to the presence of many small granite kopjes, some of which have been carved into striking
forms (Bee Annual Report for 1904, Plates II and III). At the same time, vegetation begins
to increase until close to the foot of Chunies Mountains the country resembles some aspects of the
Bushveld. This marked change is no doubt due to the great difference in the rate of erosion of the
harder Black Reef Quartzite compared with the more readily disintegrating granite.

A similar and almost as striking a change is well shown on the main road from Pietersburg
to Haenerteburg, which for the first twelve miles, to some way beyond Harais' Farm, lies in undulating,
bare, featureless country resembling high veld ; from this point eastwards there is a decided and
rapid fall towards Smitsdrift in the valley of the Malips River, which here runs through well wooded
bush country. From Smitsdrift there is a long and considerable rise to Haenerteburg, which, though
below the " Berg " is situated in comparatively high bare country composed largely of basic schists.

Rising abruptly from the tableland a few miles south and south-east of Pietersburg, there is
a well defined range of hills, generally referred to as Mount Hare, but locally this term is often
restricted to the hills lying near the common boundary of the farms Snymans Drift and Zandrivier.
Viewed from the railway line, between Zandrivier Siding and Marabastad, Mount Mare forms a straight
sharp ridge some three and a half miles in length and presenting a steep slope on its northern or
escarpment Bide, only broken through by one small poort near the eastern end of the range. Running
parallel to the most northerly and main ridge there are a few shorter and less pronounced ridges on
the south-east side. Both towards the east and west the range rapidly dies out, so that this hilly
belt is well defined topographically. Geologically it belongs to the well known belt of schists which
runB from the neighbourhood of Uitloop through the Yzerberg and Mount Mare in a north-easterly
direction. A frequent feature in tins belt iB the occurrence of conspicuous sharply created white
kopjes, due to the presence of bands of hard white quartz rock (" buck reef "). The striking kopje
occurring on Weltevreden, five miles south-east of Pietersburg, which forms the well known landmark
near that town, is a good example. Between the western extremity of the Mount Mare Range and
Uitloop on the PieterBburg railway these schists form another prominent ridge, including the
Yzerberg ; this topography will be considered later.

(B) The mountainous belt of the Strydpoort Range.

The term Strydpoort Mountains is applied to the well defined chain stretching from the Iron
Crown south of Haenerteburg in a south-westerly direction through Chuniespoort to Zebedelaa
River. A few miles further, the range bends rather sharply to the north and continues in this direc-
tion as far as Button Kop, where, both geologically and topographically, it comes suddenly to an
end, owing to a very powerful line of dislocation, as explained more fully below. In reality the chain
is merely a continuation of the Drakensberg, which begins near Tafelkop south of the Devil's Kantoor,
running thence roughly northwards as far as Haenerteburg through a distance of some 130 miles.
Along its entire length, until its final termination a little north of Button Kop, the range is made up
of quartzites of the Black Reef Series. The great thickness of this formation, together with its low
southerly and westerly dips, accounts for the persistent and pronounced escarpment which defines
the southern limit of the granite country on the north. Unlike the northern portions of the Drakens-
berg, the Strydpoort Mountains are not cut up into a number of more or less isolated peaks, but
the crest line remains at a more uniform elevation for greater distances, well seen for instance along
the portion intercepted between Malips and M'PhatleleB Rivers. Here and there certain short por-
tions of the range stand out particularly boldly, like Chunies Mountains and Button Kop. The
extremely powerful contrast, topographically, between the crest line of the Drakensberg and the
Low Country on the east has no analogy along the northern side of the Strydpoort Range ; every-
where along the latter the fall is much less, and consequently differences of level amounting to between
2,000 and 3,000 feet, Buch as frequently occur between Belvedere and the Wolkberg, are not met with
further west. It is clear that the rivers draining the eastern slopes possess the power of corrasioh
in a much greater degree than those draining the area north of the Strydpoort Range, a result probably
connected with the amount of rainfall and other climatic conditions. It is moreover significant

Annual Report, Geological Survey, 1B0S,

Fit;. l.-Wondarkop, near Malipa Drift: faulted and highly inclined Timeball Hill Quartette.

Fig. 2. -Viflw of the Yzerberg (Banded Ironstones of the Swaziland System), north-east of Potgetenrut.

65

in this connection that along the north and South portion of the Strydpoort Range the contrast is
much stronger than along the remainder of the chain.

The elevation falls steadily from the Iron Crown (6994 feet* above sea-level) towards Chunies-
poort, from which point there is a further and more rapid fall until, near Pienaara Nek, the crest line
is only a few hundred feet above the general level of the granite area on the north. North-west of
Zebedelas River the range becomes again more pronounced while the crest line rises considerably ;
thus in the north-eastern corner of Portugal No. 1,066 it attains a height of 6,573 feet*.

While on the escarpment side the Strydpoort Range is sharply defined against the underlying
Older Granite, it merges on the southern and western, or dip-slope, sides into the hilly tract made
up by the overlying Dolomite. This formation contains no quartzites or distinct beds which would
produce a definite set of topographical features ; hence the dip-slope side) of the range are made
up of very rough hilly ground, deeply dissected by numerous spruits and kloofs, making up a tract
of country very tedious to traverse and quite impassable to transport on wheels.

Sometimes the intense disturbances to which the Dolomite east of Chuniespoort has been
subjected produce striking topographical results. Thus a few miles west of Malips Drift occurs a
conspicuous conical koppie composed of Timeball Hill Quartzite, aptly known locally as " Wonderkop "
(see Plate VI, Fig. 1). The high dip of this very massive quartzite, its contorted nature, and isolated
position, due to a powerful dip fault (see Plate XVII), produce a hill type which is unique in this
area.

At several places along the strike, the Strydpoort Range is cut through by streams which
empty themselves direct into the Ohphants River. These are the Malips, M'Phatleles, Chunies
and Zebedelas Rivers, the points where they enter the range being as usual marked by more or less
well defined poorte, that of the last-named stream being known as Pienaars Nek.

Strydpoort, from which the whole range takes its name, is due to a small tributary of the
Zebedelas River. At -such points the main chain is passable to any kind of transport, with the
exception of Malips Poort (Beyers Nek) and M'Pliateles Poort, both of which form deep ravines,
impassable except on foot.

{C) The country between the Strydpoort Mountains and the Olifants River.

That part of the area, which is made up of rocks belonging to the Karroo System, the Bush-
veld Plutonic Complex, and the Pretoria Series, from Malips Drift westwards to Zebedelas River
presents a complex topography not obviously depending on any general controlling factors.

The leading quartzites of the Pretoria Series, though still in evidence west of Malips River,
become much twisted and broken and, since the dip is also rather high, " banken " scenery is less
marked ; the whole region occupied by the Pretoria Series really belongs to the Bushveld type of
country. This becomes more and more evident as the beds are traced westwards.

The whole of Zebedelas Location and the surrounding country belong to the Bushveld, and
when seen from some distance present an enormous tract of apparently level and monotonous densely
wooded country, which forms the extreme north-eastern portion of the extensive Springbok Flats.
From Zebedelas River westwards the same type of country extends as far as the neighbourhood of
Moorddrift Station, and its distinctive features are directly dependent on the distribution of the
Karroo rocks and the norite of the Bushveld Plutonic Complex.

A few miles east of Chunies River, where the Bushveld Amygdaloid dies out, the dense
vegetation gives place to more open country, while at the same time the monotony of the Flats is
relieved by scattered, isolated kopjes and short ridges. Such points form the only outcrops, all the
intervening country consisting of deep red sand. M'Phatlele's Location is highly characteristic of
this type of scenery f. The short ridges in the location are readily picked out in the landscape by
the lighter colour of the rock and the tendency to present Bharp crest lines. They are due to the
Daspoort and Magaliesberg Quartzites. Topographically the ridges are quite detached, and careful
examination alone reveals their connection across the wide gaps, such as may be found for instance
in the uppermost quartzite, which is practically featureless for several miles east of Chunies River,
It is interesting to note how Buch rocks, so frequently standing out in bold land forms, can be reduced
to an almost level surface under particular climatic conditions.

Scattered isolated kopjes are perhaps the most striking feature in the topography of
M'Phatlele's Location, especially when they occur within the area occupied by the sedimentary series.
In the northern portion and some three or four miles south or south-east of the Naauwpoort store

are found several small scattered kopjes, the position of which in the general succession is not at
once apparent. Their markedly black colour and dense vegetation, mostly made up of Euphorbia
candelabra, render them striking objects, which at once arrest die eye. Their horizon is between
the Daspoort and the Magaliesberg Quartzites and they are made up of cordierite-gneisses ; the
black colour, however, and mode of weathering into large blocks render them liable to be easily mis-
taken for norite kopjes, when seen from a little distance.

Towards the south and east, the Bushveld type of vegetation dwindles rapidly, while the
surface relief is once more that of rolling sparsely wooded country, identical in character with that
of the norite and granite zone north of Pretoria. The belt of typical norite begins along the eastern
boundary of M'Fhatlele's Location and extends thence to the Lulu Mountains of Secoecoeniland :
all over thiB area are found the usual varieties of more or less pyramidal hills, with an occasional
tendency to run along definite lines, through one of which the Olifants River has cut a sharply
defined channel at Olifantspoort. Along the southern border of M'Phatlele's Location occur two
or three lines of well marked hills usually trending nearly due east and west, which differ topo-
graphically from all the other tracts of higher ground. They form light coloured kopjes, much less
rugged and dark in appearance than those due to the norite, with some resemblance to sedimentary
rocks, as suggested by a kind of escarpment- like feature facing northwards. One set of these hills
forms an unbroken line running for nine miles from the common beacon of Turflaagte, Portje, and
Zwartbult eastwards and terminated by a conspicuous and high hill, on which stands the common
beacon of Vleip'aats, Pramkopjes, and Kopjesdam. This line runs parallel with the range of low
norite kopjes extending eastwards through the southern boundaries of Dwaalkop and Doornvlei.
Another very well defined range runs due north and south for a distance of four miles, along the
common boundary of Spelonken and Zaaikloof close to the left bank of the Olifants River, but at
its southern extremity it bends round to the west and south-west towards Adriaanskop. Enclosed
within these ranges occurs an interesting irregular group of low, very dark coloured kopjes, made up
of almost black crystalline hornfels, weathering into large blocks, so that from a short distance they
strongly resemble typical norite kopjes. It will be shown below that the hills surrounding this
hornfels belong to a highly acid phase of the Red Granite.

(D) The hilly country east and south-east of Potgietersrust.

In this neighbourhood the topography east and west of the railway line is Bharply contrasted,
for in the latter direction there iB a wide stretch of lower lying flat buBh country composed of rocks
belonging to the norite zone, while on the east aide the surface is much broken up by hills and hill
ranges.

The most striking feature is the presence of at least two separate sharply defined parallel
ranges, separated by a deep narrow valley and forming two strongly marked curvilinear and roughly
semi-circular ridges, which come to an abrupt end along the railway line a short distance north of
Potgietersrust. While towards the north these ridges tend to approach one another, in the opposite
direction they become wider apart and eventually run roughly north and south. They correspond
to the Timeball Hill and Daspoort Quartzites of the Pretoria Series, and have very regular and
extensive dip slopes, well seen from the railway line north of Potgietererust. Following them towards
the east lie two or three short ridges and isolated kopjes, which seem at first Bight to have no connection
with one another ; these belong to the uppermost or Magaliesberg Quartzite. Along the south-
eastern boundary of Oorlogsfontein two such separate masses of quartzite are found, the intervening
portions having been worn down to a level outcrop in much the same way as the quartzite behaves
in M'Phatlele's Location (see above).

On account of the folded nature of the middle quartzite and the presence of at least two minor
faults, which produce a lateral displacement along the direction of dip, such features are not at once
recognisable. - The crest line of the most easterly range becomes much higher towards the south,
and at the corner beacon in the western boundary of Doornfontein it attains an elevation of over
5,600 feet. From this point a superb view iB obtained in all directions, and the extraordinary contrast
between the scenery over the area occupied by the Transvaal System towards the north and the
" illimitable veld " of the Springbok Flats towards the south is one of the many pleasures that reward
the geologist after a long and tedious ascent.

East of Potgietersrust and north of the main road from that town to Pietersburg the leading
topographical feature is a prominent and bold range composed mainly of quartz-schists of the Swazi-
land System. Near the common boundary of the farms Uitkyk and Weenen the range becomes
composite, for the Black Reef Quartzite forms its extreme south-western portion without any marked

67

topographical break; where this occurs the range becomes much lower and turns rather sharply
towards Uitloop to the north-west.

Between the range just referred to and the conspicuous curved line of hills due to the lower-
most Pretoria Quartzites lies the somewhat broad and open fertile valley of Weenen, Makapansgat,
and De Hoop, bounded on the east and south-east by a continuous belt of broken hilly Dolomite
country. Towards the north-east the broad schist range ends at the Yzerberg, a very characteristic
feature in the landscape owing to its peculiar form, which consists of two sharply crested peaks,
resembling, especially when viewed from a little distance, the back of a double-humped camel (Plate
VI., Fig. 2).

(E) The Water-supply.

The supply of available surface water is very unequal. Over the area north of the Strydpoort
Range, occupied by the Older Granite, there is a poor supply, in spite of the not inconsiderable catch-
ment basin which feeds the rivers cutting through those mountains. Since the extensive granite
area is more or less deeply weathered and covered with much loose and easily permeable sand, a
larger proportion of the rainfall must enter the ground to feed the underground water-table. This
is one of the reasons for the characteristic appearance of such rivers as Zebedelas, Chunies, and
MThatleles, with their broad and flat sandy channels, which during the winter months contain only
a meagre trickle of running water, sometimes scarcely more than an inch in depth ; even thiB little
supply may dwindle, until the water stands in a few isolated shallow pools. Between these, in the
dry reaches of the bed, water may often be quickly obtained by digging to a depth of a few feet only,
proving the existence of a more perennial underflow. Where a harder, less permeable formation
crosses the rivet bed, this underflow is immediately brought to the surface, giving rise to a good supply
of water. All three rivers mentioned above illustrate these features extremely well ; thus the
M'Phatleles River, which is dry for some distance north of Naauwpoort, suddenly shows running
water at Krutwig's -store ; here the flow is probably forced up by the harder Timeball Hill Quartzite,
which crosses the river half a mile south of the store. Further down the stream in the Location the
same thing happens again and a good and perennial supply b obtained sufficient to irrigate a large
area of agricultural land in Molsgat and Zondernam east of the Location. Similarly south of Chunies-
poort the SlypBteen Drift sandstone brings up the underflow of the Zebedelas River after a dry Btretch
of several miles,

The belt of lower country running east and west, constituting the Springbok Flats, occupied
by the Bushveld Amygdaloid with the norite zone to the east, has a very poor water-supply, surface
water being found at only a few widely separated spots. One of these is the interesting locality
of Matlapoodi, two and a half miles north-east of Slypstee Drift, where a water-hole or shallow
depression roughly oval in shape is found, surrounded by low banks of whitish surface limestone.
Although this area is very dry, there must be considerable bodies of water underground for the
following reason. From the high ground of Dolomite, which lies west of the north and south portion
of the Strydpoort Range, several streams drain southwards towards the Flats. Some of these
eventually join ZebedelaB River, others die out soon after entering the level country, but all of them
carry a good and perennial supply, and the one running through Uitkyk, past the homestead lately
occupied by Mr. Shepstone, is a powerful stream carrying a large body of flowing water, yet two or
three miles further downstream the bed of these rivers is permanently dry. Besides these there are
several good springs issuing from the Dolomite a little north of its overlap by the Bushveld Amygda-
loid, as seen for instance on the farm Modderfontein a few yards from the house of Mr. Papenfus.

A better supply of water is met with over the area west of the Strydpoort Range where the
latter runs north and south ; thus the Dorp River, which supplies Potgietersrust, rises in several
tributaries over the high Dolomite ground of Portugal, Spanje, and Makapansgat and perennial
water runs in the main stream to about the centre of the last-named farm, but by the time tha
adjoining farm Weenen is reached the amount of water diminishes and finally disappears altogether
for a distance of several miles. No doubt this is due to swallow holes in the Dolomite of Makapansgat
and is only another instance of the numerous cases of the disappearance of water, known where-
- ever that formation has been examined in the Transvaal. Another stream rising on the common
boundary of Zwartkrans and Driefontein, also one coming from the western corner of Buffelshoek
show a similar temporary disappearance of their water. Near the centre of the farm Weenen the
water reappears in the bed of the Dorp River. Probably the underflow is here brought up by some
hard bar, such as a chert band, in the Dolomite ; but it again all drains underground after a distance
of a few hundred yards, when the bed remains dry until the spring on Planknek is reached. Here

the water reappears in a strong body and remains permanently on the surface. At this locality the
shales underlying the Timeball Hill Quartzite are well exposed and the spring rises at the base of
the former, in close proximity to the Dolomite. There can be little doubt that it fa this contact
between pervious calcareous rocks and impervious overlying shales which forces the underflow up
again. One of the reasons of this alternate appearance and reappearance is therefore to be found in
the degree of permeability of the formations occurring in and near the bed of the Dorp River. But
the hydrographic problems are further complicated by the great fault which runs from the Yserberg
south-westwards more or less along the Fotgietersrust — Pietersburg main road (see map Plate
XVII). North of the line of this break a strong stream, carrying a good and permanent supply, rises
from the quartzitic schists near the centre of the south-eastern boundary of Uitkyk and continues
southwards, until it all disappears on or close to the (approximate) position of the fault line. It is
very probable that the water drains underground along the plane of the fault and reappears in the
Planknek springs. The same fa probably the case with the water in the Driefontein Spruit, which
disappears near the point where it crosses the same fault line on the southern boundary of Zwartkrans.

The above interpretation of the Dorp River hydrography, first put forward by Dr. Humphrey
two years ago, has been demonstrated to be correct by means of the fluorescein test, which was
successfully applied by the Irrigation Department and proved the underground connection through
several miles of the main supply in the Dorp River.

It will be noticed that the general hydrographic conditions are very similar to those obtaining
in the head waters of the Aapies River at the Fountains, near Pretoria, where the upper spring
is in all probability brought up by a fault combined with a sharp juxtaposition of impervious slate
and pervious Dolomite.

Underground supplies are probably obtainable at no very great depth in the areas occupied
by the Bushveld Amygdaloid and the Older Granite, but in the regions covered by the Pretoria Series
it is more difficult to gauge the availability of such water beforehand, especially when the shales are
much altered. Experience shows that the more highly metamorphosed rocks of this type become
very hard, compact, and " tight " in a hydrographic sense. It is sometimes found that a borehole
sunk in such hard homfels may remain dry for some depth, although it fa well below the water level
in adjoining streams, which are not far removed from it. In Mr. Hamilton's borehole, for instance,
which has been sunk all the way in hard crystalline homfels between the Daspoort and Magaliesberg
Quartzites, no water was struck until a depth of about 120 feet.

In many cases the surface water, and sometimes also the underground supply, is rich in dis-
solved saline constituents, which give it a bitter taste, whereby an otherwise excellent supply is
rendered unfit for domestic purposes, as for instance in a shallow well behind Platteau's store, not
far from Zebedela's Kraal, where an abundant supply of perfectly bitter water is available twenty
feet from the surface. Frequently the edges of the river beds are lined with a soft white layer of
salts, left by the evaporation of the brakish streams. Good examples occur in Chunies River, where
it runs through the south-eastern portion of M'Phatlele's Location, and in the bed of M'Phatlele's
River a few miles north of the Strydpoort Range.

3. Structural Geology.

The geological structure of the present area, though locally rather complicated over small
portions, fa fairly simple as a whole. The distribution of the Transvaal System and the unconform-
able overlap by the Bushveld Amygdaloid furnishes the key to most of the problems, while the
structural features associated with the belt of schists belonging to the Swaziland System south of
Mount Mare furnish a special problem in themselves.

The Transvaal System, as in other parts of the central Transvaal, dips everywhere towards
the basic margin of the Bushveld Complex, that fa, roughly speaking, to the south and west.

East of Zebedelas River, the strike fa fairly regular from the Iron Crown as far as Strydpoort
in a south-westerly 'direction and the dip fa to the south-east, varying in amount from 3 degrees
to over ("'0. It fa a common experience to And that the dip decreases the further one descends in
the succession. For the Black Reef Series it rarely exceeds 6 degrees, while in the Dolomite its
average fa nearly 25 degrees. In the Pretoria Series values as high as 40 degrees are common ; but
locally the beds are inclined at a much higher angle, as in the Daspoort and Magaliesberg Quartzites
of M'Phatlele s Location, where a dip of 60 degrees was observed.

Faults are not numerous ; east of Zebedelas River the most pronounced break occurs in con'
sequence of a powerful dislocation, which runs across the central portion of M'Phatlele's Location
roughly coinciding with the direction of the dip (see map, Plate XVII). It extends for a distance of

69

at least eleven miles and has a very marked effect on the topography and on the distribution of the
boundary lines affected by it. Thus the outcrops of the quartzites of the Pretoria Series are dis-
placed through some four miles, and an approximately similar displacement affects all the associated
formations. One result of this fault is the very sharp termination and distinct bending round of
the harder beds, which come to an abrupt termination both geologically and topographically.
Another result is seen in the composite nature of the well defined quartzite range, which trends east
and west about half a mile south of Krutwig's store on Naauwpoort. While topographically this
range continues right across the fault from Naauwpoort to the Obphants River, the Timeball Hill
Quartzite terminates at the fault on the crest line near the western boundary of Kaffirnek; but the
high range persists, though now composed of faulted norite. The change in the formation is clearly
recognisable in the form, colour, and mode of weathering along the range near the break.

There is abundant evidence to show that the Transvaal System has been subjected to great ■
pressure, though it was probably less intense than that which operated over the western and southern
portions of the Haenertaburg Goldfielda to the east of our present area. Yet if the Black Reef escarp-
ment is examined both east and west of Zebedelas River, there are no signs of any marked
disturbances or folding. The short and unimportant faults at Chuniespoort and the slight anticlinal
structures near Button Kop are in no proportion to the powerful compression, of which there is
striking evidence in the higher horizons of the Transvaal System. No doubt the great thickness of
the lowermost formation enabled it to resist deformation on a large scale to a great extent. The
case is very different, however, when the Dolomite and the Pretoria Series are considered. The
complex manner in which the quartzites of the latter have been twisted and broken up is well seen
in M'Phatlele's Location and south-west of Mahps Drift, but on the map accompanying this report
the resulting features could only be indicated in a general way. Everywhere the rocks are broken
and well jointed on a small scale, while often a single bed can be Been to be repeatedly contorted,
as for instance in the Timeball Hill Quartzite west of Malips Drift, near the cone shaped Wonderkop,
where it runs out against the big fault already described. Still, the continuity of any one bed is much
more complete than is the case with the Dolomite. It would seem that the Pretoria Quartzites
behaved under pressure in a way generally similar to that of the Black Reef Series, only being very
much thinner they were able to yield to some extent by folding. The shales intervening between
the quartzites are very rarely exposed, and in M'Phatlele's Location are metamorphosed into holo-
crystalline dark coloured gneisses. The evidence of great pressure is here better seen under the
microscope, but a foliated gneissic structure is occasionally well preserved in the hand specimens,
as for example throughout the peculiar low black kopjes scattered through the northern portion of
the Location above the Daspoort Quartzite (e.g. No. 274 c,).*

East of the main road leading from Chuniespoort to Naauwpoort, the Dolomite has suffered
the greatest amount of compression and the resulting features can be studied near this road, or better
still along the M'Phatleles River from Naauwpoort to where that stream enters the Black Reef
Quartzite. The beds are thrown into a succession of anticlines and synclines, both symmetrical and
asymmetrical, some of which are on such a scale as to be readily visible on the hill slopes, even at
some distance. The folding becomes less intense the further one descends in the Dolomite ; at the
boundary between the middle and lower series, the succession is practically undisturbed, while at and
a little below the base of the Pretoria Series the folding is most marked. It is fortunate that the latter
horizon is marked by a very coarse conglomerate, extending from the Chuniespoort road eastward to the
great M'Phatlele's fault, because this rock is well adapted to preserve the effects of pressure. Over and
over again this band is cut up into sharply denned blocks by means of a set of extremely sharp joint
planes, so that the large rounded pebbles, sometimes eight inches in length, are cut clean through in
various directions, producing a kind of facetted appearance. At the same time the bedding is either
vertical or very nearly so, and a few miles east of Krutwig's Btore the beds are even overturned some
10 degrees. In this neighbourhood one sometimes finds remnants of synclines forming the highest
portions of more or less isolated and prominent kopjes or short ranges. Fig. 9 gives a general idea
of the kind of folding met with.

Turning now to the distribution of the Transvaal System near Strydpoort and towards the
north-west, the marked bend in the strike at once claims attention. This bending begins a short
distance east of Zebedelas River and takes place rapidly but steadily and without any obvious break,
apart from slight displacements at Pienaars Nek and perhaps also in Strydpoort. The final result

»[iecimens collects! during the surrey of tbifl area ftutl ilc|MMitetl in the Museum of the

ia a north and south strike, most regular in the lowest portions of the Dolomite and the Black Reef
Series ; though even these horizons have a slight tendency to turn again westwards as they approach
the Yzerberg area. The upper portion of the Dolomite and the whole of the Pretoria Series, however,
maintain a northerly strike for only a short distance and assume an extremely curvilinear distribu-
tion immediately southeast of Potgietersrust. Here the pressure must have been very great, for
individual harder beds like the Daspoort Quartzites, are thrown into powerful synclines and anticlines,
e.g. on the farm Vierentwentig Rivieren (see map Plate XVII).

Along the Pietersburg main road the whole of the Black Reef Quartzites and the greater
portion of the Dolomite come to an abrupt end against a fault running from the southern part of the
Yzerberg in a south-westerly direction. How far it extends towards Potgietersrust it is impossible
to say, but probably it does not go far into the shales underlying the Timeball Hill Quartzite. The
Black Reef Quartzites have suffered a displacement of outcrops of some eleven miles westward in
the direction of dip, while the overlying beds towards the west appear gradually to be less affected.
Where it crosses the western boundary of Makapansgat the fault can be defined fairly closely by the
great contrast between quartzite-schista of the Swaziland System, immediately north of the main
road, and the few isolated low Dolomite kopjes close to and a little south of it. Elsewhere the line
is impossible to locate accurately owing to drift. The Button Kop quartzite is picked up again on
the farm Weenen, whence it runs with a marked and rapid thinning down as far as the neighbourhood
of the Northern Lime Works at Uitloop on the railway line north of Potgietersrust. The shales of

DIAGRAMMATIC SECTION ACROSS THE TRANSVAAL SYSTEM EAST OF CHUNIES POORT.

1. Mopi/its&vj Quart}!/*. 3. Shalts. S. Datpoat Qvarti'tt. 4. Tlmtball Hill Qiiatftl/t. t. Conotamtro.lt. a Oohoiih.
7. Black R„f Sttlti. 8. North, ». O/tttr Granltt.

Fig. 0.

the Pretoria Series underlying the Timeball Hill Quartzite are well exposed at the Planknek springs,
but do not cross the railway. With the exception of isolated thin quartzites, possibly belonging to
the lowermost band, the Pretoria Series is not represented at all west of the railway. A thin bed of
banded ferruginous quartzite, exposed in the solitary railway cutting three miles north of Potgieters-
rust, occupies a doubtful position at the top of the Dolomite.

The relationship between the Pretoria Series and the overlying norite north and south of
Potgietersrust is very probably an intrusive and not a faulted one.

In the central and eastern Transvaal one always finds only the uppermost or Magaliesberg
Quartzite, or a thin development of shaly beds above that horizon, in direct contact with the norite,
so that the strike of the edge of the basic margin agrees closely and over wide distances with the
general strike of the upper beds of the Pretoria Series. Near Potgietersrust, however, the norite
cuta across the whole of the latter formation until it occupies an intrusive relationship with the
Dolomite. Immediately north-west of the Northern Lime Works, pyroxenites obviously belonging
to the noritic magma are found in contact with and intrusive in the Older Granite on the north and
the Dolomite on the Bouth. Although the Black Reef Quartzite is very much thinner in this neigh-
bourhood and appears to die out, it must have extended stilt further west, for it reappears a few miles
in that direction, though no sign is left of the intervening portion (see map, Plate XVII).

Possibly part of the formation was depressed into the magma or broken and uplifted, having
subsequently disappeared by denudation. In any case it is evident that the mechanism of the
intrusion of the Bushveld Plutonic Complex was far less simple here than elsewhere.

71

It is probable that the powerful fault alluded to continues in a north -easterly direction towards
Mount Mare and cuts off the succession of schists, phyllitea, and ironstones, belonging to the belt of
schists, which reaches from near Uitloop through Mount Mare and Paimietfontein to the Haenertsburg
main road east of Pietersburg. The banded ironstone rocks, easily recognisable and highly
characteristic of the Swaziland System, are admirably adapted for bringing out the structure and
following the beds along the strike. Three main and separate horizons are clearly recognisable in
the succession at Mount Mare, and between their position near the Yzerberg and on its south-western
side, compared with their distribution in the mountainous tract south of Marabastad, there is a
general agreement. Traced westwards from Mount Mare into Snymans Drift and Hollands Drift,
the band of " calico rock " seems to end suddenly just before reaching the main road, while between
the last-named farm and the Yzerberg they are not in evidence. Furthermore the amount of lateral
displacement shown by the Black Reef Quartzite agrees approximately with that of the ironstone
band on the Yzerberg, so that the fault probably extends through some eighteen miles.

It will be at once seen by referring to the map on Plate XVII that south of the Strydpoort
Range, the greater portion of the Transvaal System is unconformably overlain by rocks belonging
to the Karroo System ; but the limits of the Bushveld Amygdaloid are extremely ill-defined, owing
to the sandy character of the country. Still the mapping shows distinctly how, east of Zebedelas
River, each horizon of the Pretoria Series and the Dolomite runs up to and disappears below the
Bushveld Amygdaloid, reappearing again some twelve miles west of that river. The Black Reef
Series, alone, continues without interruption, though its upper portion becomes partly hidden also.
The escarpment of the Black Reef Quartzite gets gradually lower as it approaches Strydpoort from
the east and west ; where it is lowest, there also the series shows the greatest amount of overlap by
the Bushveld Amygdaloid.

Regarding the general structure of the Transvaal System and the disturbances to which it has
been subjected, few definite conclusions can be arrived at as to their age and origin. In the Annual
Report for 1907, while describing the adjoining country east of Malips River, reasons were given for
assigning to the intrusion of the Bushveld Plutonic Complex the origin of the very intense pressure
to which the Transvaal System in the Haenertsburg Goldfields has been subjected, and it was shown
that the earth movements were intimately associated with that intrusion, and could not have taken
place very long before or after the consolidation of the igneous complex. The present area merely
forms the westerly extension of that formerly described, and here the rocks have likewise been sub-
jected to more or less intense pressure, with results which differ only in degree and not in kind. The
distribution of the geological formations associated with the Yzerberg fault indicates the operation
of powerful pressure from the south-west, to which they yielded in different ways ; the thick, massive
Black Reef Quartzite was broken across, while the thinner Pretoria Quartzites adapted themselves
to the new conditions more readily by folding. The tectonic features in several respects recall those
of the folded and faulted area round Pretoria ; thus there also the main dislocations run obliquely
to the strike in the same way as the Yzerberg fault and those near the Wolkberg in the Haenertsburg
goldfields, the former, closely parallel with the schists of the Swaziland System, suggesting a giving
way of the floor underlying the Transvaal System along a direction of weakness.

The simplest and most rational conclusion therefore is that the movements are due to the
intrusion of the Bushveld Plutonic Series, and occurred either during, or shortly after, the intrusion
and before the final consolidation of the magma.

4. Descriptive Geology.
(A) General Remarks.

Within the present area a great variety of formations are represented, some covering wide
stretches, others much more restricted. The following is a list : —

Sedimentary.
Superficial Deposits
Bushveld Sandstone
Waterberg Sandstone and Quartzite
Pretoria Series

Dolomitic Limestone and Chert, etc

Black Reef Series of Quartzites and Sandstones
Quartzites, Schists, Phyllitea, and Banded Ironstones

I Karroo System.
Waterberg System.

> Transvaal System.

[ Swaziland System
\ (Moodies Series)

72

Igneous.
Bushveld Amygdaloid.
Rocks intrusive in the Transvaal System.
The Bushveld Plutonic Complex.
Older Granite and associated basic schists.
Rocks intrusive in the Older Granite.
Of these various formations the Older Granite and other igneous rocks associated with it occupy
rather more than about one-third of the entire area lying immediately north of the Transvaal System,
which runs across the country from east to west. The Karroo SyBtem, including the Bushveld
Amygdaloid, forms a well defined, but less extensive piece of country on both sides of the Zebedelas
River south of the Strydpoort Range ; to the east of it the Red Granite and norite are strongly
developed, both west and east of the Oliphants River, as far as the Lulu Mountains in western
Secoecoeniland.

Rocks belonging to the Waterberg System are confined to one small outlier a few miles
south-east of Potgietererust.

Economically the schist belt south of Marabastad on the Pietersburg railway is the most
important formation, as in it occur a number of gold-bearing quartz reefs. To the south of it lie the
alluvial workings of Eersteling and Vrischgewaagd, as well as reef gold horizons, e.g. on Waterval and
Eersteling.

The norite contains a belt of magnetic iron-ore, while chromite occurs near the extreme limit
of the area (see Annual Report for 1907). North of Potgietersrust the Dolomite has been metamor-
phosed by the intrusion of the norite and, probably as a result of this alteration, some of the beds are
suitable for the manufacture of white lime.

(B) The Older Granite with its associated basic schists and later basic intrusions.

This occupies the country lying north of the Strydpoort Range and belongs to the great mass
of this formation, which is known to cover very extensive portions of the northern Transvaal. Its
general characters do not differ essentially from those similar rocks mapped during 1905, between
Nelspruit and Sabie and described in the Annual Report for that year, to which account little need
be added.

Often the granite retains perfectly similar characters for many miles, seen not only in hand
specimens but also in the mode of weathering and the prevailing type of scenery, which forms rolling
country, sometimes fairly well wooded and covered with clean pale yellowish-pink sandy soil. Highly
characteristic are the well known low and rounded kopjes carrying large scattered blocks, which
may show all kinds of phantastic forms. Even from some distance it is easy to recognise the presence
of this granitic scenery, as for instance on reaching the Uitloop siding north of Potgietersrust. Where
schists are strongly developed, they often stand out in marked contrast to the surrounding granitic
country, as for instance near Marabastad.

The average type of the Older Granite is a fight greyish, usually rather coarse, rock, which
besides orthoclase and quartz frequently also shows muscovite to the naked eye. Often it tends to
become banded and sheared through the ferro-magnesian constituents being arranged in layers and
bands, which may alternate with coarse pegmatitic portions, so that the general appearance of the
rocks shows many variations over small areas. Such features can be well studied along the Haenerts-
burg main road at Smitedrift, where it crosses the Malips River and also for a few miles further east,
as well as over the area east of the Pietersburg railway north of Potgietersrust. Here some varieties
are banded in such a way that the quartz grains are drawn out into long thin veins with a very pale
pink colour ; but in thin sections there is very little evidence of crushing, so that in the absence of
any marked mechanical deformation, the banding must probably be regarded as a primary effect,
occurring during the consolidation as a result perhaps of differential movement in the still molten
magma, analogous to the classical example described by Geikie of the banded Tertiary gabbros in
Skye. Cases were met with, where the prevailing type of muscovite rock passes into one containing
in addition a dark greenish biotite, here and there changed into pale green secondary chlorite. Such
a variety occurs near Smitsdrift, and at the same time the banded structure is clearly noticeable in
the parallel orientation of the micas. A slightly different variety is the coarser and darker coloured
rock associated with the gold-bearing quartz reefs at Palmietfontein. Here the predominant mica
is biotite, but the ferro-magnesian constituents are only sparingly present. The orthoclase has under-
gone pronounced decomposition and becomes densely crowded with minute colourless flakes of
sericitic mica ; in addition to the usual minerals, there is also a small amount of calcite.

73

Sometimes the granite loses its greyish white colour completely and passes into bright or
pale pinkish-red varieties ; this is the case in the M'Phatleles River, a few miles north of the point
where it cuts through the Strydpoort Ranges. Another more extensive occurrence is on the farms
Driefontein and Rietvlei, where the red variety of the Older Granite occupies several square miles.
The mode of weathering is identical with that of the Bushveld Granite, but the colour in the hand
specimens has less of the brick red tint so common in the younger mass, and is more of a pale
salmon-pink shade. The rock in the bed of M'Phatleles River is a finer -grained mottled muscovite-
biotite granite, in which the darker mica is almost completely converted into chlorite. There is
abundance of extremely fresh and beautifully regular microline. Appearances in the field seem to
indicate that this red type forms a more or less well defined zone in the more normal granite, poBsibly
due to a slightly later intrusion. The rocks near the Yzerberg and north-east of Uitloop belong to
a variety of the red Older Granite usually more or less banded. Another good outcrop of this same
type of rock is in the bed of the stream south of Strydport, immediately underlying tie Black Reef
Quartzite. Here the rocks are much less basic, the ferro-magnesian constituents very poorly
represented, microcline is plentiful, and the quartz grains show signs of deformation, as if the pressure
due to the intrusion of the Bushveld Plutonic Complex had to some extent affected the floor on
which the sedimentary series was laid down.

The red variety does not occupy any well defined areas in the common grey granite,- but it
resists weathering more strongly, is usually well jointed, and appears to send felspathic and quartzoBe
veins up to three inches in thickness into the country rock granite. Probably therefore it represents
a slightly later phase of consolidation (" Nachschub ") ; it is improbable that the red type is directly
connected with the Bushveld Granite.

In several places, the Older Granite is associated with various kinds of basic schists mainly
in two places ; one of them is the belt of sedimentary and igneous schists, already alluded to, near
Marabastad. These rocks and their relationships will be discussed later, when the Moodies Series
is under consideration. Another body of schists, mainly if not entirely of igneous origin, is found
over wider areas at and round Haenertsburg. In several localities the granite is clearly seen to be
intrusive in the schists. Thus on the top of the bill overlooking Haenertsburg and close to the main
road are fine exposures of gneisses and schists veined by granitic apophyses, while one can sometimes
notice blocks of this material eighteen inches across, caught up by the granitic mass. The bulk of
the schists are dark greenish, slightly coarse-grained, lustrous greenstones or amphibolites, com-
posed almost entirely of strongly pleochroic idiomorphic hornblende, with which some patches of
serpentine are associated (198 c,). It is not always possible to discriminate in the Held between
such rocks and others, also rich in hornblende, but behaving rather as basic segregations in the
granitic magma. Near the Transvaal Police post at Haenertsburg occurs one such
doubtful mass, composed of an evenly medium-grained, greenish black rock, speckled white with
quartz and plagioclase.

Another small mass of a basic schist occurs on the main road from Pietersburg to Chunies-
poort south of Palmietfontein and must be classed as a compact dark greenish amptubolite, much
finer grained than the Haenertsburg schists.

A few yards from Mundt's old store on Kleinfontein, a thin band of another variety of schist
crosses the main road to Haenertsburg. This is a light silvery grey, hard sericitic quartz-schist,
evidently of sedimentary origin and not unlike some of the schists found associated with the Mount
Mare rocks. Macroscopically no minerals are recognisable except quartz, coated with very delicate
soft films of sericitic muscovite. The thin section shows marked evidence of mechanical deformation ;
the quartz grains are crushed into several smaller fragments, giving pronounced undulating extinction,
or they may be replaced by a fresh mosaic, while the sericite is arranged round the larger grains in
sinuous masses of slender, colourless needles. The relationship between this quartz schist and the
granite could not be seen.

Scattered irregularly through the granite are found several small masses of dark coloured
basic rocks, which form no particular feature and are very liable to be overlooked in consequence.
It is probable that these are intrusive and of much later date. Good examples occur east of Smits-
drift and on the farm Klipspruit about seven miles south of the Haenertsburg main road. The
former rock is a fine-grained pale greenish type, much altered by secondary changes and made up
essentially of hornblende (uralite) and plagioclase with dioritic affinities. The second rock is darker
in colour and weathers with the familiar brown crust, producing the spheroidal blocks so characteristic
of basic intrusive rocks in many localities. Mineralogically this rock consists of colourless" augite and
fairly fresh plagioclase in ophitic relationship. Provisionally therefore it may be referred to the diabases.

North of Strydpoort in the bed of the river occurs a dyke-like mass of a dirty greenish fine-
grained rock weathering somewhat after the manner of a schist. Its strike is roughly parallel to the
average direction of the river. The thin section is crowded with elongated lath-shaped oligoclase
and secondary hornblende, derived from augite. The rock is very much decomposed. Near the
same locality occurs another body of a less acid type, which also seems to belong to a later stage than
the consolidation of the main body of the granite. This is a dark greyish fine-grained rock, with a
Blight tendency to porphyritic structure due to larger pale flesh-coloured felspar phenocryBts. The
component minerals are quartz, orthoclase, and dark brownish -green biotite ; this rock may be
termed a quartz-porphyry. Near the Yzerberg, on the southern portion of Driefontein was found
another quartz-porphyry, lighter in colour and more coarse-grained, consisting almost entirely of
quartz and orthoclase.

Apart from the schist belt of Mount Mare, the following groups of rocks are therefore associated
with the Older Granite : —

(1) Amphibolites, greenstone schists and allied varieties, into which the Older Granite is
intrusive, e.g. at Haenertsberg.

(2) Basic rocks irregular in form, intrusive in the granite and of later age, e.g. on Klipspruit.

(3) Quartz-porphyries and pale red granite, probably belonging to the main mass of the normal
grey granite, but consolidated at a slighter later period, e.g. at Strydpoort.

(4) Dykes of basic character intrusive in and later than the granite.

(C) The Swaziland System.

Apart from the Older Granite with its associated basic schists, formations belonging to this
system are represented by a considerable ' thickness of more or less obviously sedimentary schists,
which form the well known belt south of Marabastad. The name Moodies Series has been proposed
for such types as a convenient and provisional term applied to sedimentary rocks older than the
Lower Witwatersrand.*

Although the occurrences of these sedimentary schists is generally associated with the con-
spicuous range close to Marabastad and known as Mount Mare, they form in reality a narrow but
very extensive belt reaching from close to the Pietersburg railway a few miles north of Potgietersrust
through the Yzerberg and Mount Mare in a north-easterly direction at least as far as the main road
connecting Haenertsburg with Pietersburg, a distance of some thirty-five miles. Along the whole
length of this belt varieties are found, which can be more or less accurately matched at Mount Mare,
but the rocks from the south-western termination to roughly the centre of Roodepoort are almost
wholly schists, phyllites, and other types of sedimentary origin, while south-east of Roodepoort and
especially on Palmietfontein compact greenish chloritic and other schists of igneous origin come in,
associated locally with serpentine and allied rocks. No sharp line can be drawn in the field between
these various formations, and on the map on Plate XVII, the whole belt, excepting near Uitloop, is
represented as belonging to the Moodies Series ; further south from Mount Mare lies a tract of low
and large kopjes, over the farms Eersteling, Turffontein, Waterval, and Vrischgewaagd composed
mainly of fine-grained schists of basic igneous origin, though sedimentary rocks somewhat resembling
ottrelite-schists are also represented locally. This portion is represented as " Basic Schists " on the
map.

The effect of the main belt of schists on the general topography has been indicated above,
as far as Mount Mare is concerned, which forms a well defined unit in the landscape and contrasts
strongly with the general slightly undulating surrounding country. Towards the south-west this
range rapidly dies out near the Marabastad- Potgietersrust road, and, following the latter, one traverses
several miles of fairly level ground until the Yzerberg is reached. This striking landmark forms a
high hill terminating in two peaks separated by a slight saddle, and with outlines recalling the open
wings of a butterfly, when seen from a little distance (see Plate VI, Fig. 2). The hill is composed of
banded ironstone dipping some 60 degrees towards the south-east and forming well marked dip
slopes. Passing over the nek on the north-eastern boundary of Zwartkrans, the main road steadily
descends towards Potgietersrust and on the right of it lies a continuous tract of ground forming a
striking ridge, flattened at the summit and composed of a great thickness of greyish-white more or
less strongly sheared quartzites, while the lower ground between this ridge and the road is occupied
by softer dark coloured schists, partly of igneous origin. Near the centre of the farm Weenen the
high belt rapidly diminishes, but the schists can be followed towards Uitloop, where they come to

• H. Kynaslon.- Ouol. of the Komati Poort ('ualfielil, I'returm, I Of III, |>ugo 12.

7-j

lie against the norite and Older Granite, while between the Marabastad main road and Uitloop they
are unconformably overlain by the quartzites of the Black Reef Series. To the north-weat of the
main ridge and parallel with it runs another horizon of banded ironstones closely resembling those
found on Mount Mare.

The latter range also rapidly dies out when traced in an easterly direction, before the main
spruit is reached, which drains across the centre of Zandrivier. East of this line, the schist belt
forms low lying ground, rising here and there into smaller kopjes, some due to the "calico-rock," as
on Roodepoort, others as at Palmietfontein and south of Pietersburg generally, composed of granite
or serpentinous schists. Some of these granitic kopjes are within the belt of schists and completely
surrounded by them, thus indicating in their lenticular shape an origin as apophyses from the main
mass, intruded into the schists. A very conspicuous white sharply crested kopje occurs on Welte-
vreden composed of a steeply dipping glassy quartz rock, but it is not certain whether this should be
regarded as of igneous or sedimentary origin.

Mount Mare itself affords the most comprehensive and reliable section across the schist forma-
tion and it will be necessary to consider the succession as shown south of Marabastad. The section
on Plate XIII is given as the mean result of various measurements taken across different parts of
the range, while the information thus obtained was corrected and amplified by numerous additional
traverses. Near the eastern end of the range occurs a small well defined poort cut deeply into the
lowermost ironstone band, which gives rise to the most northerly and persistent ridge. Close to
this point stood the battery of the old Eveline Gold Mining Co. About a mile north-west of Eveline
Poort lies a single low kopje covered by the debris of a whitish quartz reef and known as Witkop.
One section was taken from across this point through Eveline Poort. It is represented on Plate XIII
by the lower part of the succession up to and including the beds just above the middle ironstone
band. The overlying beds were measured on Snymans Drift and Hollands Drift, while the detailed
succession from the middle ironstones upwards to the coarse conglomerates is best seen near the
eastern end of the range at an adit facing southwards and driven into the schists in connection with
the development of Brown's Reef. The very variable series of rocks represented on Mount Mare
cannot all be fully dealt with here ; they may be roughly grouped into the following classes : —
(a) Banded Ironstones or " calico-rock."

(6) Conglomerates, ranging from coarse normal varieties down to beds in which the pebbles
have been squeezed out so as to become converted into long ribbons ; the latter
are rich in tourmaline.

(c) Quartz-schists, sheared quartzites, and other foliated rocks rich in quartz and showing

wide-spread tourmaline.

(d) Phyliiles and slaty rocks.

(e) MdamoTphic schists with either ottrelite or andalusite, or both.

The beds are all conformable to one another and show a fairly uniform dip towards the south
and south-east of about 45 degrees. On Hollands Drift and elsewhere dips may locally rise
to 60 degrees. In view of the clear evidence of the great pressure, which has affected all the rocks
more or less strongly, one has some hesitation in accepting the succession actually seen as the true
one, but there are no indications in the range of any reversal of the dip or marked folding, of which
we would expect to find some traces, if the beds had suffered isoclinal folding. Also the section given
on Plate XIII does not 8how any repetition of beds in the same order. Certainly faults do occur on
Mount Mare, notably on Snymans Drift, and as one approaches the western end of the range the
middle horizons gradually approach the lower ones until the two ridges due to the lowermost and
middle ironstones are merged into one. The general character of the rocks throughout strongly
suggests the operation of very powerful pressure, so that the possibility of reversed faulting and
thrust-planes must not be lost sight of. Along the range at Eveline Poort this compression is least
marked and consequently here the succession is more open.

The structural features are well brought out by the presence of three horizons of banded
ferruginous rock, and after their termination at the Marabastad - Potgietersrust main road three
similar bands are seen further south-west at the Yzerberg. The particular horizon of banded iron-
stone, of which this hill is composed, is best correlated with the upper one, for both here and on
Hollands Drift it is directly underlain by red slates, not found anywhere else in the succession.
Another band has been alluded to above as lying below and to the north-west of the great ridge of
quartz-schists beyond the Yzerberg. It forms here, as at Mount Mare the lowermost horizon and
is well seen on the farm Landsberghoek, while the middle band is probably represented by another
small outcrop north of the Yzerberg. The great fault which runs from Weenen more or less along

7fi

the main road cat soff the uppermost ironstone band not only where it should cross the main road oil
Zwartkrans, but probably also nearer Mount Mare on Rictvlei. The lateral displacement of the
Black Reef Quartzite from the Yzerberg to Weenan agrees approximately with that, of the upper
ironstone band from the western end of Mount Mare to near the Yzerberg, and it is hence concluded
that the schist belt from the latter locality to near Uitloop is a faulted one, forming the geological
continuation of the Mount Mare range.

The banded ironstones are the most striking and characteristic members of the Mount Mare
succession ; being readily distinguished and occupying definite positions in the sequence, they
become of great stratigraphies! importance. Three separate horizons can be marked off. Near the
base occurs the lower banded ironstone, composed of two separate bands, each about 50 feet in
thickness and separated by about 200 feet of schists. It is to these hard rocks that the well marked
and straight high ridge is due, which presents a bold escarpment to the north and trends parallel to
the railway for some miles south-westwards from Marabastad. To the south this is followed by
a deep valley bounded on one side by steep dip slopes of the upper of the two bands. The middle
ironstone band forms the crest of the second ridge. It is about twenty feet thick and consists of one
band only. Along the Eveline Poort section only these two horizons are seen, but on Hollands Drift
occurs another well defined range due to the third horizon of " calico-rock." This again is a double
band ; the uppermost is about 100 feet thick and separated by a great development of soft thinly
bedded reddish shales or slates from a second much thinner band below. These various bands
closely resemble one another and it would be difficult to identify any single bed from a hand specimen
only. The association of red shales is typical of the upper horizon, which has thinner and more
contorted bands than any of the lower ones. The rock has been referred to by other writers and
little need here be added to the description given by tiotz* who has also figured some of the contortions.
The pronounced banding due to alternate layers of compact black ferruginous material and white
or greyish granular quartz is highly characteristic throughout, and has given rise to the very suitable
designation of '" calico-rock."

Conglomerate beds occur also at fairly definite horizons, but unlike the last mentioned group,
show great variations in their petrographical characters. No description is given of them by Gcitz*

The most conspicuous and remarkable band occurs near the base of the series below the
lowermost ironstone, and is very well shown in some open workings not far west of the road which
runs past the eastern end of the main range (most northerly ridge). It is about five feet thick and
interbedded in quartzitic schists, some of which are distinctly green and, but for their markedly
sheared character, might be closely compared with the Green or Hospital Hill Quartzite of the Lower
Witwatersrand Series. The pebbles are mostly composed of finely granular or sugary-white quartzite,
but their original shape is almost entirely lost, as they are drawn out into lenticles or long rectangular
bands, the length of which may be more than ten times the thickness. These lie in a pale dirty green
impure quartzitic matrix, while cutting clean across both pebble and matrix are found numerous
dark coloured veins composed of tightly packed and very slender needles of blackish tourmaline,
such needles being always arranged with their longer axes perpendicular to the vein (see Fig. 2,
Plate VIII). These veins are up to a quarter of an inch thick and stand out sharply on the weathered
surface. Thin sections show very strongly marked cataclastic and other structures due to great
pressure, while the general appearance is suggestive of a rock which has undergone mechanical
deformation when in a more or less rigid condition. "Eye" structure is common, and the ''eyes"
of quartz are outlined by fine filmy sericite, while under crossed nicols the transparent portions are
resolved into a finely granular fresh quartz- mosaic, which sometimes forms larger continuous portions,
and sometimes winds in and out between the larger and strained quartz lenticles. It seems that
some at least of this mosaic is derived from the breaking up of the original pebbles owing to intense
strain. Undulatory extinction is wide-spread (see Figs. 4 and 5, Plate VII).

No higher horizons of conglomerates are found until about 500 feet above the middle ironstone
band; the conglomerates are here much less sheared and, in fact, the type just described is not again
met with in the entire succession. The upper series of conglomerates are not sharply separated from
the associated Bchists like the first sheared variety, but behave rather like pebbly washes with well
rounded quartz pebbles up to four inches in length. These are scattered without any definite relation
to the bedding in a pinkish-yellow quartzitic matrix and may be studied at and a little south of
Brown's adit above fie middle ironstones at the eastern end of the range. One band, about thirty

* J. Gotz.— Untersucbung einer licstciiiMiiite aus (leu (ruldfeJdurn tod Marabastad, etc. N.J., 18H4.

77

feet thick, is more sharply differentiated from its associated quartz-schists, but a short distance above
it a considerable thickness of an ill-defined conglomerate is well seen at Day's beacon ; its horizon
is no doubt identical with that seen above the Valley south of No. 3 hill on Snymans Drift, whence
the conglomerate can be traced to near the Hollands Drift fence. This is probably the band referred
to by Mr. Kynaston.* On Hollands Drift two other conglomerates were also found, but neither of
these is as much sheared as the one below the lowest " calico-rock," and it may be said in general
that the higher one ascends in the series the less pronounced does the shearing become, although the
succession as a whole is to gome extent sheared throughout. A striking contrast is thus presented
between the lowest and highest conglomerate, but it must be remembered that the character of the
latter is such, that if subjected to greater compression, it would probably assume a character more
closely resembling that of the lowest horizon.

Quartz- schinls. and sericitic sheared quartettes and allied foliated rocks rich in quarts are widely
distributed and may assume a great thickness. They are more prevalent near the lower and middle
banded ironstone, while in the upper horizons they remain quite subordinate or disappear altogether,
their place being taken by a considerable succession of soft red slates. The prominent range from
near the Yzerberg through the southern portions of Driefontein, Landsberghoek, and TJitkyb is
composed almost entirely of whitish grey slightly sheared quartzites and quartz-schists, bearing
a very close resemblance to the schists associated with the lowermost or sheared conglomerates at
Mount Mare* and belonging to horizons some way above the middle ironstone. In thin Bections the
most important feature is the widespread distribution, always in small quantities, of minute grains
of bluish tourmaline and the general appearance suggests a somewhat regional impregnation of the
rocks by this mineral.

Sometimes a section becomes crowded with fine deep blue granular tourmaline arranged like
interstitial matter round larger quartz grains, thus causing a kind of mesh structure (see Fig. 1,
Plate VIII). Tourmaline was only found in conspicuous veins at the eastern end of Mount Mare,
but its presence in thin sections was proved from several outcrops along the escarpment side of the
most northerly range and in rocks taken on Driefontein and Landsberghoek. It is significant, that
between Witkop and Eveline Foort granitic veins are found intrusive in dark green schists and
carrying abundant needles of tourmaline well seen in hand specimens, for it suggests the origin of
this mineral from an intrusive granite mass. A green colour is also a common feature of the
quartz-schists and is found at widely separated localities.

PhyUites and slaty rocks are common a little below the middle ironstone along the Eveline
Poort section and further west near the coarse conglomerates above the middle ironstone on Hollands
Drift, where a considerable thickness, mainly due to red shales or slates and amounting to over
1,000 feet, begins at the base of the uppermost " calico-rock." Some varieties of the soft argillaceous
rocks are practically indistinguishable from Bome of the red slates belonging to the Lower Witwaters-
rand Series ; thiB statement applies especially to the rocks underlying the " calico-rock " on the
northern slopes of the Yzerberg. In view of the very powerful pressure, which certainly affected
the lower horizon at Mount Mare, it is remarkable how little the rocks on Hollands Drift are
structurally changed.

Metamorphic rocks with Ottrelite and Andalttsite or only one of these minerals have been
admirably described by Gotz in the paper alluded to so that nothing requires to be added to his
petrographical accounts. The occurrence of such metamorphic schists is of great importance, and
some time ago Dr. Hatch assigned the Mount Mare beds to a series older than the Witwatersrand
System, on account of the presence of these rocks, which in his opinion suggested a thermo -metamorphic
origin due to an intrusive granite.. t The schists are found only at a single horizon above the middle
ironstone band on the Zandrivier Mynpacht about half a mile east of the south-western boundary
of Zandrivier at the end of a narrow kloof cutting through the second ridge.

Followed along the strike they soon disappear, though their presence is indicated a little west
of the mouth of Brown's adit further along the strike. Their vertical distribution is very limited,
and it is remarkable that such rocks, if of contact metamorphic origin, should not be more in evidence
at lower horizons nearer the granite, or below the uppermost ironstone, where a great development
of argillaceous rocks occurs with characters very suitable to receive and retain thermal influences
by means of new contact minerals, bearing in mind Dr. Hatch's view of regarding Mount Mare as a
belt of sediments caught up by a granite mass. Apart from this one horizon, ottrelite has also been

78

met with in one or two occurrences of quartz-schist, sparingly and in small ill-defined crystals not
visible in hand specimens. As typically developed, the normal types are silvery greyish or pale
greenish hard phyllitic schists showing ottrelite visible to the naked eye in numerous bright green
rounded flakes, less than a pin's head in size and scattered irregularly throughout the rock. Those
schists, which are very rich in or contain only andalusite as the metamorphic mineral, are much, softer
and dark slaty grey to lustrous black in colour with a kind of graphitic or carbonaceous appearance.
The weathered surfaces often Bhow the andalusite Btanding out as roundish flattened knobs
up to one-eighth of an inch in diameter, and hence the rock has been aptly termed a " knotted schist."
No crystal outlines are shown in the andalusite. A bed may show much variation along the strike
and pass rapidly into schists without visible secondary minerals, and then resemble a grey phyllite,
while vertically rocks rich in ottrelite may quickly pass into varieties showing both this mineral and
andalusite in varying proportions. Such changes are sometimes complete over a few feet and point
to an intimate association of the minerals, the origin of both being most probably due to the same
cause. Sometimes one finds very pale greenish grey varieties characterised by a faint silvery lustre
and showing no ottrelite in the hand specimens, while thin sections reveal abundant crystak of this
mineral. This is probably due to a greater local intensity of the pressure resulting in an increased
amount of squeezing out and flattening until the ottrelite flakes are merged into the sericitic films
of the groundmass and thus disappear macroscopically.

Thin sections show a thoroughly schistose structure characterised by idiomorphic and fairly
fresh ottrelite, practically colourless, but marked by very clearly defined hour-glass structure (see
Fig. 3, Plate VII), while the andalusite remains only as irregular wavy and cloudy semi-opaque patches
(see Fig. 1, Plate VII). The groundmass is mainly due to thin sinuous strands of biotite. For
further details the reader is referred to Gotz's paper.

Of other rocks found at Mount Mare, mention need only be made here of a soft highly decom-
posed garnet-schist carrying that mineral in numerous black well shaped crystals about one-eighth
to one-quarter of an inch in diameter and found as a narrow band intercalated in the phyllites below
the middle ironBtone, on the little footpath which leads across the range through Eveline Poort. It
is probable that this mineral is likewise of metamorphic origin. The groundmass of the rock cannot
be definitely determined.

The succession may be grouped aa follows : —

Approximate thickness in feet.

Uppermost Ironstone band 100,

Bed Slates 800

Upper Ironstone band .. .. 30 1 Upper Mount

Red Slates ; Phyllites ) I Mare

Pale greenish compact Schists, Quartz-schists, with thin slightly {• 1,300/ Series,
sheared conglomerates .. .. ) I 3,180 feet.

Coarse conglomerates with well rounded pebbles , . I q». |

Ottrelite and Andalusite Schists . . . . . . ) /

| Middle Mount

Middle Ironstone band 20 1 Mare

Quartzitic, Chloritic, and other Schists 600) Series,

I 620 feet.
Lower Ironstone band .. .. 50 \

Quartzitic and compact Green Schists 200 j

Lowermost Ironstone band . . 50 (Lower Mount

Sheared Quartzites and hard Compact Greenish Schists . . 250/ Mare Series,

Highly sheared conglomerates with tourmaline veins .. 5i 755 feet.

Green sheared quartzite and schists .. .. .. .. 200]

Total 4,555 feet.
Unconformity
Sheared and massive granite with basic schists.

The total thickness is therefore well over 4,000 feet, but as neither the base nor the top is
seen, the estimate is only approximate.

Regarding the mode of origin of the sheared rocks as a whole and their stratigraphical position,
Gotz regarded them as due to dynamic metamorphism, and, in spite of the fact that he was presumably

79

unaware of the proximity of granite, his view is a very reasonable one, for the petrographica!
characters of all the rocks are more or less strongly suggestive of the influence of great pressure. This
is the most striking and constant structural feature of the entire succession from the lower of the two
upper bands of banded ironstone downwards ; metamorphic minerals have to be carefully looked
for and unless a definite horizon at a single small locality is examined, even the ottrelite and andalusite
schists might easily be missed. Both these minerals have been recorded in many localities elsewhere,
for example, in the Ardennes, and shown to he due to dynamic metamorphism, without the presence
of an intrusive body being demonstratable. On the other hand ottrelite occurs in the Rustenburg
and Marico Districts under conditions which can leave no room for doubt that it owes its origin to
purely thermal metamorphism (see present volume, page 133). Hence this mineral cannot be
regarded as denoting any particular type of metamorphism, and the same is even more true in the
case of andalusite.

In assigning the Mount Mare beds to the Swaziland System, that is to a period older than the
Witwatersrand, Dr. Hatch* assumes the presence of the above-mentioned ottrelite and andalusite
schists as a proof of metamorphism in a granite contact zone ; enough has been said, however, above
to show that this occurrence by itself would scarcely be sufficient to justify the conclusion that the
Older Granite must be intrusive into the Mount Mare rocks. Dr. Hatch also points to the more
highly disturbed condition of the beds as compared with the Witwatersrand Series, and maintains
that this " is in itself an important indication of stratigraphical differences." But it cannot be
admitted that differences, due to a greater amount of tectonic disturbance in one group of rocks as
compared with another, have any value as principles controlling the relative ages of two formations.
Hence it must be concluded that the intrusive relationship of the Older Granite at Mount Mare
cannot be relied upon merely on account of the presence of ottrelite and andalusite schists. Further-
more, Dr. Hatch, satisfied with the non-intrusive relationship of the Older Granite to the Lower
Witwatersrand Series, lays stress on the absence in the latter of such metamorphic changes as are
seen at Mount Mare. Recently, however, slaty rocks with both ottrelite and cordierite have been
found near the base of the Orange Grove Series at Witpoortje, and under such conditions as to exclude
the possibility of the rocks belonging to the underlying schists, associated with the Older Granite.
A description of their mode of occurrence and petrographical characters is given elsewhere. f It
is sufficient to state here that there is the closest resemblance between the metamorphic schists from
the two localities. It does not, however, follow that the Older Granite must therefore be intrusive
at Witpoortje, though this conclusion would be quite in reason, if the presence of such metamorphic
minerals in the Lower Witwatersrand were explained in accordance with Dr. Hatch's interpretation
in the case of Mount Mare. Apart from ottrelite and andalusite, tourmaline has been shown to occur
at many horizons in Mount Mare and under wide-spread conditions and this mineral is a thoroughly
characteristic contact product, occurring quite as commonly at Mount Mare as it does in the thermally
altered shales of the Pretoria Series. Possibly the remarkable absence of chiastolite and staurolite-
slates, of hornfels and cordierite contact-fels among the Mount Mare rocks, which one certainly should
expect with thermal metamorphism due to a very extensive granite mass, is due to the chemical
composition, predominantly siliceous, of the greater part of the Mount Mare beds, which was not
suitable for pronounced molecular rearrangement. It has been pointed out in several occasions
elsewhere, how one and the same horizon of shales in the Pretoria Series may yield different varieties
of metamorphic rocks.

An actual contact is unfortunately not seen at Mount Mare under conditions which would point
to only one conclusion, though tongues of granitic rock can he seen intrusive into green soft schists
in the bed of the little spruit on the floor of the valley between Witkop and Eveline Poort. The veins
behave like apophyses from the granite, which occurs in situ close by at Witkop, and carry visible
needles of tourmaline, but the general appearance of the intruded schista is more like that of a basic
igneous rock, so that the evidence of this solitary instance of a contact is unsatisfactory.

More definite indications of an intrusive contact are seen on the farm Roodepoort, some miles
north-east of Mount Mare and four and a half miles south of Pietersburg. On the west side and not
far from the Old Smite Dorp road lies a low kopje, on which banded ferruginous ironstone similar
to that of Mount Mare is exposed in several little quarries. This locality is just within the belt of
schists, which extend from here unbroken into Mount Mare. The " calico-rock " shown in the quarries
probably is the lowermost horizon. A short distance to the south a light coloured banded fine-
grained granitic rock is seen forming a narrow tongue a few yards wide, which, traced westwards,

Trans. Cl.S.S.A., Vol. XII, 19051,

seems to run into the main mass of the granite. An actual contact cannot be seen, but the junction
between the finer-grained sheared variety of the igneous rock and the sedimentary formations on
the south can be fixed to within three or four yards. The latter rocks are soft pale greyish lustrous
mica-schists, the schistosity of which runs parallel with that of the granite. In the hand specimen
the rocks have s typically metamorphic appearance with large and numerous flakes of biotite in a
finely micaceous or phyllitic groundmass. Thin sections show the latter to be entirely recrystallised
into a finely granular and clear quartz mosaic full of filmy strands of deep brownish green biotite
(see Fig. 3, Plate VIII). Associated with these mica schists and overlying them are soft whitish or
pink highly micaceous and lustrous fine-grained altered quartzites, thin sections of which contain
abundant sericitic muscovite in a fresh groundmass of clear quartz grains. The conversion of slates
into mica schists and of quartzites into soft micaceous lustrous varieties is a common and striking
feature of the thermal metamorphism in the Pretoria Series of the Haenertsburg Goldfields and
elsewhere. Between the latter types and the altered rocks from Roodepoort there is a very close
general resemblance, and the conclusion is hence justified that the relationships seen in the latter
area point to a similar intrusive contact and consequent thermal metamorphism between the Older
Granite and the Mount Mare Series.

A third contact occurs north of the Yzerberg, some ten miles south-west of Mount Mare in
the north-western corner of the farm De Berg, where a sharp junction can be seen. While north
of this point the Older Granite has ordinary characters, it becomes sheared over the last few feet
from the contact, which shows delicate veins passing into the greenish schists. The latter resemble
some of the types seen below the ironstone range at Mount Mare and are composed of chlorite and
quartz, but not showing any metamorphic minerals identifiable with certainty. Near the floor of
the valley succeeding this locality towards the Yzerberg occur well bedded hard cherty compact
rocks closely resembling some varieties of the more siliceous and less crystalline hornfels of the altered
Pretoria Series. This evidence points likewise to an intrusive relationship, though it is less satis-
factory than that seen on Roodepoort.

Hence contact metamorphism is strongly suggested in the case of the Mount Mare succession
by the occurrence of evidently altered micaceous schists and sericitic quartzites ; biotite is by far
the commonest and most typical mineral of thermal metamorphism and the wide-spread occurrence
of tourmaline also points strongly towards the same kind of alteration, whereas ottrelite and
andalusite are not necessarily of contact metamorphic origin, though it is quite possible that they
may be due to the same agency here also, in view of the additional evidence just discussed.

If it be argued that the petrographical characters are due entirely to regional metamorphism
of a later date affecting both granite and sedimentary schists, the restriction of the foliated granite
to the junction, the preponderance of shearing in the lower horizons of Mount Mare and the great
vertical distribution of tourmaline, more especially below the Middle Ironstone range, must be borne
in mind, and are not easy to satisfactorily explain on such a supposition. The presence of a large
body of red slates, practically free from the effects of pressure and mechanical deformation, offers
a further difficulty to the theory of regional metamorphism, seeing that such rocks elsewhere readily
pass into schistose varieties.

Finally, attention may he called to the fact that several types seen in the Mount Mare
succession resemble more or less closely certain horizons of the Lower Witwatersrand Series. This
applies for instance to the green quartzites, the banded ironstones and the red slates below the
highest ironstone band ; these might be referred respectively to the Green or Hospital Hill Quartzites,
the Contorted Bed near the base of the Hospital Hill Slates, and the Red Slates below the Speckled
Bed, while the ottrelite-schists can be matched almost exactly by the occurrence recently found at
Witpoortje in the Orange Grove Series. It will be seen, however, that while taking the a trati graphical
succession of Mount Mare as it appears in the section on Plate XIII, the sequence and relative position
of the horizons alluded to do not agree with those in the Lower Witwatersrand Series.

In view of the above evidence the following conclusions seem justified :—
{1) That the Older Granite is intrusive in the Mount Mare Series.

(2) That both these formations have undergone considerable metamorphism from the

effects of pressure, which may also have been responsible for the formation of the
ottrelite and andalusite schists.

(3) That a correlation of the Mount Mare rocks with the Lower Witwatersrand Series

is not justified on the very small amount of evidence available, and that they should
therefore be retained for the present in the Swaziland System,

Annual Rtport, Geological Surrey, 1608.

*tdJ>-

Micro-sections of Metamorphosed Rocks In tbe Swaziland System ol Mount Hare, near Pietenbarg.

Fig. 1. — Ottreliie-Andalusite Slate, at the bottom ami on one siile irregular fiaichej- of Amlnlusite. Monnt Mart 1 , x 15 diam.

Fig. 2.— Ottrelite Slate. Mount Man;, x 15 diam.

Fig. 3. — Ottrelite Slate, lar^e crystal;! of Ottvclite with liour (rlahs struct lire. Mount Man;. X I- 7 ' iliam,

Fig. 4. — Sheared Conglomerate. Mount Man;, x 1"' diam.

Fig. 5. — Tbe same with Crnssol Niruls, shoivini: strain shadows and crushing.

Annual Report, (reoiogUal Surrey, 190$.

Micro-sections at Metamorphosed Rocks In the Swaziland System, near Pietersburg, and In the Pretoria Series
of tbe Rnstenbnrg District.
Fig. 1. — Quart*. Sehiet. Urannlnr Tourmaline outlining larger (|uarts jrraina. Mount Marc. X 15 <liam.
Fig. 2.— Sheared Conglomerate with a vein of Tourmaline. Mount Man?. X 12 diam.
Fig. 3. — Mica Schist (altered Shale). Large liiotid.' in a ojinplelelv lvcrvstslliscd ground mass of quarts mosaic. Koode

south of Pictersburjr. Orr.KM'd Nicol«. x liidiam.
Fig. 4.— Ottrelite Slate (altered Pretoria Series). Hear Rondavel Siding, llustenburg District, x l>"i diam.

Near the centre numerous -mall needles of Ottrelite. Lin

81

(D) The Black Reef Series.

The distribution of the quartzitea, sandy shales, shaly sandstones, and conglomerates, which
together make up the lowermost formation of the Transvaal System, is coterminous with the great
Strydpoort Range. To the north-west and to the east a bold and prominent escarpment is presented,
running across the country like a great wall and defining the southern limit of the Older Granite
area. This Strydpoort escarpment is nowhere as imposing as that of the Drakensberg, partly owing
to a decrease in thickness, but also to the climatic and hydro-graphic conditions alluded to above.
The crest line of the escarpment remains fairly level between Haenertsburg and the Malipe River,
but becomes more uneven and irregular further west, while the general level falls distinctly. East
of Zebedelas River several poorts occur, such as Strydpoort, Cnunies and M'Phatleles Poort, and
the narrow deep ravine which admits Malips River. The general characters of these gaps closely
resemble those of frequent occurrence in the quartzitea of the Pretoria Series, e.g. Wonderboom
Poort north of Pretoria, and are in every case due to the erosion of the rocks by the respective rivers.
The narrowest poort is that of M'Phatleles River ; both it and that due to Malips River are impass-
able except on foot more or less in the bed of the river. North-west of Zebedelas River the escarp-
ment becomes again much more pronounced, for the crest line rises and there is in consequence a
more striking contrast with the low lying granite country ; but the range is not cut through by any
stream here and has a more uniform elevation. About Strydpoort and Pienaars Nek, where the
Zebedelas River cuts across the south side of the range, the crest is fairly low and merges gradually
into the sandy country towards Zebedela's Location. A small purely local displacement is indicated
in the river bed below Pienaars Nek and one or two faults probably run through Chuniespoort, but
none of these breaks interrupt the general continuity to any marked extent.

The general succession making up this series can be well studied at Chuniespoort, and along
the bed of the little tributary which cuts through the range at Strydpoort to join Zebedelas River.
A section at the former locality is given on Plate XXIII of the Annual Report for 1904. The series
is here made up of a succession of quartzites, thinly bedded sandstones, subordinate shales, and
conglomerates, resting upon the grey granite, which extends some way up the northern slopes. Thin
beds of conglomerates occur here, as elsewhere near the base of the series. The summit of Chunies
Mountain is roughly 2,000 feet above the level of the river bed in the poort and is formed by massive
quartzite. On the south of the kopje along the small tributary joining the main stream from the
west, the upper horizons of the series contain intrusive sheets of diabase, but no very sharp line can
be drawn between the upper boundary of the quartzitic formation and the overlying Dolomite ;
the same feature was observed along the base of the Dolomite south of the Iron Crown. The Stryd-
poort section agrees with that -of Chuniespoort, but affords better opportunities of studying the
lowermost pebbly beds. Immediately south of this poort, some of the coarser horizons give place
to a very fine grained almost compact, bright pink quartzite, sometimes closely resembling a felsite.
This type is well shown in the bed of the river ; the thin section shows small rather angular quartz
grains set in a cloudy decomposed felspathic matrix, to which no doubt the pink colour is due.
Nearer to the underlying granite, the beds become much coarser and the flesh coloured felspathic
particles and layers are more pronounced, until a few yards from the base pebbly and conglomeratic
beds come in. These are very striking, owing to the preponderance of large bright red fragments
up to one inch in length, composed of flesh coloured orthoclase or portions of felspathic granite.
Some varieties closely resemble coarse weathered granite with slight indications of bedding. Thin
sections show well rounded quartz grains, and some large tabular orthoclases with small fragments
of microcline. The whole appearance is strongly suggestive of an arkose derived from the disintegra-
tion of a granite and rich in some of the most characteristic components of that rock. Clearly
therefore the junction between the Black Reef Series and the Older Granite iB a sedimentary one,
which bears out previous experience.

The dip of the series is higher on the east side of Zebedelas River than along the north portion
of the range. As a rule the base of the dolomite lies well up in the mountainous tract, which succeeds
the crest line on the south or west side, but where more powerful streams run along the strike near
the top of the quartzites, the latter make very striking dip slopes, well seen round Chuniespoort.

The thickness becomes gradually less from east to west and amounts to about 1,600 feet at
tile last-named poort ; from here there is a gradual but not very great thinning down towards Button
Kop near the Yzerberg ; at this point the range, and with it the entire series, is cut off by the great
fault alluded to above and the "quartzites reappear near the centre of the farm Weenen'on the north
side of and close to the main road leading from the Yzerberg to Potgietersrust. Unlike the Yzerberg,
the series now no longer makes anv marked feature, because it forms'the south-western portion of
7

82

a prominent ridge of quartzitic and other schists, which runs for some miles parallel with the main
road and on its northern side. Looking from the store on Zwartrand in the direction of Potgieters-
rust, a short krans can be observed high up -on this schist range. It denotes the faulted continuation
of the Button Kop krans on the east. From Weenen the Black Reef Series now continues unbroken
as far as the small kop lying about one mile east of the railway line at Uitloop, but the beds are
somewhat twisted and disturbed, so that one meets with several sharp curvatures in the strike. The
dip is also rather high, so that the series does not occupy any great width. There is moreover a very
rapid thinning down towards Uitloop and when the beds finally disappear the thickness must be
reduced by at least nine-tenths. What exactly happens further north-west is not clear, but it is
quite evident that for a distance of some two or three miles the entire series is missing, its presumable
position being taken by a detached mass of pyroxenite belonging to the Bnshveld Plutonic Complex,
and forming the well marked pyramidal kop about two miles north of Uitloop and a short distance
west of the railway. Very probably the intrusion of the noritic magma is the cause of this peculiar
behaviour in the manner suggested above.

(E) The Dolomite.

Overlying the quartzites of the Black Reef Series conformably there follows the great
succession of thickly bedded dolomitic limestones with chert and quite subordinate shales, which
form the middle division of the Transvaal System. Its general characters are closely similar to
those shown by the same formation throughout the Eastern Transvaal as far south as Godwan River
on the Delagoa railway ; but certain distinctive features, first noticeable near the junction of the
M'Thlapetei and the Olifants Rivers in the Haenerteburg Goldnelds become more persistent and call
for special notice. These are the presence of banded ferruginous quartzitic or jaspery rocks and
the common occurrence of bright pink or purple dolomite in certain horizons. In addition,
metamorphosed beds, some carrying contact minerals visible to the naked eye, come in more freely
west of the Malips River.

The distribution of the Dolomite formation is closely bound up with that of the underlying
Black Reef Series and coincides with the hilly and broken country following southwards and west-
wards from the main range of the Strydpoort Mountains. As there are no definite horizons, which
tend to stand out markedly owing to the differential rate of erosion, the Dolomite topography shows
no regular features, but gives rise to very irregular deeply dissected country extremely tedious to
traverse. Attention was drawn above to the striking evidence of compression between Chunies-
poort and the Malips River and the resultant highly folded character of the beds, especially in the
upper portion of the series. For this reason it becomes difficult to give a correct estimate of the
thickness.

The base of the formation, as usual, is not well defined, for a horizon of Bandy shales of no
great thickness comes in as one approaches the underlying Black Reef Series, well shown in the section
along M'Phatleles River on the farm Rivierplaate, where a small tributary joins the main stream
from the east. Similarly no sharp line can be drawn at the top, for the uppermost horizon seen east
of Chuniespoort is followed by a marked valley devoid of outcrops, the nearest overlying rock in situ
being a striking and rather thick coarse conglomerate belonging to the lowermost portion of the
Pretoria Series. The top of the Dolomite is therefore drawn a little below this conglomerate.

Banded ferruginous quartzites interbedded in the Dolomite are referred to for the first time
by Mr. Kynaston in the Annual Report for 1904, page 19, and a short account of their occurrence
Bouth of Chuniespoort was given by Mr. A. R. Sawyer. 1 * On the geological plan of Fig. 2, Plate XX,
in Mr. Sawyer's paper, one band of this rock is correctly shown. It is conformable to the dip and strike
of the associated dolomite and chert and can be well studied in the river bed close to a drift a few yards
west of the main road at the common beacon of the farms Morgenzon and Staanplaats ; this exposure
shows highly contorted rocks, which run up into the high kopjes bordering the river at this point and can
be readily traced both to the east and west. In the latter direction, the band may be readily traced
to the farm Rooiboklaagte, where it makes prominent dark coloured kopjes, such as the one on which
the south-easterly beacon of that farm is situated. Usually the banding is due to alternate layers
of hard yellowish impure jaspery quartz and black or dark deep brownish bands of hematite.
Sometimes thin layers of filmy crocidolite about one-sixteenth of an inch thick occur with the other
bands. In the other direction the rocks can be traced as far as M'Phatleles River, where they are
seen to be thrown into folds. A second and upper band outcrops at the Transvaal Police

" Tlie Geology of Chuniespoort, Transvaal." Trans. Inst. Min. Kim.'., l.cintion, ] !IOfi, u. 4 (A.C.)

83

poet on the main road from Chuniespoort to Boomplaats. It is thinner than that of the lower
horizon, which probably is some 150 feet thick. It likewise can be followed for some distance in
both directions along the strike, No doubt these two bands are identical with those described in the
last Annual Report* as occurring on the west side of Malips River in the Haenertsburg Goldfields and
probably extending thence at least as far eastwards as the M'Thlapitsi River. These banded
ferruginous rocks bear a general resemblance to the lowermost beds underlying the Timeball Hill
Quartzite of the Pretoria Series, but are separated from them by a series of dolomite and chert beds,
before the very characteristic conglomerate comes in, which denotes the base of the Pretoria Series.
Probably the same bands reappear after the strike of the Dolomite turns finally northwards west
of Zebedelas River. The strong pressure, to which they have evidently been subjected, must
undoubtedly be connected with the generally folded and highly tilted character of the whole Dolomite,
as a result of the great regional compression exerted on the Transvaal System consequent upon the
intrusion of the Bushveld Plutonic Complex.

A variation from the well ■ known lithological characters of the common bluish - grey
dolomite occurs on Rooiboklaagte and elsewhere ; here were found dolomitic rocks of a bright red
and purplish colour, weathering in the manner usually seen in dolomite, but presenting a very fine
grained or nearly compact structure on the fresh surface. Some varieties resemble the non-
porphyritic and more compact types of felsite, others have a highly acid character and might be
mistaken for quartzite. All of them show qualitatively the presence of silica, lime, magnesia, and
iron, but in varying relative amounts. Thin sections are composed mainly of calcite and quartz,
the latter scattered in countless uniform grains throughout the rock.

Other remarkable varieties occur on the prominent isolated kopje, on which stands the common
beacon of De Hoop, Weenen, and Makapansgat, on its escarpment side from the foot of the kopje
upwards. The lower horizons are more like the normal dolomite, but are overlain by several varieties
of a purplish kind, passing into yellowish or white rocks resembling coarse grained marble. The
most interesting rock in this succession is a purplish crystalline strongly marked oolitic variety,
crowded with many perfectly rounded grains, very uniform in size, ^ of an inch in
diameter and standing out clearly from the weathered surface. The structure is evidently original
and not confined to weathered surfaces. In a thin section the dark red grains show a markedly
circular outline resembling that typical, for instance, of the Carboniferous oolitic limestones of the
west of England, The material of these, one of which may, under a magnification of twenty
diameters, fill nearly the whole field of view, consists of evenly granular fresh calcite, similar to that
making up the groundmass, but stained irregularly by compact blackish irresolvable carbonaceous
matter ; sometimes a central portion is seen without any calcite and composed of similar black
pigment. In addition one section shows several long narrow slightly curved fragments, consisting
of fresh calcite, and recalling the familiar appearance of the fragmentary remains of Molluscs, and
other organisms found in fossiliferous limestones.

In other respects, the Dolomite formation in the present area is closely similar to its develop-
ment further east ; the usual characteristic narrow ledge like black kranzen, never persistent over
any great distance, the frequent layers of interbedded chert, and occasional tendency to form tall
precipitous cliffs serve to remind one of that formation as seen elsewhere in the Colony. An additional
new feature, however, deserves special mention, namely, the strong thermal metamorphism, to which
it has been subjected near the Northern Lime Works, owing to the intrusion of the Bushveld Plutonic
Complex, and a description of the resulting changes will be given below.

The average thickness of the whole Dolomite does not show much variation west from the
Malips River, but no reliable estimate can be given on account of the marked disturbances affecting
it. In the area south-east of Potgietersrust after its emergence from beneath the overlying Karroo
rocks, the beds are much less disturbed and here the thickness is near 4,000 feet ; like the under-
lying Black Reef Series, the Dolomite rapidly becomes considerably thinner towards the Pietersburg
railway line.

(F) Pretoria Series.

The uppermost division of the Transvaal System can be at once recognised, in spite of its
striking petrographical peculiarities, by the presence of the same three leading quartzites, which have
been shown on various occasions (see previous Annual Reports) to be remarkably persistent. In
the Eastern Transvaal they extend without permanent interruption from the Delagoa Bay railway

' Annvinl Rep. Oeol. Surrey for 1907, page 40.

section through Lydenburg to Potgietersrust, that is some 190 miles ; while they have been followed
from Pretoria continuously to Zeerust, another 150 miles. Their lateral extent must therefore
amount to at least 340 miles. In the last two Annual Reports a detailed account was given of the
gradual change which each horizon of quartzite undergoes when traced northwards from the eastern
railway line, until at Malips Drift on the western border of the Haenertsburg Goldfielde, one finds
three narrow prominent parallel ridges, separated by narrow and deep valleys, and giving rise to a
kind of wall-like topography owing to the high dip.

The present year's work west of Malips Drift showed these quartzites to continue in approxi-
mately the same relative positions as far as the northern railway, but it becomes increasingly more
difficult to follow and identify them, owing to the enormous pressure which has affected them
south-east of Potgietersrust ; hence no satisfactory correlation would be possible without a knowledge
of their peculiar characters and distribution in Seccoecoeniland. Thus one finds the Daspoort
Quartzite altered beyond recognition into a soft aandy sericitic schist, which can only be definitely
correlated by fallowing it along the strike into a normal quartzite east of Lydenburg. The powerful
folding seen south-east of Potgietersrust, the repeated and rapid changes in strike, together with
the fact that often a quartzite makes no feature at all on the surface for a considerable distance,
render the systematic mapping of the geological features of this part of the Transvaal System more
than usually difficult. But for the presence of such harder beds, many interesting problems
presented by the mechanism and consequent metamorphism due to the Bushveld Plutonic Complex
would remain obscure. East of the Zebedelas River the Pretoria Series forms a narrow belt about
four miles wide, in contact with the norite magma of the central plutonic basin, and producing a
sandy stretch of fairly level ground broken by more or less disconnected or isolated quartzite ridges
and kopjes, a type of scenery typically developed in and round M'Phatlele's Location. It is very
rarely that one finds any outcrops at all between the ridges of harder rock, and only close to the
Malips River can an approximately continuous section across the series be examined. North-west
of Zebedelas River a stretch of country some seven miles wide is covered bv Pretoria Series and an
excellent succession of beds is exposed, where the farms Doornfontein and Witpoort on the south
join with Pruizen and Rcoipoort on the north, for the beds between the Magaliesberg and Daspoort
Quartzites stand out as kopjes of hard crystalline homfels, while the great range of Timeball Hill
Quartzite towers on the top of an enormous thickness of slightly hardened dark coloured shales
belonging to the lowermost horizon of the series. It will be convenient to consider the mode of
occurrence and structural behaviour of each of the three quartzites as they are traced from east to
west.

T^e Lo-.rermost or Tim&att HiB. Oiiartzi'e is the thickest and most conspicuous of the three; it
consists of a single band without any marked ferruginous character, thus differing from the same
horizon at Pretoria. From the southern portion of the Haenertsburg Goldfields it passes unbroken
across the Malips River at the drift, and continues south-westwards as a very prominent wall as far
as the common beacon of Katkloof, Tygerpoort, and Leuwkop ; here its strike is sharply bent
towards the west and continues until it terminates at the great fault, which runs in a south-westerly
direction through M'Phatlele's Location. The beds all along show in a very striking manner the
effects of compression in numerous contortions and minor folds, so that a single stratum is rarely
traceable for any distance. Where this quartzite ends at the above fault, it forms the remarkable
conical kopje locally known as Wonderkop (see Plate VI), whose form is due to the high dip com-
bined with the abrupt ending of the rather sharply crested ridge. Possibly between this point and
the common beacon referred to, two or three short subsidiary faults occur also. Along the line of
the fault the outcrop has been shifted from the south-west to the north-east, bo that the quartzite
is picked up again at the north-eastern comer of Molsgat. Here it gives rise to another and less
disturbed ridge, perfectly continuous topographically, but composed of quartzite west of the fault
line and of norite east of it. Probably the close association, produced by this break, of harder
quartzite and softer norite explains the presence of a distinct notch in the crest line of the composite
range, not otherwise accountable by the hydrographical conditions there prevalent. Prom Molsgat
the Timeball Hill Quartzite extends westwards unbroken to within a mile of Chunies River and gives
rise to that very regular ridge, which at once reveals the beginning of the Pretoria Series to the
traveller who approaches the district from the Chuniespoort direction. Although not displaced,
there is abundant evidence of great pressure : thus close to Krutwig's Store on Naauwpoort the beds
are strikingly bent, twisted, and generally strained, so that no reliable dips can be measured.
Probably the average value is between 40 and 50 degrees. Proceeding further west, the quartzite
seems to end, topographically at least, just before reaching Chunies River, though it must be

85

Continuous geologically, as it forms an isolated kopje situated a mile or so on the other side of the
river. From here the beds rapidly die out as a feature and merge over a long sandy Btretch into the
extreme eastern end of the Springbok Flats. Consequently here the contrast between the " island
landscape " of the Pretoria Series and the vast expanse of the "Flats" is most marked. After being
buried for some thirty miles below Karroo beds, the Timeball Hill Quartzite reappears on Doornfon-
tein and quickly riBes into a superb kop, some 2,000 feet above the general level and presenting a bold
escarpment to the east. This conspicuous landmark forms the southern extremity of a continuous
semi-circular mountainous ridge, the crest line of which remains fairly uniform, while the remarkably
regular and wide dip-slopes distinguish it from the various subsidiary ridges seen to the south-east
of Potgietersrust. Finally, just before reaching the Marabastad main road, the range quickly falls
and the quartzite terminates just before reaching the railway line. Where one would expect to
find its continuation, only norite or other basic rocks are seen, but it is probable that certain isolated
masses of quartzite west of the railway line were originally connected with the main easterly outcrop.

The Daspoort Quartzite behaves in a general way much like the lower one, excepting that the
effects of pressure are more pronounced. West of Malips Drift it gives rise to a similar type of ridge,
only much lower than that of the Timeball Hill Quartzite and since the valley between the two often
attains a breadth of only 200 or 300 yards, the presence of a second hard band is sometimes scarcely
noticeable in the landscape. It undergoes the same sudden change in strike and is also out off and
displaced' by M'Phatleles fault. The continuation begins close to the eastern gate of the Location,
which leads through the western boundary of Molsgat, but the quartzite, unlike the lower one, is not
at once continuous, but forms two low isolated kopjes one on either side of and close to M'Phatleles
River, which it follows along the left bank as far as the drift, about one and a half miles south-east
of the Naauwpoort store, before reaching which point the ridge at one place practically disappears
as a feature, though geologically the continuity is never lost. Beyond the drift there is a straight
sharply crested ridge running parallel with that of the Timeball Hill Quartzite and enclosing one of the
characteristic kloof-like valleys frequently seen in the Pretoria Series of this neighbourhood. Near
Chunies River over the eastern portion of the farm De Beer occur several small isolated kopjes due
to this quartzite, which further west finally disappears, like the lower horizon, below the Karroo
beds of the Springbok Flats. From Malips Drift to the Chunies River this quartzite is intensely
metamorphosed and locally converted into highly lustrous micaceous varieties, hearing very little
or no resemblance to a quartzite. At the same time the beds throughout are much more bent, for,
as for instance where they approach the great fault, while local gaps in the continuity become more
frequent, the whole thickness is sometimes worn down to a level plain like that underlain by the
intermediate shale. Beyond the stretch of the Bushveld Amygdaloid the quartzite reappears near
the western boundary of Doornfontein and forms the second high range seen west of the main road
leading from Potgietersrust to Zebedelas Location. Traced northwards into the adjoining farms
Pruizen and Vierentwentig Rivieren, one immediately notices a striking anticline and syncline traceable
as a shallow trough for some two miles along the strike (see map, Plate XVII), and quite conspicuous
even from some distance. The valley between the Daspoort and the lower quartzite is once more very
deep and narrow and continues so to within a short distance of Potgietersrust. The nearer one
approaches this township the more irregular becomes the distribution of the middle quartzite ; thus
the little kop at the common beacon of Oorlogsfontein, Planknek, and the town lands must belong
to this horizon, though it cannot be directly traced into the second range. A short distance south-
south-east of the hill, on which the Transvaal Police post stands, east of Potgietersrust
Station, the Daspoort Quartzite shows the most northerly outcrop ; beyond this point it forms only
three or four small ill-defined patches, one of which is seen on the eastern slope of the Transvaal
Police hill a few yards below the fence, another at the cemetery, and another still further
north in the bed of the Dorp River close to the railway bridge. It only extends a very short distance
west of the railway and finally ends against the norite zone.

The Magaliesberg Quartzite exhibits the irregularity seen in the two lower beds even more
strongly. From Malips Drift it can still be followed westwards for a few miles, but south of Wonder -
kop it ceases to make any feature, only rising into a single low kop just before ending at the fault ;
in M'Phatlele's Location it is picked up again at a white detached piece of higher ground some three
or four miles south-east of the Naauwpoort store and striking with the middle quartzite in a southerly
direction. Towards the west there is no continuity, but the few solitary kopjes of quartzite to be
seen very probably belong to one band, worn down to the ground level of the surrounding country ;
the little feature juBt west of Chunies River on which Mr. Stewart's homestead is situated is.no doubt
Magaliesberg quartzite, dipping about 40 to 50 degrees to the south. North of the Springbok

Flats, this horizon ia first identified ae an elongated and isolated rather low kopje, on which the
common beacon of Rooipoort, Witpoort, and Grasvlei stands, but traced northwards it makes no
continuous feature. The extreme southern beacon of OorlogBfontein lies on a low circular kopje
showing a peculiar quartzite characterised by a kind of speckling due to pink felspar ; this must
no doubt be regarded as the Magaliesberg horizon. Another slightly more marked kopje lying near
the centre of the south-east boundary of the same farm is the last definite indication of die uppermost
quartzite, which here dips at a high angle. On the main road passing a mile to the south-west of
this kop occurs a very small patch on the east side of the road and exposed in some quarries a short
distance south-east of the railway line. It is the last outcrop of the Magaliesberg horizon. Further
along its strike only norite is found or metamorphosed dolomitic masses belonging to a lower horizon.
Regarding the intervening beds, which elsewhere are soft typical Pretoria shales and ahlet,
little need be said here, since the bulk of them are completely recrystallised hornfels, and very rarely
seen at the surface. True shales are only found below the Timeball Hill Quartzite along that por-
tion of the Transvaal SyBtem which strikes north and south near Potgietersrust. Here a particularly
well exposed aeries can be studied during the ascent of the prominent kop marked by the central
northern beacon of Doornfontein No. 1956 ; these beds consist of a very great thickness of thinly
bedded dark coloured slightly indurated shales, slates, sandy shales, and flagstones with one or two
intrusive sheets. The rocks are throughout unaltered or merely somewhat indurated, but without
any contact .minerals. Near the summit the overlying Timeball Hill Quartzite dips at about
22 degrees to the west and is a purplish grey gritty rock with a good deal of matrix. Thin sections
show fairly large and usually well rounded grains, the original outline of which is emphasised by a
distinct dark border, but many grains show considerable secondary enlargement, so that their optical
continuity with the new quartzitic matrix becomes very striking, Frequently the earlier grains
may show well developed strain shadows, but probably this phenomenon belongs to the rock, by
whose disintegration the present quartzite was formed, for as a rule the enlarged portion of such
grains are optically quite normal. The horizon here alluded to presents a fairly uniform succession
of shaly beds at least 1,500 feet thick ; it is interesting that it can be easily followed with identical
petrographical characters as far as the Flanknek springs some four miles east of Potgietersrust,
whereas beyond Zebedelas and especially along the Malips River, no sign of any shales was found
below the Timeball Hill Quartzite. From the easterly termination of the Bushveld Amygdaloid to
the great fault running across M'Phatlele's Location no rock in situ could be found between the
last-mentioned quartzite and the striking conglomerate at the base of the Pretoria Series, for this
stretch of country shows practically nothing except the three quartzites, the intervening strata being
covered by a thick deposit of bright red sand. East of the fault from Wonderkop to Malips Drift
the lowest Pretoria beds are banded ferruginous quartzitic slates or slaty ferruginous quartzites with
thin layers of yellow " tiger-eye " or bluish crocidolite, very similar in general appearance to the
two zones found inter bedded with the Dolomite. It seems, therefore, that a gradual change in
sedimentation took place from east to west, the harder quartzitic beds becoming more and more
shaly and less ferruginous. Where this change in facies first begins cannot be ascertained with
certainty, but the preponderance of the quartzite topographically and the wearing down of the
lowermost portion of the series to an almost level plane suggests a greater differential rate of erosion,
owing to the decreasingly quartzitic character of the beds below the Timeball Hill Quartzite ; thus
the change probably begins near the fault and is most likely complete before the Chunies River is
reached, after which a true argillaceous facies is established. Such features recall the phenomena
observed near the Steelpoort River, where the same horizons of shales, so conspicuous near Ohrigstad
rapidly become sandy and pass into banded ferruginous quartzitic rocks near the left bank of the
Steelpoort River.* A peculiar set of beds is found between the Timeball Hill and Daspoort Quartz-
ites over the southern portion of Pruizen and the adjoining farm Doornfontein close to the main
road leading over a nek at the common boundary between these two localities. They are well bedded
brownish to somewhat purplish impure quartzites containing much felspathic matrix, while on the
weathered surface the quartz grains appear rather angular like those of a grit. These beds alternate
with compact slightly micaceous and sandy yellow or dark greenish slates ; the whole series here
suggests a kind of passage between the lowermost quartzites and true shales above, but the latter
are beginning to be metamorphosed and differ from the unaltered beds below the Timeball Hill
Quartzite. Of particular interest is the presence in the Pretoria Series south-east of Potgietersrust
of a considerable thickness of rocks which resemble Dolomite in their mode of weathering. They

* Sec Ann. Rep. for 1907. p. 43.

87

belong to a definite horizon between the Daspoort and Magaliesberg Quartzites, and are best exposed
in the low hill immediately east of Potgietersrust, which has the Trarsvaal Police quarters
on its summit. This hill is at once identified by its whitish appearance due to the prevalent ground
colour of the rocks, which are metamorphosed dolomitic marls. Hitherto the presence of calcareous
rocks interbedded in this series has not been known, apart from a small occurrence mentioned by
Dr. Molengraaff* and lying near the summit of Van Lennops Kopje, near Waterval Boven on the
eastern line, at a horizon a little- above the Tunnel or Daspoort Quartzite. Near Potgietersrust
these rocks have a thickness of about 500 feet with a uniform westerly dip, but owing to the intrusive
transgression of the norite their exact position cannot be well decided at the Trarsvaal Police
hill, but follows clearly from their distribution when traced southwards along the strike
to the farm Viereutwentig Rivieren and OorlogBfontein. In the former locality there is no doubt that the
horizon is a little above the Daspoort Quartzite. The rocks occur also still further south on Witpoort
and probably form a continuous band, everywhere dipping conformably with the associated Pretoria
beds, and this fact precludes the possibility of their representing a faulted horizon of the true Dolomite,
subsequently altered by contact alteration due to the Bushveld Plutonic Series. That the latter is
the metamorphic agent follows readily when it is stated that at the station and on the railway line
just south of it, the marls are intruded by the norite and associated with black coarsely crystalline
cordierite-hornfels ; near the summit of the Transvaal Police hill similar highly altered
Pretoria shales are seen interbedded with the lighter coloured marls. These are hard finely
crystalline light greenish white rocks, showing bands of dark brownish green serpentine and frequently
traversed parallel to the bedding planes by veins of coarsely crystalline calcite. The quarries on
Oorlogsfontein show thickly bedded rocks with a very regular dip, which are used locally as building
material. The dark greyish appearance of the weathered surface strongly resembles that of ordinary
dolomite, with the difference that the colour is not so black. Apart from serpentine and carbonates,
no minerals are visible to the naked eye, but vesuvianite was found in large idiomorphic crystals
south of the station. An account of their composition and the reasons for classing this type as a
marl will be given below.t It follows that there exists in the Pretoria Series of this area a definite
horizon of considerable thickness composed of calcareous rocks, representing once more a change
of facies. The analogous phenomenon in the Dolomite has long been known ; thus repeated alter-
nations of shales and dolomite are found in the Blyde River Valley from Pilgrims Rest northwards.
Other thin intercalations of calcareous bands in shale (or altered shale) occur in the Pretoria Series
of M'Phatleles Location. f

Conglomerates are found at two distinct horizons, the upper in the lowermost portion or at
the base of the Daspoort Quartzite, the lower horizon just above the base of the Pretoria Series.
Conglomerate bands occur in the middle quartzite more or less all along the strike of that formation,
having been met with west of Malips Drift, on Chunies River, on the farm Pruizen south-east of
Potgietersrust and close to the station. In all cases the beds are rather like thin pebbly washes in
a continuous succession of arenacous rocks than like a well defined conglomerate. This is especially
well seen in the bed of the Chunies River and on the top of the sharp ridge due to this quartzite, where
it passes through the farm Katkloof a few miles east of Malips Drift. The pebbly horizons vary
from a few inches to a few feet and give rise to conglomerates, characterised by the presence of a
ferruginous matrix, in which fairly well rounded or sub-angular pebbles of glassy clear quartz or
opaque white powdery quartzite are interbedded, rarely exceeding half an inch in length. On
Pruizen a few bright red felsitic looking pebbles were found in this rock, measuring one and a half by
two inches. The peculiar persistence of the ferruginous matrix distinguishes this type of conglomerate
from those of the Waterberg System, for which it might be mistaken, where the latter formation is
represented south-east of Potgietersrust.

The lower horizon of conglomerate is considerably thicker and has very distinctive characters.
It was only found east of Zebedelas River extending from the Chunies River to within a short distance
of the fault. Characteristic and instructive outcrops may be seen on the main road from Chunies
Poort to Naauwpoort, on the latter farm, where the road passes over a low kopje about a mile from
the Naauwpoort store. It is the presence of the hardened conglomerate which gives rise to the kop.
Another good outcrop lies about a mile north of the road from Naauwpoort to the Malips River in
the northern corner of the last-named farm. The long low kopje at this point is an easterly con-
tinuation of the one west of M'Phatleles River. Measurements across this kop, which is fairly flat,

* Ann. Rep. of the Slate Geologist for 1S97, Jolifti
t Compare A. L. Hall "On Calcareous Beds in tl:
Trans. G.S.8.A., Vol. XII, 1900,

give an approximate thickness of eighty feet, but the lower boundary is not seen, as on the north side
there follows a narrow valley devoid of outcrops beyond which ia dolomite. On the south side the
upper boundary is fairly sharply denned by a dark greenish ferruginous shaly rock evidently
metamorphosed and crowded with short conspicuous needles of chiastolite. Both this and the under-
lying conglomerates strike conformably with the Timebali Hill Quartzite further south, but the dip
and its direction vary considerably. Nearer to the quartzite just mentioned the dip is from
70 degrees to 90 degrees or slightly overtilted, while nearer to the Dolomite it varies from
60 to 70 degrees to the north. Hence there is a local overturning of the beds, and no doubt
the lowermost horizons of the Pretoria Series have here shared in the general compression and folding
which, as pointed out above, affects the middle horizon of the Transvaal Series east of Chuniespoort
(see Fig. 9). The pebbles are markedly flattened, with rounded edges, and some appear squeezed
out ; all are arranged with their longer axes parallel to the strike of the beds, while the large flat
surfaces he in the plane of the dip. Sometimes the pebbles are six inches long and three inches thick,
but the majority are nearer two by one inch in size. They consist practically entirely of fresh
compact hard vitreous quartzite, and where the beds have been long exposed they form by disintegra-
tion a kind of hard gravel, much in evidence on the Naauwpoort road west of the M'Phatleles River.
The matrix is quite subordinate in amount and due to yellowish green glassy coarser quartz with
a little felspathic or ferruginous material. The most striking feature, however, is the very sharply
defined joint planes, which cut across the bedding planes through both matrix and pebble in such a
way as to produce sometimes two or three very clean facers on one pebble. The rock appears cut
through sharply as by a knife, so that the beds are spilt up into rectangular fragments, one set of
joint planes passing at right angles to the bedding planes, while a second Bet gives rise to vertical
joint faces ; smooth flat surfaces result in this way. This process can be observed east of the
M'Phatleles River, and is probably due to great pressure, consequent upon the intrusion of the Bush-
veld Plutonic Series, but no doubt aided by thermal metamorphism, which has obviously affected
the shaly beds immediately above the conglomerate. The jointing seen in the latter is analogous
to that found in Pretoria Quartzites elsewhere, which are known to occasionally assume a columnar
structure due to the influence of heat. The general appearance of this conglomerate and its position
close to the base of the Series suggests its correlation with Bevets conglomerate of the Lydenburg
Goldfields, the analogy being specially marked by the presence of the same kind of pebble with
occasional striping, and the characteristic rectangular shapes, but the present beds are much thicker
and show evidence of much greater pressure. The latter point cannot be of any great importance,
in view of. the increased efficiency of thermal and dynamic metamorphism, which one knows to have
been active in the Malips River area. The presence of this marked conglomerate horizon suggests
a certain degree of unconformity between the Pretoria Series and the Dolomite. The beds seem to
point to a kind of shingly beach deposit. The thickness of the Pretoria Series and its leading horizons
remain fairly constant, but the Daspoort Quartzite shows a slight increase, while south-east of
Potgietersrust the series as a whole is a little thicker than where it disappears below the Karroo beds
west of the Chunies River. The following table gives an approximate thickness for the major sub-
divisions : —

Magaliesberg Quartzite

Crystalline HornfelB

Dolomitic Marls (altered)

Daspoort Quartzite

Altered Slates

Shales and Flagstones

Timebali Hill Quartzite

Banded Ferruginous Quartzites

Shales, and Shaly Flagstones . .

Conglomerate

Approximate Total 7,680 „

{G) Metamorpkic Rocks of the Dolomite and Pretoria Series.

The phenomena of contact and other metamorphism seen in the present area must be studied
in conjunction with previous discussions and descriptions contained in the Annual Reports for 1906
and 1907. Many of the statements there made, more especially the character and distribution of
the very variable and numerous kinds of altered shales found east of Malips Drift in die southern

100 feet.
2,600 „
600 „
80 to 150

feet.

1,800 feet.

120 to 160

feet.

2,500 feet.

80 „

portion of the Haenertsburg Ooldfields and over eastern Secoecoeniland apply also to the adjoining
country round M'Phatlele's Location and near Potgietersrust (see Annual Report for 1907). At
the same time notable and instructive differences are found in the western area, more especially in
connection with the Dolomite and the depth to which the influence of the great plutonic basin can
be traced, while for the first time strongly altered calcareous beds at a high horizon in the Pretoria
Series are met with.

Naturally the metamorphism of the Dolomite is most pronounced near the Pietersburg railway
line a few miles north of Potgietersrust, where the intrusive norite of the Bushveld Plutonic Complex
breaks across the entire Transvaal System. Thus at the Northern Lime Works on Uitloop the
dolomite can be seen in contact with pyroxenites and other basic phases of the norite magma a short
distance west of the railway line. The altered rocks here seen are generally similar to those exposed
in the quarries at the Lime Works and on the main road connecting the latter with Potgietersrust.
The only resemblance to ordinary dolomite remains in the mode of weathering ; in other respects the
rocks are more or less crystalline lighter coloured varieties, recalling those found in Secoecoeniland
in immediate contact with intrusive sheets. A good example of this kind occurs at the bottom of
a little sluit running east and west on Uitloop and closely resembles a lime-silicate hornfels, traversed
by serpentinouB veins. Elsewhere one finds very fresh bluish holo-crystalline medium or Bomewhat
coarse-grained marble, weathering with an evenly granular grey surface ; this rock can be observed
on the road from Uitloop to Potgietersrust on the east side of the railway line and half a mile south
of the Lime Works. Tnterbedded in it are thin beds of dark coloured fine-grained altered shaly rocks,
closely resembling the more compact varieties of hornfels ; they may be seen a little south of the
Bluit just referred to. Another very wide-spread type is the one from which the white lime is derived
at the Lime Works and which is also very abundant west of the railway. This is a fine-grained whitish
grey rock crowded with minute brown specks, the result of subsequent weathering and standing out
sharply against the white crystals of calcite. When fresh it has a light bluish to pale slate-grey colour.
Thin sections show mainly calcite or dolomite and serpentine. The former, minerals are rather
decomposed and usually without twin lamellae ; probably the bulk of the rock is dolomite. Serpen-
tine is abundant both in irregular larger patches as well as in numerous smaller rounded grains
enclosed in large dolomite crystals, which may show optical continuity over the whole field of view
under a magnification of twenty diameters. The structure is that of a coarsely granular nolo-
crystalline dolomite, and the peculiar scattered occurrence of smaller uniform granular serpentine
enclosed in the large dolomite plates is analogous to the sieve structure in contact metamorphosed
shales. No doubt the numerous minute dark coloured grains seen in the hand specimen, especially
at the quarries of the Lime Works, are due to this form of serpentine. Taking the altered rocks as a
whole, it is clear that their peculiar characters are due to thermal metamorphism consequent upon
the intrusive contact with the norite ; this is expressed petrographically in their more highly
crystalline condition, generally lighter colour approaching that of white marble, in the presence of
interhedded bands of black hornfels, and the abundant occurrence of serpentine, showing that
particular relationship to the carbonates, which is typified by the pronounced sieve structure seen
in the larger crystal plates of dolomite.

Elsewhere, away from the norite, altered dolomites are found only at the topmost horizon of
the series immediately below the base of the Pretoria beds, and here their general characters are
very different, though there is good evidence, discussed below, for regarding the metamorphism
as likewise due to the Bushveld plutonic magma, in spite of the fact that the entire thickness of the
Pretoria Series intervenes. Such rocks are confined to the area north of M'Phatlele's Location, from
the great fault westwards to beyond Chunies River. Specially good localities for their study are
near the common beacon of Boomplaate, Middelkop, and Schildpat, about two miles west-north-west
of the Naauwpoort store. The rocks are fine-grained dark coloured dolomites, which give rise north
of the Chuniespoort-Naauwpoort road to a number of narrow well marked kranzeu. The hand
specimens nearly always still show the original planes of stratification and contain abundant
metamorphic minerals readily visible to the naked eye as small soft disc-shaped, dark greyish green
crystals not exceeding one-eighth of an inch in diameter. They are sometimes found on the bedding
planes, so that in direction at right angles to them the crystals are seen as short delicate needles.
This mineral is probably actinolite and represents the only constituent seen in hand specimens.
Occasionally the mineral occurs scattered sparingly throughout the rock, but is also frequently found
aggregated in certain layers about quarter of an inch thick, which tend to weather less rapidly than
the rest of the rock and stand out in a highly characteristic manner so as to give a kind of ribbed
appearance. Such features are probably the result of a Blight difference in chemical composition of

90

individual layers, owing to which some parte of the rock are more readily affected by thermal
metamorphism ; the effect is analogous to the occurrence of narrow bands of holo-crystalline garnet-
Btaurolite-hornfelB, seen in rocks otherwise much leas altered and not recrystaHised, such as were
described in the previous Annual Report from the Motae River in Secoecoeniland. Thin sections
from the outcrop near Naauwpoort show a finely crystalline calcareous groundmass with large ill-
defined plates probably of actinolite, almost colourless, practically non-pleochroic, and with pale
yellowish brown interference colours. A few slender long colourless needles of wollastonite are
also seen, with a refractive index nearly equal to that of calcite and giving grey interference colours
like those of quartz. The rocks near the south-eastern beacon of Rooiboklaagte resemble those
just described, excepting that the groundmass is more crystalline, the actinolite much smaller and
scattered more regularly throughout the mass, while in thin section a slight pleochroism and typical
hornblende cleavage can still be recognised. The second group of metamorphosed dolomites thus
differ markedly from the first, in the darker and more obviously dolomitic appearance in the hand
specimen and the presence of contact metamorphic actinolite as a macroscopic constituent, while
marble-like varieties are not represented. It is clear from the mode of occurrence and general
character that their alteration was much less intense than that of the rocks in contact with the norite
north of PotgietersruBt. In the area east of Chunies River no intrusive sheets were found in contact
with or near the'altered dolomites.

The Alteration of the quartzites is seen at all three horizons, but for obvious reasons the lowest
one appears least metamorphosed, while the Magaliesberg Quartzite is sometimes entirely re-
crystallised, highly jointed and broken up, both on a small and large scale. Though occupying an
intermediate position with reference to the intensity of metamorphism, the altered Daspoort
Quartzite is the most striking rock of the three, because it is converted for miles along the strike into
a soft extremely beautiful brilliantly lustrous and highly micaceous variety, strongly resembling a
mica-schist in some phases. This particular result is probably due to the original rock having been
slightly shaly, perhaps approaching a greywacke in composition, when it is more easily subject to
mineralogies] transformation than the purer Magafiesberg Quartzite, which shows more intense change,
mainly resulting in recrystallisation without new minerals, and is thus much less -striking petro-
graphically. Typical altered Daspoort Quartzites of the mica-schist variety occur west of Malips Drift
on the farms Goudhoek No. 492, Doornkop No. 463, and on the top of the high ridge which runs
through the farm Katkloof, also close to Chunies River in the eastern corner of the farm De Beer.
Besides alteration in mineralogical composition, the rocks here show pronounced shearing, the
pebbles of conglomeratic bands being drawn out into elongated lenticules. In addition to abundant
sericite or muscovite some rocks on the Katkloof ridge have a pronounced gnarled and knotted
appearance due to large conspicuous broadly elongated andalusite crystals. Where no larger original
constituents, such as pebbles, are present, the effect of pressure produces silvery greyish rocks, the
individual layers of which are thrown into minute delicate folds, while the parallel orientation of the
sericite films imparts to the hand specimens a striking lustre when held in favourable positions. Thin
sections of the more micaceous variety show the following minerals in approximate order of abundance:
muBcovite and sericite, quartzite, chloritised biotite, andalusite, sillimanite, and tourmaline. The softer
and more micaceous the rock the greater the amount of colourless mica, often forming large platy
crystals, while sericitic white mica occurs in crowded aggregates of short delicate needles, sometimes
arranged round the larger quartz grains. These UBually show strain shadows, and in some sections they
are associated with a kind of interlocking relationship with practically no matrix; in Buch cases also
no secondary enlargement of the grains is observable. Andalusite was found in one section only, and in
large more highly refracting colourless plates with pronounced sieve structure. Such rocks occur on
Katkloof on the summit of the Daspoort Quartzite ridge ; in some types the quartz grains become
quite subordinate, considerably more than half the slide being made up of muscovite and sericite.
Tourmaline, noticed in many rocks further east, is also well seen in the present area. It occurs in
two forms, both of small dimensions. In one of these it builds minute deep bluish grains, with the
highly typical trigonal outline, bevelled at the three corners, and showing very feeble pleochroism.
In the other kind, one finds short stout prisms sharply terminated by planes of the fundamental
rhombohedron. These give the deep blue strongly marked pleochroism, with the peculiar direction
of absorption (300 c,). Rocks showing this feature were collected close to Chunies River on the
farm De Beer. The significance of this wide-spread occurrence of tourmaline, though always as an
accessory mineral only, is of great importance in connection with the evidence of far-reaching thermal
metamorphism. South-east of Potgietersrust, the quartzites have also undergone alteration, but
their more normal characters are better shown than in the much more highly metamorphosed rooks

9)

just considered. Thus schistose micaceous varieties were not found in the latter area, no doubt on
account of the fact that here the intensity of the pressure was less, while the rocks seem to be purer.
The most obvious indications of alteration are seen in the Magaliesberg Quartzite, as for instance on the
low isolated kopje carrying the common beacon of Pruizen, OorlogBfontein and Vierentwentig Bivieren,
where the rock shows a pale Sesh coloured matrix, something like that of the so-called speckled bed
in the Lower Witwatersrand Series. Though the matrix may have been originally more felspathic
than usual, the presence of numerous fresh sericite needles and the clear glassy appearance of the
fresh quartz grains points to thermal metamorphism as a contributory cause, resulting in recrystalli-
sation of the matrix. The long and conspicuous kopje further south, on which the common beacon
of Witpoort, Kooipoort, and Grasvlei is situated, shows the Magaliesberg Quartzite to have been
subjected to great pressure, as seen in the general broken appearance of the highly jointed beds, but
owing to the greater purity of the original rock and absence of felspathic matrix, the conditions for
producing micaceous schistose rocks were not present.

The metamorphism of the dolnmitic marls in the Pretoria Series produces a type of rock which
has not been previously described from the Transvaal System. In the more important locality
south-east of PotgieterBrust the altered marls are light greyish-green, medium to fine-grained holo-
crystalline rocks, still showing the original planes of stratification and often strikingly veined by a
pale green serpentinous material parallel to the bedding planes. Such features are well seen in the
small quarries on Oorlogsfontein, where the local building stone is obtained. As a rule no other
mineral apart from dolomite or calcite can be recognised in the hand specimen, but south of the
station small pockets of idiomorphic crystals of vesuvianite are found. Other occurrences of altered
calcareous rocks lie along the southern border of M'Phatlele's Location near the common boundary
of the farms Pramkopjes and Kopjesdam, where bands of lime-silicate hornfels are found interbedded
with coarsely crystalline dark or almost black cordierite-hornfels. Since the latter ib surrounded
on all sides by a peculiar quartzitic variety of the Red Granite it is impossible to assign a definite
horizon to such bands. They are only a few feet thick and, like the marls of Potgietersrust, weather
after the manner of ordinary dolomite, thus forming a strong contrast with the associated altered shales.

Thin sections show the following minerals : — Dolomite, calcite, serpentine, chlorite (pennine),
olivine (forsterite), wollastonite, garnet, apatite, and biotite. Of these, serpentine is the moat common
in the marl, while garnet and wollastonite are characteristic of the southerly occurrence. The
minerals are associated with pavement structure and the larger crystals of dolomite show well marked
sieve structure due to numerous small granular inclusions of serpentine.

The mode of occurrence of the latter is very similar to that described from the altered
dolomites of Uitloop north of Potgietersrust, which are metamorphosed by and in direct contact
with the norite. Micro-chemical tests as described by Lemberg* and Hinden* show the calcite to
occur in subordinate quantities only. Biotite and apatite are sparingly represented. Both in
structure and mineralogical composition these rocks are typical of calcerous beds which have been
exposed to more or less intense thermal metamorphism. In case of the thin bands intercalated in
the hornfels of M'Phatlele's Location this influence was more pronounced, for the rocks are more
granular and have lost much of their bedded character, while the marls near Potgietersrust are finer -
grained and still distinctly bedded. Chemical analysis points to a calcareous rock with a small amount
of alumina. The following results were obtained from two occurrences close to Potgietersrust : —
Analysis of Altered Dolomitic Marls.

I.

II,

sio, ..

. 10-89

27-2.1

CaO ..

22-70

17-26

A1,0, . .

j

FeO ..

. 1300

6-65

Mn,0,

. |

MgO ..

12-74

17-47

CO, ..

. 26-80

21-10

H,0 ..

7-30

4-30

93 43 94 02

I. No. 498 Cj. — Quarry on Oorlogsfontein south-east of Potgietersrust Station.

II. No. 473 c,. — Quarry a short distance east of Potgietersrust Station.

• A. L. H., !,.c, cit., jk .-..

92

Rocks of the above composition can be classed neither with true dolomite nor with limestone.
The amount of alumina, included in the sesquioxides, of which FeO was present in greater quantity,
is a low one, but agrees with that found in some varieties of marls. This fact and the obvious effect
of thermal metamorphism justify the designation altered dolomitic marls. Dr. Molengraaff referred
to them briefly as marls, altered by contact with the Bush veld Plutonic Complex*

The shaies of the Pretoria Series are intensely metamorphosed over considerable areas from
Malips Drift to Potgietersrust ; the three leading types previously established in the Haenerteburg
Goldfield and Secoecoeniland can be again identified and their distribution and relative preponderance
shown to depend upon the presence of the Bushveld Plutonic Complex and the tectonic disturbances
connected with its intrusion, but not without some variation in detail. It is not possible here to do
full justice to the interesting petrography of these remarkable rocks. Further information will be
found in the last Annual Report and in Volume 28 of TschermakB Min. and Petrogr. Mitt.

Under the name Groothoek Type were embraced those highly altered shales and slates, which
have lost all their original bedding in the hand specimens, have become deep black or bluish black in
colour, and the material of which has been completely recrystallised with a resultant holo-crystalline
structure. Their original ahaly nature is only indicated by their behaviour as sedimentary rocks,
when examined in larger outcrops. Apart from this many varieties might be classed as medium-
grained plutonic rocks if seen in hand specimens only. Further detail regarding the significance of
this term can be gathered from the references given above.

Rocks of this holo-crystalline character occur from Malips Drift westwards as far as the
Pietersburg railway, where the Pretoria Series is cut off and intrusively invaded by the norito, while
vertically they range down to the top of the Timeball Hill Quartzite in that portion of the Transvaal
System which lies east of Zebedelas River; south-east of Potgietersrust, as far as their disappearance
below the Bushveld Amygdaloid, the Groothoek type is represented only down to the top of the
Daspoort Quartzite. The vertical range of each type of altered rock cannot be limited by any hard
and fast line, as in the nature of the case there must be gradations. Along one and the same horizon
above the middle quartzite, the petrographical character of a particular bed remains remarkably
uniform, though the presence of gneissic varieties complicates comparisons as applied to certain
localities, e.g. M'Phatlele's Location, a point discussed below in connection with the Malips River
type. The peculiar characters of such crystalline hornfelB are reflected in the scenery, for besides
the ridges of quartzite one finds belts of higher ground between them, showing smoother outlines
and not carved into escarpments and dip-slopes. The great contrast in colour between the former
and the black crystalline slates is very marked in M'Phatlele's Location and south-east of Potgieters-
rust. Moreover, reference to the report on portionB of the Marico and Zeerust Districts contained
in this volume will show many analogies, due to the Groothoek type, as developed in the Western
Transvaal.

The rocks here considered may be seen under very typical conditions along a section taken
from the Daspoort Quartzite on Pruizen westwards towards the Pietersburg railway, where a
practically continuous outcrop occurs, consisting of rocks, the metamorphic nature of which is con-
trolled almost entirely by thermal agencies, not complicated to any extent by pressure. The most
striking observation is the extraordinary thickness of perfectly similar rocks consisting of hard
lustrous compact or fine-grained varieties, rarely containing constituents visible to the naked eye.
Under the hammer they frequently give a characteristic metallic sound. The bedding planes can
often still be determined when larger surfaces are exposed, while in hand specimens they are usually
lost through recrystallisation. Weathering frequently gives rise to large flattened disc-shaped
masses. Sometimes one Ends intercalated a few quite thin more coarsely crystalline bands and,
on splitting the massive rock along these, layers densely crowded with long needles of andalusite
and chiaetolite can be exposed. The little kop, readily identified by a luxuriant mantle of Euphorbia
candelabra, within and close to the eastern gate of M'Phatleles Location in the fence of Molsgat,
illustrates this point. Apart from the invariable presence of minute crowded scales of biotite no
indication is given as to the mineralogical composition. An interesting exception was found in the
railway trenches a few hundred yards south of Potgietersrust Station, where black rocks occur, quite
as coarsely crystalline as some granites, with large biotite flakes and platy crystals of cordierite, up
to half an inch in diameter. No greater contrast could be imagined between such a rock and the
compact soft shale, from which it was derived. To the Groothoek type also belongs a remarkable

" G. A. V. Molengraaff.— Note on some rock specimens exhibited at the meeting of the (foot. S.S.A., Feb.. 1SHI5.
Trans. G.S.8.A.. toI. VIII, p. 66.

93

outcrop, several square miles in extent, oE black rocks near the southern boundary of M'Phatlele's
Location, which have a good dip and markedly bedded appearance, but lie quite detached from the
main area of the Pretoria Series further north, as they are surrounded by the Red or Bushveld
Granite, here represented by an unusually acid and quartzitic white phase.

In thin section rocks of the most highly altered type show the following minerals: — Biotite,
cordierite, quartz, felspar, andalusite, chiastolite, siUimanite, garnet, and tourmaline, with amphibole
and pyroxene under certain conditions.

The structure is holo-crystalline, evenly granular, and moderately coarse-grained, the former
predominating. Sometimes one or two constituents may build very large crystals and then are
often riddled with inclusions so as to produce sieve structure. Tn the absence of such pseudo-
porphyritic appearances, the whole field may be perfectly evenly granular like that of an aplite,
for example, when pavement structure becomes evident.

Such varieties might, apart from other evidence, suggest a finer and even grained slightly
basic granitic rock. The most widely spread mineral, present in all sections examined is biotite in
smaller lath-shaped or slightly rounded deep brownish or greenish crystals with intense pleochroism,
and frequently crowded with inclusions. In rocks which have been subjected to recrystallisation
during pressure the mica forms more irregular and lenticular wavy masses, feathering out at their
extremities into needles or whisps, and aggregates of very delicate fibres of siUimanite.

Another form of orientation occurs, when short lathed-Bhaped biotite crystals arrange them-
selves round and enclose large cordierites, so as to produce a kind of mesh structure. Next to dark
mica, quartz is the most common constituent, always very fresh and granular, often forming an even
mosaic in which some grains are biaxial and probably represent felspar. The mosaic is no doubt
due to the complete recrystallisation of the shaly matter in the unaltered rock. Plagioclase with
distinct twin lamellae was found very sparingly in one section from a rock from the south-eastern
corner of M'Phatlele's Location ; sometimes larger quartz grainB assume a faintly striated appearance,
which under crossed nicols produces characters somewhat like those of a plagioclase. This would
appear due to pressure. In the previous Annual Report reasons are given for regarding the plagio-
clase as albite. Cordierite is extremely common and after biotite forms the most typical contact
mineral. When the influence has been less intense cordierite forms large rounded plates only
identifiable under crossed nicols. Elsewhere it occurs in smaller granular masses resembling the
quarts mosaic. When the habit is more elongated straight extinction is observed ; a further
characteristic are the very numerous inclusions, while the biaxial character serves to distinguish it
from quartz for which it may be easily mistaken, since it is invariably colourless and does not show
the peculiar yellow halos which characterise it in other rocks. Sector twinning is poorly shown.

Other minerals pointing to thermal metamorphism are andalusite and chiastolite : these are
found only occasionally and seem confined to horizons above the Daspoort Quartzite east of Zebedelas
River. Typical rocks showing macroscopic chiastolite in needles up to half an inch in length occur
near the foot of the black isolated kop close to the Molsgat fence, leading into M'Phatlele's Location.
Garnet is fairly wide-spread, but always in small quantities and especially abundant in horizons
just below the Daspoort Quartzite west of Malips Drift. An important very widely spread contact
mineral is tourmaline, occurring in almost every section of altered shale right along the strike and
through the entire thickness. It is found only sparingly and always in minute deep blue grains or
short prisms, which in the presence of abundant deeply coloured biotite may be easily overlooked.
Its markedly isomorphic habit is like that described above in connection with altered quartzites
and consequently its recognition presents no particular difficulties. The great lateral and vertical
distribution of this mineral is important, as it is pre-eminently of purely thermal origin, whereas
some of the others are known also under conditions where the presence of a large igneous intrusion
cannot be demonstrated.

A variation from the average Groothoek type is found in a very peculiar collection of rocks
met with only here and there, but' interbedded' with the former and having the same general appear-
ance with a more compact habit. The colour is lighter and approaches whitish grey, while banding
is well shown on the weathered surface, but differs from that of the common hornfels in traversing
the rock and not being confined merely to the exposed surface. Such abnormal kinds occur on
Katkloof and in the isolated patch of gneissic hornfels found within the Red Granite zone at the
southern boundary of M'Phatlele's Location. No minerals can be recognised in the hand specimens.
Thin sections show amphibole and pyroxene in addition to cordierite, associated with a very fine-
grained'recrystallised matrix consisting of a quartz mosaic. The banding is reflected in the layers
of this mosaic alternating with others rich in pale greenish or nearly colourless hornblende, either

94

forming delicate radial aggregates of needles, or else deeper coloured sections with a typical lozenge
shape, showing common double cleavage traces of hornblende. The above features recall similar
abnormal and banded varieties described in last year's report, and found both in Secoecoeniland
and in the Rustenburg District ; in the latter area they can be clearly seen to owe their peculiar
composition to transference of material from igneous intrusions and this probably has also taken
place in the present cases. Such rocks may be classed as injected hornfels.

A second variety of altered shales and slates has been referred to as the Longsight type,
characterised by a much less intense degree of metamorphism and easily recognisable in the hand
specimen as an original shaly rock ; macroscopically visible contact minerals are always present
and the planes of stratification are not obliterated, while the colour is that of the original rock. To
this group belong andalusite and chiastolite slate, also dark coloured ferruginous actinolite slates,
confined to horizons below the Daspoort Quartzite south-east of Potgietersrust and below the
Timeball Hill Quartzite east of Zebedelas River. But throughout the present area the Longsight
type is far less common than east of the Malips River, mainly for two reasons. In the first place
the zone of hornfels down to the lowermost quartzite is succeeded in the Chuniespoort district by
banded ferruginous quartzites, the composition of which does not lend itself readily to retain thermal
influences with an altered mineralogies! composition, so that the extraordinarily variable series of
staurolite and other slates found in Secoecoeniland is not to be expected east of Malips Drift. In
the second place, although south-oast of Potgietersrust the beds below the Timeball Hill Quartzite
are typical shales the metamorphism did not extend so far down into the Pretoria Series here as further
east, and the belt of hornfels above the Daspoort Quartzite is separated from unaltered shales by
only a thin horizon of less altered rocks belonging to the Longsight type. It seems that the meta-
morphism rapidly died out below the zone of the crystalline rocks ; probably the interposition of
the middle quartzite may have prevented thermal effects from reaching further down, the more so
as tectonic disturbances, which would tend to provide greater facility for the passage of mineralizing
solutions, are less marked south-east of Potgietersrust than in the Haenertsburg Goldfields.

Rocks belonging to the second group show andalusite and chiastolite as the common contact
minerals, while actinolite is confined to one horizon well below the Timeball Hill Quartzite, a few
miles west of Malips Drift, consisting of compact dark coloured ferruginous shales. The almost
complete absence of staurolite is remarkable, but goes to support the contention that it is more
characteristic of those rocks, which have been altered while under greater pressure, as will be pointed
out below.

Besides the above two groups one can recognise a third variety, the Malips River type, which
combines the structure and composition of the Groothoek type with a more or leas pronounced
schistose or gneissic habit. These are abundantly developed east of Malips Drift, but extend in a
westerly direction only as far as the Chunies River. The best outcrops are the few low scattered
and isolated black kopjes seen in the northern portion of M'Phatlele's Location, about three-quarters
of a mile south of M'Phatleles River between the Daspoort and Magaliesberg Quartzite. The rocks
are medium or coarse-grained cordierite gneisses, first mentioned by Mr. Kynaston* who has given
an account of their microscopic character. Their stratigraphical position and metamorphic history
were subsequently studied by the writer, and it was shown that this rock as well as others with pro-
nounced gneissic and schistose habits are due to thermal metamorphism under great pressure. It
was pointed out that the unaltered shales of this horizon seen at Lydenburg can be traced continuously
as far as M'Phatlele's Location with a gradually increasing degree of alteration, until they finally
pass into biotite-silhmanite-cordierite-gneiBses.'t' Besides argillaceous rocks the quartzites of the
Daspoort horizon must be classed with the Malips River type, since north of the junction between
the Olifants and M'Phatleles Rivera they are converted into micaceous schists, of which a description
was given above. The restricted distribution of the third type is intimately connected with tectonic
disturbances, which were most pronounced in the southern part of the Haenertsburg Goldfields, with
which the Pretoria Series of M'Phatlele's Location is geologically connected.

Regarding the agents of metamorphism and the distance to which their influence can be recognised,
certain general observations of great interest can be deduced, now that the Transvaal System has
been mapped from the eastern railway through Lydenburg and Chuniespoort to Potgietersrust.
It was shown in the last two reports, that near Lydenburg and in Secoecoeniland, some of the altered
rocks, specially the Longsight type, owe their metamorphism less to the main body of the Bushveld

■ H. Kynaston. — On certain rocks associated with the Norite ami (Irnnitc of the Central Transvaal. Trans. G.S.S.A.,
I90B, vr,l. Tin. p. B9.

t See Trans. Qeol. Soc. S.A.. ml. XI. 1908, pp. IB. Hi.

95

Complex than to numerous intrusive sheets, some of which attain a considerable thickness. The
further one follows the Transvaal System towards the north-west the leas evident do these intrusions
become, until in the southern portion of the Haenertsburg Goldnelds very probably, and undoubtedly
over M'Phatlele's Location and south-east of Potgietersrust, the metamorphiBm must be almost
solely assigned to the main body of the Bushveld Complex ; thus in the last-named areas the entire
Pretoria Series shows only one or two quite insignificant sheets, and it is out of the question to allow
these more than a comparatively small share in producing the final result. Hence one may
say, that the whole of the metamorphiBm west of the Malips River must be attributed to the far
more effective central basin of plutonic rocks. The local restriction of the gneissic and schistose
rocks is significant, for these only occur close to and on both sides of the Malips River, where, from an
entirely independent line of argument, it can be shown that powerful tectonic disturbances must
have taken place. Reasons are given in the last Annual Report why such rocks cannot be regarded
as due to regional metamorphiBm of later date. Lastly, interesting observations can be made regard-
ing the width of the contact belt due to the Bushveld Complex. The latter can be more accurately
estimated by disregarding the dip and measuring the width of the aureole along the direction of
thickness. Near Lydenburg the contact rocks due to the central basin only extend a short distance
below the Magaliesberg Quartzite, while north of the Steelpoort River they reach to the top of the
Daspoort Quartzite ; along the Malips River the aureole extends to the base of the Timeball Hill
Quartzite, but east of Chuniespoort it just touches the top of the Dolomite ; lastly, south-east of
Potgietersrust the shales underlying the middle quartzite mark the outer limit of the contact belt.
Hence between Lydenburg and Potgietersrust the width of the aureole varies within wide limits.
This is probably due to a combination of several causes. Shales are well known to vary considerably
in their susceptibility to react mineralogically or structurally under the influence of thermal meta-
morphism. It is frequently noticeable both in Secoecoeniland and in the present area, that more
sandy horizons within argillaceous beds are far less altered, and conversely, intensely altered hornfels
may occur interbedded with much less altered slates. Furthermore the contact belt is thicker in
proportion to the intensity of tectonic disturbance, and east of Naauwpoort, where the Transvaal
System has been subjected to very great pressure, the metamorphism extends further down than
anywhere else in the Eastern Transvaal. If a broad view of the central basin is taken and the meta-
morphic phenomena considered as a whole and followed round the basin westwards to Zeerust it
will be found that the variations in the width of the aureole is in reality not great, when compared
with the enormous area covered by rocks belonging to the Bushveld Complex. The differences are
of an order which could be readily accounted for by local changes in chemical composition and by the
presence or absence of tectonic disturbances*

The following table shows the distribution of the leading types of metamorphic rocks and
enables the variation of individual horizons, when followed towards the north-west, to be compared :—

Lead i no Horizon.

I.YDEKBURG.

—

Haenebtsbitrg

GoI.DFIELDS AND

MThatlelb's
Location.

south-bast of

rOTftlETERSBUST.

Above tlie MajCTliesU'rE
Qnartxite

--

t'Vvstullinc hornfels
((iroothoek type).
Due to Bush. I'omp.

Trv-rallinc hornfels
(Oroothoek type).
Due to Bush. C'ump.

Mapilieaberg Quartzite...

Altered by Bush. Com.

Altered with >eridtk:

Highly altered

Altered and with treah

Slate* anil Shales

Mainly chiastoliti; si.
(Longsight type).
Due to intr. sheets.
Thin bands of Cor-
el ierite hornfels

Crystalline hornfels . Mien and sorii-ile sell,
{(troothoek type). (Malips River type)
Due Hilelv to Hush.
Comp.

Crystalline hornfels
(Oroothoek type).
Altered rlolomitir

' See Plate III in vol. 28 of Tichcrra. Min. Petr. Mil., Viel

LRAD1NO HobizON,

Lydenbero.
Nut altered

Seco«iokmilasd.

Haenertsbubo

(lOLDFIELDS AND

M'Phatlele'b
Location.

South-east op
potg i bt e b8bd st.

Dnapoort QuarUite

Altered anil slightly
sheared

Quartj; seta, (Mnlipn
River type)

Slightly altered.

Slates and Shales

Not altered, except Lungiripht type. i.e.
locally at contact antlalusite and Htaiir-
with intrusive sheets olile slates tine to

Bush. Pomp, ant! in-

MalijM River and
Long, types. Mica,
sericite schists:
stanr. and aridalosile

Slightly altered ; oc-
casional Long. type.

Timeball Hill Qnartsite

Not altered Slightly alt ent I

Slightly altereii

Not altereii.

Slates and Shales

Sot alt e ret 1, except
locally at contact
with intrusive sheets

Sand v sh ftl es 1 1 r ban . 1 et 1

altered locally into

el j last, slate

Banded ferr. quart site
with thin horizons of

Very slightly or not

Dolomite

Not altereii Sot altered

Slightly altered . Sot altered, except
(actitinlite) north of Fotgietersl
! rust due toabnorraa-
local conditious.

i

(H) Waterberg System.

Rocks belonging to this system were found in a single locality only, on the farm Fruizen some
eight miles south -south -east of Potgietersrust. The main road leading from the latter town to
Zebedela's Location crosses the outcrop just before reaching the nek, over which the road leads down
to the Springbok Flats. The Waterberg rocks form a low kopje lying in a well defined valley between
the Timebafl Hill and Daspoort Quartzite. On the west side of the road the latter forms a high
prominent ridge, and a little below the summit is found the ferruginous conglomerate which occurs
in this quartzite. Blocks of it mingle as detritus with debris from the Waterberg Quartzite kop,
and may thus be mistaken for a Waterberg Conglomerate. The major portion of the outcrop lies
east of the road and has a good dip of about 25 degrees to the north, thus making a marked
unconformity with the westerly dipping Pretoria beds. True conglomerates belonging properly to
the former were not found, the whole of the kop consisting of pale brownish purple quartzites with
characteristically Waterberg appearance. The position of this outlier in the floor of a deep valley
is peculiar and it is difficult to realise under what conditions the material was deposited there. No
doubt the old surface of the Pretoria Series must have been very uneven.

On the south Bide and near the foot of this kop occur dark coloured sandy or shaly flagstones
resembbng the argillaceous beds of the Waterberg System, but their undoubted conformity with the
Pretoria beds shows them to belong to the latter.

(/) Karroo System.

This division is represented by the Bushveld Sandstone and the Bushveld Amygdaloid. The
former is confined to a small patch lying within Zebedela's Location round Slypsteen Drift and form-
ing the extreme north-easterly end of an extensive tract of sandstone, which is seen at various
localities on the Springbok Flats (see Annual Report G. S. for 1904). The best exposure
is in the bed of Zebedelas River at the drift and a short distance down stream. The rocks are
fine-grained yellow or cream coloured soft sandstones arranged more or less horizontally. In the
river a thickness of about twenty feet is exposed carved by the stream into an interesting succession
of falls and cataracts with pot-holes and other peculiar structures due to corrasion. Away from the
river towards the east, where the rocks are found in a few scattered and small outcrops, they are

97

more highly weathered and pass into brick-red varieties with a thin film of glistening siliceous material
below the black weathered crust, due to secondary surface silicification *

The true thickness of these sandstones cannot be ascertained in Zebedela's Location and their
ureal extent is extremely uncertain owing to the abundant presence of sandy flats devoid of outcrops.
From north to south along the right bank of the river a fairly sharp change is noticeable in the nature
of the surface material, which passes from bright red or greyish coarser sand with occasional white
irregular lumps of secondary surface limestone into pale greyish brown finer sand, denoting the
presence of the Slypsteen Drift sandstone. The later is known as the Bushveld Sandstone and
correlated with the Upper Karroo or Stormberg Series.f

(J) Rock* intrusive in the Transwal System.

At several horizons both in the Dolomite and the Pretoria Series, igneous bodies are found
as sheets intrusive into the sedimentary rocks. They usually persist for a considerable distance
along the strike and may give rise to a more or less well marked feature. Compared with the neigh-
bourhood of the capital, their number and thickness is much reduced east of Zebedelas River. Only
two intrusions were found in the Pretoria Series of M'Phatlele's Location, while the Dolomite near
Chuniespoort has two or three, though none of any great importance. In the lower horizons of the
upper series south-east of Potgieterarust, there is one very persistent sheet high up on the eastern
slope of the great semi-circular range formed by the Timeball Hill Quartzite.

Such rocks are usually fine to medium-grained dark greenish blue varieties, which in their
mode of occurrence and forms of weathering closely resemble analogous intrusions, found wherever
the upper portions of the Transvaal System occur. Mineralogically the sheets in the Pretoria Series
are characterised by the abundance of hornblende, associated with highly decomposed plagioclase,
in one place recognisable as oligoclase. The ferro-magnesian mineral forms either long delicate needles
and almost colourless fibrous aggregates, or it is present in large fresh idiomorphic crystals, with
atrongly marked pleochroism sometimes approaching that of glaucophane. One section shows
hornblende in highly idiomorphic lozenge-shaped crystals with prismatic cleavage. These minerals
may lie in a colourless fresh groundmass composed of granophyric material. The rocks may be
classed as diorites. Another intrusion occurring near the base of the Dolomite, a little south-west
of Chuniespoort, belongs to a soft pale dirty greenish highly decomposed rock, which, besides the
above-mentioned minerals, shows much chloritised hornblende as well as a fair amount of ilmenite
with typically trigonal and hexagonal outlines. Lying at a slightly higher horizon in the Dolomite
is found a second sheet closely associated with quartz reefs : this will be further considered in connec-
tion with the Economic Geology.

(K) Rocks belonging to the main body of the Bushveld Plutonic Complex.

The southern portion of the area dealt with in this report from the junction of the Olifante
with Zebedelas Rivers north-eastwards iB made up of basic and acid rocks of the Bushveld Complex,
which may be arranged under the Red Granite and norite groups.

Rocks belonging to the Red or Bushveld Granite occur on both sides of the Olifante River
as far north as its junction with Chunies River. The fairly level and gentle undulating country,
relieved by isolated kopjes carrying large rounded blocks of granite, recalls the familiar features of
this formation as seen/ in many places throughout the Central Transvaal, and the petrographical
characters, both in hand specimens and in tain sections, agree with those of typical coarser-grained
Bushveld Granite described from areas further west.

Attention must be called to a remarkable variation first met with at Adriaanskop and described
by Mr. Kynaston.J The rock, of which the whole of this kop is composed is a light coloured white
and highly acid variety, closely resembbng a quartzite in the hand specimen. Recent mapping
north of this point has shown this rock to possess a much greater distribution, for it gives rise to two
well defined ranges or belts of high ground, which rise locally into sharp ridges and prominent peaks.
The northerly range begins at the eastern fence of Zebedela's Location, with the low kopje, on which
the southerly beacon of Kalkbult is situated ; from here it runs due east to the northern corner of
Pramkopjes, where it terminates in a high and conspicuous pyramidal kop. The southerly range
runs through the southern boundaries of the farms Grootkop, Kameelbult, Pramkopjes, Kopjesdam,
Stoffpoort, and Zaaikloof ; at the southern corner of the last-named locality it turns sharply

' Passarge. Die Kalahari, Berlin, 1904, cu. XXXV.

t H. Kjnanton.— Note on the correlation of the Bushveld Sandstone Series, etc. Trans. G.8.8.A., Vol. X, pp. 31-33.

I Ann. Rep. for 1904, p. 20.

9S

northwards and continues in that direction roughly along the common boundary between Spelonken
and Zaaikloof nearly as far as Chuniea River. Adriaanskop forma approximately the south-western end
of this range. Throughout these various hills the same quartzose variety is found indistinguishable
from that of Adriaanskop. The resemblance to a highly siliceous sedimentary rock is so strong that
the final recognition of its true nature was not possible, until a large number of outcrops bad been
examined, both in the field as well as in thin sections ; thus a series can be found passing from rocks
almost identical with quartzite through all possible gradations into varieties rich in flesh coloured
orthoclase with clearly defined cleavage faces. In the highly acid kinds, such as occur on Adriaanskop
and on the prominent peak in the northern corner of Pramkopjea, the rocks are light grey or white
and consist of fresh evenly granular quartz grains with small scattered specks of muscovite and
eericite ; when the rocks are more highly weathered the exposed surfaces acquire a kind of speckled
appearance due to the mica becoming reddish, and where one finds also a Urge development of intensely
altered shales, for instance, close to Pramkopjea, the quartzoee rocks might well be mistaken for
metamorphosed Pretoria Quartzite. There is nothing, however, in the geological structure to suggest
the presence of faulted Pretoria Series. Some varieties on Pramkopjea pass into highly micaceous
rocks, composed solely of quartz and muscovite and resembling a greisen. The fresh glassy trans-
parent quartz may give way locally to a variety not unlike vein-quartz. The most constant and
characteristic feature, however, is the presence of the irregular specks of white mica. Near the main
road passing through the eastern corner of Turfiaagte, and in the same zone of kopjes which contains
the speckled quartzose varieties referred to, are found phases more like granitic rocks, since the
quartz grains are surrounded by much highly telepathic matrix. Close to the latter type are found
peculiar milky white fine-grained varieties with an even-grained texture, owing to tike abundance
of white felspar. The latter kind is highly suggestive of a plutonic igneous rock. On the eastern
face of the high ridge running parallel to the western boundary of Spelonken occurs an interesting
outcrop of very large rounded blocks, in shape closely resembling the spheroidal forms of weathering
seen in diabases, etc., and these consist of the same quartzose variation, but the rocks are distinctly
striated due to a kind of flow structure ; this banding is especially well shown on the black weathered
surface, but goes through the rock ; it is suggestive of primary flow structure occurring during or
previous to consolidation. Thin sections of this rock show no signs of mechanical deformation.

Examined with regard to general field relationships, these peculiar rocks never show any
bedding planes. Although in the last-named range one finds larger outcrops with well defined
divisional planes dipping towards the west, these are never continuous along the supposed strike,
while further south in the western corner of Olifants Poort structures resembling ripple-marking are
found on closer examination to indicate subsequent movements along joint planes. Finally west
of the main road, which passes Lawrence's store in Secoecoeniland occur types of this acid kind
intimately associated with fine-grained red granitic bands.

Thin sections cut from a series of these rocks show abundant quartz in interlocking grains,
with small patches of sericitic mica, nearly always present, while several show fairly fresh orthoclase
and microcline. In one or two cases ferro-magnesian minerals are found in small quantities. The
milky white rock from Turfiaagte can certainly be regarded as of undoubted igneous origin.

It is therefore concluded that these highly quartzose rocks are an acid phase of the Bushveld
Granite, analogous perhaps to a greisen, which best agrees with their mode of occurrence and petro-
graphies! characters.

Between the Red Granite and the Pretoria Series lies a belt of norite and allied basic rocks.
West of the Olifants River the zone is only a few miles wide; but towards the east and south-east-
it broadens considerably, until the whole of the Lulu Range is included, so that a strip of country
some fourteen miles wide is made up entirely of norite, pyroxenite, and serpentine, with occasional
bands of magnetite and pegmatite. Near Potgietersrust the norite occurs both north and south of
the station and on both sides of the railway line. Its relationship to the underlying Pretoria Series
is quite different here, for along the railway from south to north it comes to He successively against
lower horizons of the Pretoria Series, indicating an intrusive transgression, whereas along the
Olifants River its distribution agrees roughly with the strike of the Hagaliesberg Quartzite, and it
is found in contact only with that horizon or else with hornfels overlying the latter, as on Diepaloot
five miles east of Naauwpoort. The norite gives rise to highly characteristic country with numerous
rugged black kopjes, especially west of the Lulu Range in Secoecoeni's Location ; some of the kopjes
are approximately pyramidal, when the close similarity to the norite belt of the Pretoria and Rusten-
burg Districts is even more pronounced.

Petrographically the norite remains very uniform over wide aress and little need be added
to the descriptions previously given. Its features east of the Lulu Range were discussed in last
year's report and apply equally well to the present area. Rocks composed almost entirely of
pyroxene, in particular enstatite or hypersthene, and showing a deep olive-green colour with coarse
texture, occur along both banks of the Olifants River near its junction with the Malips River and
near Uitloop north of Potgietersrust. One is at once struck in the former area with the regular dis-
tribution of norite rocks along certain, straight and low ranges of hills, which show dip-slopes and
escarpments, when viewed from a little way off. Several of these occur as markedly linear ridges
in the valley of the Oliphants River and are very probably due to rocks of varying composition
differentiated from the norite magma. A similar tendency, but to a lesser degree, can be recognised
in Secoecoeniland over the hilly ground close to and on the north-west side of the main road leading
past Lawrence's store to Schoonoord. All this portion of Secoecoeniland is most typical norite country.
Thus the whole of the farm Malikskraal shows only norite, magnetite, and pegmatite veins. The
rocks here found have been classed by Mr. Jorissen with the Swaziland Series.* Mr. Jorissen admits
the absence of Older Granite but refers to the presence of this formation " about eleven miles
east-south-east on the farm Clapham No. 626." This locality, however, belongs to equally character-
istic norite country, and no Older Granite was found there ; it is further stated that " the schists
are traversed by veins of pegmatite, which stand in intimate relation to the granite." Pegmatite
certainly occurs on Malikskraal, but no connection was found between them and the supposed Older
Granite of Clapham ; it is far more reasonable to regard these veins as related to the Buahveld
Granite, which forms a range of hills only about three miles from Malikskraal on the south-west side
of the main road. The rocks are closely similar both in their mode of occurrence as veins in typical
norite, as well as in their detailed characters, namely, granophyric structure and association with
copper ores, to those described in the Annual Report for ISMS from the Rustenburg District along
the Sterkstroom near the Bethanie Mission Station.

In many localities the norite assumes a banded structure, which when associated with medium
or coarse-grained textures produces rocks conveniently classed as gneissic norite ; sometimes the
banding is very pronounced and may in connection with finer-grained varieties give rise to schistose
norite. At the common boundary of Kaurnek and Zondernam a short distance north of M'Phatlele's
Location, or on Grootdraai, where the main road to Lawrence's store crosses the river, also in several
localities in Secoecoeniland including Malikskraal, every gradation can be observed from massive
coarse-grained typical norite without any signs of parallel structures to thoroughly gneissic or even
more finely schistose varieties. In describing a series of rock specimens collected by Mr. Leslie,
Professor Molengraafif drew attentionf to such banding and stated " very interesting are the
schistose varieties of amphibole norites, often much resembling amphibolites. They give evidence
of a schistosity at the bottom of the intrusive mass (laccolite) due to pressure." The banding is
expressed in the hand specimen by the orientation of the plagioclase and ferro-magnesian con-
stituents. In the coarser gneissic varieties this is not well seen in thin section, but becomes marked
in the schistose kinds. The twin lamellae of the plagioclase are sometimes gently bent and the
crystals may also show strain shadows, thus clearly proving the effects of pressure. Such features
are not associated, however, with cataclastic structures. The banding is no doubt of primary origin
and took place before the final consolidation of the magma. It is moreover extremely wide-spread
and has been noticed in the zone of norite north of Pretoria and westwards as far as the Great Marico
River. Naturally where the Transvaal System was during the intrusion of the Bushveld Complex
subjected to great pressure, there one would expect such banding to become more pronounced, which
accounts for the prevalence of foliated norite along the Olifants River and in Secoecoeniland.
Viewed in this light the occurrence of belts of norite of particular composition arranged in straight
lines parallel to the strike of the Pretoria Series may be regarded as a kind of banded structure on a
very large scale. If it is due to pressure, it should become less marked further from the margin of
the Bushveld magma. This is borne out in the field ; foliation is very rarely seen in the Red Granite,
one instance was given above ; in the main body of the granite no directional structure of this kind
is seen ; finally the widespread occurrence of more or less highly banded norite throughout the Central
Transvaal is readily understood in view of the enormous body of rock intruded under similar conditions
over very large areas.

100

(L) The Buskveld Amygdaloid.

Although a considerable area of the Springbok Plata both east and west of Zebedelas River
k underlain by the Bushveld Amygdaloid, the rock is only seen in very few and small outcrops
separated by miles of level sandy bush country, in which only blocks of white secondary surface
limestone are met with. The best exposures occur on the farm Doornfontein not far from the little
spruit draining southwards through the centre, and on the adjoining farm Groothoek in the westerly
drift, through which the main road to Potgietersrust passes. One or two additional outcrops occur
in Zehedela'a Location. The occurrence on Groothoek shows fine-grained dark coloured and highly
decomposed rock faintly speckled white, but without obvious amygdaloidal structure ; the outcrops
on the southern portion of Doornfontein are compact deep purplish red varieties with small irregular
blebs or amygdules of a white mineral resembling a zeolite. Behind Flatteau's store close to the
chiefs kraal in Zebedela'a Location, the rock brought up from the well is a very highly decomposed
variety permeated in many directions by whitish veins forming a continuous network and due
mainly to secondary carbonate of lime. This form of weathering gives the rock a characteristic
platy structure. The relationship between the amygdaloid and the Slypsteen Drift sandstone is
not seen.

(M) Superficial Deposits.

Apart from surface material due entirely or mainly to mechanical disintegration, such as the
extensive tracts of pale sand north and east of the Strydpoort Range derived from the weathering
of the Older Granite, distinctively superficial deposits are especially noticeable over the Springbok
Flats south of Strydpoort. The scarcity of running water, the predominance of low bush country
over extensive almost level areas, combined with a relatively low rainfall, result in physical and
climatic conditions not unlike those of semi-desert regions. Here therefore superficial deposits are
very common and. wide-spread, the most frequent being secondary surface limestone. While this
rock is certainly characteristic of those areas, which are underlain by the Bushveld Amygdaloid,
it is also found overlying the Bushveld Granite and the norite. The former association can be seen
on the farm Kopje Enkel a few miles north-west of Adriaanskop, while surface limestone above norite
occurs on Doornvlei, a little west of Chunies River. Such deposits usually form a kind of thin
incrustation occurring in scattered and irregular patches from a few square feet down to small lumps ;
continuous sheets are very rarely seen and the thickness is inconsiderable. In the Zebedelas River
at the main drift close to the south-eastern fence of Schaapplaate the deposit is about twenty feet
thick. A few miles north-east of Slypsteen Drift occurs a peculiar sheet of water filling a roughly
oval depression, known to the natives as Matlapoodie. The edge of this pan is marked by thin walls
of secondary limestone only a few feet thick. These rocks consist of pale greyish white fine-grained
or compact cellular limestone traversed by delicate streaks of some indefinite blackish material. A
slight development of nodular or concretionary structure can also be recognised.

Other surface deposits are the familiar earthy brown incrustations somewhat resembling
laterite, and several varieties of sand. The occurrence of the latter on formations not particularly
arenaceous such as the Bushveld Amygdaloid is no doubt partly due to the prolonged and selective
effect of wind in blowing away the lighter particles and thus increasing the originally small propor-
tion of sandy grains, and partly also to the drifting of sand from adjacent areas of sandstone or
quart rite.

Many rocks show the familiar surface change due to secondary infiltration of silica. Thus
a dull soft sandy rock is often found protected by a delicate veneer of lustrous fresh and glassy silica,
probably precipitated in the colloidal form, in which crystalline quartz would pass into solution under
the solvent influence of alkaline waters. Such results are similar in character to the gradual hardening
and cementation of roads, which have fallen into disuse for long periods.

5. Economic Geology.

The present area includes the Marabastad Goldficlds, a term applied to the schist belt a few
miles south of Marabastad, but also usually including other workings for alluvial and reef gold further
south on Waterval, Vrischgewaagd, and Eersteling. At the latter locality gold was found as far back
as the late sixties, and probably the first gold produced in the Colony was obtained from Eersteling.

At the present time little work is being done in the more southerly portions of these gold-
fields, but attention is being directed to the possibilities along Mount Mare, more particularly on
Snymansdrift and on the adjoining mynpacht of Zandrivier. The principal works on the former
locality are: — (1) No. 3 hill, near the northern foot of which stands the battery; (2) No. 4 bill a little

ioi

east of the former and separated from it by a email poort ; (3) No. 7 hill to the north of the two previous
workings and identical with the principal ridge formed by the lowermost ironstone. The earlier
workings are those of the last-named point. There are found at least three or four quarts reefs in
the schists, both interbedded and cutting across the formation. In the lower ironstone range
(No. 7 hill) a reef occura apparently interbedded at the base of the lowermost ironstone on the north
side of the range and underlain by very hard greyish green quartritic slates ; its average width is
near fourteen inches and its value about 8 dwta. In Nos. 3 and 4 hills the horizon is a little above
the middle ironstone, and above the horizon of the ottrelite and andalusite schists, to the left of the
point A in the section on Plate XIII. The foot-wall of the reef in No. 3 bill is known locally as
" Knotted Schist," but does not represent a metamorphic rock, such as is usually implied by this
term, but is a kind of greenish sheared conglomerate, not unlike that found close to Brown's adit
near the eastern end of Mount Mare, with which rock the foot-wall of No. 3 reef may roughly be
correlated. The hanging wall is a pale green more compact, sheared sericitic quartzite with meta-
morphic tourmaline. ThiB reef is interbedded and averages nine inches in width. In No. 4 hill
are found two distinct quartz reefs, one of which is interbedded with schists, also resting on sheared
conglomerates and is probably identical with that of No. 3 or Battery Hill. The other cuts across
the schistosity in a north-easterly direction, is much twisted about and may even he interbedded
for short distances. A little above the horizon of the ottrelite-schist on the Zandrivier mynpacht
(Mount Mare proper) occurs an old shaft some seventy feet deep, at the bottom of which is a drive
about 200 feet long, cutting across the strike towards the north. It exposes four interbedded reefs,
of which the most southerly near the shaft averages fourteen inches in width. At the time of the
writer's visit (July, 1908), the most northerly reef was at the face of the drive and about three and a
half feet wide. The country rock is composed of compact quartzitic schists dipping 45 degrees
to the south.

Between the lithological characters of the various reefs there is a close general resemblance,
as they consist almost entirely of fine-grained rather hungry looking whitish dull quartz, while
occasionally the foot-wall at the contact with the reef may become highly charged with pyrites.
The average yield, taking all sources into consideration, is somewhere near 9 dwta. per ton milled.
No definite conclusion can be arrived at regarding the mode of origin of these reefs, but it seems clear
that they are later than the country rock, and probably the disturbances to which the beds were
subjected gave rise to cracks and fissures, subsequently filled with vein quartz. The western portion
of Mount Mare is moreover traversed by several smaller dykes and faults, running out through the
little poorts and crossing the main ridge formed by the lower ironstone bands. Probably such dykes
are connected with the mineralisation of the quartz reefs, for one finds, as for instance on the top of
No. 7 hill the " calico-rock " carrying gold only close to its contact with the dyke.

Elsewhere, numerous gold-bearing horizons of lesser importance are found along or near the
Mount Mare range. Thus near the eastern corner of Hollands Drift occurs an interbedded quartz
reef about eight inches thick at the base of the uppermost ironstone band and underlain by red slates.
Near the eastern termination of Mount Mare on Zandrivier are numerous old workings now abandoned.
Brown's reef, situated some 200 feet above the middle ironstone band, yielded a fair amount of gold
some years ago and even in the sheared conglomerates near the base of the series patches carrying a
few pennyweights have been worked.

Further east on the north-eastern portion of Zandrivier and the northern part of Wildebeeste-
fontein, known as Noodshulp, the schists of Mount Mare continue, and auriferous quartz reefs are
found interbedded in them and in phyllites, some of which resemble Mount Mare varieties. Babe's
workings are situated close to the common beacon of Zandrivier, Roodepoort, and Wildebeestefon-
tein (Noodshulp), in nearly vertical thinly bedded compact greenish grey phyllites. ■ Interbedded
in them is found a white quartz reef up to four feet thick in places, consisting of one or more parallel
layers which may pass locally into pockets. The yield is said to be up to 20 dwta. per ton,
although the general appearance of the reef is very "hungry." In these phyllites along their strike
towards Mount Mare reefs have been located at one or two other points.

South of Mount Mare gold has long been known to occur on Eersteling, Vrischgewaagd, and
Waterval both as alluvial and reef gold. The commonest country rock is a compact and hard dark
greenish grey chloritic schist like some varieties found in the lowest horizons of Mount Mare ; to
thin belong the bulk of the dark coloured schists, such as those occurring at the old shaft of the
Waterval Gold Mining Co., and at the old " Fair Maid of Perth " Mine in the northern corner of
Vrischgewaagd, while the pale greenish and softer schists exposed in the bed of the river at Wullff's
workings resemble certain phases of the ottrelite-schist of Mount Mare. For these reasons the schistose

rocks in the southerly area are beat assigned to the Swaziland System. Close to the right bank of
the main spruit, draining Vrischgewaagd, are small earthy " paddocks " containing quartz float and
about three feet thick, resting on the last-named greenish schists. Alluvial gold has been worked
from these patches, and the underlying country rock also contains interbedded quartz reefs. At
various places on thiB farm occur irregular " blows " of opaque whitish " hungry " quartz, which may
lead to more persistent and payable deposits in depth, though further prospecting and development
is required.

On Eersteling alluvial gold was obtained as far back as 1868 close to the creek near the old
homestead. The country rock is a greenish basic schist weathering to a dirty yellowish green, not
unlike that of the Waterval Gold Mining Co.'b shaft. The alluvial gold is probably derived from
the disintegration of the auriferous quartz reefs, as the distribution of the former seems to follow
those reefs, and small blocks of quartz carrying as much as 80 ounces of gold have been found
within a few yards from the outcrop of the reef. Here again occur irregular " paddocks " of earthy
ground, two or three feet deep, carrying a little alluvial, or rather eluvial gold.

The bedded quaitz reef on Eersteling runs with the schistosity of the country rocks and is
interbedded with them in an approximately east and west direction. It is about two feet thick
and consists of impure quartz carrying from 7 to 8 dwte. Other reefs, less persistent than
the above, are found striking across the formation, but they are more pockety and uncertain.

Unlike any of the above occurrences, the country rock of the Palmietfontein Mine, about
five miles south of Pietersburg, is Older Granite, though schists mainly of igneous origin occur in
close proximity to the granite, which is surrounded by them on all sides. The mode of occurrence
of the granite over Palmietfontein and the adjoining farm Roodepoort points to an intrusive relation-
ship to the schists, and the fresh grey granite at the mine is probably a lenticular offshoot from the
main mass. The reef is a well defined body of fresh white quartz dipping 50 to 60 degrees to
the south-west and embedded in coarse fresh grey granite. On the east aide it ends abruptly against
Boft clayey schists. As a rule it is only a single extremely well defined body, but at the 185 feet level
it passes off into stringers, when a thin soft clay film is found separating the main reef from the
granitic foot-wall. The average width is eighteen inches, but on the 85 foot level, the reef
sometimes expands from six inches to four feet in three or four yards. The gold content is extremely
variable, averaging about 4 dwts. Those portions which consist of the purest white quartz contain
very little gold. At the 45 foot level a brownish red and very rich variation comes in, some-
times slightly honeycombed. Specks of very dirty gold may sometimes be seen, but as a rule the
reef consists of hungry white vein-quartz. No associated minerals and practically no pyrites occur.
Owing to the solid nature of the country rock little timbering is required, the mining being easy and
clean.

Near the eastern termination where the contact between granite and schists is found, a layer
of quartz boulders was met with, some twenty feet below the surface. The boulders are up to
fourteen inches in length, more or less rounded and mixed up with various rocks belonging to the
formations of the surrounding country. This apparently alluvial deposit was highly payable.

The presence of gold and copper in the Dolomite area on both Bides of Chunies River has long
been known, but such occurrences are not of great importance. On the farm Doornkloof occurs
an intrusive sheet in the lower portion of the Dolomite, which has slightly altered the overlying rock.
This sheet and probably others are associated with auriferous quartz reefs. The latter can be seen
to dip at about tO degrees against the direction of the Dolomite dip and in one pit two reefs are
exposed. The upper one U about two feet thick and lies entirely within the intrusive dioritic rock,
while the lower one has a thickness of about one foot and passes upwards across the boundary of
the igneous rock for about three feet into a compact hard ferruginous slate. Probably the mineralisa-
tion depends upon the presence of the intrusive bodies.

In the same district copper has been found in several localities, but none of these occurrences
are on an economic scale.

VI.

ON A PORTION OF THE BUSHVELD

BORDERING

THE CROCODILE RIVER

AND INCLUDING THE ROOIBERG TIN-FIELD.

By W. A. HUMPHREY (Oeologist).

Contents.

2. Physical Features.

(a) Surface Relief and Drainage.

3. Geological Structure.

i. Geological Formations.
(a) Transvaal System.
Dolomite.
Pretoria Series,
(b) Rooiberg Sedimentary Series.

(c) Waterberg System.

Volcanic Breccia.
Felsite.

(d) Igneous Rocks.

Red Granite.

NorHe.

Diabase.

5. Economic Geology.

Tin-Deposits.
Gold.

VI.-ON A PORTION OF THE BUSHVELD BORDERING THE CROCODILE
RIVER AND INCLUDING THE ROOIBERG TIN-FIELD.

By W. A. Humphrey (Geologist).

1. Area.
The area which is described in the present report comprises a stretch of country borderinS
the Crocodile River (Rustenburg District) and stretching from its confluence with the Elands River
northwards. It includes the lower portion of the basin of the Sand River, the Boshofberg, and the
Elandsberg, together with the country between them and the Rooiberg. To the south and east
it ib confined by the Elands and Crocodile Rivers respectively as far north as the farm Liverpool
No. 522. From this farm the boundary runs eastwards to the southern end of the Rooiberg and
then northwards, roughly following the line dividing the Waterberg and Rustenburg Districts. From
the farm Donkerpoort No. 945, the northern boundary of the area stretches roughly from Qatkop
westward to the junction of the Brak Spruit with the Crocodile River.

2. Physical Features.
(a) Surface Relief and Drainage.

Topographically the area presents a somewhat varied aspect. The Crocodile River and its
tributaries have been the formative agents giving the general outlines to the whole area, while the
complicated geological structure and extensive faulting are responsible for the splitting up of the
area into several distinct zones which differ markedly from one another in the nature of their scenery
and general characteristics. These contrasting areas are generally determined by the nature of the
rocks of which they are composed, the roughly broken country consisting of sedimentary rocks,
while the more regular undulating type of scenery marks the outcrops of the masses of igneous rock
which surround them. The transition from the one type to the other is usually sudden and very
marked.

The tributaries of the Crocodile River flow for the most part in directions almost at right angles
to the course of the main stream, descending from the high ground situated on the east and west.
The Crocodile River itself flows roughly from south to north, thus following the general slope of the
country. In the southern portion of the area the river flows through a rolling granite country, which
rises gradually from the banks of the river in gently rounded bults and flat hummocky shoulders.
The typical piled up granite kopje is noticeably absent. The courses of the tributary streams are
sharply defined, being cut out on smaller lines than that of the main stream, and give rise to
steeper banks and more varied scenery.

Occasionally along the banks of the main stream and usually near the junction of tributary
spruits wide expanses of alluvial flats are developed, but in the southern portion of the area these
occur only at comparatively long intervals, and are not so continuous as lower down.

On the farm Slagt Kraal No. 741, the Crocodile River enters a roughly broken tract of country
composed of sedimentary rocks which form prominent ranges of hills ; massive kopjes overshadow
the river, and high ridges alternate with deep valleys. On the farm Doornkloof No. 624 the river
passes through a narrow poort and immediately afterwards receives the Aapies River on its right
bank, thence flowing for some miles due west. From the bend where it resumes its northerly course
the alluvial stretches along the banks are practically continuous. Below the farm Liverpool No. 522
are extensive alluvial flats which give place on Kromdraai No. 692 to the black turf soil characteristic
of the norite, and which forms a featureless plain stretching for several miles to the west of the river.

At Wachteenbeetje Drift, two miles south of the Sand River, the Crocodile River re-enters
the sedimentary formation, and the scenery varies accordingly, the river cutting its way across a
series of ridges which trend east and west.

The only considerable tributary belonging to the area dealt with is the Sand River, which
enters the Crocodile River to the north of Wachteenbeetje Drift. It contains no surface flow during

rocks in the southerly area are beet assigned to the Swaziland System. Close to the right bank of
the main spruit, draining Vrischgewaagd, are small earthy " paddocks " containing quarts float and
about three feet thick, resting on the last-named greenish schists. Alluvial gold has been worked
from these patches, and the underlying country rock also contains interbedded quartz reefs. At
various places on this farm occur irregular " blows" of opaque whitish " hungry " quartz, which may
lead to more persistent and payable deposits in depth, though further prospecting and development
is required.

On Eersteling alluvial gold was obtained as far back as 1868 close to the creek near the old
homestead. The country rock is a greenish basic schist weathering to a dirty yellowish green, not
unlike that of the Waterval Gold Mining Co.'s shaft. The alluvial gold is probably derived from
the disintegration of the auriferous quartz reefs, as the distribution of the former seems to follow
those reefs, and small blocks of quartz carrying as much as 80 ounces of gold have been found
within a few yards from the outcrop of the reef. Here again occur irregular " paddocks " of earthy
ground, two or three feet deep, carrying a little alluvial, or rather eluvial gold.

The bedded quaitz reef on Eersteling runs with the schistosity of the country rocks and is
interbedded with them in an approximately east and west direction. It is about two feet thick
and consists of impure quartz carrying from 7 to 8 dwts. Other reefs, less persistent than
the above, are found striking across the formation, but they are more pockety and uncertain.

Unlike any of the above occurrences, the country rock of the Palmietfontein Mine, about
five miles south of Pietersburg, is Older Granite, though schists mainly of igneous origin occur in
close proximity to the granite, which is surrounded by them on all sides. The mode of occurrence
of the granite over Palmietfontein and the adjoining farm Roodepoort points to an intrusive relation-
ship to the schists, and the fresh grey granite at the mine is probably a lenticular offshoot from the
main mass. The reef is a well defined body of fresh white quartz dipping 50 to 60 degrees to
the south-west and embedded in coarse fresh grey granite. On the east side it ends abruptly against
soft clayey schists. As a rule it is only a single extremely well defined body, but at the 185 feet level
it passes off into stringers, when a thin soft clay film is found separating the main reef from the
granitic foot-wall. The average width is eighteen inches, but on the 85 foot level, the reef
sometimes expands from six inches to four feet in three or four yards. The gold content is extremely
variable, averaging about 4 dwts. Those portions which consist of the purest white quartz contain
very little gold. At the 45 foot level a brownish red and very rich variation comes in, some-
times slightly honeycombed. Specks of very dirty gold may sometimes be seen, but as a rule the
reef consists of hungry white vein quartz. No associated minerals and practically no pyrites occur.
Owing to the solid nature of the country rock little timbering is required, the mining being easy and
clean.

Near the eastern termination where the contact between granite and schists is found, a layer
of quartz boulders was met with, some twenty feet below the surface. The boulders are up to
fourteen inches in length, more or less rounded and mixed up with various rocks belonging to the
formations of the surrounding country. This apparently alluvial deposit was highly payable.

The presence of gold and copper in the Dolomite area on both sides of Chunies River has long
been known, but such occurrences are not of great importance. On the farm Doornkloof occurs
an intrusive sheet in the lower portion of the Dolomite, which has slightly altered the overlying rock.
This sheet and probably others are associated with auriferous quartz reefs. The latter can be seen
to dip at about 10 degrees against the direction of the Dolomite dip and in one pit two reefs are
exposed. The upper one is about two feet thick and lies entirely within the intrusive dioritic rock,
while the lower one has a thickness of about one foot and passes upwards across the boundary of
the igneous rock for about three feet into a compact hard ferruginous Blate. Probably the mineralisa-
tion depends upon the presence of the intrusive bodies.

In the same district copper has been found in several localities, but none of these occurrences
are on an economic scale.

VI.

ON A PORTION OF THE BUSHVELD

BORDERING

THE CROCODILE RIVER

AND INCLUDING THE ROOIBERG TIN-FIELD.

By W. A. HUMPHREY (Geologist).

108

abnormality of the junction, which can be most readily explained by the presence of a fault plane.
On the west the granite cuts out the Pretoria Series for some twenty miles and lies directly on the
banded ironstone* which overlie the Dolomite.

At the north-west corner of this group of sedimentaries the Pretoria Series quartzites again
reappear, but broken and only extending for a few miles. These are apparently normally overlain
by norite, which stretches for some distance to the north-west, while in the far distance the Witfontein
Mountains, which consist of Pretoria Series* quartzites are seen and dip roughly south-west. It
would thus seem possible that these quartzites and underlying rocks form a shallow syncline under
the norite, and that this apparently isolated mass of sedimentary rocks is continuous from this point
with the main line of the Witfontein Mountains to the north-west beneath the norite. Otherwise
the whole of this body of rocks represents a huge fragment of the Transvaal Syatem broken off from
the main mass of the formation at the time of the great subsidence of these rocks after the intrusion
of the Bed Granite.

4. Geological Formations.

The geological formations represented in the district under review are as follows, in ascending
order: —

Sedimentary Formations.

1. Transvaal System.

Dolomite.

Dolomite and Chert.

Banded Ironstones.
Pretoria Series.

Shales.

Quart Ei tea,

2. Rooiberg Sedimentary Series.

Igneous.

1. Bushveld Igneous Complex.

Granite.
Norite.

2. Interstratjfied in Transvaal System.

Diabase.

3. Felbtfe.

4. Volcanic Breccia.

(a) Transvaal System.

Rocks belonging to the Transvaal System, as before stated, are developed over two distinct
portions of the area, namely in a broad belt along the northern boundary and in a compact group
around the mouth of the AapieB River. It will be more convenient to consider these two occurrences
separately.

I The more normal and widely extending of the two occurrences is that which stretches in an

east-west direction along the northern boundary of the area. This is part of the extensive develop-
ment of these rocks which is to be found along the northern margin of the Bushveld Plutonic Series
from the Marico District eastward. They end up abruptly in the Red Granite immediately east
of the area on the farm Sandfontein No. 2155. The main formation may be continued under the
quartzites and sandstones of the Waterberg formation which is developed immediately to the north,
but there is no continuous outcrop connecting these rocks with the banded ironstones and Dolomite
near Warmbaths as has been asserted, an uninterrupted zone of Red Granite being developed from
the outcrop of the rocks belonging to the Waterberg System on the farm Zandrivier No. 2136 south-
ward to the Aapies River. There is, however, no question that these rocks belong to the same
formation as is developed north-west of Warmbaths.

The most important factor in the structure of the Transvaal System of this northern portion
of the area is the presence of a series of step faults by which the rocks to the north have been
successively let down. It thus happens that as one traverses the formations from south to north
in the contrary direction to that of the dip which is roughly to the south, one does not pass over
the rocks of the Transvaal System in descending order from Pretoria Series to Black Reef, as one

' Holm™.— "Tbe Geology of a part of the Rustenburg District." Tians. Geo). Hoc. S.A., vol. VIII, p. 1.

109

would do were a normal series in question. On the contrary, after traversing the Pretoria Series
through the Banded Ironstones into the Dolomite, instead of Black Reef, one gets a repetition of the
Banded Ironstones with a little of their underlying dolomite, then Red Granite, and lastly, felsites
and conglomerates of the Waterberg System.

It is thus evident that each abrupt change from older to younger rocks from south to north
represents a dislocation on the series, and as the down thrown side is always that to the north of the
faults, all these dislocations are probably due to the same causes. Mr. Holmes* has drawn attention
to the duplication of the Black Reef along the same horizon many miles to the west in the Witfon-
teinberg, caused by faulting on a similar scale and in the same sense as that mentioned above, with
downthrow in the north. He has also noted the duplication of the Banded Irorstones and Dolomite
west of the Crocodile River by a fault of which the most southerly of these mentioned above is the
continuation. In addition he maps a narrow band of Red Granite occupying the same horizon as
in the area under review. It thus seems fairly certain that we have here to deal, not with phenomena
merely local in their character, but with dislocations intimately connected with the general structure
of the northern rim of the Bushveld Plutonic basin, and which will probably be found to have an
important bearing upon the movements which accompanied the intrusion of the huge igneous laccolite.
Plate No. XIV illustrates the general structure as above set forth.

The three main faults mentioned above are accompanied by at least two smaller ones, each
with an insignificant throw which affects only the outcrop of Banded Ironstones overlying the
Dolomite. The outcrops of these faults run parallel to those of the more important ones, only the
throw is too small to do more than increase the thickness of the Banded Iromtonea by a duplication
of a portion of the horizon, being insufficient to duplicate the whole thickness. It is this subsidiary
faulting, affecting the outcrops of the more northerly of the Banded ironstone horizons, which is
accountable for the puzzling contradictory dips which are common along this range. Most probably
the underlying dolomite has been removed along the fault planes by percolating waters, and sub-
sidences have taken place to whose account may be placed the very varying dips met with, more
particularly between Sundays River Poort and Gatkop.

South-east of Gatkop the throw of fault No. 2 duplicating the Dolomite and Banded Ironstones
decreases gradually, and the hills formed by these two horizons gradually approach one another
until on the farm Rookpoort No. 1835 they meet and the duplication of the banded ironstones ends
in the bold hill overlooking Groothoek No. 257 on the west, where they form one massive horizon.

The Transvaal System, consisting of Dolomite, Banded Ironstones, and Pretoria Series, ter-
minates against the Red Granite on the farm Sandfontein No. 2155, although isolated patches of
sedimentary rocks are to be found to the south east.

In addition to the dislocations already enumerated the Pretoria Series is separated from the
Red Granite on the south by a fault plane which cute off the upper members of the series in succession
from west to east. The dislocations affecting the Transvaal System may be classified for con-
venience of reference as follows : —

Fault No. 1. — The fault extending from a point some nine miles west of Crocodile River over
a distance of twenty-nine miles to Groothoek No. 257. This fault duplicates the Banded Ironstones
and part of the Dolomite and has a throw of some 3,500 feet.

Fault No. 2. — To the north of No. 1, bringing the Red Granite down until it is in contact at
the surface with the Dolomite of the faulted portion down thrown by No. 1.

Fault No. 3. — To the north of No. 2 bringing the felsites and conglomerates of the Waterberg
System into contact at the surface with the Red Granite down thrown by No. 2.

Fault No. 4. — Small fault affecting only the more northerly of the Banded Ironstone horizons.

Fault No. 5. — The fault which bounds the Pretoria Series on the south and cuts off the upper
member of the series in succession from west to east, bringing their edges against the Red Granite.

Fault No. 6. — A local overthrust described in detail by Dr. E. T. Mellor affecting the country
around Gatkop. This fault breaks the continuity of faults Nob. 1 and 2 and of the northerly Banded
Ironstone horizon at Gatkop. They are continued eastward from the middle of the farm Sand-
spruit No. 953. That this overthrust is merely local is shown by the fact that the immediately
succeeding horizon to the south is unbroken and unaffected.

The Transvaal System as developed along the northern boundary of the area consists of —

1. Dolomite and Banded Ironstones.

2. Pretoria Series.

* Holmes.—" The Geology of a portion of the Rustenbmg District." Trans. Geol. Soc. S.A., vol. VIII, p. I ,

110

1. Dolomite and Banded Ironstones.

The Dolomite \b exposed along two parallel horizons neither of which can be considered
normal, as in neither case is the whole thickness of the series present. The thicker of the two parallel
occurrences is the more southerly which underlies the Pretoria Series conformably. The base of
the series has been faulted away, and the lower horizons of the Dolomite rest in all cases upon Banded
Iionstonee from which they are separated by a strike fault plane. To the west of Sundays River
Poort the succession of the beds of the whole system is more complete and less disturbed than to
the east of the poort. The thickness of Dolomite present to the west of the poort is probably about
3,500 feet. It is of the normal bluish type, and is in no way petrologically different from that found
in the Pretoria and Ruatenburg Districts. The massive bands of chert, however, which characterise
its upper horizons in the districts mentioned are not present here. Occasional isolated kopjes are
found with fairly massive chert, and thin chert bands, varying from a mere thread to an inch in thick-
ness, are found throughout its distribution. The siliceous bands in the overlying ferruginous banded
rocks are much more compact and cherty than quartzitic, and this horizon may perhaps be regarded
as the equivalent of the massive and banded chert so common in other districts where the Banded
Ironstones are not developed as a definite independent horizon. Evidence in support of this view
was gathered from the Zeerust District by the writer* from that portion of the Dolomite and over-
lying banded rocks lying west of Zeerust, where in fact these rocks first make their appearance as a
definite horizon, replacing more or less the massive chert horizons found farther to the east in the
RuBtenburg and Pretoria Districts.

A noticeable feature of these banded rocks in this district is the presence in many places
between them and the overlying shales of the Pretoria Series proper of a band of chert conglomerate
of some twenty feet in thickness. Occasionally a double band of conglomerate is found, usually
separated by massive ironstone. This conglomerate, which was frequently observed in the Marico
District, might be taken as marking the base of the Pretoria Series, thus relegating the banded iron-
stones to the Dolomite with which in their distribution both here and at the mouth of the Aapies
River, as well as in the narrow band found near Warmbaths, they are always intimately connected.
Nowhere does the banded ironstone occur away from Dolomite, while in many places it occurs witb
Dolomite where the Pretoria Series is not present.

The Dolomite contains the usual admixture of manganese, which weathering out gives the
peculiar and characteristic purplish brown colour to the Dolomite soil. In several places oaves
occur, the best known of which is that situated south of Gatkop. The opening is situated to the
west of a small knoll a short distance to the north of the main road, and is marked by trees and
bushes which obscure the entrance. The opening is somewhat funnel-shaped, but not vertical.
This cave was penetrated to a distance of some 200 feet by Dr. E. T. Mellor, who describes it as
consisting of several chambers, of which the outermost one has been used by natives, the remains
of whose goat kraals are still visible. The inner chambers which are very lofty are covered with
very thick deposits of guano which bad never been disturbed. The cave descended sharply to the
north-east and the floor became fissured, and, owing to the presence of guano masking the dangerous
places, extremely difficult of access. Dr. Mellor did not reach the water level. On the farms Sand-
river Poort No. 342 and Waterval No. 965 are Beveral chert kopjes and short ridges.

The dolomite occupies the whole of the valley of the Sand River from the centre of the farm
Sandspruit No. 953, as far as Sand River Poort No. 342, and extends almost to the summit of the
banded ironstone ridges to the north and south.

The more northerly outcrop of dolomite is of considerably less extent than that just dealt
with, and extends from the west through the farm Waterval No. 965 and broadens out north of
Sundays River Poort. It crops out on the northern slopes of Gatkop ; further to the east it is only
found in the northern slopes of the kopje situated near the western boundary of the farm Sand-
spruit No. 953 and south of the Sand River. It is always overlain normally by banded ironstones
on the south, and on the north abuts on the fault plane No. 2 which separates it from the Red Granite.
This horizon shows nothing peculiar calling for remark. Together with the banded ironstone which
overlies it, it owes its appearance here to fault No. 1 which has caused the duplication above referred to,

Banded ironstones. —An with the Dolomite just described, these rocks are duplicated and
appear as two parallel horizons constituting the main elevated ridges formed by the rocks belonging
to the TranBvaal System in this neighbourhood. The more southerly of these ridges is more typical
and shows the whole succession, while the more northerly being the faulted portion is more disturbed
and is wanting in its uppermost beds.

This horizon consists foi the greater part of bluish and brown banded rocks of great hardness.
Usually the thickness of the individual bands varies from one-eighth of an inch to half an inch, but
in the upper portions much greater thicknesses are occasionally met with. The bands consist of
white and yellow and occasionally pink compact chert, occasionally graduating into a very fine
sugary, quartzite ; interbanded with these are layers of haamatite containing some magnetite. On
the fresh fracture the ironstone bands are blue and metallic; they tend, however, to weather a dull
red brown and to stain the alternate cherty bands yellow and brown. These ridges are almost
always magnetic and deflect the needle, rendering observations taken by compass unreliable.
Occasionally the magnetite is absent in any quantity and the compass is not affected.

Towards the top of this horizon the ironstone bands become more massive and occasionally
reach an individual thickness of several feet. A typical example of this is to be seen on the
farm Clipgat No. 72, where a massive band of iron ore is to be Been overlying the banded rocks. This
is traversed in its upper portions by a double band of chert conglomerate, each approximately twenty
feet thick. The conglomerate consists of rounded chert fragments in a siliceous groundmass.

The true thickness of this banded horizon is somewhat difficult of determination owing to the
great amount of debris which always obscures the dip-slope. It probably varies between 150 and
200 feet. Much greater thicknesses may be measured both at Sundays River Poort in the bed of
the stream, and also on the escarpment at Gatkop. In the former place the increased thickness is
due to duplication caused by the small fault No. 4, and most probably more than one of these minor
strike faults traverse this northern range.

At Gatkop the whole structure is much involved, the banded ironstones developed on the
peak forming a syncline whose axis coincides with the strike of the main range. The beds dip in to
the centre of the mountain from the north, east, and south, showing a bold escarpment towards those
points of the compass. That facing towards the north can be followed along the ridge to tie south-
west, while the escarpment to the south only occupies the face of Gatkop itself, beyond the limits
of which the synclinal structure is not continued. The thickness of these ironstones is abnormal
and due partly to the small strike faults before mentioned, by which portions of the ironstones are
duplicated, probably several times, and partly doubtless owing to the reversed fault or overthrust
by which the continuity of this horizon has been broken at this point. At Gatkop the banded
ironstones end abruptly against the fault plane No. 6, the topographical continuation of the range
to the eastward consisting of rocks belonging to the Waterberg System, and not of rods belonging
to the Transvaal System.* The continuation of the banded ironstones to the east of the fault plane
is found on the farm Sandspruit No. 944 to the south of the Sand River. Their thickness here is
very insignificant, and the contrast between their development on Sandspruit' and that on Gatkop
is very striking. As before shown, however, this horizon of banded ironstones dips directly towards
and is cut off by the plane of fault No. I, and with advancing denudation tends to disappear altogether.
This it actually does on Sandspruit No. 944. In the bed of the Sand River, where denudation has
been carried furthest, the banded ironstones appear on either Bide of the river on the rising ground,
leaving a gap of about one and a half miles where the ironstones have been completely denuded.
This fact serves as corrobatory evidence to the accuracy of the relations described on the two sides of
fault No. 1. There can be no doubt that this range is the geological continuation of Gatkop, its
relation to the Dolomite and to the granite being identical with those of the Gatkop range further
to the west. The throw of the fault No. I gradually becomes less as the beds are followed east,
and consequently the two banded ironstone horizons approach one another more and more nearly
until they finally unite and form the thick mass of these rocks which occupies the prominent ridge
trending along the south-easterly boundary of the farm Rookpoort No. 1635, ending in the bluff
which overlooks the farm Groothoek No. 257 from the west.

The massive ironstone bands which form the uppermost beds of this horizon together with
the chert conglomerate are not found along the more northerly of the two outcrops. This is due
to the fact that the outcrop of the banded ironstones abuts directly against the fault plane, dipping
towards it, and the upper beds have been removed, denudation having now caused the disappear-
ance of everything overlying the main banded ferruginous rocks. In course of time these too will
disappear, leaving Dolomite abutting against Dolomite to mark the outcrop of the fault plane.

The more southerly and main banded ferruginous quartzite horizon extends unbroken from
the Crocodile River on the west across the whole of the area. It varies little in character, showing
usually the normal succession of banded rocks, massive ironstone, and chert conglomerate. Here

* Rejiort of Geological Survey for 1907. p. B4,

112

and there, however, abnormalities occur. On the bluff overlooking Sand River Poort on the east,
the dip of the banded rocks on the summit is 65 degrees to the south. Descending the southern
elope of the range the dip becomes reversed, and half-way down registers 45 degrees to the
north. At the same time the ironstone and conglomerate bands are not found here or for some
distance to the east, although the northerly dip is not persistent. Probably this marks a small local
strike fault, on one side of which the dip has been reversed owing to the formation of caves along
the fault plane of the Dolomite by the action of water and to the subsidence of the overlying iron-
stones. On the farms Rhenoeter Kloof No. 1720 and Weltevreden No. 1782 there are no signs of
the upper beds of this series, and here too the dip increases to 80 degrees. Possibly these horizons
may be intermittent along this stretch of country, but more probably a series of minor strike faults
exists throughout the whole Transvaal System as here developed, which owing to the nature of the
rooks (large areas being covered with Dolomite) are only occasionally evident.

2. Pretoria Series.

The Pretoria Series as here developed consists of three parallel ridges of quartzite with
intervening bands oi softer rock which seldom show any outcrops. These might be classed as the
Timeball Hill, Daspoort, and Magaliesberg horizons did the quartzites composing them show any
striking resemblance to those included in these horizons in the better known development of the
series in the Central Transvaal.

These rocks, however, are separated by a length of outcrop of 135 miles measured along the
strike from Linokana in Moilo's Location in the Western Transvaal. This is the nearest point
that the detailed mapping of the Pretoria Series in the Western Transvaal has reached. They are
distant also one hundred miles from Pietpotgietersrust, the nearest locality where the succession
in the eastern development of the series has been established.

It seems consequently somewhat premature to definitely correlate particular horizons of the
Pretoria Series as exposed here with the horizons which are recognised in the Pretoria, Rustenburg.
and Harico Districts. Although the succession of the Pretoria Series is very constant over great
distances in the Eastern Transvaal, and the three main horizons can be followed continuously, yet
there is evidence in the Marico District of the Western TranBvaal that the lower horizons of the series are
gaining in thickness at the expense of the upper. *Thus at Breedts Nek in the Magaliesberg, east of
Rustenburg, the thickness of the Magaliesberg Quartzites approximates to 450 feet. At Onderste Poort,
north-west of Rustenburg, they do not measure more than 150 feet, while on the Marico they have
dwindled still more and do not measure 100 feet. As contrasted with this, south of Boon's Siding on
the Krugersdorp-Zeerust railway, the Timeball Hill Shales begin to increase in thickness when followed
westward, and a new quartzite makes its appearance, below which a scries of banded shales and quartzites
gradually developes overlying the Dolomite. In the Marico District there are thus two distinct
additional horizons below the Timeball Hill Quartzites in addition to an increased thickness of shales.
Should this tendency continue of the Magaliesberg beds to decrease in thickness and the Time-
ball Hill beds to increase as the series is traversed along the strike, it is quite possible that by the
time the 135 miles separating Linokana from the Crocodile River, via the Dwarsberg and Witfon-
teinberg, are traversed, the Magaliesberg horizon may have disappeared entirely, and possibly also
the Daspoort horizon and it is quite conceivable that the series as exposed along the northern run
of the Bushveld consists only of an increased development of the Timeball Hill Series. For the
present therefore it will be better to speak of the quartzites as lower, middle, and upper, until they
can be definitely traced into one or other of the recognised and named horizons.

The individual horizons of the Pretoria Series are only intermittently developed in this area,
quartzites forming prominent ridges disappearing entirely over several miles, and reappearing again
with exactly the same petrological characteristics further on, and it is only on the farm Klipgat No.
172 that a complete and continuous section can be obtained.

The beds met with in ascending order on Klipgat are as follows :—

Chert Conglomerate 20 feet.

Grey-blue Flagstones 700 „

Diabase 210 „

Indurated Shale 200 „

* Ht-"!" 11 "' "f Heolofrie*! Surrey for IB07. |i. H*,

Lower

Quartzite

Middle
Quartzite

Red Quartzite

.. 200 feet

White Quartzite

.. 20 „

Red Sandstone

.. 25 „

Shale * (No Outcrop)

..850 „

Grey Black Quartzite

.. 40 „

Green Shale

.. 40 „

White Quartzite

.. 60 „

2,365 „

Overlying the topmost white quartzite is a wide stretch of some two and a half miles covered
by sand and alluvium and forming the immediate valley of the Sand River. Two small ranges of
white quartzite running parallel to the strike of the Transvaal System crop out here and there over
this area, the higher of which is the more persistent. It can be well seen in the bed of the Crocodile
River immediately below Waehteenbeetje Drift, and west of the Crocodile River it forms a con-
tinuous and well marked range for some distance. These quartzites have the same dip as the lower
members of the Pretoria Series enumerated above, and if the intervening space obscured by
alluvium is occupied by rocks belonging to the Pretoria Series, a thickness of some 4,000 feet must
be added to that already measured. This point will he settled as soon as the mapping is carried
further west and this horizon ceaBes to occupy the low ground of a valley.

Chert Conglomerate. — The lowermost bed of the Pretoria Scries is a chert conglomerate. It
consists of rounded chert fragments embedded in a quartzose matrix. The whole rock is very compact
and has a thickness of some twenty feet.

Shales and Flagstones.— Overlying the conglomerate on Klipgat No. 72 is a considerable thick-
ness of hard grey-blue flagstones which make up the whole of the southern slope of the ridge leading up
to the banded quartzites. They weather red and are well and evenly bedded. This horizon affords
but few outcrops, of which the one on Klipgat No. 72 is much the most extensive and best exposed.
Following the shales along the strike to the east no outcrop was found for a distance of twenty miles.

On the farm Rhenoster Kloof No. 1720, however, east of the poort where the Sand River
first breaks northwards through this horizon, good exposures are to be seen in the river bed. The
shales here show signs of contact metamorphiBm, well developed needles of chiastolite being dis-
tributed evenly through them. This is the same geological horizon as that described in a previous
report* in which it was stated that no outcrops whatever were found. It is, however, now
certain that the first development of rocks overlying the banded ironstones and their accompanying
conglomerate is a blue evenly bedded and usually soft shale which is only exposed at long intervals.
The presence of the underlying banded ironstones always forming a prominent ridge immediately
dominating these shales and shedding a large amount of detritus which obscures the slopes is
sufficient to explain the paucity of outcrops of the latter. Masses of surface conglomerate with a
ferruginous matrix derived from the banded ironstones are common along the whole range, and
more especially on the dip-slope side.

Diabase. — Intruded into this shale horizon is a thick sheet of igneous rock which shows the usual
rounded boulders along its outcrop. It is a dark blue grey rock of even grain consisting principally of
an intergrowth of pale green pleochroic hornblende and diopside. Felspar, probably bytownite, is
fairly well distributed, and small quantities of quartz and granophyric intergrowths of quartz and
felspar are to be seen. The shale below this intrusion appears somewhat more compact and horny
and shows less bedding than the upper horizons.

The Lower Quartzite. — Overlying this Bhale is the lowermost quartzite horizon, which is the
most persistent and continuous of the series, and can be traced with minor breaks from the Crocodile
River for twenty-one miles to the east.

It consists of three distinct varieties of quartzite which are constant in their mutual relations
throughout the area.

The lowermost of these is a greenish grey quartzite of fine grain. It weathers deep red brown,
and for some distance from the surface apparently fresh fractures reveal the brown colour. Its true
colour is, however, as above stated ; the outcrop of course shows the red weathered tints.

* Report of Geological Survey for 1906, p. 112.

This is overlain by a white sugary quartzite of fine grain and of true Pretoria Series type,
which is followed by a dark reddish purple micaceous sandy quart-Kite of inferior hardness.

Of these three quart zites the middle white quartzite always forms the crest of the ridge, which
rises occasionally to considerable heights. The upper micaceous sandy quartzites are usually found
at the foot of the white quartzite dip-slope forming a slightly raised feature or small prominence.
The underlying hard green grey quartzite usually forms a small subsidiary ridge to the north of the
main ridge, occasionally mounting the escarpment slope for some distance. On Klipgat and along
the eastern extension of this horizon it is evident that some softer horizon separates these two lower
quartzites. On Groothoek No. 257 in the south eastern corner of the farm this horizon is occupied
by a metamorphosed shale. Its outcrops simulate those of a basic sheet in their rounded forms.
It is probable that a soft shale separates these two quartzites across the whole area, the outcrops
always occupying a comparative depression and being always obscured by detritus from the two
neighbouring quartzite bands. The metamorphosed shale from Groothoek is a blue-black rock
appearing macroscopically almost hoi ocrystal line. It consists of large crystals of chiastolite and
flakes of me tarn orp hie biotite in a groundmass of quartz grains.

Overlying the lower quartzite zone on Klipgat is a stretch of some 400 yards without any
outcrop, and this horizon furnisher outcrops only at one locality throughout the area. On
the farm Sand Rivera Poort No. 342 and south-west of the poort the dip of the quartzite is almost
vertical, and when followed to the west the dip decreases to about 65 degrees. About two miles
west of the poort a succession of beds is found where good outcrops of chiastolite shale occur over-
lying the uppermost red quartzite. From the evidence of this outcrop it is probable that the level
ground occupying this horizon to the west and showing no outcrops is occupied by shales which must
be of considerable thickness, as the surface width of the horizon varies between 400 and 100 yards.

Middle Quartzite. — Above these shales occurs the middle quartzite which is not by any means
so continuous in its outcrops as the lower. Outcrops are found on Klipgat No. 72, Brakvlei No. 40, and
Rooiwal No. 143, while on Sandfontein No. 2155 this horizon attains its maximum development
and thickness on this area. On Klipgat the succession of beds is confined to

Grey to black quartzites • . , . . 40 feet.

Pinkish green shale . . . . . . . . 40 „

White quartzite 60 „

On Sandfontein, however, the succession is as follows : —

Quartz conglomerate consisting of small quartz pebbles in a gritty groundmass often

very micaceous and grey in colour.
Sandstone, pinkish to white, friable, slightly micaceous and blotchy, often much sheared

and occasionally very gritty, sometimes full of glassy quartz blebs.
No ou^craps. — Two hundred yards.

Quar! zite. — Very compact chert-like quartzite, pink graduating to dark blue, very hard.
Diabase. — Thin intrusion of diabase.

Shales. — Blue compact shales whose lower portions are black and graphitic.
This succession as here exposed represents a thickness of at least 1,500 feet, and it is somewhat
peculiar that these quartzites should be so much thicker than those of the same horizon further west.
Although the structure on Sandfontein is abnormal, the Pretoria Series being only developed over
a strike of seven miles and terminating abruptly in granite at each extremity, yet when traversed
against the dip in a northerly direction, these beds are seen to pass conformably down through the
lower quartzite to the banded rocks and the Dolomite.

The considerable thickness of shales which overlies the middle quartzite on the farm Sand-
fontein No. 2155 is of assistance in helping to determine the succession on Klipgat and to the west.
Here, in the lower reaches of the Sand River Valley, which is mostly covered with sand and alluvium,
the only outcrops above those of the middle quartzite are a disconnected series of ridges of white
quartzite (the upper quartzite), separated from each other by broad expanses of flat country. It
is possible that these stretches of undeterminable country, occupied by alluvium, are in reality
occupied by shale. At any rate it is pretty certain that for several hundred feet above the middle
quartzite shale is developed, above which three well defined horizons of white quartzite rise at
intervals through the alluvium of the Sand River Valley.

US

From the foregoing it will be seen that while a complete succession of outcrops through the
Pretoria Series is nowhere to be found in the district under review, yet it is possible to fill in the missing
details of one section from other sections taken at different points along the outcrop. The upper
portion of the series is imperfect, as there is a large gap as yet unaccounted for between the middle
and upper quartzites, and also the top of the series is not yet determined, the junction with the
granite being a faulted one. One traverse was made west of the Crocodile River where norite was
found apparently conformably overlying the uppermost white quartzite. °o that it is possible
that a little further work west of the Crocodile River will be sufficient to provide a complete section
across the Pretoria Series in this neighbourhood.

The second development of rocks belonging to the Transvaal System to be described is
situated in the neighbourhood of the junction of the Aapies and Crocodile Rivers.

A portion of these rocks was described in a previous report,* namely those situated on the
right bank of the Crocodile River, and the present report, while referring occasionally to these, will
deal more particularly with those situated to the west of the river.

These rocks differ from the group described above mainly in their relative development, the
lower members of the system being better developed than the upper. We have thus a more com-
plete section through the Dolomite together with the interbedded quartzite horizon which occurs
towards its base, while the Pretoria Series is only feebly developed.

The most noticeable feature of the structure of these rocks as here developed, and contrasting
with the faulted nature of those to the north, is the series of large folds into which the strata have been
thrown. In the south-eastern e~rner of the area adjoining the Crocodile River the beds are developed
in the form of an overtilted anticlinal fold. When this fold is followed along the strike to the north-
west it broadens out, the summit becomes denuded, and the two arms stretch away along increasingly
divergent lines. The most convenient horizon upon which to observe this structure is that of the
banded ironstones which overlie the Dolomite. In the south-east this horizon forms the centre of
the fold, its resistant characters giving rise to an elevated ridge, on each side of which the Pretoria
Series beds are ranged in corresponding lines all dipping to the south-west. In the north-east of
Ramakok's Kraal the banded ridge broadens out and finally becomes denuded, and the less resistant
Dolomite exposed along the centre of the anticline forms a broad valley between the two arms of
banded ironstones. The Dolomite in turn becomes denuded along the axis of the anticline and
the quartzites interbedded in the series are exposed to view.

On the farm Langrand No. 1004 the anticline above mentioned is overtilted, and the beds
of the Pretoria Series flanking the Bides all dip in the same direction roughly to the south-west, while
on Slagtkraal No. 741 the series comes to an abrupt termination against the Red Granite for a space.
On the east side of the Crocodile River after a small interval the banded ironstones are continued,
overlain by shales and one quartzite horizon dipping to the south-west. These must be taken as a
continuation of the banded ironstone and overlying beds on Slagtkraal, only on the east of the river
there is no indication of a continuance of the anticlinal structure along the axis running in a
north-west south-east direction. This is most probably due to a fault with an outcrop whose direction
is roughly east and west. On the south side of this fault one has in ascending order dolomite, banded
ironstones, shale, and white quartzites normally developed, while on the north side of the fault the same
beds are developed as an overtilted anticline. Thus the northern side shows much more disturb-
ance than the southern side.

In addition to the above fault, a second fault traverses the system from north-west to south-
east, cutting the east and west fault almost in the bed of the Crocodile River. This fault cuts off
the banded ironstones at the north-west beacon of the farm Vaalkop No. 364, and brings the
Dolomite on that farm into abnormal relationship with the felsite and underlying Pretoria Series
to the west.

The structure at this point is very complicated and involved, due doubtless to the combina-
tion of faulting, folding, and igneous intrusion which took place when this portion of the Transvaal
System was broken away from the main mass of the formation and took up its present position
during and after the intrusion of the Red Granite.

The structure is also partially obscured by the presence on and around the farm Kruidfon-
tein No. 47 of a large expanse of a red felsite and breccia which lies in the syncline of the Transvaal
System, and which is probably of Waterberg age and unconformable to the surrounding rocks of the
Transvaal System.

* Report of Geological (Survey for l!M)ii.

116

Dolomite.

The Dolomite in this locality occupies a considerable area, being exposed along the axis of
the denuded anticline of the Transvaal System, which controls the distribution of its various horizons.
It usually occupies the low ground of the area, rising up in its higher beds to the handed ironstones
which everywhere overlie it. Fetrologically it is of quite a normal type, hard and blue, and weather-
ing brownish owing to the manganese it contains.

On the farms Doompoort No. 1007, Nooitgedacht No. 1604, and extending thence in a northerly
direction is a aeries of quartzites together with a considerable thickness of shales. These might be
taken as representing the Black Beef Series exposed along the axis of the Dolomite anticline, bnt
on the northern boundary of the farm Elandsfontein No. 810, the quartzite anticline is still further
weathered and divides into two arms, exposing first shale and then Dolomite. It would thus appear
that this series of quartzites and shales is underlain by Dolomite and must be considered as an
intercalation in the series and not as marking its base. The quartzites are white to brown and fairly
compact, while the shales show considerable alteration, rudimentary chiastolite crystals being
well distributed through the rock, in which much mica has also been developed. It could be
described as a spotted schist. The whole thickness of this intercalation in the Dolomite probably
does not exceed 100 feet. Applying the evidence gained in this locality to the similar rocks observed
to the east of the Crocodile River,* it would appear that the quartzites and shales there mapped as
Black Beef in reality also belong to this horizon intercalated in the Dolomite. A similar development
of quartzite in the Dolomite has been noted by Mr. Hall in the Eastern Transvaal.

An isolated outcrop of dolomite is to be found on both banks of the Elands River at its junction
with the Crocodile River. A few beds of banded ironstones are developed here, but of only a few
feet in thickness.

The dolomite is in places quite bleached, and in parts consists of white coarse limestone in
which are embedded large brown weathered crystals of siderite or ankerite.

On the farm Langrand No. 1004 the dolomite appears on the south ride of the bill formed
by the overtiltcd anticline, its outcrops being confined on all sides by those of the overlying banded
ironstones.

The banded ironstones which here overlie the Dolomite are similar in every respect to those
described as occurring further to the north. Their higher horizons become richer in iron and in
some places fine outcrops of massive ironstone are well exposed.

From Slagtkraal No. 741 these rocks trend in a north-westerly direction, forming the crest
of an anticline. On the farm Geluk No. 279 the anticline broadens and the crest is denuded, exposing
dolomite. The banded ironstones stretch away in two well marked lines to north-west and north
respectively.

The former arm, with the exception of a break of a few miles on the farm RoodepanNo. 811,
extends northwards almost to the farm Liverpool No. 522. The latter bends to the east on the farm
Elandsfontein No. 810 and crosses the Crocodile River twice on the farm Boschkop No. 7. Here
it bends to the south and, uniting with another branch which meets it from the south, turns to the
east and crosses the Crocodile River as a Byncline overlying Dolomite on both sides. The further
development of these rocks to the east has been described in a previous report. f

Were these rocks only normally folded, then the disconnected ridges ending approximately
at the north-westerly and south-easterly corner beacons of Vaalkop No. 364 respectively should
be joined tip, and both be bent so as to form the continuations of the overtilted anticline ending at
the south-west corner of the farm. The beds, however, have been broken across in two directions
almost at right angles to each other as shown on the map, Plate XVIII.

Pretoria Series.

The Pretoria Series es here developed consists of quartzites and shales with diabase intrusions.

The shales overlying the banded ironstones of the Upper Dolomite are exposed satisfactorily
only in one place, namely, on the farm Kruidfontein No. 147. Here they are seen in the bed of Sand
Skit near the eastern boundary of the farm. They are overlain by a white compact quartzite (the
lower quartzite), but the overlying and underlying Red Quartzite and sandstones of the more
northerly development of this series along the Sand River were not observed.

The n:iddle quartzite is only poorly represented here and consists of one thin band of white
quartzite of similar type to the lower. They are separated by a massive outcrop of basic igneous

ii?

rock of diabasic type, which occasionally becomes almost entirely felspathic. No outcrops of shale
were observed between these two quartzites as in the northern series already described.

(6) Rooiberg Sedimentary Series.

These comprise the series of sedimentary rocks forming the bulk of the Elandsberg, stretching
as a thin layer of quartzites almost to the summit of the Boshoffberg. They occupy the valley
between these two ranges of hills and stretch eastward beyond the Rooiberg.

The series consists in the main of felspathic quartzites, arkose, and shales, with which are
associated some felsites and diabases, and is overlain by a thick sheet of felsite which forms the
Rooiberg and the crest of the Elandsberg, and covers a considerable area to the north of that range,

With regard to the geological position of these beds, when first examined by the Geological
Survey to the south-west of the Rooiberg, they were provisionally placed in the Pretoria Series until
further investigation should determine their true relationships*

They have been mentioned by previous observers f and in each case have they been referred
to the Pretoria Series. The mapping carried out by the Survey during the year 1908 has thrown
much new light upon the details of the series and its relationships. These sedimentary rocks are
everywhere overlain by a thick sheet of felsite. On the summit of the Elandsbergen this felsite has
every appearance of being a sheet conformable to the underlying quartzites and shales. The felsites
further bear a close resemblance to those forming the volcanic fades of the lower portion of the
Waterberg System as at present defined. If this be the position of the felsite, then the Rooiberg
Bedimentaries would occupy a position immediately below the volcanic series of the Waterberg
System.

The outcrops of the Pretoria Series in this neighbourhood are very unlike those of the same
series previously described, and individual horizons occasionally resemble some of those of the Rooi-
berg sedimentaries. The succession of beds however is in no case the same, and it is quite impossible
to consider these beds as the equivalents of the Pretoria Series as developed a few miles to the north
on the farms Groothoek and Sandfontein, or along the main Pretoria Series horizon north of the
Sand River striking east and west.

The succession of the beds of these two series will be compared below.

The extension of the Rooiberg beds to the south-west abuts against rocks of the Transvaal
System some eight miles to the north of the junction of the Aapies River with the Crocodile. The
junction is however a faulted one, the Rooiberg quartzites being cut off against the banded iron -tones
and Dolomite of the Transvaal System. The latter contains no horizon in this locality that could be
looked upon as a continuation of the Rooiberg quartzites, nor can any conformable relation be found
between any known development of the Pretoria Series and the Rooiberg beds. In view of the above
consideration, therefore, it would appear more reasonable to include these Rooiberg beds with the
Lower Division of the Waterberg System.

Comparison of the Pretoria Series

north OF the Rooiberg Area and the

Rooiberg Series.

Pretoria Series.

Feet.

Rooiberg Series.

Feet.

Chert Conglomerate ...

20

Quartiites : White, hard, pink, grey

Blue Shales and Flagstones w

Mottled Quamite

Diabase Intrusion

... 1,110

White Cherty QuarMite

. 500

f Red QuartBiW )

... 200

Pink to Red highly Felspathic Quartiite. con- ,

1 Shale f

Ukfniog Tourmaline (,'assiterite

Quartz ite

1 White Quartiite

20

Shale, Blue and Well Bedded

. 260

{ Red Sandstone

23

Diabase

10

Shale ...

... 860

Grey-black Qnartiite

[ 60

Middle

f Grey- black Quartsite

40

Qoartiite, pink

j Shale, Green

40

Bed Arkose (Qnartiite, Felspar, and Hornblende) '

Baud.

^ White-mottled Quartiite 60 ft.

1

Greyish Ouartiite

| (No Outcrop) SOOft.

Finu-grainod Felsite |

1 Cherty Quartsite

J- 1.600

Diabase i

Diabase

Shale

. 200

Shaks : Blue and Compact to

:

Coarse Breccia i

Graphitic

J

Diabase

Shale

Quartsite ]

Totals

... 3,806

1,050

* Report of the Geological Surrey, 1906, p. 113.

t Holmes.— The Geology of a part of the Rustenburg District. Trans. Geol. Hoc. of S.A., vol. VIII.

E. Jorisson.— On the Occurrence of the Dolomite and Chert Series in the north-eastern part of the Rustenburg
District. Tnuw. Geol. Soc. of 8.A., vol. VII, p. 30.

118

The base of this series rests upon the granite of the Boshof Mountains, which bears an intrusive
relationship to the quartzite. The beds along the contact are quartzites showing considerable altera-
tion, being rich in felspar, and especially in tourmaline, which often occupies the bulk of the rock
and consists of blue, brown, and colourless varieties (d, 362, R.S. No. 766 and 767). These lower
quartzites are of considerable thickness, at least five hundred feet. They consist of various distinct
types as follows, in ascending order : —

1. Whitish felspathic quartzite with tourmaline.

2. Hard grey white quartzite.

3. Pink and white fine-grained quartzite.

4. Coarse quartzite graduating to conglomerate.

5. Whitish pink quartzite with brown specks.

6. Quartzite, composite red, white, and blue, resembling fine-grained granite or aplite.

7. Hard white quartzite.

8. Pinkish sandstone, and quartzite resembling No. 6.

Of these variations one or two are worthy of special mention.

No. 1 along the actual granite contact shows clearly the intrusive nature of the granite. The
tourmaline here is disseminated through the rock in patches and grains, being clearly a result of
contact metamorphism. The larger masses of tourmaline found in higher horizons of these quartzites,
with which cassiterite is associated, while the result of pneumatobtic action, might conceivably be
the result of action subsequent to the consolidation of the main granitic mass, and is not such good
evidence of the intrusion of the granite as these sections along the actual contact.

No. 5 is a very felspathic quartzitic sandstone, the brown patches being composed of limonite
(d, 344, R.S. No. 771).

No. 6 in the hand specimen has much the appearance of a holocryst&lline granitic rock of
fine grain. Under the microscope, however, while seen to consist in almost equal parts of quartz and
felspar, the rock is undoubtedly of sedimentary origin. Biotite occurs in plates, but only sparselv
distributed (d, 339, R.S. No. 770).

No. 8 is the uppermost bed of this lower portion of the series and its outcrops show much
alteration. It is a pinkish sandstone, also appearing almost granitic in parte, and along this horizon
chiefly occur those massive outcrops of tourmaline with which cassiterite is associated so plentifully,
and in which the shafts of the Rooiberg tin mines have been sunk.

In the upper beds of these quartzites is a thin sheet of felsite of some twelve feet in thickness.
In the hand specimen it is deep blue in colour with small pink felspar phenocrysts distributed through-
out. Microscopic examination shows phenocrysts having the shape of quartz, but consisting of
beautiful micrographic intergrowths of quartz and felspar together with large quartz crystals. Very
pronounced flow structure marks this rock.

Shales. — Overlying these basal quartzites is a band of shales which are persistent across the
terrain. Outcrops can be seen in the drift between the tin mines and Upton's Store, also on the
western extension of the horizon in the bed of the most westerly of the spruits joining the Vaalwater
on the south. They are very micaceous and well bedded in the more easterly outcrops ; on the west,
however, they are hard and compact and show little signs of bedding.

Diabase. — A well marked and persistent zone of diabase overlies this horizon ; it is of ultra-basic
character.

Quartzite. — The diabase is followed by white quartzites, which on the farm Olivienpoort No. 758
are apparently overlain by a coarse-grained granite. This granite closes the succession of sedimentary
rocks in this westerly section and stretches in an unbroken line to the east, where it abuts upon the
felsite which overlies the sedimentaries.

On the Elandsberg this quartzite is followed by a series of quartzites and shales which show
good outcrops on the face of the hills on the farm Olivienbosch No. 939. In all, overlying the main
lower quartzites on this farm are three horizons of alternating shales and quartzites presenting fairly
uniform characteristics, of which the lowest has been described above.

The upper beds vary somewhat along different sections, igneous intrusions and flows being
more developed in westerly sections of the Elandsberg than to the east. The succession of these
upper beds forming the southerly slopes of the range is as follows. The section is taken on the farm
Olivienbosch No. 939 along the western boundary of the farm.

119

Pink and white felspathic quartzite of varying grain
Black to red telepathic quartzite
Diabase amygdaloidal
Red mottled sandstone

20

'■100

I
100

the ridges formed by the
It varies from a fine-grained

Pure white quartzite

Shale

Whitish grey quartzite

Felsite

Followed to the west this succession becomes less marked ;
quartzites die away and outcrops are wanting.

A horizon composed of a peculiar arkose makes its appearance,
to exceedingly coarse aggregate of quartz and red felspar with iron ores. The rock occasionally
bears a very close resemblance to granite. Microscopic examination, however, reveals the fact that
the quartz grains, which are often rounded, are surrounded by a margin of subsequently deposited
quartz, its boundaries are marked by fine lines of pigment, and it is optically continuous with the
quartz grains which it Burrounds. This structure points definitely to the plastic origin of the rock
and subsequent silicifi cation. These upper horizons cannot be traced to the west, as the detritus
from the felsite, whose outcrops begin to descend the southern face of the range on the farm Onver-
wacht No. 294, tend more and more to conceal the underlying sedimentary horizons.

This also obscures the actual relations between the felsite and the underlying sedimentary
rocks, an important point in determining the stratigraphical position of the latter, since the existence
of a marked unconformity between the two would probably indicate an independent position for the
sedimentary beds between the Waterberg and Transvaal Systems. The evidence, however, on the
whole would appear to be more in favour of the generally conformable relations.

The tin occurrences to be subsequently described fall within the area occupied by these sedi-
mentary rocks.

{c) Waterberg System.

On the farm Kruidfontein No. 47 occurs a mass of igneous rock, consisting of felsite and
volcanic breccia, which occupies a synclinal area and overlies the banded ironstones and Pretoria
Series of that farm.

These volcanic rocks have every appearance of being un conformably deposited upon the
denuded surface occupied by the rocks of the Transvaal System. They transgress in succession
over shale, quartzite, and diabase along their southern margin. They consist of two principal types.
The first is a fine-granied red felsite, showing no other macroscopic constituent apart from the felsitic
groundmasa than small crystals of fluorspar. Under the microscope the groundmass appears mottled
and indeterminate, the red ferruginous stain veiling its composition, which is apparently for the most
part quartz and felspar. There is abundance of Eaussurite, and weathered magnetite crystals are
plentiful, while fluorspar in large grains b present throughout the rock.

The other type is a red breccia consisting largely of felspar. The fragments vary in size and
consist principally of fragments of red felsite. Iron ores are seen in small patches throughout the
mass.

These igneous rocks are probably to be referred to the Waterberg Volcanic Series.

Felsite. — The felsitic rocks of the area to be described are associated with the Rooiberg Series,

The most extensive of these outcrops is that lying along the northern slopes of the Elandsberg
where the outcrop of the felsite overlying the Rooiberg Series occupies an area of some thirty to
forty square miles. It is very uniform in type. It is a blue rock containing small visible felspar
phenocrysts often of a pink colour. This rock becomes red near the surface, retaining its other
characteristics. The relationship of this felsite to the underlying sedimentary rocks has already
been discussed.

A rock of very similar type but with a pronounced flow structure and phenocrysts of micro-
pegmatite as well as felspar and quartz occurs interbedded in the Rooiberg Series above the basal
quartzites. It is only a few feet in thickness.

(d) Igneous Rocks.

1. Bushveid Igneous Complex.

Granite.
Norite.

2. Diabase.

120

1. Buahveld Igneous Complex.

Granite. — The gtanite in the area in question varies considerably in petrological character in
different parts, and is distributed in a somewhat peculiar manner.

North of the Elands River and west of the Crocodile it covers the whole of the area included
in this report as far north a? Ramakok's Mountains, which it skirts on the west. It is normal in type,
consisting for the most part of quarts, red felspar, and hornblende with a coarsely crystalline structure.

On the right bank of the Crocodile River the Red Granite occupies the greater portion of the
country between the river and the BoBhofT Mountains stretching north to the Sand River. On
its western boundary along a line coinciding roughly with the bed of the Crocodile River, but a little
to the east of it, it abuts against norite. The junction between these two rocks is not the normal
one usually found around the margin of the Bushveld Plutonic basin, but is a faulted one. This
fault ib the extension of that which cuts off the beds of the Transvaal System north of the Aspies
River and brings the Dolomite and banded quartzites against the Red Granite and Rooiberg Series.

On the north this granite area is bounded by rocks of the Pretoria Series from which it is probably
separated by a fault plane, which brings the various horizons of the Pretoria Series from the upper-
most quartzites to the lowermost in turn into contact with the Red Granite and cute them off
successively from west to east. The outcrop of this fault can nowhere be observed, though its
presence is indicated by the behaviour of the Pretoria Series along the granite junction.

The most probable explanation of the successive cutting off of the various horizons of the
Pretoria Series is by meaiiB of a fault. In this case, however, the down throw side is to the south.

On the east the granite is bounded by the quartzites of the Rooiberg Series, which form a dip
slope on the north-eastern face of the Boshof Mountains which are composed principally of Red
Granite. The contact here is undoubtedly an intrusive one, as already pointed out.

The granite of this area varies somewhat in type. That along the Boshof Mountains is a
hard dark red compact rock composed principally of a beautiful granophyric intergrowth of quartz
and felspar, while a dark green hornblende furnishes the ferro-magnesian constituent; sphene is not
uncommon in the rock.

Another type is found on the farm Olivienpoort No. 758, bounding the Rooiberg sedimentaries
on the north. This is a very coarse-grained Bpotted rock consisting of red felspar in which are larga
irregular blebs of quartz, while no other constituent is diBcernible.

Various types are found, the prevailing characteristic of which is the beautiful granophyric
intergrowths of felspar and quartz.

A narrow outcrop of granite separates the felsite which overlies the Rooiberg Series from the
Transvaal System to the north. This is usually of a very coarsely crystalline type.

Another band of granite separates the rocks of the Transvaal System from those of the Water-
berg. The outcrop of this band is very narrow west of Gatkop ; it broadens out to the south-east
of that mountain, and joins the wide expanse of granite to be found to the east of the area stretching
from the Hoekbergen to the south. South-east of Gatkop the granite shows various types, a local
phase being a blue grey variety with marked granophyric structure.

On the farm Olivienpoort No. 758 near Papagai's Kraal is a fine-grained felspathic granite
showing no visible quartz.

While the numerous dislocations which have broken up the present area to such a great extent
have served to bring to the surface various types of granite, at the same time the presence of the
different sedimentary rocks, most of which must have become detached from the main formations
to which they belong before the consolidation of the granite, has doubtless had a considerable effect
on the conditions of cooling of the igneous magma, and tended to produce divergent types.

NorUe.— Norite is only developed over a small portion of the area along the course of and to
the west of the Crocodile River. It appears to occupy its normal position with regard to the small
development of rocks of the Pretoria Series on the farm Kromdraai No. 692, and stretches away to the
west beyond the area included in the present report. It also overlies the Pretoria Series quartzites
which cross the Crocodile River at Wachteenbeetjedrift south of the Sand River, thus bearing a
normal relationship to the main development of the Pretoria Series in this part of the country. Its
junction with the Red Granite along the course of the Crocodile River is, however, most probably a
faulted one.

The rock in the hand specimen is a typical norite of medium grain. It consists of intergrowths
of diallage and hypersthene together with anorthite.

121

2. Diabase.

The rocks of the Pretoria Series in this area contain intrusive sheets of igneous rock. These
are usually of a diabasic type such as are found in the better known phases of the series, and consist
of hornblende with some pyroxene often intergrown and felspar of the bytownite or labradorite type.

On the farm Langrand No. 1004 occurs a type which approximates closely to a gabbro.

5. Economic Geology.
Tin- Deposits — The area covered by the Rooiberg Series is in parte richly mineralised, tin
being the most important economic product yet developed.

The tin occurs in two definite and contrasting modes -

(1) Associated with massive tourmaline in fissures and veins, the cassiterite occurring in
crystals disseminated throughout the tourmaline or in layers separated from each other by tourmaline.

(2) In fissures as a mixed ore without definite arrangement, consisting of cassiterite, haematite,
malachite, and azurite.

In occurrences after type No. 1, the tourmaline is always looked upon as an indicator, and
the tin is only found in association with it. At the same rime tourmaline often occurs in large masses
without carrying tin.

The whole of the area in which tin occurs in this neighbourhood is occupied by quartzites,
which are underlain by Red Granite, with an intrusive contact. The tin occurs in fissures in the
quartntes. These fissures undoubtedly extend to the underlying granite, and probably into it.

Intrusions such as that of the Red Granite are usually accompanied by outpourings of vapours
and superheated steam containing metallic Baits in solution. This outpouring of vapours some-
times takes place subsequently to the intrusion itself, and massive sedimentary rocks are often
metamorphosed and altered for distances of thousands of feet by these vapours. It is, however,
in places, where the surrounding sedimentary rocks have been broken and fissured that the vapours
most readily deposit their metallic content, the fissures serving as vents relieving the pressure and
affording a means of escape for the superheated vapours. As these become cooler by contact with
the comparatively cool walls of the fissures the various vein infUlings are deposited upon the surface
of the fissures.

1. The most important deposits associated with tourmaline in this area are those situated on
the farms Olivienbosch No. 939 and Haartebeestefontein No. 949, and known as the Rooiberg Tin
Mines.

The ore occurs in two sets of fissures running at right angles to each other in north-south and
east-west directions respectively. The country rock is a pinkish felspathic quartzite which dips at
an angle of 6 degrees in a direction north 50 degrees east. It is considerably altered and sometimes
resembles macroscopically an aplite or fine-grained granite.

The fissures dip approximately at an angle of 82 degrees, although individual lodes have a
varying dip.

The mineralisation is distributed over a considerable area, and the deposits have been opened
up at several distinct workings. In every case the ore is associated with black tourmaline, which
usually serves as an indicator on the surface, large outcrops of massive tourmaline being frequently
met with.

The fissures sometimes swell out and become broader in parts and may then dwindle to a
mere crack, with a thin seam of clay as infilling. As this widens the tourmaline and tin reappear.
The thickness of the ore-body varies in the different workings from half an inch to over four feet.
At No. 2 west shaft the fissure frequently measures six feet in width.

The central portion of the infilling in the more typical uniformly arranged portions is com-
posed of tourmaline and cassiterite, sometimes in layers, but more usually mixed up, having evidently
been deposited together.

Veins of tourmaline traverse the country rock in many places ; sometimes they are thin and
stringy, occasionally very compact when they resemble sheets or stratified rocks.

The arrangement of the ore in the various fissures differs somewhat in the different workings.

At the north end workings the deposits are true fillings, the minerals being arranged in a
regular manner, quartz and carbonates being succeeded towards the centre of the fissure by tour-
maline and tin.

At the east end old workings copper ores are common both as sulphides and as carbonates, the
latter being especially abundant at the 25 feet level, while the former are found at the 100 feet level.

122

Copper pyrites occurs here with cassiterite. The tin ore occurs here too in very rich patches
with massive stellate tourmaline. The ouklip, which is common at these workings, sometimes contains
a considerable amount of tin ore, the percentage varying from 1 per cent, to 25 per cent.

At the camp workings the assure itself is filled with tourmaline, of which the central portion
is quite barren, the ore being concentrated along the periphery. The country rock consisting of
sandstone is impregnated with cassiterite for two or three inches from the fissure. No tourmaline
has been detected in the country rock at these workings. Sometimes copper pyrites in considerable
amount is found as impregnation in the country rock, This impregnation of the surrounding sand-
stones is peculiar to these particular workings.

At the No. 2 west shaft, which is on a north and south fissure, the ore is rich and compact; the
average width of the fissure is four feet, the outside measurement being Biz feet.

2. The tin deposits on the farm Weynek No. 763 differ in their occurrence and nature from
those juBt described. Here there is a noticeable absence of tourmaline, and the ore occurs in the fissures
without any definite arrangement.

There are two parallel fissures which dip to the north-west at an angle of 16 degrees. The
dip of the quart rites which form the country rock is 10 degrees to the north-east.

The fissures are filled with a heterogeneous ungraded ore without parallel arrangement. It
consists of cassiterite, haematite, with sometimes malachite and azurite. The relative richness of tin
and copper respectively varies at different depths. When tin is abundant the copper ores diminish
in quantity, and vice versa. TbiB relative richness in copper and tin seems to occur alternately, and
is similar in occurrence to some of the Cornish mines which are worked alternately for tin and copper
as the ore changes in character.

The ore-body is as a rule very soft and easily worked, felspar and mica being common in a
decomposed state. The cassiterite occutb in small crystals disseminated throughout the lode, and
not in the country rock.

Underlying the ore-body in the westerly fissure and accompanying the lode is a band of blue
green known as the " shale body." It varies in thickness up to two feet, and sometimes overhangs
the lode, but is usually to be found between the ore and the footwall. It occasionally pinches out
and is sometimes traversed by small veins of cassiterite. Running through the generally soft ore-
body are sometimes to be found veins of hard cassiterite which die away in the main mass of mixed ore.

Frequently veins of quartz occur and sometimes leave the fissure and traverse the country
rock along its bedding-planes, usually eutting across and rejoining the lode, thus including portions
of the country rock. When this occurs the quartz often carries cassiterite.

The footwall of the lode is hard green quartzite, which near the surface becomes white and
friable, soft and sandy.

As the fissure is followed south from the manager's house the ore is always associated with
quartz and becomes more laminated.

The easterly or A lode has a thickness of from two feet six inches to one foot aix inches. The
ore is soft and friable as in the other lode, but directly bounded by the quartzite of the country rock.
The hanging wall of quartzite shows some alteration for a distance of some three feet from the lode.

Both these mines show evidence of ancient workings on an extensive scale, round stone
hammers being common in and around the workings, showing the indentations made by the gad heads.
At Weynek also iron gads were found and large earthen vessels which were evidently used for carrying
away the ore. In default of blasting materials, huge boulders were used to break up the solid rock.

A noteworthy fact about the old workings on Weynek is that the ancient miners mined for
tin, and left those parts rich in copper intact.

Work was being done on Doorndraai No. 56 and Onverwacht No. 241 and some of the
neighbouring farms in July, 1908, but was mainly of a prospecting character.

Gold.— On the farm Haartobeestvlei No. 178, in a working which was being opened for tin,
gold occurs associated with arsenical pyrites.

In a decomposed diabase veins of arsenical pyrites occur running irregularly through the
rock, sometimes in pockets and large masses, sometimes in thin strings. There is much mica in the
decomposed rock, and gold is also reported to occur in those portions where mica is most abundant.
An examination of surface indications failed to furnish any definite information with regard
to the mode of occurrence and nature of the deposit.

The writer is indebted to Messrs. Jameson and Harrington for much information in relation
to the above deposits.

VII.

THE GEOLOGY OF THE COUNTRY

BETWEEN ZEERUST AND

ZWARTRUGGENS.

By A. L. HALL (Omlogi.il).

CONTENTS.

1. Introduction.

2. Physical Features.

[a) Topographical,

(b) Hydrographical.

3. Structural Geology.

4. Descriptive Geology.

(a) General Remarks.

(b) The Dolomite.

(c) The Quartettes of the Pretoria Series.

(d) The Shales and Ironstone Bands of the Pretoria Series.

(e) Metamorphic Rocks.

(/) Rocks Intrusive in the Pretoria Series.

(g) Contemporaneous Igneous Rocks.

(A) Basic Rocks of the Bushveld Plutonic Series.

VII —THE GEOLOGY OF THE COUNTRY BETWEEN ZEERUST
AND ZWARTRUGGENS.

By A. L. Hall (Geologist).

1. Introduction.

The present report describes portions of the Marico and Rustenburg Districts lying along the Zeerust
railway from Zwartruggens Station westwards, and is based on field-work, which forms a continua-
tion of the mapping completed in 1907 south of Rustenburg as far as Tafelkop.

In the present year 798 square miles were examined and 470 miles of boundary lines traced.
The area ib bounded on the north by a line drawn roughly east and west through the Enzelberg, some
thirteen miles north-east of Zeerust, while eastwards it extends as far as a line drawn north and south
through Zwartruggens. The southern limit is that of the Zeerust Sheet (published separately from
this volume) and coincides with an east and west line running through the farm Rietpan No. 113
in the extreme north-eastern comer of the Lichtenburg District ; but west of the Great Marico River
mapping was carried south only as far as the Zeerust railway and westwards up to a line drawn
north and south through Zeerust (see map on Plate XX). In the south-western corner a few square
miles are included of the Potchefstroom and Lichtenburg Districts.

At two points mapping was extended for a few miles northwards so as to complete the
geological information regarding the Pretoria Series ; north-east of Zeerust the Enzelberg was
examined, while the valley of the Great Marico River was followed northwards so as to embrace the
whole of the Magaliesberg Quartzite and (ho lowermost portion of the norite zone. These small
areas, however, do not appear on the map on Plate XIX, accompanying this report.

Much additional information was collected regarding the influence of the Bushveld Plutonic
Complex on the Pretoria Series, and the depth, to which the resulting metamorphism can be traced.
Economically the area here dealt with is not of any great importance at present.

2. Physical Features.
(a) Topographical.

The distribution of low and high ground and the character of the scenery generally show a
direct and simple connection with the presence or absence of certain horizons of the Transvaal System.
Thus the country south of the railway as far north as the base of the Timeball Hill Quartzite is typical
high veld, bare of vegetation, excepting for small patches of short trees scattered here and there over
the rolling and monotonous tableland of the Dolomite. The latter formation gives rise to an enormous
expanse of bare and barren-looking country stretching for a great distance over the north-eastern
portions of the Lichtenburg District, but also embracing certain parte of the Rustenburg and Marico
Districts as shown on Plates XIX and XX at the end of this volume. The high plateau has no running
streams and is generally poor in surface water. Towards the north-east it is defined by the escarp-
ment of the lowermost Pretoria Quartzite, which here gives rise to a low but distinct range of higher
ground with a remarkably even crest line. Occasionally the low dip combined with the presence of
underlying softer shales gives rise to more or less completely detached kopjes of the Tafelkop type,
due to circumdenudation, aa for instance on Kwaggashoek, about three miles north of Mabaalstad
near the Jameson road.

The Timeball Hill Quartzite forms the transition from the high veld country to typical banken
scenery caused by the differential rate of erosion of harder beds in the Pretoria Series compared with
underlying softer shales. Hence from the crest line of the most southerly ridge of quartzite there
is a rapid and uniform fall towards the north-east and north down to the prominent open valley,
along which the Zeerust line runs west of Zwartruggens (see section Plate XV, Fig. 2). The floor of
this may be over 1,000 feet lower than the high quartzite ridge on the south. This great topographical
contrast is the most striking feature in the whole area, and is due to the low dip of the lower part

of the Pretoria Series, which gives rise to a very wide and regular dip-slope locally as much as five
miles in width. The sandy nature of these slopes and their suitability to retain much of the rainfall
accounts for the well wooded condition of this strip of country, running parallel to and to the south
of the railway nearly up to Zeerust. Many farms situated on these quartzite dip-slopes are in con-
sequence well watered and famous for their fruit.

The striking valley just referred to is practically the only stretch of open country, and henee
forms the most important thoroughfare for all kinds of transport. Towards the east beyond Zwart-
ruggens Station it is confined by the high ground, on which the headwaters of the Elands River
take their rise, and along the railway line from Kosters Station to Zwartniggens there is again a very
marked fall of over 1,100 feet. Along the north Bide of and some two or three miles from the line
occurs a persistent belt of high ground known as the Schurvebergen, which forms a conspicuous feature
along the entire length of the railway from Zwartniggens to near Woodbine Siding on Schietkraal,
but owing to the effective work of the Great Marico River, the Elands River, and other minor streams,
Buch as the Tolani, which cut across the valley from south to north, the range is not very regular,
receding now and then northwards for several miles. This high ground is due to the great develop-
ment of an amygdaloidal contemporaneous flow, the same as that described in connection with the
easterly continuation of the present area, and elsewhere referred to as the " Vesicular Diabase,"
while the steeper slopes facing southwards are mainly composed of the underlying chiastolite-slates.
The highest points of the range may be as much as 800 feet above the average level of the valley to
the south, and on account of this topographical barrier, communication by means of heavy trans-
port between the railway and the more northerly portions ib only possible along the Marico and
Elands Rivers. Since the formation is igneous and moreover very thick, the crest line of this high
tract is succeeded by a stretch of hilly, rugged, and generally broken country between two and five
miles in width, limited on the north by a fairly well defined escarpment, due sometimes to the
Daspoort Quartzite, and sometimes to a thick succession of underlying dark coloured quartzitic
and ferruginous flagstones. Along the rivers a great contrast is usually seen between the igneous
rock and the overlying sediment, so that black kranzen of the ferruginous beds become conspicuous,
as on Rustvoorby along the Elands River or on Rhenosterfontein along the Marico, while the Tolani
River cuts through them so as to form a very fine and deep kloof confined by superb kranzen, mainly
formed of these ironstones. Hence the escarpment of the Daspoort Quartzite is often quite incon-
spicuous, and away from the river valleys draining northwards the high ground of igneous rocks
sometimes merges gradually and almost imperceptibly into the overlying sedimentary formation.

Further northwards follows a second and wide dip-slope due to the middle Pretoria Quartzite,
well watered and wooded, It behaves topographically just like the lowermost one and forms the
transition to the low-lying BuBhveld country, occupied mainly of norite and terminated by the
Dwarsberg in the far north.

From the high ground near the base of the Daspoort Quartzite a very extensive view is obtained
over this Bushveld country, which resembles a vast monotonous plain, rising here and there into
wide and swelling " bults," due either to the Magabeeberg Quartzite or altered crystalline slates.
The very broad dip slopes fall gently and steadily towards the north, are even more densely wooded
than those of the Timeball Hill Quartzite, and carry a type of vegetation characterised by the absence
of the Acacia family and prevalence of evergreen trees.

Often the base of the Daspoort Quartzite is only faintly marked, since it rises from an already
elevated shelving plateau of ferruginous flagstones, and in the neighbourhood of the Enzelberg it
becomes almost lost topographically. This prominent landmark forms a most conspicuous feature,
visible for many miles. Its highest points are due to a small capping of Magaliesberg Quartzite,
underlain by several hundred feet of hard dark-coloured crystalline hornfels, to which the ateep'slopes
facing southwards are due. The whole of this striking kopje rests on an almost level platform of
Daspoort Quartzite.

The main road leading northward from Zeerust through Sephtons Nek illustrates also very
clearly the evident connection between the geology and the regular distribution of high and low
ground ; thuB the well marked low ridge on Rykvoorby, Doornrivier, and Nooitgedacht is a geological
continuation of the high range seen north of the railway line between Zwartniggens and Marico
Station.

(6) Hydrographtcal.

With reference to the water supply, the present area falls into three portions, two of which
contain a very poor supply, while the third lies between these and is well watered,

127

That part of the country belonging to the Dolomite and die ahaies below the Timeball Hill
Quartzite is practically without running water and the Dolomite area especially contains large dry
flats ; it is possible to travel in the adjoining portions of the Lichtenburg Districts for over twenty
miles without coming across water of any kind, and even the numerous pans are not to be depended
upon, for many of them are dry in winter or reduced to a single water -hole a few yards across. There
must, no doubt, be a considerable underground supply, for the perennial and powerful streams at the
various " eyes " of the Great Marico River bring up a supply that cannot have been derived from
the impervious Pretoria Series on the north, but must belong to the very wide catchment area of
the Dolomite. Bore-holes sunk in the latter have sometimes proved barren, and it ia probable that
the underflow is along more or less well defined channels, but there is little or no indication on
the surface of the presence of such hidden seams of water. One finds occasionally, as on Geyerspan
and near that farm, a sudden change in the character of the soil overlying the Dolomite, denoted by
a marked abundance of a particular weed, extending along a straight piece of ground of fairly uniform
width ; such stretches may be due to more fertile soil, derived perhaps from a dyke, along which a
greater portion of moisture may reach the surface. Elsewhere one notices small and well defined
patches of dense shrubs, for the presence of which in wide stretches of bare ground there seems no
apparent reason, unless they indicate the proximity of an underground supply. In attempting to
tap the latter, there is practically no indication to be followed, excepting such features as suggested
above. In the present area, as also further east, the junction between the Dolomite and the over-
lying shales of the Pretoria Series is an important water-bearing plane ; the " eyes " of the Great
Marico, especially the two on Bokkraal and that near Mr. Lombard's house on Rietspruit, issue at
points not far removed from this junction, and the last named spring is also fed by a series of smaller
eyes from the south. It is not necessary that the point of issue should be at the actual junction
seen on the surface for the underflow to be forced up, so long as the latter eventually strikes the
overlying impervious shales of the Pretoria Series.

Following this waterless tract on the north comes a belt of country, some fifteen miles wide,
from the crest line of the Timeball Hill Quartzite up to the top of the Daspoort Quartzite, over which
there is a fairly good supply, especially along the Great Marico. After the various " eyes " of the
latter have united, a powerful perennial stream results, working its way across the tremendous
succession of soft shales below the Timeball Hill Quartzite, and producing a highly dissected, rugged,
and mountainous country, gradually merging both east and west of the stream into more even and
less broken dip-slopes. From the "eyes" of the Marico, therefore, as far north as Marico Station,
the river traverses very fine scenery, further complicated by the highly sinuous course of the river.
It is possible, that the remarkable doubling back of the stream from Vergenoegd, Rondavelskraal,
and Koedoesfontein is due to a secondary deviation of the original line of drainage, which ran from
the sharp bend near the centre of Vergenoegd in a south-south-easterly direction towards Bronk-
horstfontein, where it was finally diverted through the cutting backwards of a tributary, occupying
the present position of the main stream on Koedoesfontein. There is nothing in the present dis-
tribution of harder and softer rocks to account for the remarkable degree of meandering seen in this
part of the Marico, while south of the centre of Vergenoegd occurs a deep and long valley, difficult
to account for on the prevailing hydrographic conditions. From Marico Station onwards the river
now runs in a wide, beautiful, and fertile valley, enclosed by high kopjes of amygdaloidal andesite,
but becoming much narrower where the river cuts its way through the ferruginous flagstones and
overlying Daspoort Quartzite, after which it remains in the flat open Bushveld country, but running
during the winter months at the bottom of a deep trough-like channel. Perennial running water
continues to about the northern boundary of Schuinsdrift, but even here it is already so much reduced
as to require damming back for the purpose of feeding the water furrows; by the time Koffy kraal is
reached, the once powerful current has dwindled to the slenderest trickle, hidden away at the bottom
of a gorge-like channel twenty to thirty feet deep. Most of this loss is accounted for by extensive
irrigation along both sides of the river, from the Timeball Hill Quartzite downwards, while some of
the water may join the underground flow on reaching the higher quartzite. From the southern
boundary of Koffykraal northwards the bed of the river is either dry or interrupted by occasional
small stagnant pools. This condition continues for many miles northwards, as far as the Dwarsberg.
The Elands River shows analagous features, while the only other stream carrying perennial water is
the Tolani, the supply from which is even less reliable.

Underground water can usually be obtained in the Pretoria Series, but attention must be
paid to the distribution of the highly altered hard and compact bluish black crystalline slates (hornfels).
Experience has shown that such rocks are very " tight " hydrographically, and one finds that a

128

bore-hole may not strike an underground supply until a considerable depth has been reached, even
though it is sunk well below the water-level of an adjoining running stream. This is well illustrated
by the holes put down on the farm Zamenkomst, a few miles north of Bieckersdam, close to the left
bank of the Great Marico, where water was struck in hornfels at a depth of about 120 feet, i.e. well
below the water-level in the main stream. It is sometimes assumed, that because a hole is carried
to the water-level of an adjoining body of running water, it should strike an underground supply,
but in a sub-tropical country, where solid formation is often bo near the surface, the water of a
perennial stream does not denote that the underground water-table is intersecting the surface, as
generally holds good in European countries, but merely, that such water is running in a kind of channel
lined by impervious material and the level of permanent underground water may be at a considerable
depth below the bed of the river.

The wide and wooded dip-slopes of the Daspoort Quartzite form an extensive and effective
catchment area ; hence water is found at no great depth along the top of this quartzite, which is also
covered by much pale greyish yellow sand extending northwards a little beyond into the overlying
igneous formation. Th&se features are well illustrated by the hydrography of the locations known
as Fella (Zandfontein) and Moodjadjee {Grootfontein). Such richly wooded tracts contain some
excellent fruit farms, e.g. Teerputefontein.

The third portion, north of the Daspoort Quartzite, belongs to the great tract of level bush
country and has a very poor surface supply, while too little is so far known of the distribution of
the solid rocks below to judge its underground supply accurately. The great majority of the spruits
and other drainage lines shown on the Zeerust Degree Sheet are merely dry channels for conveying
storm water during the summer, though in view of the exceptional rainfall during the early months
of the present year a larger surface supply may now be available.

3. Structural Geology.

Little need be said regarding the geological structure of the present areas. It is quite simple.
Bince only the Dolomite and the Pretoria Series are involved and both dip conformably to the north-
east or the north and usually at a low angle, varying from 2 or 3 degrees near the top of the
Dolomite to 15 degrees near the Magaliesberg Quartzite. On account of the low dip the surface
width of the Pretoria Series is considerable when compared with its distribution near the Capital.
To appreciate properly the position of the present portion of the Transvaal System in the general
geological sequence along the southern edge of the central Bushveld, the country represented on
the Zeerust and Bustenburg Sheets must be examined together. It will then appear that between
Zeerust and Zwartruggens the Transvaal System runs along a short east and west strike, which in the
tatter direction beyond Zwartruggens turns sharply southwards, only to bend back once more towards
the east near Olifants Nek in the Magaliesberg, while beyond Zeerust it becomes again northerly or
north-westerly, previous to turning completely round to the east, so that the present area belongs
to the southern edge of a synclinal basin, filled with basic norites and pyroxenites of the Bushveld
Plutonic Complex, and defined northwards by the Transvaal System along the Dwarsbergen, where
the dip is to the south.

4. Descriptive Geology.
(a) General Remarks.

The sedimentary and igneous rocks may be arranged in descending order as follows :■

Norites, Pyroxenites, and allied rocks

Magaliesberg Quartzite

Crystalline Cordierite Hornfels with thin intrusive sheets

Daspoort Quartzite

Ferruginous Quartzitic, Flagstones

Contemporaneous Flow

Thin Quartzitic Conglomerate

Chiastolite and other altered Slates with thin intrusions

Timeball Hill Quartzite

Thick body of Shales

Thin Quartzite

Thin body of Shales

Cherty Conglomerate

Dolomite and chert

Bushveld Plutonic
Complex.

Pretoria

Series.

Dolomite,

129

(6) Dolomite.

ThiB formation is confined to the high veld portion east of the GlJat Msrico River as far as
Tafelkop, but forms merely the north-eastern extremity of a very extensive tract many miles wide, and
reaching far into the Lichtenburg District.

As a rule the only outcrops consist of loose angular blocks of whitUh blue chert, separated
by much deep red soil and rising here and there into low irregular hills or short ridges. Further down
in the succession true Dolomite begins of a similar character to that previously described in other
reports. No dip can be seen, but it cannot be more than a few degrees, to judge from that of the
immediately overlying shales. At the top of the formation, as in other areas further east, occurs
a well marked conglomerate a few feet thick, and exposed along the Jameson road on Bokkraal
No. 300, as well as further south-east on Kwaggasnek No. 603, near Tafelkop, and probably belonging
to a persistent horizon. The pebbles are slightly rounded and consist of hard compact greyish chert,
their usual size being about one inch in diameter, a few attaining three inches ; the matrix is made
up of impure, very hard, siliceous material. Immediately on the top of this conglomerate cornea a
thin horizon of brownish yellow sandy shales, well seen a little east of the Tra svaal Police
Post on Rietspruit, and these may be regarded as a kind of transition between the Dolomite proper
and the Pretoria Series, the base of the latter being drawn at the bottom of the conglomerate. Its
presence indicates a slight unconformity, and this agrees with the occurrence of Bevel's conglomerate
in the Pilgrims Best District and with a very similar horizon near Chuniespoort (see page 87).

(o) The quartzites of the Pretoria Series belong to the same three horizons, recognised so far
wherever th's series has been examined.

The Lower or Timeball Hill Quarlzite consists of a single band about sixty to eighty feet thick,
forming an escarpment facing south-west or south and succeeded by an extensive and persistent
dip-slope, giving rise to a strip of beautifully wooded country with olean and sandy soil. At Tafelkop
(5,374 feet)* the rock carries abundant granular patches of black iron-ore, standing out as little knobs
on the weathered surface ; further west and north-west the ferruginous character, elsewhere highly
typical of tins quartzite in the central Transvaal, becomes practically lost and the rocks pass from
dark reddish brown varieties into lighter coloured normal Pretoria Quartzites.

Between the main lower quartzite and the Dolomite is found another and much thinner
quartzite, separated by a thin succession of soft shales from the conglomerate band referred to. It
is only some fifteen to twenty feet thick, but remarkably persistent as a nearly continuous line of
higher ground from the Great Marico River through Kwaggasnek, south of Tafelkop, to about four
miles south of Magalies Station near KrugeiBdorp, thus extending for some sixty miles. In a westerly
direction it continues to Zeerust, where it forms the good exposure seen at the polo ground. The
succession from the Timeball Hill Quartzite downwards from the Marico River to south of Magalies
Station is perfectly normal and does not show any anticlinal structures whatever, such as have been
assumed to account for the presence of this quartzite at Zeerust. It must be regarded as an additional
quartzite, beginning somewhere east of the Steenkopjes district. Topographically it behaves just
like the higher one, though producing features on a smaller scale.

Both quartzites are practically unaltered ; the main bed north of Zeerust shows in thin section
some fresh green apparently metamorphic biotite, but further east no alteration is noticeable. It
will be shown below that the thermal metamorphism due to the Bushveld Plutonic Complex very
probably extends to below the main Timeball Hill Quartzite.

The middle or Daspoort Quartzite stretches roughly along the northern part of the present
area with an easterly strike, which beyond Lindleyspoort gradually turns to the south-east in con-
formity with the behaviour of the Pretoria Series as a whole. The dip varies from 4 to 10 degrees
to the north, while the formation does not give rise anywhere to a very marked escarpment, since
it is underlain by hard ferruginous flagstones which are already in sharp topographical contract with
the more easily eroded amygdaloidal andesite. Hence the Daspoort Quartzite forms as a rule merely
a shelf-like ridge on a plateau of ironstone rocks. Such features can be well studied south of the
Enzelberg and along the main road leading from Zeerust through Zendelingsplaats to Rieckersdam,
especially over the south-western corner of the last-named farm, where the common beacon of
Eromellenboog, Rhenosterfontein, and Rieckersdam stands on a small isolated " tafelkop" capped
by the quartzite, which here dips at a low angle to the north. A very good section across the middle
horizons of the series occurs along the stream which drains the farms Qatwater, Karriefontein, and
Kortkloof. a few miles east of the Great Marico River. The same succession is shown here, that is

* Trigonometrical Survey level.

130

from 80 to 100 feet of quartzite giving rise to a sharp feature on both sides of the valley, underlain
by soft darker coloured andalusite-slates weathering red, which in turn rest on black ferruginous
quartzitic flagstones causing kranzen further south. Unlike the Timeball Hill Quartzite the middle
one is markedly altered by thermal metamorphism and produces types very different from the unaltered
rock to the east of the present area. In the Lindleyspoort neighbourhood it in sometimes a double
quartzite separated by thinner bands of sandstone. Usually the Daspoort Quartzite is softer and
more deeply weathered into wide sandy flats carrying a deWe^ vegetation. The colour of the fresh
rock is creamy yellow or greyish white and free from ferruginous admixtures.

The Uppermost or Magaliesberg Quartzite was located on the summit of the Enzelberg and along
the Great Marico section on the farm Schuinsdrift No. 363 ; at both these localities it is intensely
metamorphosed either into an extremely fresh bluish white glassy quartzite, or else into almost black
very hard and glistening varieties, weathering with a rusty brown surface and showing no longer
any resemblance to a normal Pretoria Quartzite. The uppermost horizon has become much reduced
in thickness by the time it crosses the Marico River and makes only a small insignificant feature,
indicated along both sides of the river by swelling sandy bults. In the Enzelberg the great thickness
of hard crystalline hornfels underlying it has given rise to a prominent tract of very high ground,
of which the quartzite forms the two highest peaks, seen for many miles round. The latter are
cappingB due to small outliers of the main quartzite, which begins again on the northern dip-slopes
of the Enzelberg. The thickness of this quartzite probably does not exceed 100 feet.

Details regarding the metamorphism of the two upper horizons will be given below.

[d) The shales and ironstone bands.

True shales of the type so abundantly represented in the section west of the Capital are con-
fined to horizons below the Timeball Hill Quartette east of the Marico River. West of this few
unaltered argillaceous rocks are found, while in horizons still further to the west the shales are more or
less profoundly altered down to the Dolomite. In the nature of the case no hard and fast line can
be drawn between the two.

Pefcrographically they closely resemble analagous beds further eaat, and consist of a great
succession, over 1,200 feet thick, of uniform and thinly bedded soft greyish or yellowish shales, with
sometimes slaty characters. A few varieties are dark greenish or almost black. Their mode of
weathering and effect on the topography is likewise similar to that seen further east, but from the
Capital westwards the preponderance of shales in the lower horizons of the Pretoria Series gradually
becomes more marked, the increase in thickness being specially pronounced from the Steenkopjes
onwards as far as the Marico River.

Mention must specially be made of that particular horizon below the Daspoort Quartzite,
which is marked by ferruginous beds. These are dark coloured almost black highly ferruginous and
quartzitic flagstones, corresponding in position with the well known ironstone bands seen on the southern
slopes of Meintjes Kop at Pretoria. While it is here only a few feet thick and persists with the same
thickness for some miles further west, it rapidly thickens beyond Scheerpoort and north of Boons
Siding, until along the Marico River it is at least as thick as the overlying Daspoort Quartzite and
gives rise to persistent and striking kranzen, well seen along the Tolani and Elands Rivers. It is
on account of these that the higher quartzite often forms a quite subordinate feature in the landscape.
Such ironstones rest on the amygdaloidal contemporaneous andesite and are overlain directly by a
highly characteristic andalusite-slate with large and numerous crystals of that mineral.

The relationship can be most conveniently studied along the left bank of the Marico River,
where the main road crosses the stream on the northern portion of Rhenosterfontein No. 13, and
passes a large outcrop of dark, well bedded, fresh, fine-grained, ferruginous sandy rocks.

About a mile north-north-east of the common beacon of Bokkraal No. 300, Cyferfontein
No. 299, and Brakkuil No. 318 are the remains of an old battery, which appears to have been employed
to crush a reef interbedded in the top part of the extra quartzite, occurring a little above the base
of the Pretoria Series, or lying in the shales immediately above this quartzite. The reef belongs to
the quartz-bearing varieties, found at several horizons in the Pretoria Series elsewhere, and consists
almost entirely of milky opaque quartz with cavities due to oxidised iron-ore or pyrites, while thin
shaly partings traverse the reef parallel to the bedding planes of the country rock. No reliable
information could be obtained regarding its economic value. In general appearance it resembles
the Sterkspruit Reef near Lydenburg, but the latter belongs to a much higher horizon.

(e) Metamorpkic Rocks.

Between ZwartruggenB and Zeerust the Pretoria Series shows profound thermal metamor-
phism, the distribution of which must be studied in conjunction with descriptions of the adjoining

country to the cast, given in previous Annual Reports. The experience there gathered applies also
to the present area, but the further one proceeds west the greater is the depth to which the influence
of the great Bushveld Plutonic Complex is traceable ; this extends down to the base of the Daspoort
Quartzite along the Hex River section south-east of Rustenburg, and by the time the Great Marico
River is reached, the shales down to the top of the Timeball HiU Quartzite are affected, while finally '
south of Zeerust the metamorphism reaches the uppermost beds of the Dolomite.

The alteration of quartzites is seen in the Magaliesberg and Daspoort horizons, and naturally the
former is more strongly affected ; the two little peaks lying on the summit of the Enzelberg are due
to a very peculiar bluish white metamorphic phase, showing a very fresh glassy appearance and greasy
lustre. No minerals excepting quartz are recognisable with the naked eye, but thin sections show
a fair amount of fresh sericitic mica, possibly due to recrystallisation of the original matrix. Traced
eastwards this horizon in the north-western corner of Veeplaata is represented by a very peculiar
dark greyish rock well banded on the deep brown weathered surface and consisting in thin section
of rounded fresh quartz grains of fairly even size, surrounded by a very unusual crystalline matrix,
composed almost entirely of short delicate nearly colourless needles of ampmbole, resembling
tremolite ; the resulting appearance is like finely spun glass and the general structure recalls that of
the skeletons of some varieties of siliceous sponges. Probably this rock was less pure originally and
the matrix, under the influence of thermal metamorphism, became recrystallised in the manner
described.

The metamorphism of the middle quartzite is less pronounced and produces rocks which,
though widely different from the unaltered quartzites at Daspoort, near the Capital, can still be
readily recognised as a quartzitic rock. The little " tafelkop " a short distance west of the main
drift across the Marico River on Rieckersdam illustrates most of the varieties met with. The results
obtained further east and in the Lydenburg District in connection with altered Pretoria Quartzites
also hold good here; that is, thr rocks change into soft light coloured highly micaceous varieties.
The lower beds of the Daspoort Quartzite both east and west of the Marico River seem to have been
more impure and shaly, bo that they are consequently more highly altered with locally abundant short
chiastolite needles and a soft lustrous micaceous groundmass, whereas the highest horizons, as seen
on Kortkloof and Kroomellenboog, are less micaceous and do not show any metamorphic minerals
visible to the naked eye. Thin sections have muscovite or sericite abundantly developed in fresh
delicate needles aranged film-like round large grains of quartz or scattered in a few larger crystals.
The original matrix: between the grains has been entirely recrystallised, while tourmaline is once
more a constant accessory constituent in very small rounded idiomorphic deep bluish grains.
Numerous varieties of more or less highly altered Daspoort Quartzite can be seen at the southern
foot of the Enzelberg and close to the house of the field cornet on Witpoortje.

Metamorphic shales and greywackes give rise to an extremely variable group of rocks ranging
east of the Marico River from the base of the MagalieBberg Quartzite down to the top of the Timeball
Kill Quartzite, but further west reaching down to the Dolomite. No very definite line can be drawn
at any given point, as the increase in width of the metamorphic belt must in the nature of the case
be gradual. To what extent intrusive sheets have a share in bringing about such widespread altera-
tion will be considered below. Within this belt the Daspoort Quartzite separates fairly sharply two
widely different groups of altered shales, which may however emerge into one another at certain
places.

Above the Daspoort Quartzite, the metamorphism has resulted almost entirely in a single
well defined series of altered shales — the Qrooihoek type ; the precise meaning of this term has been
discussed in the last Annual Report and elsewhere. It embraces, briefly, all such rocks as are often
referred to as hornfels, but in the great majority of cases no horny and compact appearance remain?,
for the rock is entirely recrystallised with an evenly granular medium-grained texture ; the colour
is invariably dark bluish black, while the only mineral recognisable by the naked eye is biotite in
countless minute black flakes, owing to which the rocks have a remarkably-glittering appearance.
Along the Enzelberg section one occasionally finds thin partings showing andalusite and chiastolite.
The original bedding-planes are nearly always lost on the fresh surface of a hand specimen, but remain
in the banding of the brownish weathered surface, while in the field the rocks as a whole often have
the appearance of plutonic igneous rocks, strongly suggested by their fresh and holocrystalline con-
dition ; in larger outcrops stratification can always be made out in conformity with the dip of the
Pretoria Series. Since cordierite is a very common and most characteristic mineral in these contact
rocks, the designation cordierite-contactfels was used in earlier descriptions to replace the rather
indefinite term hornfels,

m

Mineralogically the rocks are characterised by the combination of biotite and cordierite, with
one or more of the following : — Andalusite, chiastolite, quartz, felspar, sillimanite, muscovite, gamet,
and tourmaline. Under certain purely local conditions, both amphibole and pyroxene may also be
represented. All rocks are thoroughly holocrystalline with medium or fine-grained textures, while
' often the intense metamorphism has obliterated almost all resemblance to the " porphyritic "
relationship, seen between larger metamorphic minerals and the slightly recrystallised groundmass
of shales which have been subjected to less intense alteration. Pavement and sieve structures are
universal and clearly reflect the intensity of the thermal changes. Biotite is by far the most abundant
and essential contact mineral, having been found in every altered shale of this type which has so far
been examined. As a rule it occurs plentifully in short stout blades, showing few cleavage traces,
deep brown colour, intense pleochroisnt, and the usual interrupted growth due to abundant inclusions,
sieve structure. Some sections show this mica in deep greenish varieties. Occasionally fairly
symmetrical structures are produced by the mica blades arranging themselves round numerous large
cordierite crystals. The latter mineral Tanks next to biotite in order of importance and is nearly
always present ; it is extremely rare that one can recognise it in the hand specimen, only a single
instance of thii being found. Where the hornfels has not been so strongly altered, cordierite occurs
in large irregular jagged plates occasionally slightly elongated and always colourless, scarcely
differentiated from the groundmass, but standing out strongly under crossed nicols, owing to the
innumerable inclusions with which this constituent is invariably riddled in a highly typical
manner, so that sometimes the latter make up a greater area than their host. When
such plates are very large so as to occupy the entire field of view under a magnifi ation
of fifteen to twenty diameters, the continuity of scattered remnants of cordierite can only be detected
with crossed nicolB. The well known pleochroic halos associated with some inclusions in volcanic
cordierite are wanting, but the biaxial character server to distinguish the mineral from
quartz, for which it may readily be mistaken. Where metamorphism has been more intense, as for
instance on the top of the Enzelberg, the mineral is much fresher and more transparent and also
associated with a coarsely crystalline structure, not unlike that of certain acid plutonic rocks, while
the inclusions are considerably larger, so that occasionally its identification becomes extremely
difficult, especially when it lies in a clear quartz-felspar mosaic. Of the remaining minerals andalusite
and chiastolite are quite subordinate and restricted to thin layers or partings in coarsely crystalline
rock, probably due to small variations in chemical composition of the original shale. The andalusite
stands out well as large colourlesB more highly refractive plates with a marked worm-eaten
appearance, due to inclusions. Quartz is invariably present and makes up the bulk of the recrystallised
fresh granular mosaic of the groundmass. Muscovite, garnet, and sillimanite are quite subordinate,
but tourmaline is very common though requiring sometimes much looking for, especially in the
presence of abundant brown biotite laths. It is always found in small amounts only, but occurs in
deep blue rounded six-sided grains, as well as in short stumpy prisms, which then show the peculiar
direction of absorption. The presence of this mineral both in altered shales and quartzite all along
the southern and eastern margin of the Bushveld is extremely significant of far-reaching thermal
metamorphism.

Practically limited to horizons below the Daspoort Quartzite, and extending along the Zeerust
section down to the Dolomite, are found metamorphic rocks much less altered than those described
above. They have been elsewhere grouped under the name Longsight type to denote shales and
slates or allied argillaceous rocks, which contain visible contact minerals, show distinct stratification
even in hand specimens, and at once reveal their original nature ; while the groundmass is not at all
or only slightly recrystallised. This group includes all the varieties of chiastolite and andalusite
slates, which may sometimes show very uniform characters through great thicknesses, as for instance
below the Timeball Hill Quartzite north of Zeerust, where a great body of them occurs down to the
Kareespruit level. Mr. Holmes* has suggested an origin due to " pressure metamorphism," but
reasons will be given below, why this view cannot be maintained. Chiastolite-slates like the above
are also very common along the Marico River between the top of the Timeball Hill Quartzite and
the base of the amygdaloidal flow ; above the latter, between the Daspoort Quartzite and the under-
lying ironstones, occurs a remarkably persistent horizon of andalusite slate, a soft brownish grey
rock crowded with abundant plates of andalusite and biotite. It has been also recorded many miles
further east in the neighbourhood of Steinbokfontein, north of Rosters Siding. In the present area,
the best outcrops of this variety occur on Eortkloof east of the Marico River, on Eromellenboog

* (1, O, Holmes.— The Pretoria Series in Marico District. Titos. Q.S.S.A., 1905, V 1. VIII, p. 16a,

133

West of the main drift across that stream, and over the southern portion of Lindleyspoort, two oi
three yards east of the new main road connecting that farm with Zwartruggens Station ; at the last
named locality the rock is very fresh and grey, containing also ottrelite in small bluish needles, while
elsewhere it is frequently deeply weathered into a light yellowish flagstone (see Fig. 5, Plate VIII).
Special mention must be made of an unusual variety of the Longsight type, containing abundant
fresh dark green lustrous ottrelite ; such a rock found on the farm Twyfelpoort, within a few yards
of its common boundary with Rondavelskraal, and about 200 yards south of the Sterkstroom. Its
position is at or just above the top of the Timeball Hill Quartzite. It is a hard compact dark greenish
quartzitie flagstone and weathers with a rusty brown colour ; scattered throughout its mass are
countless small flakes of ottrelite forming the only mineral visible to the naked eye. In thin section
it shows fairly fresh broad faintly pleochroic lath-shaped crystals, in a slightly recrystallised
shaly groundmass. The held relations of this rock indicate its metamorphism as due to a thin intrusive
sheet, which is seen in the Sterkstroom north of the main outcrop on Twyfelpoort (see Plate VIII,

The common chiastolite . or andalusite slates contain the contact minerals in conspicuous
slender needles up to three inches in length, usually scattered irregularly through the rock, but some-
times also more densely crowded along certain thin layers. The colour is slaty grey or dark greenish
grey to black; sometimes certain beds, apparently quite unaltered in the hand specimens, show ill-
defined short weathered chiastolites in thin sections (e.g. Roodekrans, south of Marico Station).
Microscopically, biotite is a constant constituent in numerous greenish crystals with lath-shaped
outlines. RecryBtallisation is naturally more evident along higher horizons, as it will be shown
that the influence of the intrusive sheets becomes gradually subordinate to that of the main body
of the Bushveld Plutonic Complex, the further the Pretoria Series is traced westwards. Unlike the
previous group of the Groothoek type, cordierite is not represented and tourmaline rare, but in the
andalusite slate below the Daspoort Quartzite aggregates of pale bluish short slender needles of
ottrelite were observed.

Reference may here be made to a very peculiar set of metamorphic rocks found near the base
of the ironstone band on Slypsteenkop, a little north of its extreme southerly beacon and at the summit
of the high ground east of the Tolani River. Another locality is on Olievenkloof, in the bed of the
stream running across that farm from south to north, close to one of its springs. At both places are
found lightish cream coloured quartzitie sandstones a short distance above the top of the con-
temporaneous andesite, weathering pale yellow, but showing a slate grey colour in the fresh hand
specimen. Crystals of andalusite up to two inches in length are seen in the hand specimens, and in
thin section these are associated with a finely crystalline quartzoae groundmass, crowded with short
delicate blades and needles of sericite.

Abnormal varieties of altered shales in some respecls resemblirg adinoles are found along the
right bank of the Sterkstroom on the farms Vlakplaats, Doordrift, Sterkstroom, Twyfelpoort, and
Rondavelskraal, at the base of a persistent intrusive sheet (see map on Plate XIX). Their outcrop is
marked by a faint escarpment rising quite close to the right bank of the stream and traceable for several
miles along the strike ; sometimes a second and thinner intrusion is found at the foot of this escarpment,
as indicated over the farm Twyfelpoort. The rocks intervening between the two sheets are compact hard
bluish varieties resembling hornfels, but less obviously crystalline than the Groothoek type of the
Enzelberg section. No minerals are recognisable by the naked eye, but under the microscope an
interesting and unusual mineralogical composition is seen, not easy to account for. This horizon forms
one of the few cases where the metamorphism can be beyond doubt attributed directly to an intrusive
Bheet, and hence the local restriction of such rocks becomes intelligible, compared with the enormous
areal extent both vertically and horizontally of the other varieties detailed above. The altered abnormal
types are characterised mineralogically by ottrelite, the presence of which was established at four
separate localities, in all of which the hornfels was found within a few yards of the intrusive sheets.
Thin sections show besides ottrelite, also biotite, sericite, cordierite, muscovite, and, in one instance,
granular pyroxene. The latter is probably due to a local transference of basic material from the
igneous rock, and this agrees with similar experiences both here and further eist, where a local enrich-
ment in amphibole and pyroxene was evident, giving rise to varieties to which the term pyroxene-
hornfels may be properly applied. The groundmass is always entirely recrystallised. The most
interesting point is no doubt the presence, sometimes in crowded masses, of ottrelite, since the
chloritoid group is often regarded as, and has been shown to represent, a typical product of dynamic
metamorphism. It forms short stout prisms with or without basal cleavage and with a faint or
strongly marked pale indigo bluish pleochroism. Multiple twinning is common throughout. But

134

unlike the ottrelite slate described above, no sign of the presence of that mineral is seen in hand
specimens. The clearest evidence of its relationship to the intrusive sheets is seen in an exceedingly
good outcrop of fresh rocks on Vlakplaats about half mile north of the railway line, at the point where
the dip is shown on the map. A little spruit coming from the escarpment of the " vesicular diabase "
on the north cuts a little gulley into the hornfels and into the underlying intrusion a few yards south
of the main road, exposing a sharply defined contact. The altered shale is here a very fresh
ottrelite-homfels.

Enough has been said to show that an extensive development of many variable types of
altered shale are represented between Zwartruggens and Zeerust ; it now remains to consider the
agents of metamorphism and their respective spheres of influence. Since the Pretoria Series has now
been examined from the Capital westwards to beyond Zeerust, the question as to how much contact
influence must be assigned to the main body of the Bushveld Plutonic Complex, on the one hand,
and to intrusive sheets on the other can be more satisfactorily discussed. It was shown in previous
Annual Reports, that at and immediately west of the Capital contact metamorphism is very feebly
developed and almost entirely confined to intrusive sheets, the influence of which does not usually
extend beyond a hardening and slight silification of the shales. At Scheerpoort and further west,
altered shales of the Longsight type begin to come in and are almost, certainly due to the same agency.
Crystalline cordierite-hornfels does not appear until some miles west of Scheerpoort and some of these
are probably due to the main body of the Bushveld Complex ; this is certainly the case along the
Hex River from Oliphants River southwards, for nearly the entire succession from the base of the
Magaliesberg Quartzite to the top of the Daspoort Quartzite is converted into a uniform variety of
dark coloured crystalline rocks, several times as great as the total thickness of die intervening
intrusive sheets. But the latter have a certain influence of their own as seen in the predominance
of ferro-magneBian constituents other than mica over a few yards along the contacts. Hence in this
area the influence of the latter is subordinate to that of the Bushveld Plutonic Complex.

Still further west, about as far as the Marico River, the latter agency affects the Pretoria Series
to a horizon reaching approximately to the top of the Timeball Hill Series, while the local influence
of intrusive sheets is confined to the beds close to their upper and. lower boundaries. The ottrelite-
homfels described above furnishes a good example of a type of alteration solely due to such intrusions,
while the great thickness of chiastolite and andalusite slates, exceeding that of the biggest sheet,
must probably be attributed to the influence of the main body of the Bushveld Plutonic Series. This
is particularly clear in case of the conspicuous andalusite slate lying immediately below the Daspoort
Quartzite, for west of the Elands River no intrusive sheet occurs at this horizon. It must be
remembered that the very thick contemporaneous flow found near this horizon can only have pro-
duced a purely local and subordinate metamorphism on the shales below it.

Finally, from the Marico River westwards, the intrusive sheets become even less important,
and both in number and thickness are quite inadequate to account for the production of some 1,000
feet of chiastolite -slate below the Timeball Hill Quartzite north of Zeerust.

Neither can Mr. Holmes' view be admitted, which regards these rocks as due to " pressure
metamorphism," for chiastolite is well known to be pre-eminently characteristic of purely thermal
metamorphism, and the succession at Zeerust shows no great tectonic disturbances. It is true that
between Jacobsdaal and Zeerust are seen several minor undulations, but these affect only the thin
extra quartzite of the polo ground and the shales underlying it, and are moreover quite unimportant
as factors in regional metamorphism. The core brought up from the new bore-hole near the Govern-
ment School at Zeerust shows the presence of two or three thin intrusions, and at their contacts
altered shales resembling hornfels, but this example only serves to show the impossibility of ascribing
to such sheets the uniform and wide-spread metamorphism so" strikingly shown throughout the
whole of the Lower Pretoria Series in this neighbourhood.

It is therefore concluded that the beds from the base of the " vesicular diabase " down to the
top of the Dolomite, mainly represented by chiastolite-slate, are metamorphosed by the Bushveld
Plutonic Complex, and form the outer belt of the contact aureole, of which the beds from the top
of the Daspoort Quartzite upwards form the inner or hornfels zone of more intense thermal meta-
morphism.

(/) Rocks intrusive in the Pretoria Series.

At various horizons throughout the Pretoria Series occur intrusive sheets of no very great
thickness and closely resembling in their mode of occurrence, topographic features, weathering, and
general petrographies! characters, analogous rocks described from the easterly extension of the present

135

area ; an account of these will be found in previous Annual Reports and very little need be added
here. No rocks of this kind were found in the Dolomite, and only two rather persistent but thin
sheets below the Timeball Hill Quartzite ; above the latter at least two were found, close to and
along the right bank of the Sterkstroom, the upper of which is the thickest of all and well seen along
the railway line from Zwartruggens to Marico Station, where it is sometimes associated with a little
escarpment due to harder overlying hornfels. Good exposures are also found in several stream
sections east of Rondavel Siding. This sheet attains a thickness of 150 to 300 feet. Above the
Daspoort Quartzite at least two thick sheets are found, usually much hidden by surface deposits ;
the one immediately overlying the quartzite is very persistent, extending from Lindleyspoort to near
the Enzelberg, and probably further ; it is very well exposed in the drift across the Great Marico
River on Rieckersdam, on which farm the contrast between the soil due to it and the underlying
quartzite is very striking. The higher sheet probably runs up into the southern face of the Enzelberg
and there gives rise to a clearly defined krans, seen as a sharp linear feature, even at some distance.

The lowermost sheets are very well exposed on Cyferfontein and at Mabaalstad, close to the
Jameson road. They consist of dull medium-grained greenish rocks made up essentially of augite
and decomposed plagioclaBe, the former sometimes in square plates with typical pyroxene cleavage.
Provisionally they may be classed as diabases with affinities to gabbro, on account of occasional
diallage. Above the lowest main quartzite are the two sheets running parallel to one another along
the right bank of the Sterkstroom, The upper one is well seen at Rondavelskraal and shows pro-
nounced opbitic structure, due to numerous lath-shaped plagioclase crystals with low extinction,
wrapped round by pyroxene, also seen in several long colourless needles passing at their edges into
pale green secondary hornblende. This diabase sheet is separated by a fair thickness of ottrelite
hornfels and other altered shales from a thinner and lower sheet, with the composition of hornblende
diabase.

Immediately above the Daspoort Quartzite lies the persistent sheet already referred to. It
is also well exposed in the bed of the Tolani River at the old drift on the southern portion of Vrede.
The rocks are coarser-grained and light greenish grey ; in thin section abundant hornblende with
plagioclase is seen, without ophitic relationship. The rocks may be classed as diorites. The same
name must also be given to the conspicuous intrusion seen on the southern slopes of the Enzelberg,
which is rich in hornblende.

(a) Contemporaneous igneous rocks.

It was shown in the last Annual Report that the most persistent and thickest body of igneous
rocks occurring below the Daspoort Quartzite is a contemporaneous flow and it was pointed out
how it follows a very definite horizon for many miles westwards from the Capital, and also in the
eastern Transvaal. From Zwartruggens to Zeerust this rock forms a continuous sheet of great width,
averaging some 1,200 feet in thickness, with characters usually the same over many miles along the
Strike, but showing great variations vertically. It forms the greater portion of the Schurvebergen
and causes a billy and rugged type of scenery, sharply defined against the overlying kranzen of iron-
stone on the north, while on the south a faint but well denned line sometimes forming a small krans,
denotes its base, and is caused by a remarkably persistent quartzitic conglomerate, first located by
Dr. Humphrey on Brakfontein along the Elands River, and referred to in the Annual Report for
1907. The contemporaneous rock is highly characteristic, owing to the presence of white amygdales
of quartz. These are not clearly confined to particular horizons, and it is possible, as Mr. Holmes*
points out, that there may be several distinct flowB in the belt, represented by a single colour on the
map (Plate XIX). Often the amygdales become arranged in oriented bands and drawn out into long
tubular structures highly suggestive of volcanic action. Such features are well shown on Gatwatcr
a few miles north-east of Marico Station. The colour 1b uniformly bluish or greyish green, but the
vesicular or amygdaloidal phases alternate with more thoroughly crystalline darker coloured zones.
The matrix or groundmass between the amygdales is finely crystalline or compact, giving rise to a
hard rough surface on a fresh fracture ; thin sections show large grains of quartz as the. main con-
stituent of the amygdales, but they are often crowded with abundant short slender needles of born-
blende, of which the more compact groundmass is mainly composed. This also shows isotropic portions
and others, resolved under crossed nicols into a very finely crystalline quartz mosaic. Owing to the
intense me ta morphia m, which has likewise affected this flow, secondary devitrification and recrystal-
lisation is to be expected, seeing how readily glass reacts to changes of temperature and pressure.

136 1

Hence isotropic areas are not so common as one might expect in such a body of volcanio rocks. The
most persistent and abundant ferro-magnesian constituent iB amphibole and the term hornblende-
andesite may he retained for this variable mass.

At several localities were found coarse fragments! rocks suggestive of volcanic agglomerates.
These may be seen on the main road leading from Marico Station northwards along the Marico River
on the southern portion of Doomkraal and a few yards east of the road. They belong to the lower-
most portions of the flow ; other localities are Welverdiend, about a mile north of the Klein Marico,
and Rykvoorby north of Sephtons Nek near Zeerust. At the last named locality Buch rocks may
show a fragmental habit only on the weathered brown surface, while the fresh rock shows scarcely
any signs of it. Along the Great Marico River the foreign fragments are subangular and up to two
or three inches long ; some appear to be baked shales ; the agglomerates are probably identical
in origin with similar rocks found at Pretoria in the railway cutting near Pretoria West Station.

Occurring in the amygdaloidal rock at some distance above its base are found small outcrops
of very hard bluish compact rocks behaving like hornfels in the field in showing a bedded appearance.
Their relationships to the associated contemporaneous rock is not clear. Good examples are found
in the eastern corner of Zendelings Plaata No. 69, north-east of Zeerust, close to the main road, and on
Gatwater east of the Marico River near some old kraals. The former variety in thin section closely
resembles a crystalline hornfels and contains large plates of calcite with very pronounced sieve
structure. The rock from Gatwater is also a fine-grained hornfels with a typical quaxtzose ground
mosaic, but containing much hornblende. Such varieties may be regarded as sedimentary inclusions
caught up and metamorphosed by the magma, which gave rise to the andesite ; or they may represent
original partings between two successive flows, but in either case the agent of metamorphism was
probably the same.

Finally, mention must be made of the thin quartzite with conglomerate bands which can be
seen to underlie the contemporaneous rocks for many miles, It seems to be first definitely recognis-
able in the Steenkopjes neighbourhood, but does not become a permanent feature until the Elands
River is reached. From here one can trace it westwards at least as far as the main road leading
northwards from Zeerust, and although not represented continuously along the strike (see Plate XIX),
it is no doubt persistent, but only clearly traceable, where the base of the amygdaloidal sheet runs
along higher ground. The best exposures are found along the Tolani River on the farm Tolaniesfon-
tein, about 100 yards east of the road leading northwards across that farm. It is also well seen all
along the southern face of the range of hills which runs along and to the north of the railway line
from Zwartruggens Station to the Marico River and beyond ; for example, on Sterkstroom and
Modderfontein. The quartzite is from ten to twenty feet thick and thus sometimes gives rise to a
definite feature in the slopes. At the very good first-named outcrop the rock is a fine-grained dark
greenish grey quartzite with conglomeratic layers containing well rounded pebbles of lighter coloured
quartzite up to three inches long, while some of the bedding planes show ripple-marking very clearly ;
this suggests, as pointed out in the Annual Report for 1907, the prevalence of a slight unconformity
and agrees with the mode of occurrence of the analogous vesicular rock in the Ohrigstad Valley of
the Lydenburg District.

On Tolaniesfontein the conglomeratic quartzite is seen to rest on greyish indurated shales,
otherwise practically unaltered. They are well exposed in the main road close to the little ridge due
to the conglomerate. On the farm Rykvoorby, north of Zeerust, and a little east of the main road,
the quartzite forms the upper portion of a well defined ridge extending both east and west. The
rock is here very compact and almost black in colour and causes much debris along the northern side
of the little stream flowing across Rykvoorby towards the east. This abnormal type is a highly
metamorphosed rock, originally quartzitic, but the matriz between the grains has been altered into
bright green hornblende and a little colourless pyroxene. These features are probably due to the
influence of the overlying basic flow, with perhaps a slight transference of material to the underlying
sedimentary rock.

(k) Basic rocks belonging to the main body of the BushveU Plutonic Complex.

The southern edge of die complex was met with on the Great Marico River near the southern"
boundaries of the farms Straatsdrift and Kopjeskraal, whence it extends northwards for many miles
about as far as the Dwarsberg. The rocks represented at the two first-named localities differ from
those usually found above the Pretoria Series in having amphibole as the main ferro-magnesian con-
stituent, but their general appearance and micro-structure is otherwise like that of medium-grained
norite. The colour is also greyish, but in the hand specimen a fairly distinct banding is seen, similar

137

to that shown at many localities, where the norite has been examined elsewhere. It is due to
consolidation under pressure.* Coarse light coloured veins rich in hornblende occur with these ;
also dark coloured and deep greenish peridotites are found on Straatsdrift, rich in olivine. These
belong to the group of rooks described by Dr. Hatch from the Marico District. t

VIII.

THE GEOLOGY OF THE SOUTH-
WESTERN PORTION OF THE
MARICO DISTRICT.

By W. A. HUMPHREY (Geologist).

CONTENTa

1. Area.

2. Physical Features,

(a) Surface relief.

(6) Drainage and Water Supply.

3. Geological Structure.

4. Geological Formations.

(a) Old Granite,

{b} Ventersdorp System.

(c) Transvaal- System.

1. Black Reef.

2. Dolomite.

3. Pretoria Series.

5. Economic Geology,

(a) Zinc and Lead.

{b) The Malmani Goldfield.

VIII —THE GEOLOGY OF THE SOUTH-WESTERN PORTION OF THE
HARICO DISTRICT.

By W. A. Humphrey (Geologist).

1. Area.

The area comprised within the limits of the present report includes the south western portion
of the Marico District and a portion of the north-western section of the Lichtenburg District.

It is bounded roughly on the north by the main road from Wonderfontein through Zeerust
to Pitsani, on the south by the 26th parallel of latitude, and on the east and west by the Marico River
and the Transvaal border respectively.

It includes the township of Zeerust, the Malmani Goldfields, and tne various mines and workings
connected with the lead and zinc deposits which are being opened up in this district.

In parts it is thickly populated, more particularly to the immediate south-west of Zeerust
itself, along the sandy well wooded slopes lying to the south of the main road from Zeerust to Wonder-
fontein No. 41, along the valleys of the Great and Little Marico Rivers, and along the courses of the
Malmani and Molopo Rivers.

Large stretches of the country are practically without inhabitants. Taking the stretch of
country from the eye of the Malmani east-south-east to the source of the Marico and Dtukerfontein
No. 392, over an area of some 330 square miles, the only permanent water is at Wonderfontein No. 78
and Klipkuil No. 210. In consequence of this scarcity of water there are no homesteads, and
though apparently a good grazing country, with the exception of the farms mentioned, no stock of
any kind was seen.

2. Physical Featureb.
(a) Surface Relief.

Physically the country to be described may be divided into two distinct and well defined
portions, namely, that to the north-east of the area which is varied and broken and covered with
rocks belonging to the Pretoria Series, and that to the south and west which is on the whole flat and
monotonous, and is covered by the Dolomite to the south and the Ventersdorp System to the west.

These two zones are separated from each other by the very rough broken and picturesque
country formed by the basal quartzites of the Pretoria Series and the broken and contorted banded
cherts of the Upper Dolomite.

The greater portion of the inhabitants of the area live in the north-easterly zone, where water
is plentiful and irrigated crops can be grown, while for the most part the rest of the area is waterless,
except in the neighbourhood of the rivers Molopo and Malmani, and one or two large local springs
whose water after flowing for a short distance disappears again in the Dolomite.
Mt The noticeable feature of the north-eastern portion is the striking escarpment of the Timeball
Hill Quartzites facing to the south-west, and their long and well wooded dip-slope trending in the
opposite direction. This dip-slope is cut into and through by numerous rivers and spruits which
have carved out deep channels for themselves, forming fertile, sheltered, and picturesque valleys
which are thickly populated.

The soles of these valleys are covered with alluvial soil which can be placed under water from
the perennial streams which traverse them. The steep kloofs afford a sheltered pasturage in winter,
while the elevated and well wooded dip-slope and plateau above the southerly escarpment are cool
and bracing even in summer. Lying to the south of the range formed by this escarpment is a broad *
valley which from Zeerust eastwards to Rietvlei No. 5 forms the basin of the Little Marico River.
Further to the east this horizon is traversed by various spruits flowing from the high ground around
Zyferfontein No. 85 through steep and rugged valleys to the Great Marico River. To the north-west
of Zeerust this depression is continued westwards into Moilo's Location, dominated by the escarp
ment to the north and shut in by ranges of hills to the south,

142

From this valley the country rises as a general rule gradually to the south, passing over a dip-
slope of quarteite, from the edge of whose escarpment the rough and broken scenery characteristic
of the banded ironstone shales and banded chert of the upper beds of the Dolomite commences.

Sometimes, as in the more easterly portion of the area this escarpment directly overlooks the
Dolomite ; from Kafnrskraal No. 214 westwards, however massive kopjes and ridges composed of
banded ferruginous shales and chert interpose between this escarpment and the Dolomite proper.
This continues increasingly to be the case as the strike is followed to the west, where the banded iron-
stones and cherts increase in thickness and become more continuously massive. They here form a
high and continuous range equal in height and importance to those formed by the main quartzite
horizons of the Pretoria Series.

The southern and western portions of the area present features differing widely from those
just described. For the greater part it is covered by Dolomite, which extends in wide flats or gently
undulating contours, broken here and there by kopjes and ranges of white and blue chert. The
general level of this portion of the country is high, ranging from 5,036 feet at Welverdiend No. 249
on the southern boundary of the area to 4,500 on the farm Doornfontein No. 289. Probably the
average elevation of this portion of the area is about 4,800 feet.

There is a somewhat considerable fall down to the beds of the rivers Molopo and Malmani,
but the general level of the surrounding country is very constant.

About seven miles to the west of Ottoshoop rocks of the Venteredorp SyBtem take the place
of the Dolomite. These are for the most part of basic igneous origin and give a somewhat character-
istic aspect to the scenery. Over a portion of this area, namely, the more southerly, surface time-
stone is largely developed, but north of the Molopo wide extending grassy flats alternate with thickly
wooded bults, giving place a few miles south of the Ramathlabama Spruit to evenly wooded flat or
gently undulating country with a red sandy soil.

North of Ramathlabama Spruit the surface becomes more broken, granite kopjes standing
out in conspicuous groups. This type graduates further north into the sandy well wooded type of
country before mentioned, which is separated again from the wide monotonous flats which characterise
the outcrops of the Dolomite in the north-west by a small well wooded ridge formed by the Black
Reef Quartzites.

(b) Drainage and Water Supply.

The area is divided into three principal drainage basins: — (1) Great Marico River, (2) Molopo,
(3) Notwani. Of these the most important is that of the Great Marico. This river by means of its
tributaries, the Little Marico and Malmani Rivers, drains the whole of the central and north-eastern
portions of the area, while streams mouthing directly in the main river occupy the eastern borders.
The flats and undulating country of the south-east form the catchment area which feeds the strong
springs constituting the head-waters of the river.

The drainage basin of the Molopo occupies the south-western portion of the map and is con-
fined to a somewhat restricted area of Dolomite. The Ramathlabama Spruit further to the north
also belongs to this system and drains the sandy bushy country covered by rocks of the Ventersdorp
SyBtem. In the north-western corner of the area the spruits flow in a general northerly direction
into the Notwani River, which eventually finds its way into the Great Marico River. This system
of drainage, while independent in the area under consideration, really belongs to the basin of the
Great Marico River.

(1) The basin of the Great Marico. — The present area is bounded on the east by the Great
Marico River, and consequently only the western half of the upper portion of its basin is included in
this report.

The river rises in five principal springs of which that on Grootfontein supplies the most water.
This is the only one of the five which comes within the limits of the present report. It rises in a large
circular pool of great depth, whose one side is bounded by dolomite and the other by dolomite and
shale. The water thus probably issues on the dolomite. It is, however, certain that the impervious
shales overlying the dolomite at this point have the effect of Btopping the underground flow through
* the dolomite and of bringing it to the surface.

The bed of the Great Marico has a very considerable fall in the first few miles of its course.
This is doubtless due to the fact that the powerful springs which supply most of its water all issue
within an area of a few square miles, and the streams thus formed soon unite, thus supplying a body
of water whose erosive power is very considerable. A deep gorge has been cut through the Time-ball
Hill shale horizon, below which the valley widens, and rich alluvial flats border the banks of the stream,

143

The river receives comparatively little water on its left bank and no considerable tributary
joins it until, on the farm Zamen Komst No. 88, it is joined by the Little Marico. At this point very
little water is found in either river in the dry season.

The Little Marico River drains a large area. It rises on the farm Rietvlei No. 5 and Groote
Afdeeling, and flows in a north-westerly direction to Zeerust, receiving various tributaries on both
banks. Immediately below Zeerust it is joined by the Karree Spruit and turns to the north-east,
in which direction it persists until it joins the Great Marico.

Numerous fine springs rise in the range of hills which trends due north and south to the south-
west of Zeerust, of which the largest are those on Vergenoegd No. 3 or near the common beacon
of Vergenoegd No. 3, Uitvlucht No. 63, and Wolvekoppies No. 104. Others rise on Klaarstroom
No. 55, Stinkhoutboom No. 269, and Buffelshoek No. 264. All these issue on the dolomite some
distance from the junction with the shale. They probably owe their existence to the presence of a
persistent horizon of chert, which is very prevalent over that area, and which would act in the same
way as the shales upon the underground flow of water.

The springs on Rhenosterfontein No. 50 while on the junction between shale and dolomite
are also on the line of a local fault plane.

Of the streams which flow into the lower reaches of the Little Marico from the north, most
of them rise in the upper portions of the Timeball Hill shales and flowing to the north receive affluents
from the broad horizon of the Timeball Hill Quartzites. All these streams, while perennial in their
upper reaches, lose their surface flow in winter before reaching the Little Marico. The water is
largely used for irrigation purposes along their upper reaches in the dry season.

Of the water belonging to the Marico Basin only that of the Malmani remains to be con-
sidered. The Malmani River rises some eight miles above Ottoshoop. The discharge in 1906*
measured 5,000,000 gallons daily. This water is most probably brought to the surface by the most
easterly of the quartz veins which traverse the Ottoshoop Goldfields. This reef has a continuous
outcrop from the farm Windheuvel No. 205, through the intervening farms south-south-eastwards
to the farm Naauwpoort No. 102, and the direction of strike if continued for a distance of about a
mile from the point where the distinct outcrop disappears would bring it directly to the above-
mentioned spring. It is thus moBt probable that this reef has the effect of damming back the water
flowing underground through the Dolomite from the east and bringing it to the surface. Should
this reef be a continuation of that to the south-east found on Wonderfontein No. 78 and Elipkuil,
then most probably the water issuing and then disappearing again on Wonderfontein No. 78 is part
of that which forms the source of the Malmani.

This water flows past Ottoshoop and spreads out into vleis below the village, where it disappears
on the farm Wonderhoek No. 192, slightly over a mile below the village.

On Rietpoort No. 95 the water reappears and forms a long vlei. Some of this is led in furrows
to Kamrkraal No. 93 and Faardevallei No. 62. The main stream, however, disappears into the vlei
on Rietpoort. The surface course of the Malmani from Rietpoort is through the farms Kamrkraal
No. 93 and Faardevallei No. 62, crossing the main road from Zeerust to Ottoshoop immediately to
the south of Commandant Botha's house. Some three miles to the east it joins the Little Marico River.
There is no flow, however, in this portion of the river bed, which is dry. The water from Rietpoort
evidently takes some other course underground. Some of the water must be lost by evaporation in
the numerous vleis between Malmani's Oog and Rietpoort, but a considerable amount should still
be left. The vleis along the course of this river are caused by the presence of numerous quartz and
reefs which traverse the Dolomite in this neighbourhood. Seven well denned reefs exist on the
southern side of the road leading through the village. As these are followed to the north their out-
crops disappear in the flats which near the river become vleis. These are undoubtedly caused by
a quartz reef damming up the river water a short depth below the surface at the end of the vleis.
From the Ottoshoop vlei to the point where the Rietpoort vlei ends the water probably flows in an
underground channel in the Dolomite between two reefs, being brought up to the surface at that
point, more likely by a reef crossing the bed of the stream lower down. The probability is that
several reefs cross the river below this point, although no definite outcrops are to be seen. Outcrops
were found further to the north-west, which are possibly the extension of the more persistent of these
Ottoshoop reefs ; and should this be so it is quite possible that the Malmani water follows an under-
ground channel to the south-west of these, and being prevented by these reefs from escaping to the
north-east into the basin of the Little Marico. This direction would eventually bring it to

* Rqxnt nf Mr, Karlsoti, Urn-eminent Hydrogrepbic Surveyor,

144

the Dolomite area south of Linokana (Tkalafyti) . It is probable that this water forms a considerable
portion of the supply which issues from the Linokana spring. On the other hand it is equally possible
should the reefs be intermittent that the Malmani water flows eastward and nourishes the springs
on Vergenoegd and Uitvlucht and Klaarstroom

In the basin of the Marico must be included the strong spring issuing on Wonderfontein No.
78. The water issues over the space of several acres where it saturates the surface soil and is drained
away in furrows forming a strong stream. After flowing about a mile it disappears again in the
Dolomite. An extensive acreage of ground is cultivated from this spring.

On Duikerfontein No. 128 is a channel in which after rainy seasons a strong stream flows for
several miles and continues to run until a continued drought sets in. The water in this stream is
sufficient to float a small boat. In dry seasons no water issues at all, and the farm is dependent
on wells. *Thus from 1862-1866 there was a large sheet of water on the farm and a strong running
stream. After 1866 it gradually dwindled, and in 1877 the spring ceased to flow. In 1890 the how
restarted and increased until 1893. Between 1899-1902 the water disappeared again, and in October,
1908, no running water existed on the farm. This appears to be a good example of a true intermittent
spring.

(2) The basin of the Mohpo is confined to the western portion of the district. The Molopo
itself rises directly in the Dolomite like the Malmani by means of a strong spring forming vleis. The
water is brought to the surface by quartz reefs which traverse the Dolomite in this neighbourhood.
The river flows with a strong stream over certain reaches and disappears wholly in others. The
Ramathlabama Spruit carries no permanent Burface flow, water being collected in large dams along
its course in the rainy Beason.

(3) The Notieani River. — The Notwani River is fed from a spring in the Dolomite on the
farm Tweefontein No. 64. It is probable that this water is brought to the surface by a diabase dyke,
of which a few boulders were found in the poort to the north. In a line from the poort, past the spring
along the road which leads to the location gate on the way to De Futten No. 4, is a narrow and per-
sistent band of thorn trees contrasting forcibly with the grasslands and occasional bushy kopje of
the surrounding country. This possibly marks the line of an igneous dyke and would afford an
explanation of the spring. No outcrops were, however, found south of the poort.

The other important spring in this drainage area is that of Linokana, which rises on the
junction between banded shales and dolomite. The spring in 1906 had a yield of 2,140,000 gallons
per diem, and the water at the present time, as was then the case, is dissipated by means of hundreds
of smaller channels which carry it along the streets of the large native Stadt of Ikalafyn or Linokana.
Mr. Karlsen, in a report dated November, 1906,+ supplies some interesting data
with regard to the levels of the various springs in the neighbourhood of Zeerust. The heights were
taken in comparison with a fixed aneroid in Zeerust, assuming Zeerust to be 3,960 feet above
Bea -level.

Malmani River at Ottoshoop 4,515 feet.

Rietpoort 4,472

Vergenoegd Spring 4,351

Karee Spruit 4,382

Tweefontein (approx.) . . . . , . 4,380
Linokana 4,292

It will thus be seen that the level of the water in the Malmani River, where it finally disappears,
is 90 feet above the level of the highest of the hillside springs which flow north and east from the
Dolomite, and 180 feet higher than the spring at Linokana, so that either of the explanations
given above with regard to the underground course of the Malmani water from Rietpoort is perfectly
feasible. The question really depends upon the continuity or otherwise of the quartz reefs from
Rietpoort No. 95 to the north-west.

Some interesting facts relating to water supply were recently disclosed by the borehole
recently put down by the Irrigation Department at the town school in Zeerust.

Small supplies of water are obtained from wells at the Transvaal Police Station,
and also at the Resident Magistrate's house near the school. This water was obtained at the junction
of the surface shale with an underlying diabase sheet. At the school, however, a borehole was put

* Information supplied by Mr. W. Froude. (■ root fori teiii.

t Report to the Irrigation Department Hnted 8th November, 19t»ti. "On the Zoerust Springs." by A, Kurisen,
tiovernment Hydrographii' "'

143

down and practically no water was found for more than five hundred feet, which is a considerably
lower level than the Karree Spruit. The rocks in the immediate neighbourhood of the school are
slightly folded, the axes of folding running east and west. It may thus be that the wells immediately
to the south of the school are situated along a synclinal axis on which is a flow of water, while to the
north along the arm of the succeeding anticline no water can collect. The two small faults between
which Zeerust lies in all probability divert the underground flow coming from the south and west.
If this be the case the count ry immediately to the south and west of these two faults should be favour-
able positions for boring.

3. Geological Structure.

The structure of the country under review taken on broad lines is extremely simple.

On a denuded surface represented on the west of the district by rocks belonging to the Old
Granite and its unconformably overlying Yentersdorp System, the rocks of the Transvaal System
were laid in order. Their outcrops stretch across the area mapped from north-west corner to the
south-east, and consist of Black Reef, Dolomite, and the lower portion of the Pretoria Series.

On the whole the detailed structure of the Transvaal System is not involved. To the south
of Zeerust a flattening of the dip, accompanied by folding, has taken place which has affected the
horizons below the Upper Timeball Hill shales. The thick diabase at the base of this horizon and
the quartzite and shales overlying the Dolomite have all partaken of this movement ; a well marked
anticline being developed, whose axis is roughly along the quartzite horizon running north and south
through the farm Vergenoegd No. 46 and along the western boundary of Kruisrivier No. 154. This
is followed to the west by a syncline consisting of the overlying shale and diabase, and the quartzite
again reappears in normal sequence on the farms Vergenoegd No. 3, Uitvlucht No. 63, and to the
south of the latter farm dipping normally to the east or north-east, and normally overlying the iron
stone shales and Dolomite.

The stress to which these rocks owe their folding has found relief further south in two small
dislocations whose outcrops run almost at right angles to one another through the farms Winterhoek
No. 287 and Rhenosterfontein No. 50. A similar pair of faults in the immediate neighbourhood of
Zeerust limits the folded area on the north. Of these, one passes northwards from Karee Spruit
through the western extension of the town to Sefton's Nek, cutting off the prominent ridge
of diabasic rock which crops out on the western side of the spruit which flows from the north into
the town. The continuation of this diabase horizon was found in the town school borehole, showing
that the side of the fault on which Zeerust stands is the down thrown side. The second fault referred
to passes more or less along the course of the Little Marico River, having an insignificant throw in
the Karree Spruit abreast of the town, but being easily and distinctly seen on the farm Kalkdam
No. 72, where the Timeball Hill Quartzites having formed a long dip-slope down to the river from
the heights north of Zeerust cross the stream and form a prominent bluff to the north of the main
road and between the road and river. Here they end abruptly against the fault plane abutting
against the underlying shales. Their continuation on the opposite Bide of the fault is found a mile
and a half further to the north-east in the spruit which flows into the Little Marico from Klein Marico
Poort No. 71, through the south-eastern portion of Kalk Dam. Outcrops of the quartzite can be
well seen where the railway crosses this spruit. These faults have a surface extension of only a
few miles and are thus merely local phenomena, and are probably intimately connected with the
folding of the strata of the lower Timeball Hill beds.

Previous writers have explained the presence of quartzites south of Zeerust by an anticline
of the Timeball Hill Quartzites, thus denying their existence as a separate horizon. This horizon,
which crops out normally on the crest of the escarpment overlooking Zeerust on the north, has, how-
ever, no connection with the quartzites south of Zeerust. These can be traced as an independent
horizon from the Great Marico River as far west as Linokana. Mr. Hall has noticed them as far
east as the longitude of Magaliesberg Station. South of ZeeruBt this horizon has experienced folding,
but these quartzites belong to a new horizon some 1,200 feet below the main Timeball Hill Quartzites,
and overlying the Dolomite by only some few feet.

The greatest abnormality in the district under review is the extraordinary surface develop-
ment of the Dolomite. From the farm Winterhoek No. 287 measured across the strike to the southern
edge of the map, Dolomite extends for some twenty-eight miles without a break, and continues
beyond the limits of the present mapping. In the north-western portion of the sheet the surface
width of the Dolomite is from seven to eight miles, measuring across the srrike of the series through
Linokana in Moilo's Location. From thiB line the surface development of the Dolomite increases
to the south-east. This is probably due to the south-westerly extension of the folding noticed at the
11

146

base of the Pretoria Series to die south of Zeerust. In the Dolomite this folding is impossible to
follow out, but a folding of the strata in a series of the thickness of the Dolomite in this area would
result in a great increase in its surface development, and this explanation is borne out when the over-
lying beds in this particular locality are examined and found to be affected in the same way.

The axes of the folds in the Pretoria Series run approximately north and south, and it ia along
the continuation of these lines that the abnormal development of the Dolomite is to be seen. The
lower beds of the Timeball Hill horizon increase in surface development across the strike from two
and a half miles west of Zeerust to seven miles of measured to the south of the town. This is owing
to the folding above mentioned, so that with a lessening of the intensity of the folding to the south
would result in a corresponding flattening out of the series and would explain an even greater surface
extension than that before quoted.

To the west of the Harico River on the farm Draaifontein No. 64 there are consistent dips to
the east over several square miles, showing that some local folding or dislocation has taken place.
Isolated patches of quartzite also occur on high kopjes, which can only be referred to the main Time-
ball Hill horizon. In a deep kloof on the north-western portion of Draaifontein No. 84 the lower
quartzite and its underlying Dolomite are exposed. From the dips this appears to be a denuded
anticline. The quartzite cappinga on the hills in the centre and east of the farm may thus represent
portions of the synclinal area. In this way their relatively low position when compared with their
normal theoretical position obtained by following the dip of the Timeball Hill Quartzitee into the
air is not difficult to explain.

4. Geological Formations.
The geological formations exposed in this area are as follows, in ascending order : — ■

(a) Old Granite.

(b) Ventersdorp System.

(c) Transvaal System.

(1) Black Reef

(2) Dolomite.

(a) Dolomite and Chert.
(6) Banded Ironstones.

(3) Pretoria Series.

(«) Timeball Hill Beds.

(b) Interbedded Diabase.

(a) The OUt Granite.

Granite rocks are confined in the area under review to its western portion, and are chiefly developed
along the course of the Ramathlabama Spruit and for a few miles to the north of it. The granite
varies in type. Close to the Transvaal boundary on the road south of the spruit is an outcrop of
granite, consisting of quartz, felspar mostly microcline, and muscovite. It is almost white in colour,
and appears a fairly normal example of the Older Granite. Further to the north and east, however,
the granite is of a very different type. It forms rugged kopjes, and petrologically cannot be dis-
tinguished from the granophyric portions of the Bushveld Igneous Complex. It varies somewhat
in grain, some outcrops being coarse, while others are fine-grained. All specimens are characterised
by beautiful granophyric intergrowths of quartz and felspar. Sometimes no other minerals are
present, when the rock becomes a true granophyre. Hornblende is the ferromagnesian constituent
in the more granitic outcrops.

The striking difference existing between this Red Granite and that found further down the
Ramathlabama Spruit, which belongs undoubtedly to Older Granite and is continuous on the Cape
side of the border, raises a doubt as to whether they are phases of the same "massiv" or whether
the red granite iB a later intrusion in the older rocks. Both thin sections and hand specimens of the
granitic types strikingly resemble specimens obtained from the Red Granite area in the neighbour-
hood of the Boshofberg and Elandsberg, while the gTanophyres are very similar to those obtained
by the Director of the Geological Survey at Crocodile Poort. This is an interesting point, and no
reason exists why the Red Granite of the Bushveld Igneous Complex should not be found in contact
with older rocks than the Transvaal System, such as the Ventersdorp System or Older Granite;
seeing that the Bushveld laccolite can only be a portion of the whole mass of the Red Granite, and
must have connection with possibly still larger masses of the same rock more deeply buried, which
acted as the reservoir during the intrusion of the Bushveld laccolite,

147

(6) The Venlersdorp System.

Rocks belonging to this system occupy almost the whole of the country mapped along the
Transvaal border. They consist principally of diabasic rocks, amygdaloids being well developed.
Porphyries and acid lavas are found along certain lines, the latter associated with tuffs and breccias.
Probably all these rocks fall within the limits of the Pniel Series as constituted in the Cape Colony.
Very little structural information is to be obtained from the development of the system in this portion
of the Transvaal, but some interesting penological varieties were met with. These rocks are divided
into two distinct outcrops by a development of granite which extends from Ramathlabama eastwards,
and which disappears underneath the Black Reef immediately to the east of the farm De Putten
No. 4.

Of these two divided outcrops the more southerly extends roughly from the Ramathlabama
Spruit to Rooigrond. For the most part this area is covered by basic igneous rocks of various types,
forming flat ground. Across this plain and stretching from De Egge No. 19 eastward to Klippan
No. 194 and thence southward on to the farm Buhrmansdrift No. 105 is a ridge which forms the
most noticeable feature in the immediate neighbourhood. It rises 100 to 150 feet above the surround-
ing country and is covered with bush of various kinds. The rocks forming this ridge are acid lavas,
tuffs, and breccias. They dip to the north and then along the southern extension to the east and
south-east, their outcrop forming an irregular semicircle.

The lavas are mostly ryolitea consisting of a very finely divided groundroass of quartz and
felspar grains with phenocryste of quartz. Cavities filled with volcanic glass are common. This
has sometimes been devitrified or replaced by quartz. In the hand specimen they vary somewhat.
On the farm Klippan No. 194 are found white and grey slightly banded compact cherty looking rocks.
They graduate into a dark grey, almost black, type which shows to the naked eye small quartz, crystals,
and occasional felspars as phenocryste. The rock presents the appearance of a blue chert on the polished
and weathered surface. Interstratified with these are tuffs, and a rock in which the only discernible
mineral is quartz in small blebs. The rest of the rock is aoft and weathered.

On the farm De Egge No. 19 these lavas are somewhat different from those just described from
Klippan No. 194.

The most common type is a blue rock which shows abundance of quartz-filled amygdales of
a blue devitrified groundmass. Under the microscope thin felspar laths and grains of iron ore are
seen in a groundmass which now consists principally of chlorite and calcite. The amygdales are filled
with quartz and calcite. The vesicules are frequently drawn out into long shapes and flattened.
Associated with this tuff is a massive volcanic breccia. The fragments range in size up to one foot
in diameter, and consist of qnartidte and jasper and basic igneous rocks.

To the south of this ridge of lavas are developed rocks of a more basic type which represent
successive igneous flows. They vary from holocrystalline to amygdaloidal. On Buhrmansdrift
No. 105 a pale green rock consisting principally of epidote is developed. Felspar is present and
cavities and amygdales are filled with quartz. The epidote occurs plentifully in the groundmass of
which it forms the bulk. It also occurs in larger individuals with rounded outlines. The rocks to
the south of the ridge above mentioned contain markedly amygdaloidal varieties. Macroecopically
the principal type appears dark olive green in colour, filled with vesicules which contain whitish quartz
or red jasper; sometimes the jasper forms the outer filling enclosing transparent quartz. Under the
microscope this rock is seen to consist of chlorite, for the most part probably the decomposition product
of either hornblende or augite. A rhombohedral carbonate is found throughout the rock.

When determinable felBpar occurs in these rocks, which is usually only in the non -amygdaloidal
varieties, it is either oligoclase or albite. Sometimes a colourless augite makes up a considerable
portion of the rock.

North of the ridge the rocks are less inclined to be amygdaloidal, although such are occasionally
found, the prevailing type being a close-grained green rock consisting of felspar and augite.

The second outcrop of rocks of the Ventersdorp System lies to the north of the granite zone
developed on the northern bank of the Ramathlabama Spruit. They consist of two principal types,
each occupying a well defined zone. The more southerly abutting on the granite consists of a blue
porphyritic rock. The groundmass is very finely divided and appears to consist of felspar with a
little quartz. Phenocryste of oligoclase and microcline occur in well developed crystals. Iron-ores
occur plentifully in the groundmass. This rock is a ceratophyr, a sub-section of the orthoclase
porphyries.

Along the bed of the more northerly spruit on Karreelaagte No. 81 this rock appear to be
sheared and shows apparent bedding planes, dipping to the north-east.

To the north of this porphyry zone is developed a close-grained green diabase similar in some
respects to that developed further south. It consists principally of felspar and chlorite with iron-
ores and titanite as accessories. In the outcrops of this rock there is an entire absence of amygda-
loidsl structure, and the rock has all the appearance of an intrusive rock. It cannot be classed with
the Venteradorp lavas and amygdaloids, all of which as at present known are contemporaneous
lavas and flows. While probably of different age from those diabases intruded into the Transvaal
System, it has been thought best to include it in the group of " intrusive diabases " for purposes
of mapping.

(c) Transvaal System.

Rocks belonging to this system cover the greater portion of the area mapped. They extend
from the north-western corner of the map to the south-east and are developed over the whole of the
eastern boundary. They consist of —

1. Black Reef.

(a) Quartzites.
(6) Shales.

2. Dolomite.

(a) Dolomitic Limestone and Chert.

(b) Banded Ironstones

3. Pretoria Series.

{a) Timeball Hill Beds.

(1) Lower Shales.

(2) Lower Quartzites.

(3) Upper Shales.

(4) Upper Quartzites.

1. Black Reef.

The lowest beds of the Transvaal System consult here, aa is usually the case, of quartzites and
shales. Their outcrops form a continuous and well marked ridge extending from the north-western
corner of the map in a general southerly direction. The northern portion of its outcrop forms the
watershed separating the R&mathlabama Spruit and its tributaries from the basins of the Notwani
and Halmani Rivers. Southeast of the Ramathlabama Spruit the dip decreases, and the outcrops
of the Black Reef cease to form a prominent feature. On the right bank of the Molopo exposures
occur, but for some miles to the north of the river no outcrops are to be seen. South of the river
the first outcrops are found about a hundred yards north of Rooigrond and continue to the south,

The quartzites are white and sugary in the more northerly outcrops, underlain by a fine and
compact reddish brown variety which shows bands of slightly coarser and lighter coloured grains.
Coarse gritty quartzites also occur. The thickness of the quartzites is probably about 30 feet in the
more northerly outcrops ; south of Zondagspan No. 44, however, the outcrops become very scanty
and no estimation of the thickness can be made.

The shales which overlie these quartzites are about 150 feet thick ; they are well bedded and
usually dark blue grey to black in colour. Occasionally they are pitted owing to the decomposition
and weathering out of some mineral, probably small crystals of iron-ore.

On the farms Klippan No. 194 and Zondagspan No. 44, a series of shales underlain by yellowish
quartzite occur below the Black Reef. They dip in conformity with the rocks of the Black Reef
and have been mapped in as part of the Black Reef Series. These shales are yellow in colour, slightly
micaceous and well bedded. Between them and the Black Reef was found a thin horizon of a felsitic
banded igneous rock of greyish colour. This is probably an interstratified tuff, and these rocks are
only developed locally on the above-mentioned farms, and "their position beneath the main
quartzites of the Black Reef is somewhat peculiar. At present, however they can only be referred
to the Black Reef Series.

2. Dolomite.

(a) Dolomitic limestone and chert. — This rock occupies slightly more than half the area to be
described. In the north-western portion of the area it has a surface width of between seven and
eight miles, and dips uniformly to the north-east at an angle of from 25 degrees to 30 degrees. As
the horizon is followed to the south-east, ite outcrop becomes much wider owing to flattening of the

149

dip and to a shallow folding of the strata. In the south-western corner of the area the dip of the
Black Reef and overlying Dolomite varies from 2 degrees to 5 degrees and is in a due easterly direc-
tion, while the chert bands which are found in the Dolomite along the southern and central portions
of its outcrops have dips which vary much both in degree and direction. This Dolomite area is on
the whole deficient in surface water, wide expanses being unoccupied owing to its scarcity. It serves
as the catchment area which supplies the strong springs forming the head waters of the Greater
Marico, Malmani, Molopo, and Notwani Rivers.

Chert kopjes and ridges are developed, being more massive towards the top of the series,
appearing however owing to undulations of the strata along a line running roughly parallel to the
northern margin of the scries through the farms Wonderfontein No. 78, Christina's Home No. 324,
Holpan No. 10, and Doornplaat No. 305. Thin bands of interbedded chert are found more or less
throughout the series.

Very noticeable in certain portions of the area are series of quartz reefs which cut through the
Dolomite. All these reefs agree approximately in strike, the direction varying from 20 degrees to
30 degrees west of north. There are two localities where these quartz reefs are specially prominent,
namely, in the immediate neighbourhood of Ottoshoop where they carry gold, and form the Malmani
Goldfield, and in the area to the south-south-east of the farm Molopo's Oog No. 286. They are also
found at intervals from Molopo's Oog to Duikerfontein No. 393, having the same general strike as the
others. Undoubtedly the springs which form the sources of the Molopo and Malmani Rivers owe
their origin to the action of these reefs in stopping the underground flow of water in the Dolomite
and bringing it to the surface.

It is impossible to say whether any of the reefs in the Ottoshoop area are continuous into the
Molopo area, joining up with the reefs there, as there is a considerable stretch of country of some
seven to ten miles in breadth lying between the two in which no outcrops occur.

It seems, however, likely that the most easterly of the Ottoshoop reefs, known locally as the
Koh-i-noor or Golden Calf, is continuous through Kaalplaats No. 97, and most likely extends to the
eye of the Malmani River. Possibly this reef is continuous underground, although no outcrops are
visible, and reappears again on Rlipkuil No. 210. If this were so it would to some extent explain the
great disparity in the yield of the springs at Malmani's Oog and Molopo's Oog, as it would cut off
all water to the east of it from the Molopo, at the same time leading it along the junction with the
Dolomite to Malmani's Oog. This water would include that which appears and after a few miles
again disappears on the farm Wonderfontein No. 78.

These reefs consist of fine white milky quartz for the most part, although zones occur where
manganese veins split up the quartz and form a network through it. Cavities showing fine quartz
crystals are common, and various chalcedonic forms of silica are present of yellow and amethyst
colours.

Associated with several of these white quartz reefs are companion reefs of yellow jasper which
accompany the quartz reefs for considerable distances. They dip from 60 degrees to the east-
south-east to vertical.

The upper beds of the Dolomite show a considerable amount of metamorphism, tremolite and
wollaetonite being found very abundantly in some thin sections. The metamorphism appears to
be most widely developed in the western portion of the area described, and certain extreme phases
of it occur in the neighbourhood of the lead and zinc deposits. The general metamorphism is
probably due to the same causes whose subsequent operation produced the ore deposits, and extends
uniformly over large areas. This fact indicates that it was caused by powerful regional agencies
acting from a distance uniformly upon great masses of sediments, while the more pronounced
metamorphism in the neighbourhood of the ore deposits was the result of the later and exhausted
phases of the same energy localised in particular areas.

Banded ironstones. — Overlying the Dolomite and often merging insensibly into the banded
chert horizons which occur in its upper beds is a horizon of banded ironstone rocks of great hardness.
They occur as a definite horizon in the north-west of the area and form the conspicuous range of
hills which marks the limit of the Dolomite to the south of Linokans.

They frequently appear to replace the massive chert horizon at the top of the Dolomite
altogether, as immediately to the south of Linokana, where they rest directly upon Dolomite, the
line between the two being a prominent escarpment facing south-west. Sometimes the upper beds
of the Dolomite are very cherty, either massive or intercalated in thin layers with the Dolomite.
In the latter case this banded chert and Dolomite gives place to chert and shale. As one ascends
the series the shale becomes ferruginous and the chert may become quartzitic, graduating into the

massive banded ironstones and cherty quartzites which are characteristic of this horizon. The
junction with the Pretoria Series has been fixed at the base of a thin band of chert conglomerate,
which represents the lowest bed of the Pretoria Series in this district.

(3) Pretoria Series.

The Pretoria Series as developed in the area to be described is Limited to its lowest beds,
namely, those of the Timeball Hill horizon. The following is the succession in ascending order : —

i Chert Conglomerate.
Ferruginous ShaleB -
Lower Quartzite.
Shales.
Upper Quartzites.
The chert conglomerate which has been taken as the base of the Pretoria Series doei not always
show an outcrop in the present area, although it occurs at frequent intervals showing the same
characteristics. It is a hard siliceous rock composed of rounded and oblong fragments of chert
embedded in a cherty siliceous cement. It is seldom more than from ten feet to fifteen feet thick and
frequently measures about five feet. The base of the Pretoria Series where this conglomerate does
not appear consists of blackish ferruginous shales usually of a banded character. These pass imper-
ceptibly into rocks in which banding consists of alternate ironstone shale and quartzite. This latter
gradually becomes cherty and the iron content of the shale diminishes, and the shale bands give place
eventually to Dolomite. The bands become thick and develop into beds and eventually normal
Dolomite occurs with the usual thin cherty partings.

There is thus in many places along the contact between the Dolomite and Pretoria Series
a perfectly gradual passage from somewhat ferruginous shales to typical Dolomite, and it is difficult
to draw a hard and fast line between the two. The presence of this conglomerate bed, however,
which marks a distinct change in the sedimentation if not an actual break is convenient as fixing
a definite limit to the Pretoria Series. This grouping includes the massive banded ironstones with
the Dolomite series, and in view of the fact that in northern developments of the Transvaal System
these ironstones are invariably found in association with the Dolomite, this appears to be the most
natural line of demarcation between the two series. In some places beds of chert conglomerate
occur in the Dolomite, but this particular bed is always overlain by ferruginous shales. Good out-
crops of this basal bed can be seen a few hundred yards to the west of the boundary fence of Moilo's
Location, some seven miles to the south-east of Linokana. From this point it can be seen at intervals
along this horizon to the Great Marico River. Mr. Hall has noted its presence for some miles to the
east of that river*

This horizon together with the ferruginous shale and quartzite which overlie it are not found
in the succession of the Timeball Hill beds in the Pretoria District nor in the eastern portion of the
Rustenburg District. The most easterly point at which they were noted is about the latitude of
Magaliesberg Station, where Mr. Hall refers to the quartzite horizon. From this point westward
they gradually become a definite and continuous part of the Timeball Hill beds, and increase in
thickness as they are followed westwards. The shale is black to blue in colour, somewhat micaceous
and uniformly well bedded. In the section along the Great Marico River they resemble shales of
the normal Pretoria Series type. Traced westwards along the strike they become ferruginous, and
on Winterhoek No. 287 definitely magnetic. Further west banding makes its appearance and they
assume a type somewhat similar to that of the banded ironstones which underlie the conglomerate.
They are, however, considerably softer and are shaly throughout.

The Lower Quartzite. — This quartzite shows a continuous outcrop across the whole of the
terrain. It consists of a hard white somewhat sugary quartzite underlain by a soft brown ferruginous
sandstone. To the south of Zeerust the beds immediately above and below have partaken of the
folding to which thiB horizon owes its double outcrop. At first sight this portion of the geology of
the Pretoria Series appears obscure and difficult, varying dips and repeated horizons rendering inter-
pretation difficult. With the help of the normal undisturbed succession, however, as exposed to the
west between Zeerust and Linokana, and to the east along the upper reaches of the Little Marico,
the reading of this portion becomes comparatively easy.

The quartzites exposed near the Zeerust Station and to the south of that point are not, as
suggested by Mr. Holmes,'!' a folded portion of the Upper Timeball Hill quartzite brought down by

i5i

means of an anticline from the summit of the escarpment north of Zeerust. The succession of beds
both above and below the said quartzites, differing entirely in character from those associated with
the upper quartzites, render such a reading of the geology impossible of acceptance. These quartz-
ites are folded portions of the lower quartzites and practically continuous outcrops can be followed
from Zeerust Station along the Karee Spruit to the west, also along the Ottoshoop road, through
Kameeldoorn, Uitkyk, Vergenoegd, and Uitvlucht to the south, connecting these Zeerust Quartzites
with the basal quartzite in its normal position. They outcrop on the farms Kruisrivier No. 154,
Vergenoegd No. 46, and Bufielsfontein No. 190, forming the axis of the anticlinal fold before mentioned;
thiB is followed to the west by a syncline where the beds overlying the quartzite are exposed. The
latter emerges from the syncune on the farms Vergenoegd No. 3, Uitvlucht No. 63 and the adjoin-
ing farms lying to the south. On these farms the quartzite crops out in its normal position overlying
the ferruginous shales and Dolomite. Its continuity is twice broken to the south by the two faults
mentioned before, but from the farm Rhenosterfontein No. 50 its outcrops are practically con-
tinuous to the eastern boundary of the area.

SAofes.— The massive shales which lie between the upper and lower Timeball Hill Quartzite
horizons in this district exceed in thickness all the rest of these beds put together. They form a
horizon of several miles in breadth into which are intruded several considerable sheets of diabase.
The shales are uniformly well bedded and are blue black, grey, red, and yellow in colour, the two
former colours predominating. Along the course of the Great Marico River they are deeply dissected
and form very broken country. On the farm Draaifontein No. 84 is an inlier of the lower quartzite
and Dolomite of small extent.

From the farm Cyferfontein No. 85 westwards these shales form the broad valley of the Little
Marico River. In the neighbourhood of Zeerust the shales occupy the face of the hills overlooking
the town on the north. They extend down to the Karree Spruit, but this does not include their fuD
thickness, but only the thickness as far as the top of the thick diabase intrusion which outcrops along
the Little Marico River above the town. They show signs of considerable metamorphism, chiastoli te
needles being distributed throughout their whole thickness. The question of the metamorphism
of the Pretoria Series has been dealt with in another portion of thiB Report* by Mr. A. L. Hall, and
it is sufficient to say that the uniform metamorphism of such a considerable thickness of shales must
be attributed to the wide reaching influence of the Bushveld Plutonic Basin, and not to local
intrusions, of which the thickest only measures 250 feet. He has also given reasons why this mete- >
morphism cannot be due to dynamic agencies.

To the west of Zeerust these shales occupy a wide depression bounded north and south
respectively by the Upper and Lower Quartzites. The igneous intrusions become less and less marked,
while the shales immediately underlying the Upper Quartzite occasionally become a true hornfels
of the Qroothoek type.t showing plentiful andalusalite in thin sections.

Upper Quartzites. — The Upper or Main Timeball Hill Quartzites occupy a broad and important
stretch of country extending right across the terrain.

They show a continuous escarpment to the south and an extended dip-slope to the north.
They consist of two main beds of massive white to red rather coarse-grained sugary quartzite.
Occasionally a thin intrusion of diabase was noticed about thirty feet above its base. As the outcrop
of this band is only a few feet thick and usually occurs on the escarpment face it was not mapped ; it is
in fact doubtful whether it is continuous over any considerable distance. Outcrops were noticed
on the farms Rietvallei No. 91, and Driefontein No. 90, and a well marked shelf or ledge is frequently
found along the escarpment, pointing to the fact that a layer of Borne rock with a resistance to
denuding agencies different from that of quartzite is present.

5. Economic.
(a) The Zinc and Lead Deposits.

Deposits of zinc and lead ores are found in this district extending intermittently over a con-
siderable area.

The great majority of these occurrences are situated in the upper horizons of the Dolomite,
and extend as at present known from Klaarstroom No. 55, at intervals, to Bokkraal on the right bank
of the Great Marico and a few miles from the river.

These occurrences have been known for some considerable time. The most authentic account
of their discovery places the time as in the early seventies when John Volschenk accidently

1M

discovered galena on the farm Rhenosterhoek No. 211. As a result of this, galena was mined at what
ia now called " The old lead mine " on Rhenosterhoek No. 211. The oie was smelted on the farm
and the lead sold for Is. 6d. per lb. Owing to difficulties connected with the supply of fuel and fall
in the local price of lead, work was discontinued. In the meantime galena had been discovered on
Kuilfontein and Doornhoek No. 32, but little work was done. After 1886 no work was done at the
old lead mine until the latter end of 1907.

The modern history of the lead and zinc mines commences with the opening of the Krugers-
dorp-Zeerust railway in 1907. Since then galena has been sold from mines on the farms Rhenoster-
hoek No. 211, Doornhoek No. 32, Euilfontein, and Buffelshoek No. 264, and zincblende from Witkop
No. 288* while prospecting has been carried on on almost all the farms along the contact between
the Pretoria Series and the Dolomite. In most cases galena and sometimes zincblende have been
found in varying quantities, but nowhere except on the four farms above mentioned has the work
been carried beyond the prospecting stage.

As before stated, the ore appears to follow the upper beds of the Dolomite as a general rule.
On the farms Strydfontein No. 267, Kaalplaats No. 97, Naauwpoort No. 102, and Oog van Mahnanie
No. 101 indications, and in some cases galena and zincblende, have been found, and these farms are
situated some miles measured across the surface from the contact between the Dolomite and the
Pretoria Series. It must, however, be borne in mind that the area in question has been subjected to
gentle folding, as has been shown above, f and it is thus quite possible that the true geological position
of the farms referred to may be one comparatively near the top of the series, which owing to low and
undulating dip appears at the surface at a considerable distance from the contact. It is thus by no
means certain that the occurrences on these farms do not belong to a geological horizon very slightly
removed from that of Buffelshoek No. 264 and Doornhoek No. 32.

It is not therefore considered advisable to divide these ore deposits into an upper and a lower
horizon, but to consider them as belonging all to the same system and caused by the same or related
agencies.

At the same time it cannot be denied that more southerly deposits, to take the Witkop Mine
as an example, differ in several important particulars from the northerly or upper belt. The latter
are chiefly deposits of galena invariably associated with soft manganese earth, and frequently with
a green soft mass of chloritic material, and may for convenience be referred to as the Doornhoek
, type. As contrasted with these, the Witkop type shows more constantly the presence of tremolite
needles in the Dolomite near the contact with the ore. This mineral may be so strongly developed
as to convert the Dolomite practically into a tremolite rock.

A short description of the more important known occurrences will perhaps be the best way of
bringing out similarities and differences.

Doornhoek No. 32.

This mine is situated in the Dolomite formation a few hundred feet only from its junction with
the Pretoria Series. The lower quartzite of the Timeball Hill horizons occupies the summit of the
bold escarpment overlooking the Doornhoek Valley on the north. The ore consists of galena
associated with cerussite, vanadinite, and pyromorphite. The latter occurs chiefly in greenish yellow
" mass like " aggregates, and seldom in normal well crystallised forms. The vanadinite occurs in
beautiful aggregates of ruby red crystals. Cerussite occurs well crystallised in cavities and clustered
on the galena.

The galena occurs in irregularly shaped masses varying in size from small pellets up to blocks
weighing several tons. These are usually covered with a thin coating of red oxide of lead, which
on assay is said to yield a small percentage of mercury. The galena is embedded in a soft black
manganese earth, which has undoubtedly replaced the Dolomite along certain lines. This is shown
by the fact that thin lines of chert traverse the manganese passing into the Dolomite. These bands
are identical with those traversing the Dolomite, and owing to their highly resistant character with-
stood the metamorphosing agencies which converted the Dolomite into manganese earth. Blocks
of Dolomite of varying dimensions occur in the manganese ground, the largest yet met with weighing
many hundreds of tons. These probably represent unaltered portions of the original Dolomite rock
which escaped the metamorphosing influence of the mineral waters, either owing to the composition
of the rock or on account of their situation relative to the fissures along which the waters were flowing.

* Up to Decemlier. 1908.

Aanua! Hfi»Tt, Geotogicul Surrey, Iim$.

Fir. 1. -TimeMII Hill Shales, overlooking Zeerust.

Pig. -1. Manganese ground, Doomhoek Mine.

Annvitl Report, Geological Surrey, 1S08.

Doom hoe k Lead Mine.

153

On the southern wall of the open workings at Doornhoek is a band of chloritic rock against which
the ore-carrying manganese earth ends abruptly. This is apparently a faulted junction, as the chert
bands traversing the Dolomite end abruptly against the chloritic rock. The whole mine shows many
signs of disturbance. According to information received at the mine the ore and manganese earth so
far removed have occupied an anticlinal position, dipping away to the east and to the west. The
general dip of the series in this neighbourhood is almost due north. Anticlinal folds whose axes run
due north and south would be due to stress acting in the direction of the strike, i.e. east and west.
This will be referred to later in a paragraph dealing with the probable origin of the deposits.

Slickensiding is met with in all parts of the workings. Owing to the exceedingly treacherous
nature of the manganese earth, which is unable to support the heavy masses of galena and to prevent
them falling into the drives, most of the work done so far on this mine has been in the nature of a large
open working, out of which the ore is trucked on inclines. An incline shaft was started in November,
1908, trending due east with a gradient of 1 in 2. The indication on the surface which marks the
outcrop of this manganese horizon is a band of contorted greyish chert eaten out into cavities and
running in wavy veins; the cavities contain manganese ores which also permeate the chert in
veins and patches. This band of contorted chert trends here almost due east and west.

Fig. 10.

The galena varies a little in character, but is mostly of a coarsely crystalline variety.

Along a line running from the mine to the south for some fifty or sixty yards small nodules are
found of a finely crystalline galena which is said to carry a good percentage of silver.* This type of
ore has not yet been located in situ The average output from the mine has been 175-200 tons
per month up to February, 1909, from the time of the first consignments.

Old Lead Mine, Rkenasterhoek No. 211.

This mine, which was the first to be worked for lead in this district is situated on the right
bank of the Great Marico River close to the northern boundary of the farm. The occurrence here
is in many respects similar to that on Doornhoek. The galena usually occurs in irregular lumps
embedded in manganese earth which has replaced the Dolomite, forming huge pockets and veins
in which the galena is distributed. The galena also occurs in veins running through the
manganese earth as well as in the more common fantastically shaped lumps. It is also found in veins
traversing the unaltered Dolomite and sometimes in fissures with a calcite vein matter. The mode
of occurrence of the ore is more varied in this mine than in any other in the district; a marked feature
is perhaps the occurrence of calcite as vein matter and in strings and leaders. Calcite veins are

* Assay in laboratory of Ueologicnl Kum-y gave 1 on. 2 <lwt

1 grfi. of si

154

locally considered an indication of ore and are followed where other indications fail. The ore varies
from fine to coarsely crystalline, the former being richer in silver than the latter. Cerussite occurs
fairly plentifully. The mode of occurrence of the ore a more easily observed than at Doornhoek, as
the great mass of manganese earth at the latter mine renders it difficult to observe its relationships
to the surrounding country rock, the work being carried on almost entirely in the manganese ground.

Buffelshoek No. 264.

Both zincblende and galena are found on the above farm. This occurrence is marked by the
presence of fiuorite and calcite as the minerals usually accompanying both the zincblende and galena.

Work had been discontinued on this farm before the deposits were examined in the month
of October, 1908, when a survey of the surface indications was all that was possible.

An outcrop of cherty somewhat contorted rock charged with manganese marks a zone where
galena is found associated with fiuorite. Small pits have been dug along this outcrop showing galena
in most cases, and two shafts have been sunk.

ROUGH SECTION THROUGH WITKOP MINE.

j

!

T"* 1 f ^

T

4
■T

Lor^

\

&

/AL CL Boundary waff of 7remo/iH.

\
\
\

1 )( ^*" b. Or* Sod,. (ZmeUtndt)

\
\

Iff C. Ore chamber.

\

f\

\

\
\

\

Galena also occurs further to the south-east associated with calcite. This latter splits along
the rhombohedral cleavage planes into regular forms and is milky white in colour. Zincblende is
found associated with fiuorite, and calcite occurs in considerable quantity.

Large masses of fiuorite are found whose relationship to the country rock is somewhat difficult
of determination. The strike of the outcrops is roughly to the east-north-east, that is practically
parallel to the strike of the Dolomite at this point.

The indications on this farm are extensive, but the relationships are involved, and can scarcely
be satisfactorily worked out in the absence of more detailed information.

Withsp. — In many respects the occurrences on the farm Witkop No. 288 are the most interest-
ing in the district. The main ore body in this case is zincblende, although galena has also been found
in small quantities.

The ore is confined within a more or less circular area defined by an outcrop of highly altered
Dolomite. The metamorphism has practically converted the Dolomite into a tremolite rock, which

155

outcrops as a alight ridge enclosing a more or less circular area measuring roughly 170 feet across
from north to south and some 160 feet from east to west. The tremolite rock which forms the walls
of a cone-Bhaped pipe is only a few feet in thickness, merging into the less highly altered Dolomite
on the outside. On the inside, however, it is usually followed by a band of zincblende of varying
thickness. The accompanying text figure illustrates roughly the method of occurrence of the ore.
The walls of tremolite at the 50-foot level dip outwards from the circle at an angle of 75 degrees, the
area included within the containing walls of tremolite thus increasing with the depth. The zinc-
blende which follows the tremolite hanging wall very consistently varies in width from four inches to
two feet, and is sometimes composed of a double layer of ore between which tremolite occurs. This
zone of blende sends off veins towards the centre of the circle. Whether these pinch out or lead to
large pockets and lenses of ore has still to be proved. By following a thick vein of zincblende leading
towards the centre of the circle at about the 20-foot level a large chamber of massive ore was found,
which was worked out and found to be of circular shape with no continuation {as far as blende is
concerned) towards the centre of the main circle. The walls of this chamber consist of very coarsely
crystalline calcite which breaks along the rhombohedral cleavage faces, giving well defined rhom-
bohedral outlines. This calcite is usually white in colour, but sometimes it is dark grey still retaining
its coarsely crystalline form. Embedded in. this calcite are masses of Dolomite which apparently
make up an appreciable part of the contents of the cone.

The zincblende is dark to golden brown in colour, while a considerable amount of black blende
is met with. An analysis of the ore by A. Heyman, Johannesburg* showed —

Zinc contents 60' 1 per cent.

Sulphur 310 „

In the western portion of the mine is a small development of galena which carries a consider-
able percentage of silver. Its occurrence is irregular and uncertain.

In addition to zincblende and galena the carbonates of zinc and lead are found in considerable
quantities together withf luematite, mispickel, marcasite, chalcopyrite. The average monthly
output from the mine since starting to ship has been some 75-100 tons.J

Galena and zincblende have been found on several farms in addition to the above described
occurrences, of which a full list will be given later on.

The horizon of soft manganese earth can be recognised in several places where up to the present
no ore has been found. Thus on Rhenosterfontein No. 83, a well denned band makes a persistent
horizon at the top of the Dolomite. This horizon can be followed into the farm Kuilfontein where
a considerable body of galena was taken out. The greater portion of the ore on this farm occurs in
irregular lumps in the manganese earth, but occasionally it is found in veins and in the Dolomite
without any vein matter. The metalliferous horizon here dips with the bedding planes of the
enclosing formation. Masses of Dolomite occur in the manganese zone here as elsewhere which have
escaped metamorphism. A peculiar brown rock consisting of manganese ore with a fine network
of cherty quartz giving rigidity to the whole is said to contain a high percentage of silver. § On
Uitvlucht No. 63 and Paardevlei No. 62 galena is found associated with cherty manganese not
far from the boundary of the Pretoria Series. Occasionally, as on Vergeuoegd No. 46, galena h
found associated with these quartz veins in shales of the Pretoria Series, while at Klaarstroom No.
55 it is associated with quartz in the Dolomite.

Conclusions. — From the foregoing account it will be seen that galena has been formed at
intervals along a line some forty-five miles in length, extending from the farm Klaarstroom No. 15 in
a wide curve to the eastern Bide of the Great Marico River. At every place where galena has been
found it is more or lees intimately associated with manganese ores. In many places where galena
has not been found the zone of manganese earth is still continuous, so that we have an almost unbroken
horizon of manganese holding rocks extending along the whole stretch of country above mentioned.
This zone occupies an almost constant geological position at the top of the Dolomite series, and in .
seeking for an explanation of the formation of the ores this is a most important factor which must be
kept constantly in mind.

The slight folding experienced by the overlying Timeball Hill beds of the Pretoria Series, where

* S. A. Mining Journal, 17th October, 1908, p. SIS.

t Report by Col. H. Bottomley, Deputy Inspector of Mines, Krugerslorp, to the Government Mining Engineer
llth November, 1907.

1 Dp to February, 1909.

S Assay in laboratory of Geological Survey gnv« I ilwt. .") grs. of silver |ier Lou.

156

anticlines and synclines are developed whose axes trend in a north and south direction, together with
the anticlinal structure in the Doornhoek Mine where the axis of folding also trends north and south,
point to pressure which has been exerted along east and west lines which has modified the normal
structure of the system. Added to this, if one follows the line of junction between the Dolomite and
the Pretoria Series, one cannot help being struck with continual abnormalities, more particularly
in the relationships which exist between the massive chert beds of the former and the lowest shales
of the latter.

In Borne places as on Kuilfontein the massive chert beds in the upper portion of the Dolomite
are practically missing, while on Knoffelfontein No. 150 and Bronkhorstfontein No. 33 they form
several parallel ridges. This abnormality occasionally extends to the beds of the Pretoria Series ;
thus on Grootfontein the lower quartzite of the Timeball Hill beds is duplicated over a short distance,
and further to the west still on the same farm has a break in its continuity. These facts appear
to point to the existence of series of small strike faults along the junction between the two series,
sometimes in the Pretoria Series, but more usually in the Dolomite. That this junction has suffered
considerable stress is evident from the frequency with which the banded shales of the Pretoria Series
and the chert bands of the Dolomite show signs of contortion and varying dips. This is especially
to be noticed on the farm Groote Afdeeling, where at the eastern corner of the farm the banded shales
and quartzites of the Timeball Hill bed have experienced a sharp double fold. On the farm Zen-
deHngspoBt No. 263, the banded shales overlie the Dolomite, dipping at high angles showing the result
of disturbance. These faults, owing to the nature of the Dolomite in which they occur, cannot be
traced, and their existence can only be deduced from the facts above mentioned. The upper beds
of the Dolomite have suffered alteration by the action of mineral solutions possibly following pneuma-
tolitic action, which have advanced along the junction of the Pretoria Series and the Dolomite. In
the Dolomite, after depositing the lead and zinc ores along lines of fracture or weakness, the action
has continued, gradually replacing the surrounding Dolomite with impure oxides of manganese.
This latter process has gone on longer in some places than in others, and probably represented the
last phase of the " post vulcanisch " activity to which the ores owe their deposition.

In view of the fact that the contact aureole of the Bushveld Plutonic Series can be shown
to extend in this district well into the Dolomite Series* the most reasonable theory would be to con-
sider it the source from which these ores were derived. Assuming the laccolitic nature of the Bush-
veld igneous intrusion, a theory put forward by Molengraaff and since accepted by geologists in this
country, one must suppose that at the time when the rocks of the Transvaal System sank in the magma
of the laccolite (now exposed over the Red Granite area and underlain by the Transvaal System),
a very considerable amount of dislocation and fracturing of the rocks belonging to that system most
have taken place. This is a necessary result of the stresses which are always exerted when a rigid
body such as the rocks of this system represent undergoes violent changes of position over large areas.
The destructive action of the igneous magma itself on the sedimentaries must have been very great,
moving in response to the intratellural agencies which were forcing it up from below.

That this fracturing of the Transvaal System was a fact is amply demonstrated by the numerous
isolated masses of quartzite shales and Dolomite which occur scattered sporadically throughout the
area covered by rocks of the Bushveld Igneous Complex. Such rocks occur to the north of Rnsten-
burg, at the junction of the Aapies River with the Crocodile River, also at the junction of the Elands
River with the Crocodile River. The latter is specially instructive, as it consists of a piece of Dolo-
mite, covering some five or six square miles in area which is entirely surrounded by Red Bushveld Granite.
Such occurrences have been noted at many other localities, and point to a wide spread fracturing
of the sedimentaries of the Transvaal System which are supposed now to underlie the Bushveld
Igneous Complex. Contemporaneous with and subsequent to the phase of igneous activity repre-
sented by these rocks, vast outpourings of gases and superheated steam, must have taken place,
followed by hydro thermal action. These products of volcanic energy had to find a passage for them-
selves to regions of less pressure, and the Dolomite, being peculiarly sensitive to and more easily
dissolved and carried away by vapours and heated waters than the more resistant shales and quartz-
ites of the rest of the system, formed the line of least resistance and became the channel by which
these mineralised gases and solutions found a vent. If to this be added the probability that a
system of minor fractures existed at the junction of the Dolomite with the Pretoria Series, ideal
conditions for the deposition of ores existed.

The zincblende deposits of Witkop are unlike those occurring along the manganese belt in

* A. L. Hall, |i, 134. Molcngniaff. — "Geology nt tl

157

many respects. They appear to be confined to the space within the tremolitic circle above described.
It is possible that the circular nature of the deposit and of metamorphism in the Dolomite is the
result of the thermal action having proceeded not so much from an extended fissure, but from one
point in a fissure, as the metamorphic changes around the edge of the circle are more or less uniform
at equal distances from the centre. This points to the fact that a line drawn down through the centre.
of the circle would represent the locus of the metamorphosing agency and the original source of the
ore- bringing solutions.

There is no reason for supposing that this occurrence represents a volcanic pipe or blow, no
volcanic minerals being found in the neighbourhood or in the circle, and the term "blow" so often
used is misleading.

It must be borne in mind that the Dolomite of this area shows metamorphic minerals, not only
near the known ore deposits, but more or less uniformly along the whole belt from Knoffelfontein
No. 150 to the westward, for some distance from the actual contact with the Pretoria Series.

That the galena deposits are due to replacement of the Dolomite along definite lines of fracture
is most probable, and the peculiar shapes of the galena go to support this theory. The manganese
moat certainly represents a further replacement of the Dolomite. In support of this one has the
fact that chert bands identical with those occurring in the Dolomite traverse the manganese earth.
Specimens of rock partially altered to manganese earth are not uncommon showing well defined
crystals of tremolite still embedded in the rock which have resisted the metamorphosing agencies
of the ore-bringing solutions (d, 408).

The general metamorphism shown by the upper beds of the Dolomite, while related to that
due to the immediate action of the ore-bringing agencies, cannot be attributed to them alone. It is
too general and uniform in character over large areas, and must be attributed, as shown in a previous
paragraph, to the wide reaching effect of the main mass of the Bushveld Igneous Complex which
acted through the whole thickness of the Pretoria Series, in this district reaching the upper beds
of the Dolomite. The metamorphism in the immediate neighbourhood of the ore deposits, while
due to the same genetic cause, is to be referred to a particular phase of that activity and not to its
general influence. The deposition of the ores and subsequent local metamorphism is most probably
to be referred to the period when the igneous magmas had ended their main aggressive, disturbing,
and dislocating phase, and had entered upon the era of gradual cooling and consolidation, their gases
and heated waters finding a vent through the channels already prepared by the previous violent
movements. This phase probably extended through long periods of time.

The thanks of the writer are dne to Messrs. Poynter, Froude, Raffaelowitz, Scott, and Ruslen
for information and assistance during the inspection of these deposits.

The Malmani Goldfield.

Under the above title is included the series of gold carrying quartz reefs which outcrops along
the course of the Malmani River from Malmani's Oog to Kaffirskraal No. 93. The village of Ottos-
hoop lies roughly about in the centre of these occurrences if the Kaalplaats mine be included on the
south. The number of reefs is greatest, however, at Ottoshoop itself and further to the north on the
farm Rietpoort No. 95.

The history of the Malmani Goldfields dates back to the year 1875, although little work was
done until 1876 when several reefs were worked, notably Kelly's Reef and Ladies' Reef. The latter
was very rich in parts, large nuggets being found. Other reefs were opened up and received local
names, such as Little Wonder, Pioneer, Mitchell's Reef, Elizabeth Reef, etc. At one time two
thousand miners are said to have been working on the fields. At the end of 1908 the numbers could
be counted on the fingers of one hand.

The gold occurs in quartz reefs which traverse the Dolomite in a direction north-north-west: —
south-south-east. This is practically the direction of Btrike of the Dolomite in this locality. The
dip of the Dolomite is low, varying from 5 degrees to 15 degrees to the east-north-east.

The reefs are fissure veins which dip at angles varying from 80 degrees to vertical, the former
direction of dip being towards west-south-west. Systematic mining does not appear to have been
carried on except on Kaalplaats No. 97, where a well timbered shalt has been sunk and some
driving done, and on Windheuvel No. _205. The other reefs have been worked open cast and appear
to have been abandoned when the water-level was reached.

There are seven principal reefs on the farm Rietpoort No. 95 and Stinkhoutboom No. 269,
some of which disappear to the south in thevlei north of Ottoshoop; the more easterly and westerly
reels, however, are more continuous. At Ottoshoop itself are several reefs which are undoubtedly

158

continuations of those occurring on Rietpoort, their continuity of outcrops having been broken
by the marsh lands and vlei bordering on the river. The reefs have been named locally and occur
from west to east as follows : —

Rietpoort.

1. Crystal Reef

2. Pillar Reef.

3. Daisy Reef (Greathead).

4. Green Cutting (Golden Butterfly).

5. Elizabeth Reef.

6. Magazine Reef.

7. Koh-i-noor (Golden Calf).

Ottoshoop.
Crystal Reef.

I Little William.
Colorado Reef.
Ladies' Reef.
i Little Wonder.
I Pioneer Reef.
I Kelly's Reef.
Mitchell's Reef.

PLAN

OF THE OTTOSHOOP GOLD REEFS.
Reefs ihcwn thih — — — -* "~^~—

*___* warn, e, rmt **<•

7 \ ***,„ 1 *»* /

""""-"-L #Y

MA \\

ELIN65P05
298

BUFFERS Hfltlt., / \

3 «* \ / \

KXALyJ^LAATS

/ w \-

/

/ \ V

/ PAARDEVfcLLEI \
/ 6Z \ \ ^

\^- WINOHEUWEL^' \ _^^\ OOORNPIAa/

X s

OUT BOOM \^^^__et»-=^ :: ^l3?- a1 t , " H<>UT -\ M/ ,-

^3_—->-<^v_. — - * \ -, eoow \ 2V /

^"WINTER HO EK ^Tt^- JWpWygTnT^Vri
\ / ,-•*! lEEKOEVAL^-EI

f 108 /

RIETPOORT

t-"—^**** - *^™™^ 1 \***s.

^^-V^T»06eiik;ht

D0ORN*FONTE1N \,

***tL / l8 °

■sag

_1 ^1 L ^-^^ Jl

/ W— —

Fig. 12.

These reefs differ somewhat from each other in character, continuity, and thickness.

The Crystal Retj is the most westerly, and sometimes attains a thickness of close on a hundred
feet. It consists of several distinct quartz horizons separated from each other by a dull opaque
yellow jasper. The quartz is somewhat sugary and weathers into round pebbles. It is white and,
in miners' parlance, " hungry." Visible gold occurs in several places along this reef and work has
been done here and there, but to no great extent. It is also known as the Witrantje. The barren
appearance of the quartz has resulted in comparatively little prospecting being done.

With regard to the other reefs it will be better in order to avoid obscurity to deal with the
series in order as occurring on the farm Rietpoort No. 95, and then with those which outcrop in the
immediate neighbourhood of Ottoshoop, merely mentioning the probable identity of pairs on the two
groups when evidence seems to point that way.

159

The Greathead Reef consists of somewhat coarse white quartz containing irregular cavities
which persist in layers running parallel to the Bide of the fissure. These cavities vary in size from
that of a good sized bean downwards. They are filled with acicular quartz crystals or soft manganese
earth and limonite. The quartz in the zones where these cavities occur is usually discoloured and
stained brown or dirty grey. This is probably due to the presence of numerous cavities throughout
the quartz of small or microscopic dimensions ; associated with the quartz of this reef is a thick band
of the yellow jasper noted as occurring in the Crystal Reef. This is part of the infilling of the fissure,
of which the dimensions are considerably in excess of those given above of the quartz reef proper.

The Green Cutting or Golden Butterfly is the name of a persistent reef situated on the farm
Rietpoort on the right bank of the Malmani, and is the one nearest to the river. It has been worked
down to the water-level over a considerable portion of its length, and is reported to have given values
equal to any on the fields. It consists of one main quartz reef from which a network of stringers
extend on either side, joining up with the main reef and with each other. This quartz is not solid
like the two previous reefs, but is a good deal perforated and full of cavities. Manganese and limonite
are constituents of the greater part of the reef, and occur in the cavities before mentioned, and in
thin wavy lines traversing the quartz, which gets distinctly chertyin parts. Figure 13 illustrates the
stringers in this occurrence.

SECTION ACROSS "MAGAZINE REEF"
0TT0SH00P.

Fig. 13.

The Elizabeth Reef appears at present to be one of the most promising in the neighbourhood.
It varies somewhat in character. It has eighteen inches to two feet of fine white quartz which
graduates to grey. The cavities in this white quartz are frequently filled with beautiful amethystine
quartz. The bulk of the reef, however, is of a somewhat varying character, stringers of vein quartz
alternating with portions in which the rock is honeycombed with negative crystals of iron pyrites.
In this type of rock visible gold in blebs is frequently seen, the general character of the rock being
dirty brown and earthy. Copper ores, more particularly malachite are frequently met with, and
these portions are usually rich in gold. This type of rock consists of pure white quartz with irregular
masses of malachite and a yellow ocherous mineral ; the quartz makes up perhaps half the bulk of the
rock. The yellow jasper occurs in this reef in patches among the quartz and frequently traversed
by veins of quartz. The Magazine and Moonlight reefs call for no special mention, being quartz reefs
with more or less vein matter filled with negative crystals, running in stringers of dirty brown earthy
rock.

The Koh-i-noor or Golden Calf, which is the most easterly of the Malmani gold reefs, is the most
continuous and among the thickest. From Kaffirskraal No. 93 it trends southward through Stink-
houtboom No. 269 and Windheuvel No. 206. Here it is interrupted, having probably been displaced

160

laterally by a email fault. Its continuation can be found slightly to the west of its termination on
Windheuvel, and from thence it runa without a break to Kaalplaats No. 97. On Windheuvel
No. 206 the reef is being worked. It here measures thirteen feet, of which three feet was running
18 dwts. to the ton in October, 1908. The continuous reef, which carries practically no gold, is
called locally " the Buck Reef," and at intervals along its length are lenses of quartz adjoining the
Buck reef, which carry gold. In such places the whole reef carries some gold.

On Kaalplaats No. 97 the width of the reef is eighteen feet. A timbered shaft has been sunk,
land a considerable amount of gold was taken out. In October, 1908, no work was being done
anywhere on the fields except at Windheuvel and intermittently at the Elizabeth Reef.

Of the reefs in the immediate neighbourhood of Ottoshoop, the two earliest worked were
Kelly's Reef and Ladies' Reef, the former on the right bank and the latter on the left bank of the river.
No work is being done on any of these properties, and very little information is obtainable about them.

The Pioneer Reel on the right bank of the river contains copper ores, both azurite and malachite,
and resembles in many respects the Elizabeth Reef on Rietpoort No. 95. A slightly banded
amethystine cherty quartz occurs along the Pioneer Reef. The copper bearing portions of the reef
are said to have been rich in gold.

The total yield of the Ottoshoop mining district for the year ended June, 1908, was 50~417
ounces of fine gold valued at £214,* which is eloquent of the fallen fortunes of this once busy goldfield.

Very little attempt seems to have been made, apart from the Koh-i-noor and Mitchell's Reefs,
to cope systematically with the difficulties of the situation. The reefs, many of which have paid
handsomely for a time, have usually been worked open cast from the surface, and abandoned when
the ground water-level was reached. After dry seasons the workings are all dry, but doubtless in
wet seasons considerable quantities of water would have to be dealt with.

The mynpacht on Rietpoort has recently been thrown open for pegging. All official records
of output, etc., perished in the war. From old residents it would appear, however, that the average
monthly output amounted to some 400 ounces, and that at one period £13,000 worth of gold was taken
out of the farm Windheuvel in fifteen months.

From information received since the district was geologically surveyed it would appear that
batteries are about to be erected on Kaalplaats of 10 stamps and a similar one on Windheuvel.
These together with the 10 and 3 stamp milk now running on Rietpoort, and the 5-etamp mill on
Syferfontein No. 299 should materially increase the annual output of gold from its present insignificant
figure.

The thanks of the writer are due to Mr. W. Dyke Poynter, District Registrar of Mining Rights,
whose assistance in locating the various disused reefs was invaluable, also for information and access
to official plans and diagrams.

* Annual Kejiort of tbe Government Mining Engine

IX.

ADDITIONS

TO

THE MUSEUM AND LIBRARY.

IX. ADDITIONS TO THE MUSEUM AND LIBRARY.

1. u

ST OF AD

IN COLLKCTrOt

DITIONS TO MUSEUM COLLECTION.

DONOR. NO.

DESCRIPTION.

LOCALITY.

J. M. Bowker

. IV. 448

Spinel (Pleonaste)

Hoogedoorn No. 2, Magala-
kwin River, Waterberg.

J. L. Grobler

. IV. 449

Corundum and Ilmenite. .

Zwankop No. 335,

Mapochsland.
D'Jelele Copper Syndicate.

W. J. Jaffray

. IV. 451

Molybdenite

J. F. Prellar

. IV. 452

Monazite

Bishopsberg, Waterberg.

G. Brown

. IV. 453

Caasiterite in Matrix

Enkeldoorn, Pretoria.

. IV. 455

Tin ore . .

Olievenhoutbosch, Rooi-

berg.
North Eaap River , Bar

Allan Molliaon

. IV. 456

. Chalcopyrite

berton.

W. Sonnenberg

. IV. 457

Vesuvianite

Town Lands, Potgietersrust,

Waterberg.
Carolina.

M. F. Arehibold & Co.

. IV. 458

Massicot in Quartz

E. J. A. Bierman

. IV. 459

Altered Quartz and Tourmaline

Naauwkloof No. 499,
Pretoria.

E. C. Clarke

. IV. 460

Manganese ore

Jachton No. 535, Lyden-

burg.
Rietfontein, Middelburg.

J. J. Robinson

. IV. 461

Antimon fahlerz. . .

L. Schultis. .

. IV. 462

Hematite rich in Tin

Elandsfontein No. 2149,

Waterberg.
Marabona No. 2550, Zout-

pansberg.
Lot No. 206, Haenerteburg.

C. de Vries

. IV. 463

Schist rich in Gold

L. A. Zeederberg ■

. IV. 464

Bed Granite with Pyrites con-

taining Cobalt

J. Gillespie

. IV. 465

Fibrous Hornblende

Klipfontein, Waterberg.

A. Barnes

. IV. 466

Asbestos

Rooikop, Standerton.

Gen. C. F. Beyers

. IV. 467

Molybdenite in Granite

Rietfontein, Waterberg.

G. L. Beechy

. IV. 468

Barytes

Donkerhoek No. 160,
Middelburg.

8. P. J. Daniel . .

. IV. 469

Molybdenite

Mapochsgr onden .

. IV. 470

Tondeldoos Spruit, Middel-
burg.

D. B. M. E., Pietersburg

. IV. 471

Caasiterite in Quartz

Delagoa, Waterberg.

J. van Walbeek . .

. IV. 472

Haematite and Limonite with

Goklenkoppie, Ruigte-

Fluorspar (rich in gold)

poort, Waterberg.

W. J. Hepplewhite

. IV. 473

Copper Glance and Vesuvianite

Stirum, Waterberg.

P. J. Porgieter . .

. IV. 474

Molybdenite

Zandfontein No. 2155,

Waterberg.
Eenzaamheid, Waterberg.

Mentz and Andrews

. IV. 475

Galena with Barytes

H. L. de van Scholtz

. IV. 476

. Molybdenite

Rietfontein No. 223, Pre-

T. G. LiUiecrona . .

. IV. 477

Amazon Stone

Waterberg.

M. Wagner

. IV. 478

.Coal

Wonderfontein Station,
Middelburg.

E. Cooper

. IV. 480

Molybdenite in Quartz

Rietvlei, Zoutpansberg.

T. Booth

. IV. 481

Molybdenite in Amphibotite

Ala map alios kraal, Pretoria.

L Garbe . .

. IV. 482

Cassiterite in Matrix

Mel moth, Zululand.

G. Weigel

. IV. 483

Antimonite in Quartz

Murchiaon Go Id fields.

H. J. Ford

. IV. 485

Sulphur Springs, Zoutpans-

164

DONOR. NO. IN COLLECTION. DESCRIPTION. LOCALITY.

H. 0. V. Jensen . . .. IV. 486 .. Limestone Witklip No. 29, Wol-

maransstad.

A. R. Osmond IV. 487 . . Magnetite and Htematite . . Kwaggershoek No. 759,

Rustenburg.
W. E. Oldfield IV. 488 . . Copper ore containing 300 ozs. Gevonden, Carolina.

Ag., 2 ozs. Ail per ton, and
11 per cent. Cu.

B. T. Bouike . . IV. 489 . . Nickel ore Mount Aliff, East Griqua-

land.

Cooper and Loch IV. 490 Corundum Rietvlei, Zontpansberg.

\V. 3. Whitworth . . . . IV. 492 . . Concentrates from Diamond Koffyfontein Diamond

washing containing Garnets Mine, O.R.C.
and llmenite

L. Handley . . IV. 492 . . Zincblende Edendale.

R. Recknagel — Specimens of Cassiterite, Rooiberg Tin Mine.

G. A. Troye — Old native ingot of Tin, picked up on Buffelspoort, north of the Rooiberg.
— Bar of Tin, smelted from cassiterite, from Vlaklaagte.
A series of rocks from the Premier Diamond Mine, kindly donated by the Company, illustrating
the occurrence of the Diamond on that property.

The collection includes specimens of the various kinds of " blue ground."

Micro-Sections.
The collection of Micro-sections was increased by the addition of 158 specimens, the whole of
which were cut and prepared in the Laboratory of the Geological Survey.

SAMUEL M. TWEDDLLL,

Curator, Geological Survey.

2. ADDITIONS TO THE LIBRARY.

Europe.
United Kingdom—

Summary of progress of the Geological Survey for 1907.

England and Wales —

Memoirs :— Nob. 126, 254, 283.

Scotland—

The Geology of the small Isles of Inverness-shire.

Ireland —

Parte of Sheet 28, 29, 36, 37, 143, 144, 186, 187, 194, 195, Drift Edition.

Sheet Maps 3, 5, 6, 7, scale 4 miles to 1 inch.

„ 16, 230, 231, 232, 246, 247, Drift Edition.
„ 230, 231, 246, 247, Solid Edition.
„ „ 125, 126 (236, 237, 253, 254), 254, 267, 268, 295 (351, 358).

Australia.
Victoria —

Geological Records. Vol. II, Part 4.

Western Ausiralia—

Geological Survev Annual Report of Progress, 1907-08.
„ " Bulletins 31, Part 1, 34.

South Australia —

A Review of Mining Operations, No. 8.

Asia.
India —

Records, Vol. XXXVI, Parte 3 and 4.
„ Vol. XXXVn, Part 1,

Socth Africa.

Transactions of the Geological Society of South Africa.

Cape of Good Hope —

Annals of the South African Museum, Vol. V, Part 6.

„ „ ,. VoL VI., Parts 1 and 2.

Vol. VII, Parts 1 and 2.
Geological Commission. Twelfth Annual Report for 1907.
Geological Commission. Sheet Maps 33, 41.

Rhodmo—

Sixth Annual Report of the Rhodesia Museum, 1907-08.

America.
United States—

Geological 8urvey Bulletins : —

No. 207. Surface Water Supply of the Upper Mississippi River and Hudson Bay
Drainages.

209. Surface Water Supply of Lower Western Mississippi River Drainage.

210. Surface Water Supply of Western Gulf of Mexico and Rio Grande Drainages.

211. Surface Water Supply of Colorado River Drainages above Huma.

212. Surface Water Supply of the Great Basin Drainage.

213. The Surface Water Supply of California.

214. Surface Water Supply of the North Pacific Coast Drainage.

215. Geology and Water Resources of a portion of the Missouri River Valley
in North- Eastern Nebraska.

216. Geology and Water Resources of the Republican River Valley.

217. Water Resources of Beaver Valley, Utah.
21S. Water Supply Investigations in Alaska, 1906-07.
319 Summary of the Controlling Factors of Artesian Flows.
322. Geology and Oil Resources of the Santa Maria Oil District.

326. The Arkansas Goldfield.
325. A Study of Four Hundred Steaming Tests.

327. Geologic Reconnaissance in the Matannska and Talksetua Basins, Alaska.

330. The Date of Geochemistry.

331. Portland Cement Mortars and their Constituent Materials.

333. Coal Mine Accidents ; their causes and prevention.

334. The Burning of Coal without smoke in Boiler Plants.
336. Washing and Coking Tests of Coal and Cupola Tests of Coke.
339. The purchase of Coal under Government and Commercial Specifications on

the basis of its Heating Value. Water Supply and Irrigation Papers.
Topographical Maps.

Oklahoma.

Bulletin No. 1.- Preliminary Report on the Mineral Resources of Oklahoma.

Geological Society of America, Vol. XVIII,

Acquired by Purchase.
Annales de Mines, Vol. XII.
Bulletin of the Imperial Institute, VI.
Centralblatt filr Mm. Geo. Pale, 1908.
Comptes Rendu, VoL CXLVL
Geologisches Centralblatt, VX.
Geographical Journal, Vol. XXXI.
Jahrbuch fur Min. Geo. Pale, 1908.
Journal of Geology, Vol. XVI.
Mines and Minerals, Vol. XXVIII.
Nature, Vol. LXXVII.

Quarterly Journal of the Geological Society, 1908.
Mineralogies! Magazine, Vol. XVI.
Geological Magazine, Vol. V.

Teche.rmak'e Hitteilungeo, VoL XXVI.
Pieterman's Mitteilungen.
Zeitschiift fiir Krystallographie, Vol. XLV.
Zeitschiift fiir Piaktische Qeologie, Vol. XVI.
Zeitschiift fur Analytisobe Chemie, Vol. XLVII.

SAMUEL H. TWEDDILL,

Curator, Geological Survey.

INDEX.

Note,— Tilt names of [arms, with the districts in which thej ore situated, i

Accessory minerals with tin . . 49

Actinolite 16, 89

Adinole 133

Adrians Kop 63, 68. 97, 98

Agents of metamurpbiBm 94, 134

Amphibole 93. 133

Amygdaloid • 147

Amygdaloidal f elsites 38

.l-il.il.i-..*.' 90. 93, 94

schists, Mount Mare . . 78

„ slate . 94. 132, 133

Andnes Spruit. SO

Anticline in Waterberg System . . 34

Aplito . . 20, 22

Apfnagndan WO, Wattrbtrg 5. 21. 29, 33. 36. 37. 38. 39. 40. 49

Ardennes. . 79

Area completed during the year 7

Aikose 1 10

Arsenical pyrites 49, 122

Atnhte 122

Backtu 830. Wattrbtrg 28

Banded ferruginous quarttite in dolomite 82

Pretoria Series 86

Ironstone* 110, 149

„ „ Monnt Mar* 76

Banken Scenery 60, 126

Bavtaamklool 2012, Wattrbtrg . 30. 40

Belvedere 04

Bethanie 99

Berets Conglomerate 88. 129

Beyers Nek 6a

BtUo*gfa*tn* 2840, Wattrbtrg 13,18

Biot lie .eiflimanite .cord ipnte- gneiss 94

Black Beef Series 13, 81, 148

Bt.nhmttr 707. Wattrbtrg 11. 20. 22. 29. 42

Bbxmhof 1672. Wattrbtrg 13

Blyde Hirer Valley. 87

ii-i t-j.i. ; 300. Manto. 127. 129. 130. 151

botpoort 37$. Wattrbtrg 28

BoompiaaU 1956. Zoutpatubtrg K3. v.i

Boone Siding 130

Boshof Mountains 108, 118

Breccia 37. 118. 119. 147

Brown's Adit 101

.. Reef 75.101

"Buck Reef " 64. 160

BugtUdoontA 1773. Wattrbtrg 30,31

Bifftttftrntttn 190. Marieo 1S1

BufftUhatk 264, Marieo 143. 132. 164

799. Hutttabwg 56

MAS, Wattrbtrg 67

fafftUUoof 704. 3 1 . 40

■V-.i ■-„:■•., Drift 105. .1/ ■<".<. 147

Building Stone 23

Busbveld Amygdaloid 31, 33. 3b. 46. 67. 96. 100

Granite 97

Plutonic Complei 17. 33. 41, 97, 120

Sandstone 31, 33, 3ft. 4ft, 46

Button Kop 64, 69, 81

Csloite 49

Cakite in the Dolomite 16

.. »eins 154

■Calico" rock 71,75

Cassiterite 4, 5, 22, 39. 49, 00. 121

Caves no

Ceratophyr . 147

Osruasite 152

Cbalcopjnite 150

Chert . . 149, 153

„ conglomerate 110, 113,150

Chiaatolite 151

Slate 84. 132. 133

Chritttna* Homt 524. l.irJurnburg 149

Chuniea Mountains 64. 65

Poort 63. 64. 65, 81, 82

Clapham 625, Lydtrtburg 99

■•Colorado" Reef 158

Conglomerata llo, 113

Conglomerates in Pretoria Series 87

„ Mount Mare . 76

Waterberg System 31, 33, 34, 64. 08

„ .. „ boulders with :• ir

maline in 41

„ .. „ contact with Gra-
nite, Uatkop 58

i, .. „ granite pebbles in 40

on Oroothnek 946 59

Cuntact-fels 17

Contemporaneous rucks 135

Copper 99. 102

.. ores of 2!

.. pyrites 49, 122

Oordisrite 17

contact.fels 131

Eieisses 66

imfeU 1 34

in altered I'ret.ina Shales 93,94

Crocodile River 105

Crystal Reef 158

CyltrfonUtn 85. Maruo 151

299. Marieo 130

Daisy Reef 158

Daapoort Quart lite 15, 17. 86, 129

Dt fleer 61.1. Zoetpaiubrrg 90

Dt Berg 1668. Wattrbtrg. ... SO

Dt /■>-• 19. '■/■».', 147

Dt Hnop 1670. Wattrbtrg 67. 83

Dt Puttrn 4, Marieo 144. 147

Dtlagoa 1337. Wattrbtrg 1 1. 22

Devils Kantoor 64

Day's Beacon 77

Diabase 21. 22,34. 43.44, 113. 148, 151

Dlallage IB

Diopaide 1 6. 20

niorita 136

Distribution of me (.amorphic rock 96. 96

Dolomite 13. 15. 17. 82. 110. 116, 129. 148. 151

Oatkop 69. 60

.. metamorphism of 88

Dolomittc marl. Potgietersrtut 91

PAGE

Pnntrrp—Tt 94S. RmOtalmrg 34. 55, 56. 58

Doorndraai 56. „ 182

379. Waltrbtrg 28. 34. 36

dam on 28, 34

Doorndr,/! 293, Rntltrtfmrg 133

Doomfontein 1956. Waterbtrg 66. 84. 85. 86, 100

Doomhotk 32, Maitro IffS

896, Waterbtrg 5. 31, 34, 36. 37, 39, 40, 41, 4». 50

Doomhoek fault 35

896. shales on 35

lin mine 35

DoonJtop 463. Zmlpatubrry 90

Doornttoof SOT. 102

Doontkraol 6, Martco . 136

Downpoort 1007. Rutlenburg 116

Ooornnvitr 65. Martco 126

fhomttodt 828. Waterbtrg 28

AjonutfVi 612. 7, Aitpa-neberg , 66, 100

Dorps River, Potgietersrunt . 67

Draaifontctn 84, Marieo 146. 151

Drainage. Marieo District 142

Drakeneberg 64, 81

Dritfantti* 90, Marieo 161

1547. Waltrbtrg 87. 73, 74. 77

Drusy cavities 49

Dutkerfonlei* 128. [.ieJutnburg 144

Dwaalbop S38. Zmtipantbtrg 66

Dwarebere 126. 127. 128

Dykes on Welgevonden 1772 and .Ian II... .r. Jan 481 45

Eastern Xeooecoenil&nd 89

Beitlem 1866. Waterbtrg ...22, SB, 42

Economic Geology 4. 5. 22. 46, 121, 151

BtrelegtMc 378, Waterberg 35

Striding 87. Zoolpanebtrg 72. 74. 100. 101. 102

Elandsberg 106. 117

Btande/otitei* 810. Rutltnburg 116

1782. Waltrbtra B

Elands River 126. 127

" Elisabeth " Reef 168. 159

Kostatite 19

Bnnelberg 1J0. 130. 131

Epidote 147

Euphorbia candelabra . . . 66. 92

Eveline Poort 7ft 77. 79

■' Fair Maid of Perth " . . 101

Fault, Doomboek 896 35

overt h rust, Gatkop 56, 57

,. Welgevonden 1772 36

Faulting 109. 115, 120, 146

Fault* 34, 36

Feleito 115. 118, 119

physical features. 30

Felsitas, amygdalotdal 36

Hoekbexgen 58

Lower Waterberg 32

Ferruginous shales 150

Field work, distribution ami general review of . , 3

Fine-grained granitt 42

Fluorescein test 68

Fluorite 49

Folding 1 16. 145

Forsterite . 16

Oabbro ..19. 20, 22

Oaten*. 23. 49. 152. 154

QarMt 16, 20

„ in altered dolnroitin marls. Pot gieters rust . . 91

„ in Pretoria shales S3

schists. Mount Mare 78

Gatkc-p A. 6. 40. 64. 106, 111

.. magnetite qoartxitee on 66, 68

'Tohmfer 201. Ruetenburg 129, 136

QeOtie 72

Geluk 279. Rustenburg 118

., 897. Waterbtrg 34. 41. 43

Oemtbokfontein 2123. Waterbtrg 36

Gtytrtpan 54. Rueteubvrg 127

Uodwan River K

Qotdknp 780. Waterbtrg 30. 11

Odd 50. 101, 102. 123

"Golden Butterfly " Reef 158. 159

" Golden Oalf " Reef 149. 158

Qoldfields, Malmani 5, 149. 157, 158

„ MarabaaUd 5, 100

03U 76, 78

Qondhntk 492. Zovtpatubtrg 90

Graoite. dykes of 16. 18. 22

Aoe -grained 42

kopjes 30

rilder 12. 13. 22, 146

prbblm of, in conglomerate* 40

pecudo- bedded 29

ranges 29

Red.6. 11. 17. 18, 20, 32, 33. 33. 41. 44. 47.97,120

„ contact with conglomerates 58

„ erteot of fid

., intrusive. Uatkop 55

relationship to falsites 42

.. Waterberg System . . 43

., tones In 20.21,42

Granophyre 20, 21, 146

on Romwal 317 44

(Sranophyric granite. 42

Orostifei 1094. Walerherg 86.91

Great Marieo River 99. 126. 127. 142, 149

"Greathead" Reef 158

" Green Cutting " Reef 168. 159

(froenfonltm 195. Waltrbtrg 34

871. .. 5,21.22,29.30.39.42.44.47,49

iTrorarfr. 810. Waletberg 21. 29. 3«. 37, 38. 12, 49

(hootdram 880. Zmttpaiubtrg 99

Croats Afdttling. Marieo 143

'■'■Ml:*!. 82. 142

301. Ruttenbvrg 128

(hoothotk 2S7. Waterbtrg 111. 114

■.'■,-

., 946, Ruetenburg 59

1992. Waterbtrg 100

Grnolhoek Type . . 9!

QrooOop 1961. ZvHtpnnebtrg 97

I -root oog van Marieo 142

ffroohnwer 2321. Waterbtrg 38. 60

Haenrt iMltorir 64, 81

GuldSelds 89

Hamilton 68

Hsnglip 28. 29. 30. S4, «

Hans Masibi's Location 12. 13, 18

llarUoeeHtonttm 949. RtuUnburg 59. 121

HartebtetUiUi 174. RvUnburg 122

Hatch 77, 79. 137

Hamatite 153

Hendrik Masibi's Ixxation 17, 20-22

Hex River 131

Hinden 91

Hoekbergen *.'

„ correspondence to ranges \ vv. of

Potgietersrust. "

„ geology of 5*

., topography 5!

Holland* Drift 113, Zoutpnnebtig 71, 75. 77, 101

Holmes 109. 132, 134. '35

HolmtaUigk 1349. Waterbtrg 16,18

//.. -,-i ».f -i • 706. ■ 21, 2!

Hornblende 17. IB, »

andesile '36

diabase 135

HomfeU 15. 17. 131. 151

Igneous Complex, Bushveld 17, 33, 41, 97. I»

Injected hornfels **

" Inselberg-Landsohaft " 85

Intrusive rocks in Transvaal System 97, 154

PAI1E

Iron carbonate 49

Iron Crown 64, «5. 68. SI

Iron Ore* 23, 36, 111

Ironstones, banded .110, 130. 149

Island landeoape 86

Jaoobadal 184

Jameson Road. 186

Jan flown -'■..< 481. Waltrbtrg 28

Jasper 149

Joriasen 99

KaalUtgtt 615. Waltrbrrg 34

Kaaiptaat* 97. Monro 14«. 152. 167. 160

Kaffrrtkraal 93 143. 169

Kafftrtkraal 214. .. 142

Ka/imrk 605. Zoutpantl-rg 99

KalkbuU 97

KametlbuU 551 97

Karrttlonltin 540. /(i*»' m j 129

Karrttlaaglt 81. Marico 147

Karros Spruit 132, 161

Karroo System 33. 35. 46, 96

Kaliktoof 563. Zoutpantbtrg 84. 87. 90

"l£eUy^■• Reef 169

Ktaaratroom 55. Manco 143. 144. 161. 166

KJein Marico River 136. 142. 143

KUiitfontti* 632. ZoHtpatubtrg 73

Kltpfanttt* 116. Waltrbrrg 46

773 28

719. 28

KUpgat 72. RvMtnburg 111. 113

Klxppan 194. Marteo 147. 148

KUpplattdnft 609. Waltrbrrg 29

Kt.pknil 210. Marieo. 142, 143. 149

Kttptpna 267. Waltrbtrg 40

„ 351. Zotttpatulvtff 73

Ki.-.i w schist*. Mount Mare 101

Kotdot»fonltin 309. Waltrbtrg 34. 35. 38. 46

„ 676, Runttnburg ; 127

K;(fykranl 619. Rvtttnburg 127

" Koh-i-noor " Reef 149. 168

Kopje Enkrt 533. Zoulparuhtrg , 100

Kopjes near Nabooraapruil 31

Kopjtedam 609. ZoHlpantbtrg 06, 91. 97

Kvpjtakraal 272. Marin, 136

K<*tUoot 639. Rtuttnburg 129, 131. 112

Roster Siding 126

Krondraai 692. Ruttrntmrg 120

KromeUtnboog 133. Marits, 129, 131, 132

Kromkloof 1637. Waltrbtrg 6

Krugersdorp 129

Kruidlonlttn 47. Ruiltnburg 106. 115. 119

Krvurintr 154. Maricn 146, 161

Kruiwege Store 67. 69. 84

h nUfonk ,:-. 82, Marten. . . 15!

Kvaggatkotk 243. Jtutltnburg 126

Kwaavatntk 603. PoUhtfMroom . 129

'Indies" Reef 158

Landtbwghotk 1633. Waltrbtrg 75.77

Langramt 1004. ftu.i..-.w , 115. 121

Latent* 46

Lava 147

Lewrstioe 98, 99

I«ad and iinc deposit*, ilincuiwion 155, 166

Lead ores fi. 23. 161, lfi2

Utuvjtop 528. Zoulpanthtrg . . 84

Lemberg 91

Leslie 99

Uydtn 580. Waltrbrrg 42

Limestone, surface 20. 23

LindUytpoort 102. Rualtnburg 130, 133

Linokana In. 149

Lttbon. Wutrrhrn 11. 42, 43

Little Hsrieo Rivor 13B. 142. 143

" Uttle William " Reef 168

" Little Wonder " Reef 168

Liverpool 522. Rnrttnbxrg 106, 116

Lombard 127

Longsight Type 94. 132

Loser Divuion. Walerbeig System 32, 36

Lulu Mountains ..66, 72, 98

Lydenburg , . . 84. 94. 95, 130

Machaohaan's Location 11

Magalakwin River.. ..11, 12. 18, 20, 21. 27. 28. 29. 32. 33

Magabee Station 129. 150

Magalieeberg Quartsite .15. 85. 112, 130

" M&gainte " Reef 158

Magnetic 14. 20. 23

qunrtxiti 13. 17. 19. 22

quartzitng of (Jatkop 56, 58

Makapans Mountain" 29

Makaparugal 1667. Waltrbtrg (17. 70. S3

Malachite 122

Mahktkraat 406. l#dtxburg. . 99

Mabps Unft 65. 84. 86. 88. 90. 92. 93. 95

Poort 66

River 81. 87. 88, 94

„ ,. type 94

Malmani ii.:"..l:..-!.1 5. 149. 157, 16B

River 142, 143. 149

Manganese earth. 162

Mapping, summary of 6

Maps issued during the year 7

Marabastad 78, 126

Goldfield 6. 100

Maraia' farm 61

Maroasite 166

Manco Distriot 141

„ „ lead and tine in S

mapping in 6

,. River 99. 126, 127. 142. 149

.. Station. 126

Matala a Location . . 11. 12

Matlapoodi 67. 100

Meinljea Kop 130

Memoirs issued daring the year 7

Mercury 162

Melamorphiam 114. 149. 161. 166

contact, of dolomite of Pretoria
Series-- 15, 16. 87. 90, 92, 94, 131, 132, 133, 134

Metamorphosed ahalee lit. 161

MtddtJkap 1861. ZotKpaastero 89

Mistnckel 165

" Mitrhelle " Reef 1 68

Moddtrlonltin 1803, Waltrbtrg. . 67. 136

Molengrsaff 87, 92, 99

Molopo River 142, 144. 148. 149

Moitgat 1660, Zoatpambrrg 67. 84. 93

Molybdenite 49

Mooiiie* Series. Mount Mare 74

Moodjadjee 128

-tfoortoe* 2044. Waltrbtrg 21. 22

Moorddr,/i 2098. 30. 33, 34. 37, 43, 63

Station , 28. 66

Morgtnzon 1876, Zoulpantbtrg BS

Motae River. . 90

Mount Mare 6, 63. 64, 68. 71. 101

Mount Mari Series 78

,. .. etratigraphy of . ,. 79,80

., ., structure of 74, 76

.. HuccexaioD 76 to 78

Mojtmbaiw 1336. Watrrhrrg 82

M'Phalleles fault. 68. 69

Location 66, 66, 92, 93. 96

Poort 65.81

River 64,66

MThlsuetsi 82.83

Mundt'a Store 73

V .,.,.,., *(,.. ' 616. Waltrbtrg 37

.Voouiopwrf 102. Manco 143. 152

521. ZoUpantbttg HA. 67, <t», 84, 87

Naboomlonttin 118. Waltrbtrg 41

Naboonupntit 2297, Watttbcrg 30, 31, 36. 46

k-p|-JiT l\r*t 31

Nelspruit. . . ...'.... 72

Nooithulp 89, Zondpnrubtrg 101

Sootlgtdacht 155. Marieo . . . 126

1604. Hu/Unlw, 116

S ••wwkt 816. WaUrbrrg 30. 34, 42

Norite 15. 17. 20. 97. 115. 120. 136

gneiasio structure in 99

.. magnetite in 90

Northern Lima Works 70, 83, 89

Notwani River 142, 144, 149

Nyl River 33

., Valley 31

Nylstrooro 6. 31

Ohrigstad 86

Valley 136

Old Lead Mine on Rhenosterhoek 211 153

Old Smiladorp . . 79

Older flranile 12. 13. 22, 72, 73. 74. 146

(Hievtvbonek 939, RutUnburg I IB, 121

Olievenkloof 320. „ 1 33

OlttvenpooH 758, „ 118. 120

OlifanUklip 1325. WaUrberg 46

Olifanta Nek 128

Poort 66. 98

River 63, 66

Olivine 16. 20

Omvrwarto 294. Rv»t*nb*rg 119. 122

Oog van Malmanie 101. Ma'%co 1(12

Oolitic dolomite 83

OorlngtfiHttein 1S17. WaUrbrrg 66. 85. 86. 87. 91

Orange drove Series 79, 80

Ortblte 49

Ottnahoop 143. 149. 157

Otlrelite Wnfela 134

in altered shales 183

- I.i-'-. Mount Mar* 77

Ouklip 46

Paarddritt 1017. WaUrberg.
Paardevallei 62. Marieo- . . .

Palala River

Palmiel/onUin

Pepenius

Paasarge 66,

Pegmatite 18, 20. 21. 42.

Phlogopite

PhylAtee. etc.. Mount Mare

Pjenaars Nek 65. 69. I

Pilgrims Reel 87. 1!

" Pillar '• Reef II

Pipes, tin-bearing <

PUmhtri 1544, WaUrberg 67, I

., springs I

Platleau s Store 68, 11

Pnlet Series )■

Poogola River 27, I

Porphyry !•

Portjt 1968. Zovtptmtberg I

Portugal 1066. WaUrberg 65. I

Potgietersruat 5. 11, 15. 17. 19. 23. 29. 42. t

Pramkop I

Pronlcopjt* 607. Zoutparuberg 66, 91. 97, 1

Pretoria Qnartzite oountry between Zeerust and

Zwartruggens II

Pretoria Quarui!e, tnetamorphism of !

Series 13. 15, 17. 20. 40. 83. 88. 112. 116. 117. If

„ „ conglomerates in I

„ tnetamorphism of I

Shales, pavement structure in I

„ pyroxene in altered 93, II

unaltered f

Praiitn 1536. WaUrberg 85.86.87,91,1

Pyrites, arsenical •

| Pyrites copper 49

I Pyroroorpbite 151

I Pyroxenite 15, 17, 19

I Publication* "

Quart! Reefs 143, 149. 158

I ., schiste, etc., Mount Mar* 77

1 Quart-ale 112. 113. 118, 151

Zeerust 145, 160

Quartiites, Blaok Reel 148

Quartettes of Pretoria Series IS

Rabe 101

Ramaknka Mountains 120

Stadl 107

Ramatuiebama Spruit 142, 146

Red Granite, tue, 5, 11, 17, 18.20.32.33.36.41,44,47,97,1%

extent of 60

„ „ intrusive. Oatkop 38

,. relations of the 4. 42

„ ,. relationship to (elsites 4!

rVaterberg System... 43

aonee in 20. 21. 4!

HktnmterfoHlttn 13. Marieo .129, 130

143. I4S

38

..71. 72, 73. 75. 102

83.

\x

RhenotUrhotk 211. Marteo 152, 153

Rhenotter poort 2145. WaUrberg S. H

Riekertdam 203. Marin, 128. 129. 131

Hu-v.i«;.-'.if 1713. WaUrberg U

RiHfouUin X 31. 35. 38. 4*. 4«

176. „ . 29. 44

1426. 30. 34

1638. ,. 31. 32. 33. 34, 35. 39. 41, 4fl

„ „ warm spring on 33

1669, „ 13,20,2:

Ritlpoort 95. Maneo 143. 157. IS"

RitUpntit 304, Maneo 127, IS

Rtrfirt 5, Marito 143

.. 94. Zoutpantberg 73. 7»

.. 1348. WaUrbrrg *i

RtetvatUi 91. Maneo 131

Ripple^ marking 38

Rinerploat* 674. Zovlpatubera K

Road metal 23

RvndavrUkmal 420. RxtUnbary 127, 133

Rondepvort 8S. Zoutparuberg 74. 75. 79, 80, 101. 10!

813. WaUrberg 21,30.33.38

819. 37.39

2148 41.64

Rooiberg S

Sedimentary Series H'

Tin Mine* 1*1

Rooibotlaagtt 504. Zoutpaneberg 93

Rooigrond 1*'

Rooipoorl IS. WaUrberg 31, 39, 40

,. 1639 86.91

R-oiwal 143. Rudenlntrg H*

317. WaUrberg 29. 34, «

„ granophyre on **

Rookpnnrt 1835. WaUrbrrg M

Ruatenbuig . '3'

Dislncl, mapping in the *

Rijkvovrbij 73. Marieo 120, I3«

Ryolite W

Sabie 75

Sand River. Rustenburg District ' w

vinaV,-, - Poort 342. RiuUnbvrg HO

Suiufeprtitf 944. Martto 1"

Sandfontrin 2115. WaUrbtrg 108, 1"

Sandstones. Waterberg 28. 29, 30. 38. 41. «■?

Sawyer *•

' ■ 736. Znwpaiubrrg ™

Schaapptaat* 7
feheetit*

Seheerpoort 260. Pretoria

Sehirtkraat 332, Maneo

HthMpad 520. Zoulpaiuberg.

PAGE
Schoonoord 99

.VrAl .iiWm" 2S.J, .1/ inc. 130

Schnrre Hergeti 126

Secnecoeni's \, - .1 .< « 08

Sccnecoeniland 84, 94

Scphtnns Nek 126, 136, 146

Serpentine 15. 17, 19. 20

Shales 113. 151

alteration of. in Dolomite of l*reU)ria Svnes 16. 17

„ Black Reef 148

metamorphosed 114

on Doornhoek 896 36

Lower Walerherg 32. 37

I'pper 69

Khepstone ... 67

Sieve structure in altered dolomite 69

Sillimanite 90. 93. 94

Slagtkraal 741. Al-.w.w i 106. 115, 116

SUckriwidn ">3

Klynsteen Drift 67. 96

., sandstone . 97,100

Slypntetn Kup 3t!3. !(:-t- :!■«■ > 133

Smits Drift 64,72

Sngmamdrift 4. Zoulpantbtrg ,. ,64. 71. 75, 77, 100

Solomon* Temple 1014. Waterberg .. 21. 29. 30. 37. 38. 42. 43

Spanje 1067, Waterberg 67

Speeularite. 60

Spelonttn 1872. Zoalpand. ;• , . . 66

Springbok Plate 30. 31. 33. 36. 66. 66. 96. 100

„ water supply. 67

Spring" 143. 144. 149

Staanpiaats 368. Zoutpaiuberg 82

Steel poort River 86

Kteenkopjea District 129. 130

Stembotfontem 570. Ruttenburg 132

Sterklemtem 1531. Waterberg 28

Sterk River, Rnstenburg Diterict 106

Waterberg .. 11.20.27,28,29.30.32,38

.. Shot** 38. 41

Sterkspruit Reel 130

Sttrkttroom 372. Waterberg 2n. 30. 34, 36. 41

897. Rtutenburg 133

Sterleaater 1549. Waterberg .20. 22. 30

Sterkwater Spruit 30

Stewart 85

StiKthouleoom 269. Marieo 143. 168

Stof poort 550, Zoutpaiuberg 97

>!. ; nibcrg Series 97

StroaUdn/t 977. It;.:. :>..■•■,■ 136, 137

UlrydfonUm 267. Marteo 152

Slrydpoort Range 63. 64, 66. 81

Superficial deposit* 100

Swaziland System 5. 6, 74

Syncline in Waterberg System 34

Tafelkop (Devils Kantoor) 64

(Ruatenburg District) 129

TesJL Dr.. referred to 16

Tt'pvltlontrtn 209. Rudenburg 128

Timebat] Hill Beds 112. 143. 149

,. quartsite 15. 17. 84. 129

Tin 22. 42. 46. 47. 121

„ accessory minerals with 49

Deposit*, probable period of origin of 43

Tobias Spruit 31. 46

/■ -ifliiM'-'i/M-t 228. Rvrtenbiay 136

Toiani River 126. 127

Tnpai 40

Tourmaline 37. 49. 60. 76. 77. 90. 93. 121. 131. 132

Toumlali nidation 37

Transvaal System ..12. 13. 17. 68. 69. 70. 108. 148

Tremolite 16. 140. 152. 155

Tufa 37. 147

Turfaugtr 1967. Zoutpaiuberg 66, 74

Tme/onlein 64. Marieo 144

1033. Waterbtrg 13, 10, 16. 19. 20, 23

Tay/elpoort 899. BtuUnburg 133

Tygtrpoort 542. Zuulparuberg 84

Uitkgk 1793, Waterbtrg 66, 67. 77

Vitloop 1543,

13, 14. 16. 17, 19. 22. 23. 64. 07. 74. 75. 82, 89, 90

I'illoop lime Works 63

I'uvltuM 63, Marieo 143. 144. 145, 161

I'noonformily in Waterberg System. 33. 39

Vndergruund water 143

I'pper division, Waterberg System. . , , 32, 33

Vaalkop 384. Jtuetenburg 118

1439. Waterberg 12. 13. 18, 20, 22

Vaalwatex 106

Valtyn Macapsana [location. 18

Van Lennope Kopje 87

Vanadimte 152

Vetplaau 143. Marteo 131

Vegetation. Bushveld 107

Ventersdorp System 7. 147

Vergenotgd 3. Marieo 143. 144. 145. 151

„ 46. „ 146, 161. 165

" Vesicular Diabase " 12ft

Vitre*turintig Rtviertn 1540. Waterberg 70. 85. 87. 91

Vitthoat 2121. Waterberg 32

TlelfmUin 2187, Waterberg 45

Vlakptnatt 621, Ruttmburg 133

Vleiptaatt I960. Zoutpantberg 6fl

Vleis 143

Vriechgticaagd 100. 7*mtpa<Mbeeg. . . .72. 74, 100. 101. 102

Voghs 49

Waebienbtetje Drift 326, Ruelenburg 105. 120

Warm springs. 32, 35

Waterberg District, mapping in the 6

Plateau 27. 28. 29. 31. 32

Sandstone .28. 29. 30, 38. 41. 42

System ..35, 116, 119

., anticline in 34

.. composition of 39

., conglomerates 31. 33. 39. 64. 58

„ above the base. . 40

,. .. boulders with

tourmaline in.. 41

.. granite pebbles in 40

vertical in 34

,. feWtes 32. 36. 68

Lower Division 32.36

,. overtilted on Rookponrt 1835... 64

Waterval 965 68

relations of the tipper and Lower

Divisions 39

relations of, to Red (iranite 4

section on Waterval 966 53

,. shales 32. 35. 36. 37

syuoline in 34

unconformity in 33. 39

I'pper Division 32.33

Water supply 12, 31, 67, 6B. 126. 127, 142

WaUrvai 211. ZoBtpatubero 72.74. 100. 101

316. Waterberg 2fl

720 30. 31. 40, 43

965. Ruttrnburg 41. 56. 58. 110

Waterval Boven 87

Weenert 1427. Waterberg 66. 67. 74. 75. 83

Welgelegen 357. Wattrberg 29. 30. 33, 44, 4B

Welgtvottden 777. 30.43.44.49

7772. 31, 32. 34. 36, 37. 41. 49. 60

„ ,. warm apring on.. 32

dykes on 45

Welgevonden tsull 35

Weltevreden 140. Zuutpnntberg 64. 75

Wtlverdiend 48. Marieo 136

Weynek 763. Riutenimrg 6. 122

Wildebeettfonleia 89. Zoutptmtberg 101

Windheuvel 205. Maneo 143. 167. 159

WiMerhotk 287. 146, 160

Witfontein Mountains 108

WHforUrin 371. Waterberg 38

Witkop 76. 77

Wtikap 298. Uanco... 152, 154

Wtlpoort 1909. WaWbtrg S4, BB, 87. 91

Witpoortje 79, SO

WUpoonjt 167. Mane, 131

Wolframite 49

Wolkberg 64, 71

Wollaatooite 16, 90

Wohxtoppttt 104. Atarteo 143

HWer/m/fui 78 Ml. 143, 144. 149

WowUrhotk 192. 143

Wonderkop 06, «». 84. 86

Woodbine Hiding 120

Wulfl 101

\MtberR 63. 64. 67. 70. 81

fanlt 70

tuatUool 631. Zonlpatubtrg 66. 97

Zaaiptaatt 2086. Walerbtra. ,5. 21. 22, 37. 3K, 39. 42. 47. 40

ZanenJeomt 88. Matieo 128, 143

ZandfatiUm 390. KvHrnt.w, 128

Eandnner 139. Zovlpaiuhtrg 64, 76, 100

PACE

Zandri vier Siding 64

Zandrivier Mynpachl 136. Zoutpaiubtrg 77, 101

Zandfprt.it 782. Watrrbtrg 37. 43, 44

953. KuAlrnfiur,/ 60

Zebedelae Kraal 68

„ Location 65

Riw .'...63. 64. 65, 81

Zeeniat 7. 84. 143. 144, 161

„ quartiitea 146, 160

ZtnddingipinaU 69. Marin, 129, 136

Zino 161

Zinc and lead drpomta. diacusaiun 165. 156

Zino-blcnde 49. 152. 154

Zondagajian 44. Martco 1 48

Zondtrnaam 52S. ZotUpaiubtrg. 67. «»

Zones in th.- R«l Granite 20. 21, 42

ZoutpanaberR Dint net, muppinR in the 6

Zwarlbult 1675. Zoittpa**b*rg 66

Ztmrtfonttin 1174. Walerhtrg 18, ID

Zioartkraiu 109$, 67, 74, 76

Zwartruggen* 126, 126

Zyletjcmttin 85. Maria, 141

173

PUBLICATIONS OF THE GEOLOGICAL SURVEY OF THE TRANSVAAL.

The publications of the Geological Survey of the Transvaal consist of —
- (1) Annual Reports.

(2) Memoirs.

(3) Sheet Maps (with accompanying explanations).

77w following Annual Reports hare been issued: —

Annual Report for 1903— 48 pp. and 24 plates (including one map) — deals with portions
of Pretoria and Middelburg Districts. Price 7s. 6d.

Annual Report for 1904 — 79 pp., 24 plates, and two maps — deals with portions of Pretoria,
Middelburg, and Waterberg Districts. Price 7s. Gd.

Annual Report for 1905 — 114 pp. and 31 plates (including five maps)— deals with portions
of Pretoria, Middelburg, Lydenburg, Rustenburg, Waterberg, and Barberton
Districts. Price 7s. fid. Out of print.

Annual Report for 1906 — 140 pp. and 37 plates (including seven maps) — deals with
portions of Pretoria, Witwatererand, Middelburg, Lydenburg, Carolina, and
Rustenburg Districts. Price 7s. fid.

Annual Report for 1907 — 116 pp. and 10 plates (including three maps) — deals with
Potchefstroom, Lydenburg, Zoutpansberg, Rustenburg, Witwatersrand, and
Waterberg Districts. Price 7s. fid.

Mem nir» published; —

No. 1. The Geology of Pretoria and Neighbourhood. By A. L. Hall. 55 pp. and
14 plates, with map. Price 10s.

No. 2. The Geology of the Komatipoort Coal-field. By H. Kynaston. 55 pp.,
7 plates and 2 maps. Price 7s. fid.

No. 3. The Geology of the Transvaal Coal-measures, with special reference to the Witbank
Coal-field. By E. T. Mellor. 60 pp. and 16 plates (including map). Price. 7s. fid.

Report on a Reconnaissance of the N.W. Zoutpansberg District. By T. G. Trevor and
E. T. Mellor. 40 pp. and 16 plates (including one map). Price 2a. fid.

No. 4. The Geology of the Waterberg Tin-fields. By H. Kynaston and E. T. Mellor,
with a chapter on their Economic Aspects by U. P. Swinburne. 124 pp. and
14 plates (including two maps). Price 7s. fid.

Sheet Maps are published in uniform size (16 x 24 inches) and are each accompanied by
an Explanation. The following have been issued : —

No. 1. Pretoria (with Explanation by H. Kynaston).

No. 2. Piensars River (with Explanation by H. Kynaston).

No. 3. Middelburg (with Explanation by E. T. Mellor).

No. 4. Rustenburg (with Explanation by W. A. Humphrey).
Price 5s. each complete.

The above publications are to be obtained from the Government Printer, Pretoria,

ANORIES SPRUIT

SPRINGBOK FLATS.

+ + -!- + -» + +- + + + + + + +■ + '-+ + +^»^^^^

TS near NABOOMSPRUIT.

T»T«¥ ">- » "¥

+ +■*-+ + ■*■ + + + * *- + + + + + ♦ V4 + 4- + + + + + + + + + + + + + + VT

* | t . 4 . T .'",'." t .*.*.''. < .-.* -' f -'*' .• f . + l*4- -4- .f- * 4- ■> .. fj X * ■+ * + «■ * * * + *

OORNHOEK 896.

Ooornhoak 896 -

rn

+ H RcdGrrmte

!
I

red

■nera

(ON BETWEEN THE BUSHVELD IGNEOUS COMPLEX AND

;TION 111>i MILES.

MagaJlMbrg, Magali— Mo*.
*bottg#jfa ri t stunkoneb 101.

Horitontal Soalt 1:476000. (Jeppe't map.)

9 10 I*

Vtrtical nlitf sketched.

ST.

NNE.

NOOITOEDACHT IBB.
haiumt. unu *4»m£o Aivtm.

e

"*""^*^r^T^S

EIGHBOURHOOD OF LINDLEY'8 POORT.

8. N.

BOKKRAAL 300! OIIOOTFONTEtH SOI.

"o'u rotuim mim**m-ttt*im

D \ ;

Ff r "^fer^U^

[_ ^LW < ^

t=^ '*S

Ooltmlto Mr,

frirtt/t* 1 a

SWH. I |

i I JJbrW *•**■ f 1 F 1 (IMMM. f 1 OMut.
I | S/iitot *«* bur J k. i I 1

Gmtlanmt*. J

"6. f a

i/ !

noBr ■ '

DATE DUE

L_

STANFORD UNIVERSITY LIBRARIES

STANFORD, CALIFORNIA

94305

3 110S 013 111 151

•573
i*S."-?n ?1KTW scwnci-s Usui*? > c iC) < ?

H

Ms.7j^l<£

3 tlDS D13 111 tSt

\ J

"S13

4

Ms.7|2 l il<£